Systematic Thermal Error Compensation Technology For The Spindle And Cutting Tool Of High Speed Machining Center

With the development of technology in manufacturing, people pay more and more attention to how to improve the machining accuracy and quality of NC machine tools which are the basic equipments in modern manufacturing. There are two methods to ensure the machining accuracy of parts and they are error prevention and error compensation respectively. Error prevention depends on the design accuracy of machine itself to ensure the machining accuracy, that is to say, it should make the accuracy of machine higher than that of machining requirement of part. Generally speaking, the accuracy of common parts is guaranteed by this way. Error prevention is the most basic and effective method to ensure the machining accuracy, which is the widely proved method. This method is to ensure machining accuracy by improving design accuracy, manufacturing accuracy and assembly accuracy and controlling the temperature and environment strictly to reduce and eliminate the error resources in system. By this means, the original errors can be reduced to a certain extent, but there exists some limitations. And it is very difficult and costly to produce the machine tool whose accuracy is higher than the part, or even that machine tool can not be produced. In order to produce the parts whose accuracy is higher than the machine tool, error compensation should be adopted.In this paper, the thermal error compensation of the spindle and cutting tool of machine tool is studied to improve machining quality and efficiency, and the feasibility and validity of the proposed error compensation model is verified in actual processing, the proposed error compensation can provide guide to actual processing.The main research contents are shown as follows:(1) In this paper, the heat transfer mode and the basic theory of spindle are analyzed and calorific value of spindle used as thermal load is calculated and applied on the spindle system. 3D modeling software Pro/E is used to build the 3D model of the spindle in a certain high-speed milling center, and the simplified solid model of spindle is introduced into the finite element analyzing software Analysis. The finite element method is used to analyze the thermal characteristics and the temperature field of spindle to obtain the variation of temperature field and status of thermal deformation to provide guide for the structure design of machine tool in the future.(2) In accordance with the main factors which effect thermal error of spindle, the distribution of temperature field is analyzed. In order to optimize the measuring points to reduce the number of measuring points, a set of data measuring and collecting equipment is developed which can measure the temperature field and collect data in real time. By using the gray correlation analysis in the gray theory, the gray correlation analysis model is built. The measuring points are regarded as effect factor, and the effect of these measuring points on thermal error is analyzed. After comprehensive evaluation of the measuring points, these points are arranged in the order of the importance of their effects on the thermal error in Z direction to find the key measuring points. In this way, the number of measuring points in actual error compensation model is reduced effectively and the robustness of model is improved.(3) After measuring points optimization, on Matlab operation platform, the least square support vector machine is used to build the prediction model of thermal error to predict the thermal error in Z direction. In modeling process, RBF is selected as the kernel function. The grid method is used to optimize the parameters ? and ? of the kernel function, namely RBF. The grid method is the most frequently and directly used method which can find the optimal parameters of kernel function in the widest range of parameter. The model built with the optimal parameters can improve the prediction accuracy to meet the prediction accuracy in actual error compensation. After initial training the original data, the obtained model is lake of sparsity and the robustness of model is weakened, so the trained model needs to be trained again to get rid of the points with sparsity and increase robustness of model. After several times training, the prediction model has both the sparsity and robustness. In this way, the accuracy of model in improved and application scope is expanded.(4) The thermal deformation of cutting tools in NC machining is measured. Because the cutting tool is rotating in high-speed manufacturing, the traditional contact measurement can not measure the cutting tool deformation. Accordance with this trait of cutting tool measurement, the technology of non-contract measurement based on image information is proposed to measure the thermal deformation of cutting tool. The system of image acquisition and measurement is developed and the processing system is designed. The improved gravity method is used to locate the tool tip in sub-pixel level. The edge of cutter of sub-pixel location can meet the measurement requirement more accurately in this paper. The coordinate of the vertex of tool tip can be calculated by using twice curve fitting method. By this means, the thermal deformation of cutting tools can be measured.(5) To solve the problem that how to improve machining efficiency on the premise of ensuring the machining quality, the thermal error compensation in preheating is mainly considered. The NC machine tool needs preheating before machining every time, or else the spindle system can not reach a thermal balance that will lead to the obvious decrease of processing quality and the preheating of spindle not only is a waste of time but also virtually increases processing cost. Therefore, the thermal error compensation in preheating is very significant. In this paper, based on the obtained thermal error compensation model and the thermal error data of cutting tool, the gray theory and the least square support vector machine are used to compensate the thermal error in preheating period to decrease the preheating time, even the preheating can be skipped . In this way, the efficiency can be improve on the premise of ensuring quality, the waste of manpower, material and time can be avoided and the machining cost of part in actual processing can be decreased significantly. The method proposed in this paper is meaningful in actual engineering application.