A set membership theory based parameter and state of charge co-estimation method for all-climate batteries

Abstract

State of charge estimation of the battery is one of the core functions in the battery management system. Accurate and reliable state of charge estimation under wide temperature range is critical for the application of all-climate electric vehicles. The main work of this paper is as follows: (1) To achieve accurate closed-loop state estimation, a temperature dependent battery model is proposed; (2) The common model-based state of charge estimation methods using filters like Kalman filter assume that the state perturbations and measurement noise are white and Gaussian noises, which is not realistic in practical application. To solve this problem, set membership method which holds that the noises are unknown but bounded is used for state of charge estimation. Based on the established temperature dependent battery model and the set membership method, battery parameter and state of charge co-estimation algorithm is proposed for all-climate battery state estimation; (3) The proposed method is fully verified at −10 °C–40 °C and the comparison between the proposed method and extended Kalman filter is conducted to illustrate its superiorities. Furthermore, the validity and real time performance of the co-estimation method are verified in a hardware-in-loop test bench. Results show that the proposed co-estimation method has excellent robustness and the state of charge estimation error is bounded to 5% under wide temperature range.