Optimization of Laser Cut Quality Characteristics Considering Material Removal Rate Based on Pareto Concept
Stainless steels are one of the most important engineering materials widely used in the industry. This paper presents multi-objective optimization of CO2 laser cutting of stainless steel considering different cut quality characteristics and material removal rate (MRR). Laser cutting experiment trials were conducted based on Taguchis L27 experimental design by varying the laser power, cutting speed, assist gas pressure and focus position at three levels. Using obtained experimental data, six mathematical models for the prediction of surface roughness, kerf width, kerf taper angle, width of heat affected zone, dross height and MRR were developed using artificial neural network (ANN). The developed mathematical models were taken as objective functions for the multi-objective optimization using genetic algorithm based on Pareto concept. As a result of multi-objective optimization, five 2-D Pareto fronts were generated covering all combinations of cut quality characteristics and MRR. It was observed that the mathematical relationships in the Pareto fronts between MRR and cut quality characteristics are in some cases linear and in another nonlinear.