Optimal design for dual laser beam butt welding process parameter using artificial neural networks and genetic algorithm for SUS316L austenitic stainless steel

Abstract

Gas porosity is very critical factor to affect the welding performance and mechanical properties of weld bead. Dual laser beam welding, as a new laser welding process, is helpful to reduce the porosity number in weld bead because of several advantages like enlarging the molten pool, increasing the fluid speed, stabling the keyhole geometry. However, dual laser beam welding requires consideration of more parameters (beam spacing, energy distribution ratio) than single laser beam welding. To optimize dual beam welding process, a 16 groups Taguchi approach together with artificial neural networks (ANN) and genetic algorithm (GA) has been applied to obtain the best welding parameters of laser beam welding for 316L austenite stainless steel during dual beam laser beam welding. X-ray detection results are regarded as the basis for evaluation of porosity number and welding quality. After optimizing, the porosity number is significantly decreased compared to original weld bead. The comparison of microstructure, and yield strength between original parameters and optimized one is carried out to verify optimal results.