Research And Realization Of The Fault Diagnosis System For The Rolling Bearing Of A Certain Aero-generator

The rolling bearings of aero-generators are time-consuming parts,i.e.components are replaced when the service time is reached.However,during the actual investigation,bearing faults occur before the replacement time is reached.This paper takes the X-4612 rolling bearing of an aero-generator as the research object,collects the vibration signal of the aero-generator,extracts fault characteristic quantities by wavelet packet decomposition,and combines RBF network optimized by genetic algorithm to design and implement rolling bearing fault diagnosis system of the aero-generator.Firstly,the vibration mechanism,fault form and fault characteristic frequency of rolling bearing are discussed,and the common identification and judgment methods and fault characteristic extraction methods of rolling bearing are discussed.It is proposed that the vibration signal of rolling bearing is used as the research object,and the characteristic quantity of vibration signal is extracted by wavelet packet decomposition and EMD decomposition methods.Secondly,the wavelet threshold denoising method is used to denoise the vibration signal data,and the effect of multiple wavelet threshold denoising is compared,and the wavelet soft threshold function limit threshold denoising method is selected.The wavelet packet decomposition method and the EMD decomposition method are used to extract the vibration signal fault characteristic quantity after the noise reduction processing,and the data sample set is constructed.Thirdly,RBF neural network is selected to establish wavelet packet-RBF and EMD-RBF fault diagnosis model by combining data sample set;on this basis,genetic algorithm is used to optimize SPREAD parameters of RBF network,and wavelet packet-GA-RBF and EMD-GARBF fault diagnosis models are constructed by combining wavelet packet decomposition and EMD decomposition.By comparing the diagnostic accuracy of the four models with the test set,it shows that the RBF neural network model optimized by genetic algorithm has a great improvement in the recognition accuracy.Meanwhile,the wavelet packet-GA-RBF fault diagnosis model with the highest diagnostic accuracy is selected for the fault diagnosis system.Finally,based on the above research and analysis,a certain type of aero-generator rolling bearing fault diagnosis system is designed using Lab VIEW and MATLAB joint programming,and related experimental equipment is built.The system is used to diagnose the rolling bearing in different states and use the system.The data is collected to verify the diagnosis result,and the result proves that the system can effectively identify the fault state of the rolling bearing of the aero-generator.