An integrated evolutionary approach for modelling and optimization of laser beam cutting process

This paper presents a new integrated methodology based on evolutionary algorithms (EAs) to model and optimize the laser beam cutting process. The proposed study is divided into two parts. Firstly, genetic programming (GP) approach is used for empirical modelling of kerf width (Kw) and material removal rate (MRR) which are the important performance measures of the laser beam cutting process. GP, being an extension of the more familiar genetic algorithms, recently has evolved as a powerful optimization tool for nonlinear modelling resulting in credible and accurate models. Design of experiments is used to conduct the experiments. Four prominent variables such as pulse frequency, pulse width, cutting speed and pulse energy are taken into consideration. The developed models are used to study the effect of laser cutting parameters on the chosen process performances. As the output parameters Kw and MRR are mutually conflicting in nature, in the second part of the study, they are simultaneously optimized by using a multi-objective evolutionary algorithm called non-dominated sorting genetic algorithm II. The Pareto optimal solutions of parameter settings have been reported that provide the decision maker an elaborate picture for making the optimal decisions. The work presents a full-fledged evolutionary approach for optimization of the process.