Multi-Objective Welding-Parameter Optimization Using Overlaid Contour Plots and the Butterfly Optimization Algorithm

Area of Science:

• Materials Science
• Mechanical Engineering
• Computational Optimization

Background:

• Welding processes often introduce distortion and residual stress, negatively impacting component dimensional accuracy and fatigue life.
• Minimizing distortion and residual stress are critical objectives in welding, but often conflict, necessitating a multi-objective optimization approach.

Purpose of the Study:

• To perform a multi-objective optimization to simultaneously minimize welding-induced residual stress and distortion.
• To investigate the conflicting nature of parameters that minimize distortion versus those that minimize residual stress.
• To apply the butterfly optimization algorithm (BOA) for the first time to a multi-objective welding problem.

Main Methods:

• Selected welding parameters: current, voltage, and welding speed.
• Utilized response surface methodology to generate contour plots for distortion and residual stress.
• Employed an overlaid contour plot approach to identify a feasible region for compromise parameters.
• Applied the butterfly optimization algorithm (BOA) within the identified feasible region.

Main Results:

• Identified that parameters minimizing distortion differ significantly from those minimizing residual stress.
• Demonstrated that minimizing distortion can intensify residual stresses, highlighting the need for a compromise.
• Successfully applied the BOA to find optimal parameters balancing both objectives.

Conclusions:

• The developed framework, integrating response surface analysis and BOA, effectively addresses the multi-objective challenge in welding optimization.
• Weld simulation and experimental validation confirmed the efficacy of the optimized parameters in managing distortion and residual stress.
• This study presents a novel and effective computational framework for optimizing welding processes.