Modeling and multiobjective optimization of thermal effects of fiber laser cutting of Inconel 600 sheet by employing the ANN and multi-objective PSO algorithm

In this study, the experimental results of fiber laser cutting of Inconel 600 was modeled and optimized by combining artificial neural networks (ANNs) and particle swarm optimization (PSO). The impact of cutting criteria on the temperature adjacent to the cut kerf and roughness of the cutting edge was experimentally evaluated. The independent variables are the cutting speed, focal length, and laser power. The fiber laser cutting characteristics are modeled at different cutting conditions by the ANN method according to the experimental data. The findings indicated that the ANN is performing reasonably well in dealing with the training and test datasets. Also, the multiobjective PSO has been developed to effectively optimize the laser cutting procedure parameters in order to achieve the maximum temperature (the temperature upper than 370 C-degrees) and minimum roughness (lower than 3 mu m) simultaneously in order to improve the laser cutting efficiency. Based on the PSO results, the optimal laser power gained at a laser power of 830 and 1080 W at cutting speed ranges from 2 to 4 m/min and maximum focal length ranges between 0.75 and 0.8 mm where the lowest amount of roughness was created. The optimum temperature ranges were between 370 and 419(degrees)C. At a laser power of 1000 W and speed of 4 m/min, the smooth cutting edge at minimum roughness was gained without any defects. Transmission of the focal point up to 1.5 mm below the top surface of the sheet improved the roughness of the cutting edge and the cut quality by producing the smooth surface without slags.