Scholarlink

With the rapid development of electric vehicles (EVs), the reducer, as a key component of the electric drive transmission system, plays a critical role in ensuring vehicle reliability and safety. This study focuses on the gear components of the reducer within EV electric drive systems. By employing a multi-scale composite attention model and a hybrid LSTM-DNN model, combined with two different aging label generation methods—namely, the linear RUL (Remaining Useful Life) labeling and the dimension-reduced RUL labeling—this work achieves accurate prediction of degradation trends. Public bearing datasets and a self-constructed non-full-lifecycle reducer dataset were utilized for validation. The proposed approach offers new solutions for the health management of complex mechanical systems.

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

Degradation Trend Prediction of Components in Electric Vehicle Reducers

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Zizhen Qiu,
  • Xiaoyu Li,
  • Xu Liu,
  • Jialin Li,
  • Lingxiao Zhao,
  • Yang Kang,
  • Fang Wang

摘要

With the rapid development of electric vehicles (EVs), the reducer, as a key component of the electric drive transmission system, plays a critical role in ensuring vehicle reliability and safety. This study focuses on the gear components of the reducer within EV electric drive systems. By employing a multi-scale composite attention model and a hybrid LSTM-DNN model, combined with two different aging label generation methods—namely, the linear RUL (Remaining Useful Life) labeling and the dimension-reduced RUL labeling—this work achieves accurate prediction of degradation trends. Public bearing datasets and a self-constructed non-full-lifecycle reducer dataset were utilized for validation. The proposed approach offers new solutions for the health management of complex mechanical systems.