System Reliability Analysis Considering Time-Dependent Correlation of Failure Modes Based on Mixed Copula
Accurate quantification of failure modes is the premise of system reliability analysis. However, diversity and time evolution of failure modes remain obstacles to the improvement of system reliability. To address this issue, this paper proposes a novel adaptive time-dependent mixed copula framework, and a two-stage system reliability analysis framework is built. First, to overcome the limitation of traditional copula in capturing asymmetric tail dependence, a rotation transformation technique is introduced to construct a mixed copula model comprising 0°, 90°, and 180° rotated components. Second, to quantify the evolution of time-dependent correlation of failure modes, a time-dependent parameter estimation strategy based on Bayesian Markov chain Monte Carlo (MCMC) inference and sliding window is proposed. Furthermore, to suppress the random noise in the estimation process, a Kalman filter is applied to adaptively smooth the time-dependent parameter trajectory. Finally, a system reliability analysis framework based on the time-dependent mixed copula model is presented. The results show that the proposed framework can provide accurate system reliability estimates while accounting for time-dependent dependence among failure modes.
Authors
- Zhonglai Wang (ORCID: https://orcid.org/0000-0002-1708-4197)
- Pengpeng Zhi (ORCID: https://orcid.org/0000-0003-1537-8455)
- Suiping Tang
- Hu Wan
Institutions
- University of Electronic Science and Technology of China (CN)
Publication Details
- Journal
- ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems Part A Civil Engineering
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1061/ajrua6.rueng-2109
- Primary Topic
- Reliability and Maintenance Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00