Research On Fault Diagnosis Method Of Rolling Bearing Based On EEMD-GSSA-SVM

As a very important part of electrical machinery and equipment,rolling bearings often work in complex working conditions and harsh conditions,making rolling bearings the most vulnerable to damage,resulting in an important impact on the operation of the entire equipment.If the rolling bearing of the motor is abnormal or malfunctions,it is very likely to cause the equipment to deviate from the normal working state,and even cause downtime,shutdown,mechanical damage and other failures.Therefore,the research on the fault diagnosis method of rolling bearing has important practical significance.Based on Matlab software and programs,this thesis studies rolling bearing feature extraction and fault diagnosis methods.First of all,this article reviews the research background of motor rolling bearings,analyzes the working principle,common failure forms and causes of motor rolling bearings,and the failure mechanism of motor bearings in detail,and introduces the fault diagnosis simulation experiment platform.Through the use of Empirical Mode Decomposition(EMD)and Ensemble Empirical Mode Decomposition(EEMD)algorithms and the theory of energy moments,it can be concluded that EEMD can effectively suppress modal aliasing.Secondly,the principle of the sparrow search algorithm is discussed and the shortcomings of the implementation are improved and verified.Furthermore,using the fault diagnosis simulation experiment platform,using the improved sparrow search algorithm and the support vector machine(Support Vector Machine,SVM)classification principle combined,respectively using the sparrow search algorithm(Sparrow Search Algorithm,SSA)and improved sparrow The Global Sparrow Search Algorithm(GSSA)of the search algorithm optimizes the penalty factor C and the kernel function parameter g of SVM.Finally,the classification accuracy of different combinations is compared,and the better GSSA-SVM method is selected as the research model.Perform experimental verification,that is,use the K-CV cross-validation method to compare the diagnosis and prediction results of EEMD-SSA-SVM and the application of EEMD-GSSA-SVM in this thesis to compare and diagnose the experimental signals.The experimental results show that the method in this thesis is used to extract and diagnose the vibration signals in the experiment.The experimental results verify that the EEMD-GSSA-SVM rolling bearing fault diagnosis method can quickly and accurately diagnose rolling bearing faults,which is a more accurate and faster rolling bearing fault diagnosis Provides basic research methods.There are 26 graphs,7 tables,and 63 references in the thesis.