Investigations on quality characteristics in gas tungsten arc welding process using artificial neural network integrated with genetic algorithm

Abstract Gas tungsten arc welding (GTAW) technology is widely used in industry and has advantages, including high precision, excellent quality, low equipment cost. However, the inclusion of a large number process parameters hinders its application on wider scale. Therefore, there need to implement prediction optimization models that effectively enhance performance GTAW different applications. In this study, five-factor five-level central composite design (CCD) matrix was conduct experiments. AISI 1020 steel blank as substrate; UTP AF Ledurit 60 68 were materials two tubular wires. Further, an artificial neural network (ANN) simulate then combined with genetic algorithm (GA) determine can provide optimal weld. experiments, five current levels, speed, distance nozzle, angle movement, frequency wire feed pulses used. Using GA, determined: = 222 A, speed 25 cm/min, nozzle deflection 8 mm, travel 25°, pulse Hz. The determination coefficient (R 2 ) RMSE value all response are satisfactory, R data remained higher than 0.65.