Scholarlink

<p>Gas metal arc welding is widely used for depositing corrosion-resistant austenitic stainless steel claddings on structural steel substrates, where clad bead geometry significantly influences performance. In this study, a Levy flight-based enhancement of the whale optimization algorithm is proposed to model and optimize bead geometry characteristics, including bead width, penetration depth, reinforcement height, and percentage dilution. Experimental data were generated using a structured design of experiments and used to develop predictive models. The results indicate that the proposed method significantly improves prediction accuracy, achieving coefficient of determination values above 0.91 for all output parameters, compared to approximately 0.80 obtained using the conventional approach. In addition, the root mean square error was reduced from 0.8796 to 0.6313 for clad bead width and from 0.7987 to 0.5398 for dilution. Although both methods satisfy the acceptable dilution range (10–15%), the proposed method achieves a value of 10.29%, indicating better precision and parameter tuning compared to 11.00% obtained using the conventional method. These findings confirm that the proposed optimization method provides superior predictive performance, reduced error, and improved process control, leading to enhanced cladding quality, improved mechanical performance, and more efficient material utilization.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Levy flight trajectory based whale optimization algorithm for prediction and optimization of weld clad geometry applications

  • 期刊论文

  • 发表日期：

    2026年6月17日

查看全文
  • T. Kannan,
  • R. S. Jayaram,
  • Shyni P. Nair

摘要

Gas metal arc welding is widely used for depositing corrosion-resistant austenitic stainless steel claddings on structural steel substrates, where clad bead geometry significantly influences performance. In this study, a Levy flight-based enhancement of the whale optimization algorithm is proposed to model and optimize bead geometry characteristics, including bead width, penetration depth, reinforcement height, and percentage dilution. Experimental data were generated using a structured design of experiments and used to develop predictive models. The results indicate that the proposed method significantly improves prediction accuracy, achieving coefficient of determination values above 0.91 for all output parameters, compared to approximately 0.80 obtained using the conventional approach. In addition, the root mean square error was reduced from 0.8796 to 0.6313 for clad bead width and from 0.7987 to 0.5398 for dilution. Although both methods satisfy the acceptable dilution range (10–15%), the proposed method achieves a value of 10.29%, indicating better precision and parameter tuning compared to 11.00% obtained using the conventional method. These findings confirm that the proposed optimization method provides superior predictive performance, reduced error, and improved process control, leading to enhanced cladding quality, improved mechanical performance, and more efficient material utilization.

Graphical abstract