Reliability Assessment of Double-Sided Linear Switched Reluctance Generator System Based on Hierarchical Markov Model

Abstract

This paper presents a hierarchical Markov model to assess the reliability of a double-sided linear switched reluctance generator (DLSRG) system. The proposed reliability model can be applied to find the balance point of reliability assessment accuracy and model complexity. First, the dominant fault modes of the DLSRG system are highlighted to quantitatively obtain the operation states, survival or failure, with a novel failure criterion. Second, a hierarchical Markov model is adopted to build a systematic reliability model of the DLSRG system, wherein a state transition diagram can be built in the detailed description to constitute the graphical Markov reliability model. And then the solving process to get state probability matrix P ( t ) is intensively investigated. Notably, a reliability block diagram model is also built to compare with the hierarchical Markov model, whose results verify that the Markov model can capture the effect of fault-tolerance capability on system reliability. Moreover, it is expected that the one-level Markov model is the most suitable model for the DLSRG system for its ability of decreasing the model complexity and improving the assessment accuracy. Analytic investigation in conjunction with simulation and experimental measurements demonstrates the feasibility and validity of the proposed model.