Prediction And Analysis Of Structural Parameters Of Solid End Mill For Machining Thin-walled Parts

The advantages of heat resistance, corrosion resistance and high specific strength of titanium alloy have been widely concerned by the industry. Titanium alloy thin-walled parts not only has the advantages of titanium alloy, but also has the characteristics of light weight and cost savings, titanium alloy thin-walled parts in the aerospace industry has been widely used. But at the same time, titanium alloy is a kind of typical difficult-to-process material with complex shape, low rigidity and high degree of coordination, and vibration deformation is easy to occur during the machining process. Tool structure parameter is one of the main factors that affect machining deformation in milling process of titanium alloy thin-walled parts, so it is very important to predict the performance of tool structure parameters. This paper takes the cantilever plate structure typical of Ti-6A1-4V thin-walled parts side milling as the research object, the milling force and milling temperature as measure target, combined with theoretical analysis, three-dimensional physical simulation, BP neural network and experimental study, the performance prediction and analysis of the structural parameters of the integral end milling cutter are carried out, which provide the theoretical support for the structural optimization of the end milling cutter.With AdvantEdge FEM finite element simulation software platform,3D physical simulation milling model is established, the several selected important tool parameters (rake angle of helix edge, flank angle of helix edge, helix angle, number of teeth). Firstly, the milling force and temperature change caused by the single factor structure parameter change are analyzed, the optimum value of structure parameters of milling cutter are obtained. The design orthogonal experiment design table, based on the physical simulation of milling process to obtain the milling force and the temperature data needed in the performance prediction of multi factor structure parameters of the cutting tool, through the established BP neural network prediction model, the prediction between the multi factor structure parameters and the maximum milling force and the maximum milling temperature are realized, optimum structure parameters combination is obtained. Finally, the cutting force, temperature, dynamic characteristics and deformation of the workpiece in the machining process were compared and verified before and after optimization.(1) The 3D physical simulation model is established, which can provide technical support for the analysis of single factor structure parameters and the performance prediction of multi factor structure parameters.(2) According to the single factor analysis for the cutting tool rake angle of helix edge, flank angle of helix edge, helix angle, number of teeth of the structure parameters of, comprehensive comparison between the maximum cutting force and milling temperature value, choose the side milling cutter structure parameters adopted to cantilever plate structure typical of Ti-6Al-4V thin-walled parts.(3) Combined with the nonlinear characteristics of BP neural network, the BP neural network is used to predict the performance of multi factor structural parameters of the cutting tool. Construct the two layer BP neural network model suitable for the study of this paper. The training and simulation data are obtained through the milling simulation, and the BP neural network prediction model is established with the help of Matlab. The accuracy of the BP neural network model is verified by comparing the prediction model with empirical formula. Finally, through the comparison of the predicted values, the optimal structure parameters are obtained.(4) Establish the vertical milling cutter before and after optimization of 3D model and carry out transient dynamic analysis of the tool, and the performance of the optimized cutting tool is analyzed from the aspects of stiffness, strength and stress, respectively. Milling force, milling temperature and dynamic characteristics were compared before and after the optimization of the cutting tool. The maximum deformation of the workpiece is analyzed, and the performance of the optimized tool is verified and analyzed.Through the prediction model based on BP neural network, the theoretical model is provided for the optimization of machining tool of titanium alloy.