Research On Fault Diagnosis Of Rolling Bearings Based On Non-stationary Signal Statistical Index Feedback Optimization

Rolling bearing is an important component of rotating machinery and one of the main fault sources.It is of great significance to carry out on-line condition monitoring and fault diagnosis for rolling bearing.In this paper,a bearing fault diagnosis method based on statistical index feedback optimization of non-stationary signals is proposed to solve the signal distortion and spectrum aliasing problems brought by non-stationary working conditions for bearing acoustic vibration fault signals.The specific research contents are as follows:Acoustic detection technology in view of the road in the doppler effect caused by bearing acoustic signal distortion problem is put forward to statistic feedback optimal adaptive signal distortion correction algorithms,but by the signal itself statistical parameters automatically identify rail edge,and based on the parameters of the automatic identification and signal resampling technology realize the rail edge adaptive doppler signal distortion correction.The critical signal-to-noise ratio(SNR)with kurtosis,skewness,peak factor,pulse factor,waveform factor and margin factor as feedback indexes is analyzed and compared.Further combined with genetic algorithm to achieve fast and accurate extraction of rail edge parameters.The proposed method is a signal-driven method and requires no other sensor except the microphone.Simulation and experimental signals are used to verify the effectiveness of the proposed method.Aiming at the frequency aliasing phenomenon of bearing signals in the motor variable speed condition,an accurate estimation algorithm of speed based on the Hall speed measurement signal and the feedback optimization of signal statistics index was proposed to realize the accurate resampling of fault signals.This method utilizes the algorithm to make up for the low accuracy of the Hall speed measurement.Only the Hall sensor single-turn speed measurement signal can be used to accurately estimate the speed and realize the fault diagnosis.It has the advantages of low cost,convenient installation and debugging,especially suitable for the occasions where the encoder is inconvenient to install,the working environment is harsh or the Hall sensor has been installed,and the method has good adaptability.Simulation and experimental signals are used to verify the effectiveness of the proposed method.