Research On Error Modeling Of Cutting Force And Analysis And Application Of Three-Dimensional Heat Critical Point Of CNC Machine Tools

The geometric errors of CNC generated in the process of machining are related to the cutting force of CNC. But the influence of the cutting force is neglected in most of the existing geometric error modeling method of CNC. Grating measuring system for CNC machine tools in the traditional fixed mode, numerical control machine tool thermal deformation of lathe bed will be through the set screw is passed on to the grating measuring system, the additional impact on the grating measuring system, it is difficult to through error compensation theory to establish precise zero error of the grating measuring system and the model of error phenomenon.In order to improve the modeling precision of CNC geometric error and the error compensation effect, the main cutting force was simulated and measured by using the dynamometer which was loaded to the CNC. The pitching angle error and the deflection angle error were measured by using laser interferometer measurement instruments. Trained by the practical measurement error data, the force-geometric error predicting model based on PSO-SVM was established, whose key parameters were optimized by using the particle swarm optimization algorithm (PSO).According to form thermal deformation theory, the paper puts forward a kind of CNC machine tools based on thermal deformation point new way of fixed grating measuring system, The three axis CNC machine tool thermal deformation point as the direction of the axis lathe bed on fixed point of grating measuring system, adopt the way of floating support limit its terminals. The method can reduce or even eliminate the numerical control machine tool lathe bed of thermal deformation on the additional effects of thermal deformation error of grating measuring system. Through simulation analysis, the method can build more precise grating measuring system of the thermal deformation of the zero error and the error compensation model of value. In order to improve the comprehensive compensation effects of all kinds of CNC machine tools error, a mathematical model of comprehensive error of CNC machine tool is presented for CNC machine tools thermal deformation error synthesis model provided the powerful basis.The verification experiments result shows that the maximum error of the deflection angle error between the output of PSO-SVM model and the actual measured error data is0.6μrad, and the maximum error of pitching angle error is0.21μrad. Compared with the force-geometric error predicting model built by using BP neural network and SVM whose parameters was optimized by using conventional method, the prediction precision of the PSO-SVM error model was greatly improved. So the force-geometric error predicting modeling method based on PSO-SVM can be used to compensate the geometric error of CNC machine tool in real-time with high-precision.