Robust State of Charge estimation for Li-ion batteries based on Extended State Observers

This article proposes a novel State of Charge estimation approach for Li-Ion batteries, using Extended State Observers of Generalized Proportional Integral type. The nonlinear relationship established between the Open Circuit Voltage and the nominal state of Charge is simulated via a set of interpolating polynomials, providing a more accurate description, meaningfully improving the classic piecewise linearization results, particularly at low state of charge conditions (0 to 0.1). The proposed method is experimentally tested in a 20 [Ah] LiNiMnCoO₂ battery at different discharge C-rates and noise presence. Likewise, the estimator is compared against a sliding mode estimation scheme and an Extended Kalman Filter, showing competitive results as found by an integral error based performance index measurement with a reduction of up to 35%.