PROJECT TITLE :

Dynamic Uncertain Causality Graph Applied to Dynamic Fault Diagnoses of Large and Complex Systems

ABSTRACT:

Intelligent systems for on-line fault diagnoses will increase the reliability, safety, and availability of enormous and complicated systems. As an intelligent system, Dynamic Uncertain Causality Graph (DUCG) could be a newly presented approach to graphically and compactly represent complicated unsure causalities, and perform probabilistic reasoning, which can be applied in fault diagnoses and alternative tasks. However, only static evidence was used previously. During this paper, the methodology for DUCG to perform fault diagnoses with dynamic evidence is presented. Causality propagations among sequential time slices are avoided. Within the case of method systems, the basic failure events are classified as initiating, and non-initiating events. This classification can increase the potency of fault diagnoses greatly. Failure rates of initiating events can be used to exchange failure probabilities while not affecting diagnostic results. Illustrative examples are provided to illustrate the methodology.