关键词: 流向图, 非朴素贝叶斯推理, 滚柱轴承, 故障程度识别, 节点约简

Abstract: To address the issue of poor reasoning ability and the high computational burden of flow graphs,a fault severity identification method of roller bearings using flow graph and Non-naive Bayesian inference (NNBI) was put forward.A normalized flow graph constructed according to fault features of roller bearings extracted from training samples was used to represent the causal relationship among attributes in an intuitive manner.Then a node reduction algorithm based on SDCAN was developed to delete redundant condition attribute nodes,which could reduce computational complexity of mode identification.An NNBI algorithm based on flow graph was presented to identify roller bearing conditions in test samples.Experimental results demonstrated that the proposed method could intuitively and accurately recognize fault severities of roller bearings.

Key words: flow graph, non-naive Bayesian inference, roller bearing, fault severity identification, node reduction