Optimizing welding assembly operations in automotive body through using neural networks and genetic algorithm

The body of a vehicle is made up of several hundreds of stamped components which are joined together by spot-welding process. Overall quality of the car body (BIW) and quality of the sub assemblies, apart from quality of each stamped part, depends remarkably on quality of the welded joint. This paper considers optimization of the welding parameters to enhance quality of the joint resulting in improving overall quality of the body in white. The most important welding parameters in spot welding of the body components, are welding current, welding time and gun force. In this research first the effects of the aforementioned parameters on the deformation of body sub-assemblies are experimentally investigated. Then neural networks and the genetic algorithm are applied to select the optimum values of the welding parameters.