Thermal Error Modeling Method of Machining Center Linear Axis for Heat Conduction Mechanism

Abstract

AbstractTo address the thermal deformation of machine tool components, a thermal error prediction model based on the ROA-LSSVM network was proposed. First of all, the heat transfer mechanism of the linear feed system was analyzed. By analyzing temperature distribution characteristics during the heat transfer process, the best temperature measurement point position was determined to ensure that the thermal error could be accurately predicted. Secondly, in order to build a prediction model with high accuracy and strong robustness, Raccoon optimization algorithm (ROA) was proposed to optimize the hyperparameters of the least square support vector machine (LSSVM) network model, which was difficult to determine the kernel function and penalty function. Finally, the experiment was measured on a VDL-600A machining center, and the accuracy and practicability of the proposed thermal error prediction model were verified by the thermal deformation in the measurement process. The experimental results show that The ROA-LSSVM model reduces the RMSE by 42% compared with the LSSVM network and 45% compared with the SVM network.