| 引用本文: | 张朝龙,赵筛筛,何怡刚.基于集成经验模态分解与集成机器学习的锂离子电池剩余使用寿命预测方法[J].电力系统保护与控制,2023,51(13):177-187.[点击复制] |
| ZHANG Chaolong,ZHAO Shaishai,HE Yigang.Remaining useful life prediction method for lithium-ion batteries based on ensemble empiricalmode decomposition and ensemble machine learning[J].Power System Protection and Control,2023,51(13):177-187[点击复制] |
|
| 摘要: |
| 准确预测储能锂离子电池剩余使用寿命(remaining useful life, RUL)对于电力系统的安全性与可靠性至关重要。针对锂离子电池老化轨迹呈现非线性变化的问题，提出一种基于集成经验模态分解(ensemble empirical mode decomposition, EEMD)和集成机器学习的锂离子电池剩余使用寿命预测方法。首先，利用集成经验模态分解算法分解锂离子电池老化数据。其次，分别利用集成的长短时记忆神经网络与相关向量机对分解得到的残差数据序列和本征模态数据序列建模预测。最后，融合预测的残差数据序列和本征模态数据序列，综合计算锂离子电池未来寿命老化轨迹。采用储能锂离子电池老化数据进行验证，结果显示所提出的锂离子电池RUL预测方法具有更好的鲁棒性与非线性跟踪能力。 |
| 关键词: 锂离子电池 剩余使用寿命预测 集成经验模态分解 相关向量机算法 长短时记忆神经网络 |
| DOI：10.19783/j.cnki.pspc.221746 |
| 投稿时间：2022-11-03修订日期：2022-12-07 |
| 基金项目:国家重点研发计划项目资助(2020YFB0905905)；金陵科技学院高层次人才科研启动基金项目资助(jit- rcyj-202202) |
|
| Remaining useful life prediction method for lithium-ion batteries based on ensemble empiricalmode decomposition and ensemble machine learning |
| ZHANG Chaolong1,2,ZHAO Shaishai3,HE Yigang2 |
| (1. College of Intelligent Science and Control Engineering, Jinling Institute of Technology, Nanjing 211169, China;
2. School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, China; 3. School of Electronic
Engineering and Intelligent Manufacturing, Anqing Normal University, Anqing 246011, China) |
| Abstract: |
| A precise prediction of the remaining useful life for energy storage lithium-ion batteries is critical to the safety and reliability of power systems. To solve the problem of serious nonlinear changes of the aging trajectory of lithium-ion batteries, this paper proposes an ensemble empirical mode decomposition (EEMD) and an ensemble machine learning-based RUL prediction method. First, the measured raw lithium-ion battery aging data are decomposed using the EEMD algorithm. Then, a long short-term memory (LSTM) neural network and the relevance vector machine (RVM) algorithm are integrated and applied to model and predict the residual sequence and the intrinsic mode sequences obtained by decomposition. Finally, the future lifespan aging trajectory of the lithium-ion battery is acquired by fusing the predicted residual sequence and the intrinsic mode sequences. The aging data of energy storage lithium-ion batteries are employed to validate the proposed method. The results show that the proposed RUL prediction method for lithium-ion batteries has better robustness and nonlinear tracking ability. |
| Key words: lithium-ion battery remaining useful life prediction ensemble empirical mode decomposition RVM algorithm LSTM neural network |
