Fault diagnosis of rolling bearings in multiple conditions based on EMD and PSO-SVM

Abstract

∗In order to improve the accuracy of fault diagnosis of rolling bearings, a fault diagnosis method based on empirical mode decomposition and particle swarm optimization support vector machine is proposed. After preprocessing the rolling bearing data under multiple operating conditions, the vibration signal is decomposed by empirical mode decomposition (EMD). The appropriate intrinsic mode function is selected to construct the energy feature vector as the rolling bearing fault feature vector. The particle swarm algorithm is used to optimize the parameters of the support vector machine, and the single fault depth and multiple fault depth data samples are respectively selected for the training and testing of SVM and Particle Swarm Optimization – Support Vector Machine (PSO-SVM). MATLAB experiment results show that PSO-SVM significantly improves the accuracy of rolling bearing fault diagnosis under multiple operating conditions compared with SVM, and the average fault diagnosis accuracy is 99.32% and 95.43%, which can be used as the evaluation standard of rolling bearing fault diagnosis.