Scholarlink

<p>DP1000 dual-phase steel is widely used in automotive lightweighting, but heat-affected zone softening and welding deformation during welding severely impact vehicle structural safety and dimensional accuracy—a common challenge in ultra-high strength steel welding. To address this, we designed experiments via central composite design (CCD), obtained TIG welding data of DP1000 steel through finite element software, and established nonlinear models between process parameters and welding quality (yield strength, deformation) using response surface methodology (RSM). The models exhibit high reliability (<i>P</i>-values &lt; 0.0015, R<sup>2</sup> = 0.9103 and 0.9885). We compared the multi-objective grey wolf optimizer (MOGWO) with NSGA-II, MOPSO, and MOEA/D, then optimized MOGWO using logistic chaotic mapping. Simulation Results show MOGWO outperforms other algorithms in multi-objective optimization, and chaotic mapping further improves its performance (SP reduced by 12.21%, Δ increased by 18.16%, ART shortened by 11.53%), yielding 79 Pareto optimal solutions. This RSM-MOGWO strategy provides an efficient method for ultra-high strength steel welding optimization, offering references for similar material quality control and intelligent algorithm engineering applications.</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

Research on quality optimization strategy of DP1000 dual phase steel TIG welding based on RSM-MOGWO

  • 期刊论文

  • 发表日期：

    2026年5月30日

查看全文
  • Zhu Yubin,
  • Meng Xiangli

摘要

DP1000 dual-phase steel is widely used in automotive lightweighting, but heat-affected zone softening and welding deformation during welding severely impact vehicle structural safety and dimensional accuracy—a common challenge in ultra-high strength steel welding. To address this, we designed experiments via central composite design (CCD), obtained TIG welding data of DP1000 steel through finite element software, and established nonlinear models between process parameters and welding quality (yield strength, deformation) using response surface methodology (RSM). The models exhibit high reliability (P-values < 0.0015, R2 = 0.9103 and 0.9885). We compared the multi-objective grey wolf optimizer (MOGWO) with NSGA-II, MOPSO, and MOEA/D, then optimized MOGWO using logistic chaotic mapping. Simulation Results show MOGWO outperforms other algorithms in multi-objective optimization, and chaotic mapping further improves its performance (SP reduced by 12.21%, Δ increased by 18.16%, ART shortened by 11.53%), yielding 79 Pareto optimal solutions. This RSM-MOGWO strategy provides an efficient method for ultra-high strength steel welding optimization, offering references for similar material quality control and intelligent algorithm engineering applications.

Graphical abstract