Research on the fault classification method of rolling bearings based on envelope entropy screening criterion and vmd fusion SVM

As a critical component in mechanical equipment, the fault diagnosis of rolling bearings is of great significance. Due to the presence of substantial noise in vibration signals, extracting fault features becomes particularly challenging. This paper proposes a method combining Variational Mode Decomposition (VMD) and Support Vector Machine (SVM) based on the envelope entropy selection criterion for rolling bearing fault diagnosis. First, the VMD method is used to decompose the vibration signal into several Intrinsic Mode Function (IMF) components, and the envelope spectrum of these components is obtained through envelope analysis. Then, the envelope entropy is adopted as the IMF component selection criterion to select the IMF components that best reflect the fault characteristics. The envelope energy of the selected IMF components is used as feature indicators and input into the SVM model to achieve fault type recognition. This method is applied to the fault diagnosis experiment of rolling bearing vibration signals, and the results show that the method can accurately extract fault features and effectively recognize fault types. The experimental results indicate that the fault recognition accuracy of this method reaches 99.5%. By comparing with other methods, the superiority and practical application value of this method in rolling bearing fault diagnosis are verified.