Reliability analysis of subsea control module based on dynamic Bayesian network and digital twin

The reliability evaluation of the subsea control module(SCM) is the key to ensure the safety and stability of subsea oil-gas production. The failure probability and reliability of SCM components are dependent on time and working conditions. To analyze the SCMs' reliability considering varying working conditions, this paper proposed a new digital twin and dynamic Bayesian network(DBN) based model utilizing historical working condition data in reliability analysis. In the proposed framework, critical working condition data is obtained by sensor-based Digital Twin(DT) simulation and used for dynamically updating the parameters in the DBN reliability analysis model. The reliability evaluation of an actual SCM electric system was carried out. The results revealed the most probable failure mode and the most vulnerable components in the system. Finally, the fault prediction based on the back-forward analysis capacity of the proposed method was conducted to predict the probability of the faulty device when unexpected failures occur.