基于机器学习方法的锂离子电池RUL预测方法  

The Lithium-Ion Battery Remaining Useful Life (RUL) Prediction Method Based on Machine Learning

作　　者：刘佳琛 张东 LIU Jia-chen;ZHANG Dong(School of Electric Power,Shenyang Institute of Engineering,Shenyang Liaoning110136,China;School of New Energy,Shenyang Institute of Engineering,Shenyang Liaoning110136,China)

机构地区：[1]沈阳工程学院电力学院,辽宁沈阳110136 [2]沈阳工程学院新能源学院,辽宁沈阳110136

出　　处：《机电产品开发与创新》2024年第4期108-111,115,共5页Development & Innovation of Machinery & Electrical Products

摘　　要：随着新能源汽车和电力系统电化学储能发展的需要,锂离子电池正在得到社会的普遍关注。然而,锂离子电池由于其内部化学反应所导致的储能能力会逐渐减弱,鉴于此,本研究将焦点放在了储能电池剩余寿命的问题上,并以神经网络为基础进行深入研究。首先,文章介绍了基本的神经网络结构,然后利用神经网络对大量数据进行了训练和学习。通过这一过程,成功构建了储能电池剩余寿命的预测模型,为电池寿命问题提供了一种全新的解决途径。通过建立预测模型,我们不仅能够更好地理解储能电池的寿命特性,还为其使用和管理提供了有效的参考依据。这一研究成果对电网调峰调频中储能电池的合理运用具有重要意义。With the development of electrochemical energy storage in new energy vehicles and power systems,lithium-ion batteries are getting widespread attention from the society.However,the energy storage capacity of lithium-ion batteries will gradually weaken due to their internal chemical reactions,so this study focuses on the remaining life of energy storage batteries and conducts in-depth research based on neural networks.First,the basic neural network structure is introduced,and then the neural network is used to train and learn from a large amount of data.Through this process,a prediction model of the remaining life of the energy storage battery is successfully constructed,which provides a new way to solve the battery life problem.By establishing a prediction model,we can not only better understand the life characteristics of energy storage batteries,but also provide an effective reference for their use and management.The research results are of great significance for the rational use of energy storage batteries in power grid peak regulation and frequency regulation.

关 键 词：锂离子电池 神经网络 寿命衰减 CNN LSTM 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]