Dynamic failure assessment of well barrier system of underground gas storage using object-oriented Bayesian network

Abstract

With the sustained growth in natural gas demand, the construction scale of UGS facilities has been progressively expanded. As critical infrastructure in UGS, the failure of well barrier system may result in natural gas release or blowout incidents when compromised. In this study, a methodology for dynamic failure assessment of WBS in UGS wells is developed using OOBN. The root causes of failure are identified based on the operation mechanisms of each well barrier component, and failure scenarios for subcomponents are constructed using BN. These are integrated into OOBN framework for failure of the entire WBS. The critical contributing factors to WBS failure are subsequently determined. By utilizing monitoring data, the occurrence probability of WBS failure is updated continuously. The applicability of the methodology is illustrated through a case study, and it is observed that the model can be a useful tool for integrity management of WBS in UGS.