Optimizing spot welding parameters in a sheet metal assembly by neural networks and genetic algorithm

Abstract

Deformation of the spot-welded sub-assemblies in assembly operations and the gap between the matching sub-assemblies have been quality concerns specifically in the automotive industry. Overall quality of the car body and its sub-assemblies, apart from quality of each stamped part, depends markedly on the welding process. This paper considers optimization of three important process parameters in the spot welding of the body components, namely welding current, welding time, and gun force. In this research, first the effects of these parameters on deformation of the sub-assemblies are experimentally investigated. Then neural networks and multi-objective genetic algorithms are utilized to select the optimum values of welding parameters that yield the least values of dimensional deviations in the sub-assemblies. Welding sub-assemblies with the optimized set of parameters brought all of them into the tolerance range. The proposed approach can be utilized in manufacturing sub-assemblies that can fit and match better with adjacent parts in the automotive body. It enhances quality of the joint and will result in improving overall quality of the body in white.