Tese Jpm Final 20062014

Iv-Oil & Gas courtesy

Self-Installing Offshore Platform Investigation of Feasible Structural Design Improvements

João Pedro Pedroso de Lima Ferreira de Matos

Thesis to obtain the Master of Science Degree in

Structural Engineering

Supervisors: Prof. Dr. José Joaquim Costa Branco de Oliveira Pedro MSc. Etienne Hubert Boender

Examination Committee President: Prof. Dr. António José da Silva Costa Supervisor: Prof. Dr. José Joaquim Costa Branco de Oliveira Pedro Member of the Committee: Prof. Dr. Pedro António Martins Mendes

June 2014

This page is intentionally left blank

Thesis Acknowledgement Foremost, I would like to express my sincere gratitude to my supervisor Etienne Boender for all his trust when he give me the chance to be part of Iv-Oil & Gas and the opportunity to work in an enthusiastic thesis subject; and my co-supervisor Professor José Oliveira Pedro for accepting the challenge to guide me and help me to navigate into unknown waters. I am beyond words for their continuous support on my work, encouragement, patience, motivation, enthusiasm and enormous help and sacrifice.

Besides my supervisors, I would like to thank Stefan Beukers for his great knowledge and experience in my assistance, guidance, patience and sacrifice in reviewing my report in his free time.

I thank my co-workers for their help and stimulating discussions. “It has been truly gezellig” .

I would also like to thank Oene Dijkstra and Mark Haarsma (from SPT Offshore) for their insight help in the Self-Installing Platform (SIP) and the suction buckets.

Last but not least, my special gratitude and love goes to my parents, my sister and my girlfriend that never stop supporting me, in particular during the toughest times. Without them my stay in The Netherlands and conclusion of this Thesis would have not been possible. Their patience, help, understanding, encouragement and love is my biggest treasure.

i

Abstract Offshore platforms have been used since the 60’s (and impulse in the 70’s by the oil crises) in the exploration and production of oil and gas in the North Sea. Offshore structures are usually fixed platforms supported on pile foundations, built primarily on steel. These structures have not been economical feasible to explore marginal fields in the North Sea, mainly due to its high costs of installation and removal. Thus, an innovative self-installing platform , capable of being easily relocated when the field runs dry, was designed and built for the Dutch sector of the North Sea. However, as usually occurs in the developing projects, the self-installing platform still faces some engineering issues, mainly when it comes to support the impact of cyclic environmental loads, such as winds and waves. The lack of legs bracing and the fact that the entire structure (topside and legs) work together as a portal frame results in: (a) a decrease of the first frequency modes that affects the structure fatigue life; and also that (b) the fatigue loads go through to the structure topside. This study presents and analysis several variant structural solutions for the self-installing platform, using the F3-FA platform project layout, both the design basis and the geometrical configuration; and incorporating different key structural solutions in the design with the aim of improving its dynamic behaviour and, by consequence also, its fatigue performance.

Keywords:

Offshore structure, Self-installing platform, SACS software, wave action, spectral fatigue analysis.

ii

Resumo Plataformas offshore têm sido utilizadas desde os anos 60 (e impulsionadas nos anos 70 pelas crises do petróleo) na exploração e produção de petróleo e gás natural no Mar do Norte. Estruturas offshore são tipicamente plataformas fixas fundadas em estacas, constituídas essencialmente por aço. Estas estruturas não têm sido economicamente viáveis para explorar campos com acumulações marginais de petróleo e gás natural no Mar do Norte, principalmente devido aos elevados custos de instalação e remoção. Assim, uma inovadora plataforma auto-instalável ( self-installing platform ), com a capacidade de ser facilmente relocalizada quando o campo de petróleo e gás natural secar, foi projectada e construída para região Holandesa do Mar do Norte. Porém, como normalmente acontece em projectos em desenvolvimento, a plataforma auto-instalável enfrenta ainda alguns problemas de engenharia, sobretudo quando sujeita às acções cíclicas do ambiente, tais como ventos e ondas. A falta de pés contraventados e o facto de toda a estrutura (topside e pés) funcionar em conjunto como um pórtico têm como consequência: (a) uma diminuição dos primeiros modos de vibração afectando a vida da estrutura à fadiga; e também (b) a inclusão de acções de fadiga no topside da platforma. Este estudo apresenta e analisa várias soluções estruturais para a plataforma auto-instalável, utilizando como base o projecto da plataforma F3-FA, quer as bases de projecto e a configuração geométrica; e incorporando diferentes soluções estruturais chave com o objectivo de melhorar o seu comportamento dinâmico e, por consequência, o seu desempenho à fadiga.

Palavras-chave:

Estrutura offshore, Self-installing platform, SACS software, acção das ondas, análise spectral da fadiga.

iii

Table of contents 1

2

3

Introduction .....................................................................................................................................1 1.1

General considerations .......................................................................................................... 1

1.2

Main objectives....................................................................................................................... 4

1.3

Thesis organization ................................................................................................................ 6

F3-FA platform general description .................................................................................................9 2.1

Introduction............................................................................................................................. 9

2.2

Geometry.............................................................................................................................. 10

2.2.1

Topside......................................................................................................................... 11

2.2.2

Substructure ................................................................................................................. 14

2.2.3

Foundation.................................................................................................................... 15

2.3

Suction buckets .................................................................................................................... 16

2.4

F3-FA leg-topside connection .............................................................................................. 18

2.5

Construction, mating, transportation and installation............................................................ 19

Re-built F3-FA model ....................................................................................................................27 3.1

General considerations ........................................................................................................ 27

3.2

Model geometry.................................................................................................................... 28

3.2.1

Topside structure.......................................................................................................... 30

3.2.2

Substructure ................................................................................................................. 32

3.2.3

Leg-topside connection ................................................................................................ 33

3.2.4

Bucket foundation......................................................................................................... 35

3.2.5

Sections........................................................................................................................ 35

3.3

Design actions...................................................................................................................... 38

3.3.1

Introduction................................................................................................................... 38

3.3.2

Permanent loads .......................................................................................................... 39

3.3.3

Live loads ..................................................................................................................... 43

3.3.4

Environmental loads ..................................................................................................... 45

3.4

Basis of analysis................................................................................................................... 55

3.4.1

General bases .............................................................................................................. 55

3.4.2

In-place analysis........................................................................................................... 62

iv

3.4.3

Fatigue analysis............................................................................................................ 62

3.4.4

Load combinations ....................................................................................................... 72

3.5

4

3.5.1

Frequency modes......................................................................................................... 79

3.5.2

Loads............................................................................................................................ 82

3.5.3

Fatigue analysis results ................................................................................................ 86

3.5.4

Conclusions............ ...................................................................................................... 91

Study of possible structural improvements....................................................................................93 4.1

General considerations ........................................................................................................ 93

4.2

Knuckle joint ......................................................................................................................... 93

4.2.1

Introduction................................................................................................................... 93

4.2.2

Model without the knuckle joint and centric suction buckets ........................................ 94

4.2.3

Results ......................................................................................................................... 98

4.3

6 A.

v

Model with different leg diameters...................................................................................... 101

4.3.1

Leg diameter increased based on the slenderness ratio.......... .................................. 101

4.3.2

Results ....................................................................................................................... 101

4.4

5

Comparison results .............................................................................................................. 79

Leg-topside connection ...................................................................................................... 103

4.4.1

Model with adapted leg-topside connection ............................................................... 104

4.4.2

Model adapted for transportation on barge ................................................................ 107

4.4.3

Models with different leg diameter.............................................................................. 110

4.4.4

In-place analysis......................................................................................................... 114

4.4.5

Fatigue analysis.......................................................................................................... 120

4.4.6

Final remarks and conclusions ................................................................................... 122

Conclusions.................................................................................................................................123 5.1

Final considerations ........................................................................................................... 123

5.2

Future work perspectives ................................................................................................... 124

References ..................................................................................................................................127 Wind loads ..................................................................................................................................129 A.1

Wind speed and wind force calculations ............................................................................ 129

A.2

Wind load results................................................................................................................ 130

A.3 B.

Wind area loads ................................................................................................................. 131

Wave Loads ................................................................................................................................132 B.1

Airy’s linear wave theory .................................................................................................... 132

B.2

Stream function wave theory............ .................................................................................. 135

B.3

Static wave analysis ........................................................................................................... 137

C.

Hydrodynamic loads calculation..................................................................................................150

D.

API buckling and stress check of cylindrical members................................................................163

vi

Table of figures Figure 1.1: Jacket platform example [LEFT] & SIP platform example [RIGHT] .............................................................................. 1 Figure 1.2: F3-FA platform .............................................................................................................................................................. 2 Figure 1.3: F3-FA platform installation sequence:........................................................................................................................... 3 Figure 1.4: Jacket platform .............................................................................................................................................................. 4 Figure 1.5: Self-installing platform (SIP).......................................................................................................................................... 4 Figure 1.6: Geometry overview of the F3-FA platform (Iv-Oil & Gas courtesy)............................................................................... 6 Figure 2.1: Location of the F3-FA platform (adapted from [/28/]) .................................................................................................... 9 Figure 2.2:F3-FA platform 3D overview (Iv-Oil & Gas courtesy) ................................................................................................... 11 Figure 2.3: F3-FA topside SACS................................................................................................................................................... 12 Figure 2.4: F3-FA topside side view (Iv-Oil & Gas courtesy) ........................................................................................................ 12 Figure 2.5: F3-FA topside principal decks plan view & longitudinal side view .............................................................................. 13 Figure 2.6: F3-FA suction bucket and ........................................................................................................................................... 14 Figure 2.7: F3-FA suction buckets, transition frame and truss frame (SACS model).................................................................... 14 Figure 2.8: F3-FA substructure leg in yard (Iv-Oil & Gas courtesy) .............................................................................................. 15 Figure 2.9:F3-FA suction bucket (Iv-Oil & Gas courtesy).............................................................................................................. 15 Figure 2.10: Gullfaks C GBS platform [LEFT] and Draupner E jacket platform [RIGHT] (adapted from [/26/] & [/24/]) ................ 16 Figure 2.11: Suction bucket installation......................................................................................................................................... 17 Figure 2.12: Leg-topside joint ........................................................................................................................................................ 18 Figure 2.13: F3-FA top deck – leg connection .............................................................................................................................. 19 Figure 2.14: F3-FA superbolts....................................................................................................................................................... 19 Figure 2.15: F3-FA cellar deck – leg connection........................................................................................................................... 19 Figure 2.16: F3-FA clamp sy stem being installed ......................................................................................................................... 19 Figure 2.17: F3-FA topside construction (Iv-Oil & Gas courtesy).................................................................................................. 20 Figure 2.18: F3-FA topside without corner nodes ......................................................................................................................... 20 Figure 2.19: F3-FA corner nodes built separately ......................................................................................................................... 20 Figure 2.20: Suction bucket top plate............................................................................................................................................ 21 Figure 2.21: Suction buckets transport.......................................................................................................................................... 21 Figure 2.22: Knuckle joint .............................................................................................................................................................. 21 Figure 2.23: Suction buckets attached to the knuckle joint ........................................................................................................... 21 Figure 2.24: F3-FA legs attached to topside (Iv-Oil & Gas courtesy ............................................................................................. 22 Figure 2.25: F3-FA platform being transported to the Heerema H-541 barge (Iv-Oil & Gas courtesy) ......................................... 22 Figure 2.26: Self-propelled modular trailers .................................................................................................................................. 23 Figure 2.27: F3-FA platform rolling over to the BOA 35 barge...................................................................................................... 23 Figure 2.28: Matador 3 with suction bucket and knuckle joint (Iv-Oil & Gas courtesy) ................................................................. 23 Figure 2.29: F3-FA platform completed (Iv-O il & Gas courtesy) ................................................................................................... 24 Figure 2.30: F3-FA platform during sea transport (Iv-Oil & Gas courtesy).................................................................................... 24 Figure 2.31: F3-FA platform tender during installation (Iv-Oil & Gas courtesy)............................................................................. 25 Figure 2.32: F3-FA platform during installation (Iv-Oil & Gas courtesy)........................................................................................ 25 Figure 2.33: F3-FA platform superbolts connection (Iv-Oil & Gas courtesy)................................................................................. 26 Figure 2.34: F3-FA clamp system installation (Iv-Oil & Gas courtesy).......................................................................................... 26 Figure 2.35: Drilling rig Noble Scott Marks drilling first well in the F3-FA sector [/23/].................................................................. 26 Figure 3.1: Original F3-FA SACS model [left] and Re-built F3-FA SACS model [right] ................................................................ 28 Figure 3.2: (a) Re-built F3-FA SACS model with main parts and critical joints highlighted........................................................... 29 Figure 3.3: F3-FA SACS model primary framing structure............................................................................................................ 30 Figure 3.4: Re-built F3-FA topside SACS model main decks: cellar deck [right], main deck [center] and top deck [left] highlighted ....................................................................................................................................................................................................... 31 Figure 3.5: Re-built F3-FA SACS model longitudinal truss row 1 (3D view and section view)...................................................... 31 Figure 3.6: Re-built F3-FA SACS model transversal truss row A1 (3D view and section view).................................................... 31

vii

Figure 3.7: Re-built F3-FA SACS model transversal truss row A [left] and row C [right] .............................................................. 31 Figure 3.8: Original F3-FA topside SACS model with mezzanine deck and intermediate deck highlighted [in red] ..................... 32 Figure 3.9: Re-built F3-FA substructure SACS model .................................................................................................................. 32 Figure 3.10: Re-built F3-FA SACS model leg-topside connection [different perspective views]................................................... 33 Figure 3.11: Plot of the leg topside connection (e.g. leg E1)......................................................................................................... 34 Figure 3.12: Re-built F3-FA SACS bucket foundation................................................................................................................... 35 Figure 3.13: Environmental analysis dir ections (adapted from [/2/]) ............................................................................................. 38 Figure 3.14: Crane operational load directions ............................................................................................................................. 45 nd

Figure 3.15: Comparison between the linear wave model with a non-linear wave model (in this case the Stokes 2 order wave) ....................................................................................................................................................................................................... 48 Figure 3.16: Regular wave shape ................................................................................................................................................. 49 Figure 3.17: F3-FA tides and storm surges................................................................................................................................... 51 Figure 3.18: API & ISO procedure for calculation of deterministic static wave & current forces (taken from [4]).......................... 52 Figure 3.19: Current force diagram on the wave crest and wave trough ...................................................................................... 53

Figure 3.20: Typical double sided V-groove weld ......................................................................................................................... 58 Figure 3.21: Typical wall thickness transition ................................................................................................................................ 59 Figure 3.22: Spectra fatigue analysis: fatigue life estimation of a joint hotspot............................................................................. 65 Figure 3.23: Significant wave height (Hs) / Spectral peak wave period (Tp) – Scatter diagram (all year) (adapted from [/3/]) .... 66 Figure 3.24: Number of points (hotspots) checked around the joint for fatigue assessment ........................................................ 68 Figure 3.25: Design HSE-P curves (S-N curves) for welded plates in air and seawater with 45 and 150 mm wall thickness...... 70 st

Figure 3.26: Re-built F3-FA SACS model 1 mode shape............................................................................................................ 80 Figure 3.27: Re-built F3-FA SACS model 2

nd

mode shape ........................................................................................................... 80

rd

Figure 3.28: Re-built F3-FA SACS model 3 mode shape ........................................................................................................... 80 Figure 3.29: Orthogonal coordinates of the F3-FA SACS models – in plan view.......................................................................... 82 Figure 3.30: Leg joints in SACS model ......................................................................................................................................... 88 Figure 3.31: Bucket and transition frame joints in SACS model.................................................................................................... 88 Figure 4.1: F3-FA knuckle joint (Iv-Oil & Gas courtesy) ................................................................................................................ 94 Figure 4.2: Example of a c entric bucket ........................................................................................................................................ 94 Figure 4.3: Solution with centric bucket side by side with the barge [left] and with centric bucket underneath the barge [right] .. 95 Figure 4.4: F3-FA site location (adapted from [/23/])..................................................................................................................... 96 Figure 4.5: Foundation plan view of the model with centric buckets underneath the barge.......................................................... 97 Figure 4.6: Side view during transportation of the model with centric buckets underneath the barge .......................................... 97 Figure 4.7: SACS Model with centric buckets underneath the barge (SACS software) ................................................................ 98 st

Figure 4.8: Model without the knuckle joint and with centric buckets 1 mode shape ................................................................ 100 nd

Figure 4.9: Model without the knuckle joint and with centric buckets 2 mode shape ............................................................... 100 rd

Figure 4.10: Model without the knuckle joint and with centric buckets 3 mode shape .............................................................. 100 Figure 4.11: Structural incompatibility on the leg-topside connection with a leg diameter increase (SACS software)................ 103 Figure 4.12: Top deck and cellar deck– Leg connections (Iv-Oil & Gas courtesy)...................................................................... 103 Figure 4.13: Free space on the ................................................................................................................................................... 104 Figure 4.14: Re-built F3-FA SACS model with former and new leg-topside connection [sleeves highlighted in green] ............. 104 Figure 4.15: General plan view of the topside design modifications for the models with centric buckets underneath the barge 105 Figure 4.16: Plot of the new leg topside connection (e.g. leg E1) ............................................................................................... 106 Figure 4.17: A view of part of the F3-FA gril lage during load-out operations .............................................................................. 108 Figure 4.18: Platform on barge during sea transportation........................................................................................................... 108 Figure 4.19: Comparison between the previous SACS model and the one adapted with extra columns for grillage support .... 109 Figure 4.20: Possible leg/sleeve – extra column connection ...................................................................................................... 109 Figure 4.21: General plan view of the topside design modifications with the extra columns ...................................................... 109 Figure 4.22: Leg/sleeve shifted in N-S direction for the different leg diameter models (e.g. 3250 and 5000 mm) ..................... 111 Figure 4.23: SACS model topside members maximum UC’s colour represented....................................................................... 114 Figure 4.24:SACS model leg members maximum UC’s colour represented .............................................................................. 116

viii

Figure 4.25: SACS model leg A3 principal loads for the load condition C045 ............................................................................ 117

Figure A.1: Wind area l oads (adapted from [/2/]) ........................................................................................................................ 131

Figure B.1: Linear wave theory wave shape ............................................................................................................................... 134 Figure B.2: API & ISO procedure for calculation of deterministic static wave & c urrent forces (taken from [/4/]) ....................... 137 Figure B.3: Diagram for the selection of an appropriate periodic wave theory (adapter from [/11/])........................................... 139 Figure B.4: Current diagram force stretch to wave surface by two different methods................................................................. 140 Figure B.5: Flow in a bluff body (a) and in a streamlined body (b) (adapted from [/14/]) ............................................................ 143 Figure B.6: Wake amplification factor for drag coefficient as a function of the K, for K<12 (adapted from [/6/])......................... 146 Figure B.7: Wake amplification factor for drag coefficient as a function of the K/Cds, for K>12 (adapted from [/6/]) ................. 146 Figure B.8: Drag and inertia coefficients for a smooth cylinder, with Re= =11525 (adapted from [/15/]) .................................. 148 Figure B.9: Drag and inertia coefficients for a rough cylinder, with Re=

=6833 and 14200 (adapted from [/15/]) .................... 148

Figure C.1: Wave horizontal velocity distribution methods above wave c rest for the linear wave theory ................................... 151 Figure C.2: Wave horizontal velocity distribution, per elevation, for a phase angle of 0 deg...................................................... 152 Figure C.3: Rectangle method applied to calculate the maximum drag and inertia forces ......................................................... 152 Figure C.4: Base shear and overturning moment ....................................................................................................................... 153 Figure C.5: Base shear dis tribution per phase angle .................................................................................................................. 156 Figure C.6: Wave surface comparison for different theories ....................................................................................................... 157 Figure C.7: Base shear distribution, per phase angle, comparison between linear wave calculations and SACS results ......... 160 Figure C.8: SACS Models studied for the calculation of the hydrodynamic base shear force and overturning moment ............ 162

ix

Table of tables Table 3.1: Substructure leg members in the splash zone ............................................................................................................. 33 Table 3.2: Topside section properties of the re-built F3-FA SACS model..................................................................................... 36 Table 3.3: Substructure section properties of the re-built F3-FA SACS model ............................................................................. 37 Table 3.4: Re-built F3-FA model topside permanent loads distributed per deck (nett weights, i.e. without contingencies) ......... 40 Table 3.5: Re-built F3-FA model topside permanent loads checked against the srcinal F3-FA model ....................................... 40 Table 3.6: Re-built F3-FA model substructure permanent loads checked against the srcinal F3-FA model ............................... 42 Table 3.7: Re-built F3-FA model total permanent loads checked against the srcinal F3-FA model ............................................ 43 Table 3.8: Live loads (adapted from [1])........................................................................................................................................ 44 Table 3.9: Summary of Live loads................................................................................................................................................. 44 Table 3.10: Crane operational loads (adapted from [2])................................................................................................................ 45 Table 3.11: Wind velocity (m/s) (adapted from [/2/]) ..................................................................................................................... 46 Table 3.12: Wave data (adapted from [/2/])................................................................................................................................... 50 Table 3.13: Design still water depth (adapted from [/2/])............................................................................................................... 50 Table 3.14: Current velocities (m/s) (adapted from [/2/])............................................................................................................... 51 Table 3.15: Drag and inertia coefficients used in the F3-FA design.............................................................................................. 53 Table 3.16: General primary steel classification for the F3-FA structure ...................................................................................... 57 Table 3.17: Steel yield stress based on wall thickness (WT) ........................................................................................................ 58 Table 3.18: Marine growth layer thickness for the F3-FA design .................................................................................................. 60 Table 3.19: Hydrostatic head calc ulation for the F3-FA s tructure ................................................................................................. 61 Table 3.20: Parameters of the S-N curves .................................................................................................................................... 69 Table 3.21: Permanent loads without SACS model self-weight (load case DDDD)...................................................................... 73 Table 3.22: Load combinations for the modeshape, in-place and fatigue analysis (topside loadings) ......................................... 74 Table 3.23: In-place analysis load combinations for operational conditions with maximum vertical load ..................................... 75 Table 3.24: In-place analysis load combinations for operational conditions with minimum vertic al load ...................................... 76 Table 3.25: In-place analysis load combinations for survival conditions with maximum vertical load........................................... 77 Table 3.26: In-place analysis load combinations for survival conditions with minimum vertical load............................................ 78 Table 3.27: First mode shapes natural periods / frequencies comparison between the re-built and the srcinal F3-FA model.... 81 Table 3.28: Higher natural periods up to a contribution of 75% of the live loadfor the re-built and the srcinal F3-FA model ... 81 Table 3.29: Higher natural periods up to a contribution of 50% of the live loadfor the re-built and the srcinal F3-FA model ... 81 Table 3.30: SACS models load case results ................................................................................................................................. 83 Table 3.31: SACS models load case summary report for vertical forces (in Z direction) .............................................................. 84 Table 3.32: SACS models load case summary report for forces X direction ................................................................................ 85 Table 3.33: SACS models load case summary report for forces Y direction ................................................................................ 86 Table 3.34: Maximum fatigue lives [yrs.] comparison between the Original and the Re-built F3-FA SACS model – Substructure leg butt weld joints ......................................................................................................................................................................... 89 Table 3.35: Maximum fatigue lives [yrs.] comparison between the Original and the Re-built F3-FA SACS model – Substructure transition frame and bucket butt weld joints .................................................................................................................................. 90 Table 4.1: Differences in the structural steel nett weight between the re-built F3-FA model and the model without knuckl e joint 98 Table 4.2: Natural period’s: Model without the knuckle joint and with centric buckets underneath the barge [50% live load]...... 99 Table 4.3: Natural frequencies: Model without the knuckle joint and with centric buckets underneath the barge and with the legs diameter increased to 4000, 4500 and 5000 mm vs. Re-built F3-FA Model [50% live load]....................................................... 102 Table 4.4: Structural main steel weight savings: Models without the knuckle joint and with centric buckets underneath the barge and with legs diameter increased to 4000, 4500 and 5000 mm vs. re-built F3-FA model .......................................................... 102 Table 4.5: Natural frequencies: Model without the knuckle joint and with centric buckets underneath the barge and with legtopside connection adapted vs. Re-built F3-FA Model [50% live load] ....................................................................................... 107 Table 4.6: Natural frequencies: Models without the knuckle joint and with centric buckets underneath the barge and with topside modifications vs. Re-built F3-FA Model....................................................................................................................................... 110 Table 4.7: Model legs shifted values to be compatible with barge BOA 35 ................................................................................ 111

x

Table 4.8: Natural frequencies: Model without the knuckle joint and with centric buckets underneath the barge, former and new leg-topside and with legs diameter up to 4000, 4500 and 5000 mm vs. Re-built F3-FA Model.................................................. 112 Table 4.9: Structural main steel weight savings: Model without the knuckle joint and with centric buckets underneath the barge, adapted leg-topside connection and with legs diameter up to 4000, 4500 and 5000 mm vs. Re-built F3-FA Model ................. 113 Table 4.10: Maximum topside members UC’s – model with the srcinal size of leg diameter (3250 mm).................................. 115 Table 4.11: Maximum topside members UC’s – comparison between models with different leg diameter ................................ 115 Table 4.12: Maximum leg A3 members maximum UC’s – model with the srcinal size of leg diameter (3250 mm)................... 118 Table 4.13: Maximum leg A3 members maximum UC’s – comparison between models with different leg diameter ................. 119 Table 4.14: Maximum fatigue lives [yrs.] – Substructure butt weld joints for 50% Live Load (LL) [in 8 wave directions]............ 121 Table 4.15: Variant solutions overall results................................................................................................................................ 122

Table A.1: Wind parameters........................................................................................................................................................ 130 Table A.2: Wind forces ................................................................................................................................................................ 130

Table B.1: Wave parameters without the Doppler effect ........................................................................................................... 138 Table B.2: Wave parameters with the Doppler effect ................................................................................................................. 138 Table B.3: Wave steepness and relative water depth calculation ............................................................................................... 139 Table B.4: Reynolds number calculation (adapted from [/2/]) ..................................................................................................... 143 Table B.5: Criteria for the drag & inertia c oefficients – D = 3450 mm ......................................................................................... 144 Table B.6: Criteria for the drag & inertia coefficients – D = 4200 mm, 4700 mm & 5200 mm..................................................... 144 Table B.7: Keulegan-Carpenter number and factor K/Cds ........................................................................................................... 145 Table B.8: Drag coefficient based on the wake amplification factor according to API [/4/] ......................................................... 147 Table B.9: Drag coefficients us ed in the F3-FA design ............................................................................................................... 147

Table C.1: In-place survival condition, with maximum water depth, wave data for the F3-FA site ............................................. 150 Table C.2: In-place survival condition current velocities for the F3-FA site................................................................................. 150 Table C.3: In-place survival condition drag and inertia coefficients for the F3-FA design........................................................... 150 Table C.4: Marine growth considered in substructure legs for the F3-FA design ....................................................................... 150 Table C.5: Intrinsic wave period calculation for the in-place survival condition, with maximum water depth, for the F3-FA site 151 Table C.6: Base shear and overturning moment calculation....................................................................................................... 154 Table C.7: Base shear and overturning moment comparison with SACS [bigger values marked in green] ............................... 158 Table C.8: Base shear (F) and overturning moment (M) comparison with different SACS models ............................................ 161

Last page italic numbers

xi

1 Introduction 1.1

General considerations 1

The F3-FA platform is a pioneer structure, one of a kind in the North Sea, built to be a

re-usable gas

production platform, using an innovative structural concept for offshore structures called the selfinstalling platform (SIP) (Figure 1.1).

Offshore platforms are supporting structures placed in marine environment. The most common type of offshore platform is a fixed platform that extends from the seabed to a certain height above the sea surface, often consisting of a steel tubular structure called the

jacket. This solution is frequently the

most economical for the shallow waters of the North Sea (i.e. up to 100 meters).

After being transported and positioned on its final destination, the jacket faces a critical stage when it is free-standing in the seabed while piles are driven into the soil to support the structure. After this risky operation the topside module can be installed on top of the jacket, as shown in Figure 1.1.

Figure 1.1: Jacket platform example [LEFT] & SIP platform example [RIGHT] (Chevron B13 platform installation on left and F3-FA platform on right, Iv-Oil & Gas courtesy)

Usually, these kinds of platforms in the North Sea are installed by means of heavy-lift vessels. However for big offshore structures the size of the lifts can be reduced by making the construction modular, but with an increase of the man-hours spent offshore and the need for a crane vessel for lifting and installing the components, this is also not the most economical solution. 1

The offshore platform it’s called after is initial design location site.

-1-

Chapter1:Introduction

Pres nt companies co nc rn

ith e on o ic an

of sh re platf rms ha e rai ed t e ch llenge

environm nt l issues and the nee for f

tu yi g alternati e soluti ns t

ig er

t e continue

establis m nt of install tion p roced res for offs or stru tures.

S lf-installi g latform SI )

In this

ontext, the self-i st lling

latform (SI ), as the

3- FA pl tform,, c n be an ideal s lution

(Fiigure 1.2 . It consists of a latfor hi h is t an ported on sucti n uckets j ck d-up an a tache .

T e compan

I - il & Gas [/20/] ha

installing plat or 4,00

fl t t p cargo ba ge with the legs and

for the F -FA

2

T , in a tot l

d evelo ed a s lf-

rojec with th

t pside of

ei ht of 9,000 m , including legs and

b ckets (Figure 1.2). This

tructure was d esiign d for a 40

meters wa er depth, f r the se er environ ment l con itions of the

orth

ea, n ar th Dog er Bank,, a d bei g prepared

for r -u age.

On the desti ation of th

platfor , the su cti n bu ke s

fillled with wa er, w ile the leg

re

are lo er d. Af er ard, he

b ckets are suc ed in to he seab d. Ther after, th

t psiide

wiill li b lo g the le gs to the requi red h ig t si g trand jack

con ected to t e to

dec

nd th e ba ge c n be

i m di tely r m ved a fter the t pside lift off .

On t e op of he legs, a b olt d onne tion 6 m te -lo g bolts

it

as pr vided using

a di m te of 2 0 milllim te s, called

s perb lts . C la ps w re al o i st lled on t he bo to

preven la er l

deck to

ov m ents o the l gs.

Figure 1.2: F3-FA plat or (Iv-Oil & Gas co rte y)

Figure 1.3 pr sents a simple ill strati n f t he F3 FA pl tform i st lla ion sequen e.

2

T e I m etric sy te

uniity o f m ss equal to 10 0 k ilograms is the onne r epr se ted by he ymbol “ t”. IIn th e o fsh re i ndustry

to void c nfusion wit the s ort ton an th lo g t n i t is also kn wn as

m tri to wi h th e a br viat ion “mT”.

3

Superbol s or m lti-jack olt tensioners are an alte rna ive saf s luti n to a sin le no mo s size bolt ed joint by the us

se eral s aller jackb lts red cing the torqu force

nd siz . F r the F3 -F

the topsid .

Ch pt r 1 : Intro uction

of

platform 16 superbol s M240 w re sed to sup ort

-2-

Figure 1.3: F3-FA platform installation sequence: Transportation on barge [LEFT]; Legs lowering into the seabed [CENTER]; Installed [RIGHT] (Iv-Oil & Gas courtesy)

Currently the F3-FA platform is already in service after a successful installation. The big benefits of this offshore platform design solution are: i.

the unnecessity of expensive heavy lift vessels during the transportation and installation of the platform on the offshore site;

ii.

the platform (substructure and topside) is installed in only one piece;

iii.

the reduction of the offshore installation period of the platform;

iv.

the use of suction technology instead of piles for the foundation design, reducing environmental issues concerning the energy spent driving the piles to the target penetration depth and the disturb once caused to marine life due to the noise associated with pile driving;

v.

the potential of re-use of the self-installed platform, revolutionizing the economic panorama of the offshore oil and gas fields, reclassifying the so called “marginal fields”.

4

Engineering issues

One of the main differences of the SIP platforms with respect to the traditional installed platforms is the fatigue loads going through the topside of the structure.

In a traditional jacket platform the vertical loads due to weights and live loads are transmitted through the legs to the sea floor and the later al horizontal loads , mainly due to waves, are supported by braces, connected to the legs, which in turn will transmit the forces to the sea floor. 4

Marginal fields refers to an oil or gas field that may not produce enough net incom e to make it worth developin g at a given time,

special due to technical and economic conditions.

-3-


Admitting that the topside is stress free for lateral loading, in general only the design of the legs and braces will be conditioned by the dynamic horizontal loads affecting the fatigue life of the jacket.

In a SIP platform, however, the entire structure (topside and legs) will work together in a behavior similar to a portal frame with no stabilizing braces. This means that also the topside is loaded with cyclic wave loads which result in a fatigue design required for the topsides.

Figure 1.4 and Figure 1.5 present a very simple static example and the resulting bending moment diagram (BMD), comparing the behavior of a traditional offshore structure with a SIP platform, when submitted to lateral loads. The topside of the porta l frame will be subjected to stresses, due to a horizontal force applied, unlike the cantilever structure (simulating the behavior of the jacket platform).

Figure 1.4: Jacket platform

Figure 1.5: Self-installing platform (SIP)

Also, the lack of braces restraining the legs (substructure stiffness is lower) will decrease the first frequency modes of the structure resulting in a lower fatigue life when compared to a jacket platform. As a result SIP platforms are more likely to be sensitive to fatigue issues and therefore need a careful design with respect to this important structural design requirement.

1.2

Main objectives

Due to the innovation of the solution combined with the tight planning of the srcinal project, the F3-FA design of the SIP has not been developed fully efficiently. The structural layout and the dynamic behavior are the main aspects under investigation in the master thesis; the two combined may significantly improve the fatigue performance of the structure. The impact of environmental loads in an offshore structure will generate considerable time-variable stresses, which may lead ultimately to fatigue failure. Indeed, the dynamic behavior is decisive in the fatigue performance and design of the structure.


-4-

Therefore, the main objective of the master thesis is to improve the dynamic behavior of the F3-FA platform, by proposing solutions to its design layout that can minimize the dynamic effects due to wave loading and therefore, at the end, reduce fatigue issues of the platform. The characterization of the wave loading and the study of the fatigue assessments of an offshore structure are also goals of this work.

To achieve the main objective: 

The F3-FA platform layout is used: basis of design, structural model build with SACS

5

software [/10/] and as-built structural drawings; 

A simplified structural model of the F3-FA is re-built based on the srcinal F3-FA SACS model;



The re-bu ilt F3-FA SACS model is compared with the srcinal mode l, to guarantee the feasibility of the study;



Different key structural alternative solutions are incorporated, in the geometric structure configuration of the re-built model, with the aim of improving the dynamic sensitivity of the platform to wave loading;



The re-built F3-FA model is used as a model of reference to compare the different structural solutions, and two main results are compared: frequency modes and structural weight.

The main alternative design solutions studied are (Figure 1.6): (a) the exclusion of the transition frame (also known as knuckle joint) existing between the legs and the suction buckets. Will allow a better geometric configuration in the leg-topside connection, due to the restraints necessary during transp ort for the knuckle joint, over the smaller leg buckling length existing; (b) the increase of legs diameter , to increase the frequency of the modes of the structure that will affect the fatigue behavior of the platform; (c) a new leg-topside connection , to provide a more stiffer and robust link between the substructure and the topside, avoiding many expensive hours of offshore work to accomplish the present connection in the F3-FA platform.

A quick overview of the F3-FA platform geometry is shown in Figure 1.6.

5

SACS Offshore Platform Structural Design & Analysis Software and abbreviated to SACS [/10/] is a structural engineering

software that supports the design and analysis of offshore structures.

-5-


e t s ta k

o si e

e -t psid egs

o n ction

ransition fra e u tion u ke Figure 1.6: Ge metry ver vie

of the F3-FA latform (Iv-Oil

Gas cour esy

Af er re ching the final model soluti n , design stu ie will ls take int a count th in-pl ce analysis as w ll s ini er ice fatig e n lysis; co si eriing dead, live a nd envir n ental loa ds quantified and combined, ccording to th F -FA tructu ral Basis of

e ig [/1/].

The in-place nalysis is perf r ed by a static analysis, the M rison equation and t e ef ec s

ave forces are c lc lated throu h the

a e in m tic s are ba ed on non-li ne r

re easily incorporated in t e sta tic analy is by in lu in

ave t eories.

he w ve dyna ic

the tructure res po se in th fluid

fo ce equatiion.

The fati ue asse s ent is done by m ans of a spectral fatigu anal si , f cusing on the fatigu li es of th b tt

eld j ints i the platfor

to the struc ur l

1.

substructure. In ac , t ese stru tu al le e ts ar v ry se sitive

odific ati ns pr pose a d an gi e n overvi w of he platfo m fati gu b havior.

T esis or aniza tion

The master t esis is i organize

in fi e m in chapters. The in ro uctio

overvie

and a compari on with the traditio al ja ke pllatfor .

of t e F3 F

platfor

Ch pt r 1 : Intro uction

cha ter p es nts

brief

ls , he

-6-

engineering issues involved in this pioneer platform are referred, and finally the main objectives and thesis organization are presented.

The second chapter presents the F3-FA platform, namely its structure geometric characteristics, structure foundations, leg-topside connection, and platform construction, transport and installation procedures.

The third chapter describes the structural model based on the F3-FA platform re-built, the loads and analysis used in the study. The geometry is described with plots of the respective model. The loads are summarized and detailed explanation is given about the environmental loads calculation. The inplace analysis and fatigue analysis are covered with a more comprehensive description of the spectral fatigue analysis. The combined loads with the respective load factors are listed in tables. Finally, the model re-built is tested through a comparison study with the srcinal structural model, approved and used in the F3-FA platform design and analysis.

Chapter four includes the analysis of the different structural alternative design solutions, and presents the final layout solution, its dynamic and fatigue analysis.

Finally, the fifth chapter presents the conclusions of this study and indicates some possible future developments.

Appendix A presents a detailed explanation of the calculation of the wind loads.

Appendix B describes the wave loads, with a small introduction to the wave theories used and a description of the static wave analysis method.

Appendix C includes an example of the calculation of the hydrodynamic loads (wave and current).

Appendix D presents a spreadsheet example of the calculation of buckling and stress check for cylindrical members according to the API standards [/4/].

-7-



-8 -

2 2.1

F3-FA platform general description Introduction

The F3-FA platform was built to operate in the F3 block of the Dutch sector in the North Sea.

Early in the 70’s, the F3-FA gas field was discovered and found to be economically not viable to build an entire structure to process, compress and transport to shore the raw material by the oil and gas companies, due to the field small size.

The field is situated in a remote location on the Dutch sector in the southern part of the North Sea, as shown in Figure 2.1, well known by its severe environmental conditions.

Figure 2.1: Location of the F3-FA platform (adapted from [/28/])

In March 2008, Venture Production (later acquired by Centrica Energy [/23/]) purchased the field rights with the intention of installing a production platform.

-9-

Chapter 2 : F3-FA platform general description

Usually the installation of an offshore platform in the North Sea is a highly expensive activity. When the well is dry, the production company is also responsible for the process of “decommissioning” the well, the platform and equipment used to support the production, a process equal or more expensive than the initial installation of the platform.

For the project to be feasible, the F3-FA platform engaged a new offshore structure concept called the Self-Installing Platform (SIP).

The F3-FA platform is revolutionary in the way that the costs for installation are rather small compared to the traditional North Sea offshore platform and, as soon as the field runs dry, the structure can be re-located rather quickly to other fields. These features are related to its transportation and installation procedures, as well as to its SIP support structure.

As explained in chapter 1, the SIP is a four leg portal frame using a suction bucket foundation, transported on a flat top cargo barge. Strand jacks are used to lower the legs into the seabed and jacked up the topside to the design elevation from his rest position on the barge.

This pioneer design engineering project was developed by SPT Offshore (SPT) and Iv-Oil & Gas (IvOG) [/8/] in a combined effort. IvOG was responsible for the entire platform project starting at the bucket’s cover plate upwards, while SPT task was to develop the bucket-platform interface.

Besides the buckets project, SPT responsibility covered the transport and installation analysis of the complete platform, including the design of the grillage and sea fastening.

Since F3-FA is a rather unique offshore platform, this chapter intends to present its main aspects. The F3-FA main structure geometry is described in section bucket foundation is presented in section

2.2 and the methodology behind the suction

2.3. The innovative SIP platform and the F3-FA structure

exceptional size required a special connection between the substructure and the topside, which is illustrated in section 2.4. Finally, in section 2.5 all the steps in the installation sequence are presented in more detail.

2.2

Geometry

The F3-FA platform can be divided in three main parts, as shown in

Figure 2.2: topside, substructure

and foundation.


- 10 -

Topside

Subs ru ct re

Foun a io ig re .2: 3-FA p atform D vervie

2. .1

The

as ou tes )

To si e

3-FA to side is a three dimensi na l port l f a e tructu re si ilar to a re tan gular bo

2. ). Th

ai steel st uctur c m rises tw longi udinal fr m e t us es, in ro

trans ersal ones i n t e ro s 1, A,

11 -

(Iv-Oil &

,

a nd F ( Figure 2.3 a nd Figure 2.5).

Ch pt r 2 : F -F platf rm ge er l d escription

( ig re

1 a d ro 3, and five

The t ree prin ip l dec s, celllar de k, main eck and to

dec ,

o pl te the

orizo tal elements

f he

primary ra in g.

The decks onsist o a wellded bea

rid wi h n

E

pr fil a d tiffen d pla ing o the u pper flange of he beams.

et e en he m in gri s, striingers with an I E

pr fil , an el e to he prin ip l pr vi ed to su port the pla ed deck.

eams,

re

The i te ra ion of th legs wit t e t op id is m de by fo r t b lar sl eve sectiion inco po rat d et e n he to

dec

and th

ai

d ck, at t e co ners of he

to si e ( Figur 2.3 and Figure 2.5).

An o en spac e i left et een deck due to th Fi ur 2.3: F -FA topsid S CS model D vervie

T o se onda y

ecks co exist be ween the

i st lla io

ai n ec and cellar of t he pl tform ( ee

section .4).

ell r

nd main d cks

re pecti el . The e e ks ar c ar ct riz ed by a grated flo r o n top of

nd the

ai n

nd top

ecks,

welded be am gri sup orted

on the trus colu n of th prim ry fra in g.

The pl tform also c nt in

an elevate

helic pt r l nding d ck , located on the sout si e, wi h a

teel plated deck on top o f

bea

g id uppo te b c ntilevered trusse

in ro

1 an r w

( ig re 2. ).

The top id also eatur s n he north fac e w at er d ck (s o n in Fi ur over th

t p deck, n ar le

pr p rly s ace for th drilling e quipment.

or

2.4) plac 3, to al lo

over a d th

Fi ur 2.4: F -FA topsid si e v iew Iv- il

Further or , the to d ck supp rts, t e edes al crane on collu n

Gas c urt sy)

3 an t e en t s ac o colu n

C1 ( ig re 2.3 and F igure 2.5).

Cha ter 2 : F3-FA platfor

gen ral de cri tion

- 2-

Topside key dimensions: 

Length = 40.630 m (distance between rows A1 and F);



Width = 25.250 m (distance between columns farther apart);



Height = 20.500 m (distance between cellar deck and helideck);



Longitudinal spacing between legs circumference axis = 36.430 m;



Lateral spacing between legs circumference axis

= 20.750 m.

Figure 2.5: F3-FA topside princip al decks plan view & longitudinal side view

- 13 -


2.2.2

Substructure

Large tubular sections, diameter 3250 mm, formed the four legs on the platform corners. The legs are named after their close position to the intersection of rows A1, A3, E1 and E3, as shown in Figure 2.5.

At the sea bottom, each leg is connected to a suction bucket. A transition frame (also known as knuckle joint) allows the bucket circumference to be eccentric from the leg axis, by distributing the platform loads to the buckets.

The buckets were required to be eccentric, to stand side by side with the barge during transportation. The transition frame is needed to reduce the leg unbraced length and support it from the perimeter of the bucket, distributing the platform Figure 2.6: F3-FA suction bucket and knuckle joint (Iv-Oil & Gas courtesy)

loads to the bucket walls and, consequently to the sea floor (Figure 2.6).

In addition, a horizontal truss frame was provided connecting leg A1 with A2 and leg E1 with E3, to fix the buckets orientation during installation (see Figure 2.7).

Figure 2.7: F3-FA suction buckets, transition frame and truss frame (SACS model)


- 14 -

The legs head size, illustrated in Figure 2.8 , supports the strand jacks during the installation operations (i.e. legs lowered into the seabed and topside jacked up to the design elevation) and fix the legs to the topside through the superbolts, as mentioned in chapter 1.

Figure 2.8: F3-FA substru cture leg in yard (Iv-Oil & Gas courtesy)

Substructure key dimensions: 

Leg diameter

= 3.250 m;



Leg height

= 78.020 m;



2.2.3

Knuckle joint height

= 18.370 m.

Foundation

The F3-FA platform foundation is composed by four suction buckets with 15 meters diameter and 13 meters elevation height, penetrating into the sea bottom. On the top of the bucket, a plate closes the caisson and a frame distributes the leg loads through the bucket’s walls.

Figure 2.9:F3-FA suction bucket (Iv-Oil & Gas courtesy)

These gian t steel structures represent a big inves tment in material but avoid the use of a pile foundation that requires expensive man-work-hours offshore, heavy-pile drive equipment and a large crane vessel.

- 15 -


2.

Suctio b cket

The uc ion b ckets a e n ffsho e fou datio that esemble s an i verte b cket, usual y of greater di e si ns due t t eir aspe t ratio b ing less than a unit y (L/D and

1, w er L is he le gt of the s irt

is th diameter o f the buc et).

C m ared to the tr dit ional ile foun ati on th re over during “platf r

s ction buck ts ar quic er to in tall a d ea ier to

d com is ioning ” and mi ht preve t t he exten iv u e r e ve a y se of an

expe si e eavy-lift ve sell for b th op ra io s.

Hi s t ri c al ackg o n In th 8 ’s, researc o t e uctio

b cket echn lo y ul in ated, in 1989, n he in tallati n of the

G llf ks C latform, in he Norw gi n ector of the N rth Sea, using 16 co cr te bu kets f 28 di m ter p n tra tin

22

eters in o the seab d [/2 /]. The

ullfa s

lat or

st uc ur ( B ) con tr ct d f steel r inf orced co cr te helld i n place by gravity wit h 380

eters, w hic 217 meter a e b elow

Fig re .1 : G llfaks

et rs

i s a g avity bas ed to al height of

ater /25/].

G S latf rm [LEFT] and Dr up er

ja ket plat for

[RIGHT] ( da te fro m [/ 26/] & [ 24/]

In 1994, the first ja ke platf rm u in a sucti n ucket foun ation, the

ra pner E jac et platfo m,

w s installed nd th a pplica ions of this ol tion hav s re d by the entire gl b .

ore r centl , with

6

Allso no n a s s ction pi les, suction caisso s or suction a chors.

Cha ter 2 : F3-FA platfor

gen ral de cri tion

- 6-

the development of the offshore wind industry, the bucket foundation is been considered in the installation of wind turbine towers.

Principle The suction bucket principle and installation sequence, illustrated in Figure 2.11, is the following [/19/]: i.

Bucket is lowered in the water and gravity will help the bucket to sink into the seabed;

ii.

After the bucket walls reach the seabed, a valve on top of the bucket enables air and water to be pumped out;

iii.

The pressure inside the caisson will drop causing a massive suction force downwards;

iv.

The suction force, caused by the di fference between the buck et inside pressure and the hydrostatic pressure, will induce the bucket to penetrate into the seabed.

Figure 2.11: Suction bucket installation

When the bucket penetrates the soil, the skin friction is mobilized along the bucket wall surface and the soil tip resistance in the wall sleeves toe. Under tension, the suction bucket resistance is given by the skin friction resistance mobilized along the bucket wall surface and the bucket self-weight [/19/].

- 17 -


B simply ev rsing th

proces , t e bu ket can be ra idl r emov d y puttin

caisson, th o erpress re cr at d intact to b

p ressure i si e he

ill ca se a vertic l li tin f rc a d he sucti n

r used. T hu , it is a more pr ctical and

u ket is re o ed

cono mi al solution or th

F3- A platfo m,

designe to b remova le an reinstalled in iff re t fields duri ng her servi e liife.

The ucket penetration is limi ed by th re latiion be ween the di ameter of th b cket (D) and t e len th of th s irt (L);; a d suall the pen tr tin g r tio L/D i s aller th n .

s, a con eque ce, the ucket

di m ter n eds t b o conside able s ize.

2.

F -F

leg-to si e o necti n

The 3- A platfo m is i onne te in tw diifferent elev tions bet ween the to si e nd the sub tructure, as shown i Figure 2.1 .

On t e top, l gs a e rigid

onne te

to t he to si e using

pr te si n olts – known as uperbolt . The e par icular bolts, due to thei c n idera le size small r

olts, nown a

force an e sure

Each le

re fixed by a number of 36

jack olts,

ase the nece sa y torque

stiffe r c n ec ion (Figure .13).

has a set of 4 s perbolts M 40 to a oi

co n cti n th t inevit bl

a weld

o ld require more exp nsive

worki g off hore hours, l ss a curacy

nd no

bility to be

dismantl d an r -installled in a later pha e (Fi ur 2.14). su erbolts re 6

eters lo g

a ension nut with

eighing appro x.

50 k . In he e d he e tir

he

T nd carry t pside is

hanging in 16 uperbolt p eload d o 1 2000 kN each [/2 /].

Fi ure 2.12: L g-tops ide oint (Iv-Oil

G s courtesy

To g id

t e legs, from the top o the m ain d ck, tubular

sleev s are bond d o t he to side by four ve rti al hear pla es transferriing the vertical l oa s o t e e k f a in .

O t e cellar an l we st de k, du t r st iction o t e sp ce a ail bl , a c nn ec ion of shore is acco plished wit a si ple clla p, o voiid lateral move e ts of he le s due to the wave impa t and trans erring only orizo tal lo ds (Figu e .1 ).

O ly on the offshore si e, aft r t e leg l s b eing l w re into he seab d nd the t ps ide climbed along th legs,, the clla ps could be in tallled. T is sy te

r quire c nsiderable ff hore

o king hou s t at

should b e a voide in a ut re de ig ( igure 2.16). Cha ter 2 : F3-FA platfor

gen ral de cri tion

- 8-

Finally, on both elevations, cellar deck and top deck, rubber blocks filled the gap between the leg and the clamp system or the sleeve to smoothly distribute the horizontal loads.

2.5

Figure 2.13: F3-FA top deck – leg connection

Figure 2.14: F3-FA superbolts

(Iv-Oil & Gas courtesy)


Figure 2.15: F3-FA cellar deck – leg connection

Figure 2.16: F3-FA clamp system being installed



Construction, mating, transportation and installation

The SIP concept was applied to the F3-FA platform and, basically, consists of an offshore structure which is transported on a flat top cargo barg e with the legs and sucti on buckets jacked-up and attached. The topside, the substructure legs and the foundation by suction buckets are built separately due to their size dimensions.

The topside construction followed a building technique called the “pancake method” [/29/], where all decks are built separately and assembled together through vertical stacking. (Figure 2.17). The major advantage, compared to the traditional construction method of building the entire topside together, is the possibility to weld the structural steel beam grids while still in the ground level improving

- 19 -


productivity and safety among the working force, with an obvious indirect impact on the construction 7

costs [/30/].

Figure 2.17: F3-FA topside construction (Iv-Oil & Gas courtesy)

The F3-FA project and construction followed close schedules, for that reason the corner nodes that would connect the four legs with the topside were fabricated separately and parallel to the topside and integrated in a later stage to be more time effective in the construction ( Figure 2.18 and Figure 2.19).

Figure 2.18: F3-FA topside without corner nodes

Figure 2.19: F3-FA corner nodes built separately



While the topside was being built by Heerema Fabrication Group (HFG) [/22/] in Vlissingen, SPT Offshore [/21/] designed and fabricated the suction buckets ( Figure 2.20). After completed, the caissons were transported to HFG yard and attached to the stiffening transition frame, i.e. the knuckle joint (Figure 2.21, Figure 2.22 and Figure 2.23).

7

In a typical offshore production platform, as the F3-FA, non-structural items, such as mechanical equipment, vessels and

piping equipment are also more easily and safely installed using the construction “pancake method”.


- 20 -

Meanwhile the legs were built horizontally, upended and attached to the topside after being complete and transported outside the building hall. With the legs integration on the topside, the strand jacks are installed, that will enable the legs to be lowered and the topside to climb to the top of the legs (

Figure

2.24).

- 21 -

Figure 2.20: Suction bucket top plate

Figure 2.21: Suction buckets transport



Figure 2.22: Knuckle joint

Figure 2.23: Suction buckets attached to the knuckle joint




Figure 2.24: F3-FA legs attached to topside (Iv-Oil & Gas courtesy

By now, the semi-finished platform can be transported to the barge, in a process known in the offshore industry as loadout. It is the first step on the final journey that will end with the installation on the site and start production of the platform.

For the F3-FA platform a more complex

loadout was necessary since the BOA 35 transport barge

could not withstand the bending moments produce by an eccentric F3-FA on the boat. Thus, the Heerema H-541 barge was used as an intermediate barge from where the platform could roll over to the final transport barge, as shown in Figure 2.25.

Figure 2.25: F3-FA platform being transported to the Heerema H-541 barge (Iv-Oil & Gas courtesy)


- 22 -

The topside and legs are transported together with the upper grillage by self-propelled modular trailers (SPMT), as indicated in Figure 2.26 . To better distribute the platform weight through the barge stressed skin structure, the upper grillage is fit on top of a lower grillage, previously installed on the barge (Figure 2.27). The upper grillage was previously welded on the topside bottom steel and the lower grillage was also welded to the barge upper steel. In the end, both grillages are welded together in a total height of 6,5 m.

Figure 2.26: Self-propelled modular trailers

Figure 2.27: F3-FA platform rolling over to the BOA 35 barge



The final part in completing the F3-FA platform is to add the suction buckets to the platform. This phase is called mating and required the use of the floating crane Matador 3 ( Figure 2.28). The lifting vessel picks up the suction bucket with the knuckle joint from the quay and transported to the exact position in the F3-FA where they are welded together, as presented in Figure 2.29.

Figure 2.28: Matador 3 with suction bucket and knuckle joint (Iv-Oil &

- 23 -


Gas courtesy)

Figure 2.29: F3-FA platform completed (Iv-Oil & Gas courtesy)

The fully assembled F3-FA platform is now ready to be sea transported to its final destination (

Figure

2.30).

Figure 2.30: F3-FA platform during sea transport (Iv-Oil & Gas courtesy)

During sea transport, the barge had to return to shore due to the bad weather conditions. The closest safe haven with enough draught to receive the platform was the Rotterdam port. However, once arrived, the authorities refused to allow the barge into the port. So, the platform had to withstand for a period of 24 hours waves up to 6 meters height. When the storm ended the platform had suffered no visible damage and a second transport attempt, this time successful, was made [/29/].

On the arrival, the barge was positioned, with the help of the tug boats, the suction pumps were installed on the caissons and the F3-FA platform legs were lowered into the seabed using the strand jacks, after the sea fastenings were released, as represented in Figure 2.31 and Figure 2.32. Once the Chapter 2 : F3-FA platform general description

- 24 -

suction buckets were fixed the topside could be jacked to the final elevation, following the cut out of the welds linking topside and grillage.

Figure 2.31: F3-FA platform tender during installation (Iv-Oil & Gas courtesy)

Figure 2.32: F3-FA platform during installation (Iv-Oil & Gas courtesy)

The final stage was the installation of the superbolts (Figure 2.33), such that the strand jacks could be decommissioned, and the setting up of the clamp system filled in with rubber blocks to absorb wave impact loads (Figure 2.34). Also, a weather deck was installed on top of leg E3 to support the drilling work.

- 25 -


Figure 2.33: F3-FA platform superbolts connection (Iv-Oil & Gas courtesy)

Figure 2.34: F3-FA clamp system installation (Iv-Oil & Gas courtesy)

The F3-FA platform was installed in September 2010 and in the following month the first well was opened with the drilling rig Noble Scott Marks (Figure 2.35) and the first gas came on January 2011.

Figure 2.35: Drilling rig Noble Scott Marks drilling first well in the F3-FA sector [/23/]


- 26 -

3

Re-built F3-FA model

3.1

General considerations

For this proposed study a new model called

re-built F3-FA model has been created in order for a

better understanding of the geometry and structural behaviour of the F3-FA structure.

The re-built F3-FA model is not a full copy of the srcinal F3-FA model, since a simplified geometry and applied loads has been adopted to create a basic reference model for the objectives here purposed. The new model is compared to the old one in order to certificate it has been correctly build.

Using SACS software [/10/], the new model was built based on the as-built drawings, srcinal F3-FA SACS model, Basis of Design (BoD) [/1/], in-place analysis [/2/] and fatigue analysis [/3/] reports.

All actions considered are described and a simplified calculation for the hydrodynamic loads, waves and currents, for a single column, is undertaken and compared to SACS software

[/10/] for a better

understanding of the main actions applied to the structure.

Two structural analyses are studied in the current report, in-place analysis and fatigue analysis. The in-place analysis covers the overall main structure (topside and substructure) for the normal operational and the survival conditions. The fatigue analysis concentrates on the substructure (legs and transition frame) butt welds fatigue lives, since they are critical joints and are very sensitive to structural modifications. From this particular joints, the fatigue analysis results provide a general fatigue evaluation for the overall structure.

Excluded from this study are: 

Tilt effects due to legs deviation from the vertical, during installation, through the gap between the legs and the sleeves;

-27-



Foundation settlements;



Local design analysis (e.g. plating, stringers and other secondary structures);



Joint can sections design;



Deck span deflections and horizontal deflections.

Chapter3:Re-builtF3-FAmodel

3.2

Model geometry

As already mentioned, the re-built F3-FA model geometry is based on the F3-FA as-built drawings and original SACS model. The main difference between both models is the mezzanine deck and intermediate deck, modelled in the srcinal model but not in the re-built model.

Figure 3.1 presents an overall view of the two models.

8

Figure 3.1: Original F3-FA SACS model [left] and Re-built F3-FA SACS model [right]

8

Different colours were used in the re-built F3-FA SACS model for a better understanding of the topside geometry.


-28-

(a)

(b)

(c)

(d)

(e)

(f)

Figure 3.2: (a) Re-built F3-FA SACS model with main parts and critical joints highlighted [from top left to bottom right: (b) topside structure, (c) substructure, (d) leg-topside connection, (e) transition frame and (f) bucket foundation]

-29-


The F3-FA model can be divided in two main parts: the topside structure and the substructure. In addition, three critical and complex joints are referenced: leg-topside connection, transition frame (knuckle joint) and foundation (bucket), as represented in Figure 3.2.

3.2.1

Topside structure

The primary topside framing of the F3-FA structure includes the longitudinal truss row 1 and row 3 and the transverse truss frames in rows A1, A, C, E and F, as shown in Figure 3.3.

Figure 3.3: F3-FA SACS model primary framing structure

The re-built F3-FA structural model includes the horizontal main framing (H-beam members) and deck plating for shear stiffness against horizontal forces, located in the three main decks (cellar, main and top deck) (Figure 3.4).


-30-

Figure 3.4: Re-built F3-FA topside SACS model main decks: cellar deck [right], main deck [center] and top deck [left] highlighted

The primary vertical framing is presented in Figure 3.5, Figure 3.6 and Figure 3.7. Structural truss row 3, row E and row F are identical to truss row 1, row A and row C, respectively (disregarding the helideck support).

Figure 3.5: Re-built F3-FA SACS model longitudinal truss row 1 (3D view and section view)

Figure 3.6: Re-bui lt F3-FA SACS model transvers al truss row A1 (3D view and section view)

Figure 3.7: Re-bu ilt F3-FA SACS model transversal truss row A [left] and row C [right]

-31-


The mezzanine deck and intermediate deck are structural floors supported at the columns midspan (see Figure 3.8). In the srcinal F3-FA SACS model these decks were modelled with the respective loads. However, theirs stiffness does not have a relevant contribution to the global behaviour of the platform since the beam members are considered hinged (moment free). So, in the re-built F3-FA model the secondary decks are only modelled as loads to account for their weight.

Figure 3.8: Original F3-FA topside SACS model with mezzanine deck and intermediate deck highlighted [in red]

Other secondary decks and structures were considered in the re-built F3-FA SACS model the same way as in the srcinal model. The weather deck (superimposed on top deck) and the crane pedestal are modelled as loads to account for their weight; on the other hand the ventstack is modelled with line elements.

3.2.2

Substructure

The four legs and the transition frames on the platform corners are modelled with line members, with tubular section properties (Figure 3.9). The leg line members have an average length of 4 meters, with the node s representing the butt weld joints.

The outer surface of the substructure legs in the splash zone (between El. -4.000 m and El. +5.000 m) is considered corroded with a decrease in wall thickness of 5 mm, however the steel density of these members is conservatively modified to account for their total weight without corrosion, as shown in Table 3.1.

Figure 3.9: Re-built F3-FA substructure SACS model

Appurtenances, such as the conductors, risers, J-tubes, water caissons running internally through the legs are modelled as loads in the leg structure.


-32-

Table 3.1: Su str ctu e l g

Type

em ers in t he pla h z one

Section (m )

A ea (c

2

)

Ste el den ity (m /m )

eight (m /m)

Original ectiion

3 50 x 5

43 3,6

7, 49

34 1,

C rroded section

3 30 x 4

35 4,6

9, 23

34 1,

3. .3

Le -topside co necti n

The onne tion o the t pside to the le is m d at tw levels: t he cellar de k nd th e t p ec ( ig re 3.10). The co necti n arr nge en in S C odel is i a e by du my e b rs (s e ig re 3.11) and as been deter ined du ing th F3- A ro ec with detaile d fiinite ele ent a al si . I the re-built m del no c anges hav b en m de in the le -t psiide conn cti on.

O t e op,, the du m me bers re weig tless, with 00 m m len th and link d he sl eves to he le s, as sh w in Fi ur 3..11. Seconda y c lu ns

Secondar columns

Fig re .10: Re -bu lt F -F

S CS model eg-topside con ect ion [differe t p rspecti e v iew ]

The le ve are onne te t the t pside through ve tic l plat s in f ur differ nt dir ections, s ho n in Figur 3.10; th plates are co nected t “sec ndary” c lu n s that ill trans it the lo ds to the m in truss s, on the topsi e, by a ra in s ste m.

O the ottom, the du m mem be s are weiigh less,

it 2 0

le gth and are di rectly linked to the

to si e, bu the con ectio o ly tra sf rs horiz nt l f rc s nd the

at rial p operti es are adjusted to

th cla p r b er pa h ori on al tiffne s. An illustr tion f t e onne tion is pres nt d i n Figure 3.11.

33 -

hapter 3 : R -b ilt 3- A

o el

Figure 3.11: Plot of the leg topside connection (e.g. leg E1)


-34-

3.2.4

Bucket foundation

The soil stiffness matrix, which specifies the structural behavior of the bucket and the soil/bucket interface, is reduced down to a single boundary joint, per leg, located at center of the bucket area.

Four dummy beam members radiate horizontally from the boundary joint to the interface joints with the leg and transition frame. The dummy beam members are weightless and have a consistent stiffness with the deflections and rotations at the bucket edge and at the knuckle joint, obtained with detail finite element analysis developed during the F3-FA project. In the re-built model no changes have been made in the bucket foundation.

Figure 3.12: Re-built F3-FA SACS bucket foundation

3.2.5

Sections

The section properties used in the re-built F3-FA SACS model are listed in Table 3.2 and Table 3.3.

The SACS software [/10/] does not accept sections with zero weight density, thus a density of 0.0001 3

mT/m was considered for the dummy members.

The section RC0 represents the leg members considered to be corroded in the analysis.

-35-


Table 3.2: Topside section properties of the re-built F3-FA SACS model

Name

Section

Steel density

(mm)

(mT/m )

Description

3

Topside beams H1A

HE1000A

7,85

Main steel topside H-beam

H5A

HE500A

7,85


H6A

HE600A

7,85


PG2

PLG-990*500*30*40

7,85

Main steel topside plate girder beam

PG3 HB1

PLG-700*300*25*25 BOX-990*100*25*25

7,85 7,85

Main steel topside plate girder beam Cellar deck beams located in the leg-topside connection

HB2

BOX-590*100*25*25

7,85

Main deck beams located in the leg-topside connection

HB3

BOX-990*100*25*25

7,85

Top deck beams located in the leg-topside connection

Topside columns and braces 16C

Ø 406,4x12,7

7,85

Columns and bracing members supporting Helideck

20E

Ø 510x15

7,85

Columns and bracing members supporting Helideck Joint can member at truss frame supporting Helideck

20F

Ø 510x20

7,85

20G

Ø 510x25

7,85

Column connecting all three main decks at row D

24F

Ø 610x20

7,85

Most common topside column section

24G

Ø 610x25

7,85

Joint can member at truss frame supporting Helideck

26F

Ø 660x20

7,85

Bracing member at row 1, row 3 and row A and row E

26G

Ø 660x25

7,85

Bracing member at row 1, row 3 and leg-topside connection

26H

Ø 660x30

7,85

Bracing member at row 1 and row 3

28G

Ø 710x25

7,85

30G

Ø 760x25

7,85

Bracing member at rows A1, A, E & F and leg-topside conn. Ventstack column member

30I

Ø 760x35

7,85

Columns at rows A1, A, E & F (leg-topside connection)

42G

Ø 1070x25

7,85

Ventstack column member

42L

Ø 1070x50

7,85

Column member at rows A & E (leg-topside connection)

42Z

Ø 1070x25 / 760x25

7,85

Ventstack cone transition member

54G

Ø 1370x25

7,85


54J

Ø 1370x40

7,85


54K

Ø 1370x45

7,85

Ventstack joint can column member

54L

Ø 1370x50

7,85

Ventstack joint can column member

54Z

Ø 1370x25 / 1070x25

7,85

Ventstack cone transition member

1XG

Ø 2540x25

7,85

Crane pedestal column member

1XM

Ø 2540x55

7,85

Crane pedestal joint can column member

1XN

Ø 2540x60

7,85

Crane pedestal joint can column member

SLV

Ø 3520x60

7,85

Sleeve Sleeve member: 1


st

2.60 m segment length section Ø3490x45

-36-

Table 3.3: Substructure section properties of the re-built F3-FA SACS model

Name

Section

Steel density

(mm)

(mT/m )

Description

3

Leg members R45

Ø 3250x45

7,85

Leg member with 45 mm wall thickness

R55

Ø 3250x55

7,85


R60

Ø 3250x60

7,85


R70

Ø 3250x70

7,85


R75 R80

Ø 3250x75 Ø 3250x80

7,85 7,85

Leg member with 75 mm wall thickness Leg member with 80 mm wall thickness

R85

Ø 3250x85

7,85


R95

Ø 3250x95

7,85


R1X

Ø 3250x135

7,85


RC0

Ø 3230x45

9 ,6 2 3

Leg member with 55 mm wall thickness, but considered corroded (-5 mm wall thickness)

Transition frame RD0

Ø 2600x85

7,85

Upper bracing member

RD1

Ø 2600x50

7,85

Upper bracing member

RD2

Ø 1750x25

7,85

Lower bracing member

RH1

Ø 1200x20

7,85

Horizontal bracing member

RTF

Ø 800x20

7,85

Truss frame beam member

RTG

Ø 650x15

7,85

Truss frame bracing member

Truss frame

Dummy leg member connected to topside DUC

Ø 3250x45

0,0001

Cellar deck leg dummy member

DUT

Ø 3250x45

0,0001

Top deck leg dummy member

RBU

Ø 1500x45

0,0001

Bucket member

RIN

Ø 1500x450

0,0001

Bucket dummy members

Bucket

-37-


3.

Design ac io s

3. .1

Introd ction

In this section,, the design actions i the r e-buil F -F fo the

esign of the

combined i a load

A S

o el ar p es nted. Alll relev nt lo ds

latfor , su h as permanent loads, li e loa l s an

environ m nt l l ads

re

atrix an c n e u m rized as follows:

- er anent l ads (referred t s ction 3.3.2): a.

str ct re self-w eight;

b.

de d

ei ht of no - odelled ite s.

- ive lo ds (r ferred to section .3.3): a.

loads e er te d ri g the servic e li e f t e latfor ;

b.

crane op ra io al oa s.

- nviro m nt l l ads (ref. to se tions 3.3.4): a.

wind l a s;

b.

hy ro ynamic wave an c rr nt) lo ds.

In a practi al an

con er ative approac , the

envir n ental lo d

a re co si er d col lin ar

and o bi ed wi h theiir th

axim m value .

r ason th t the to side

and

g o etry

is

not

or

eight di tri ution y m tri al

cove

di ec ions

ind r se, acc rding to

ro nd the

ig t

the

envir n ental l ads,

dif er nt

Fi ure 3 .13.

The

3-FA pl tform h s

e n design fo r t o

different c nditions: o er tional a d su vival. The peration l con iti n is

eant o on fir

if

th o fs or structur , uring he e tir s rvice lif , is

ithiin he appr priate d si n limi ts for

th pre- st blish d peration ( .g. pr du cti n, drilling, tc); f r t is purpose the live l ad s re combined with

n iro nmental lo ds w ith a

Figure 3.13: Env ironme tal nalysis dir cti ns (ad pted fr m [ 2/])

re ur p riod o f 1 ye r.

Cha ter 3 : Re built F -F mod l

- 8-

The survival condition is defined as a condition during which the platform may be subjected to extreme environmental loadings based on a 1/100 annual probability of exceedance, i.e. 100 year return period. In these circumstances the platform is shut down due to the severity of the actions, and so a smaller portion of the live loads are considered.

3.3.2

Permanent loads

As mentioned before, the permanent loads consist of the structure self-weight and the dead weight of non-modelled items, such as structural secondary/tertiary steel, architectural items, mechanical equipment’s (crane weight included), electrical and instrumentation items and piping.

a. Structure self-weight

The structure modelled line elements generates a self-weight based on a unit weight of the steel of 3

7850 kg/m .

b. Dead weight of non-modelled items

A document called the Weight Control Report (WCR) [/8/] is made during the project, providing all permanent loads (total weights and center of gravity). The srcinal F3-FA SACS model has been checked against the WCR and the main permanent loads were modelled (as concentrated forces or distributed loads) on their correct location.

In the re-built F3-FA model, the dead weight of non-modelled items has been modelled as distributed loads, with no respect to their location on the deck.

The two secondary decks (intermediate and mezzanine decks) were not modelled in the re-built model, as mentioned earlier, and the corresponding loads have been applied in the deck below. This results in a vertically different location of the center of gravity. The sensitivity of this difference has been checked by comparing the results, in section 3.5, and no significant impact has been found.

3.3.2.1

Topside


The topside self-weight includes the main steel beams, braces, columns and secondary steel, such as plating, stringers, handrailing and kick plates. The plating density has been increased to account for stringers, handrailing and kick plates.

-39-



The additional non-modelled weight of secondary steel and tertiary steel and non-modelled items, such as architectural, mechanical equipment, piping, electrical and instrumentation have been modelled as uniformly distributed loads in the re-built model. Except for the crane dead weight, that is applied as a concentrated load in the crane pedestal.

The non-modelled items weight are divided in nett and gross weights. The nett weights refers to dry equipment, without fluids and turned-off. The gross weights are measured with the operating equipment’s and all fluids necessary for operation, also they account also for uncertainties.

c. Results

The re-built model perm anent loads are presented in the table belo w in a load summary and distributed per deck.

Table 3.4: Re-built F3-FA model topside permanent loads distributed per deck (nett weights, i.e. without contingencies)

Permanent loads Structureself-weight

Cellar deck

Load case DDDD

1)

STEA

138,4

TOTALstructuralsteel

--

--

Architectural

ARCH

30,4

MECI

178,2

3)

Equipment operational

3)

Pipingdry

MECO PIPL

2)

(mT)

--

Non-modelled structural steel

Equipment dry

Main deck (mT) --

110,1

137,1

2 54,7

112,1

194,0 76,8

(mT) --

13,3

-94,5

214,7 44,2

E&I

42,5

30,1

33,6

TOTALnon-structuralsteel

--

--

--

--

411,1

2 4 1 2 ,4

--

--

1,4

--

2,9 4,2

Total nett weight (mT)

2001,2 12,3

--

151,8

Electrical & Instrumentation

1) Cellar deck includes weight and loads from the Mezzanine deck;

Ventstack

(mT) --

--

114,3

Helideck

(mT)

--

--

48,1

Top deck

-6,6

3,0

--

n/a 179,9

--

--

237,1 446,7

109,2

972,9

2) Main deck includes weight and loads from the Intermediate deck;

3) The equipment weight is divided in two different loads: one for dry weight and other for operational weight. n/a – not applicable.

The topside permanent loads from the re-built F3-FA SACS model have been checked against the srcinal F3-FA SACS model, as shown in Table 3.5.

A contingency factor is applied to account for uncertainties and to convert the nett weights in gross weights.

Table 3.5: Re-built F3-FA model topside permanent loads checked against the srcinal F3-FA model


-40-

SACS

Permanent loads

model

STEA

--

12,9

411,1

12,9

ARCH MECI

Equipmentoperational

MECO

Pipingdry

PIPL

--

n/a 179,9

(mT) 2259,4 464,1

2 7 2 3 ,5

237,1 446,7

Gross weight

27,3 --

301,8 n/a

11,8 15,0

744,8 206,9

E&I

109,2

12,4

122,8


--

9 7 2 ,9

--

1 3 7 6 ,3

TOTAL topside permanent loads

--

3 3 8 5 ,3

--

4 0 9 9 ,8

Topsideself-weight

1

2029,4

12,9

2291,2

STEA

372,4


--

Architectural

ARCH

Equipmentdry

2 4 0 1 ,8

MECI

Equipmentoperational

MECO

Pipingdry

PIPL


--

2001,2

2 4 1 2 ,4

Equipmentdry


F3-FA

factor (%)



Original

Contingency

(mT)

DDDD

Architectural

F3-FA

Nett weight

case

Topsideself-weight


Re-built

Load

E&IL

12,9

--

2 7 1 1 ,6

237,2 446,3 n/a 179,9

27,3 -11,8 15,0

109,2

12,4

--

9 7 2 ,9

--

TOTAL topside permanent loads

--

3 3 7 4 ,7

--

--

(%)

+ 0 ,3 1 %

301,9 n/a


Difference

420,4

--

745,2 206,9 122,8

1 3 7 6 ,8 4 0 8 8 ,4 + 0 ,2 8 %

Results show that no significant difference is found in the topside permanent loads between the rebuilt and the srcinal model.

Except for the structure self-weight and non-modelled structural steel loads, where the secondary decks (intermediate and mezzanine decks) are modelled in the srcinal model but in the re-built model are not. As, a consequence, the non-modelled structural steel loads have been increased, in the rebuilt model, to include the secondary decks.

-41-


3.3.2.1

Substructure


The substructure self-weight includes the legs, truss frame and leg transition frame (i.e. knuckle joint). The weight of the bucket is not included in the F3-FA model, because the weight of the foundations is only required for the design of the foundations.


Substructure non-modelled permanent loads are included in the F3-FA model and are generated by risers, pumps, padeyes, jacket anodes, anchor blocks, strands, etc.

c. Results

The substructure permanent loads from the re-built F3-FA SACS model have been checked against the srcinal F3-FA SACS model, as shown in Table 3.6.

Table 3.6: Re-built F3-FA model substructure permanent loads checked against the srcinal F3-FA model

SACS

Permanent loads

model

Re-built F3-FA

factor (%)

Substructure non-modelled loads

LXXW

4 6 3 8 ,0

LXXW

-(%)

2653,9 1)

12,0

--

-0,05%

Gross weight (mT) 2994,7 2223,8

5 2 1 8 ,5

12,9

1985,5

4 6 3 9 ,4 --

12,9

1985,5

1)

1

Substructure non-modelled loads

Difference

2652,5

--

TOTAL --

Contingency

(mT)

DDDD

Substructureself-weight F3-FA

Nett weight

case

Substructure self-weight

TOTAL Original

Load

12,0

--

2996,3 2223,8

5 2 2 0 ,0 --

-0,03%

1) LXXW = LA1W, LA3W, LE1W and LE3W.

Results show that no significant difference is found in the substructure permanent loads between the re-built and the srcinal model.

3.3.2.2

Total

The total permanent loads from the re-built F3-FA SACS model have been checked against the srcinal F3-FA SACS model, as shown in Table 3.7.


-42-

Table 3.7: Re-built F3-FA model total permanent loads checked against the srcinal F3-FA model

SACS

Permanent loads

model

Re-built F3-FA

Original F3-FA

--

Nett weight (mT)

Center of gravity

Gross weight

X

Y

Z

(m)

(m)

(m)

(mT)

Substructure permanent loads

4638,0

5218,5

Topsidepermanentloads

3385,3

4099,8

--

--

--

--

TOTAL permanent loads

8 0 2 3 ,3

9318,3

1 0 7 ,5

1 1 4 ,5

Substructure permanent loads

3374,7

4088,4

--

--

Topsidepermanentloads

4639,4

5220,0

TOTAL permanent loads

8 0 1 4 ,1 +0,11%

Difference

9308,4 + 0 ,1 1 %

--

--

1 0 7 ,5

1 1 4 ,4

0

+ 0 ,1

---

0 ,7 ---

0 ,8 -0,1

Results show that no significant difference is found in the total permanent loads between the re-built and the srcinal model.

3.3.3

Live loads

Live loads are all the loads expected to act in the topside during the platform service life.

a. Loads generated during the serv ice life of the platfo rm

The live loads include the weight of personnel, the weight of transportable equipment’s (e.g. vessels and containers) and of temporary access systems used during operations and maintenance (e.g. scaffolding).

For simplicity of the live loads set-up: 

the loads are spread-out over the whole area of the decks, and do not consider the span-tospan load distribution;



the live loads are multipl ied by a factor correspo nding to the percentage of the actual free deck space.

-43-


A resume of the uniform distributed live loads used, per area, is shown on Table 3.8.

Table 3.8: Live loads (adapted from [1]) 2 1)

Area

Uniform load (kN/m )

Helideck

2)

2,00

Top deck

15,00

Maindeckandcellardeck

10,00

Secondary decks (mezzanine and intermediate deck)

5,00

External walkways, stairs and fire fighting platforms

3,00

Laydownareas(cellarandmaindeck)

20,00

Laydown areas (mezzanine and intermediate deck)

10,00

Roofs(ifaccessible)andmusterarea

5,00

1)

The decks are also designed locally for concentrated live loads;

2)

The helideck is also designed, according to CAP [/34/] and ICAO [/35/] and [/36/] guidelines, for the emergency landing of a single rotor helicopter equivalent to the Sikorsky S-61 (with a Maximum Take-off Weight of 9,3 mT and a maximum capacity of 30 passengers).

For the structure global analysis only live loads applied on the cellar deck, main deck and top deck are considered.

A summary of the live loads applied per deck is presented in Table 3.9.

Table 3.9: Summary of Live loads

Sum of live load

Area (kN)

(mT)

Top deck

5607,9

571,8

Main deck

2619,3

267,1

Cellardeck

3297,4

336,2

The remaining live loads applied on the helicopter landing deck (“helideck”), weather deck, intermediate deck and mezzanine deck are used in local design analyses.

b. Crane operational loads

The platform will also accommodate a crane with a dead weight of 82,4 mT capable of operating a 14,5 mT cargo at 37,5 m reach [/1/] & [/2/]. During stormy weather (survival conditions) the platform crane will not operate and is positioned in boom rest, where the vertical dead load is (conservatively) distributed between the crane pedestal (605 kN) and the crane boom (219 kN).

The crane operational loads act as concentrated forces and moments acting on the crane pedestal.


-44-

In operational environmental conditions, the crane pedestal is checked for forces produced by the crane loads throughout the full 360º range (8 directions equivalent to environmental loads directions), as shown in Figure 3.14.

Figure 3.14: Crane operational load directions

The crane operational loads are considered in the operational condition and are implemented at 8 different directions (following the environmental loads) simulating a revolving crane [/1/] & [/2/], as presented in Table 3.10.

Table 3.10: Crane operational loads (adapted from [2])

Crane direction FX 0 29,7 90 150,3 180 209,7 270 330,3

3.3.4

3.3.4.1

Forces(kN) FY

47

0

40,8

23,3

0

47

-40,8

23,3

-47

0

-40,8 0 40,8

-23,3 -47 -23,3

FZ

-298 -298 -298 -298 -298 -298 -298 -298

MX 0

Moments(kNm) MY 12405

-6146,2 -12405 -6146,2 0 6146,2 12405 6146,2

475

10775 0

CRN1 475

475 475 475

-10775 0 10775

CRN2 CRN3

-10775 -12405

Load Case

MZ

CRN4 CRN5

475 475

CRN6 CRN7

475

CRN8

Environmental loads

Wind loads

Wind is stochastic by nature and their properties, i.e. the wind speed, vary in space and time. For the analysis required, it’s sufficient to describe the wind parameters in simplified statistical terms.

The mean and standard deviation of the wind speed and direction, taken over durations of the order of an hour, do not vary horizontally (only vertically with elevation). Therefore, periods over one hour or

-45-


less are suitable for the averaging of wind speeds and directions [/11/], and the wind velocity has to be classified by its elevation (at 10 meters above mean sea level is used as a standard reference height) and its duration [/4/] & [/5/].

For the overall design of the topside and substructure, API

[/4/] and the ISO [/5/] recommend the use

of sustained wind speeds. In the F3-FA platform a sustained wind speeds with an averaging period of 1.minute has been used for the global analysis of the structure.

9

The Metocean criteria for F3-FA platform retrieved the values listed in Table 3.11 , for the basic average wind velocities at 10 meters above mean sea level.

Table 3.11: Wind velocity (m/s) (adapted from [/2/])

All year wind speed at 10 meters above sea level Condition averaging period of 1 hour

averaging period of 1 minute

Operational (1 year return period)

24,7 m/s

29,8 m/s

Survival (100 years return period)

33,7 m/s

42,0 m/s

In the global analysis the wind loads are based in projected area loads. For all wind calculations a shape factor of Cs = 1.50 was used, except for the vent stack where a Cs = 1.00 was adopted.

The expressions used to evaluate the wind loads are based on the API [/4/] and ISO [/5/], [/6/] codes, as presented in appendix A, together with the results obtained and the area loads used.

For self-installing offshore platform in the North Sea it is not common practice to consider the dynamic effect in the structure global analysis. In fact, the static approach with the wind coefficient pressure and the shape factor of 1.5 is usually a safe scenario. So it is assumed that the F3-FA platform and is structural components have no appreciable dynamic response for wind induced loadings due to any of the structural modifications that will be study in Chapter 4.

3.3.4.2

Hydrodynamic loads

Waves combined with the current are usually the most important environmental action in the design of offshore structures. Besides creating very large loads, they have a significant influence on the dynamic behaviour of the structure and consequently on the fatigue life of the platform.

Waves are influenced by many factors, such as the wind stress on the surfa ce of the sea, tidal currents and local thermal sources. Thus, waves can be classified as random in nature [/12/]. 9

For the design of individual structural elements a wind gust speed with an averaging period of 3 seconds was used.


-46-

Current velocity is caused by the tidal and residual currents. The tidal currents are regular and can be anticipated since they precede or follow the highest and lowest astronomical tides, HAT and LAT. The residual currents are generated by the ocean’s circulation, storm-generated currents and other occurrences, such as density gradients (temperature), wind stress and internal waves [/11/].

The hydrodynamic loads are represented by distributed load diagrams, applied to the platform substructure legs, which increase exponentially with the water column elevation. The wave loads are calculated based on a theoretical wave model that best describes the sea state motion. The current velocities are collected on site, per elevation, and added to the wave velocities.

Wave kinematics Two basic models are typically applied on the design and analysis of offshore structures [/11/]: I.

deterministic periodic model;

II.

probabilistic random model.

I - Deterministic periodic model

“The periodic wave model is based on idealized wave forms (…). Several theories have been developed, depending on the ratio of wave height to wave length (…) and on the ratio of water depth to wave length. ” (J.H. Vugt, 2002, p.4-12) [/11/].

It’s common practice in the offshore industry to adopt a horizontal sea bottom and infinite horizontal free surface as boundary conditions and for the wave a two dimensional model with infinitive long crest without changing form. With these properties a wave can be described by a number of common accepted periodic regular wave theories, such as: Airy (linear) wave theory, Stokes 2

nd

& higher order

theories, Stream Function theory and Cnoidal wave theory.

The linear wave theory, presented in appendix B.1, is very usefully, combined with Morison’s equation, for the calculation of fatigue loads or preliminary ultimate loads for offshore structures. However, in general, it is not the desir ed design wave for fixed platforms in extreme conditions due to their limitations for shallower waters or waves too steep. It is shown in appendix B.1 that regular wave theories are based on an ideal fluid with two boundaries: the sea floor and the sea surface. In the sea floor, the water particle velocity normal to the bottom is zero. At the sea surface there is a kinematic and a dynamic condition to be satisfied.

-47-


The theories differ in respect to the free surface boundary (that is, the nonlinear condition), and are either linearized in one of the two conditions (kinematic or dynamic) or include nonlinear terms to a particular order (see appendices B.1 and B.2).

Thus, for example starting from the Linear theory and changing to the Stokes 2 adding a non-linear term – 2

nd

nd

order theory (by

sinusoidal component – to the wave surface equation) one is

progressively correcting the linear wave model for waves with steeper crests and flatter troughs, as presented in Figure 3.15.

Figure 3.15: Comparison between the linear wave model with a non-linear wave model (in this case the Stokes 2

nd

order wave)

With the increase order of the Stokes wave theory, not only we are increasing the complexity of the mathematic equations but also the accuracy of the sea state to certain conditions. Not always the increase of the wave theory order implies that the results are more realistic, and that applies to all the regular wave theories, since they depend on three main factors: wave period (T), wave height (H) and water depth (d).

The Fenton’s Stream Functio n theory uses a finite number of Fourier series to generate complex nonlinear wave equations capable of solving the wave motion problem [/13/], being a good alternative in shallow waters to the Airy’s linear and Stokes wave theories. II - Probabilistic random model

Besides the periodic deterministic model, the wave motion can be characterized by a random probabilistic model. The random wave model is based on the addition of infinite linear wave


-48-

components; each wave component is described by a periodic linear wave theory (Airy theory) with a different amplitude, frequency, direction and phase angle [/11/] & [/12/].

The random wave model is obviously a more realistic illustration of the sea with three-dimensional waves. In general, it is used in more deeper waters and dynamic sensitive structures (e.g. tension-leg platforms) [/11/], for the reason that it shares the same limitations of the Linear wave theory for shallow and/or steeper waves.

In Iv-Oil & Gas (inc. F3-FA platform) it is common practice to use only the deterministic wave method, with the linear wave theory implemented for the calculation of fatigue loads and the Stream Function th

11 Order for the in-place analysis. Appendix B.1 and B.2 present simplified formulations of the Linear wave theory and the Stream Function theory.

Environmental data and air gap a. Wave data

The regular wave theories have a period such that each cycle has exactly the same form. To describe a regular wave three parameters are needed (Figure 3.16): 

Wave period ( )



Wave height ( )

time between two successive wave crests to pass a fixed point; vertical distance between the wave crest and the following trough;



Water depth ( )

vertical distance between LAT and the sea floor.

Figure 3.16: Regular wave shape

For the F3-FA platform, the wave parameters are taken from the Metocean data analysis based on the forecast of extreme wave parameters [/2/]. Two different conditions are established: operational and survival. The operational condition is based on a wave with 1 year return period and the survival condition on a 100 years wave (Table 3.12). -49-


Table 3.12: Wave data (adapted from [/2/])

Analysis

(s)

(m)

(m)

Operationalcondition

12,4

14,9

Survivalcondition

15,1

20,7

40,300 40,300

For the analysis a maximum and minimum design still water depth (also given by the Metocean data) is considered. The maximum and minimum still water depths are taken from the environmental data presented in the in-place analysis report [/2/], as shown in Table 3.13.

Table 3.13: Design still water depth (adapted from [/2/])

Analysis

(m)

(m)


41,990

40,300

Survivalcondition

42,950

40,300

The maximum still water depth is based on the highest astronomical tide (HAT) and the storm surge.

b. Air gap

When designing an offshore platform, an air gap between the elevation of the lowest element on the platform topside and the wave height of an hypothetical extreme wave is considered. So that the risk of wave impact on the topside (bigger wave force in the structure), damage to platform equipment and consequently pollution of the sea is minimal. Two criteria’s are considered to define the platform elevation: 

The crest elevation of an extreme wave with 10.000 years return period: LAT (+) 19,40 m



The crest elevation of 100 years wave plus 1,50 meters of air gap: LAT (+) 15,00 m + 1,50 m

The maximum determined 10.000 years extreme crest elevation is LAT (+) 19,400 m and the bottom of steel for the lowest deck equals LAT (+) 19,810 m, allowing 0,410 m for subsidence and settlement (Figure 3.17).


-50-

Figure 3.17: F3-FA tides and storm surges

c. Current data

Currents vary in space and time at scales much larger than those for wind so, in general, currents are admitted to be a horizontal uniform flow field of constant velocity, which is only a function of depth. The current velocities provided by Metocean criteria for F3-FA platform were determined for still water conditions and are presented in the Table 3.14 [/2/].

Table 3.14: Current velocities (m/s) (adapted from [/2/])

Elevation above the sea floor

Analysis 1.00d

0.75d

0.50d

0.30d

0.10d

0.05d

1 meter


0,80

0,80

0,80

0,80

0,73

0,68

0,63

Survivalcondition

1,04

1,04

1,04

1,04

0,95

0,88

0,81

Note: d is the still water depth

Wave analysis Wave loads are dynamic by nature, however they can be satisfactorily represented by static equivalent forces based on the dynamic behaviour of the fixed platform.

-51-


Static wave analysis

The procedure to calculate the static wave loads for a deterministic wave model are given by the API [/4/] and ISO [/5/] & [/6/] codes and presented in Figure 3.18.

Figure 3.18: API & ISO procedure for calculation of deterministic static wave & current forces (taken from [4])

In appendix B.3, a simple description and application for the F3-FA structure is given for all the steps involved in the static wave analysis prescribed in API [/4/] and ISO [/5/] & [/6/] codes.

In this section a summary is made of all the parameters and steps used in the analysis of the F3-FA srcinal model and re-built model: a.

The Doppler effect (wave stretched due to collinear current) is considered numerically in the Stream Function theory for the in-place analysis; th

b. Wave kinematics: Stream Function 11 Order theory applies; a.

Wake kinematics factor is set to 0,95;

b. Current blockage factor is set to 1,00; c.

Current velocity profile is extended vertically from the mean water level to the wave cre st and truncated above the wave trough, since the current speed is constant in the majority of the water column extent, as shown in Figure 3.19;


-52-

Figure 3.19: Current force diagram on the wave crest and wave trough

d. Marine growth thicknesses on the radiu s around the sect ion applies as follows: - from seafloor till LAT -10 meters (-40,300 m to -10,000 m):

50 mm;

- from LAT -10 meters till LAT + ½ HAT (-10,000 m to +0,305 m):

100 mm.

3

A dry density of 1150 kg/m is considered for the marine growth e. For the hydrodynamic forces on the substructure the Morison equation applies; f.

The drag and inertia coefficients are facto red to account for the turbulence caused by the interruption of the flow, as shown in Table 3.15.

Table 3.15: Drag and inertia coefficients used in the F3-FA design

Drag coefficient (Cd)

Inertiacoefficient(C

m)

Condition smooth members Standard

rough members

0,65

1,05

In-placeoperational

1,12

1,28

In-placesurvival

1,02

1,10

Fatigue

0,50

0,80

smooth members 1,60 1,60 1,60 2,00

rough members 1,20 1,20 1,20 2,00

g. Conductor shielding factor is set to 1,00; h. No hydrodynamic models are consi dered for appurtenances, since the jacket anodes and conductors are on the inner surface of the legs.

The hydrodynamic forces are calculated through Morison’s equation; that is valid for a submerged rigid vertical cylinder with a slenderness of

, diameter relative to wave length, and is given

by: (3.1)

-53-


Where,

is the hydrodynamic force acting normal to the axis of the member; is the inertia force and is given by: (3.2) is the drag force and is given by: (3.3) is the horizontal water particle velocity; is the horizontal water particle acceleration; is the density of the water; inertia coefficient; drag coefficient; leg diameter (inc. marine growth).

Dynamic wave analysis

As mentioned before, the self-installing platform response to wave loading is more sensitive than a typical jacket platform. Due to a less stiffer substructure the platform has an higher natural period, closer to the wave period, resulting in greater deformations and stresses due to the wave forces. A practical approach to quantify the dynamic response of the structure is to use a factor named the dynamic amplification factor (DAF). The DAF factor is a relation between the dynamic response and the static response of the structure.

In the F3-FA project, the dynamic load effect is considered through the structure modal response, adjusting Morison’s equation in the calculation of the fluid forces on members.

The Morison equation expression is modified to account for the structure relative velocity by adjusting the drag force expression, according to API [/4/], as follows:



(3.4)

Where, the horizontal water particle velocity is subtracted from the structural velocity compon ent normal to the axis of the leg members,

(m/s).


-54-

For more information over the dynamic wave analysis, and effect on the F3-FA structure, reference is made to appendix C.

Global hydrodynamic loads

It is common to refer to the global hydrodynamic loads on the structure, which are: 1) The total applied horizontal load at the sea floor, often called the base shear; 2) The total moment about the sea floor, named the overturning moment.

For a better understanding of the hydrodynamic loads involved, an estimation of the global loads is made. The wave and current loads are calculated, through Morison equati on, based on the wave kinematics determined with Airy’s linear wave theory for a single F3-FA column, and presented in appendix C.

It is clear from the preceding appendices B and C that the calculation of hydrodynamic loads on a space frame structure is a complex and laborious process.

The difficulty to design a 3D frame structure with non-linear wave kinematics formulations and to account for the dynamic structural response of such loads, in a short period of time, is a problem that, other than for the simplest geometries, can only be solved adequately by powerful software tools.

It is however imperative not to lose the insight in structural behaviour, only this way can the Engineer be able to detect input human errors, question the program solving methods and results, and achieve creative engineering solutions to different problems. To achieve this propose the preliminary design and structure description play an important role.

In the F3-FA design, the calculation of hydrodynamic loads is always simulated using SACS thoroughly verified computer program in the offshore industry.

3.4

Basis of analysis

3.4.1

General bases

The F3-FA structural design involves two main stages [/1/]:

-55-



Fabrication, transport and installation;



Operation and survival conditions.


[/10/], a

The first stage concerns the most critical phases during the platform construction, assembly, transport and installation. The second stage involves the entire life of the platform, after the installation, exposed to the design environmental conditions and operational loads.

In the operation and survival conditions, to ensure the safety of the entire structure and to satisfy the Client and certificatory requirements, three main global analyses are considered in an offshore platform design and thus also for the F3-FA design: 

In-place analysis;



Fatigue analysis;



Accidental assessment (boat collision and post-impact analysis).

The in-place analysis covers the stress and buckling check of all structural members and the strength and stability integrity of the global structure for the design operational and survival conditions. The fatigue analysis evaluates the fatigue damage caused by the cycling wave loads for the overall main structure.

The boat collision analysis evaluates the direct and indirect structure response to the impact of a supply boat. Based on the imposed leg deformation caused by the accident, a post-impact analysis covers the load distribution and corresponding stresses in the remaining intact frame members for a specific environmental condition combined with topside operational loads.

In the current report only the in-place analysis and fatigue analysis are covered. In both of the analyses (in-place and fatigue), besides the modelling process (described in section 3.2) and the design actions (described in section 3.3), other important parameters are introduced below. a. Site specific data

The Metocean environmental data, soil data and water depth are determined for a fixed location, in this case the F3-FA site (as shown in Figure 2.1).

For the in-place analysis the data used to characterize the environmental actions are the following: 1) Wave period, wave height and water depth for the wave load; 2) Current and wind speed for both loads. (According with section 3.3.4.2). In the fatigue assessment only wave loads are considered and the hydrodynamic action is described by the number of wave occurrences, in one year, with a specified wave height and period.


-56-

The soil conditions are modelled as a set of non-linear strings, as referred in section

3.2.4, based on

the soil assessment and suction bucket design. SPT Offshore (co-partner of Iv-Oil & Gas on the F3-FA design) provided a soil stiffness matrix.

b. Materials

The steel grade used for the primary structure is the S355 following the European standard specification, designed for the offshore sector, EN 10225 [/17/]. Steels with the denomination “G” follow this code. In the F3-FA design, the primary steel material is classified in different types, as shown in Table 3.16.

Table 3.16: General primary steel classification for the F3-FA structure

Materialtype

Description

Standardsteel

V

Primarysteel

S355G7+M, for plates S355G11+M, for sections S355G14+N, for hollow seamless **

VI

Primary steel + TTP*

S355G10+M, for plates t ≤ 100 mm S355G10+N, for plates t > 100 mm S355G12+M, for sections S355G15+N, for hollow seamless **

* TTP = Through Thickness Properties ** Seamless hollow section : hollow long product, where the shape is made by piercing process.

The denominations +N and +M are relative to the delivery conditions. “N” stands for a normalizing rolling process, where the final product (shape) retains the srcinal mechanical properties. And “M” stands for a thermo-mechanical rolling or quenching process. In both cases the final product mechanical properties are modified and cannot be achieved by heat treatment alone. Moreover, the steel is not homogenous due to the redistribution of the alloy grains in the final product, the TTP (Through Thickness Properties) quality assures a more homogeneous steel structure.

The F3-FA overall main structure steel (topside and substructure) is type V. The legs joint can’s (in the connections with the transition frame and bucket) are type VI, due to significant transverse forces.

To satisfy EN 10225 [/17/] material toughness requirements, in all steel types samples tests should be carried out for Charpy V-notch impact at -40 ºC (degree Celsius) with a minimum average impact energy of 50 J (Joules).

For structural resistance evaluation, and still according with EN 10225, the steel yield stress depends on the wall thickness, as shown in Table 3.17.

-57-


Table 3.17: Steel yield stress based on wall thickness (WT)

Steel grade S355 2 fyield (MPa or N/mm )

WT (mm) 0 < WT ≤ 16

355

16 < WT ≤ 25

345

25 < WT ≤ 40

345

40 < WT ≤ 63

335

63 < WT ≤ 80

325

80 < WT ≤ 100

325

100 < WT ≤ 150

320

Finally, other general mechanical characteristics of the steel adopted are: 3

-

Steel density:

7850 kg/m

-

Young modulus:

200 GPa or 2,0 x 10

-

Shear modulus:

80 GPa or 8,0 x 10

-

Poisson ratio:

0,25 (defined indirectly by the shear modulus)

4

5

2

N/mm 2

N/mm

c. Welds and wall thickness transition

It is common practice in offshore projects, to define construction good practices to ensure a good performance for the entire structure in harsh offshore conditions. When designing and analysing a structure it is important to define or/and be conscious of such practices.

In the design, all welds are defined to have a full penetration, unless noted otherwise. In the current study, attention is paid to the legs tubular joints in the fatigue analysis. These major joints are completed through a butt weld configuration with double sided V-groove welds, as shown in

Figure

3.20.

Figure 3.20: Typical double sided V-groove weld

This type of welding offers, in general, a smaller amount of weld volume, reduced welding time and costs and, consequently with less weld metal in theory one has less residual stresses, longitudinal shrinkage and weld defects.


-58-

The welding runs are made in the “down hand” technique, which means that the welder starts at the top and works downward in vertical welds. The benefit of this technique is that the metal deposit and base metal blend better and that most of the slag will float to the surface and will not be included in the weld. Thereafter, in the offshore industry usually submerged arc welding method is avoided due to the possibility of potential dangerous residues be left behind affecting the weld quality.

Section wall thickness transitions along the leg full length are very common, to keep up with the member stresses and provide an economical solution. On these particular joints sharp edges are avoided and the wall thickness transitions have a chamfered ratio 1:4, to prevent any local structural weakness, as shown in Figure 3.21.

Figure 3.21: Typical wall thickness transition

d. Marine growth and corrosion

In both analyses, in-place and fatigue, a marine growth profile is considered in all members from the mudlevel up to half of highest astronomical tide (HAT) and a corrosion effec t is adopted in the members located in the splash zone.

The marine growth effect is reflected in the hydrodynamic loads and legs mass.

The hydrodynamic loads, are influenced by the marine growth through the drag and inertia coefficients in Morison equation. These values are selected depending if the member is considered “smooth” or “rough” (see appendix B and C for example). All members from the mudlevel (-40,300 m) up to +0,305 m (½ HAT) are considered to be “rough”.

The leg mass considers the marine growth mass by increasing the member’s weight, assuming an arc 3 element, with a dry density of 1150 kg/m , surrounding the legs section and with the radius given in Table 3.18.

-59-


Table 3.18: Marine growth layer thickness for the F3-FA design

Range

Layerthickness

from sea floor till LAT-10 meters

-40,300 m to -10,000 m

from LAT -10 meters till LAT + ½ HAT

-10,000 m to +0,305 m

LAT = Lowest Astronomical Tide;

50 mm 100 mm

HAT = Highest Astronomical Tide.

The corrosion is reflected in the legs stiffness and strength by reducing the legs wall thickness. The outer surface of the substructure legs in the splash zone (between -4,00 m and +5,00 m) is considered corroded with a decrease in the wall thickness of 5 mm. However, the steel density of these members is conservatively modified to account for their total weight without corrosion, as already referred in section 3.2.

e. Design (structural codes and design lif e)

The F3-FA structure was designed for a service lifetime of 20 years . The member stresses and stability are verified according to the American Petroleum Institute (API) standard code API RP 2AWSD [/4/]. API only addresses to substructure tubular members and reference is made to the American Institute of Steel Construction specification code ANSI/AISC 360-05 [/7/] for topside members.

The Working Stress Design (WSD) method, implemented in the API [/4/] code, was developed to ensure that the structure stresses are well below the elastic limit strain for the design loads, through working (or allowable) stresses determined by multiplying the material yield stress by a safety factor.

Until 2010, it was common practice in the North Sea to design an offshore structure using the API [/4/] code and the WSD method. Nowadays, the Limit State Design method 11

ISO

10

is being enforced by the

19900 standards, e.g. ISO 19901 [/5/] and ISO 19902 [/6/], meant to replace the API [/4/].

f. Buoyancy and hydrostatic collapse

The F3-FA legs and buckets are flooded, which means these elements are not subjected to significant hydrostatic pressures. However, all other members underwater are not flooded and thus need to be checked for hydrostatic collapse, based on the design hydrostatic head,

, calculated according to

API [/4/], as follows: (3.5)

10 11

Also known as the Load & Resistance Factor Design (LRFD). ISO = International Organization for Standardization.


-60-

Where, – depth below still water surface including tide, adopted as 42,95 m; – water depth, adopted as 40,30 m; – wave height; – wave number, equal to 2π/λ; λ – wave length; 3

3

– seawater density, adopted as 1,028 mT/m = 0,01008 MN/m .

In the F3-FA design, the hydrostatic head was determined based on the worst situation, that is survival condition with maximum water depth. Table 3.19 presents the calculation for the hydrostatic head.

Table 3.19: Hydrostatic head calculation for the F3-FA structure

Conditions Operational min water depth Operational max water depth Survival min water depth Survival max water depth

wave

water

wave

wave

Intrinsic wave

Actual wave

period

depth

number

height

period

length

head

(m)

HZ (m)

T(s) 12,4

12,4

d(m) 40,300

41,990

k

H(m)

0,0290

14,9

0,0286

Ti (s)

14,9

Hydrostatic

12,985

216,757

4 4 ,5

12,978

219,360

4 6 ,3

15,1

40,300

0,0222

20,7

15,965

283,577

4 7 ,6

15,1

42,950

0,0217

20,7

15,946

290,024

5 0 ,3

Ti is the wave period combined with a collinear current, see appendix C.

The transition frame and truss members are not flooded and are design to withstand the hoop stress (

) due to the hydrostatic pressure (

).

g. Buckling factor

The F3-FA substructure legs are not braced as a conventional jacket and are analysed with a design buckling factor of 2.0, i.e. the legs unbraced length is double of the legs free length.

h. Dynamic sensitivity

The F3-FA platform is sensitive to dynamic excitation, for that reason a dynamic analysis is performed. The dynamic analysis consists of a modal analysis to determine frequency modes and a dynamic wave analysis to determine the equivalent (enlarged) static load, as referred in section

3.3.4.2. The

dynamic wave analysis is performed with a design damping factor of 2%, to account for the

-61-


hydrodynamic damping, i.e. the decrease of the platform structure oscillations due to the friction forces in the legs caused by the hydrodynamic drag pressure.

3.4.2

In-place analysis

The in-place analysis involves the stability and strength determination of the structure when subjected to the normal operational condition and the extreme (or survival) condition. For the F3-FA platform, the normal operating condition covers the 1 year return period and the survival condition the 100 years return period for wind, current and wave actions.

The wave forces are calculated through the Morison equation and based on non-linear wave kinematics theory, the Stream Function 11

th

order. The wave position considered in the in-place

analysis is based on the maximum base shear forces and overturning moments calculated (according with, appendix C).

The in-place environmental actions are, conservatively, considered collinear (i.e. wind, current and wave) and cover 8 different directions around the wind rose, as shown in

Figure 3.13. The topside

permanent and live loads are combined with the environmental actions, referred to section

3.4.4. All

structural members are checked according to API [/4/] and AISC [/7/], as previously presented.

More attention is given to the legs design, and an example of the buck ling and stres s check for cylindrical members, according to API [/4/], is given in appendix D.

3.4.3

3.4.3.1

Fatigue analysis

Introduction

All steel structures have material imperfections, which can fracture under cyclic loads, such as waves. Normally, the fatigue assessment is based on a stress analysis of the structure and can be summarized as follows: 1) For a given cyclic load a certain stress range can be determined; 2) The number of cycles for failure can be specified based on a logarithmic relation with the stress range, known as S-N curve; 3) The expected damage is given by the ratio between the number of cycles likely to occur for the cyclic load in the operational life of the platform and the number of cycles for failure; 4) The total damage is obtained by the sum of the expected damage for all cyclic actions considered; 5) The design fatigue life is the inverse of the total damage.


-62-

The fatigue assessment is performed on the offshore structure joints and welds, where the material imperfections are critical. To account for stress concentrations the nominal stress ranges are multiplied by a stress concentration factor (SCF).

For the self-installing platform, in the fatigue assessment a spectral analysis is used, mainly because waves are stochastic and should not be defined as regular with a constant period.

The spectral fatigue analysis is divided in four different parts, as explained on section 3.4.3.2: a. Hydrodynamic loading model, in which the structure resp onse to wave loadi ng and the wave action (seastate) are characterized; b. Joint stress model, referring to the calcu lation of the stress range for a specific sea state and 12

joint hotspot ; c.

Fatigue damage model, for determining the expec ted damage occurrence in a specific seastate and hotspot, based on the number of cycles expected to occur for the seastate and the number of cycles necessary for failure of the joint;

d. Fatigue life of a joint, calcu lated based on the maximum damage expected from all the specified joint hotspots.

The cycling loading considered in this assessment is limited to the environmental wave action, thus no wind and current forces are adopted in this analysis, as is common practice in the offshore industry.

In the F3-FA design, a spectral fatigue analysis was performed for the overall main structure. The current study do not pretend to make an exhaustive fatigue analysis for all structure joints, but rather to focus on critical welds sensitive to fatigue, due to the structural modifications proposed in Chapter 4. In fact, the substructure leg butt welds are critical joints in the platform and are very sensitive to structural modifications, thus the fatigue assessment focus on them.

The load combinations used in the fatigue assessment are specified in section 3.4.4.

3.4.3.2

Fatigue spectral analysis

A briefly description of the spectral analysis is done, following Figure 3.22 that presents a flow diagram showing the fatigue life estimation of a single joint hotspot.

12

The hotspot is a particular location in a joint where cracks may occur due to fatigue.

-63-


a. Hydrodynamic Loading Model

To determine the fatigue life first the wave loading and the structure response must be estimated.

Wave response transfer functions

The maximum stress range at a given joint of the structure model is calculated based on 16 different 13

crest positions, per wave height, according to Morison equation and Airy linear wave theory, as described in section 3.3.4.2 and appendices B and C.

A number of wave heights values are carefully chosen to cover the highest structure frequency modes. The wave height is considered to have a constant wave steepness

14

of 1/18 and thus, the

relation between the wave height and the wave period is known, providing the variation of the calculated wave force in a frequency domain.

Through the calculated wave forces, the relationship between the nominal stress range and a wave period can be then determined for all the structure joints. This relation is known as wave response transfer functions

and is calculated for eight different wave directions, as specified in section 3.3.

However, the varia tion of the wave force with wave height (or frequency) is not linea r since the Morison equation drag term has a second order term

. The drag force may be linearized but, for

the performed calculations, is perfectly reasonable to normalize the transfer functions per unit of wave height (by dividing the nominal stress range per wave height) [/31/].

13

Offsets of the crest with respect to the srcin of the coordinate system (wave incidence).

14

Wave steepness is a ratio of wave height divided by wave length ( H / λ ).


-64-

Figure 3.22: Spectra fatigue analysis: fatigue life estimation of a joint hotspot

-65-


Wave spectrum

The different sea states occurring during the F3-FA design life are represented by an Omni-directional scatter diagram, also normalized for one year, and presented in

Figure 3.23. In this scatter diagram

the annual frequen cy of the significant wave height against the peak wave period is represented, providing the probability of a significant wave height

15

with an associated wave period event.

Figure 3.23: Significant wave height (Hs) / Spectral peak wave period (Tp) – Scatter diagram (all year) (adapted from [/3/])

To provide a realistic simulation of the water free surface profile, based on the scatter diagram data, a wave spectrum is used. The wave spectrum is an empirical distribution, per frequency, of the energy of the waves in the sea. In the North Sea offshore industry two wave spectrums are commonly used: 16

JONSWAP

and Pierson Moskowitz [/32/].

The JONSWAP wave profile is the Pierson-Moskowitz spectrum multiplied by an peakness parameter, , and given by [/31/]: (3.6)

2,

[m

s]

Where, – is a constant, equal to

;

15

The significant wave height (Hs) represents the average height of the highest one-third waves in a wave spectrum.

16

JONSWAP = Joint North Sea Wave Project.


-66-

– is the wave significant height (the mean height of the highest third of the waves) (m); 2

– is the acceleration of gravity (m/s ) – is the angular wave frequency, equal to

-1

(s );

– is the wave period (s); – is the angular spectral peak frequency, equal to

-1

(s );

– is the peak wave period or significant wave period

(s);

– is the spectral width parameter equal to 0,07 if

and 0,09 if

;

– is the peakness parameter.

And the Pierson-Moskowitz spectrum results from de former adopting

. For the F3-FA site the

Pierson-Moskowitz prove to be the more accurate representation of the sea according to the Metocean report [/3/] and thus is used in the fatigue analysis.

b. Joint Stress Model

When the wave transfer function,

, and the spectral density function,

are defined, the

nominal stress range of a specified joint can be obtained.

Nominal stress range

Assuming that the design fatigue waves prescribed have a Gaussian and stationary random process, i.e. a certain probability distribution does not change with the length of time, the nominal statistic stress range for a given seastate in a specified joint of the structure is given by [/31/]: (MPa)

The nominal stress range,

(3.7)

, is calculated for eight (8) points around the circumference of the

members connected, called hotspots, covering the entire butt welding around the tubular members, as shown in Figure 3.24.

-67-


Figure 3.24: Number of points (hotspots) checked around the joint for fatigue assessment

Every

has an associated zero up-crossing wave period,

,

17

given by:

(s)

(3.8)

Stress concentration factor

The actual local stresses in a tubular connection can be higher in certain hotspots due to construction and fabrication “discontinuities”. For this reason, the maximum local hotspot stress range is obtained by multiplying the nominal stress by a stress concentration factor (SCF). For the butt weld joints the SCF is given by the parametric formulae called the modified Burdekin expression: (3.9) Where, – is the eccentricity due to the wall thickness change, given by:

(mm);

– is the thinner wall thickness (mm); – is the thicker wall thickness (mm); – is the fabrication tolerance(mm).

For the F3-FA butt welds the fabrication tolerance is considered to be 6 mm and the minimum SCF adopted in the fatig ue analysis is 1,30. The hots pot statistic stress ran ge for a specific seas tate

and

for a given joint hotspot can then be determined by: (MPa)

17

(3.10)

The zero up-crossing perio d is the mean period of all the waves considere d in the analysis.


-68-

c. Fatigue Damage Model

The fatigue failure of a welded joint primarily depends on the stres s range and thei r number of occurrences.

Number of cycles for failure (S-N curves)

Based on a large number of cyclic tests for a specific connection detail a relation between the stress range and the number of cycles to failure can be found. Usually the ratio between both values is formulated by an S-N curve. For the F3-FA design it was imposed that the S-N curves presented by the Health Safety Executive (HSE) Offshore Technology Report 2001/015 [/9/] would be used. These curves consist of a linear relatio nship between logarithmic functions of the stress range ( ) and the number of cycles to failure ( ), as given below: (3.11) Where, –

is the stress range (MPa);

–

is the number of cycles to failure;

–

is a constant;

–

is the inverse slope of the S-N curve;

–

is the wall thickness of the member (mm);

–

is wall thickness reference value, equal to 16 mm;

–

is the thickness exponent factor, equal to 0,30 for welded joints.

The expression has a thickness correction for plates thicker than 16 mm.

The parameters for the S-N curves are given by the HSE Offshore Technology Report 2001/015

[/9/]

and presented in Table 3.20.

Table 3.20: Parameters of the S-N curves

Environment

(MPa) 12,182

Air Seawater FC

15,637 11,705

FC = Free Corrosion and

-69-

15,637

53

5

-

3

CP = Cathodic Protection

are the initial values in the S-N curve


7

-

3 5

10 -

-

11,784

Seawater CP

(cycles)

3

84 -

1,02x10 -

6

igure 3.25: Des ign HSE-P curves (S-N cu rves) or welded lates in ai and sea ater ith 45 and 150

Cha ter

: R -built F3- FA

odel

m all thickne s

- 70 -

In this table three different environment conditions are considered: air (members not wet), seawater free corrosion (members in the splash zone) and seawater cathodic protection (members from the mudlevel up to the splash zone). The members defined in the splash zone have no endurance, i.e. the S-N curve represented by the seawater free corrosion has only one slope (more conservative).

Figure 3.25 presents the S-N curves, according to the HSE Offshore Technology Report 2001/015 [/9/] for welded plates with 45 and 150 mm wall thickness, the range of plate sizes used in this study.

Number of cycles expected to occur

From

the number of cycles expected for a given seastate can be determined: (3.12)

Where, – is the design lifetime of the specific joint; – is the fraction of design life associated with a specific seastate, determined through the wave scatter diagram data.

As referred before the F3-FA platform service life was established in 20 years. However, in the F3-FA project a design fatigue factor (DFF) applies to the tubular welded joints depending on the structural importance (functionality/criticality) and inspectability of the connection. For the substructure butt welded joints, a DFF of 5 applies [/3/] thus the design lifetime (DL) is 100 years.

Expected damage

Having the number of cycles expected to occur during the service life of the structure and the number of cycles for failure, of a specific joint hotspot, the resultant ratio between both values provide the expected damage for a specific joint hotspot and for a given seastate.

However, as formerly stated, the design fatigue waves are assumed to be a random process, and therefore the response is also a random process.

In this case, it is assumed that, the peak stress ranges are Rayleigh distributed and the expected fatigue damage for a specific joint hotspot and a given seas tate is equal to:

(3.13) for a given stress range

-71-

(MPa).


Finally the cumulative damage caused by a fin ite number of

cycles of stress range

from all

seastates considered, over the operational life of the F3-FA platform, is obtained by the PalmgrenMiner rule:

where is the number of critical seastates.

d. Fatigue Life

To conclude the calculation the fatigue life for a specific joint hotspot is obtained by:

(3.14)

The fatigue life of a particular joint is the minimum value of all the eight joint hotspots specified.

3.4.4

Load combinations

The load combinations for the F3-FA in-place and fatigue analysis are presented in this section.

Table 3.21 presents the permanent loads without the structure self-weight used in the load combinations. Table 3.22 presents the topside loads (permanent, live and wind loads) used in the modeshape analysis and for both in-place and fatigue analyses.

Finally, Table 3.23 to Table 3.26 present the in-place load combinations for the operational and survival conditions and with maximum and minimum topside loadings.

3.4.4.1

Effective live loads

In the global analyses, all main decks (cellar, main and top deck) are considered to be subjected to live load but only a fraction of the loading is applied on the topside, as is common practice in the offshore industry.

The fraction of the live load applied on the topside, per analysis, is considered as follows: -

75%

operational conditions in the in-place analysis;

-

50%

survival conditions in the in-place analysis;

-

0%

tension load cases (minimum vertical load effect) in the in-place analysis;

-

25%

fatigue analysis in general;

-

50%

fatigue analysis for critical joints on request of the certifying authority.


-72-

3.4.4.2

Load combination tables

The permanent loads applied in the SACS model are combined according with

Table 3.21. The load

combinations used for modeshape, in-place and fatigue analysis are resumed in Table 3.22.

Table 3.21: Permanent loads without SACS model self-weight (load case DDDD)

n o ti a in b m o C d a o L

l e e t s lr a tu c ru t s d le e d o m n o n e d i s p o T A E T S

l ra u t c e it h c r A

n o ti a t n e m u tr s n I & l a c rit c le E

H C R A

I & E

y r d g in p i P

ry d t n e m ip u q E l a ic n a h c e M

l a n tio a r e p o t n e m ip u q E l a c i n a h c e M

s m te i d le e d o m n o n 1 A g e L

s m te i d le e d o m n o n 3 A g e L

s m te i d e l e d o m n o n 1 E g e L

s m te i d e l e d o m n o n 3 E g e L

L IP P

I C E M

O C E M

W 1 A L

W 3 A L

W 1 E L

W 3 E L

AMIN

1.000

1.000

1.000

1.000

1.000

--

1.000

1.000

1.000

1.000

AMAX

1.129

1.273

1.124

1.150

--

1.118

1.120

1.120

1.120

1.120

With: AMIN = Minimum permanent loads for tension cases, excluding model self-weight; AMAX = Maximum permanent loads for compression cases, including contingencies and excluding the model self-weight.

-73-


Table 3.22: Load combinations for the modeshape, in-place and fatigue analysis (topside loadings)

l e e t s l a r u t c u rt s d e l e d o


M

D D D

m u im in M

s d a o l t n e n a m r e p m u im x a M

X + n io t c e ir d k c a tt a d in W

Y + n io t c e ri d k c a tt a d n i W

-X n io t c e ri d k c ta t a d in W

-Y n io t c e ir d k c a tt a d in W

I N M A

A X M A

1 N I W

3 N I W

5 N I W

7 N I W

s d a o l t n e n a m r e p

d a lo e v li e d i s p o T

E IV L

NOPE

--

0.750

--

1.000

--

--

--

--

NFAT

--

0.500

--

1.000

--

--

--

--

WMIN

0.950

--

0.950

--

--

--

--

--

WOPE

1.129

0.750

--

1.000

WSUR

1.129

0.500

--

1.000

WFAT

1.129

0.500

--

WND1

--

--

--

WND2

--

--

--

WND3

--

--

--

WND4

--

--

--

WND5

--

--

WND6

--

--

WND7

--

--

WND8

--

--

-----

1.000 ---------

1.000 0.869 -----0.869

--

--

0.495 1.000 0.495 -----

--

---

---

0.869 1.000

---

0.869 0.495 --

1.000 --

0.495

Being: NOPE = Modeshape analysis for the in-place analysis; NFAT = Modeshape analysis for the fatigue analysis; WMIN = Minimum vertical load for tension cases in the in-place analysis; WOPE = Maximum vertical load for operational conditions in the in-place analysis; WSUR = Maximum vertical load for survival conditions in the in-place analysis; WFAT = Maximum vertical load for the fatigue ana lysis; WND1 to WND8 = Survival wind load for 8 different attack directions.


-74-

Table 3.23: In-place analysis load combinations for operational conditions with maximum vertical load d a o l l a ic rt e v . x a M

n o it a n i b m o C d a o L

E P O W

A001

1

A002

1

A003

1

A004

1

A005

1

A006

1

A007

1

A008

1

A041

1

A042

1

A043

1

A044

1

A045

1

A046

1

A047

1

A048

1

-75-

Operational waves loads in 8 attack directions with minimum water depth

1 0 L W

2 0 L W

4 0 L l W

3 0 L W

5 0 L W

6 0 L W

Operational waves loads in 8 attack directions with maximum water depth

7 0 L W

8 0 L W

1 0 L X

2 0 L X

3 0 L X

4 0 L X

5 0 L X

6 0 L X

Wind operational loads In 8 attack directions

Crane operational loads In 8 different directions

7 0 L X

1 N R C

8 0 L X

1

2 N R C

3 N R C

4 N R C

5 N R C

(Note: WND1 to WND2 is calculated with the survival wind speed)

6 N R C

7 N R C

8 N R C

1 1

1 D N W

2 D N W

3 D N W

4 D N W

5 D N W

6 D N W

7 D N W

½ 1

1

½ 1

1

½ 1

1

½ 1

1

½ 1

1

½ 1

1

½ 1

1

1 1

½ 1

1

½ 1

1

½ 1

1

½ 1

1

½ 1

1


½ ½

1 1

8 D N W

½ 1

½

Table 3.24: In-place analysis load combinations for operational conditions with minimum vertical load

n o ti a in b m o C d a o L 1

d a lo l a c tir e v . n i M

Operational waves loads in 8 attack directions with minimum water depth

1 0 L W

IN M W

2 0 L

3 0 L

W

W

4 0 lL

W

5 0 L

W

6 0 L

W

Operational waves loads in 8 attack directions with maximum water depth

7 0 L

W

8 0 L

1 0 L X

W

2 0 L X

3 0 L X

4 0 L X

1B001

5 0 L X

1 B006 1 1

B007 B008 1B041 1 B042 1 B043 1 B044

1 B045 1 B046 1

B047

1

B048

8 0 L X

1 N R C

2 N R C

3 N R C

4 N R C

5 N R C

6 N R C

1 1 1 1 1

3 D N

W

4 D N

W

5 D N

W

½ ½

1

½ 1

½ 1

½ 1

½ 1

1

W

½ 1

1

1

2 D N

½ 1

1

W

½ 1

1

1 D N

½ 1

1

8 N R C

½ 1

1

7 N R C

½ 1

1

(Note: WND1 to WND2 is calculated with the survival wind speed)

½ 1

1 B0031

1 B005

7 0 L X

1

11B002

1 B004

6 0 L X

Wind operational loads In 8 attack directions

Crane operational loads In 8 different directions

½ 1

1


½ 1

½

-76-

6 D N W

7 D N W

8 D N W

Table 3.25: In-place analysis load combinations for survival conditions with maximum vertical load d a o l l a ic rt e v . x a M


Survival waves loads in 8 attack directions with minimum water depth

R U S W

1 0 L Y

2 0 L Y

3 0 L Y

4 0 L Y

5 0 L Y

6 0 L Y

Survival waves loads in 8 attack directions with maximum water depth

7 0 L Y

8 0 L Y

1 0 L Z

2 0 L Z

3 0 L Z

4 0 L Z

1 C001 1

5 0 L Z

1C004 1C005 1 C006 1 C007 1 C008 C041 1 C042 1 1 C043 1C044 1C045 1 C046 1 C047 1 C048

-77-

7 0 L Z

1 D N

8 0 L Z

2 D N

W

W

1

C002 1 1 1 C003

6 0 L Z

Wind survival loads In 8 attack directions

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1


1

3 D N

W

4 D N

W

5 D N

W

6 D N

W

7 D N

W

8 D N

W

Table 3.26: In-place analysis load combinations for survival conditions with minimum vertical load d a lo l a c tir e v . n i M


Survival waves loads in 8 attack directions with minimum water depth

IN M W

1 0 L Y

2 0 L Y

3 0 L Y

4 0 L Y

5 0 L Y

6 0 L Y

Survival waves loads in 8 attack directions with maximum water depth

7 0 L Y

8 0 L Y

1 0 L Z

2 0 L Z

3 0 L Z

4 0 L Z

1 D001 1

5 0 L Z

1D004 1D005 1 D006 1 D007 1 D008 D041 1 D042 1 1 D043 1D044 1D045 1 D046 1 D047 1 D048

7 0 L Z

1 D N

8 0 L Z

2 D N

W

W

3 D N

W

4 D N

W

5 D N

W

6 D N

W

7 D N

W

8 D N

W

1

D002 1 1 1 D003

6 0 L Z

Wind survival loads In 8 attack directions

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1

1 1


1

-78-

3.5

Comparison results

The re-built F3-FA SACS model is compared with the srcinal (and approved) F3-FA SACS model to guarantee the feasibility of the report and removing “human errors” on the input data.

Three parameters of high importance for the study, are used for comparison of the srcinal with the rebuilt F3-FA SACS models: frequency modes, loads and fatigue analysis results.

3.5.1

Frequency modes

The main comparison point between the two models is the natural periods and corresponding mode shapes.

It is important to refer that the live loads applied in the topside are considered only partially for the global (in-place and fatigue) analysis, as referred in section 3.4.4. So, for the F3-FA in-place and fatigue analysis were adopted respectively 75% and 50% of the total applied live load. That applies also to the modeshape analysis for the in-place and fatigue analysis. st

Figure 3.26 to Figure 3.28 present the 1 mode shapes of the re-built F3-FA SACS model.

Table 3.27 to Table 3.29 present a comparison between the model natural periods of the re-built F3FA model and the srcinal F3-FA. The first mode shapes present natural periods close to the values shown by the srcinal F3-FA model.

The later mode shapes are more sensible to local changes on the topsi de and, thus lose some precision, mostly due to the simplification when applying the vertical loads on the re-built F3-FA topside. However, the energy contain ed within each later mode is rather small, which means that these resonant modes will not induce the structure into resonance, regardless of the dynamic load applied.

As a result, the re-built F3-FA model reproduces well the dynamic behaviour of the F3-FA platform (more precisely it is close to the dynamic results of the srcinal F3-FA model).

-79-


st

Figure 3.26: Re-built F3-FA SACS model 1 mode shape

Figure 3.27: Re-built F3-FA SACS model 2


nd

mode shape

Figure 3.28: Re-built F3-FA SACS model 3

rd

mode shape

-80-

Table 3.27: First mode shapes natural periods / frequencies comparison between the re-built and the srcinal F3-FA model

Mode shape --1 2 3 4 5 6 7

Natural period (s) / Frequency (Hz) OriginalF3-FAmodel

Re-built F3-FA model 75% live load

Mode shape characterization

50% live load

75% live load

--

50% live load

2,73 s 0,366 Hz 2,70 s 0,370 Hz 1,94 s

2,67 s 0,375 Hz 2,64 s 0,379 Hz 1,91 s

2,74 s 0,366 Hz 2,71 s 0,369 Hz 1,96 s

2,67 s 0,374 Hz 2,64 s 0,378 Hz 1,92 s

0,514 1,43 s Hz 0,698 Hz 1,41 s 0,708 Hz 0,76 s 1,313 Hz 0,72 s 1,386 Hz

0,523 1,43 Hz s 0,698 Hz 1,41 s 0,708 Hz 0,76 s 1,320 Hz 0,72 s 1,392 Hz

0,511 1,44 Hz s 0,696 Hz 1,42 s 0,707 Hz 0,76 s 1,312 Hz 0,72 s 1,382 Hz

0,519 1,44 Hz s 0,696 Hz 1,42 s 0,707 Hz 0,76 s 1,318 Hz 0,72 s 1,388 Hz

-Sway mode shape: Single curvature in Y direction Sway mode shape: Single curvature in X direction Torsion mode shape Ventstack curve in X direction Ventstack curve in Y direction Sway mode shape: Double curvature in X direction Sway mode shape: Double curvature in Y direction

Note: X direction is the South-North direction and Y direction is the West-East direction.

Table 3.28: Higher natural periods up to a contribution of 75% of the live load for the re-built and the srcinal F3-FA model

Mode shape

Natural period (s)

Re-built F3-FA

Mode shape

Original F3-FA

Natural period (s)

Re-built F3-FA

Mode shape

Original F3-FA

Natural period (s)

Re-built F3-FA

Original F3-FA

8

0,694

0,695

17

0,305

0,303

26

0,218

0,225

9

0,667

0,670

18

0,289

0,299

27

0,213

0,217

10 11

0,576 0,545

0,576 0,547

19 20

0,275 0,271

0,281 0,272

28 29

0,197 0,188

0,204 0,198

12

0,409

0,410

21

0,263

0,264

30

0,182

0,197

13

0,401

0,402

22

0,258

0,260

31

0,178

0,193

14

0,370

0,371

23

0,255

0,256

32

0,172

0,186

15

0,364

0,365

24

0,243

0,251

33

0,154

0,180

16

0,321

0,321

25

0,241

0,241

34

0,153

0,174

Table 3.29: Higher natural periods up to a contribution of 50% of the live load for the re-built and the srcinal F3-FA model

Mode shape

Natural period (s)

Re-built F3-FA

Mode shape

Original F3-FA

Natural period (s)

Re-built F3-FA

Mode shape

Original F3-FA

Natural period (s)

Re-built F3-FA

Original F3-FA

8

0,694

0,694

17

0,304

0,302

26

0,215

0,221

9

0,665

0,669

18

0,287

0,298

27

0,209

0,214

10 11

0,576 0,545

0,576 0,546

19 20

0,274 0,270

0,280 0,272

28 29

0,195 0,188

0,202 0,198

12

0,409

0,409

21

0,263

0,263

30

0,178

0,196

13

0,401

0,401

22

0,258

0,259

31

0,175

0,192

14

0,370

0,371

23

0,254

0,25

32

0,171

0,185

15

0,364

0,365

24

0,239

0,250

33

0,154

0,177

16

0,313

0,313

25

0,235

0,236

34

0,153

0,173

-81-


3.5.2

Loads

Another aspect of comparison is the load case summary, reported by SACS software [/10/], providing the actual force values and center of gravity compared to the structural srcin in X, Y and Z directions (orthogonal).

Figure 3.29 presents a plan view of the F3-FA SACS model, with orthogonal coordinates for a better understanding of the center position of the loads applied in the structure.

Figure 3.29: Orthogonal coordi nates of the F3-FA SACS models – in plan view

Looking at Table 3.31, the following main differences between both models (srcinal and rebuilt) are listed below: -

Load cases DDDD (modelled structural steel) and STEA (non-modelled structural steel);

-

Load cases AMIN and AMAX (maximum and minimum permanent loads with a contingency factor, excluding the structure self-weight);

-

Load cases NOPE and NFAT (permanent and live loads for the modeshape analysis);

-

WMIN, WOPE, WSUR and WFA T vertical load combinations for the in-p lace and fatig ue analyses.


-82-

a. Load cases DDDD and STEA

Load cases DDDD and STEA present higher differences due to the fact that in the re-built F3-FA SACS model the intermediate decks are not modelled, and to compensate the loads were increased in load case STEA. Combining the load cases, DDDD and STEA, the re-built F3-FA SACS model has a force 113 kN higher than the srcinal F3-FA SACS model, which corresponds only to 0,23% of the total permanent vertical load.

b. Load cases AMIN and AMAX

The load cases AMAX and AMIN consider only the permanent loads with and without a contingency factor and exclude the structure self-weight and thus, the differences are bigger than for the STEA load case.

c. Load cases NOPE and NFAT

The load cases NOPE and NFAT, used in the modeshape analysis, consider the load case AMAX and a percentage of the load case LIVE and consequently, the differences are bigger than for the AMAX load case.

However, the modeshape analysis combines these load cases with the self-weight of the structure and so, no big differences are noted in the frequency modes, as mentioned before in section 3.5.1.

d. Load cases WMIN, WOPE, WSUR and WFAT

The load cases WMIN, WOPE, WSUR and WFAT represent the vertical loads combined, with or without contingency factors, and are used in the in-place and fatigue load combinations. Their differences are mainly explained by the inconsistencies in the load cases DDDD and STEA.

Table 3.30 presents a comparison between the srcinal and the re-built models of the total vertical forces produced by the load combinations used in the analyses covered by this work.

Table 3.30: SACS models load case results

Original F3-FA SACS model

Re-built F3-FA SACS model

Z-DIRECTION

LOAD LABEL

FORCE

X

Difference between both models

Z-DIRECTION FORCE

X

Z-DIRECTION

Y

Z

Y

Z

FORCE

WMIN

kN -58308.9

m m m 107.5 114.2

kN 8.9

m m -58419.4

m 107.5

kN 114.3

m m m 8.8 110.5

X

Y

0.0

0.1

WOPE

-80448.5

107.3

114.4

11.9

-80619.7

107.4

114.4

11.8

171.2

0.1

0.1

-0.1

WSUR

-77583.3

107.3

114.3

11.3

-77738.5

107.4

114.4

11.1

155.2

0.1

0.1

-0.1

WFAT

-77583.3

107.3

114.3

11.3

-77738.5

107.4

114.4

11.1

155.2

0.1

0.1

-0.1

-0.2

More detailed information on the load combinations is given by Table 3.31, Table 3.32 and Table 3.33.

-83-


Z

Table 3.31: SACS models load case summary report for vertical forces (in Z direction)

Original F3-FA SACS model LOAD LABEL


Z-DIRECTION FORCE

X

kN

m


Z-DIRECTION Y

m m

Z

FORCE

kN

m m

X

Z-DIRECTION Y

m

Z

kN

FORCE

m m

m

X

Y

DDDD

-28717.1

107.4

113.8

26.2

-28449.9

107.5

113.8

26.1

-267.2

0.0

0.0

0.0

STEA

-3651.7

106.7

116.4

31.8

-4031.8

107.5

113.8

29.8

380.1

0.8

-2.6

-2.0

ARCH

-2325.7

E&I

-1071.1

108.1

107.2

108.0

101.8

29.8

31.0

-1071.0

-2324.8

107.2

107.2

114.6

30.8

-1.0

1 13.9

28.2

0.0

0.0

12.8

-0.3

-0.9

6.0

-1.6

PIPL

-1764.2

107.3

117.7

31.0

-1764.7

107.5

1 13.8

29.6

0.5

0.2

-3.9

-1.4

MECI

-4376.8

106.7

118.0

28.6

-4380.8

106.2

1 14.5

28.7

4.0

-0.5

-3.5

0.1

MECO

-6536.3

106.3

117.5

28.5

-6533.1

106.5

1 14.5

28.5

-3.2

0.2

-3.0

0.1

LA1W

-4922.3

121.8

92.2

-29.7

-4922.3

121.8

92.2

-29.7

0.0

0.0

0.0

0.0

LA3W

-4847.0

93.1

92.1

-30.3

-4847.0

93.1

92.1

-30.3

0.0

0.0

0.0

0.0

LE1W

-5321.9

121.5

138.0

-27.6

-5321.9

121.5

1 38.0

- 27.6

0.0

0.0

0.0

0.0

LE3W

-4380.0

92.7

138.9

-37.8

-4380.0

92.7

138.9

- 37.8

0.0

0.0

0.0

0.0 0.0

LIVC

-3297.9

107.1

114.5

20.5

-3297.4

107.1

1 14.5

20.5

-0.5

0.0

0.0

LIVM

-2555.6

105.0

113.9

28.7

-2619.3

104.8

1 14.3

28.7

63.7

-0.3

0.4

0.0

LIVT

-5607.1

105.6

116.6

36.5

-5607.9

105.6

1 16.6

36.5

0.8

0.0

0.0

0.0

LIVE

-11460.6

105.9

115.4

30.2

-11524.6

105.8

CRN1/8

-298.0

94.5

100.3

42.3

-298.0

94.5

115.5 100.3

30.2 4 3.9

64.0 0.0

-0.1 0.0

0.1 0.0

Z

0.0 1.6

AMIN

-32660.7

107.6

114.6

-6.3

-33044.2

107.6

114.8

-6.2

383.5

0.0

0.1

0.1

AMAX

-39431.4

107.5

114.6

-3.8

-39856.3

107.6

114.8

-3.7

424.8

0.1

0.1

0.0

NOPE

-48026.9

107.2

114.8

2.3

-48499.7

107.3

114.9

2.3

472.8

NFAT

-45161.7

107.28

1 14.73

0.54

-45618.6

107.4

114.8

0.5

456.8

0.1

0.1

0.0

WMIN

-58308.9

107.5

114.2

8.9

-58419.4

107.5

1 14.3

8.8

110.5

0.0

0.1

-0.2

0.1

0.1

0.0

WOPE

-80448.5

107.3

114.4

11.9

-80619.7

107.4

114.4

11.8

171.2

0.1

0.1

-0.1

WSUR

-77583.3

107.3

114.3

11.3

-77738.5

107.4

114.4

11.1

155.2

0.1

0.1

-0.1

WFAT

-77583.3

107.3

114.3

11.3

-77738.5

107.4

114.4

11.1

155.2

0.1

0.1

-0.1

Tables legend: DDDD = Modeled structural steel; STEA = Topside non-modeled structural steel; ARCH = Architectural; E&I

= Electrical & Instrumentation;

PIPL

= Piping dry;

MECI

= Mechanical Equipment dry;

MECO = Mechanical Equipment operational; LA1W = Leg A1 non-modeled items; LA3W = Leg A3 non-modeled items; LE1W = Leg E1 non-modeled items; LE3W = Leg E3 non-modeled items; LIVC

= Live load in cellar deck;

LIVM

= Live load in main deck;


-84-

LIVT

= Live load in top deck;

LIVE

= Total live load in topside;

CRN1 to CRN8 = Crane operational loads 8 different directions; AMIN

= Minimum permanent loads for tension cases, excluding model self-weight;

AMAX = Maximum permanent loads for compression cases, inc. contingencies and exc. model self-weight; NOPE = Modeshape analysis for the in-place analysis; NFAT = Modeshape analysis for the fatigue analysis; WMIN = Minimum vertical load for tension cases in the in-place analysis; WOPE = Maximum vertical load for operational conditions in the in-place analysis; WSUR = Maximum vertical load for survival conditions in the in-place analysis; WFAT = Maximum vertical load for the fatigue ana lysis.

Table 3.32: SACS models load case summary report for forces X direction



X-DIRECTION FORCE

X

kN

m

CRN1

47

CRN2

40.8

Y m


X-DIRECTION Z

FORCE

m

kN

m

100

115.24

39.84

47

100

115.24

39.84

40.8

m

X m

X-DIRECTION Y

Z

FORCE

kN

m

m

100

115.24

40.75

0.0

100

115.24

40.75

0.0

X

Y

m 0.0 0.0

0.0

0.9

0.0

0.9

CRN3 CRN4

-40.8

100

115.24

39.84

-40.8

100

115.24

40.75

0.0

CRN5

-47

100

115.24

39.84

-47

100

115.24

40.75

0.0

0.0

CRN6

-40.8

100

115.24

39.84

-40.8

100

115.24

40.75

0.0

0.0

0.0

0.9

100

115.24

39.84

115.24

40.75

0.0

0.0

0.0

0.9

0.0

0.0 0.0

0.9 0.9

CRN7 CRN8

40.8

40.8

100

WIN1

2879.5

100.09 1 11.95 3 2.12

2878.5

99.16

111.96 3 2.15

1.0

-0.9

0.0

0.0

-2879.5

114.6

-2878.5

115.7

111.96 3 2.15

-1.0

1.1

0.0

0.0

WND1

2879.5

100.09 1 11.95 3 2.12

2878.5

99.16

111.96 3 2.15

1.0

-0.9

0.0

0.0

WND2

2502.29

100.09

32.12

2501.42

99.16

111.96

0.9

-0.9

0.0

0.0

WIN3 WIN5

111.95 3 2.12

WIN7

111.95

32.15

WND3 WND4

-2502.29

114.6

111.95

32.12

-2501.42

115.7

111.96

32.15

-0.9

1.1

0.0

0.0

WND5

-2879.5

114.6

111.95

32.12

-2878.5

115.7

111.96

32.15

-1.0

1.1

0.0

0.0

WND6

-2502.29

114.6

111.95

32.12

-2501.42

115.7

111.96

32.15

-0.9

1.1

0.0

0.0

2502.29

100.09

111.95

32.12

2501.42

99.16

111.96

32.15

0.9

-0.9

0.0

0.0

WND7 WND8

-85-


Z

Table 3.33: SACS models load case summary report for forces Y direction



Y-DIRECTION FORCE kN

X m

Y m

m


Y-DIRECTION Z kN

FORCE m

m

X m

Y-DIRECTION Y

kN

Z m

FORCE m

X

Y

m

CRN1 CRN2

23.3

100

115.24

39.84

23.3

100

115.24

40.75

0.0

CRN3

47

100

115.24

39.84

47

100

115.24

40.75

0.0

0.0

CRN4

23.3

100

115.24

39.84

23.3

100

115.24

40.75

0.0

CRN5 CRN6

-23.3

100

115.24

39.84

-23.3

100

115.24

40.75

0.0

CRN7

-47

100

115.24

39.84

-47

100

115.24

40.75

0.0

CRN8

-23.3

100

115.24

39.84

-23.3

100

115.24

40.75

0.0

1623.86

107.72

96.92

33.4

1623.86

107.72

96.93

33.46

0.0

0.0

0.0

0.1

-1623.92

107.72

128.34

33.4

-1623.92

107.72

128.34

33.45

0.0

0.0

0.0

0.1

WND2

803.81

108.94

96.92

33.4

803.81

108.94

96.93

33.46

0.0

0.0

0.0

WND3

1623.86

107.72

96.92

33.4

1623.86

107.72

96.93

33.46

0.0

0.0

0.0

WND4

803.81

106.5

96.92

33.4

803.81

106.5

96.93

33.46

WND6

-803.84

108.94

128.34

33.4

-803.84

108.94

128.34

33.45

0.0

0.0

0.0

WND7

-1623.92

107.72

128.34

33.4

-1623.92

107.72

128.34

33.45

0.0

0.0

0.0

0.1

WND8

-803.84

106.5

128.34

33.4

-803.84

106.5

128.34

3 3.45

0.0

0.0

0.0

0.1

0.0 0.0

0.0 0.0 0.0

0.0

0.9

0.0

0.9

0.0

0.9

0.0

0.9

0.0

0.9

0.0

0.9

WIN1 WIN3 WIN5 WIN7 WND1

0.0

0.0

0.1

0.0

0.1 0.1

WND5 0.1

Tables legend:

CRN1 to CRN8 = Crane operational loads 8 different directions; WIN1

= Wind attack direction +X;

WIN3

= Wind attack direction +Y;

WIN5

= Wind attack direction -X;

WIN7

= Wind attack direction -Y;

WND1 to WND8 = Survival wind load in 8 different attack directions.

3.5.3

Fatigue analysis results

The third part of comparison between the original and the re-built model are the fatigue analysis results. As mentioned before, one of the main aims of this study is to achieve a better fatigue behaviour of the F3-FA platform through alternative structural solutions for the structure. Thus, it is fundamental that the re-built F3-FA SACS model, that serves as a reference for the F3-FA platform


-86-

Z

and for comparing the alternative structural models to be studied, can accomplish the same fatigue analysis results as the srcinal F3-FA SACS model.

It is common practice in the offshore industry to display the fatigue analysis results in terms of the fatigue lives estimation for each of the welded joints being considered. To reach these values a spectral fatigue analysis has been done, as explained in section 3.4.3.2.

Table 3.34 and Table 3.35 present a comparison between the srcinal and the re-built F3-FA SACS model for the fatigue analysis results of the welded joints considered in this study, i.e. substructure butt weld joints.

The stress concentration factor (SCF) used for each joint is also presented in the table s and is calculated based on the members connected wall thickness (t

1

and t 2), the fabrication tolerance (ft)

and the resultant eccentricity (e) through the Burdekin expression, as mentioned in section 3.4.3.

The respective study butt weld joints are illustrated in

Figure 3.30 and Figure 3.31 The leg joints are

order ascendant in regards to elevation, as shown in the tables and figures.

The substructure butt weld joints SCF is multiplied with a weld classification factor, according to the Offshore Technology Report 2001/015 [/9/]. The following weld classification factors are used: -

Double sided butt wel ds welded with an auto matic process, other than sub merged arc and welded in down hand position =

1.00;

-

Double sided butt welds

=

1.14;

-

Single sided butt welds

=

1.52.

The joint welded during mating, when the legs are linked with the bucket and the transition frame (knuckle joint) does not have the same quality as the other butt welds, therefore the SCF is multiplied by a welding factor of 1.14 (Table 3.34). Others yard welds mentioned in Table 3.35 are also multiplied by the appropriate weld classification factor of 1.14 or 1.52, respectively in the case of a double sided or a single sided butt weld is used [/3/].

In the srcinal F3-FA SACS model the fatigue analysis results with a contribution of 25% effective live load are reported and, for the most critical joints, a contribution of 50% effective live load is also considered. For that reason, not all fatigue analysis results are available (see tables with values: “n/a”, i.e. not available). Therefore, the fatigue analysis was re-done for the srcinal F3-FA SACS model for the most severe condition, that is with a 50% of effective live load contribution.

-87-


Finally, the differences between the srcinal and the re-built model in the fatigue analysi s results (based on the 50% effective live load) are presented in the left column of Table 3.34 and Table 3.35.

Figure 3.30: Leg joints in SACS model

Figure 3.31: Bucket and transition frame joints in SACS model


-88-

Table 3.34: Maximum fatigue lives [yrs.] comparison between the Original and the Re-built F3-FA SACS model – Substructure leg butt weld joints

Joint number

t1

Elevation

t2

(m)

ft

e SCF

Weld

SACS

classification

applied

(mm)

factor

Original F3-FA model

SCF

Reported in [/3/] FL FL 25% LL 50% LL


Re-analysis FL 50% LL

FL 50% LL

Ratio between both model results

Substructure leg xx09

-35.670

70

85

6.0

13.5

1.49

1

xx11

-31.870

70

70

6.0

6.0

1.26

1

xx12

-29.310

70

85

6 .0

13.5

1 .49

1

1.49

8145

xx13

-26.750

85

135

6.0

3 1.0

1 .73

1

1.73

2683

xx14

-24.325

135

1 35

6 .0

6.0

1.13

xx15

-21.900

135

135

6.0

6.0

1

1.13

1.14

1.49 1.30

1.30 1.48

2 0 7 7 81

1 9 5 31 9

1.06

36245

4 1 21 7

0.88

n/a

5480

6 5 51

0.84

n/a

1786

2 1 94

0.81

13820

1 7 04 2

0.81

>100000 (1)

n/a 52571

n/a

(1)

20832

n/a

(2)

732

468

468

484

0.97

xx20

-20.700

95

135

6.0

2 6.0

1 .61

1

1.61

234

149

148

154

0.96

xx21

-17.300

75

95

6 .0

16.0

1 .53

1

1.53

397

243

241

251

0.96

xx22

-13.300

55

75

6 .0

16.0

1 .67

1

1.67

355

213

208

218

0.96

xx23

-9.300

55

55

6 .0

1.33

1

1.33

4420

n/a

2662

2 7 81

0.96

xx24

-5.300

45

55

6 .0

11.0

1 .62

1

1.62

3024

n/a

2085

2 1 66

0.96

xx25

-1.300

45

55

6 .0

11.0

1 .62

1

1.62

9074

n/a

7875

8 0 19

0.98

xx26

2.700

45

55

6 .0

11.0

1 .62

1

1.62

14404

n/a

11396

1 1 95 5

0.95

xx27

6.700

45

55

6 .0

11.0

1 .62

1

1.62

5689

n/a

3873

4 2 76

0.91

xx28

10.700

45

45

6 .0

6 .0

1.40

1

1.40

2937

n/a

1838

1 9 52

0.94

xx29

14.700

45

55

6 .0

11.0

1 .62

1

1.62

401

247

243

258

0.94

6 .0

xx30

18.700

55

60

6 .0

8 .5

1.43

1

1.43

479

290

285

301

0.95

xxCD

20.305

60

60

6 .0

6 .0

1.30

1

1.30

688

413

405

429

0.94

xx40

22.700

55

60

6 .0

8 .5

1.43

1

1.43

681

409

401

425

0.94

xx41

26.700

45

55

6 .0

11.0

1 .62

1

1.62

1251

745

731

777

0.94

xx42

30.700

45

45

6 .0

6 .0

1.40

1

1.40

58191

n/a

28575

xx43

34.700

45

45

6.0

6.0

1.40

1

1.40

>100000

n/a

> 1 0 00 0 0

> 1 00 0 0 0

--

xxTD

36.305

45

80

6.0

23.5

1.93

1

1.93

>100000

n/a

> 1 0 00 0 0

> 1 00 0 0 0

--

3 3 97 2

FL – Fatigue Lives;

LL – Live Load (represented in percentage of effective live load considered in the analysis).

(1) Minimum SCF applicable is 1.30;

(2) Weld classification factor of 1.14 (SCF is multiplied by 1.14) due to the yard weld, that is the bottom leg + bucket connection with the remaining leg part.

n/a – Not available.

-89-


0.84

Table 3.35: Maximum fatigue lives [yrs.] comparison between the Original and the Re-built F3-FA SACS model – Substructure transition frame and bucket butt weld joints

Joint number

Elevation

t1

t2

ft

e SCF

(m)

Weld

SACS

classification

applied

(mm)

factor

Original F3-FA model Reported in [/3/] FL 25% LL

SCF

Re-analysis


FL 50% LL

FL 50% LL

FL 50% LL

Ratio between both model results

Horizontal and lower diagonals in transition frame xx61/62

-30.507

20

20

6.0

6.0

1.90

1.52

2.89

1294

n/a

10 5 0

88 2

1.19

xx53/54

-31.435

25

25

6.0

6.0

1.72

1.52

2.61

2986

n/a

26 2 2

26 5 1

0.99

xx55/56

-31.070

25

25

6.0

6.0

1.72

1.52

2.61

2349

n/a

15 6 3

21 3 6

0.73

xx06

-39.070

25

25

6.0

6.0

1.72

1.52

2.61

98257

n/a

8 89 0 3

9 97 7 4

0.89

xx08

-39.070

25

25

6 .0

6 .0

1.72

1.52

2.61

799

542

54 6

81 2

0.67

Columns on side of bucket xx51/52

-32.01

60

60

6.0

6.0

1.30

1.52

1.98

22574

n/a

1 44 1 7

1 00 5 1

1.43

xx59/60

-34.97

35

60

6.0

1 8.5

1 .98

1.52

3.01

1752

n/a

14 0 3

11 1 9

1.25

xx63/xx64

-26.445

50

85

6.0

2 3.5

1.88

1.14

44 8

51 8

0.86

Upper diagonal in transition frame

FL – Fatigue Lives; (1) Minimum SCF applicable is 1.30;

2.14

654

445

LL – Live Load (represented in percentage of effective live load considered in the analysis). (2) Weld classification factor of 1.14 or 1.52 (SCF is multiplied by 1.14 or 1.52) due to the yard weld.

n/a – Not available.


-90-

3.5.4

Conclusions

The main conclusion is that the re-built F3-FA SACS model is a good reference model of the F3-FA platform, based on the comparison results with the srcinal F3-FA SACS model.

As a result, the re-built model can be used in the study of alternative solutions for the F3-FA structure, presented in Chapter 4. The frequency modes, loads and fatigue analysis results show good approximation values between both models (srcinal and rebuild) and will be used as reference values in Chapter 4.

-91-



- 92 -

4

Study of possible structural improvements

4.1

General considerations

This chapter presents the investigations regarding possible structural improvements, presenting and discussing variant solutions for the srcinal F3-FA platform. Two key goals were initially proposed: 1) to improve the dynamic behaviour of the platform by increasing its natural frequencies for a better performance to fatigue loading; 2) to come up with structural alternative solutions to design features that faced some issues during the F3-FA construction / installation.

To attain the two goals a simplified model of the F3-FA platform was built and analysed with SACS software [/10/], to simulate the structural behaviour of the current structure, as described on Chapter 3. This model was called the “re-built F3-FA model” . Then, structural modifications were implemented and analysed individually. The comparison analysis is concentrated on the first mode shapes and natural period’s values, as well as on total structural weight of the variant solutions.

After consideration of the main variables and most sensitive construction parts, the following items have been considered to investigate its influence on the design: - Knuckle joint; - Different leg diameters; - Leg-topside connection.

In the follow paragraphs all items will be explained what the pros and cons are. All items combined will result in a set of possible designs (variant solutions). These designs will be checked against strength, stability and fatigue at the end of this chapter.

Finally the overall results will be summarized in a table.

4.2

4.2.1

Knuckle joint

Introduction

On the legs bottom a transition frame, known as knuckle joint, supports and distributes the leg loads to the eccentric suction buckets, as stated in Chapter 2 and illustrated in Figure 4.1.

- 93 -

Chapter 4 : Study of possible structural improvements

The knuckle joint was primarily required to avoid a deep water seaport during the platform assembly and transport, trough eccentric buckets. Plus it provided the following benefits to the platform: -

Increase the legs strength and stiffness, by

-

Improve the F3-F A structure dyna mic behaviour,

decreasing their unbraced length;

through stiffness and lateral support.

Figure 4.1: F3-FA knuckle joint (Iv-Oil & Gas courtesy)

During the F3-FA construction and installation, the knuckle joint proved not to be the desirable structural solution. The disadvantages of the knuckle joint were: -

Extensive steel weig ht, welding work and lift requirements, resulting in considerable construction time and cost;

-

Substantial expensive offshore work, for the current F3-FA leg-topside connection to be compatible with the knuckle joint;

-

Fatigue critical tubular weld joints in the chord and bra ce members of the knuckle joint, according to the fatigue analysis report [/3/].

4.2.2

Model without the knuckle joint and centric suction buckets

The F3-FA suction buckets are very large steel structures with a 15 meters diameter and 13 meters elevation height. Removing the knuckle joint requires the buckets to be centred with the legs, for the leg loads to be properly distributed to the bucket.

Centric buckets would greatly diminish the moments applied on the foundation, partly due to centric loads ( Figure 4.2 ). As a result, the bucket size might be reduced and become a more economical solution. Figure 4.2: Example of a centric bucket (Iv-Oil & Gas courtesy)

Also, no tubular joints are required without the knuckle joint, dropping the number of fatigue sensitive connections.

However, as mentioned before, the F3-FA platform was transported and installed using a large flat top barge BOA 35 with a 124 meters long, 31,5 meters wide and 8 meters depth. The space between the legs and buckets was carefully selected together with the cargo barge to avoid future conflicts.

For the buckets to be centred with the legs, and without changing the transportation, two structural alternatives are possible, as shown in Figure 4.3:


- 94 -

(A) Buckets underneath the barge and shifted position to be centric with the legs; (B) Buckets side by side with the barge (as before) and won’t change position, since the legs axis will shifted to be centred with the buckets.

Only the first design, with buckets underneath the barge, will be studied in the current report. The second design would have involved a significant modification on the topside, since the platform would have to be “stretched”, with an immediate increase of the structural steel weight.

The first design however, requires also some structural adjustments that will be introduced further.

B

A Figure 4.3: Solution with centric bucket side by side with the barge [left] and with centric bucket underneath the barge [right]

The bucket underneath the barge would require a seaport with at least 21 meters of water depth

18

.

The F3-FA platform was transported from the port of Vlissingen (Zeeland Seaports) which can handle vessels up to 16,5 meters draught. Actually, one of the reasons for the eccentric buckets was the intention to build it on Heerema Fabrication Group (HFG) Vlissingen yard or Hartlepool yard (near Middlesbrough in UK).

The regional principal seaports in the area, such as Amsterdam, Terneuzen (Zeeland Seaports) and the Belgian ports of Bruges-Zeebrugge, Antwerp and even France’s northern ports of Dunkirk and Le Havre have a limited draught capacity of 13 to 17,8 meters. The only possible nearby solution is the Port of Rotterdam capable of receiving vessels with a draft of 22,5 meters. Where the topside, legs

18

21 meters water dept h = 5 meters barge under water + 14 meters bucket + at

- 95 -

least 2 meters for gap and tidal changes.


and buckets can be built and separately transported by boat from nearby construction yards, such as Schiedam, Zwijndrecht or even Vlissingen. Building the platform in Norway, where the seaports have deeper drafts, is also a possibility, even if the transport distance of the platform would be longer, as shown in Figure 4.4 ; and as a result, the F3-FA platform would experience bigger transport fatigue damage and a larger weather window would be required.

F3-FA site location

Figure 4.4: F3-FA site location (adapted from [/23/])

Since the buckets are underneath the barge, during mating the buckets will have to be welded to the legs underwater using a welding process called the hyperbaric welding. Underwater conditions make it harder to assure the welds integrity and to detect defects in such a crucial connection. An alternative is to attach, on yard, a longer portion of the legs to the suction buckets, enough to allow a dry welded connection during mating, with the use of a scaffolding (Figure 4.6).

The resume of the substructure proposed changes are presented on Figure 4.5 and Figure 4.6.


- 96 -

Figure 4.5: Foundation plan view of the model with centric buckets underneath the barge

Figure 4.6: Side view during transportation of the model with centric buckets underneath the barge

In the SACS model, a rigid connection between the legs and buckets is no longer required, since the suction buckets are now centred with the legs axis. The truss frame, built to fix the buckets orientation, change configuration but continues to be connected to the leg through dummy rigid and weightless members, as shown in Figure 4.7.

- 97 -


Dummy members connecting truss frame and legs

Figure 4.7: SACS Model with centric buckets underneath the barge (SACS software)

4.2.3

Results

As mentioned before, two items are studied and compared with the re-built F3-FA model, for each variant solution, the structural steel weight and the first mode shapes.

Furthermore, in the model without the knuckle joint two solutions are compared: a) with the srcinal F3-FA leg wall thickness; b) with a modified leg wall thickness based on the results obtained in the fatigue analysis for the model with the final variant solution and the srcinal leg diameter (i.e. 3250 mm), that will be presented in a later stage (ref. to section 4.4).

The removal of the knuckle joint (alone) spares almost 1000 mT of structural steel (19,7% of the total 19

structural steel nett weight used on the F3-FA platform, excluding buckets ), as illustrate in Table 4.1.

Table 4.1: Differences in the structural steel nett weight between the re-built F3-FA model and the model without knuckle joint

Elements Knucklejoints(transitionframes) Legs wall thickness redesign (based on fatigue analysis results)

TOTAL

Nettweight(mT) -988 +568

-420

Note: The weights differences regarding the truss frame and possible influences on the bucket’s plates wall thickness are neglected, since the bucket was not a part of the Iv-Oil & Gas work scope and the truss frame has practically no influence in the structure behaviour.

The first modeshapes natural periods are presented in Table 4.2 and illustrated on Figure 4.8, Figure 4.9 and Figure 4.10. 19

Re-built F3-FA SACS model structure self-weight and non-modelled structural steel = 4654 + 411 = 5065 mT (section 3.3.2).


- 98 -

Table 4.2: Natural period’s: Model without the knuckle joint and with centric buckets underneath the barge [50% live load]

Mode shape

Naturalperiod

Mode shape characterization

Re-built F3-FA model without the knuckle joint and with centric buckets --

Original legs wall thickness

Structure 1) self-weight

3642mT

Modified legs wall thickness 4210mT


--

2)

4654mT

--

1

3,27s

2,79s

2,42 s

2

3,25s

2,77s

2,41 s

3

2,17s

1,90s

1,74 s

Torsion mode shape

4

1,43s

1,43s

1,43 s

Ventstack curve in X direction

5

1,41s

1,41 s

Ventstack curve in Y direction

1,41s

6

0,56 s

a)

7

0,55 s

b)

Sway mode shape: Single curvature in X direction Sway mode shape: Single curvature in Y direction

0,52 s

a)

0,61 s

Sway mode shape: Double curvature X direction

0,51 s

b)

0,58 s

Sway mode shape: Double curvature Y direction

1) Truss frame weight is included and secondary/tertiary non-modelled structural steel and bucket weight is excluded; 2) Modified leg wall thickness determined based in the fatigue analysis results presented in section 4.4. a) Non-sway mode shape: double curvature Y direction;

b) Double curvature in X and Y direction.

From Table 4.2, two remarks can be made: a) The F3-FA platform proved to be very sensitive to variations in the legs wall thickness, when it comes to the platform weight and the first modeshapes natural periods. However, the improvement was not enough to match the F3-FA re-built model natural periods; b) The differences between models are more noticed on the first three mode shapes, with a higher impact in the dynamic response of the platform to wave loading.

- 99 -


Figure 4.8: Model without the knuckle joint and with centric buckets st

1 mode shape

Figure 4.9: Model without the knuckle joint and with centric nd

buckets 2 mode shape

Chapter4 : Study of possible structuralimprovements

Figure 4.10: Model without the knuckle joint and with rd

centric buckets 3 mode shape

- 100 -

4.3

Model with different leg diameters

In the self-installing platform the fatigue loads are the principal action affecting the design. Thus, our first aim is to lower the natural periods of the mode shapes on the new design by increasing the leg diameter, in order to improve the fatigue behaviour of the structure.

4.3.1

Leg diameter increased based on the slenderness ratio

The leg unbraced length with the knuckle joint is 43,1 meters and the buckling length is given by:

The leg section has a 3250 mm diameter and an average thickness between 45 to 55 mm (radius of gyration approx.

), and therefore the slenderness ratio

is equal to 76,3.

Without the knuckle joint the leg unbraced length increased to 59,6 meters and the buckling length to 119,2 meters. Thus, the slenderness ratio will rise to 105,5. To preserve the initial slenderness ratio the leg diameter has to increase up to 4475 mm. Based on this simple analysis, three models without the knuckle joint and with centric buckets underneath the barge are studied using a leg diameter of 4000, 4500 and 5000 mm. With the leg diameter increase, the sleeves will have also to change size. In the F3-FA platform a gap of 75 mm between the internal wall of the sleeve and the external wall of the leg, have been used. For this analysis the gap of 75 mm was preserved, so the inner diameter of the sleeve will be 150 mm bigger than the external diameter of the legs.

4.3.2

Re sults

Table 4.3 presents a comparison of the 1

st

mode shapes natural periods of the model without the knuckle

joint, with centric buckets underneath the barge, with different leg diameters and with two alternative wall thickness solutions; one with the original F3-FA leg wall thickness and other with a modified leg wall thickness, that was determined based in the fatigue analysis results presented in section 4.4.

- 101 -


Table 4.3: Natural frequencies: Model without the knuckle joint and with centric buckets underneath the barge and with the legs diameter increased to 4000, 4500 and 5000 mm vs. Re-built F3-FA Model [50% live load]

Natural period

Structure

Models

self-weight


1)

Mode shape 1 2,42s

4654 mT

Mode shape 2 2,41s

Mode shape 3 1,74s

Mode shape 6 0,61s

Mode shape 7 0,58s

Models without knuckle joint and with centric buckets underneath the barge Leg diameter 3250 mm

Leg diameter 4000 mm Leg diameter 4500 mm Leg diameter 5000 mm

Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick.

3642 mT

3,27s

3,25s

2,17s

0,56s

0,55s

4210 mT

2,79s

2,77s

1,90s

0,52s

0,51s

4057 mT

2,66s

2,64s

1,78s

0,54s

0,53s

4436 mT

2,39s

2,38s

1,63s

0,53s

0,52s

4333 mT

2,41s

2,40s

1,63s

0,55s

0,53s

4533 mT

2,26s

2,25s

1,54s

0,54s

0,52s

4610 mT

2,24s

2,24s

1,53s

0,55s

0,54s

4784 mT

2,12s

2,12s

1,47s

0,55s

0,54s

Note: Mode shapes 4 and 5 are skipped from the table because they are relative to the buckling of the ventstack and are of no interest for our study. 1) Truss frame weight is included and secondary/tertiary non-modelled structural steel and bucket weight is excluded; 2) Modified leg wall thickness determined based in the fatigue analysis results presented in section 4.4.

st

With the increase of the leg diameter, the 1

three natural periods decrease substantially, to values lower

than the re-built model. Therefore, these models have a variant solution expected to demonstrate a better dynamic behaviour than the srcinal design, and thus obtain improvements on the fatigue analysis results. The variant solutions demonstrate smaller structural steel weight when compared to the re-built model (except for the model with the leg diameter of 5000 mm), as presented in Table 4.4. Table 4.4: Structural main steel weight savings: Models without the knuckle joint and with centric buckets underneath the barge and with legs diameter increased to 4000, 4500 and 5000 mm vs. re-built F3-FA model

Structural steel nett weight differences from the re-built F3-FA model (mT)

1)

Models TOTAL Leg diameter D = 3250 mm

-420 (-8,3%)

Leg diameter D = 4000 mm

-193 (-3,8%)

Knuckle joints

-97 (-1,9%)


+155 (+3,1%)

Legs wall thick.

Sleeves

-988

2)

- 988

+ 383

+380

+ 32

- 988

+ 638

+200

+ 53

- 988

+ 894

+175

+ 74

2)


Legs diameter

2)

2)

0

+568

0

1) The weights differences regarding the truss frame and possible influences on the bucket’s plates wall thickness are neglected, since the bucket was not a part of the Iv-Oil & Gas work scope and the truss frame has practically no influence in the structure behaviour; 2) % of the re-built F3-FA SACS model structure self-weight and non-modelled structural steel (excl. buckets) = 5065 mT (section 3.3.2).

Though, the increase of the leg diameter and consequently of the sleeve section, some structural incompatibilities were exposed on the topside, that can be seen on Figure 4.11. In fact, due to the resize of


- 102 -

the sleeve diameter, the braces are no longer compatible with the leg-topside design. However, the structure was not modified since a complete new leg-topside connection design is analysed in section 4.4.

Figure 4.11: Structural incompatibility on the leg-topside connection with a leg diameter increase (SACS software)

4.4

Leg-topside connection

The topside connection with the legs is made at two levels, the cellar deck and the top deck. On the top, the legs are bolted to the topside, transferring vertical loads and bending moments. A sleeve guides the legs between the top and the main deck, inside which a bearing support allows horizontal loads to be transmitted. The sleeves are linked to the topside through vertical plates in four different directions, as shown in

Figure

4.12, the plates are connected to “secondary” columns that will transmit the loads to the main trusses on the topside by a bracing system.

Figure 4.12: Top deck and cellar deck– Leg connections (Iv-Oil & Gas courtesy)

On the lowest deck, between the main and the cellar deck, space was left available for the knuckle joint during construction and transportation, as shown in Figure 4.13.

- 103 -


After the buckets are sucked into the seabed and the topside reached is final height by using the strand jacks, clamps were installed on the cellar deck to absorb the wave impact loads. These clamps, prevent lateral movement of the legs and, as a consequence, the instability of the entire structure.

Considerable offshore hours were required to assemble the clamp connection, due to the dimensions of the elements and the conditions available, turning out to be an expensive activity.

In fact, offshore construction is very expensive since it relies on specialized equipment, workers and safety parameters; and becomes unpredictable due to the vulnerability of the weather conditio ns on Figure 4.13: Free space on the

open sea.

F3-FA topside (Iv-Oil & Gas courtesy)

Thus, it is reasonable to propose the variant solution for this connection.

4.4.1

Model with adapted leg-topside connection

In this next study the starting point is the model with the srcinal leg diameter of 3250 mm and with centric buckets underneath the barge (and, of course, no knuckle joint), as mentioned before. From this model a new-leg topside connection is made.

The new leg-topside connection will connect the topside main trusses directly with the legs, through the sleeves, and the sleeves will expand to the lowest deck (cellar deck).

Figure 4.14 presents an overview of the structural modifications that have taken place on the topside model.

Figure 4.14: Re-built F3-FA SACS model with former and new leg-topside connection [sleeves highlighted in green]


- 104 -

Figure 4.15 presents a general plan view of the topside modifications. The center to center (ctc) of the legs stayed the same and the deck extended to the center of the legs.

The longitudinal row trusses 1 and 3 shifted almost 3 meters to the edge of the deck area, including the ventstack and the crane boom pedestal; allowing a straight connection with the sleeves, and are now called rows 1’ and 3’. Also, the row trusses A1, A, E and F, located on the corners of the legs, are replaced by centred ones joining the two legs and named A’ and E’.

The F3-FA strand jacks, meant to lower the legs and jack-up the topside during installation, are supported in the exterior secondary column in the leg-topside connection (see Figure 4.13 ). In the new leg-topside connection the removal of this support, as shown in

Figure 4.15, could be solved by attaching the strand

jacks in the sleeves, a solution tested before in the first SIP concepts [/21/].

The connection of the topside with the legs is still made at two levels: the cellar deck and the top deck.

Figure 4.15: General plan view of the topside design modifications for the models with centric buckets underneath the barge

- 105 -


In the SACS model, the connection arrangement is made by dummy members (as before), see Figure 4.16. The dummy members are weightless, with 200 mm length and they link the sleeves to the legs. The sleeves are connected to the topside through the topside main trusses (Figure 4.14 and Figure 4.15).

Figure 4.16: Plot of the new leg topside connection (e.g. leg E1)


- 106 -

The natural mode shapes and periods are then determined in the new model and compared with the previous models with the former leg-topside connection, as presented in Table 4.5.

Table 4.5: Natural frequencies: Model without the knuckle joint and with centric buckets underneath the barge and with leg-topside connection adapted vs. Re-built F3-FA Model [50% live load]

Modeshape

Naturalperiod

Modeshapecharacteriz.

Re-built F3-FA model without the knuckle joint and with centric buckets underneath the barge

New leg-topside conn.

--

Original legs wall thick. Structure 1) self-weight

Re-built F3-FA

Former leg-topside conn.

Modified legs 2)

wall thick.

Original legs wall thick.

2)

3131mT

3699 mT

4210mT

4654mT

1

3 ,1 6 s

2 ,7 1 s

3,27s

2,79s

2,42s

2

3 ,0 9 s

2 ,6 4 s

3,25s

2,77s

2,41s

Sway mode shape: Single curvature in X direction Sway mode shape: Single curvature in Y direction

3

2 ,3 6 s

2 ,0 7 s

2,17s

1,90s

1,74s

Torsion mode shape

4

1 ,4 4 s

1 ,4 4 s

1,43s

1,43s

1,43s

Ventstack curve in X direction

5

1 ,4 2 s

1 ,4 2 s

1,41s

1,41s

1,41s

6

0,56 s

7

0,55 s

a) b)

3642 mT

wall thick.

--

model

Modified legs

a)

0,56 s

a)

0,52 s

a)

0,61 s

0,51 s b)

0,55 s

b)

0,51 s

b)

0,58 s

0,52 s

Ventstack curve in Y direction Sway mode shape: Double curvature X direction Sway mode shape: Double curvature Y direction

1) Truss frame weight is included and secondary/tertiary non-modelled structural steel and bucket weight is excluded; 2) Modified leg wall thickness determined based in the fatigue analysis results presented in section 4.4. a) Non-sway mode shape: double curvature Y direction;

b) Double curvature in X and Y direction.

With the new leg-topside connection there is a minor improvement in the natural periods of the two first rd

mode shapes. On the other hand, the results show a “weaker” torsion mode shape (3 mode shape), where the replacement of the four transversal trusses located on the legs corners (rows A1, A, E and F) by two centred trusses with the legs (rows A’ and E’) shows to be a “weaker” solution for torsion.

Nevertheless, the main progress is the ability for a more integrated and less built expensive connection between the topside and the substructure that can be confirmed in the SACS model, that is approximately 511 mT lighter (10,1% of the total structural steel nett weight used on the F3-FA platform, excluding 20

buckets ).

4.4.2

Model adapted for transportation on barge

During sea transportation, a framing structure, known as grillage, is installed on top of the barge. The grillage is intended to support the platform and distribute the vertical loads to the barge.

20


- 107 -


The height of the grillage (6,5 meters) was considerably high to reduce the free space necessary on the topside to make it compatible with the knuckle joint and, most of all, to accomplish a better load distribution of the platform weight to the barge stressed skin structure. Because the legs are located outside the barge, in order for the grillage to support the platform, extra columns must be added, as presented in

Figure 4.17,

Figure 4.18 and Figure 4.19.

Extra column for grillage support

Figure 4.17: A view of part of the F3-FA grillage during load-out operations (Iv-Oil & Gas courtesy)

The extra column is preferably located on the edge of the barge, similar to the srcinal design where the barge is better reinforced, and connected to the sleeves through plates that will transfer the loads from the topside to the legs and vice-versa.

Figure 4.18: Platform on barge during sea transportation


- 108 -

Figure 4.19: Comparison between the previous SACS model and the one adapted with extra columns for grillage support

A possible solution for the connection between the sleeves and the extra columns is given in Figure 4.20. In SACS the connection is modelled through plates between the two tubular members axis.

Figure 4.20: Possible leg/sleeve – extra column connection

Figure 4.21: General plan view of the topside design modifications with the extra columns

- 109 -


Figure 4.21 represents the deck plan structural modifications for the added extra columns. As before, the mode shapes and natural periods of the structure are calculated and compared to the previous models, as presented in Table 4.6.

Table 4.6: Natural frequencies: Models without the knuckle joint and with centric buckets underneath the barge and with topside modifications vs. Re-built F3-FA Model

Natural period

Structure

Models

self-weight


4654 mT

1)

Mode shape 1 2,42 s

Mode shape 2 2,41 s

Mode shape 3 1,74 s

Mode shape 6 0,61 s

Mode shape 7

0,58 s

Models without knuckle joint & with centric buckets underneath the barge Former leg-topside connection

e id w s e p tN o g e l

n io t c e n n o c

No extra columns

Extra columns on row 1’ and row 3’

Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick.

3642 mT

3,27 s

3,25 s

2,17 s

0,56 s

0,55 s

4210 mT

2,79 s

2,77 s

1,90 s

0,52 s

0,51 s

3131 mT

3,16 s

3,09 s

2,36 s

0,56 s

0,55 s

3699 mT

2,71 s

2,64 s

2,07 s

0,52 s

0,51 s

3336 mT

3 ,2 3 s

3 ,1 3 s

2 ,4 2 s

0,55 s

0 ,5 4 s

3904 mT

2 ,7 6 s

2 ,6 7 s

2 ,1 1 s

0,52 s

0 ,5 1 s


The model with extra columns on the longitudinal rows 1’ and 3’ (necessary to not compromise the current F3-FA transport) have minor influence in the structure natural period’s, as shown in Table 4.6.

As for the structural weight, adding the columns on rows 1’ and 3’ will increase the weight in 205 mT (4,0% 21

of the total structural steel nett weight used on the F3-FA platform, excluding buckets ).

4.4.3

Models with different leg diameter

Like before, based on the structural adjustments made, three additional different models are studied using a leg diameter of 4000, 4500 and 5000 mm.

21



- 110 -

4.4.3.1

Incompatibilities with barge BOA 35

The cargo barge used for the installation and transportation of the F3-FA platform is named BOA 35 and has the following dimensions: 124 x 31,50 x 7,93 m. The topside is arranged on top of the flat barge and the legs and suction buckets are attached to the structure.

In the N-S (longitudinal) direction the legs are spaced from each other 36,43 m. With the srcinal size of the legs diameter, 3250 mm, the free space available between leg walls is 33,18 m. Increasing the legs diameter to 4000, 4500 and 5000 mm will reduce the walls distance and, thus, the gap between the sleeve and the barge will also diminish.

In order to maintain the platform compatibilities with the barge BOA 35, the legs are shifted away, depending on their diameter, to maintain the srcinal distance between the sleeve and the barge, accord ing to the values presented in Table 4.7 and exemplified in Figure 4.22.

Table 4.7: Model legs shifted values to be compatible with barge BOA 35

Legs diameter (mm) Models 3250 Legsshiftedvalue

0mm

4000 375mm

4500 625mm

5000 875mm

Figure 4.22: Leg/sleeve shifted in N-S direction for the different leg diameter models (e.g. 3250 and 5000 mm)

- 111 -


In the end the influences on the structure frequency modes are rather small and only the results of the models with the legs shifted are presented.

4.4.3.2

Results

The natural periods of the first mode shapes are compared with the re-built F3-FA model and the models without any topside changes, in Table 4.8.

The first half of the table indicates the results for the models without the knuckle joint and with centric buckets underneath the barge. The second half of the table presents the results of the same models but with the new leg-topside connection (incl. the extra columns in rows 1’ and 3’).

Table 4.8: Natural frequencies: Model without the knuckle joint and with centric buckets underneath the barge, former and new legtopside and with legs diameter up to 4000, 4500 and 5000 mm vs. Re-built F3-FA Model

Natural period

Structure

Models

self-weight


4654 mT

1)

Mode shape 1 2,42s

Mode shape 2 2,41s

Mode shape 3 1,74s

Mode shape 6 0,61s

Mode shape 7

0,58s

Models without knuckle joint & with centric buckets underneath the barge (ref. to section 4.3)

n o ti c e n n o c e d i s p o -t g e l r e m r o F

n o ti c e n n o c e d i s p to g le w e N

Legs diameter 3250 mm ) .3 4 n io t c e s o t f. e r (

Legs diameter 4000 mm Legs diameter 4500 mm Legs diameter 5000 mm

’ 3 d n a ’ 1 s w o r n o s n m u l o c a tr x e d n a

Legs diameter 3250 mm

Legs diameter 4000 mm Legs diameter 4500 mm Legs diameter 5000 mm

Original legs wall thick. Modified legs 2)

wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick. Original legs wall thick. Modified legs 2) wall thick.

3642 mT

3,27s

3,25s

2,17s

0,56s

0,55s

4210 mT

2,79s

2,77s

1,90s

0,52s

0,51s

4057 mT

2,66s

2,64s

1,78s

0,54s

0,53s

4436 mT

2,39s

2,38s

1,63s

0,53s

0,52s

4333 mT

2,41s

2,40s

1,63s

0,55s

0,53s

4533 mT

2,26s

2,25s

1,54s

0,54s

0,52s

4610 mT

2,24s

2,24s

1,53s

0,55s

0,54s

4784 mT

2,12s

2,12s

1,47s

0,55s

0,54s

3336 mT

3,23s

3,13s

2,42s

0,55s

0,54s

2,67s

2,11s

0,52s

0,51s

3904 mT

2,76s

3797 mT

2 , 65 s

2 , 55 s

2,05 s

0,54 s

0,53 s

4177 mT

2 , 40 s

2 , 30 s

1,86 s

0,52 s

0,52 s

4104 mT

2 , 43 s

2 , 32 s

1,91 s

0,54 s

0,53 s

4303 mT

2 , 28 s

2 , 17 s

1,79 s

0,53 s

0,53 s

4411 mT

2 , 28 s

2 , 17 s

1,82 s

0,55 s

0,54 s

4585 mT

2 , 16 s

2 , 05 s

1,72 s

0,55 s

0,54 s



- 112 -

The models analysed revealed a successful decrease on the natural periods with the increase of the legs diameter and with the selected wall thickness for the fatigue calculations, as we have seen before in section 4.3.

Matching the results presented from the models with the former and new leg-topside connection referred, it is possible to observe smaller deviations in the first two mode shapes natural periods, around 0,10 seconds. Thus, the improvements for the leg-topside connection on the SACS models structure dynamic behaviour are not significant.

The structural steel weight differences can be seen in Table 4.9.

Table 4.9: Structural main steel weight savings: Model without the knuckle joint and with centric buckets underneath the barge, adapted leg-topside connection and with legs diameter up to 4000, 4500 and 5000 mm vs. Re-built F3-FA Model

Structural steel nett weight differences from the re-built F3-FA model (mT) Models

Knuckle joints

TOTAL


- 726 (-14,3%)


- 453 (-8,9%)


- 97 (-1,9%)


- 44 (-0,9%)

2)

Legs diameter

Legs wall thickness (ref. section 4.4.5)

Sleeves

-988

0

+568

+154

-988

+383

+380

+220

-448

2)

-988

+638

+200

+263

-440

2)

-988

+894

+175

+307

-431

2)

1)

New leg-topside connection Topside

-460

1) The weights differences regarding the truss frame and possible influences on the bucket’s plates wall thickness are neglected, since the bucket was not a part of the Iv-Oil & Gas work scope and the truss frame has practically no influence in the structure behaviour; 2) % of the re-built F3-FA SACS model structure self-weight and non-modelled structural steel (excl. buckets) = 5065 mT (section 3.3.2).

The new models had a considerable reduction on the structural steel weight with the removal of several columns and braces on the topside, despite the weight increase with the sleeves length and the extra columns on the longitudinal rows 1’ and 3’.

With this last structure modif ication all models, with different leg diameters, have a total structural steel weight lower than the re-built F3-FA model.

Based on the variant solutions proposed for the F3-FA platform, an in-place and fatigue analysis will take place to evaluate if the models can withstand the F3-FA operational and survival conditions.

Further on, only the models without the knuckle joint with centric buckets underneath the barge (ref. section 4.2), with different leg diameters (ref. section 4.3), with the new leg-topside connection and extra columns on rows 1’ and 3’ (ref. section

4.4), and with the modified leg wall thickness defined based on the fatigue

analysis results (see section 4.4.5), are considered.

- 113 -


4.4.4

In-place analysis

An in-place analysis, as described in section 3.4.2, is performed to check if the F3-FA variant solutions can withstand the operational and survival loads during its entire service life. All members considered to be main steel and modelled in SACS are checked for resistance and stability (buckling) resistance.

The final models achieved, from section 4.4, are checked for the in-place analysis. That is, the models without the knuckle joint, with centric buckets underneath the barge, with a modified leg-topside connection, including the extra columns on rows 1’ and 3’ for grillage support during transport, and with a leg diameter of 3250, 4000, 4500 and 5000 mm.

The methods used on the design and analysis of the steel elements are the same used for the F3-FA structure and are presented on the American codes, API [/4/] & AISC [/7/], as mentioned in section 3.4.2. In appendix D a spreadsheet illustrates an example of the buckling and stress check for cylindrical members according to API [/4/].

4.4.4.1

Topside members

For the different leg diameter models, in the in-place analysis, some beam members sections were adapted for the new structural design reality. Although great structural changes were made, the adjustments on the topside were concentrated on the corners, i.e. the leg-topside connection.

In Figure 4.23, the maximum UC ratios for the topside members are presented (reference is made to the model with the srcinal legs diameter, 3250 mm).

Figure 4.23: SACS model topside members maximum UC’s colour represented (model without the knuckle joint, with centric buckets underneath the barge and with a modified leg-topside joint, plus the extra columns on rows 1’ and 3’ for grillage support during transportation and finally with legs diameter srcinal size 3250 mm)


- 114 -

The critical members with maximum Unity Check ratio (UC) reported, per location, are presented in

Table

4.10 and a comparison of the critical members UC’s for all models is shown in Table 4.11.

Table 4.10: Maximum topside members UC’s – model with the srcinal size of leg diameter (3250 mm)

Location

Joint numbers

1)

Critical condition

Member type

Load 2) combination

Maximum UC

Topside beams Cellardeck(Z=20,505)

156-168

Main deck (Z=28,705)

HE1000A

208-205

T+B

HE1000A

0,82

T+B

A001

1,09

3)

A048

3)

Top deck (Z=36,505) Helideck(Z=40,255)

327-333 427-403

HE1000A HE500A

C+B T+B

C008

Row1’(X=117,875)

120-272

Ø660x30

C+B

0,54

A047

Row3’(X=97,125)

233-327

Ø660x20

C+B

0,64

A043

1,12 0,58

A007

Topside diagonals

RowA’(Y=97,025)

CA3L-208

Ø710x25

C+B

0,67

C005

RowC(Y=115,240)

168-200

Ø710x25

C+B

0,43

A001

RowE’(Y=133,455)

CE3L-203

Ø710x25

C+B

0,61

C045

0,69

A002

Topside columns Verticalcolumn

203-303

1) Locations according to Chapter 2;

Ø610x20

S

2) Load cases according to section

3.4.4.2.

3) Some H-beam members require local reinforcement with diamond plates or brackets.

Label: C+B T+B

 

= =

S

Combined load compression force and bending moment; Combined load tension force and bending moment; Shearforce.

=

The values differ with a higher leg diameter, due to an increase of the wave forces.

Table 4.11: Maximum topside members UC’s – comparison between models with different leg diameter

Location

Joint numbers

1)

Maximum UC

Member type

3250 (mm)

4000 (mm)

4500 (mm)

5000 (mm)

Topside beams Cellardeck(Z=20,505)

156-168

HE1000A

Main deck (Z=28,705)

208 – 205

HE1000A

Top deck (Z=36,505)

327 - 333

Helideck(Z=40,255)

427-403

0,82

HE1000A HE500A

0,83

1,09

2)

1,12

2)

0,58

0,84

1,09

2)

1,12

2)

0,59

0,85

1,09

2)

1,04

2)

1,12

2)

1,12

2)

0,60

0,60

Topside diagonals Row1’(X=117,875)

120-272

Row3’ (X=97,125)

233-327

RowA’

CA3L-208

RowC(Y=115,240)

168-200

RowE’

CE3L-203

Ø660x30 Ø660x20 Ø710x25 Ø710x25 Ø710x25

0,54 0,64 0,67 0,43 0,61

0,53 0,63 0,68 0,44 0,64

0,55 0,66 0,71 0,45 0,67

0,61 0,71 0,73 0,45 0,70

Topside columns Verticalcolumn

203-303

1) Locations according to Chapter 2;

- 115 -

Ø610x20

0,69

0,72

0,74

0,76

2) Some H-beam members require local reinforcement with diamond plates or brackets.


The differences are more visible in the topside diagonals and vertical columns part of the main trusses in the topside, that are meant to withstand horizontal loadings.

Due to the re-built F3-FA model topside loads simplification, i.e. the loads are uniformly distributed in the deck areas, with no respect to their correct location, some topside beams are more stressed than before and, as a result, have UC higher than 1, as shown in

Table 4.10 and Table 4.11. In those cases the higher

stresses are located at the beam’s end and require a local reinforcement.

4.4.4.2

Leg members

In the models studied the expansion of the unbraced length of the legs, due to the removal of the knuckle joint, not only increased the structure natural periods but also decreased the buckling resistance of the legs (with a higher buckling length). As a result, some parts of the leg section were undersized. Also, some parts of the leg section were oversized, e.g. the former location of the connection of the knuckle joint, no longer required a thicker section.

Figure 4.24 illustrates the maximum UC ratios for the leg members of the model with the srcinal size of leg diameter (3250 mm) but with the modified leg wall thickness for the fatigue calculations.

Leg E3 Leg E1 Leg A3 Leg A1

Figure 4.24:SACS model leg members maximum UC’s colour represented (model without the knuckle joint, with centric buckets underneath the barge and with a modified leg-topside joint, inclusive the extra columns on rows 1’ and 3’ for grillage support during transportation, and finally with legs diameter srcinal size 3250 mm)


- 116 -

The substructure leg A3 shows the higher stresses values and the most unfavourable load combination is C045. Figure 4.25 presents the critical load diagrams for the leg A3 when submitted to the load combination C045 of the models checked for the in-place analysis.

Figure 4.25: SACS model leg A3 principal loads for the load condition C045

From Figure 4.25 one can conc lude that, the large r the legs section diameter the bigg er will be the hydrodynamic loads. Even, considering that the structure excitation from the wave motion for the models with smaller legs diameter is greater and, as a consequence, the quasi-static wave loads will have a higher dynamic amplification factor (DAF).

However, in Figure 4.25 the bending stresses are also shown to demonstrate that, the increase of the legs section resistance is bigger than the increase of the hydrodynamic loads in the models with a bigger leg diameter, since the bending stresses are lower.

The critical condition for all leg members is the combined stress, axial compression and bending.

- 117 -


Table 4.12 presents the maximum unity checks (UC) for leg A3 of the model with srcinal of the leg diameter.

Table 4.12: Maximum leg A3 members maximum UC’s – model with the srcinal size of leg diameter (3250 mm)

Location Z (m) +34,700 to +36,305

Joint numbers

Former member section

A343 – A3TD

+30,700to+34,700

A342–A343

+26,700to+30,700

A341–A342

New member section

Maximum UC

Ø3250x45 Ø 3250 x 45

0,26

Ø3250x55

0,26

Ø3250x70

0,31

C045

A340 – A341

0,34

+20,305 to +22,700

A3CD – A340

0,39

+14,700to+18,700 +10,700 to +14,700 +6,700to+10,700 +2,700 to +6,700

A330–A3CD A329–A330 A328 – A329 A327–A328

A325–A326

-5,300to-1,300

A324–A325

-13,300to-9,300

Ø 3250 x 85

Ø3250x55 Ø 3250 x 45

0,51

Ø3250x55

A323 – A324 A322–A323

Ø 3250 x 55*

0,84

Ø 3250 x 55

Ø3250x70

0,45

Ø 3250x95

0,40

Ø 3250x95

Ø 3250x 135

0,33

A319 – A320

-26,750to-24,325

A313–A314

-29,310to-26,750 -31,870 to -29,310

A312–A313

A310–A311

-35,670to-33,100

A309–A310

-39,070to-35,670

A308–A309

A006 C045 C045 C045

0,31

Ø 3250 x 135

Ø3250x85

Ø 3250 x 150

C045

0,32

C045

0,33 0,36

C045 C045

0,39

C045

0,43

A311 – A312

-33,100to-31,870

A005 0,46

Ø 3250x75

A3KN–A315 A314-A3KN

C005 C006

0,70 Ø3250x55

A320– A321

-22,600to-21,900 -24,325to-22,600

C045 C045

0,96 Ø 3250 x 55*

A321– A322

A315–A319

C045

0,78

-20,700to-17,300

-21,900to-21,000

C045

0,62

-17,300to-13,300

-21,000 to -20,700

C045

C045

0,51 Ø3250x70

A326 – A327

-1,300to+2,700

-9,300 to -5,300

Ø 3250 x 60

C045 C045

+22,700 to +26,700

+18,700to+20,305

Load case

C045 0,47

Ø 3250 x 70

0,49 0,53

Ø3250x85

0,58

C045 C045 C045 C045

* - Marine growth (section wall thickness is reduced in 5 mm, thus in the calculations the wall thickness is 45 mm).


- 118 -

Table 4.13 presents a comparison of the maximum UC’s for the leg A3 for the different models studied.

Table 4.13: Maximum leg A3 members maximum UC’s – comparison between models with different leg diameter

Location Z (m)

Joint numbers

+34,700 to +36,305

A343 – A3TD

Wall thick.

+30,700to+34,700

A342–A343

+26,700to+30,700

A341–A342

Alternative solution models

F3-FA platform

45

D = 3250 mm Wall thick.

UC

45

0,26

55

D = 4000 mm Wall thick.

70

0,31

55

+22,700 to +26,700

A340 – A341

0,34

+20,305 to +22,700

A3CD – A340

0,39

+18,700to+20,305 +14,700 to +18,700 +10,700 to +14,700 +6,700to+10,700 +2,700 to +6,700

A329 – A330

A324–A325

A322–A323

-17,300to-13,300

A 321–A322

-21,000 to -20,700

A315–A319 A3KN–A315

-24,325to-22,600

A314-A3KN

-26,750to-24,325 -29,310 to -26,750 -31,870 to -29,310

55 75 95

0,45

95

0,40

135

0,32 135

A309–A310

-39,070to-35,670

A308–A309

0,37 95

85

0,49

55

0,43 0,52

0,46 0,42 55* 0,37

0,27

0,25 0,24 55 0,31

70

0,35

0,37

0,30 70

0,31

0,32 0,28

95 0,31

0,30

0,24

95 0,26

0,24 0,30

0,33 1500,34

0,38

60 0,41

0,35 0,23

0,47 70

0,37

0,51

0,37

0,43

55

45 0,49

0,55

0,32

0,27

0,39

150

85

0,48 0,45

55 0,29

135 0,25

0,36

0,41

0,39

0,24

0,33

A311 – A312

-35,670to-33,100

70

0,19

0,35 0,36

0,30

55 0,37

0,33

0,30

55

UC

0,25 45

55* 0,45

0,43 0,46

70

45 0,26

0,59

0,31

A312 – A313

A310–A311

45

55* 0,51

0,70

A313–A314

-33,100to-31,870

0,53

Wall thick.

0,21

0,44

0,50

55* 0,84

55

UC

60

0,55 55

0,78

A319 – A320

-22,600to-21,900

0,44

0,96

A320 – A321

-21,900to-21,000

60

0,62

22

55*

A323 – A324

-13,300to-9,300

55

0,51

A326 – A327

-5,300to-1,300

0,32

D = 5000 mm

0,38

0,51

70

Wall thick.

0,22

0,51 45

A327–A328

A325–A326

-20,700 to -17,300

85

55

A328 – A329

-1,300to+2,700

-9,300 to -5,300

60

A330–A3CD

UC

45 0,25

0,26

D = 4500 mm

0,20

0,26 135 0,30

0,29

0,23 135 0,26

0,53

0,37

0,33

0,29

0,58

0,41

0,37

0,32

0,25

* - Marine growth (section wall thickness is reduced in 5 mm, thus in the calculations the wall thickness is 45 mm).

The minimum wall thickness for the leg members used in the F3-F A platform is 45 mm and has been adopted for all the models studied. Although the bending moments and axial forces are higher the UC’s are still lower for the models with bigger leg diameter.

Since the fatigue analysis is the critical condition in the F3-FA substructure design as well as for the variant solutions studied, the legs sections in all models are oversized for the in-place analysis.

- 119 -


4.4.5

Fatigue analysis

A dynamic spectral fatigue analysis, as described in section 3.4.3, is performed and the damage (or fatigue life) of the legs butt welds is calculated to check if the structure new designs can fulfil the F3-FA fatigue requirements.

The final models achieved, from section 4.4, are checked for the fatigue analysis. That is, the models without the knuckle joint, with centric buckets underneath the barge, with a modified leg-topside connection, including the extra columns on rows 1’ and 3’ for grillage support during transport, and with a leg diameter of 3250, 4000, 4500 and 5000 mm.

The F3-FA platform structure has a design lifetime of 20 years. The substructure butt welds have a design safety factor equal to 5, as referred in section 3.4.3. Thus, the expected butt welds fatigue life must be equal or higher than 100 years.

Table 4.14 contains the SCF calculation and the critical fatigue life, per butt weld joint elevation, for the F3FA re-built model and the variant final solutions introduced in section 4.4.

As previously referred, the leg wall thickness was modified in order for the leg butt weld joints to have a calculated fatigue life of at least 100 years. The maximum wall thickness adopted was 150 mm.

For the alternative model with the srcinal leg diameter of 3250 mm, in two joints the minimum fatigue life could not be achieved, even with a considerable increase of the leg wall thickness. On the other hand, the models with an increase of leg diameter size were able to satisfy the fatigue requirements in all joints.

Looking back to the results of Table 4.8, the first three mode shape natural periods of the alternative model with the srcinal leg diameter are considerable higher than the re-built F3-FA model. While, for the other models a better dynamic behaviour of the structure was obtained.

Thus, the models leg wall thickness was optimized based on the fatigue analysis results to reduce the structural main steel weight, as shown in Table 4.14.


- 120 -

Table 4.14: Maximum fatigue lives [yrs.] – Substructure butt weld joints for 50% Live Load (LL) [in 8 wave directions] Models without the knuckle joint and centric buckets underneath the barge, plus new leg-topside connection solution


Elevation

Leg diameter 3250 mm

Joint (m)

t1

t2

e

xx09

-35.670

70

85

xx11

-31.870

70

70

6.0

xx12

-29.310

70

85

13.5

FL

SCF

(mm) 13.5

(1)

41217 150 150 6551

1.49

135 31 .0

e

-26.750

85

xx14

-24.325

135 13 5

6.0

1.30

1.73 (1)

xx15

-21.900

135 13 5

6.0

1.48

(2)

xx20

-20.700

95

135 26 .0

150 150

1.30

(1)

6.0

1.30

6.0

1.30

6.0


FL

t1

(years)

t2

e

1.30

(1)

177

135

135

6.0

1.30

(1)

132

(1)

85

150 150

6.0

1.30

(1)

252

135

135

6.0

1.30

(1)

392

135

135

6.0

1.30

(1)

288

(1)

144

150 150

6.0

1.30

(1)

443

135

135

6.0

1.30

(1)

709

135

135

6.0

1.30

(1)

514

150 13.5 1.3 0

(1)

1.30

(1)

(1)

960

150 150

6.0

1.30

257

135

135

135

6.0

1.30

17042 150 150

6.0

1.30

(1)

470

135 135

6.0

1.30

(1)

1135 95

135 2 6.0

1.61

303

95

135 2 6.0

1.61

6.0

6.0

1.30

(1)

2262

95

18.5

1.61

231

70

95

18.5

1.61

(1)

982

70

70

6.0 1. 30

476

55

70

13.5

(2)

2291

55

12993

150 150

1.30

(1)

909

135 135

1.61

154

135 150 13.5 1.3 0

(1)

952

95

135 2 6.0

1.61

335

70

95

18.5

1.61

(2)

285

55

70

13.5 1 .82

16.0

1.53

251

95

135 26.0

6.0

1.67

218

70

95

18.5 1 .84

xx23

-9.300

55

55

6.0

1.33

2781 55

70

13.5

1.60

1424

55

55

6.0

xx24

-5.300

45

55

11.0

1.62

2166 45

55

11.0

1.62

3453

45

55

11.0

xx25

-1.300

45

55

11.0

1.62

8019 45

55

11.0

1.62

2503

45

55

xx26

2.700

45

55 11 .0

1.62

11955 45

55

11.0

1.62

800

45

xx27

6.700

45

55 11 .0

1.62

4276 45

55

11.0

1.62

211

45

xx28

10.700

45

45

6.0

1.40

1952 55

70

13.5

1.60

149

xx29

14.700

45

55

11.0

1.62

258

70

85

13.5

1.49

xx30

18.700

55

60

8.5

1.43

301

85

85

6.0

429

85

85

6.0

1.61

601

135

70

135

6.0

375

70

70

6.0 1. 30

1357

206 152 (1)

640

(2)

1528

421

55

70

13.5

416

55

55

6.0 1. 52

1.33

8007

55

55

6.0

1.33

55

55

6.0

1.33

9688

1.62

3952

45

55

11.0

1.62

3711 45

55

11.0

1.62

2986

11.0

1.62

3111

45

55

11.0

1.62

1874 45

55

11.0

1.62

1760

55

11.0

1.62

1996

55

11.0

1.62

1313 45

55

11.0

1.62

1571

55

11.0

1.62

903

45

55

11.0

1.62

714

45

55

11.0

1.62

957

45

55

11.0

1.62

751

45

55

11.0

1.62

342

45

45

6.0

1.40

990

167

55

55

6.0

1.33

810

55

55

6.0

1.33

761

45

55

11.0

1.62

235

(2)

45

1.60

55

309

1.60

6.0 1. 52

1.30

(1)

256

55

60

8.5

1.43

257

55

60

8.5

1.43

247

55

60

8.5

1.43

410

1.30

(1)

191

60

60

6.0

1.30

398

60

60

6.0

1.30

386

60

60

6.0

1.30

662

1.30

(1)

394

60

60

6.0

1.30

813

60

60

6.0

1.30

818

55

60

8.5

1.43

616

700

55

60

8.5

1.43

2327

55

60

8.5

1.43

2579 45

55

11.0

1.62

908

11920

45

55

11.0

1.62

45

6.0

1.40

24957

xx40

22.700

55

60

8.5

1.43

425

85

85

6.0

xx41

26.700

45

55

11.0

1.62

777

70

85

13.5

1.49

xx42

30.700

45

45

6.0

1.40

33972 55

70

13.5

1.60

xx43

34.700

45

45

6.0

1.40

> 10

5

45

55

11.0

1.62

> 10

45

xxTD

36.305

45

80

23.5

1.93

> 105

45

80

23.5

1.93

4363

45

- 121 -

FL

6.0

95

FL – Fatigue Lives

SCF (years)

135

75

(1) Minimum SCF applicable is 1.30;

e

135

75

Note:

t2

(mm)

118

484

1.30

t1

(1)

55

6.0


FL SCF (years)

1.30

-17.300

60

e

6.0

-13.300

60

t2

(mm)

150 150

xx22

20.305

t1

42

xx21

xxCD


FL SCF (years)

(mm)

(1)

2194

xx13

SCF

(mm)

195319 150 15 0

1.49 1.30

t1 t 2

(years)

4880 5

45

55 45 80

11.0 6.0 23.5

1.62 1.40 1.93

5

> 10

45

22696

45

45 80

6.0 23.5

1.40 1.93

16846 5

> 10

45 45

43815 45

45 80

6.0 23.5

1.40 1.93

5

> 10

48022

The green values represent the wall thickness differences for the re-built F3-FA model. (2) Weld classification factor of 1.14 (SCF is multiplied by 1.14) due to the yard weld, that is the bottom leg + bucket connection with the remaining leg part.

Chapter 4 : Study of possible structuralimprovements

4.4.6

Final remarks and conclusions

A new design was proposed for the F3-FA structure by removing the knuckle joint. The alternative structural solutions are first applied to the substructure and then combined with adjustments in the legtopside connection.

On the substructure, besides removing the knuckle joint, the suction buckets are centred with the legs and considered to be underneath the barge, during transport. On the topside a new leg-topside connection is proposed.

To compensate the worst dynamic behaviour of the structure to wave loading without the knuckle joint, as noticed in the first mode shapes natural periods, the leg diameter and wall thickness were modified. In the end, three different models with three different leg diameters were studied for fatigue (4000 mm, 4500 mm and 5000 mm), and results were compared with the srcinal leg’s diameter (3250 mm).

Except for the varia nt model with the srcinal leg diam eter, it was possible to satisfy in all othe r solutions with the fatigue requirements for the substructure butt weld joints. Without the knuckle joint, tubular joint welds with relatively lower fatigue life estimation between the transition frame (i.e. knuckle joint) and the legs were avoid.

All final variant models have a total structural weight lighter than the srcinal F3-FA platform, mostly due to the removal of the knuckle joint. Based on Table 4.9 and Table 4.14 a final table for the overall results is presented below.

Table 4.15: Variant solutions overall results

Structural steel nett weight differences 1) from the re-built F3-FA model 2) Δ (mT) Δ (kN) Δ (%)

Variant solutions LegdiameterD=3250mm

-726

-7122

Design fatigue life 3) (years)

42

-14,3

LegdiameterD=4000mm

-453

-4444

-8,9

118

LegdiameterD=4500mm

-97

-952

-1,9

177

LegdiameterD=5000mm

-44

-432

-0,9

132

1) The weights differences regarding the truss frame and possible influences on the bucket’s plates wall thickness are neglected, since the bucket was not a part of the Iv-Oil & Gas work scope and the truss frame has practically no influence in the structure behaviour; 2) % of the re-built F3-FA SACS model structure self-weight and non-modelled structural steel (excl. buckets) = 5065 mT (section 3.3.2); 3) The minimum design fatigue life for the substructure butt welds is 100 years.

In Table 4.15 the design fatigue life is based on the lowest fatigue life estimation from the substructure butt weld joints and the structural steel weight on the re-built F3-FA SACS model values. From the overall results, presented in Table 4.15, the models with the leg diameters 4000 and 4500 mm are the most recommended for a variant solution of the F3-FA platform.


- 122 -

5 Conclusions 5.1

Final considerations

The F3-FA platform is an innovative solution for a fixed platform in the North Sea capable of returning a profit in oil or gas fields with a small lifetime production expectation. When the field is “dry” it is designed to be relocated in record time without having considerable decommissioned and installation costs, as frequently occurs with the existing offshore platforms.

In a pioneering project, with a tight schedule, the platform structure was not fully efficiently designed in the matters of fatigue performance. Thus, the main purpose of the master thesis was to improve the structural layout and dynamic behaviour under wave loading of the F3-FA platform by proposing different structural solutions to be adapted in future updates of platform design.

The first aspect investigated was the possibility of adopting a structure without the knuckle joint, the transition stiffening frame between the suction buckets and the platform legs. With this improvement, considerable structural steel weight is reduced (around 1000 mT). But this improvement has a direct impact in reducing the first structure frequency modes due to an increase of the flexibility of the all structure and, consequently, a worse fatigue performance.

Furthermore, without the knuckle joint, fatigue critical joints of the srcinal design are avoided and centred suction buckets are considered, as proposed, instead of the eccentric ones of the actual F3FA platform, that require the need of the knuckle joint to distribute the leg loads to the foundation.

To solve the reduction of the first natural structure frequencies, three alternative solutions were investigated by enlarging the legs diameter. Besides the srcinal leg diameter (3250 mm), the following leg diameters were considered: 4000, 4500 and 5000 mm.

Also, new leg wall thickness was proposed to optimize the weight and improve the structural stiffness. The first frequencies were therefore increased, with respect to the srcinal design, with the increase of the leg diameter (and leg wall thickness redesign) but, at the same time, the structural steel weight was increased. However, at the end, only the structural model with the leg diameter of 5000 mm turns out to be heavier than the actual F3-FA platform.

The former leg-topside connection is made at the two levels. At the top deck, legs are bolted to the topside transferring shear and moment stress and a sleeve (fill inside with a bearing support) guides the legs between the top and the main deck and transmit horizontal loading. At the cellar deck, space is left available, due to the knuckle joint during construction and transport. For that reason, the connection is built offshore through a clamp system to absorb horizontal wave impact loads,

-123-

Chapter5:Conclusions

preventing lateral moveme nt of the legs. Since, this shows not be the perfect soluti on, with more offshore working hours during installation than expected, and also because with the increase of the leg diameter the connection is not possible without considerable changes in the topside, an alternative solution was proposed in section 4.4. In this new solution, a straight connection is accomplished with the sleeve going throug h the entire topside . The corner trusses are also replaced with truss rows directly connected to the legs and compatible with the transport barge used. This connection required less steel to build and no significant change was observed in the structure frequency modes.

With the final models completely defined by introducing the former improvements, variant solutions were analysed and checked for in-service and fatigue conditions, as presented in section 4.4.4 and section 4.4.5. For the most demanding fatigue analysis, the legs joints proved to satisfy the fatigue requirements, except when using the srcinal leg diameter size.

Therefore, the structural solution proposed can contained improvements to be implemented in future designs of the F3-FA platform, being the models with a leg diameter 4000 or 4500 mm the more balanced solutions, with a structural steel weight reduction to the srcinal model of around 450 or 100 mT, respectively (9% or 2% of the total structural steel nett weight used on the F3-FA platf orm, excluding buckets).

5.2

Future work perspectives

Although, the long extension of this work, some topics were not investigated, and can be addressed to future works in the same area:

(1) Centric buckets side by side wi th barge In the curre nt study, the solution starts from the configuration of a platform with a centric bucket foundation underneath the barge, that implies considerable restriction on the seaport draught, that needs to be 21 meters depth. However, the centric bucket could have been considered side by side with the barge, as before, with substantial increase on the topside plant size. However, this topic needs future investigation.

(2) Centric bucket design The benefits of the centric bucket were not fully considered. In fact, this enormous caisson with 15 meters diameter by 13 meters height could be probably reduced due to the eccentric vertical leg force acting on the foundation that no longer exists. Naturally, this assumption needs to be checked.


-124-

(3) Different transport barge A different transport barge, besides the flat top barge BOA 35, could imply in a more efficient and economical solution and probably mandatory in different world regions, e.g. Gulf Coast of Mexico. This practical issue needs further research to explore its repercussions.

(4) Structural engineering guide to offshore structures This work intended to enter in a new topic to Civil Engineering mainstream area of activity, that is the design of a fixed offshore platform.

Several of the design issues of this activity are considered as good practices commonly, accepted in Oil & Gas industry and implemented in computer software’s refined over many years of offshore experience. That is the case of the wave loading and the spectral fatigue analysis. At the end, it was believed that a good clarification on these subjects was achieved. But still, further works can explore this topics and go further in the investigations to better assist the young engineers who give the first steps in this area.

-125-



- 126 -

6 References [/1/]

Iv Oil & Gas B.V., doc. no. FA-IV-RP-01-001 Rev. 05, Structural Basis of Design, Venture F3th

FA, 24 June 2009. [/2/]

st

Iv Oil & Gas B.V., doc. no. FA-IV-S-RP-01-003 Rev. 03, In-place Analysis, Venture F3-FA, 21

April 2010 and respective Addendum 03 for ISO Hmax wave & Plaxis Lower Soil Matrix (Rev. th

01, 10 May 2010). [/3/]

Iv Oil & Gas B.V., doc. no. FA-IV-S-RP-01-004 Rev, 03, Fatigue Analysis, Venture F3-FA, 3

rd

August 2010. [/4/]

API RP 2A-WSD, Recommended Practice for Planning, Designing and Constructing Fixed Offshore Platforms – Working Stress Design, American Petroleum Institute, 21

st

Edition,

December 2000 and respective Errata and Supplement 1 (Dec. 2002), 2 (Oct. 2005) and 3 (Oct. 2007). [/5/]

ISO 19901-1:2005, Petroleum and Natural Gas Industries – Specific Requirements for Offshore Structure – Part1: Metocean Design and Operating Considerations, 1

st

th

Edition, 15

November 2005. [/6/]

ISO 19902:2007, Petroleum and Natural Gas Industries – Fixed Steel Offshore Structures, 1

st

st

Edition, 1 December 2007. [/7/]

ANSI/AISC 360-05, Specification for Structural Steel Buildings, March 2005.

[/8/]

Iv Oil & Gas B.V., doc. no. FA-IV-S-RP-01-002 Rev. 12, Weight Control Report, Venture F3th

FA, 30 June 2010. [/9/]

HSE, Steel, Offsh ore Technology Report 2001/015, Health and Safety Exec utive, BOMEL Ltd., HSE Books, 2002 (http://www.hse.gov.uk/research/otopdf/2001/oto01015.pdf).

[/10/]

SACS executive offshore 5.5, ©2013 Bentley Systems Inc.

[/11/]

J.H. Vugt, Handbook of Bottom Founded Offshore Structures, Volume I & II, January 2002.

[/12/] N. Haritos, Introduction to the Anal ysis and Design of Offshore Structures – An Overview, EJSE Special Issue: Loading on Structures, 2007. [/13/] M.M. Rienecker, J.D. Fenton, A Fourier approximation method for steady water waves, Journal of Fluid Mechanics, vol. 104, Cambridge University Press, p.119-137, March 1981. [/14/]

The Free Dictionary by Farlex Website: http://encyclopedia2.thefreedictionary.com/streamlining+portal+blood+flow.

[/15/]

T. Sarpkaya, In-Line and Transverse Forces on Smooth and Rough Cylinders, in Oscillatory Flow at High Reynolds Numbers, Naval Postgraduate School, Rep. NPS69-86-003, July 1986.

-127-

Chapter6:References

[/16/]

EN 10025, Hot Rolled Products of Structural Steels. Technical Delivery Conditions, November 2004.

[/17/]

EN 10225, Weldable Structural Steel for Fixed Offshore Structure. Technical Delivery Conditions, September 2001.

[/18/]

O. M. Faltinsen, Sea Loads on Ships and Offshore Structures, Cambridge University Press, 1990.

[/19/]

F. Gerven, Optimising the Design of a Steel Substructure for Offshore Wind Turbines in Deeper Waters, T.U. Delft, October 2011.

[/20/]

Iv-Groep website: http://www.iv-groep.nl/.

[/21/] SPT Offshore: http://www.sptoffshore.com/. [/22/]

Heerema Group: http://www.heerema.com/.

[/23/]

Centrica Energy: http://www.centrica.com/.

[/24/]

T. Langford & P. Sparrevik, Skirted Caisson Foundations for Offshore Structures, www.ngi.no.

[/25/]

Gulfaks C in Structurae: http://structurae.net/structures/data/index.cfm?ID=s0003250.

[/26/]

The Sleipner A Failure: http://matdl.org/failurecases/Other_Failure_Cases/Sleipner_A.

[/27/]

F3-FA superbolts: http://www.nord-lock.com/products/superbolt/nut-styletensioners/introduction/.

[/28/]

Ivormatie Magazine, Volume 25, October 2011, p. 26-28 (publication of Iv-Groep).

[/29/]

Offshore Holland Magazine, 1 Edition, May 2011, p. 6-8.

[/30/]

G. Jergeas, J. Van der Put, Benefits of Constructability on Construction Projects, Journal of

st

Construction Engineering and Management, July/August 2001, p. 281-290. [/31/]

N. S. Galgoul, Fatigue Analysis of Offshore Fixed and Floating Structures, Talks on Structural Engineering, http://www.tuhh.de/sdb/vortraege/Vortraege.en.html, Technische Universität Hamburg-Harburg, January 2007.

[/32/]

Ocean Wave Spectra: http://www.wikiwaves.org/Ocean-Wave_Spectra.

[/33/]

Dean Stream Function theory: http://www.orcina.com/SoftwareProducts/OrcaFlex/Documentation/Help/Content/html/Waves, DeanStreamFunctiontheory.htm.

[/34/]

CAA, “CAP 437, Offshore Helicopter Landing Areas – Guidance on Standards, August 2005.

[/35/]

ICAO, Annex 14 – Aerodromes, Volume II – Heliports, Second Edition July 1995.

[/36/]

ICAO, Heliport Manual (Doc 9261-AN/903), 3 edition 1995.

rd

Chapter6:References

-128-

A. Wind loads

A.1

Wind speed and wind force calculations

In storm conditions the mean wind speed profile can be described by a logarithmic profile and may be expressed by the following expression: (A.1) Where, is the 1 hour mean wind speed at height

above mean sea level (m/s);

is the 1 hour mean wind speed at 10 m above mean sea level (standard ref.) (m/s); is a coefficient given by: (A.2) is the height above mean sea level.

For the same storm conditions, the design wind speed, for averaging times shorter than 1 hour is given by: (A.3) Where, is the design wind speed at height

above mean sea level and corresponding to an

averaging time period (m/s); is the turbulence intensity at height

above mean sea level and is given by: (A.4)

is the time averaging interval (s),

;

is the standard reference time averaging interval (1 hour = 3600 s) (s).

The wind drag force on an object may be calculated as: (A.5) Where, is the wind action on the object; is the mass density of air; is the design wind speed; is the shape factor; is the area of the object.

129 - -

Appendix A

A.2

Win d load results

Table A.1: Wind parameters

Operational condition U0 (m/s) 24,7

t0 (s)

t (s)

(1hour)

(1min.)

3600

60

Survival condition U0 (m/s) 3 3,7

C 0,12429

t0 (s)

t (s)

(1hour)

C

(1min.)

3600

60

0,14100

Table A.2: Wind forces

Operational condition Wind areas

d a o ln d i W

th u o S to th r o N

d a lo d in W

ts a E o t ts e W

1

Location

Cellar deck

Elevation (m)

22,000

Cs

1,5

Height (m)

6,0

Width (m)

30,0

U(z) (m/s)

27,1

Iu(z)

0,104

u(z,t) (m/s)

31,9

Survival condition

Wind force (kN/m)

Total wind force (kN)

5,59

U(z) (m/s)

Iu(z)

u(z,t) (m/s)

Wind force (kN/m)

Total wind force (kN)

0,124

45,2

11,27

Maindeck

29,000

1,5

8,0

30,0

28,0

0,098

32,6

7,79

168 234

37,4

2

38,8

0,116

46,3

15,77

338 473

3 4 5 6

Topdeck Helideck Ventstack Crane

36,500 41,250 62,500 45,000

1,5 1,5 1,0 1,5

7,0 2,5 40,0 8,0

30,0 25,0 2,5 6,0

28,7 29,1 30,3 29,3

0,093 0,091 0,083 0,089

33,2 33,5 34,5 33,7

7,07 2,57 29,22 8,34

212 64 73 50

39,9 40,4 42,4 40,8

0,111 0,108 0,098 0,106

47,2 47,7 49,4 48,1

14,36 5,23 59,78 16,99

431 131 149 102

1 2 3a 3b 4 5 6

Cellar deck Maindeck Topdeck Topdeck Helideck Ventstack Crane

22,000 29,000 38,000 36,500 41,250 62,500 45,000

27,1 28,0 28,8 28,7 29,1 30,3 29,3

0,104 0,098 0,092 0,093 0,091 0,083 0,089

31,9 32,6 33,3 33,2 33,5 34,5 33,7

5,59 7,79 10,16 7,07 2,57 29,22 8,34

308 429 356 141 64 73 50

37,4 38,8 40,0 39,9 40,4 42,4 40,8

0,124 0,116 0,110 0,111 0,108 0,098 0,106

45,2 46,3 47,4 47,2 47,7 49,4 48,1

11,27 15,77 20,65 14,36 5,23 59,78 16,99

T o taw l i n dl oa dN o r tht oS o ut h 1,5 1,5 1,5 1,5 1,5 1,0 1,5

6,0 8,0 10,0 7,0 2,5 40,0 8,0

55,0 55,0 35,0 20,0 25,0 2,5 6,0

80 1

T o taw l i n dl o a dWe s t oE a s t

1624

1421

620 867 723 287 131 149 102 2879

Note: The wind loads direction South to North and East to West use the same values but opposed direction of the wind loads North to South and West to East respectively.

Appendix A

- 130 -

A.3

in a e a lo ds

igure .1: Win d area l ad (a apte d ro

- 1 31 -

A pendi A

[/2 ] )

B. Wave Loads

B.1

Airy’s linear wave theory

The Airy’s linear wave theory is the foundation of all wave theories, composed by a relative simple wave model adopting sine functions to better describe the wave motion.

To successful solve the water wave motion (kinematics of the water particles) some basic considerations must be assumed [/13/] & [/18/]: -

the seawater is incompressible and inviscid (ideal fluid); the fluid motion is irrotational.

When a fluid is irrotational a velocity potential can be used to describe the fluid velocity vector. Using a Cartesian coordinate system in three dimensions the fluid velocity vector is then given by (i, j and k are unit vectors along x, y and z axis, respectively) [/18/]:

(B.1)

Kelvin’s states that when a inviscid fluid is initially irrotational then it will remain always irrotational. In a incompressible flow, the changes in temperature and pressure are sufficiently small to neglect any deviation in the fluid density, i.e.

[/18/].

Thus, the velocity vector must satisfy the Laplace equation [/18/]:

(B.2)

To simplify the problem, an unsteady flow (fluid properties are timeless) is adopted and the Bernoulli’s principle (that states: for any inviscid flow the incre ase of speed is match with the decre ase in pressure of the fluid) is used [/18/].

To solve Laplace equation boundary conditions must be specified.

Appendix B

- 132 -

For a fixed body in a fluid, a kinematic and a dynamic boundary condition on the free surface of the fluid can be defined [/18/]: -

kinematic boundary condition states that the body is impermeable: (B.3) with

-

the free surface equation;

dynamic boundary condition states that the water pressure is equal to the atmospheric pressure on the free surface: (B.4)

Both equations are non-linear and a linear theory can be used to linearized the free surface 22

conditions.

The linear theory assumes that the velocity potential –

– is proportional to the w ave

amplitude, , what can only be valid if the wave amp litude is small enough, when compared to the wave length and the body dimensions in contact with the flow [/18/].

One of the problems that need to be solved is the position of the free surface that is a part of the solution. Using Taylor’s expansion, we can transfer the free surface boundary condition from the free surface position

to the mean free surface, defined as

[/18/]:

-

kinematic condition:

(B.5)

-

dynamic condition:

(B.6)

The two linearized boundary conditions are then combined and the equation can be re-write to include an harmonic oscillation of the flow with a circular frequency

:



with

(B.7)

.

Adopting a horizontal sea bottom and infinite horizontal free surface, it is possible to achieve the formulation of the Linear wave theory and finally, the velocity potential equation can be determined: (B.8)

From the equation B.8 the linear wave can be characterized. 22

The quadratic terms on the free surface boundary conditions are neglected.

133 - -

Appendix B

The wave surface elevation with an a mplitude

at the position

and instance

for a wa ve

propagating along the positive x-axis [/12/] is given by: (B.8)

The wave has the shape of a sine curve has shown in Figure B.1.

Figure B.1: Linear wave theory wave shape

The along wave

and vertical

velocities are given by

[/12/]: (B.9) (B.10)

Where, is the wave amplitude; is the wave angular frequency,

;

isthewavenumber,

;

isthewavecelerity,

;

is the wave length.

The relation between the wave number and the wave angular frequency, according to the linear wave theory is given by: (B.11) where g is the gravity acceleration [/11/] & [/12/]. This solution is derived from the free surface boundary condition and the Laplace equation.

Finally, the along wave

and vertical

accelerations are approximately given by [/12/]: (B.12) (B.13)

Appendix B

- 134 -

The linear wave theory is very usefully, combined with Morison’s equation, for the calculation of fatigue loads or preliminary ultimate loads for offshore structures though, in general, it is not the desired design wave for extreme conditions due to their limitations for shallower waters or waves too steep.

B.2

Stream function wave theory

It has been shown in section B.1 that regular wave theories are based on an ideal fluid with two boundaries: the sea floor and the sea surface. In the sea floor, the water particle velocity normal to the bottom is zero, i.e.

with

(

is the water depth). In the sea surface there is a

kinematic and a dynamic condition to be satisfied [/33/].

The theories differ in respect to the free surface boundary (that is, the nonlinear condition), and are either linearized in one of the two conditions (kinematic or dynamic) or include nonlinear terms to a particular order.

A stream function is a vector field

which satisfies [/33/]: (B.14) (B.15)

for the particular case of a wave motion

and

are the vertical and horizontal velocity fluid

components.

Dean develop the initial concept of the stream function wave theory by a continuous correction of the problem’s solution to minimize the error of a wave already known (by lab results). Based on this idea Rienecker & Fenton improved the numerical method to solve the nonlinear wave problem [/13/].

According to Fenton’s Stream Function theory, in a incompressible fluid and irrotational motion [/33/]: 

the Laplace’s equation is satisfied: (B.16)



and the following boundary conditions are valid: attheseabed,

;

o

atthefreesurface,

.

o



the Bernoulli’s equation is satisfied: (B.17)

135 - -

Appendix B

To satisfy the Laplace’s equation and the seabed boundary condition [/33/]:

(B.18) Where, is the wave surface elevation; is the wave number; is the order of the stream function. The constants

are chosen to approximately satisfy the Bernoulli’s equation and the boundary

condition at the free surface by means of a numerical process.

For practical use Fenton’s Stream Function theory can be used to determine the wave velocities [/13/]: with

(B.19)

with

(B.20)

Appendix B

- 136 -

B.3

Static wave analysis

API [/4/] and ISO [/5/] & [/6/] codes present the calculation sequence for a static deterministic design wave force, as shown in Figure B.2.

Figure B.2: API & ISO procedure for calculation of deterministic static wave & current forces (taken from [/4/])

In this appendix a simple presentation is made of the steps involved in the static wave analysis:

a. Doppler effect In the analysis the current is considered to be in line with the wave direction, the consequences of their co-existence will lead for the wave length to stretch – Doppler effect. According to API

[/4/] and

ISO [/5/] the correct wave period to be used in all the regular periodic wave theories is the

intrinsic

period –

–, that takes into account the wave-current iteration.

The wave data available for the F3-FA project was measured from a stationary point on site, that is the apparent wave period, .

For a first order wave model (linear wave theory) a good procedure for determining the intrinsic wave period is to solved the following three nonlinear equations, according to API [/4/] and ISO [/5/]:

(B.21)

137 - -

Appendix B

Where, is the wave length; is the effective in-line current speed; is the component of the steady curr ent profile at elevation

in the wave direction;

is the point elevation (negative under the storm mean water level); is the storm water level; isthewavenumber,

;

is the wave angular frequency,

;

isthewavecelerity,

.

In the next two table s it is possible to compare the wave parameters estimated for the desi gn conditions of F3-FA platform with and without Doppler effect based on the expressions above: Table B.1: Wave parameters without the Doppler effect

Conditions

wave

water

wave

wave

wave

wave

wave

period

depth

height

circular freq.

number

length

celerity

(m)

c (m/s)

T(s)

d(m)

H(m)

k

Operational min water depth

12,4

40,300

14,9

0,507

0,0309

20 3 ,2 90

16 ,3 9 4

Operational max water depth

12,4

41,990

14,9

0,507

0,0305

20 5 ,7 21

16 ,5 9 0

Survival min water depth

15,1

40,300

20,7

0,416

0,0238

26 4 ,4 63

17 ,5 1 4

Survival max water depth

15,1

42,950

20,7

0,416

0,0232

27 0 ,6 14

17 ,9 2 1

(rad/s)

Table B.2: Wave parameters with the Doppler effect

wave period Conditions Operational min water depth Operational max water depth Survival min water depth Survival max water depth

T(s)

water depth

wave number

effective in-line current speed

Ti (s)

intrinsic wave circular freq.

intrinsic wave celerity

(m)

(rad/s)

ci (m/s)

k

12,4

40,300

0,0290

0,787

1 2, 98 5

2 16 ,7 57

0 ,4 84

1 6 ,6 93

12,4

41,990

0,0286

0,788

1 2, 97 8

2 19 ,3 60

0 ,4 84

1 6 ,9 03

15,1

40,300

0,0222

1,018

1 5, 96 5

2 83 ,5 77

0 ,3 94

1 7 ,7 62

15,1

42,950

0,0217

1,018

1 5, 94 6

2 90 ,0 24

0 ,3 94

1 8 ,1 88

st

I

(m/s)

actual wave length

d(m)

Note: This expressions are valid for 1

V

intrinsic wave period

order waves, i.e. Airy linear wave theory, and based on the

dispersion equations for a two dimensional sine wave propagating in one direction, as seen in appendix B.1. For higher order waves, such as the Stream Function theory, the dispersion relationship has to be calculated numerically.

Appendix B

- 138 -

b. Wave ki nematic s (appropriat e model) In Iv-Oil & Gas is common practice to use the deterministic Airy’s linear wave theory for the fatigue loads calculation and the Stream Function 11

th

Order theory for the in-place analysis. However, an

appropriate method for the wave theory selection can be made through the diagram presented on the figure below. The diagram displayed is presented in the API [/4/] and ISO [/6/] codes and represents a parabolic relationship between the wave steepness

and the relative water depth

.

Figure B.3: Diagram for the selection of an appropriate periodic wave theory (adapter from [/11/])

Calculating the wave steepness and relative water depth dimensionless factors:

Table B.3: Wave steepness and relative water depth calculation

Conditions

Ti(s)

( 1)

(m)

(-)

(m)

(-)

Operational – min water depth

12,985

40,300

14,90

0,00901

0,02437

Operational – max water depth

12,978

41,990

14,90

0,00902

0,02542

Survival – min water depth Survival – max water depth

15,965 15,946

40,300 42,950

20,70 20,70

0,00828 0,00830

0,01612 0,01723

(1) Intrinsic wave period (i.e. with the Doppler effect)

Looking at Figure B.3 and Table B.3, for the conditions given, the Stream Function 7

th

order or higher

is the most suitable wave model for the conditions given. Also the waves site depth can be classified as intermediate depth waters. 139 - -

Appendix B

c. Wave kinematics factor The regular wave’s models do not account for the irregularity of the wave profile. A practical solution for the problem is to multiply the horizontal wave velocities and accelerations by a factor in the range 0,85 to 1,00. For the F3-FA structure a wave kinematics factor of 0,95 was used.

d. Cur rent blockage factor The presence of a fixed body, e.g. jacket structure, in the direction of the flow will cause the stream to diverge and go around the structure, therefore the current speed is reduced. However, since the SIP platform is not a standard jacket structure and no braces exist connecting the legs, the current blockage factor is conservatively adopted as 1,00.

e. Combine d wave plus current kinematics Wave kinematics should be combined with the current profile. Looking at the Metocean data [/1/], the current profile is only specified to the storm mean water level and must be stretch to the wave surface. A conservative procedure is to extend the current profile vertically from the mean water level to the wave crest and to truncate the current profile above the wave trough so that no current effects are included above the wave surface.

A different approach is to use a linear stretching technique instead of extended the current profile vertically above the storm mean water level to the wave crest, according to API [/4/] & ISO [/5/] codes: (B.22) Where, d – storm mean water depth; η – water surface elevation above the storm mean water depth;

z – elevation of the current point above the storm mean water depth; z’ – effective stretched elevation of the current point above the storm mean water depth.

Figure B.4: Current diagram force stretch to wave surface by two different methods

Appendix B

- 140 -

f.

Marine g rowth

All substructure members (structural and non-structural) should be increased to account for marine growth thickness and classified as “smooth” or "rough” depending on the amount of marine growth they are expected to accumulate during the platform service life.

For the F3-FA legs the following marine growth thickness’s on the radius around the section applied: -

from seafloor till LAT -10 meters (-40,300 m to -10,000 m): from LAT -10 meters till LAT + ½ HAT (-10,000 m to +0,305 m):

50 mm; 100 mm.

3

For the increase of weight a dry density of 1150 kg/m is considered for the marine growth.

g . Mori s on equation Wave forces depend, on hydrodynamic conditions, has stated before, but also in the geometry (relative to the wave length) and compliancy (if the element is rigid or not) of the structure where the wave load is acting [/12/].

If the structural element is a large volume body the flow pattern when crossing the “obstacle body” will change, affecting the wave field by diffraction and reflection. The wave forces on these structures have to be determined by diffraction theories with complex and time-consuming numerical calculations, that will not be discussed further on this thesis.

In the other hand the F3-FA substructure legs can be considered structural slender elements, based on the leg diameter (D) and on the wave length ( λ) [/4/]. Thus, the slenderness of the legs will have no significant impact on the global flow pattern and a more simple expression to determine wave forces can be applied [/13/].

This expression is called Morison’s equation, is valid for a submerged rigid vertical cylinder with a slenderness of

, diameter relative to wave length, and is given by: (B.23)

Where, is the hydrodynamic force acting normal to the axis of the member; is the inertia force and is given by: (B.24) is the drag force and is given by: (B.25)

141 - -

Appendix B

is the horizontal water particle velocity; is the horizontal water particle acceleration; is the density of the water; inertia coefficient; drag coefficient; leg diameter (inc. marine growth).

h. Drag a nd Inertia coeff icients According to Morison’s equation to calculate the wave forces on a fixed body there are two force components. The drag force acting on the direction of the flow related to the fluid pressure on the fixed object (depends on the flow velocity) and the inertia force acting perpendicular to the direction of the flow and associated with the inertia effects on the viscous fluid (depends on the flow acceleration).

As the magnitudes of the inertia and the drag forces do not decrease/increase with depth at the same rate, since the inertial load leads the drag load by 90º, i.e. a quarter of a period hydrodynamic coefficients, which are inertia and drag coefficients

and

[/11/], two empirical respectively, are

introduced. Both values generally lie between 0,8 and 2,0 and as a rule it can be stated that decreases as

increases and vice versa [/12/].

The drag and inertia coefficients depend upon surface roughness (i.e. marine growth), Reynolds number, Keulegan-Carpenter number, wave velocity and member orientation [/4/].

Roughness

One of the causes of the marine growth is the increase of the effective diameter of the legs section but also the increase of roughness that will affect the inertia and drag coefficients.

For most situations wave forces can be calculated using two separate values for sections affected and not affected by marine growth [/4/] & [/6/]: and

,

for smooth members

and

,

for rough members

In the tidal range structural elements are considered rough down to the sea floor, outside the tidal range the marine growth accumulation is small enough to be neglected and all elements above the tidal range are considered to be hydrodynamic smooth [/5/].

Appendix B

- 142 -

Reynolds Number

The Reynolds number (Re) measures the ratio between inertial forces and viscous forces and, consequently characterize different flow regimes, as shown in equation B.26. (B.26) Where, is the maximum velocity (inc. current) normal to the cylinder axis; is the effective diameter (inc. marine growth); is the kinematic viscosity of the water.

Laminar flow occurs where viscous forces are dominant (low values of Re) and turbulent flow occurs where inertial forces are dominant (high values of Re). The viscosity of the fluid will cause the flow to separate from the body surface resulting in the formation of a turbulence region called wake [/14/].

Figure B.5: Flow in a bluff body (a) and in a streamlined body (b) (adapted from

[/14/])

The force coefficients of streamlined bodies are independent of Reynolds number due to the small wake formation (small region of disturbed flow). While, the legs of the F3-FA platform have a circular section that will induced in a bigger wake (consequently in a bigger drag pressure on the legs) and thus, could be Reynolds number dependent [/4/].

For the in-place operational and survival conditions the Reynolds number is determined as follows: Table B.4: Reynolds number calculation (adapted from [/2/])

Conditions

(1)

(2)

(m/s)

(m)

Operational – min. water depth

6,63

Operational – max. water depth Survival – min. water depth

6,47 10,04

Survival max. water level

9,49

(0º C)

(25º C)

2

(m /s)

2

(-)

(m /s)

(-)

6 12,8 x 10

6 25,6 x 10

6

(1)

3,450

1,785 x 10

-6

12,5 x 10 6 19,4 x 10 18,3 x 10

6

º

25,0 x 10 6 38,8 x 10 36,7 x 10

6

to 25 Celsius based on the minimum and maximum temperatures on º

the sea surface for a return period of 100 years.

143 - -

-6

Wave particle veloc ities were determi ned using SACS Softwa re for the survival condi tions. The kinemat ic viscosity was determined for a temperature range of 0

(2)

6

0,893 x 10

Section diameter inclu ding marine growth thickn ess (worst situa tion: tidal range zone).

Appendix B

As for most offshore structures, the flow in the F3-FA legs is in a turbulent regime where the drag coefficient is independent of Reynolds number, as shown in the table above, Re > 4000.

Keulegan-Carpenter Number

The force (drag & inertia) coefficients standard values, presented before, are only valid for waves with . Where,

is the maximum horizontal particle velocity at storm mean water level under

the wave crest for the two-dimensional wave kinematics theory;

is the intrinsic wave period; and

the cylinder diameter (inc. marine growth) [/4/].

Table B.5: Criteria for the drag & inertia coefficients – D = 3450 mm

Conditions

(m/s)

(1)

(s)

Operational–min.waterdepth

6,63

12,4

Operational–max.waterdepth

6,47

12,4

23,8 23,3

Survival–min.waterdepth

10,04

15,1

43,9

Survival–max.waterdepth

9,49

15,1

41,5

(1)

Using the (apparent) wave period ( ) instead of the intrinsic wave period ( ) is conservative.

Besides the srcinal leg diameter of 3250 mm (+ 2 x 100 mm marine growth), the drag coefficient is calculated for the diameters values of: 4000 mm, 4500 mm and 5000 mm (these values are important for our study of alternative design solutions for the F3-FA platform).

Table B.6: Criteria for the drag & inertia coefficients – D = 4200 mm, 4700 mm & 5200 mm

Conditions

(m/s)

(s)

D = 4000 mm + 2 x 100 mm = 4200 mm Operational–min.waterdepth

6,625

12,4


6,467

12,4

Survival–min.waterdepth Survival–max.waterdepth

25,3 24,7

10,039

15,1

46,6

9,487

15,1

44,1


6,625

12,4


6,467

12,4

Survival–min.waterdepth Survival–max.waterdepth

18,3 17,8

10,039

15,1

33,7

9,487

15,1

31,8


6,625

12,4

16,4


6,467

12,4

16,0

Survival–min.waterdepth

10,039

15,1

Survival–max.waterdepth

9,487

15,1

(1)

30,3 28,7

(1)

Although under 30 the value was calcula ted without the current effect.

Appendix B

- 144 -

For the F3-FA platform substructure legs, the above criteria is not met for the operational conditions because the waves are not larger enough, compared to the cylinder diameter, to neglect the “wave encounter” phenomenon.

The size of the legs is not enough to interrupt the global flow pattern (that could cause significant wave diffraction), but locally, as the fluid pass through the legs, a wake is created, when the flow reverses the fluid particles will impact the legs with greater velocity resulting in a bigger hydrodynamic force [/4/], as shown in Figure B.5.

The API [4] provides a method to determine a more accurate value to the force coefficients due to “wake encounter” based on the Keulegan-Carpenter number.

The Keulegan-Carpenter number is a parameter that also describes the unsteadiness of the flow, i.e. if the fluid properties are time-dependent, and is given by: (B.27) Where, is the max. velocity (inc. current) normal to the cylinder axis in a half wave cycle; is the duration of the half wave cycle; is the effective diameter (inc. marine growth).

Using the Keulegan-Carpenter number, API [/4/] and ISO [/5/] allow the calculation of an wake amplification factor as a function of K/C ds (where Cds is the standard drag coefficient). Table B.7: Keulegan-Carpenter number and factor K/C ds

Conditions

(m/s)

(s)

smooth members

rough members

D = 3250 mm + 2 x 100 mm = 3450 mm Operational – min. water depth

6,63

12,4 / 2 = 6,2

23,0

36,6

22,7

Operational – max. water depth

6,47

12,4 / 2 = 6,2

22,5

35,8

22,1


6,63

12,4 / 2 = 6,2

23,0

30,1

18,6


6,47

12,4 / 2 = 6,2

22,5

29,4

18,2


6,63

12,4 / 2 = 6,2

18,9

26,9

16,6


6,47

12,4 / 2 = 6,2

18,5

26,2

16,2


6,63

12,4 / 2 = 6,2

15,3

24,3

15,0


6,47

12,4 / 2 = 6,2

14,9

23,7

14,7

145 - -

Appendix B

Figure B.6 and Figure B.7 , were taken from ISO

[/5/] and, present implicitly the drag coefficie nts

values measured in laboratory and in situ from several authors, from where the wake amplification factor (

) is calculated, as shown in Table B.8.

Figure B.6: Wake amplification factor for drag coefficient as a function of the K, for K<12 (adapted from [/6/])

Figure B.7: Wake amplification factor for drag coefficient as a function of the K/Cds, for K>12 (adapted from [/6/]) [line 1 is field data; line 2 is lab data]

Appendix B

- 146 -

From the previous figures, we can easily reach the following values for the wake amplification factor and the design drag coefficients for operational conditions:

Table B.8: Drag coefficient based on the wake amplification factor according to API [/4/]

Diameter (mm)

Wake amplification factor (Cd/Cds) smooth members

rough members

Final Drag coefficient (C smooth members

d)

rough members

Standard

--

--

0,65

1,05

3700

1,14

1,27

0,74

1,33

4200 4700

1,17 1,22

1,34 1,42

0,76 0,79

1,41 1,49

5200

1,25

1,45

0,81

1,52

Note: Marine growth included on diameter of the legs.

The Keulegan-Carpenter number can also suggests which of the components of the hydrodynamic force is predominant. For typical jacket structures, in storm conditions, K is greater than 40 and the drag force is predominant over inertia force. On the other hand, for large-diameter columns, if K is less than 10 the inertia force is dominant [/4/]. In the F3-FA platform, can be stated that the drag force will be predominant, however in any case the inertia force cannot be neglected.

In the srcinal F3-FA design the following drag coefficients were used conservatively:

Table B.9: Drag coefficients used in the F3-FA design

Condition

Drag coefficient (Cd) smooth members

Standard

0,65

Operational

1,12

Survival

1,02

rough members 1,05 1,28 1,10

Although we have obtain bigger drag coefficient value, in the operational condition, for rough members. The most unfavourable condition is the survival condition, whose drag coefficient values did not require to be upgraded.

Also, the fatigue analysis is prevalent in the substructure design. And the drag and inertia coefficients 23

assumed for fatigue calculations are considered conservative , i.e. Cd = 0,50 and 0,80 for smooth and rough members and Cm = 2,00 for both, according to API Figure B.8 and Figure B.9.

23

[/4/] for small waves, as shown in

Based on the scatter wave diagram, see section 3.4.3, the most common waves are rather small and have a Keulegan-

Carpenter number between 1,0 < K < 6,0.

147 - -

Appendix B

Figure B.8 and Figure B.9 are taken out of Sarpkaya study of hydrodynamic forces in smooth and rough cylinders [/15/] that influence the wake amplification factor graphics showed in API [/4/] and ISO [/6/] codes; and exemplify clearly the drag and inertia coefficients for smooth and rough cylinders, independent of Reynolds number (Re), as a function of the Keulegan-Carpenter number.

Figure B.8: Drag and inertia coefficients for a smooth cylinder, with Re=

Figure B.9: Drag and inertia coefficients for a rough cylinder, with Re=

Appendix B

=11525 (adapted from[/15/])

=6833 and 14200 (adapted from[/15/])

- 148 -

i.

Condu ctor s hielding factor

Conductors, near the legs, have a positive effect on the wave force since they will create “shield”, through a separation and deviation of the flow. Thus, a reduction factor may be applied to the drag and inertia coefficients depending on the configuration of the structure and the number of well conductors [/4/]. The F3-FA platform is not a typical jacket structure and conservatively no conductor shielding factor is considered.

j.

Hydrodynamic models for appurtenances

Due to appurtenances in the platform substructure such as, boat landings, bumpers, walkways, stairways, grout lines and anodes, the global hydrodynamic forces may increase substantial. Furthermore, special consideration must be paid for local design situations [/4/].

For a typical jacket structure the corrosion protection, from the mud level till the splash zone, is made by jacket anodes. Thus, a practical approach is to consider a 5% increase on the drag coefficient for rough members.

In the particular case of the F3-FA platform the sacrificial anodes are on the inner surface of the legs and inside the buckets, so no additional considerations need to be taken into account.

149 - -

Appendix B

C. Hydrodynamic loads calculation The goal of the appendix is to determine the horizontal distribution of the wave and current loads over the F3-FA platform substructure, in a periodic design wave. Thus, allowing the reader an overview of the basis of the hydrodynamic loads. The hydrodynamic loads, here estimated, are calculated through Morison equation, based on the linear wave velocities and accelerations normal to the submerged members. To simplify the study only a single F3-FA column is considered.

The F3-FA in-place survival condit ion with maximum water depth parameters are chosen for the calculations, as presented in Table C.1, Table C.2, Table C.3 and Table C.4.

Table C.1: In-place survival condition, with maximum water depth, wave data for the F3-FA site

Condition

wave

water

wave

period

depth

height

T (s) Survival Max water depth

15,946

d(m) (1)

wave

H (m)

42,950

wave

amplitude (m)

20,7

wave

circular freq.

0,394

wave

length

k

(rad/s)

10,35

wave

number

factor

c (m/s)

(m)

0,0217

kinematic

celerity

270,614

(-)

1 8,188

1,00

(1) See Table C.5.

Table C.2: In-place survival condition current velocities for the F3-FA site

Elevation above the sea floor

Condition 1.00d Survival

0.75d

1,04

0.50d

1,04

0.30d

1,04

0.10d

1,04

0,95

0.05d 0,88

1 meter 0,81

Note: d is the still water depth.

Table C.3: In-place survival condition drag and inertia coefficients for the F3-FA design

Drag coefficient (Cd)

Condition

smooth members Survival

Inertiacoefficient(C

rough members

1,02

smooth members

1,10

m)

rough members

1,60

1,20

Table C.4: Marine growth considered in substructure legs for the F3-FA design

Condition

Legs diameter DINITIAL (m)

Marine growth layer #1 zINF (m)

zSUP (m)

Marine growth layer #2

Wall thick. increase (cm)

DFINAL (m)

zINF (m)

zSUP (m)

Wall thick. increase (cm)

Nomarinegrowth

3,25

-40,3

-10,0

0

3,25

-10,0

0,305

0

Marinegrowthlayer

3,25

-40,3

- 10,0

5

3,35

-10,0

0 ,305

10

DFINAL (m)

3,25 3,45

For more information about the parameters see appendix B.3.

Appendix C

- 150 -

The combined effect wave and collinear current will stretch the design wave, affecting the wave period – intrinsic wave period –, as shown in Table C.5.

Table C.5: Intrinsic wave period calculation for the in-place survival condition, with maximum water depth, for the F3-FA site

Condition

wave period

water depth

T(s) Survival max water depth

15,1

d(m) 42,950

wave number k 0,0217

V

effective in-line current speed I

(m/s)

1,018

intrinsic wave period

wave length

Ti (s) 15,946

intrinsic wave circular freq. (rad/s)

(m)

290,024

, 30 94

intrinsic wave celerity ci (m/s) 1 8, 18 8

An excel sheet has been developed to calculate manually the wake kinematics, based on the Airy’s linear wave theory, and the corresponding fluid forces, through Morison equation, for a single pile.

Some corrections had to be done for the wave horizontal velocity distribution above the wave crest, because the linear wave theory is only valid till the sea mean water level. To solve this problem, three methods are proposed: (a) To use a linear stretchi ng technique, similar to the one referred for the current velocity profile, see appendix B.3; (b) To extended the wave profile vertically from the mean water level to the wave crest; (c) To extended the wave profile exponentially using Airy’s wave kinematic expression.

Figure C.1: Wave horizontal velocity distribution methods above wave crest for the linear wave theory (method (a) on left; (b) on the center; (c) on the right)

In the wave calculations method (c) applies, since it is used by SACS software [/1/].

The horizontal wave velocities were calculated for the three different methods and are presented in Figure C.2.

151 - -

Appendix C

7.0 )s / 6.5 m ( n o it 6.0 u b rit si 5.5 d y itc 5.0 o l e v l a t 4.5 n o zi r 4.0 o h e v a 3.5 W

Method (c) Method (b) Method (a)

l e v e l n a e m r e t a W

3.0 10.350

-42.950

Wave depth - z (m) Figure C.2: Wave horizontal velocity distribution, per elevation, for a phase angle of 0 deg.

The water column elevation was divided in 200 points, from where the drag and inertia distribution forces are calculated with the respective wave kinematics (horizontal velocity and acceleration), current velocity and hydrodynamic coefficients (drag and inertia). All these parameters vary with elevation due to the linear wave theory formulations, current data and marine growth thickness variation (affecting the object size and converting a smooth surface in rough). With the distribution of the hydrodynamic forces, per elevation, an integrated numerical technique, called the rectangle method, is used to calculated the maximum drag and inertia forces.

The hydrodynamic force function is divided into 200 subintervals and the average drag and inertia force for each subinterval is determined, as shown in Figure C.3, from where the base shear force and overturning moment can be determined.

Figure C.3: Rectangle method applied to calculate the maximum drag and inertia forces

Appendix C

- 152 -

This process is repeated for different phase angles

, for Airy’s linear wave theory formulation,

see appendix B.1.

The magnitude of the hydro dynamic effects can be measured by the high base shear force and overturning moment produce by the lateral loads, see Figure C.4.

Figure C.4: Base shear and overturning moment

To calculate the maximum base shear and overturning moment the hydrodynamic forces are calculated for a phase angle range between 0 to 360 degrees with a step of 5 degrees.

For the different phases angles the wave surface elevation will fluctuate, according to the expression in the linear wave theory, as well as the contribution of the drag and inertia force components for the total hydrodynamic force.

Table C.6 presents the results obtain for the base shear and overturning moment, per phase angle.

153 - -

Appendix C

Table C.6: Base shear and overturning moment calculation Position

Surf. elev.

TotalBaseShear D+C(kN)

F I (kN)

TotalOverturningMoment

F D(kN)

0.000

10.350

2244.5

3291.2

0.0

3291.2 70528.1

101092.0

0.0

101092.0

5.000

10.311

2224.3

3266.3

-101.4

3164.9 69827.4

100229.0

-3046.3

97182.7

10.000

10.193

2164.2

3192.2

-201.3

2990.9 67748.6

97662.5

-6032.9

91629.6

15.000

9.997

2068.7

3074.2

-298.4

2775.8 64495.8

93652.9

-8893.6

84759.3

20.000

9.726

1940.4

2915.0

-390.9

2524.1 60119.4

88224.7

-11575.9

76648.8

25.000

9.380

1784.4

2720.1

-478.0

2242.2 54843.0

81636.2

-14041.2

67595.0

30.000

8.963

1606.4

2495.7

-558.4

1937.3 48887.2

74129.2

-16243.2

57886.0

35.000

8.478

1414.6

2251.4

-630.4

1621.0 42571.0

66087.7

-18120.0

47967.6

40.000

7.929

1215.0

1993.6

-694.2

1299.4 36098.8

57730.9

-19685.6

38045.2

45.000

7.319

1014.4

1729.8

-748.9

980.9

29705.6

49325.1

-20916.9

28408.2

50.000

6.653

819.8

1468.2

-794.5

673.6

23627.9

41155.1

-21826.3

19328.8

55.000

5.937

637.1

1215.4

-830.5

384.9

18046.0

33433.2

-22406.0

11027.3

60.000

5.175

472.0

978.5

-857.0

121.5

13133.4

26391.3

-22638.2

3753.0

65.000

4.374

328.3

762.1

-873.1

-111.0

8960.6

20110.3

-22570.5

-2460.2

70.000

3.540

209.4

570.5

-879.4

-308.9

5601.7

14711.6

-22205.8

-7494.1

75.000

2.679

116.5

405.3

-878.1

-472.8

3050.1

10183.4

-21644.9

-11461.5

80.000

1.797

50.9

269.0

-867.8

-598.8

1305.5

6571.5

-20842.6

-14271.2

85.000

0.902

12.5

161.8

-851.2

-689.4

312.3

3824.8

-19880.5

-16055.7

90.000

0.000

0.0

83.1

-828.7

-745.6

0.0

1882.1

-18853.1

-16971.0

95.000

-0.902

-11.7

31.8

-802.9

-771.1

-278.3

671.0

-17828.7

-17157.6

100.000

-1.797

-44.7

5.4

-771.5

-766.1 -1039.5

98.2

-16715.5

-16617.2

105.000 110.000

-2.679 -3.540

-96.0 -162.0

-2.3 -17.8

-736.6 -696.8

-738.9 -2174.3 -714.6 -3576.6

-52.2 -422.2

-15557.9 -14358.1

-15610.0 -14780.3

115.000

-4.374

-239.5

-49.2

-654.0

-703.1 -5151.6

-1104.1

-13153.7

-14257.8

120.000

-5.175

-324.6

-92.7

-609.0

-701.7 -6811.3

-2002.6

-11948.1

-13950.7

125.000

-5.937

-414.3

-145.2

-561.3

-706.4 -8485.1

-3034.5

-10758.7

-13793.2

130.000

-6.653

-505.7

-203.3

-512.2

-715.5 -10117.1

-4130.6

-9603.4

-13734.0

135.000

-7.319

-595.7

-264.3

-462.0

-726.3 -11659.3

-5232.3

-8482.5

-13714.8

140.000

-7.929

-681.5

-325.2

-411.4

-736.5 -13072.5

-6291.0

-7397.0

-13688.0

145.000

-8.478

-761.9

-384.0

-359.9

-743.9 -14342.3

-7276.3

-6356.2

-13632.5

150.000

-8.963

-834.7

-438.8

-308.3

-747.1 -15454.5

-8163.2

-5359.5

-13522.8

155.000

-9.380

-898.3

-487.5

-256.6

-744.2 -16389.8

-8928.0

-4397.3

-13325.3

160.000

-9.726

-951.8

-529.2

-205.0

-734.2 -17155.7

-9564.8

-3472.4

-13037.2

165.000

-9.997

-994.0

-562.5

-153.8

-716.3 -17751.4

-10066.5

-2578.5

-12645.0

170.000

-10.193

-1024.9

-586.9

-102.5

-689.4 -18177.3

-10428.2

-1706.5

-12134.8

175.000

-10.311

-1043.6

-601.9

-51.2

-653.1 -18434.1

-10647.3

-849.7

-11496.9

180.000

-10.350

-1049.9

-606.9

0.0

-606.9 -18519.6

-10720.4

0.0

-10720.4

185.000

-10.311

-1043.6

-601.9

51.2

-550.7 -18434.1

-10647.3

849.7

-9797.6

190.000

-10.193

-1024.9

-586.9

102.5

-484.5 -18177.3

-10428.2

1706.5

-8721.7

195.000

-9.997

-994.0

-562.5

153.8

-408.6 -17751.4

-10066.5

2578.5

-7487.9

Appendix C

MD(kNm)

M

D+C(kNm)

M I (kNm)

M (kNm)

η(x,t) (m)

º

F

F (kN)

(wt - kx) ( )

- 154 -

Table (cont.): Base shear and overturning moment calculation Position

Surf. elev.

Total BaseShear D+C(kN)

F I (kN)

Total OverturningMoment

MD(kNm)

M

D+C(kNm)

M I (kNm)

M (kNm)

η(x,t) (m)

F D(kN)

200.000

-9.726

-951.8

-529.2

205.0

-324.1 -17155.7

-9564.8

3472.4

-6092.4

205.000

-9.380

-898.3

-487.5

256.6

-230.9 -16389.8

-8928.0

4397.3

-4530.6

210.000

-8.963

-834.7

-438.8

308.3

-130.4 -15454.5

-8163.2

5359.5

-2803.7

215.000

-8.478

-761.9

-384.0

359.9

-24.1

-14342.3

-7276.3

6356.2

-920.1

220.000

-7.929

-681.5

-325.2

411.4

86.2

-13072.5

-6291.0

7397.0

1106.0

225.000

-7.319

-595.7

-264.3

462.0

197.7 -11659.3

-5232.3

8482.5

3250.2

230.000

-6.653

-505.7

-203.3

512.2

308.8 -10117.1

-4130.6

9603.4

5472.8

235.000

-5.937

-414.3

-145.2

561.3

416.1

-8485.1

-3034.5

10758.7

7724.2

240.000

-5.175

-324.6

-92.7

609.0

516.3

-6811.3

-2002.6

11948.1

9945.5

245.000

-4.374

-239.5

-49.2

654.0

604.8

-5151.6

-1104.1

13153.7

12049.5

250.000

-3.540

-162.0

-17.8

696.8

678.9

-3576.6

-422.2

14358.1

13935.9

255.000

-2.679

-96.0

-2.3

736.6

734.3

-2174.3

-52.2

15557.9

15505.7

260.000

-1.797

-44.7

5.4

771.5

776.9 -1039.5

98.2

16715.5

16813.7

265.000

-0.902

-11.7

31.8

802.9

834.6

-278.3

671.0

17828.7

18499.7

270.000

0.000

0.0

83.1

828.7

911.8

.0 0

1882.1

18853.1

20735.2

275.000

0.902

12.5

161.8

851.2

1012.9

312.3

3824.8

19880.5

23705.4

280.000

1.797

50.9

269.0

867.8

1136.8

1305.5

6571.5

20842.6

27414.1

285.000

2.679

116.5

405.3

878.1

1283.3

3050.1

10183.4

21644.9

31828.2

290.000

3.540

209.4

570.5

879.4

1449.9

5601.7

14711.6

22205.8

36917.4

295.000

4.374

328.3

762.1

873.1

1635.1

8960.6

20110.3

22570.5

42680.8

300.000

5.175

472.0

978.5

857.0

1835.5 13133.4

26391.3

22638.2

49029.5

305.000 310.000

5.937 6.653

637.1 819.8

1215.4 1468.2

830.5 794.5

2045.9 18046.0 2262.7 23627.9

33433.2 41155.1

22406.0 21826.3

55839.2 62981.4

315.000

7.319

1014.4

1729.8

748.9

2478.7 29705.6

49325.1

20916.9

70242.0

320.000

7.929

1215.0

1993.6

694.2

2687.9 36098.8

57730.9

19685.6

77416.5

325.000

8.478

1414.6

2251.4

630.4

2881.8 42571.0

66087.7

18120.0

84207.7

330.000

8.963

1606.4

2495.7

558.4

3054.1 48887.2

74129.2

16243.2

90372.3

335.000

9.380

1784.4

2720.1

478.0

3198.1 54843.0

81636.2

14041.2

95677.4

340.000

9.726

1940.4

2915.0

390.9

3305.8 60119.4

88224.7

11575.9

99800.5

345.000

9.997

2068.7

3074.2

298.4

3372.5 64495.8

93652.9

8893.6

102546.5

350.000

10.193

2164.2

3192.2

201.3

3393.4 67748.6

97662.5

6032.9

103695.3

355.000

10.311

2224.3

3266.3

101.4

3367.7 69827.4

100229.0

3046.3

103275.4

º

F

F (kN)

(wt - kx) ( )

Where, (wt - kx) - is the phase angle (in deg.); (x,t) - is the wave surface elevation for the linear wave theory (in meters); FD FD+C FI F MD MD+C MI M

155 - -

- is the total drag force (in kilonewton); - is the total drag force with the current velocity sum to the wave velocity; - is the total inertia force (in kilonewton); - is the total base shear force (in kilonewton); - is the total overturning moment due to the drag force (in kilonewton meter); - is the total overturning moment due to the drag force + current; - is the total overturning moment due to the inertia force (in kilonewton meter); - is the total overturning moment (in kilonewton meter).

Appendix C

The base shear values obtain are compiled in Figure C.5.

3500 3000 2500

Drag Drag+Current

) 2000 N k ( s e rc 1500 o F r a e 1000 h S e sB 500 a e v a W 0

Inertia Total

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

320

340

360

-500 -1000 -1500

Wave Position [º] Figure C.5: Base shear distribution per phase angle

Appendix C

- 156 -

From Figure C.5 , it can be seen that the drag (without current) and inertia forces are symmetric in relation to an invisible vertical line drawn over the axis value 180 degrees and the drag and inertia maximum static values do not occur at the same time.

The results obtained were compared with a SACS static wave analysis [/10/] by means of two different wave kinematic theories: Airy´s linear wave theory and Stream Function 11

th

order theory. Table C.7

presents a comparison of the base shear force and overturning moment, per phase angle, for the linear wave calculations and the SACS linear wave and Stream Function calculations.

For the 360 degrees phase angle range, the worst wave position, for the structure, is chosen based on the maximum base shear force and/or overturning moment obtained (not always the two maximum values occur at the same wave position due to the shea r force arm lengt h ascension relative to maximum shear force).

The linear wave theory results between manual and SACS calculations have an average difference of 2%, what can be considered as acceptable results. Meanwhile, as expected, the Stream Function 11

th

order results show considerable higher peak values for the base shear force and overturning moment and smaller values for the low points. That is due to the “correction” made to the linear sinusoidal function of the wave surface equation with steeper crests and flatter troughs, as shown in Figure C.6.

In appendix B.3 it was proposed a selection of the adequate wave regular theory, according to API [/4/] and ISO [/6/] codes. For the survival condition the most appropriate wave was the Stream th

Function 7 order. Iv-Oil & Gas general practice is to use the Stream Function 11

th

order theory, that

for the F3-FA wave parameters presents practically the same results.

60 SACS linear wave SACS Stream Function 11th order

55

water mean level

) 50 m ( e c a rf 45 u S e v a W40

35 30 Figure C.6: Wave surface comparison for different theories

157 - -

Appendix C

Table C.7: Base shear and overturning moment comparison with SACS [bigger values marked in green] Position

Surf.elev.

TotalBaseShear

(wt - kx) ( )

η (x,t) (m)

0.000

10.350

3291.2

238.6 3

4266.8

101092.0

02456.7 1

153017.6

5.000

10.311

3164.9

086.1 3

3890.0

97182.7

7482.1 9

137510.1

10.000

10.193

2990.9

934.7 2

3527.8

91629.6

2565.1 9

15.000

9.997

2775.8

696.4 2

2952.1

84759.3

84600.3

99329.9

20.000

9.726

2524.1

463.9 2

2433.8

76648.8

76903.2

78935.3

25.000

9.380

2242.2

164.0 2

1866.1

67595.0

66935.7

57665.2

30.000

8.963

1937.3

877.1 1

1375.6

57886.0

57544.9

40216.7

35.000

8.478

1621.0

1548.5

923.3

47967.6

46891.0

25236.5

40.000

7.929

1299.4

1241.0

540.9

38045.2

37125.4

13329.8

45.000

7.319

980.9

18.1 9

225.7

28408.2

27051.6

4317.9

50.000

6.653

673.6

35.5

19328.8

18102.0

2639.7

55.000

5.937

384.9

228.8

11027.3

9649.0

7261.5

60.000

5.175

121.5

383.6

3753.0

420.8 2

10640.8

65.000

4.374

-111.0

143.0

481.9

-2460.2

796.2 3

12425.3

70.000

3.540

-308.9

335.8

553.4

-7494.1

833.7 8

13491.6

75.000

2.679

-472.8

490.9

587.2

-11461.5

12644.9

13771.6

80.000

1.797

-598.8

613.7

602.6

-14271.2

15434.6

13640.2

85.000

0.902

-689.4

695.8

589.7

-16055.7

17042.0

12953.2

90.000

0.000

-745.6

749.4

564.3

-16971.0

17853.3

12006.9

95.000 100.000

-0.902 -1.797

-771.1 -766.1

770.8 770.1

527.4 491.7

-17157.6 -16617.2

17945.6 17514.9

10964.7 10029.9

105.000

-2.679

-738.9

739.7

460.7

-15610.0

16441.9

9284.1

110.000

-3.540

-714.6

715.0

437.2

-14780.3

15578.7

8700.4

115.000

-4.374

-703.1

696.7

417.7

-14257.8

14901.3

8236.6

120.000

-5.175

-701.7

693.0

401.9

-13950.7

14547.8

7867.1

125.000

-5.937

-706.4

689.7

388.7

-13793.2

14232.4

7557.8

130.000

-6.653

-715.5

696.7

378.4

-13734.0

14125.5

7306.1

135.000

-7.319

-726.3

698.4

367.9

-13714.8

13939.2

7070.6

140.000

-7.929

-736.5

707.1

359.1

-13688.0

13882.4

6868.2

145.000

-8.478

-743.9

705.4

348.8

-13632.5

13665.9

6653.6

150.000

-8.963

-747.1

708.1

339.5

-13522.8

13528.9

6457.2

155.000

-9.380

-744.2

696.7

327.6

-13325.3

13182.5

6227.4

160.000

-9.726

-734.2

687.9

316.2

-13037.2

12887.0

6006.7

165.000

-9.997

662.7

301.3

5734.6

-10.193

638.7

286.7

-12645.0 -12134.8

12352.6

170.000

-716.3 -689.4

11847.8

5466.1

175.000

-10.311

-653.1

596.9

267.7

-11496.9

11083.5

5128.5

180.000

-10.350

-606.9

555.4

248.7

-10720.4

10329.9

4791.6

185.000

-10.311

-550.7

495.8

224.5

-9797.6

9300.6

4367.0

190.000

-10.193

-484.5

435.9

200.3

-8721.7

8261.4

3942.0

º

FSACS LINEAR (kN)


FMANUAL (kN)

623.3 37.2 3 3.7 8

FSACS STF 11 (kN)

Appendix C

MMANUAL (kNm)

MSACS LINEAR (kNm)

MSACS STF 11 (kNm)

122837.9

- 158 -

Table (cont.): Base shear and overturning moment comparison with SACS [bigger values marked in green] Position

Surf.elev.

TotalBaseShear


(wt - kx) ( )

η (x,t) (m)

FMANUAL (kN)

195.000

-9.997

-408.6

358.8

170.1

-7487.9

6936.5

3408.9

200.000

-9.726

-324.1

281.2

139.8

-6092.4

5580.4

2875.7

205.000

-9.380

-230.9

188.6

102.7

-4530.6

3939.7

2212.6

210.000

-8.963

-130.4

95.7

65.7

-2803.7

2251.2

1551.0

215.000

-8.478

-24.1

8.3

-920.1

302.5

737.9

220.000

-7.929

86.2

111.5

22.7

1106.0

1697.9

70.1

225.000

-7.319

197.7

220.1

74.6

3250.2

3892.8

1048.3

230.000

-6.653

308.8

325.7

125.2

5472.8

6114.6

2015.2

235.000

-5.937

416.1

428.9

183.2

7724.2

8406.0

3161.1

240.000

-5.175

516.3

525.6

239.1

9945.5

10659.0

4285.2

245.000

-4.374

604.8

610.7

299.7

12049.5

12787.9

5574.4

250.000

-3.540

678.9

684.6

357.8

13935.9

14757.8

6826.0

255.000

-2.679

734.3

736.7

417.7

15505.7

16340.1

8197.7

260.000

-1.797

776.9

782.5

472.2

16813.7

17728.2

9497.5

265.000

-0.902

834.6

835.7

524.6

18499.7

19311.5

10835.3

270.000

0.000

911.8

914.6

579.1

20735.2

21613.9

12200.4

275.000

0.902

1012.9

1013.2

642.1

23705.4

24626.8

13798.7

280.000

1.797

1136.8

1140.4

719.9

27414.1

28491.2

15767.2

285.000

2.679

1283.3

1279.0

815.0

31828.2

32805.8

18286.0

290.000 295.000

3.540 4.374

1449.9 1635.1

1448.4 622.1 1

934.8 1083.0

36917.4 42680.8

38114.4 43697.3

21510.4 25797.5

300.000

5.175

1835.5

826.4 1

1265.0

49029.5

50339.2

31160.5

305.000

5.937

2045.9

022.1 2

1473.8

55839.2

56876.8

37693.2

310.000

6.653

2262.7

245.8 2

1727.1

62981.4

64445.2

45818.3

315.000

7.319

2478.7

442.9 2

2002.6

70242.0

71302.4

55243.1

320.000

7.929

2687.9

662.7 2

2333.5

77416.5

79035.5

66893.5

325.000

8.478

2881.8

834.3 2

2665.3

84207.7

85272.5

79461.8

330.000

8.963

3054.1

021.6 3

3059.8

90372.3

92136.3

94859.8

335.000

9.380

3198.1

139.9 3

3402.5

95677.4

6699.6 9

109368.3

340.000

9.726

3305.8

266.9 3

3803.4

99800.5

01609.3 1

126798.6

345.000

9.997

3372.5

306.3 3

4046.6

102546.5

03469.6 1

138619.2

350.000

10.193

3393.4

348.6 3

4320.4

103695.3

05439.4 1

152114.0

355.000

10.311

3367.7

293.5 3

4293.4

103275.4

03947.5 1

152553.5

º

FSACS LINEAR (kN)

FSACS STF 11 (kN)

21.2

MMANUAL (kNm)

MSACS LINEAR (kNm)

MSACS STF 11 (kNm)

Where, (wt - kx) (x,t) FMANUAL FSACS LINEAR FSACS STF 11 MMANUAL MSACS LINEAR MSACS STF 11 159 - -

- is the phase angle (in deg.); - is the wave surface elevation for the linear wave theory (in meters); - is the total base shear calculated based on the linear wave theory (kN); - is the total base shear SACS linear wave theory (kN); th - is the total base shear SACS Stream Function 11 order (kN); - is the total overturning moment based on the linear wave theory (kNm); - is the total overturning moment SACS linear wave theory (kNm); th - is the total overturning moment SACS Stream Function 11 order (kNm). Appendix C

The base shear values from the manual and SACS calculations are also compared in Figure C.7.

5000

4000

) N 3000 k ( s e rc o F r a e 2000 h S se a B e v a 1000 W

Linear wave calculations SACS linear wave SACS Stream Function 11th order

0 0

-1000

60

120

180

240

300

360

Wave Position [º] Figure C.7: Base shear distribution, per phase angle, comparison between linear wave calculations and SACS results

Appendix C

- 160 -

The maximum base shear and overturning moment results for the F3-FA single leg are compared, in Table C.8, together with two other models presented in

Figure C.8 : F3-FA leg plus transition frame

and F3-FA complete structure.

Table C.8: Base shear (F) and overturning moment (M) comparison with different SACS models

F3-FA leg plus transition frame

F3-FA single leg Linear wave theory Manual

SACS

SACS Stream Function TH

SACS Stream Function TH

F(kN)

calculations 3393.4

3348.6

11 order 4320.4

11 order 6823.3

M (kNm)

103695.3

105439.4

153017.6

168990.0

F3-FA platform Static analysis

Dynamic analysis

SACS Stream Function

SACS Stream Function

TH

11 order 20249.8 583163.4

DAF

TH

11 order 22202.6 1.096 700043.9

1.200

The results demonstrate the great effect of the transition frame (knuckle joint) on the fluid forces applied to the structure (comparison between the F3-FA leg alone and the F3-FA leg plus transition frame), due to an increase on the substructure area submitted to the hydrodynamic pressure.

Meanwhile, for the entire F3-FA structure (group of 4 legs) a dynamic wave analysis, according to section 3.3.4.2, is considered. Based on the static and dynamic results a dynamic amplification factor (DAF) can be determined.

161 - -

Appendix C

Figure C.8: SACS Models studied for the calculation of the hydrodynamic base shear force and overturning moment

If we divided the total base shear force on the F3-FA structure by four (the number of legs), we came to the conclusion that the hydrodynamic loads are lower than for the single F3-FA leg plus transition frame; due to the fact that the different orientation and inclination of the four transition frame members will traduce in diverse horizontal loadings but also in vertical loadings.

The dynamic amplification factor (DAF) is commonly reflected, in the offshore industry, on the base shear force and overturning moment for a fixed platform. Thus, the dynamic amplification factors are determined by the ratio between the static and the dynamic shear force and overturning moment. ISO [/6/] proposes a simplified method to calculate the DAF to give insight on the structural behaviour of the structure: (C.1) Where,

is the natural period of the structure of interest; is the wave frequency; is the critical damping ratio associated, considered as 0,02 (2%).

This expression can be in practice simplified to

and, thus, we can say it is independent

from factor .

The two first mode shapes of the F3-FA structure are characterized by a sway mode shape with single curvature in the two translational directions of the portal frame and have the following natural periods: 2,731 and 2,702 seconds.

The wave period for the survival condition is 15,946 seconds (with the current effect).

Thus, given a maximum

of 0,171 and, consequently a DAF of 1,03.

In the other hand, the SACS results for the DAF, presented in

Table C.8, are respectively 1,10 and

1,20 for the base shear and the overturning moment. The expression given in ISO [/6/] code is commonly used in offshore jacket structures, but should be checked for validity in more non-traditional fixed platforms, as in the case of the F3-FA platform.

In the F3-FA project, the SACS dynamic wave analysis method, prescribed in the API [/4/] and referred in section 3.3.4.2, is used and the DAF factors are considered adequate.

Appendix C

- 162 -

D. API buckling and stress ch eck of cylindr ical member s

163 - -

Appendix D

Appendix D

- 164 -

165 - -

Appendix D

Last page

Appendix D

- 166 -

Tese Jpm Final 20062014

Recommend Documents