An improved state of charge estimation method based on cubature Kalman filter for lithium-ion batteries. (1st November 2019)

• Record Type: Journal Article
• Authors: Peng, Jiankun; Luo, Jiayi; He, Hongwen; Lu, Bing
• Is Part Of: Applied energy. Volume 253(2019)
• Publication Date: 2019-11-01
• Subjects: Cubature Kalman Filter (CKF) -- Equivalent circuit model -- Fractional order model -- State of charge -- Fuzzy controller; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2019.113520
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: Cubature Kalman filter based on fuzzy controller for SOC estimation is proposed. Compared study for thevenin model and fractional order model is carried out. Compared study for Cubature Kalman filter and Extend Kalman filter is carried out. Proposed improved method is robustness and minimize 2% estimation error specially. Abstract: In this paper, an improved state of charge (SOC) estimation method of Lithium-Ion battery is developed based on a cubature Kalman filter (CKF) supported by experimental data. Firstly, a first-order RC model and corresponding fractional order model are established to evaluate the estimation accuracy of different models. Secondly, model parameters are identified through a custom Hybrid Pulse Power Characteristic (HPPC) experiment based on the Sequential Quadratic Programming (SQR) method. Then, a CKF algorithm is used to estimate the battery SOC under different battery models with no prior knowledge of initial SOC. The results show that the proposed CKF method has a better estimate robustness rather than Extended Kalman filter (EKF) and the fractional order model can achieve higher accuracy while it consumes more computing resources compared with equivalent circuit models. SOC estimation error of CKF algorithms is less than 3%. Thirdly, a battery management unit in the loop approach is applied to verify the accuracy of estimation. Last but not least, in order to reduce the estimation error due to battery degradation and battery model errors, a fuzzy controller is constructed to modified the gain coefficient of Kalman. The proposed improved method can minimize the estimation error of SOC by 2%.