Research On Thermal Error Compensation Technique Of The Key Parts For NC Machine Tools

The modern machinery manufacturing technology is developing toward high efficiency, high quality, high precision,high intelligent and high integration, precision and ultra-precision processing technology has become the most important part of the high-end equipment manufacturing industry. With the rapid development of precision and ultra-precision NC machine tools, more and more people attach importance to factors affect the machining accuracy of NC machine tools, studies have shown that thermal error of high accuracy NC machine tools have accounted for 40%-70%, which is a major reason affect the machining accuracy of NC machine tools. Therefore, the research of thermal error compensation technique for NC machine tools has great significance to improve the machining precision.Motorized spindle and ball screw of LGMazak VTC-16 A Vertical machining center is studied, finite element software is used to analyze the temperature distribution of these two objects, the method combine fuzzy clustering and gray correlation is used to find out the best temperature measuring point. The thermal error model is built based on GA-BP network, through error compensation system, the influence of temperature fluctuations to machining accuracy of NC machine tools is reduced. The research content is as follows:(1) The basic forms of heat transfer is introduced. The finite element equation of the steady temperature field and transient temperature field is deduced based on heat transfer theory. The finite element analysis models of hot-structure coupling of motorized spindle unit and ball screw are established. Finite element method is used to study the change law of temperature field under the specific speed. The points which are sensitive to temperature changes is measured, the error between the simulation results and the experimental results obtained by calculation is about 4%, so the simulated temperature field for hot state characteristics study of motorized spindle unit and ball screw is feasible.(2) The temperature change of eleven points of the key parts is measured by experiment. The method which combine fuzzy clustering and gray correlation analysis is used to select three of the best temperature measuring points. Optimization of temperature measuring point can cut down the number of temperature sensors, reduce the coupling effect among the sensors, shorten the time needed for modeling.(3) The BP neural network and the GA-BP neural network is used respectively to establish thermal error prediction model, and the prediction results of two methods are compared to get the outcome, GA-BP prediction model has a shorter modeling time, higher precision and faster convergence speed. Thermal error of NC machine tools is compensated based on GA-BP network model. The experiment proved that the compensation rate of this model is %46.64, and compensation effect is good.