Novel method for modelling and adaptive estimation for SOC and SOH of lithium-ion batteries. (June 2023)

• Record Type: Journal Article
• Authors: Li, Zuxin; Shen, Shengyu; Zhou, Zhe; Cai, Zhiduan; Gu, Weimin; Zhang, Fengying
• Is Part Of: Journal of energy storage. Volume 62(2023)
• Publication Date: 2023-06
• Subjects: State of charge -- State of health -- Lithium-ion battery -- Improved fractional-order model -- Endogenous immune algorithm -- Adaptive dual square root Kalman filtering; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2023.106927
• Languages: English
• ISSNs: 2352-152X

Abstract

Abstract: Equivalent circuit models of lithium-ion batteries are the foundation of accurately estimating state of charge (SOC) and state of health (SOH). To further describe the performance of the batteries between working and idling states, we present for the first time the analysis of an improved fractional-order model (IFOM). Moreover, an endogenous immune algorithm (EIA) is proposed to identify the parameters of the model. Meanwhile, considering the aging effect, an adaptive dual square root Kalman filtering (ADSRCKF) with the dormancy zone is proposed to achieve the estimation. The first filter in ADSRCKF estimates SOC and SOH in which the SOH is characterised by the ohmic resistance. The second filter adaptively updates the drifted model parameters according to the dormancy zone. Finally, the proposed method is evaluated with the random walk charging and discharging test. The experimental results illustrate that the adaptively updated parameters can better match the battery status in real time, and the proposed ADSRCKF can achieve high accuracy with strong robustness. For the initial SOC error, the root mean square error (RMSE) of SOC is less than 0.3%, and the RMSE is approximately 1% even when the initial capacity is also deviated. Highlights: An improved fractional-order model is established in working and idling states. An endogenous immune algorithm is proposed to identify the model parameters. A co-estimation method is developed by adaptively updating the drifted parameters.