Simulation Research On Optimization Of Fine Blanking Process Parameters Based On NSGA-?

The intelligent optimization algorithm has been greatly developed and applied in recent years.Based on numerical simulation and artificial intelligence algorithm,this dissertation comprehensively optimizes the six parameters of the stamping process.A multi-objective optimization model with optimized wear target and bright strip length is provided.The main research content includes:(1)Proe is used to design molds that meet the structural and forming requirements according to the characteristics of the fine blanking parts.The precision blanking process and die blanking process were studied,and the die failure mode was analyzed.(2)Comprehensively consider the die cavity clearance,cutting edge fillet radius,blanking speed,friction coefficient,blank holder force and top thrust of these six aspects,and the brightness of the length of the standard as a fine evaluation of fine stamping parts and die wear This quantity is used as a tool life evaluation standard.The larger the bright band value,the better the forming quality and the smaller the amount of wear,which indicates that the die life is longer.The Deform-2D module in Deform finite element analysis software was used to study the influence of various factors on the quality of section forming in the manner of control variables,and Deform-3D module was used to study the influence of various factors on the section forming quality.Wear of the mold through orthogonal experiments.The data analysis obtained is very poor.(3)Combining the CAE technique and the orthogonal experimental design method,the influence of the size on the forming quality and the amount of die wear was sorted by grey correlation analysis.The analysis shows that the reaction force and die clearance are the main causes of die wear,followed by the friction coefficient,blank holder force,edge radius and punching speed.Embossing gaps and reaction forces are also the main factors affecting the quality of this part.Speed,edge radius,blank holder force and friction coefficient are secondary factors.(4)Using the die wear amount and the length of the bright strip as the evaluation indexes of the die life and the cross-sectional quality respectively,using the BP neural network intelligent predictive model to train the data samples corresponding to different forming processes to establish the “black box” model,and then use the multi-targets..The genetic algorithm is further optimized to obtain the "optimal" Pareto solution set.It can effectively control the forming quality and die life of fine blanks obtained by combining neural network and genetic algorithm optimization process parameters.