Monte-Carlo Simulation for the Reliability Analysis of Multi-status Network System Based on Breadth First Search

In general engineering network system, the transformation of the working state, from safety to failure, of unit and system is gradual. And there is an intermediate state between the states of being completely safety and completely failing. On the basis of "safety-intermediate-failure" three-stage working mode, this paper divides it into two-stage mode: "safety-non safety" and "non failure-failure", and realizes the calculation of connective reliability for network system by traditional disjoint algorithm. The concept of reliability vector is put forward, which includes in three indexes such as probability of safety, probability of intermediate and probability of failure for engineering structures. In order to solve the non-polynomial increase hard problem of reliability calculation for super-large network system with intermediate state, according to the breadth first search technology of graph theory, an algorithm for estimate of the network connectivity is presented to evaluate reliability of network by Monte Carlo simulation. The analysis of an example in a certain network system demonstrates the effectiveness and applicability of the established algorithm.