CNC Milling Process Mechanism Analysis And Simulation

NC machining geometry simulation could not predict the real process of NC machining accurately, so it can not provide reasonable machining parameters for users. Excessive cutting force, cutting heat can cause the damage of machine and poor precision of parts, in order to avoid this phenomenon, machining operators often sacrifice efficiency and choice conservative cutting parameters, it severely restricts the efficiency of NC machine. The physical simulation can reveal the physical nature of the machining process and avoid waste of time and cost of the actual cutting experiment on machine tools.In milling process, the modeling and simulation of cutting force and cutting heat simulation is a basis and core of physical simulation of NC machining. In the process of cutting, cutting force and cutting heat have important influence for tool wear,machining efficiency and surface quality, at the same time, they are also one of the symbols that measure the workpiece and cutting tool materials processing performance. So before the actual processing, the prediction of milling force of milling process and thermal deformation of process system has an important significance.Firstly, the orthogonal metal cutting principle, the basic theory of finite element method, the theory of elastic-plastic deformation theory and the temperature field theory must be mastered. Then we analyze milling force and cutting heat from the orthogonal cutting process of thin-walled parts. Finally, on the basis of above theories, the milling force prediction model and finite element model of temperature field are established.The milling force prediction model is the premise that reasonable controls milling parts machining error, tool wear, tool fracture and machine tool vibration, and it has the vital significance for optimization of processing parameters and the quality of parts processing. Then through multiple linear regression and BP neural network to establish the forecast model of milling force to predict the milling force, Getting fitting curve of the predicted values and the experimental value. Through the linear regression theory to remove the abnormal milling force data that collected by experiment, then using the experimental data to train the BP neural network model and predict the milling force, Getting fitting curve of the predicted values and the experimental value. Result shows that the BP neural network prediction model is more suitable for milling force prediction, and removing the abnormal points by the linear regression theory makes the BP neural network prediction model more accurate. Based on the theory of metal cutting and heat transfer, analyzing the milling heat theory, Considering the thermal coupling calculation complexity of three-dimensional cutting process, so the coupled thermo-mechanical finite element model was established, Through using finite element software DEFORM3D, three dimensional model of workpiece and cutter is set up to carry on the simulation, Then obtaining the stress distribution, strain distribution and the situation of the temperature field distribution of workpiece and tool and predicting the cutting force. At last, through cutting experiment validate the result of the finite element simulation.