内阻
健康状况
电池(电)
恒流
锂离子电池
电压降
遗传算法
电压
荷电状态
工程类
计算机科学
控制理论(社会学)
电子工程
汽车工程
电气工程
功率(物理)
人工智能
机器学习
物理
控制(管理)
量子力学
作者
Ning Li,Fuxing He,Wentao Ma,Ruotong Wang,Lin Jiang,Xiaoping Zhang
出处
期刊:IEEE Transactions on Vehicular Technology
[Institute of Electrical and Electronics Engineers]
日期:2022-12-01
卷期号:71 (12): 12682-12690
被引量:7
标识
DOI:10.1109/tvt.2022.3196225
摘要
Lithium-ion battery state of health (SOH) estimation technology is an important part of the design of a battery monitoring system (BMS) for electric vehicles. People often use battery capacity and internal resistance as SOH estimation indicators. However, due to actual working conditions, it is difficult for electric vehicles to achieve complete charge and discharge, so the battery capacity and internal resistance cannot be monitored online. In view of the above questions, this article proposes an indirect SOH estimation method for online EV lithium-ion batteries based on arctangent function adaptive genetic algorithm combination with back propagation neural network (ATAGA-BP). Firstly, constant current drop time (CCDT), constant current drop capacity (CCDC) and maximum constant current drop rate (MCCDR) in constant voltage charging stage are used as health indicator (HI) to evaluate battery SOH in order to indirectly quantify the degradation process of lithium-ion batteries. Error point optimization and correlation verification are also carried out. Secondly, an ATAGA-BP algorithm is proposed to establish the relationship between HI and available battery capacity, and SOH estimate is made for lithium-ion batteries according to the proposed algorithm. Finally, simulation results with NASA data show the correlation between the proposed HI and lithium-ion battery capacity is above 85%, the error of SOH estimation method proposed is 3.7%, and the iteration efficiency is increased by 17.8%.
科研通智能强力驱动
Strongly Powered by AbleSci AI