Fault Sampling of Complex System Under Imperfect Maintenance

Abstract

When conducting simulation for evaluating complex system reliability or availability, stochastic fault sample method is applied to simulate fault occasions of complex system. Current research of fault sample method generally assumes maintenance activity restores systems to â??good-as-newâ? and â??bad-as-oldâ?, however, system usually keeps its function through imperfect maintenance that restores system to â??partly goodâ? in reality. A stochastic fault sample method for availability evaluation for complex system under imperfect maintenance based on monte-carlo simulation principle is proposed. Firstly, single component fault occurrence process subjected to imperfect maintenance is analyzed and the formula for fault occurrence occasion is given. Secondly, fault sample becomes complicated when maintenance object converts to system with complex reliability structure. Not only fault occurrence occasions but also fault components in the system should be derived from the sample method. Fault sample for parallel system is especially difficult, so the Markov state transition process is embedded into the method to solve this problem .Finally, a numerical case using fault sampling method under imperfect maintenance is given and the number of faults and their occurrence time are obtained . Validity and feasibility of the proposed method is verified.