Scholarlink

Based on the Independent component analysis (ICA) algorithm in unsupervised learning, this paper deeply mines the latent statistical information of the dynamic response of the high-speed train bogies and realizes the independent component extraction of the measured loads and stresses of the line. In this paper, the independent components of structural loads and fatigue critical point stresses under three tests covering different operation variables are obtained. By comparing the independent components between loads and stresses, it is found that the independent component correlation of the vertical loads and the fatigue critical point stresses is significantly higher than that of the lateral loads, reaching a strong correlation level. Meanwhile, the different operation variables tested do not affect this conclusion. This conclusion can establish an underlying connection for the structural dynamic responses of the high-speed train and provide a new basis for the life optimization and assessment of the fatigue key structures.

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

Study of Dynamic Response Correlation of High-Speed Train Bogie Based on ICA Algorithm

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Zheng Yuan,
  • Xianjia Chen,
  • Lijun Ma,
  • Xiaolong Zou,
  • Zhenxian Zhang,
  • Qiang Li,
  • Shouguang Sun,
  • Yujie Wei

摘要

Based on the Independent component analysis (ICA) algorithm in unsupervised learning, this paper deeply mines the latent statistical information of the dynamic response of the high-speed train bogies and realizes the independent component extraction of the measured loads and stresses of the line. In this paper, the independent components of structural loads and fatigue critical point stresses under three tests covering different operation variables are obtained. By comparing the independent components between loads and stresses, it is found that the independent component correlation of the vertical loads and the fatigue critical point stresses is significantly higher than that of the lateral loads, reaching a strong correlation level. Meanwhile, the different operation variables tested do not affect this conclusion. This conclusion can establish an underlying connection for the structural dynamic responses of the high-speed train and provide a new basis for the life optimization and assessment of the fatigue key structures.