Scholarlink

To improve the accuracy of lithium-ion battery State of Health (SOH) estimation in optical-storage-charging systems, this paper combines the strengths of LSTM and Transformer architectures to propose a hybrid LSTM-Transformer-based SOH prediction method. The method leverages LSTM to extract local temporal features while utilizing Transformer to capture global dependencies, thereby achieving a more comprehensive understanding of sequential data. Experiments conducted on the CALCE battery dataset from the University of Maryland demonstrate that the LSTM-Transformer model outperforms standalone LSTM and Transformer models in predicting SOH degradation trends.

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

Lithium-Ion Battery SOH Estimation Based on LSTM-Transformer

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Li Lei,
  • Sang Lingfeng,
  • Lin Xiongjie

摘要

To improve the accuracy of lithium-ion battery State of Health (SOH) estimation in optical-storage-charging systems, this paper combines the strengths of LSTM and Transformer architectures to propose a hybrid LSTM-Transformer-based SOH prediction method. The method leverages LSTM to extract local temporal features while utilizing Transformer to capture global dependencies, thereby achieving a more comprehensive understanding of sequential data. Experiments conducted on the CALCE battery dataset from the University of Maryland demonstrate that the LSTM-Transformer model outperforms standalone LSTM and Transformer models in predicting SOH degradation trends.