Reliability analysis of systems with mixed uncertainties based on Bhattacharyya distance and Bayesian network

Abstract

The lack of data due to the high cost of experiments, limited knowledge of system structure, and coupling failure mechanism make it necessary to consider uncertainties when analyzing the reliability of complex systems. This paper provides a comprehensive work for system reliability analysis under consideration of mixed uncertainties. First, Bayesian networks (BNs) are used to model the structure of complex systems. Second, mixed uncertainties are described via a probability box and accurately quantified by the Bhattacharyya distance, which can synthetically represent the mean and variance of two data sets. Then, the importance of components with mixed uncertainties is defined and analyzed. Finally, a comprehensive sensitivity index is proposed for sensitivity analysis and epistemic uncertainty tracing of data sources based on the Bhattacharyya distance. This method is used to analyze the reliability of an auxiliary power supply system of a train and verifies the feasibility and accuracy of the proposed method.

Evidential network‐based system reliability assessment by fusing multi‐source evidential information

“…In other words, the developed methods can only handle whether aleatory uncertainty of the collected data or epistemic uncertainty of the elicited knowledge. However, aleatory and epistemic uncertainties oftentimes coexist and couple together, often called hybrid uncertainty or mixed uncertainty, in engineering practice 18,19 . For instance, the mean time to failure (MTTF) of a population of systems is stochastic due to the random degradation profiles of these systems.…”

“…However, aleatory and epistemic uncertainties oftentimes coexist and couple together, often called hybrid uncertainty or mixed uncertainty, in engineering practice. 18,19 For instance, the mean time to failure (MTTF) of a population of systems is stochastic due to the random degradation profiles of these systems. However, the estimation of the MTTF of these systems may be imprecise due to limited time-to-failure data and data uncertainty.…”

A reliability analysis technique for complex systems with retaining fuzzy information

“…Numerous reliability analysis models have been developed to address the challenges posed by intricate failure behaviors in safety‐critical systems. Some widely adopted models include Fault Tree (FT), 2 Dynamic Fault Tree (DFT), 3 Reliability Block Diagram (RBD), 4 Stochastic Petri Nets (SPN), 5 Bayesian Networks (BN), 6 and Dynamic Bayesian Networks (DBN) 7 . Fault Tree Analysis (FTA) focuses on the identification and quantification of failure events that may lead to system failures 8 .…”