BINARY DECISION DIAGRAMS APPLIED TO FAULT TREE ANALYSIS

摘要

Fault Tree Analysis (FTA) is employed in Remote Condition Monitoring for detecting and identifying faults. FTA model is a graphical representation of logical relationships between events. In this paper a Binary Decision Diagram (BDD) approach is used in order to provide an alternative to the traditional cutest-based methods for FTA. BDD could be defined as a method which determines the output value of the function by examining the inputs. The BDD method does not analyse the fault tree directly, but converts the tree to a BDD that represents the Boolean equation for the top event. The conversion process from FTA to BDD presents several problems. The variable ordering scheme chosen for the construction of the BDD and has a crucial effect on its resulting size. In order to facilitate improvement efforts and resource deployment in an existing system, proper and accurate prioritisation of the basic events is necessary. This paper evaluates different methods. They are AND criterion, Structural Heuristic, Weigh, Top-Down Left-Right, Fussell-Vesley, Depth First Search and Birnbaum.