Performance Optimization of Ti-Steel Laser-Welded Joints Based on Optimized Support Vector Machine and Multi-Objective Salp Swarm Algorithm

<div>The dissimilar welding of titanium to steel enables the integration of the advantageous properties of both metals, facilitating the design of lightweight, corrosion-resistant, and high-strength multifunctional composite structures. However, significant differences in their thermophysical properties pose substantial technical challenges in practical welding scenarios, necessitating careful selection of process parameters to enhance the quality and performance of the weld joint. This article establishes a support vector machine (SVM) model with laser power, welding speed, and laser spot diameter as independent variables, and the maximum residual stress and minimum yield strength of the weld joint as dependent variables. To improve prediction accuracy, the SVM model is optimized using the beluga whale optimization (BWO) algorithm. Taking the established model as the objective function, the multi-objective salp swarm algorithm (MSSA) is employed to optimize the laser welding process parameters for titanium–steel dissimilar metal welding. Simulation experiments validate the efficacy of this optimization approach.</div>