Reliability assessment of competing failure processes with correlated multi-performance characteristics: A case study of subsea Christmas tree hydraulic power unit

Complex systems typically involve multiple performance characteristics (PCs), each reflecting the failure status of the system over time. The interactions among different PCs significantly impact the accuracy of system reliability assessments. To investigate the dynamic variations in degradation parameters in failure models under the influence of external factors and to explore the impact of interactions between different PCs on reliability during competing failure processes, this paper establishes a competing failure reliability model based on the correlation of multiple PCs. A degradation failure parameter updating method is proposed using an extended Kalman particle filtering (EKPF) algorithm. Additionally, the D-vine copula function is employed to describe the correlation between different PCs under competing failure conditions. Taking the hydraulic power unit (HPU) of subsea Christmas trees as an example, a competing failure model considering the correlations of PCs is obtained. Sensitivity analysis of the degradation parameters of external environmental conditions is conducted based on the updating of the degradation failure parameters. The study investigates the correlations of PCs and the impact of different shock intensities on system reliability. Considering these correlations, the model is extended to three different shock modes: extreme shock, delta-impulse, and m-impulse.