Development of Mathematical Model with a Genetic Algorithm for Automatic Arc Welding Process

Abstract

Gas metal arc (GMA) welding process has widely been employed due to the wide range of applications, cheap consumables, and easy handling. In order to achieve a high level of welding performance and quality, a suitable model is required to investigate the characteristics of the effects of process parameters on the bead geometry in the GMA welding process. This paper is to represent new algorithms to predict process parameters on top-bead width in robotic GMA welding process. The models have been developed, linear, curvilinear, and intelligent model, based on full factorial design with two replications. Regression analysis was employed for optimization of the coefficients of linear and curvilinear models, while genetic algorithm (GA) was utilized to estimate the coefficients of an intelligent model. Not only the fitting of these models was checked and compared by using a variance test (analysis of variance (ANOVA)) but also the prediction on top-bead width using the developed models was carried out based on the additional experiments. The developed models were employed to investigate the characteristic between process parameters and top-bead width. Resulting solutions and graphical representation showed that the intelligent model developed can be employed for prediction of bead geometry in GMA welding process.