Leveraging the potential of the modified WOA to achieve optimal hardness through appropriate parameter selection in SAW

Abstract

Submerged Arc Welding (SAW) is a widely used welding process in heavy-duty industrial applications. Weld hardness is a critical factor influencing the mechanical properties and performance of welded joints. However, achieving optimal hardness can be challenging due to the complex interplay of process parameters. This research highlights the potential of a modified approach based on the Whale Optimization Algorithm (WOA) as a reliable optimization tool for optimizing hardness settings in a SAW process, thereby contributing to advancements in welding technology and industrial applications. The suggested method introduces several unique features: a dynamic search space adjustment mechanism to adaptively refine the search area, and the incorporation of inertia weights to balance the algorithm’s exploration and exploitation phases. These novel modifications significantly enhance the algorithm’s performance, leading to faster convergence and more precise solutions. The novelty of the proposed method lies in the introduction of a dual-phase adaptive control strategy that adjusts both the search step size and the influence of elite solutions over time. This allows the algorithm to balance global and local searching more effectively. Additionally, a parameter sensitivity feedback loop is incorporated to dynamically refine the influence of each welding parameter based on real-time performance trends, significantly enhancing the robustness and adaptability of the optimization process. To validate the effectiveness of the modified WOA, extensive experiments were conducted on a SAW machine using various welding parameters as input variables and weld hardness as the primary objective function. The modified WOA was compared with the standard WOA and other commonly used optimization algorithms (Particle Swarm Optimization). The results demonstrate that the modified WOA effectively identifies optimal hardness by selecting appropriate parameters, significantly outperforming the other algorithms, resulting in higher weld quality and stronger mechanical properties.