Robust design of sheet metal forming process based on adaptive importance sampling

Optimization methods have been widely applied in sheet metal forming area. However, the existence of variations during manufacturing processes significantly may influence final product quality, rendering non-robust optimal solutions. In this paper, experiments were conducted to investigate how a stochastic behavior of noise factors affected stamping quality. Robust design models for sheet metal forming process integrated adaptive importance sampling with response surface method, in order to minimize impact of the variations and achieve reliable process parameters. Support Vector Machine with nonlinear capability in both pattern recognition and regression was adopted to map the relation between input process parameters and part quality. A cup drawing example was employed to verify the feasibility of the proposed method. Comparisons were conducted between different optimization models to demonstrate robustness of the adaptive importance sampling method. Final results showed that the stamping part quality was optimized under the specified constraint requirements.