Research On Tolerance Robust Optimization Of Sheet Metal Forming Based On The Improved Response Surfacr Model

Sheet metal forming is widely used in the field of automobile, mould and aviation, which is a common processing method in the modern industrial production. The numerical simulation of the sheet metal forming technology has become a hotspot in the field of the sheet metal stamping with putting forward of the sheet metal forming simulation software. The forming defects can be explored in the early design through the finite element simulation of the numerical simulation technology. With the forming defects, the design parameters of the mould can be modified, which can reduce the defective index of the forming parts and the cost of the design.The forming results can be forecasted with the sheet metal forming numerical simulation software, but the numerical simulation needs to consume a certain amount of time, especially for the complex forming parts need a lot of time. In order to save time, the result of the simulation can be obtained by a small amount of experiment design, and an agent model can be constructed with the similar accuracy of the original model. At the same time, because of the fluctuation of the factors of the technological parameter, the external environment and the processing methods in the process of the sheet metal forming, the quality of the worked products will be affected by these factors. In order to reduce the effects of these fluctuations, the robust design, which makes the fluctuations of the factors insensitive to the quality of the products, must be applied to optimize the factors.The mapping relationship, between the forming goals and the design variables in the sheet metal stamping forming, can be built with the agent model of response surface model. In the case of a small amount of sample points, the response surface method can be improved by the interpolation of the radial basis function with the residual error of response surface model. The improved response surface model can be used in the nonlinear function. Compared with the unimproved model, the accuracy of the model is increased.In order to improve the robustness of sheet metal forming, the ideas of tolerance can be applied to the robust design. The robust models, which include fixed tolerance robust model, unfixed tolerance robust model and two step tolerance robust model, are summarized by the ideas of tolerance. The fixed tolerance robust design model and the two step tolerance robust model can be applied to the robust optimization of the square box and the fender respectively. The relationship of the forming and constraint index and design variables can be built by the response surface method. Considering the interaction of the controllable factors and uncontrollable factors synthetically, the interaction can be used to robust optimization with the square box and the fender. The mean and variance can be obtained by Monte Carlo Simulation. Compared with deterministic optimization, the result shows that the robustness of the robust model of considering the interaction is improved significantly and the fluctuation is even smaller.