Quantitative analysis of Dynamic Fault Tree by probabilistic approach
Abstract
Dynamic block diagram and dynamic fault tree are introduced to take into account the dynamic behavior of the system and failure mechanisms. The purpose of this paper is to present a methodology for safety analysis of complex systems and the evaluation their reliability. Therefore the system is at first, modeled by Dynamic Reliability Bloc Diagram (DRBD, introduced in the new version of IEC Std. 61078). Then an automatic conversion into Dynamic Fault Tree is applied. The probability of the top event of the dynamic tree is evaluated through a Monte Carlo simulation-based approach. The proposed algorithm gives the top event probability as well as the probabilities of all intermediate nodes in the DFT parameterized by the failure rates of its primary events, the mission time, the half-width of confidence interval and the accepted error rate. Another aspect of this paper relates more particularly, the influence of the primary entrance to the PAND gate that allows assessing the probability of the top event.