A Bayesian Network Model for Risk Analysis of Deepwater Drilling Riser Fracture Failure

Yuanjiang Chang,Changshuai Zhang,Xiangfei Wu,Jihao Shi,Guoming Chen,Jihua Ye,Liangbin Xu,Anti Xue
DOI: https://doi.org/10.1016/j.oceaneng.2019.04.023
IF: 5
2019-01-01
Ocean Engineering
Abstract:Drilling risers are crucial connection of subsea wellhead and floating drilling platform. Fracture failure of deepwater drilling risers is the most serious accident in offshore drilling, which would lead to disastrous consequences. This paper presents a Bayesian Network (BN) model to conduct risk analysis for fracture failure of deepwater drilling riser. A Bow-Tie (BT) model is developed to identify the risk factors associated with fracture failure and the potential consequences. Subsequently, evidence theory is used to calculate the prior probability of the input event, and the developed BT model is mapped into BN to carry out risk analysis of fracture failure of drilling riser. Finally, the probability updating is implemented using forward reasoning in BN model when new evidence is obtained, and a dynamic risk profile of fracture failure and consequence status are performed using probability adaption of BN with the occurrence of the new identified critical factors over a period of time. The application of the developed model is demonstrated through a case study, and some suggestions drawn from the investigation are presented to further mitigate the risk and the severity of accident consequences of drilling riser fracture failure during drilling operations.
What problem does this paper attempt to address?