A robust observer based on the nonlinear descriptor systems application to estimate the state of charge of lithium-ion batteries

• Published in: Journal of the Franklin Institute 2023-11, Vol.360 (16), p.11397-11413
• Main Authors: Meng, Shengya, Meng, Fanwei, Chi, Heng, Chen, Haonan, Pang, Aiping
• Format: Article
• Language: English
• ISSN: 0016-0032; 1879-2693
• DOI: 10.1016/j.jfranklin.2023.08.037

•There is an uncomplicated nonlinear descriptor system for the lithium-ion battery, which does not involve additional derivation operations.•The nonlinear characteristics of the lithium-ion batteries are retained. Furthermore, the estimation accuracy of SOC is improved.•The dynamic characteristics of input current are preserved, which can significantly reduce the chattering caused by sudden current changes.•Fewer parameter adjustments bring a more convenient observer design process.•The proposed observer has excellent robustness and can deal with various uncertainties. A robust observer based on nonlinear descriptor systems is applied to estimate lithium-ion batteries’ state of charge (SOC). We creatively take the relationship between the battery’s open circuit voltage (OCV) and SOC as a nonlinear algebraic constraint, thereby modeling the battery as a nonlinear descriptor system. This model can effectively trade off accuracy and complexity. The robust convergence, even in the presence of uncertainties, is proven through a combination of the Lyapunov stability theorem and linear matrix inequality (LMI) techniques. Experimental results showcase the effectiveness of the proposed SOC estimation technique, offering enhanced convergence speed, estimation accuracy, and robustness compared to conventional sliding mode observers.