Fatigue Assessment of Subsea Wells for Future and Historical Operations Based on Measured Riser Loads

Abstract

Operators in the North Sea have recently strengthened their efforts in documenting the integrity of subsea wellhead systems. As a part of this effort, fatigue damage estimation of subsea wells in service has been performed. Fatigue damage estimation on subsea wells due to drilling riser dynamic loads was carried out by the use of analytical model results. The applied analytical methodology is based on a decoupled approach, where global load analyses and local stress calculations are carried out prior to a SN based fatigue accumulation. Applying such methodology on safety critical systems the analytical philosophy should ensure conservative fatigue damage. For cases where the fatigue calculations returned unfavorable estimates, one corrective action has been to measure the actual riser response and to monitor the development of fatigue damage closely. For this purpose a methodology for fatigue estimation based on measured riser response was needed. In this approach of estimating the fatigue damage, the global load analysis results are replaced by measured dynamic load time series. By combining direct riser response measurements with local stress calculations, a revised SN based fatigue accumulation can be performed. The fatigue damage derived from measured riser response is compared to the fatigue damage based only on analytical results. From this comparison the conservatism in the analysis methodology for the global riser response is shown to be significant. As this method relays on measurements, it will only yield historical fatigue damage and at best it can return updated fatigue capacity usage on the fly. Forecasting fatigue damage still have to be established based on global riser analyses results, resulting in a conservative forecast. This paper suggests an updated methodology using actual measured response to both asses fatigue damages of historical operations and forecast fatigue damages based on historic operations. By cycle counts of measured response time series (one hour response) a link between this cycle count and the coexistent significant wave height and spectral peak period can be established. This relationship between observed weather and measured response is representative for the rig and riser system on which the measurements were performed. Then forecast and measurements of the weather conditions can be used to estimate the historical damage and the future fatigue damage respectively. The paper will present results from the suggested approach by use of examples from a real North Sea well in shallow water.