动力锂离子电池关键参数估计方法研究进展
Research progress of key parameter estimation methods for power lithium-ion batteries
李远茂 1刘桂雄1
作者信息
- 1. 华南理工大学机械与汽车工程学院,广东广州 510640
- 折叠
摘要
锂离子电池热失控、电化学模型参数和荷电状态等关键参数无法直接通过设备或无损方法进行直接测量,需通过复杂模型或算法进行估计,在参数估计方面仍然面临着挑战.该文对动力锂离子电池关键参数进行系统综述,包括动力锂离子电池热失控扩散建模、电化学模型参数估计、荷电状态参数估计等方法的基本原理、相关研究进展与应用,比较分析各类方法的优缺点及其适用对象,研究指出:研究电池包层级的热失控仿真技术有助于电池包的热扩散特性分析、电池热管理系统设计优化;电化学模型参数估计通常需应用元启发式算法,但选择合适的元启发算法仍需要深入研究、加强;深度学习网络+注意力机制、元启发式算法优化网络超参数等均为提升荷电状态参数估计性能重要方法.
Abstract
Key parameters such as thermal runaway,electrochemical model parameters,and state of charge of lithium-ion batteries cannot be directly measured by devices or non-destructive methods.They require estimation through complex models or algorithms,posing challenges in parameter estimation.This paper provides a systematic review of key parameters of lithium-ion batteries,including thermal runaway diffusion modeling,electrochemical model parameter estimation,and state of charge parameter estimation.It elaborates on the basic principles,research progress,and applications of various methods,comparing and analyzing their advantages,disadvantages,and applicability.It is pointed out that research on thermal runaway simulation technology at the battery pack level contributes to thermal diffusion characteristics analysis and optimization of battery thermal management system design.Electrochemical model parameter estimation typically requires the use of metaheuristic algorithms,but further research and strengthening are needed to select appropriate metaheuristic algorithms.Important methods to enhance state of charge parameter estimation performance include deep learning networks with attention mechanisms and metaheuristic algorithm optimization of network hyperparameters.
关键词
锂离子电池/热失控/电化学模型参数/荷电状态Key words
lithium-ion battery/thermal runaway/electrochemical model parameters/state of charge引用本文复制引用
基金项目
广东省重点领域研发计划项目(2019B090908003)
出版年
2024
NETL
NSTL
万方数据