Optimisation of the GTA welding process using the Taguchi method and a neural network

Abstract

There are many parameters that affect gas tungsten arc (GTA) welding process quality. Conventionally, the Taguchi method has been widely used in engineering; however, with this method the desired results can only be obtained with the use of very discrete control factors, thus leading to uncertainty about the real optimum. In this study, the Taguchi method was used for the initial optimisation of the GTA welding process parameters. A neural network was then used to construct the relationships between the welding process parameters and weld pool geometry of each weldment. The optimal parameters of the GTA welding process were determined by simulating significant process parameters obtained by the Taguchi method. Experimental results are provided to illustrate the proposed approach.