Improving battery state estimation accuracy through the addition of a series capacitor. (December 2020)

• Record Type: Journal Article
• Authors: Xu, Chu; Cleary, Timothy; Fathy, Hosam
• Is Part Of: Journal of energy storage. Volume 32(2020)
• Publication Date: 2020-12
• Subjects: Estimation accuracy -- Fisher information -- Ultracapacitor–battery; Energy storage -- Periodicals; Energy storage -- Research -- Periodicals; 621.3126
• DOI: 10.1016/j.est.2020.101948
• Languages: English
• ISSNs: 2352-152X

Abstract

Abstract: This article is motivated by the theoretical bounds on battery state of charge estimation accuracy, particularly in the presence of biased and/or noisy current/voltage measurements. The main goal is to show that the addition of a series ultracapacitor improves these theoretical bounds, enabling more accurate state estimation. The authors derive this conclusion analytically by employing Fisher information analysis to show that the Cramér–Rao bounds on estimation accuracy are significantly smaller for a hybrid ultracapacitor–battery system compared to a battery-only system. The addition of series capacitance furnishes this improvement in estimation accuracy by increasing sensitivity of the ultracapacitor–battery open-circuit potential to changes in stored charge. This makes the proposed pack hybridization concept particularly attractive for battery chemistries where the slope of battery voltage versus charge can be very small, e.g., LiFePO4 cells. Monte Carlo simulation studies support this theoretical insight, for both linearized and nonlinear LiFePO4 battery models. The simulation results also provide the important added observation that in situations where model mismatch introduces estimation bias, the use of ultracapacitor–battery hybridization reduces this bias. Moreover, experimental validation is performed in this article. Both the simulation and test results confirm the improvements in estimation variance predicted by Fisher analysis for unbiased estimators. Highlights: Improved battery state estimation accuracy through hybridization. Studied the estimation accuracy improvements using Fisher information analysis. Validated the estimation variance improvements through simulation and experiments.