Fault Diagnosis Method for Rolling Bearings Based on EWM-PCA-improved SVM Model

In order to further enhance the accuracy of rolling bearing fault diagnosis, a rolling bearing fault diagnosis method based on EWM-PCA improved SVM model is proposed to achieve high-precision recognition of rolling bearings. Firstly, the Entropy Weight Method (EWM) is used to perform weight correction on the extracted vibration signal features of rolling bearings. Then, Principal Component Analysis (PCA) is used to perform dimensionality reduction on the modified features. Finally, the Seagull Optimization Algorithm (SOA) is used to optimize the feature parameters of the Support Vector Machine (SVM) to obtain the diagnostic model. By using the model to identify normal, inner ring faults, outer ring faults, and rolling element faults of rolling bearings, the results show that the EWM-PCA-improved SVM model has a better rolling bearing fault recognition rate, with an average of 99.61%.