Research On Thermal Error Experiment And Modeling Of CNC Machine Tool Spindle

Nowadays the manufacturing industry is moving towards the direction of precision.Thermal error has become an important factor affecting the machining precision of high precision CNC machine tools,and the thermal deformation of machine tool spindle is the main source of thermal error of machine tools.Therefore,reducing the thermal error of spindle becomes an important method to improve the machining accuracy of machine tool.In this paper,VMC1165B vertical machining center is taken as the experimental object,the temperature and thermal error detection system of machine tool is built,and the temperature and thermal error changes in the machining process of machine tool are measured and analyzed.Fuzzy Clustering Analysis(FCA)and Grey Correlation Analysis(GCA)were used to determine the best temperature-sensitive points.Principal Component Regression(PCR)analysis and Multiple Linear Regression(MLR)analysis were used to establish the thermal error model of the machine tool spindle,and the fitting and generalization ability of the two models were analyzed.The main research contents are as follows:(1)A thermal error experimental platform was established.Based on the ISO 230-3 international standard for the determination of machine tool thermal effects,multiple operating conditions were designed for the machine tool,and the machining temperature of the machine tool was measured using a temperature sensor.At the same time,the thermal error data of the machine tool spindle was measured using a displacement sensor with the five-point method.The influence of operating time and operating conditions on machine tool temperature and spindle thermal error were analyzed according to the measurement data.(2)The temperature-sensitive point of the machine tool were selected.Based on the measured temperature data and spindle Z direction thermal error data,firstly,fuzzy clustering analysis is used to divide each temperature measurement point into several categories;then,grey relational analysis is used to calculate the grey relational degree between each temperature measurement point and the spindle Z direction thermal error.The temperature measurement point with the highest grey relational degree value is selected in each category of temperature measurement points.Finally,the best temperature-sensitive point is selected based on the adjusted determination coefficient.(3)The spindle thermal error model was established.Taking the temperature data of the best temperature sensitive point and the spindle speed as input variables,the thermal error prediction model of machine tool spindle in Z direction is established by principal component regression analysis and multiple linear regression analysis respectively.The two models are used to predict the thermal error of machine tool spindle under multiple operating conditions,and the fitting ability and generalization ability of the two models are analyzed by the prediction results.(4)The prediction effect of the thermal error model was verified.The temperature-sensitive points of machine tools were selected again with different data processing methods,and the temperature rise data and spindle speed of the selected temperature-sensitive points were taken as input variables to establish the principal component regression model and the multiple linear regression model,respectively.By analyzing the fitting ability and generalization ability of the models,The prediction accuracy of two thermal error prediction models is verified when the temperature sensitive points of machine tools change.