Robust modeling and prediction method of "symmetric mapping" for thermal error of CNC machine tools

In view of the obvious differences in thermal characteristics of different types of machine tools with different mechanical structures, the established thermal error prediction model is beyond the range that the model algorithm can accurately predict, which involves the algorithm mechanism. Based on this, a "symmetric mapping" robust modeling prediction method is proposed for the different conventional thermal characteristics of CNC machine tools exhibited by the research object in this paper, which solves the problem that the special thermal deformation law does not match the robustness of the existing model algorithm. This effectively improves the prediction accuracy and robustness of the model. At the same time, the temperature sensitive points of CNC machine tools change with the seasonal changes, which leads to the change of the accuracy of the prediction model and the decrease of the robustness of the model. A method of using K-means combined with gray correlation (KWG) is proposed to optimize the robustness of the model. The temperature sensitive point with high stability can effectively reduce its variability and improve the accuracy and robustness of the model. In this paper, Vcenter-55 CNC machining center and leaderway-450 CNC machining center are taken as the research objects, combined with the experimental data of the whole year and the thermal error data in the Z-axis direction as an example, the residual Z-axis thermal error prediction model is finally established. The standard deviation is increased from 11um before data processing to about 4.1um, which improves the accuracy by 63%. At the same time, with the weakening of the variability of temperature-sensitive points, the application period of the model is increased to a validity period of 6 months, which greatly improves the robustness of the model.