Rolling bearing fault diagnosis based on adaptive VMD and IAO-SVM
A new optimization strategy addressing the issues of parameter sensitivity and redundancy in feature selection encountered is proposed,when applying Support Vector Machines(SVM)for rolling bearing fault diagnosis.By integrating Adaptive Variational Mode Decomposition(IVMD)with the PE method,the fault feature vectors of rolling bearings are extracted,successfully capturing key indicators of bearing health.To enhance the performance of SVM,an improved Aquila Optimizer(IAO)algorithm was utilized,which incorporates Logistic mapping,elite reverse learning strategies,and survival rate strategies from the DOA algorithm,forming a multi-strategy IAO.This method was used to finely tune the parameters of SVM.The experimental results on the Case Western Reserve University bearing dataset show that the fault detection rate of this method reached 99.17％,which is an improvement of 5％and 0.8％over the traditional SVM and AO-SVM methods,respectively.These achievements not only demonstrate the effectiveness of the optimization strategy but also provide a more reliable technical approach for diagnosing faults in rolling bearings.
NETL
NSTL
万方数据
