Research On Milling Process And Cutting Parameters Inversion Of Splicing Die With Different Hardness

The design and manufacture of the die of automobile cover panel accounts for about 2/3 of the automobile development cycle,which has become the main restrictive factor of automobile modification.Because of its large self-weight and size of the large-scale automobile panel die,if the whole quenching and cutting process is directly carried out,not only the processing is very difficult,but also the profile accuracy is difficult to control effectively.The problems mentioned above can be effectively solved by integral milling of conbination hardened steel die after splicing.Compared with general milling,splicing die milling has two characteristics: seam impact and abrupt hardness change.Seam impact and abrupt hardness change will cause a step change in cutting force and vibration,thus accelerating tool wear and reducing the surface accuracy of the workpiece.It is an effective way to effectively reveal the tool wear characteristics,vibration characteristics and influence mechanism in milling process to solve the difficulties of splicing die processing.Aiming at the practical problems encountered in the production process of automobile panel splicing die,the prediction models of cutting force,cutting vibration,milling stability,surface topography and roughness were established.The multi-objective optimization and inversion of milling parameters were realized by combining NSGA-II,KPCA,MPGA and ANN algorithms.The specific research contents include:According to the cutter-workpiece contact relationship in the splicing area and the different hardness of the material,the instantaneous chip thickness in the splicing area was established,and the shear force model of rake face was obtained by combining the instant rigidity force model.According to the distribution of normal stress and shear stress in the wear area of the flank,the friction model of the flank was obtained by integral method.By introducing a single-degree-of-freedom Italic collision model and combining with Hopkinson pressure bar test to obtain the elastic deformation of tool-to-work collision,the impact force in the splicing area was obtained.Combining shear force,friction force and impact force model,the prediction model of milling force in the slotted area was established.The good consistency between the experiment and simulation under the two conditions proves the correctness of the milling for ce model.The research results can effectively solve the problem of accurate characterization of milling force in splicing area.The dynamic shear force on the tool rake face was built as a function of the dynamic chip thickness,and the dynamic contact force on the tool flank face caused by wear was built as a function of the total volume of extrusion material.The dynamic equation considering the impact of through-seam and tool flank wear was obtained by combining the impact force in the splicing area.The state term,time delay term and time periodic coefficient term in the state space form of dynamic equation were discretized by multi-order Lagrangian interpolation,and a fourth-order complete discretization method was proposed.Based on the proposed fourth-order fully discrete method,the multi-period milling stability lobe was obtained,and the milling stability lobe band in the splicing region was obtained by combining the maximum and minimum envelope method.The correctness of the dynamic model was verified by the time-frequency domain method such as Fourier transform and milling vibration and milling force data collected from the machining process.The research results can effectively solve the problem of non-stationarity in the milling process of splicing die.Based on the instantaneous impact force obtained by the single-degree-offreedom slant impact model and the damped vibration equation,the impact vibration of the milling system was obtained.The fourth-order Runge-Kutta method was used to solve the dynamic equation of the milling system,and the cutting vibration caused by the varying chip thickness of the milling system was obtained.The cutting edge equation was modified according to the tool flank wear,and the prediction model of machined surface topography and roughness considering the dynamic characteristics of cutting system and tool flank wear was obtained by Z-MAP algorithm.A series of experiments were carried out to verify the correctness of the model.Based on Taguchi method and grey relational analysis method,the sensitivity of surface roughness to process parameters(forward inclination angle,side inclination angle and cutting parameters)was obtained.The research results can solve the problem of profile accuracy prediction in splicing area.Taking the splicing die of convex surface with fixed curvature as the research object,a multi-objective optimization model was established,which takes surface roughness,tool life and material removal rate per unit time as objective functions,and takes milling parameters and milling stability as constraints.Multi-objective optimization of milling parameters was realized by using NSGA-II algorithm.KPCA algorithm was used to analyze the optimized milling parameters by principal component analysis.Taking the optimized milling parameters as sample data,taking surface roughness,tool life and material removal rate per unit time as input,spindle speed,axial cutting depth and feed per tooth as output,and combining MPGA-ANN algorithm,the milling parameters were inverted.At the same time,the simulation platform of milling process in splicing area was built by using MATLAB/GUI module.The research results can solve the uncontrollable problem of the surface accuracy in splicing area.