Practical SOH Estimation Based on Feasible Feature Weighted Li-ion Battery Model

State-of-health (SOH) is one of the most important state that reflects the battery health status, which plays an important role in guaranteeing normal operation of cells in battery management system (BMS). Complicated aging mechanisms and working conditions make it difficult for accurate and efficient SOH estimation. In order to realize real-time accurate SOH estimation in actual application, this paper proposes a SOH-related health feature (HF) -weighted method for SOH estimation. The proposed SOH estimation method first extracts HFs from some easy-to-access data in actual application. These features comprehensively reflect the battery aging mechanisms. Then, a weighted SOH estimation model is developed from the extracted HFs of the battery. With the HFs extracted from the battery data, SOH can be estimated by the weighted SOH estimation model. Accelerated battery aging experiments are conducted to validate the proposed SOH estimation method. Experimental results show the effectiveness of the proposed method in accurately estimating the battery SOH.