Identification and optimal selection of temperature-sensitive measuring points of thermal error compensation on a heavy-duty machine tool

Thermal error compensation is considered as an effective and economic method to improve the machining accuracy for a machine tool. The performance of thermal error prediction mainly depends on the accuracy and robustness of predictive model and the input temperature variables. Selection of temperature-sensitive measuring points is the premise of thermal error compensation. In the thermal error compensation scheme for heavy-duty computer numerical control (CNC) machine tools, the identification of temperature-sensitive points still lacks an effective method due to its complex structure and heat generation mechanisms. In this paper, an optimal selection method of temperature-sensitive measuring points has been proposed. The optimal measuring points are acquired through three steps. First, the degree of temperature sensitivity is defined and used to select the measuring points with high sensitivity to thermal error. Then, the first selected points are classified with fuzzy clustering and grey correlation grade. Finally, the temperature-sensitive measuring points are selected with analysis of location of temperature sensors. In order to verify the method above, an experiment is carried out on the CR5116 of flexible machining center. A novel temperature sensor, fiber Bragg grating (FBG) sensor, is used to collect the surface temperature of the machine. A thermal error compensation model is developed to analyze the prediction accuracy based on four sequences of measuring points, which are generated by different selection approaches. The results show that the number of the measuring points is reduced from 27 to 5 through the proposed selection method, and the thermal error compensation model based on the optimum temperature-sensitive measuring points has the best performance of prediction effect.