Research On Optimization And Simulation Prediction Of Blanking Process Parameter For Sheet Metal

Blanking is one of the basic method of metal plastic forming,which is widely used in automobiles because of its stability of dimension and high productivity.The area of blanking deformation is concentrated in the narrow die clearance,and blanking is intricate because the analysis of the blanking needs to consider material nonlinearity,contact condition nonlinearity and geometrical nonlinearity.Therefore,to study the length of shear zone is also relatively complicated.With the development of finite element simulation technology,finite element software is used to simulate the blanking process of sheet metal,it can optimize process parameters of blanking,and it improves section quality of blanking parts.Therefore,it has an important guiding significance and economic significance on the actual production of enterprises.Driven plate is taken as the research object,and DEFORM-2D is used to simulate the blanking of driven plate(the material is QSTE460 TM and the thickness is 1.5 mm),and the influence of blanking process parameters of driven plate on shear zone are analyzed.The blanking process parameters of driven plate are optimized based on response surface method.BP neural network is employed to predict the shear zone of driven plate,and progressive die is designed to verify the simulation value,the main research can be summarized as follows:(1)The finite element model of driven plate is established based on the analysis of key technology for simulation,DEFORM-2D is used to simulate blanking of driven plate,and the stress and strain state of blanking process is analyzed.The influence of blanking process parameters of driven plate on the length of shear zone is analyzed.(2)The simulation experiment of driven plate is designed based on the response surface method,and the mathematical model between the blanking process parameters and the length of smooth zone is established.The interactive influence of stamping process parameters on length of shear zone is analyzed,the conclusion is drawn that die clearances and radius of punch have the most interactive effect on length of shear zone.In addition,the optimized parameters of blanking for driven plate is obtained by response surface method.(3)BP neural network with three layers of 3-11-1 is established based on the simulation data of the response surface method,and it includes structural design,training and testing of BP neural network.In addition,the progressive die of driven plate is designed based on the process analysis of the part,and the simulation prediction value of the BP neural network and finite element simulation value are verified by stamping die.The relative errors between the experimental values and the predicted values of BP neural network are 14.81% and 7.74%,and the relative errors between the experimental values and the finite element simulation values are 6.43% and 14.78% respectively.The error is within the acceptable range of error,which verifies the correctness and feasibility of the BP neural network and finite element simulation applied to simulate the prediction of shear zone of driven plate.