Multi-objective process parameters optimization of Laser-magnetic hybrid welding combining Kriging and NSGA-II

Laser-magnetic hybrid welding (LMW), which utilizes the stirring effect of magnetic field on molten pool to restrain the welding defect and increasing welding penetration, provides a promising way to improve the quality of welding joint. The process parameters of LMW have crucial effects on the welding seam profile which related to the quality of welding joint. This paper presents an integrated methodology by combining Kriging metamodel and Non-dominated sorting genetic algorithm-II (NSGA-II) for process parameters optimization of LMW. Firstly, a three-factor, five-level experiment using Taguchi L25 orthogonal array is conducted considering magnetic flux density (MF), laser power (LP) and welding speed (WS). Secondly, Kriging metamodel is introduced to establish the relationships between LMW process parameters and welding seam profile. A set of actual tests was carried out to verify the prediction accuracy of the constructed Kriging metamodels. Thirdly, NSGA-II is employed to finish multi-objective process parameters optimization and Pareto optimal solutions searching. Finally, the obtained optimal process parameters are validated by macro-weld profile, microstructure and micro-hardness in the confirmation tests. Results illustrate that the proposed integrated methodology is helpful for reducing welding defects and obtaining high-quality joints for LMW in practical production.