Scholarlink

<p>Fatigue life assessment and structural optimization of hydraulic excavator working devices remain challenging due to the complex loading conditions and computational inefficiencies of traditional approaches. To address these limitations, this study introduces an advanced machine learning fusion model specifically designed for hydraulic excavator working components. Finite element analysis (FEA) is first conducted under critical operating conditions to generate high-fidelity data, which are used to construct a precise surrogate model incorporating objective evaluation criteria and stability assessment methodologies. This model enables quantitative fatigue reliability analysis, enhancing predictive accuracy for structural optimization. Subsequently, the framework identifies and integrates optimal machine learning algorithms tailored to the design variables' characteristics, ensuring robust performance. Finally, the fusion model is systematically combined with the NSGA-II optimization algorithm to achieve multiobjective reliability optimization, establishing a comprehensive evaluation framework that bridges fatigue analysis, algorithm selection, and structural design. The proposed approach systematically realizes fatigue reliability analysis, algorithm identification, and multi-objective optimization for hydraulic excavator components through progressive implementation stages.</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

Reliability evaluation and optimal design of hydraulic excavator working device using integrated optimal machine learning algorithm

  • 期刊论文

  • 发表日期：

    2025年7月2日

查看全文
  • Yu-Xin Liu,
  • Hong-Zhong Huang,
  • Yuan Lu,
  • Tudi Huang,
  • Zhe Deng

摘要

Fatigue life assessment and structural optimization of hydraulic excavator working devices remain challenging due to the complex loading conditions and computational inefficiencies of traditional approaches. To address these limitations, this study introduces an advanced machine learning fusion model specifically designed for hydraulic excavator working components. Finite element analysis (FEA) is first conducted under critical operating conditions to generate high-fidelity data, which are used to construct a precise surrogate model incorporating objective evaluation criteria and stability assessment methodologies. This model enables quantitative fatigue reliability analysis, enhancing predictive accuracy for structural optimization. Subsequently, the framework identifies and integrates optimal machine learning algorithms tailored to the design variables' characteristics, ensuring robust performance. Finally, the fusion model is systematically combined with the NSGA-II optimization algorithm to achieve multiobjective reliability optimization, establishing a comprehensive evaluation framework that bridges fatigue analysis, algorithm selection, and structural design. The proposed approach systematically realizes fatigue reliability analysis, algorithm identification, and multi-objective optimization for hydraulic excavator components through progressive implementation stages.