In many harsh environment or high current regions (e.g. West of Shetlands, East Africa and GoM) wellhead fatigue during drilling or workover activities can be a major concern. As a result extensive wellhead and conductor fatigue assessments are required in order to predict likely fatigue damage prior to landing the BOP on to the wellhead. These pre-drilling or predictive studies are based on a number of assumptions regarding actual environmental and soil conditions. In addition the uncertainty associated with the input data requires safety factors of 10 and 20 for wave and VIV effects respectively. As a result of these assumptions the predicted levels of fatigue damage to the wellhead may be highly conservative. In cases where wellhead and conductor fatigue life is a concern the operator may choose to deploy motion monitoring systems on the BOP stack so that actual fatigue loads recorded during drilling can be calculated. These systems can be configured to provide ‘real-time’ updates on fatigue damage accumulation or can be used to record data for ‘post-drilling’ assessment. In each case the methodology applied to calculate stresses at fatigue hotspots is critical to the calculation of overall fatigue damage.
The calculation of fatigue loads based on measured data is typically carried out by applying the measured BOP motions to a global finite element model of the drilling riser and wellhead system. However due to the large amounts of data involved this can be a very time consuming exercise and is not conducive to ‘real-time’ presentation of results (e.g. for on-board systems). A proposed solution to this problem is to develop stress transfer functions (STFs) that relate BOP motion to stresses at critical fatigue hotspots. Thus ‘real-time’ BOP motions and be instantly converted to ‘real-time’ stresses. However, as is outlined in this paper, the frequency content of the system response can have a significant impact on the levels of stress calculated. If frequency dependent response is not accounted for a significant under-prediction in fatigue loads may occur.
The objective of this paper is to outline a detailed methodology to derive STFs for critical fatigue hotspots in the wellhead and conductor. This methodology accounts for capturing the non-linear frequency dependence of the system and incorporating this into the STFs. In addition a methodology for rapidly calculating ‘real-time’ fatigue damage accumulation in the wellhead and conductor system and presenting this data on-board is also outlined.
