Temperature-sensitive point selection of thermal error model of CNC machining center

Abstract

The temperature-sensitive point selection of computer numerical control (CNC) machine tools is crucial to thermal error modeling and compensation. Using the comprehensive analysis method which is a combination of fuzzy clustering, gray correlation, stepwise regression, and determination coefficient optimizes temperature measuring points. First of all, using fuzzy clustering and F statistic classifies temperature variables. Secondly, according to the gray correlation degree between the temperature variables and thermal error, the key temperature variable of each class is selected. Then, the significance of regression equation and parameters of thermal error model are tested, based on the stepwise regression analysis and the non-significant variables are excluded. Finally, the selected temperature variables are arranged to simple permutation and combination, and compares determination coefficients to determine the optimal temperature-sensitive points. The above method is verified on the Leaderway V-450 of CNC machining center. The thermal error prediction model is established. The accuracy and robustness of the model are analyzed. The results show that the temperature measuring point number is reduced from 10 to 2, the fitting accuracy of thermal error prediction model is high, and the model can achieve a good prediction effect and strong robustness under different conditions of spindle speeds and ambient temperature.