2016 3rd International Conference on Information Science and Control Engineering (ICISCE 2016)

92 / 2016-04-12 10:26:57

Research on Intelligent Fault Diagnosis Method Based on Decision Tree and Relevance Vector Machine

9810

In view of the problems of fault diagnosis, such as small sample, nonlinear and multi fault treatment, and the existing problems of traditional intelligent diagnosis method, cause the fault diagnosis accuracy is not high. To compensate for this deficiency, on the basis of comprehensive analysis of the respective advantages of the decision tree(DT) and the relevance vector machine(RVM), an intelligent fault diagnosis method based on DT and RVM is constructed. In this paper, the multi class classification problem is decomposed into two kinds of classification problems by constructing two fork trees; in each decision node, RVM is used to classify two categories, so as to realize the multi class classification of RVM. Theoretical analysis and experimental results show that the proposed method has better performance in sparse and diagnosis efficiency while keeping high accuracy compared with support vector machine(SVM) method, which makes it more practical; and that the proposed method has a better training efficiency compared with OAR-RVM and OAO-RVM.