Effect of vibration and welding parameters on spot weld resistance: modeling integrating PSO-ANN and GA algorithm

• Published in: Multiscale and Multidisciplinary Modeling, Experiments and Design Vol. 7; no. 3; pp. 1755 - 1767
• Main Authors: Bousnina, Kamel, Hamza, Anis, Ben Yahia, Noureddine
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01-07-2024
• Subjects: Characterization and Evaluation of Materials; Engineering; Mathematical Applications in the Physical Sciences; Mechanical Engineering; Numerical and Computational Physics; Original Paper; Simulation; Solid Mechanics; Spot welding resistance; GA; Vibration; PSO-ANN; ANOVA
• ISSN: 2520-8160; 2520-8179
• DOI: 10.1007/s41939-023-00300-7

The estimation and improvement of the strength of 304L stainless steel spot welds were quantitatively analyzed using a hybrid method. This method combines the PSO-ANN algorithm and the genetic algorithm. This article aimed to study the effect of vibration and welding parameters on spot welding resistance. P value tests and PC% contributions were performed to assess the level of significance of the input parameters on the output response. Vibratory frequency is the most important parameter affecting displacement, with a contribution of 47.20%. The results highlight the PSO-ANN model's considerable ability to quickly establish a precise correlation between predicted values and experimental data. The optimum input parameters determined by the genetic algorithm were a welding time of eight cycles, a welding current of 3100 A, and a vibration frequency of 29.55 Hz. These values resulted in a 93.1% improvement over the maximum value.