U-shaped Body Stamping Numerical Analysis And Prediction Of Springback

In the autobody panel stamping process, the rule of sheet metal deforming is complicated and difficult to be analyzed because of the complex autobody panel structure and various forming parameters affecting the stamping process of autobody panel. With the development of the auto industry, the springback defect of the formed autobody panel becomes more and more intractable, which gets to be the interest of both auto industry and academia. Consequently, the prediction and control of the springback becomes one of the hottest and most difficult subjects in autobody panel manufacturing.U-shaped panel is a basic and classic type of autobody panel. In this thesis crashbox, a U-shaped autobody panel, is studied as an example. By the numerical simulation analyses of the stamping process and the real stamping experiment of crashbox, numerical simulation technology of autobody stamping process, springback prediction and analyses of the U-shaped auto panel and the optimal selection of forming parameters are studied.Firstly, the characteristics of autobody structure and its deforming status in the stamping process are discussed. As the theory basis of the further numerical simulation analyses in this dissertation, the key technology and theory of numerical simulation of autobody panel stamping process are discussed and studied.Based on the discussion and the study above, the parametric modeling technology applied in geometrical modeling of crashbox improves the flexibility of the modeling in the numerical simulation of the stamping process of crashbox. With the stamping process numerical simulation software, the forming process of crashbox is simulated successfully many times under various combinations of different levels of forming parameters. By means of forming process numerical simulation, the relationship between the forming parameters and the formability and the one between the forming parameters and stamping force are studied respectively.Springback is one of the primary forming defects of U-shaped autobody panel. Based on the crashbox stamping process numerical simulation experiment, orthogonal design of experiment is applied to qualitatively studying the relationship between the forming parameters (blank holder force, sheet thickness, the condition of lubrication and friction) and the springback of the crashbox flange.Based on the orthogonal springback experiment of crashbox with numerical simulation method, BP (Back Propagation) neural network method is applied to approaching the complex nonlinear relationship between the forming parameters and the springback of the crashbox flange. With the established BP neural network, the springback of crashbox flange can be respectively predicted under any combination of forming parameters given. In order to verify the feasibility and stability of the BP neuralnetwork which is applied to predicting the springback, based on the real stamping experiment of crashbox, an improved BP neural network is adopted to efficiently predict the springback distribution in the crashbox flange.In conclusion, it is a new approach to study and resolve the springback prediction and its control problem that applying the BP neural network to studying the complex and nonlinear springback prediction and analysis of autobody panel and tentatively establishing a springback prediction and analysis model. It has much important realistic meaning to improve the dimensional accuracy of the autobody panel, reduce the costs of tool manufacture and shorten the time cycle of tool development.