Accurate estimation of the State of Charge (SOC) and internal temperature of a battery is pivotal for enhancing its performance and safety. The precision of the battery model and the efficacy of the estimation algorithm play critical roles in this context. This paper introduces a multiparameter thermoelectric coupling model for cylindrical lithium-ion batteries, considering the interplay between SOC and temperature fluctuations. It employs an enhanced entropy heat coefficient experiment to determine reversible and irreversible heat generated during battery operation. For parameter identification, the Variable Forgetting Factor Recursive Least Squares algorithm is utilized. The accuracy of the proposed multiparameter electrothermal coupling model is corroborated by comparing the SOC and internal temperature estimation results with those of standalone electrical and thermal models. The findings indicate that our model achieves an improvement in estimation accuracy exceeding 70％ over the conventional electric heating model. Furthermore, we developed a Singular Value Decomposition-based Adaptive Unscented Kalman Filter algorithm for real-time joint estimation of SOC and internal temperature, which was experimentally validated under dynamic stress test conditions. Comparative analysis with Extended Kalman Filter and Unscented Kalman Filter algorithms demonstrates the superior accuracy of our method in SOC and temperature estimations, with average errors of 5％ and 0.2°C, respectively.
关键词
可逆热/SOC和温度联合估计/多参数电热耦合模型/SVD-AUKF算法
Key words
reversible heat/joint estimation of SOC and temperature/multi parameter electrothermal coupling model/SVD-AUKF algorithm
维普
万方数据