Abstract:
This paper presents a methodology for springback compensation in sheet metal stamping using a successive response surface method. The process leverages an optimization method that employs a D-optimality criterion for experimental design and successive linear response surface approximations to minimize the deviation between simulation results and the target design. The procedure, implemented in the software LS-OPT and integrated with the simulation package LS-DYNA and the parametric preprocessor TrueGrid, utilizes design variables to modify the tool geometry and process parameters. The NUMISHEET 96 S-Rail is used as a benchmark to demonstrate the effectiveness of this approach, with a converged design achieved in four or five iterations. The study highlights how modifying tool geometry based on a parametric model and an iterative optimization scheme can effectively compensate for springback.
