Research On Fault Diagnosis Of Rolling Bearing Of Combined SOM Motor Based On EEMD

Motor is important power equipment in modern life,especially in industrial production occupies a dominant position.As an important part of the motor equipment,the working condition of the rolling bearing will directly affect the stable operation of the motor equipment.Therefore,it has theoretical research value and practical application significance for fault diagnosis of motor bearings.This paper mainly analyzes and studies the feature extraction and state recognition of vibration signal in motor bearing fault,the main work is as follows:Firstly,the failure form of rolling bearing is analyzed,the frequency characteristics of bearing are deduced with wear as the main fault state,and the Simulation experiment platform is introduced.Because of the non-stationary vibration signal when the motor bearing fails,the set empirical modal decomposition is selected to decompose the signal,and compared with the traditional EMD,it is proved that the set empirical modal decomposition can effectively suppress the modal aliasing,and the extraction accuracy of the signal is significantly improved,and the set empirical modal decomposition has a great advantage.A method of bearing fault feature extraction based on EEMD and energy moment is proposed.In this method,EEMD is used to decompose the fault signal,and the energy moment of each order model is obtained as the eigenvector,and the feature extraction of the fault signal is realized.The self-organizing competitive neural network is selected as the identification method of fault diagnosis,and its basic principle is introduced.Set modal decomposition and energy moment tectonic fault eigenvectors are used as samples of SOM neural network for training and testing.And compared with EEMD-BP and EEMD-RBF diagnostic prediction results,the experimental results show that the method in this paper can quickly and accurately diagnose rolling bearing faults under the condition of limited training samples.