Sequential diagnosis for rolling bearing using fuzzy neural network

In the case of fault diagnosis of the plant machinery, knowledge for distinguishing faults is ambiguous because definite relationships between symptoms and fault types cannot be easily identified. So this paper presents a sequential diagnosis method for rolling bearing by a fuzzy neural network with the features of a vibration signal in time domain. The fuzzy neural network is realized with a developed back propagation neural network, by which the fault types of a bearing can be automatically distinguished on the basis of the possibility distributions of symptom parameters sequentially. The non-dimensional symptom parameters which reflect the features of signal measured for the diagnosis are also described in time domain. The faults that often occur in a bearing, such as the outer race flaw, inner race flaw, and roller element flaw, are used for the diagnosis. Practical examples of diagnosis for a rolling bearing used in rotating machinery are shown to verify the efficiency of the method.