Novel tool condition identification based on finite element modelling and support vector machine

Abstract

Timely and accurate recognition of tool wear condition could reduce the machining cost greatly. However, the problem of sample missing and imbalance affects the classification accuracy of AI models seriously. A novel strategy based on finite element method (FEM) and support vector machine (SVM) is proposed to overcome the above problem. Firstly, several tool wear experiments are carried out to obtain experimental samples. Then, a FEM model is established and verified through experimental samples. Based on the verified FEM model, several simulated sample of missing and imbalanced samples in experiments could be supplemented to compose a complete training set. The SVM model is trained by the complete training set to tool wear condition identification. Milling tool experiments demonstrate that the proposed method can obtain above 96% identification accuracy with a small number of experiments, which is 28% higher than that based on experimental samples. [Submitted 21 February 2022; Accepted 28 July 2022]