GE Oil & Gas Impact of Input Parameters on Wellhead Fatigue, and Alternative Solutions for Life Extension Hussain Hashemizadeh Vidar Strand 6 th December 2012 LEADING PROGRESS TOGETHER
Introduction The subsea wellhead is the main structural component which supports the loads generated during drilling and production operations. Whilst drilling or work-over is taking place the wellhead is the main support item in the completed well. The wellhead system is also part of the well barrier, both in itself and as the main anchor point for the BOP. The primary environment loading on the wellhead occurs during the drilling phase. The varying load is mainly caused by sea current, wave loads and drilling vessel movements. The static load is due to subsea stack-up weight and casing program. Varying bending moment induced by environmental loading is transmitted through the riser onto the wellhead housing and then to the conductor housing/casing/formation. 2
Impact of Input Parameters on Wellhead Fatigue Use of accurate input data in the fatigue assessment is essential Conservative parameter assumptions need to made if accurate input data is not available. This can often lead to significant over-predictions of fatigue damage. Fatigue damage calculation is sensitive to input data. Input data includes but is not limited to following: o o o o o o o Soil/Foundation, Cement Riser Stack-up Vessel Parameters & Motion Response Operators (RAO) Environmental Data (Wave & Current Scatter Diagram) Wellhead System Configuration (Location of welds, connectors etc.) Welds Details SCF and Appropriate S-N Curve 3
Soil/Foundation A family of P-Y curves will be required to model the conductor casing/soil interaction at various depths below mudline. Data provided by operators or 3rd party; two options: o o P-Y curves for different soil layers (Preferable) Generic soil strength data. Constructed P-Y curves based on the recommendations given in API RP 2A-WSD If there is a lack of site-specific soils data, generic assumptions may underestimate the soil s beneficial effect on fatigue performance. 4
Effect of Appropriate Environmental Data Annual Wave scatter diagram of significant wave height (Hs) and spectral peak period (Tp) Spectral peak period (Tp) (s) HS (m) 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 21 22 22 23 Sum 17 18 0 16 17 1 2 3 15 16 3 3 14 15 4 6 10 13 14 4 7 13 24 12 13 2 8 23 12 45 11 12 7 31 37 26 Seastates 101 10 11 Seastates 3 29 108 82 36 Generating >12' 258 9 10 Generating <12' 4 13 131 229 134 44 Heave 555 8 9 Heave 6 14 115 402 386 150 58 1131 7 8 2 22 108 606 798 608 309 103 2 2558 6 7 7 76 192 684 1416 1303 948 371 126 5 5128 5 6 12 144 443 965 1961 2116 2177 1464 463 180 9 3 9937 4 5 36 282 955 1709 2496 2902 3667 3264 1651 446 97 22 17527 3 4 6 778 1589 2379 2974 3920 5646 6171 3516 1195 270 55 19 3 28521 2 3 6 607 3997 2961 3692 5716 8310 8504 5049 1850 632 173 59 17 5 41578 1 2 18 680 2705 4238 5317 8485 8483 6216 3057 1237 535 230 82 29 12 41324 0 1 49 187 322 1306 938 709 331 171 86 72 25 15 2 1 4214 Sum 67 873 3640 10355 11099 16371 19734 22298 22966 20465 14039 7509 0 2554 0 850 86 0 11 0 152917 Annual Wave Scatter Diagram (with Drilling Riser Heave Limit) An initial assumption would be to apply the full annual scatter diagram A more considered approach may be to apply specific seasonal or monthly weather based on planned operations. Considered connection limit and directionality. Hs Tp Wave Scatter Diagram; July Spectral peak period (Tp) (s) HS (m) 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16 17 17 18 18 19 19 20 20 21 21 22 22 23 Sum 17 18 0 16 17 0 15 16 0 Seastates 14 15 0 Generating >12' 13 14 Heave 0 12 13 0 11 12 0 10 11 Seastates 0 9 10 Generating <12' 0 8 9 Heave 0 7 8 0 6 7 1 1 3 5 5 6 1 3 17 11 8 40 4 5 1 39 58 47 28 17 190 3 4 8 44 112 209 163 146 72 42 4 800 2 3 60 356 387 492 695 560 390 119 19 8 2 3088 1 2 3 199 646 1007 1286 1899 1582 667 219 42 14 5 3 7572 0 1 13 54 110 462 266 166 78 25 13 9 2 3 1201 Individual Month Wave Scatter Diagram (with Drilling Riser Heave Limit) 5
Effect of Subsea Stack-up Height & Tension The trend towards bigger/heaver subsea stack potentially has an adverse effect on wellhead fatigue. Larger stack combined with the higher tension results in the largest bending moment range. 6
Effect of Vessel RAO & Subsea Stack-up In this example difference in bending moment response due to vessel RAO and subsea stack-up is illustrated The difference in system response in this case is largely due to vessel RAO and Subsea Stack-Up size 7
Effect of Annular Cement Level The cement top level within the conductor/surface casing annulus can effect fatigue life. In this example it can be seen that the limiting component with respect to minimum life changes as the assumed cement level is lowered. 8
Possible Fatigue Life Improvement Housing Welds are often Fatigue Critical. Enhanced welding can lead to better S-N curves and reduced SCF resulting in dramatic Fatigue Life Improvement 9
Summary/Conclusion Following actions can improve the fatigue life estimates: Drilling in better weather based on planned operation. Include directional distribution of the environment. Consider planned riser disconnection limits (Heave limit). Record the actual scatter diagram based on measured wave data during the operation. Input Parameter Impact Level of Conservatism Cement High High Soil Low Medium Scour Low Low Environmental High Medium Conditions Subsea Stack High Low Vessel Motion High Low (RAO) Casing Load Low Medium Weld Spec. High Low Template Stiffness Low Low 10
Wellhead Interface Base Design Heavy Duty Lock Down Latch Heavy Duty Anti- Rotation 2 off landing tapers removing clearance 11
Wellhead to extension transition Geometrical Improvements 25% reduction in stress level 12
Optimized housing extensions Bigger OD Bigger wall thickness Higher yield material C1 welds oversize and machine Coating weld seems Applicable for both LP & HP housings Low Pressure Housing 914 mm C1 Welds w/machined surfaces Extension sub w/higher yields and oversized wall thickness Example from BP Skarv 13
Suction Can for side support NeoDrill Reduces movement and bending moment in conductor Other benefits Increases axial load bearing Allow for shorter conductor Pre-conduct option FE model stress level in conductor/can interface at 1.5 x design load Running cunductor with gimbal 14 14
One piece conductor 32m long Welded connections only Removes conductor connector from stress peak area about 10 m below sea bed Other benefits Optimized for driving and pre-conduct No need for tongues 15
The optimized WH system 36 DW Wellhead w/ms-700 landing tapers and heavy duty lock down Anti rotation between housings Optimized housing to conductor transections 36 x 2 Heavy wall conductor, 18m long w/x65 X80 yield 21 x 1.25 Heavy wall housing extension w/x65 X80 yield High fatigue connectors C1 welds oversized and machined Coating on weld seems and general corrosion protection 16
Thank you for listening! 17
LEADING PROGRESS TOGETHER 18