Optimization of Surface Roughness in Laser Cutting Process of Mild Steel Using RSM and GA Method

In the laser cutting process, cutting parameters play a crucial role in determining the quality of the cut surface. In this study, a fiber laser machine was used to cut mild steel S355 with a thickness of 12 mm. The response surface method (RSM) and genetic algorithm (GA) were employed to optimize the cutting parameters assuming the surface roughness as the response of the process. The main parameters investigated are cutting speed, laser power, gas pressure, focus position, nozzle diameter, and stand-off. Due to the importance of the surface roughness of structural steel elements cut by means of thermal processes in agreement with the standard EN ISO 9013 the correlation between the input parameters and the surface roughness was investigated. In detail, analysis of variance (ANOVA) was used to assess the model and investigate significant parameters and their interactions on the surface roughness. The nonlinearity of the problem and the complex influence of the parameters and their interaction on the roughness was studied considering all six cutting parameters which were not studied well by others before. Eventually, after optimization process and finding the optimal parameters, the validation of the study was carried out on additional specimens cut using predicted-optimized laser cutting parameters. The good agreement between experimental and predicted surface roughness highlights the capability of the approach to drive the choice of the laser cutting parameters.