Research on BO-CNN Based Tool Wear Status Monitoring Method

A tool wear condition monitoring method based on Bayesian (BO) optimization Convolutional Neural Network (CNN) was proposed to solve the problems of weak feature, difficulty in extraction. In the method, CNN model is used to construct the nonlinear mapping relationship between tool wear features and tool wear values, and the Bayesian optimization method is introduced to optimize the hyperparameters in the CNN prediction model to further improve the prediction effect and model fitting efficiency of the tool condition monitoring model. The results show that the CNN model based on Bayesian optimization can effectively monitor the tool wear state, with the determination coefficient R2 value of 0.9937, the root mean square error (RMSE) value of 2.0059, and the mean absolute error (MAE) value of 1.5287. Compared with other prediction models, evaluation indicators value is the best, which can more accurately complete the monitoring and intelligent warning of tool wear state.