Prediction of optimal cut quality characteristic of Inconel 718 sheet by genetic algorithm and particle swarm optimization

Selection of an appropriate optimal range of cutting parameters during the machining of Inconel-718 is quite essential to achieve the high-quality cut and is a challenging task within this domain of study. The aim of this research is to develop a robust prediction model which can suggest the desired range of cutting parameters for accomplishing better cutting quality, precision, and geometrical accuracy. Experiments have been performed on a 300 W (CNC-PCT 300) pulsed Nd: YAG laser cutting system at various levels of input cutting parameters, viz., gas pressure, standoff distance, cutting speed, and laser power. Thereafter, the artificial neural network technique has been adopted to develop mathematical models in terms of aforementioned input cutting parameters for geometrical quality characteristics: Top Kerf Width, Bottom Kerf Width, and Kerf Taper. These developed models have been validated by comparing the predicted values with the experimental ones. Furthermore, these models have been optimized using a multiobjective genetic algorithm and particle swarm optimization techniques, in order to ascertain the optimal range of cutting parameters pertaining to better quality cut with high precision and geometrical accuracy.