Stamping Parameters Optimization of an AA5754 A-pillar by Response Surface Methodology

Abstract

In this paper, an A-pillar was selected as an example to investigate the effect of stamping parameters on the parts forming quality of AA5754 sheet. A finite element model was established using commercial stamping software PAMSTAMP2G. Barlat2000 yield function was used to describe yield behavior of the material. Stamping experiment was conducted to validate the reliability of the model. The studied parameters are blankholder force (100-700 KN) and drawbead’s geometrical variables, including two fillet radii R1 (8-12 mm), R2 (4-8 mm) and the height of drawbead D (2-6 mm). The central composite experiment design method has been employed to design the simulation matrix. In order to obtain stamped parts with optimal forming quality, response surface methodology was used to establish the relationship between stamping parameters and forming quality (rupture and springback). The non-domination sorting genetic algorithms II (NSGA-II) was adopted to conduct an optimal calculation of the models. A pareto optimal solution set in the solution space was obtained. A reasonable optimized scheme was selected. The optimum blankholder force is found to be 700KN with the drawbead’s geometrical parameters R1, R2 and D of 12mm, 6.6mm and 3.8mm, respectively.