Research On Fault Diagnosis Of Rolling Bearings Based On Signal Processing And Deep Learning

Rolling bearing fault diagnosis consists of fault signal acquisition,fault feature extraction and fault feature recognition.The reliability of fault feature extraction and the accuracy of fault feature recognition are the core problems that affect the accuracy of rolling bearing fault diagnosis.The application of signal processing technology to accurately extract fault signal features is a research hotspot in the field of fault diagnosis,which has been widely used in rolling bearing fault diagnosis.In addition,deep learning has been widely applied in the field of fault diagnosis in recent years due to its powerful feature learning ability and data processing ability.In this paper,a rolling bearing fault diagnosis method based on signal processing and deep learning is proposed to improve the accuracy of rolling bearing fault diagnosis.The main research contents of this paper are as follows:A rolling bearing fault feature extraction technology based on CEEMDAN-Hilbert was proposed.Firstly,Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)was used to decompose the vibration signals of the rolling bearing.Intrinsic Mode Function(IMF)components are obtained,and then the instantaneous frequency spectrum of IMF components is obtained by Hilbert transform.The characteristic information of the original signal is extracted completely.In this paper,the CEEMDAN-Hilbert fault feature extraction method is proposed,and the simulation and example experiments show that the method can greatly improve the mode mixing problem of traditional Hilbert-Huang Transform(HHT),and has a higher accuracy in fault diagnosis.A rolling bearing fault diagnosis technology based on deep learning is proposed.The characteristic data of rolling bearing vibration signals were preprocessed,and a certain proportion of the data were randomly selected as the training set and the remaining data as the test set.The data were input into a Convolutional Neural Network(CNN)for fault diagnosis.Aiming at the problem that CNN network structure affects the performance of CNN,a GA-CNN rolling bearing fault diagnosis technology based on Genetic Algorithm(GA)was proposed to select CNN network structure adaptively.Experimental results show that compared with manual selection of CNN network structure,this method can not only improve the fault diagnosis efficiency,but also improve the accuracy of rolling bearing fault diagnosis to a certain extent.