Research on Fault Prediction of Electric Spindle in Five-Axis Machining Center Based on Bayes-SVM

Motorized spindle is an important functional part of the five-axis machining center, and its performance directly affects the quality of the workpiece. Therefore, this paper proposes an electric spindle fault prediction method based on Bayes-SVM model to solve the problems of low prediction accuracy of motorized spindle. In this method, the vibration signals of the motorized spindle under different states are extracted, and the SVM model is used to construct the nonlinear mapping relationship between the fault characteristics and the fault types. At the same time, the Bayes algorithm is introduced to optimize the hyperparameters in the model to improve the prediction effect. The results show that the Bayes-SVM model proposed in this paper can effectively predict the fault type of the motorized spindle, and the accuracy rate is as high as 99.17%. Compared with the BP model, CNN model and SVM model, the accuracy is increased by 10.84%, 6.67% and 4.17% respectively, which shows that the fault prediction model proposed in this paper has higher fault recognition ability and can effectively improve the maintenance strategy.