Fault diagnosis of rolling bearing based on WHO-VMD-SVM

A feature extraction method based on WHO optimization algorithm to optimize variational mode decomposition (VMD) was introduced, and support vector machine (SVM) was used for fault diagnosis of rolling bearings. In this method, the number of intrinsic modal function (IMF) and penalty factor α in VMD algorithm are used as optimization variables, and the variables are adaptive optimization by using WHO optimization algorithm. The mean value, variance, standard deviation and sample entropy are selected as feature vectors from the modal components after optimization and decomposition into the SVM classifier for fault diagnosis. The results show that, compared with the traditional fault diagnosis model, the WHO-VMD-SVM model has higher classification accuracy and diagnostic accuracy in the fault diagnosis of rolling bearings, which proves the effectiveness and superiority of the method.