Method Of Fault Diagnosis For Rolling Bearing Based On BP Neural Network

Rolling bearing is one of the most ordinary parts in mechanical machine, its running state can influence the performance of the whole machine directly, and the fault of bearing is one of the main factors that lead to the fault of machinery equipment during the running process. Fault diagnosis for rolling bearing is the key to assess the manufacturing quality of bearing, so it is important to study the technology of fault diagnosis for rolling bearing.The common fault diagnosis methods can hardly reflect vibration characteristics completely by only using time or frequency domain analysis based on vibration signal, and because of the relation between fault diagnosis model and eigenvector for rolling bearing is complex nonlinear, so there is no good solution for this problem just depending on experience. According to the fault diagnosis of precision engineer bearing, this paper probes into the application of the BP neural network technology in the fault diagnosis for rolling bearing by using the neural network's capacity of mapping to nonlinearity, self-learning, self-organizing and self-adapting.On the basis of analyzing the fault mechanism and vibration signal characteristics of rolling bearing systematically, and after analyzing and processing the vibration signals of right and fault state of rolling bearing, partial appropriate feature parameters are selected as the input of the neural network according to the time and frequency domain characteristics of parameters in this thesis. And then, by the approximation principle for unknown movement of neural network technology, the specific structure of neural network is designed to aim at the specific characteristic of fault diagnosis for rolling bearing, and the fault diagnosis system for rolling bearing based on BP neural network is built up. Finally, the training set of right and fault states of rolling bearing is built up by using the measuring datas of rolling bearing from former research, after that, the neural network model is trained on the platform of Matlab software, the operating state of rolling bearing has been diagnosed by using the above network which has been trained well.The simulation result shows that the BP neural network which is built up for diagnosing the fault of rolling bearing can identify the fault types accurately according to the datas.