Multi-objective optimization with automatic simulation for partition temperature control in aluminum hot stamping process

Hot stamping of sheet metals with partition temperatures can effectively improve the forming quality of products. However, the varying temperature distributions also significantly raise difficulties in their design and optimization. This study proposed a new multi-objective optimization approach with automatic simulation to solve this problem in hot stamping with partition temperature control. First, a finite element model (FEM) was established to simulate the hot stamping of an aluminum box-shaped part. Then, a multi-objective optimization approach was proposed based on a non-dominated sorting genetic algorithm (NSGA-II). In this optimization, 100 sets of process parameters were generated, and they were automatically simulated in the FEM software using python code. The maximum thinning and thickening rates of the simulation results were then judged by the optimization approach. The relatively good sets would be temporarily retained, while the other sets would be optimized to generate new process parameters. All these sets of process parameters would enter a new loop until the formability of the part could not be improved. At last, a series of optimal parameters were obtained after 16 generations of iterations, and a total of 1600 simulations were conducted. This optimization approach showed a much better performance than the response surface methodology (RSM)-based optimization. It can be adopted in other simulation optimizations as it can deal with a large number of parameters with good accuracy.