New Approaches on Automated Wrinkle Detection in Sheet Metal Components by Forming Simulation

In production of passenger cars, geometry complexity of deep drawn body panels increases constantly. For that reason, sheet metal components are analyzed within finite element analysis (FEA) with regard to their feasibility in production and expected quality before production equipment, such as drawing dies, is manufactured. Main criteria for characterizing component quality are cracks and sidewall wrinkles. In particular, cracks occur due to local overload in sheet metal plane caused by inadequate process parameters such as too high friction or forming forces. In contrast, sidewall wrinkles are caused by an inadequate level of compressive stress in component areas without contact between sheet metal component and drawing die. In FEA, failure by cracks can be analyzed evaluating scalar values of thinning or strain distribution in forming limit diagram with regard to forming limit curve. In contrast, detecting sidewall wrinkles often requires a manual and visual inspection of simulation results by the user. Therefore, a procedure to detect sidewall wrinkles in an automated manner is presented in this paper. The presented method determines occurrence of sidewall wrinkles based on strain distribution in forming limit diagram. Utilization of the disclosed calculation strategy allows estimation of cracks and sidewall wrinkles simultaneously after one run of simulation code. The presented approach for automated detection of sidewall wrinkles in combination with multivariate statistics shows a tool for virtual engineering to optimize deep drawing processes. Prior to die manufacturing, optimization with regard to both sides of the process window is possible. Hence, an increase in design efficiency, design space and reduction of development time and costs can be achieved at the same time.