State of Charge Estimation of Lithium-ion Battery Considering Aging Degree and External Factors Based on Gradient Boosting Regression Tree

Abstract

To solve the inaccuracy of online State of Charge (SOC) estimation of lithium-ion battery caused by the problem that the changes of aging degree of battery and external factors cannot be reflected to SOC estimation in real time, a new method based on data driven and ensemble learning is proposed in this paper. The standard capacity of lithium-ion battery is estimated during the discharge stage. The standard capacity and ambient temperature are used to determine actual capacity of lithium-ion battery in real time. Then, the standard capacity and some external factors are used together to estimate SOC. Using the corresponding data sets obtained from the battery test stage to train the Gradient Boosting Regression Tree (GBRT) model. The effectiveness of the proposed method is verified by several sets of experiments.