Research of Tool Wear Predictive Technique Based on Support Vector Machine

The cutting tool condition monitoring technology is very important to the automated production, which is a small sample but very complicated system on account of the limited experiment data. In this paper, with feature extracted from the workpiece surface image, the tool wear predictive model is built based on SVM. The proposed method use Genetic Algorithm adjusts SVM kernel parameter. Experiment results show that the proposed model has a good recognition performance. Compared with BP ANN method, GA-SVM provides a better generalization capability and a lower prediction error.