Joint estimation of the state-of-energy and state-of-charge of lithium-ion batteries under a wide temperature range based on the fusion modeling and online parameter prediction. (25th August 2022)

• Record Type: Journal Article
• Authors: Xia, Lili; Wang, Shunli; Yu, Chunmei; Fan, Yongcun; Li, Bowen; Xie, YanXin
• Is Part Of: Journal of energy storage. Volume 52:Part C(2022)
• Publication Date: 2022-08-25
• Subjects: lithium-ion batteries -- Equivalent-circuit model -- State-of-energy -- State-of-charge -- Parameter identification; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2022.105010
• Languages: English
• ISSNs: 2352-152X

Abstract

Abstract: Accurate remaining mileage prediction is still a challenge for electric vehicles. State-of-energy and state-of-charge are the state parameters used to represent the remaining endurance and charge of lithium-ion batteries respectively, which are related to the remaining mileage forecast of electric vehicles. In the application of lithium-ion batteries, the ambient temperature cannot be constant. The temperature has a great influence on the state-of-energy and state-of-charge estimation. To obtain a high precision mathematical description and state parameters of lithium-ion batteries, the novel fusion equivalent-circuit model of lithium-ion batteries considering the influence of temperature is proposed. For the estimation of the state-of-energy and state-of-charge, this paper adopts an adaptive noise correction-dual extended Kalman filtering algorithm to realize the state estimation, this algorithm can solve the noise influence of Kalman filtering. The experimental results show that the estimation error of the method proposed in this paper of state-of-energy and state-of-charge are within 1.83 % and 1.92 % at different working temperatures and conditions. The estimation results prove the efficiency of the co-estimation method of state-of-energy and state-of-charge. Highlights: The improved Thevenin model is used with high modeling accuracy. The FFRELS algorithm is used to realize accurate model parameter identification. An adaptive noise correction method is proposed to solve the noise influence of the EKF algorithm. An adaptive noise correction-dual extended Kalman filtering algorithm to realize the SOC and SOE co-estimation.