Estimation and Comparison of Welding Performances using MRA and GMDH in P-GMAW for ASTM 106 Material

Abstract The Pulsed Gas Metal Arc Welding (P-GMAW) process is one of the most significant arc welding processes, used in high-technology industrial applications. In order to understand and control the P-GMA welding process parameters, it is necessary to determine the input and output relationship of the welding processes. P-GMAW is widely used process, especially in thin sheet metal industries. It offers an improvement in quality and productivity over regular Gas Metal Arc Welding (GMAW). The process enables stable spray transfer with low mean current and low net heat input. This paper describes the estimation and comparison of welding process parameters viz., current, gas flow rate and wire feed rate on ultimate yield strength, ultimate tensile strength, percentage of elongation and hardness. Experiments have been performed based on Taguchi’s L 27 standard orthogonal array. Estimation of welding performances have been carried out using sophisticated mathematical models viz., MRA and GMDH, and, compared. The GMDH algorithm is designed to learn the process by training the algorithm with the experimental data. Three different criterion functions, viz., regularity, unbiased and combined criterions were considered for estimation in GMDH. Different GMDH models can be obtained by varying the percentage of data in the training set and the best model can be selected from these, viz., 50%, 62.5% & 75%. Estimation and comparison of welding performances were carried out using MRA and GMDH techniques.