Volumetric Error Prediction and Compensation of NC Machine Tool Based on Least Square Support Vector Machine

A novel method based on Least Square Support Vector Machine (LS-SVM) to predict the volumetric errors of NC machine tools is presented. Using the laser vector diagonal step measurement, the volumetric errors of NC machine tools were measured. Then LS-SVM was used to establish the predictive model of volumetric errors. Based on structural risk minimization, least linear system was used as loss-function, and grid search method was adopted to optimize the LS-SVM parameters, finally the volumetric error model was obtained for prediction and compensation. The experiment results show that the LS-SVM model of volumetric errors is more precise than the Artificial Neural Networks (ANN) model. And after compensation, the accuracy of the machine is improved more than 90.38%. Hence, the LS-SVM volumetric error model is feasible and effective to enhance the performance of NC machine tools.