Parameters optimization in plasma arc cutting of AISI 1020 mild steel plate using hybrid genetic algorithm and artificial neural network

Abstract

The aim of this study was to optimize the cutting parameters such as cutting speed, standoff distance, cutting current and gas pressure of the CNC plasma arc cutting process that affected the material removal rate, surface roughness and nozzle diameter change after cutting performed on AISI 1020 mild steel plate. Three levels of variation were taken to the four cutting parameters that were chosen. Twenty-seven trial experiments were carried out using L27 orthogonal array of Taguchi design. In this experimental investigation, the highest material removal rate (MRR) of 8.96 g/s, Ra surface roughness (SR) of 15.734 µm and nozzle orifice diameter (ND) of 1.4637 mm were achieved, whereas the lowest obtained values of MRR, SR and ND were 2.324 g/s, 5.98 µm and 1.2114 mm, respectively. For modeling the plasma arc cutting process experimental input parameters and responses' results, a hybrid ANN-GA model was constructed. This model was used to forecast and optimize MRR, SR and ND, as well as the control factors that go with it. The results indicated that the ANN-GA model could predict the output responses with a mean square error of 1.06885e–1. During optimization, a 4-9-3 network trained with neural network of back propagation by Levenberg-Marquardt algorithm was used to have the greatest prediction capability, with optimum values of MRR, SR and ND of 7.0032 g/s, 4.2062 µm and 1.3142 mm, respectively. From the confirmation tests, the average results of 6.9247 g/s of MRR, 4.3429 µm of SR and 1.3703 mm of ND were obtained. The percentage of errors between the ANN-GA predicted optimal responses' results and the confirmatory experimental results were found 1.121%, 3.250% and 4.269% for MRR, SR and ND, respectively.