Fault diagnosis based on integrated neural network and D-S evidential reasoning

Abstract

For the reasons of low fault diagnosis accuracy of traditional diagnosis methods, a fault diagnosis method fusing BP neural Network and multi-sensor information fusion technique based on D-S evidence theory was presented to realize fault diagnosis. On the base of integrated neural network, importing evidential reasoning, a fault diagnosis technique which combine neural network and D-S evidential reasoning (NN -DS diagnostic technique) is proposed. It uses BP neural network local diagnosis respectively from different symptom field, and each network receives respective result, then D-S evidential reasoning will be used for global diagnosis to gain a unified result. At last an exemple is given to indicate it's validity.