Research on Thermal Error Compensation Technology of CNC Machine Tool Under Real Cutting Condition

Improving machining accuracy is the ultimate goal of NC machine tool improvement. In the process of mass production, The dimension accuracy and process capability index (CPK) of parts are low, which is mainly affected by the thermal error of machine tool. In this paper, on the basis of the trial cutting experiment, the data of the temperature measurement points of the machine tool are analyzed by cluster analysis, and the primary temperature is divided into groups. The correlation coefficient between the variables in each group and the change of the workpiece inner diameter Δ DN is compared to select one of them. Then, based on the CPK index of the product, the key temperature measurement points affecting the thermal error of the machine tool are selected. Finally,Combined with the data of key temperature measuring points and the change data of workpiece inner diameter, the radial thermal error model of CNC machine tool spindle is constructed by using B-P neural network model. The experimental results show that the dimensional accuracy of the workpiece after compensation meets the technical requirements, the tolerance of the workpiece is reduced from 12 μm to 6 μm, and the CPK index is increased from 0.79 to 1.39. The compensation effect is obvious.