BIW Optimization by Means of an Automated CAE Process for the Reduction of Welding Points

摘要

The reduction of vehicle manufacturing costs is one of the most important targets in the development phase. One way to reduce this cost is by means of the reduction of welding points. Another challenge lies in reducing the development time. Both for reduction of costs as well as development time, simulation is increasingly important. This paper explains how both objectives can be achieved by means of spot weld optimization applying the finite-element method. The BIW has to satisfy certain requirements. The most important requirements concern fatigue, stiffness and crashworthiness. Some requirements may cause conflicting solutions, for example, a good solution for stiffness (such as a high eigenvalue), could be an unacceptable solution for crashworthiness. For this reason, joints or other parts are sometimes oversized. In order to reduce the cost of the project and simplify the manufacturing process, an optimization study is performed. In this project, all spot welds used in the manufacturing process of the BIW will be analysed with optimization software (PAM-OPT) for all load cases mentioned earlier (crashworthiness, stiffness and fatigue). The objective is to minimize the number of welds and improve the defined targets. After this spot weld reduction, the BIW must still meet all structural targets defined in the project. This means that the target in BIW stiffness (first eigenvalue), crash targets (pulse, deformation pattern, intrusions) and fatigue targets must be respected. In order to perform the optimization, it is necessary to define the target (e.g. spot weld reduction), the constraints for each load case (structural targets), the best optimization algorithm to be used and design parameters, as well as considering issues such as optimizing CPU usage. This study is limited by the amount of available CPU's. Analysing different load cases for a complete BIW, implies a significant calculation capacity, especially for load cases calculated with the explicit method (offset crash, side crash, etc.). The limited availability of CPUs could mean that only a partial study of the vehicle is feasible.