Use of Bayesian Network for Risk-Based Fatigue Integrity Assessment: Application for Topside Piping in an Arctic Environment

摘要

In the North Sea, hydrocarbon release (HCR) from offshore topside piping that is inherently degrading because of vibration-induced fatigue (VIF) has been a significant challenge. This challenge is further exacerbated when the operating assets are located in Arctic environments, especially because of higher temperature gradients and other operational challenges. The result is a significant increase of VIF failure of topside piping, which causes a higher risk of potential HCR in Arctic environments. The "zero oil spill" requirement in the Arctic region requires operators to follow stringent guidelines to mitigate the risk of VIF failure of topside piping. In this regard, this paper demonstrates the application of the Bayesian network (BN) for risk-based fatigue integrity assessment of the topside piping. The value of the calculated risk depends upon the values of likelihood of failure (LoF) and consequence of failure (CoF). The value of LoF in turn depends upon the measured vibrational velocity (coupled with the vibration assessment criteria of the Energy Institute guidelines) and probabilistically estimated remnant fatigue life (RFL) of topside piping. Likewise, the CoF has been assessed from safety, environmental, and business perspectives. The BNs have been modelled using GeNIe and have been utilized to estimate the risk. The outcome of the fatigue integrity assessment using the proposed approach is the categorization of piping into various risk groupings. Piping in the high-risk category is thereby prioritized for inspection, thus preventing HCR in the fragile Arctic region. An illustrative case study, demonstrating the usability of the proposed approach, is presented.