Lithium-ion battery state-of-health estimation in electric vehicle using optimized partial charging voltage profiles

• Published in: Energy (Oxford) Vol. 185; pp. 1054 - 1062
• Main Authors: Meng, Jinhao, Cai, Lei, Stroe, Daniel-Ioan, Luo, Guangzhao, Sui, Xin, Teodorescu, Remus
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 15.10.2019; Elsevier BV
• Subjects: Aging; Batteries; Battery chargers; Classification; Cobalt; Electric potential; Electric vehicle; Electric vehicle charging; Electric vehicles; Genetic algorithms; Information systems; Lithium; Lithium-ion batteries; Lithium-ion battery; Manganese; Nickel; Non-dominated sorting genetic algorithm; Optimization; Partial voltage range; Power management; Product safety; Rechargeable batteries; Sorting algorithms; State of health estimation; Voltage; State of health estimation; Lithium-ion battery; Partial voltage range; Electric vehicle; Non-dominated sorting genetic algorithm
• ISSN: 0360-5442; 1873-6785
• DOI: 10.1016/j.energy.2019.07.127

Lithium-ion (Li-ion) batteries have become the dominant choice for powering the Electric Vehicles (EVs). In order to guarantee the safety and reliability of the battery pack in an EV, the Battery Management System (BMS) needs information regarding the battery State of Health (SOH). This paper estimates the battery SOH from the optimal partial charging voltage profiles, which is a straightforward and effective solution for the EV applications. In order to further improve the accuracy and efficiency of the SOH estimation, a novel method optimizing single and multiple voltage ranges during the EV charging process is proposed in this paper. Non-dominated Sorting Genetic Algorithm II (NSGA-II) is applied to automatically select the optimal multiple voltage ranges, while the grid search technique is used to find the optimal single voltage range. The non-dominated solutions from NSGA-II enable the SOH estimation at different battery charging stages, which gives more freedom to the implementation of the proposed method. Three Nickel Manganese Cobalt (NMC)-based batteries from EV, which have been aged under calendar ageing for 360 days, are used to validate the proposed method. •The SOH estimation accuracy is improved in an optimization manner.•The optimal multiple voltage ranges are automatically selected by NSGA-II and grid search.•Various solutions at different battery charging stages are provided for SOH estimation.•Three NMC-based batteries are aged for 360 days to validate the proposed method.