Estimation Error Bound of Battery Electrode Parameters with Limited Data Window

Abstract

Advanced battery management system, which leverages an in-depth understanding of the battery state of health, can improve efficiently and safely. To this end, we introduce the electrode-level battery state of health (eSOH) estimation problem with open-circuit voltage (OCV) data. In real-world applications, collecting the full-range OCV data is difficult since the battery is not deeply discharged. When data is limited, the estimation accuracy deteriorates. In this article, we quantify the uncertainty of the electrode parameter estimation with partial data based on the Cramer-Rao bound and confidence interval. By introducing a voltage constraint in the estimation problem, the positive electrode parameters can be estimated with sufficient accuracy over a wide range of state of charge. However, the estimation accuracy of the negative electrode parameters is more sensitive to the depth of discharge. The proposed framework can be used as a guideline for selecting proper data windows and understanding the impact on parameter estimation.