Modeling Method And Compensation For Thermal Error Of Large Gantry Five-Sided Machining Center Based On Deep Learning

Thermal error occupies a large proportion in the machine tool error source.Thermal error compensation is an economical and effective method to reduce thermal error and improve machine tool processing accuracy.The large gantry five-sided machining center was taken as the research object.The machining precision was studied from the aspects of hard structure analysis,heat source analysis,temperature measurement point optimization,thermal error model establishment and experimental compensation verification.The installation locations of the temperature sensors were determined by heat source and structure analysis.The fuzzy clustering algorithm and the grey correlation degree theory were used to optimize the temperature measurement points.The deep learning method was introduced into the thermal error modeling.The thermal error data and the optimized key temperature measurement point data obtained by the temperature and thermal error detection system were used to accomplish the construction and optimization of the thermal error model.Finally,the validity of the model established by the method was verified by the thermal error experiment The temperature measurement point optimization and thermal error modeling system about the large gantry five-sided machining center was developed.The main research contents are as follows:1)The large gantry five-sided machining center was taken as the research object.The structure and heat source were analyzed to obtain an understanding of the temperature field distribution to determine of the installation positions of the temperature sensors.Then the fuzzy clustering algorithm and the grey correlation degree theory were introduced respectively.These two methods were combined for the optimal selection of temperature measuring points.2)According to the problems existing in the current thermal error compensation technology,the deep learning method was introduced into the thermal error modeling.The deep auto-encoder method was used to automatically extract the temperature data features and build the thermal error model of the large gantry five-sided machining center combined with the neural network training algorithm.3)In order to facilitate the data processing and thermal error model construction optimization,MATLAB GUI was used to develop the temperature measurement point optimization and thermal error modeling system of the large gantry five-sided machiningcenter based on the above research content.The hardware platform of temperature and thermal error detection system was designed with sensors and data acquisition card to gather data of temperature and thermal error.Based on the data,the thermal error model was optimized and the thermal error compensation experiment was verified.