Improved compression inference algorithm for reliability analysis of complex multistate satellite system based on multilevel Bayesian network

Bayesian Network (BN) is a powerful tool for analyzing system reliability. However, for the complex multistate satellite system, the state combination explosion makes the reliability analysis based on BN computationally unaffordable. To solve this problem, compression algorithm and inference algorithm have been proposed. The original compression algorithm classifies all the phrases with same composition into the same type. The original compression algorithm classifies all the phrases with same composition into the same type. However, cases tested in this paper show that the results of inference can be incorrect when some node probability tables have special formulations. To solve the above defects, the compression algorithm and sequential inference algorithm are improved in this paper. Based on improved compression algorithm and improved inference algorithm, the improved compression inference algorithm (ICIA) is formed and extended to the application of multistate nodes with independent binary parent nodes. Besides, the multilevel BN (MBN) models are used to construct the BN models of complex multistate satellite systems so as to further improve inference efficiency. Cases tested in this paper show that the proposed algorithms are more efficient than the Bayes Net Toolbox and the AgenaRisk software for the reliability analysis of complex multistate satellite systems.