Aiming at the requirements of wide temperature, wide power, and wide discharge depth scenarios for electric vehicles (EV), while focusing on multi-coupled states joint estimation in EV power batteries and the accurate expression of the electro-thermal complex coupling relationship (E/T-CCR), a multi-coupled states joint estimation algorithm for lithium batteries based on multi-domain hybrid model (MDHM) is proposed. Accurate modeling for E/T-CCR is proposed. That is, the characteristics of the battery in the electrical domain and thermal domain are accurately expressed through a bidirectional mapping model established in the coupling domain to form the MDHM. Furthermore, based on the coupled state space equations derived from the MDHM, considering the advantages of filtering methods in terms of stability, robustness, generalization ability, complexity, and data dependence, a dual-filter structure based on adaptive square root unscented Kalman filter (ASRUKF) is proposed to achieve multi-state joint estimation and coupling domain stability correction. Meanwhile, the state of power (SOP) estimation considering multi-constraints including state of temperature (SOT) is realized. The proposed algorithm is compared with 2 typical multi-state joint estimation algorithms based on electro-thermal coupling model (ETCM) in 12 temperature scenarios under 3 dynamic operating conditions including aging for 2 types of batteries. The results show that the proposed algorithm has better accuracy and dynamic tracking performance. Compared with 3 deep network models (DNM) algorithms. The results show that the proposed algorithm has a good balance between accuracy and complexity. It can achieve joint estimation of the 4 key battery states.
扫码关注我们
求助内容:
应助结果提醒方式:
