基于混合模型的锂离子电池热失控预判方法

Prediction of Lithium-ion Battery Thermal Runaway Based on Hybrid Model

锂离子电池在储能系统中已得到普遍应用,但其会由于热失控而产生自燃、爆炸等引发安全事故,如何对储能锂离子电池热失控故障风险进行超前预测和判定是当前研究的热点问题。将电热物理模型与深度学习模型长短期记忆模型(LSTM)相结合,提出一种基于混合模型的储能锂离子电池热失控预判方法。通过收集电池运行数据,利用电池的电热耦合模型进行电池内部温度、荷电状态(SOC)的估算;同时,将电池表面温度、电池电压、电池电流等参数共同作为LSTM的输入,利用混合模型精确预测电池的表面温度和内部温度。通过阈值方法判定热失控的发生并确定诱发原因,从而实现对电池热失控的准确预测。基于公开数据集的实验结果表明,提出的混合模型进行热失控预判具有较好的精确性和快速性,在实际工程应用中有着较好的应用前景。

Lithium-ion battery has been widely used in energy storage system,but it tends to give rise to spontaneous combustion,explosion and other safety accidents due to thermal runaway.How to predict and determine the failure risk of thermal runaway of energy storage lithium battery in advance is a hot issue in current research.In this paper,a prediction method of thermal runaway of energy storage lithium-ion battery based on hybrid model is proposed,combined the electrothermal physical model with the deep learning model LSTM(long short-term memory model).According to the collected operation data of the battery,the internal temperature and SOC(state of charge)of the battery are estimated with the battery electrothermal coupling model;meanwhile,with the battery surface temperature,battery voltage,battery current and other parameters taken as the input of LSTM,the hybrid model is used to accurately predict the surface temperature and internal temperature of the battery.Then the threshold method is used to determine the occurrence of thermal runaway and determine the causes,so as to realize the accurate prediction of the battery thermal runaway.The experimental results based on public data sets show that the proposed hybrid model has high accuracy and efficiency,and has a good application prospect in practical engineering application.