Simulation and Optimization of U-bending Springback Using Genetic Algorithms

Springback during unloading affects the dimensional accuracy of sheet metal parts. This paper proposes a finite element model to predict springback with contact evolution between the sheet and dies. The underlying formulation is based on updated Lagrangian elastoplastic materials model. The solutions validated with experimental data of NUMISHEET'93 show more accurately. The effects of the variable blank holding force (VBHF) on springback results are investigated based on genetic algorithms (GAs) for the determination of the parameters in blank holding operations. It has been found that the GAs based optimization technique is very effective in solving this kind of problem. The difficulty of choosing correct starting values for the constants in the traditional optimization techniques has been completely overcome and the GAs technique provides a better chance to converge to the global minimum.