Scholarlink

<p>RV reducers are extensively employed in heavy-duty robotic joints such as robot waist. Under heavy-load conditions, the crankshaft-bearing pairs are vulnerable to fatigue failure due to the high stress between the needle roller and the crankshaft surface. The complex mechanical modeling method and parameter optimization design of RV reducers are worthy of investigation. Firstly, considering the intrinsic mechanism of contact fatigue failure in the RV reducer crankshaft-bearing system, this paper develops an integrated approach combining mechanical analysis, TEHL simulation, and subsurface stress analysis to make accurate life predictions. Next, Morris global sensitivity analysis method and Sobol variance sensitivity analysis method are applied to quantitatively evaluate the effects and interactions of design parameters, laying the groundwork for multi-stage optimization. Finally, kriging active learning is employed to achieve efficient global search of design variables. The results demonstrate that, with the proposed method, the optimization design efficiency can be improved by up to 300%. Additionally, the structural optimization, which leads to a 12.38% reduction in the maximum subsurface stress of the crankshaft in the case study, significantly alleviates the fatigue failure problem.</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

Optimization design of crankshaft-bearing system in RV reducers based on fatigue failure analysis and kriging surrogate model

  • 期刊论文

  • 发表日期：

    2026年2月23日

查看全文
  • Yunfei Cai,
  • Zehua Han,
  • Shengrong Chen,
  • Zhaoliang Cui,
  • Jianqi Guan,
  • Xianghui Meng

摘要

RV reducers are extensively employed in heavy-duty robotic joints such as robot waist. Under heavy-load conditions, the crankshaft-bearing pairs are vulnerable to fatigue failure due to the high stress between the needle roller and the crankshaft surface. The complex mechanical modeling method and parameter optimization design of RV reducers are worthy of investigation. Firstly, considering the intrinsic mechanism of contact fatigue failure in the RV reducer crankshaft-bearing system, this paper develops an integrated approach combining mechanical analysis, TEHL simulation, and subsurface stress analysis to make accurate life predictions. Next, Morris global sensitivity analysis method and Sobol variance sensitivity analysis method are applied to quantitatively evaluate the effects and interactions of design parameters, laying the groundwork for multi-stage optimization. Finally, kriging active learning is employed to achieve efficient global search of design variables. The results demonstrate that, with the proposed method, the optimization design efficiency can be improved by up to 300%. Additionally, the structural optimization, which leads to a 12.38% reduction in the maximum subsurface stress of the crankshaft in the case study, significantly alleviates the fatigue failure problem.