由於電池模型由快、慢動態阻抗組成，本文提出了解耦遞迴最小平方法(Decoupled Recursive Least Square, DRLS)，通過分別處理電池快、慢動態阻抗參數的成本函數，提高實時電池荷電狀態(State of Charge, SOC)估測準度。另一方面，時域參數提取(Time Domain Parameters Estimation, TDPE)方法可以從實驗電壓數據中獲取所有必要的模型參數，而無需迭代優化參數，既簡單又省時。為了同時兼具TDPE識別速度快和DRLS準確率高的優點，本文提出了一種結合上述兩種方法的參數識別技術。將TDPE方法與DRLS技術相結合，可以有效簡化DRLS估計方法的複雜度，提高其估計精度。此外，還提出了考慮快速和慢速動態參數的直流電阻(DCR)補償項，以進一步提高DCR的識別精度，從而提高SOC估計準確度。
根據模擬結果顯示，在聯邦城市駕駛計劃(Federal Urban Driving Schedule, FUDS)和動態壓力測試(Dynamic Stress Test, DST)下，本文所提方法與DRLS技術相比，建模誤差之平均絕對百分比誤差(Mean Absolute Percentage Error, MAPE)分別改善78.7%/73.5%，SOC之MAPE分別改善72.3%/7.1%。實驗結果顯示，建模誤差之MAPE分別改善9.0%/5.3%，SOC之MAPE分別改善18.8%/8.3%。這些結果驗證所提方法的有效性和正確性。

Since a battery model is comprised of fast and slow dynamic response impedance, the conventional least-squares (LS) method is believed to be subject to numerical problems and low precision. Therefore, decoupled recursive least squares (DRLS) technique is proposed to conquer these problems by separately estimating the parameters of the battery's fast and slow dynamic response impedance, and improving the real-time SOC estimation accuracy. On the other hand, time-domain parameter extraction (TDPE) methods can obtain all necessary model parameters from experimental voltage data without optimizing the parameters iteratively; hence it is simple and time-effective.
To have both the advantages of fast identification speed of TDPE and high accuracy of DRLS at the same time, this study proposes a parameter identification technique that combines the two methods mentioned above. Integrating the TDPE approach and DRLS technique can effectively simplify the complexity of the DRLS estimation method and improve its estimation accuracy. In addition, a direct current resistance (DCR) compensation term considering fast and slow dynamics parameters is also proposed to further improve the identification accuracy of DCR, which further enhances the SOC estimation precision. Comparing with the original DRLS technique, the mean absolute percentage error (MAPE) of the modelling error obtained by the proposed method can be improved by 78.7 %/73.5 % under federal urban driving schedule (FUDS) and the dynamic stress test (DST) test patterns for simulated results, and enhanced by 9.0 %/5.3 % under FUDS and DST test profiles for experimental results. In addition, the MAPE of the SOC error is reduced by 72.3 %/7.1 % under FUDS and DST test patterns for simulated results, and lowered by 18.8 %/8.3 % under FUDS and DST test profiles for experimental results. Those results validate the effectiveness and correctness of the proposed method.

參考文獻
[1] M.U Hassan, S. Saha, M.E. Haque, et al., A comprehensive review of battery state of charge estimation techniques J. Sustainable Energy Technologies and Assessments 54 (2022) 102801.
[2] I. Babaeiyazdi, A.R. Zare, S. Shokrzadeh, State of charge prediction of EV Li-ion batteries using EIS: A machine learning approach, J. Energy 223 (2021) 120116.
[3] A. Manoharan, K.M. Begam, V.R. Aparow, et al., Artificial neural networks, gradient boosting and support vector machines for electric vehicle battery state estimation: A review, J. Energy Storage 55 (2022) 105384.
[4] Y. Jianga, J. Xua, M. Liua, et al., An electromechanical coupling model-based state of charge estimation method for lithium-ion pouch battery modules, J. Energy 259 (2022) 125019.
[5] I. Sgura, L. Mainetti, F. Negro, et al., Deep-learning based parameter identification enables rationalization of battery material evolution in complex electrochemical systems, J. Computational Science 66 (2023) 101900.
[6] A.H. Ranjbar, A. Banaei, A. Khoobroo, et al., Online estimation of state of charge in Li-Ion batteries using impulse response concept, IEEE Transactions on smart grid, 2012, vol. 3, no. 1
[7] A. Lievre, A. Sari, P. Venet, et al., Practical online estimation of lithium-ion battery apparent series resistance for mild hybrid vehicles, IEEE transactions on vehicular technology, 2016, vol. 65, no. 6.
[8] T. Kalogiannis, M.S. Hosen, M.A. Sokkeh, et al., Comparative Study on Parameter Identification Methods for Dual-Polarization Lithium-Ion Equivalent Circuit Model, Energies 2019, 12, 4031.
[9] C. Unterrieder, C. Zhang, M. Lunglmayr, et al., Battery state-of-charge estimation using approximate least squares, J. Power Sources 278 (2015) 274-286.
[10] Y. Che, J. Jia, Y. Yang, et al., Parameter Identification of 3R-C Equivalent Circuit Model Based on Full Life Cycle Database, J Electr Eng Technol.2018; 13(4): 1759-1768
[11] M. Verbrugge, E. Tate, Adaptive state of charge algorithm for nickel metal hydride batteries including hysteresis phenomena, J. Power Sources 126 (2004) 236–249.
[12] Z. Wei, T. Lim, M.S. Kazacos, et al., Online state of charge and model parameter co-estimation based on a novel multi-timescale estimator for vanadium redox flow battery, Applied Energy 172 (2016) 169–179.
[13] M. Elmarghichi1, M. Bouzi, N. Ettalabi, Robust parameter estimation of an electric vehicle lithium-ion battery using adaptive forgetting factor recursive least squares, International Journal of Intelligent Engineering and Systems, 2020, vol.13, no.5.
[14] Q. Song, Y. Mi, W. Lai, A novel variable forgetting factor recursive least square algorithm to improve the anti-interference ability of battery model parameters identification, IEEE Access, 2019, vol. 7.
[15] X. Liu, Y. Jin, S. Zeng, et al., Online Identification of Power Battery Parameters for Electric Vehicles Using a Decoupling Multiple Forgetting Factors Recursive Least Squares Method, CSEE Journal of Power and Energy Systems, 2020, vol. 6, no. 3.
[16] S. Liu, J. Wang, Q. Liu, et al., Deep-Discharging Li-Ion Battery State of Charge Estimation Using a Partial Adaptive Forgetting Factors Least Square Method, IEEE Access, 2019, vol. 7.
[17] L. Zhang, H. Peng, Z. Ning, et al., Comparative Research on RC Equivalent Circuit Models for Lithium-Ion Batteries of Electric Vehicles, Appl. Sci. 2017, 7, 1002.
[18] H. He, X. Zhang, R. Xiong, et al., Online model-based estimation of state-of-charge and open-circuit voltage of lithium-ion batteries in electric vehicles, J. Energy 39 (2012) 310-318.
[19] M. Nawaz, J. Ahmed, G. Abbas, Energy-efficient battery management system for healthcare devices, J. Energy Storage 51 (2022) 104358.
[20] V.H. Duong, H.A. Bastawrous, K. Lim, et al., Online state of charge and model parameters estimation of the LiFePO4 battery in electric vehicles using multiple adaptive forgetting factors recursive least-squares, J. Power Sources 296 (2015) 215-224.
[21] T. Kalogiannis, M.S. Hosen, M.A. Sokkeh, et al., Comparative study on parameter identification methods for dual-polarization Lithium-ion equivalent circuit model, Energies 2019, 12, 4031.
[22] C. Zhang, W. Allafi, Q. Dinh, et al., Online estimation of battery equivalent circuit model parameters and state of charge using decoupled least squares technique, J. Energy 142 (2018) 678-688.
[23] F. Wen, B. Duan, C. Zhang, et al., High-accuracy parameter identification method for equivalent-circuit models of Lithium-ion batteries based on the stochastic theory response reconstruction, Electronics 2019, 8, 834.
[24] T. Ouyang, P. Xu, J. Chen, et al., Improved parameters identification and state of charge estimation for lithium-ion battery with real-time optimal forgetting factor, J. Electrochimica Acta 353 (2020) 136576.
[25] Y. Ye, Z. Li, J. Lin, et al., State-of-charge estimation with adaptive extended Kalman filter and extended stochastic gradient algorithm for Lithium-ion batteries, J. Energy Storage 47 (2022) 103611.
[26] F. Guo, G. Hu, S. Xiang, et al., A multi-scale parameter adaptive method for state of charge and parameter estimation of lithium-ion batteries using dual Kalman filters, J. Energy 178 (2019) 79-88.
[27] S. Wang, C. Fernandez, C. Yu, et al., A novel charged state prediction method of the Lithium-ion battery packs based on the composite equivalent modeling and improved splice Kalman filtering algorithm, J. Power Sources 471 (2020) 228450.
[28] N. Wassiliadisa, J. Adermanna, A. Frericksb, et al., Revisiting the dual extended Kalman filter for battery state-of-charge and state-of-health estimation: A use-case life cycle analysis, J. Energy Storage 19 (2018) 73–87.
[29] Z. Chen, L. Yang, X. Zhao, et al., Online state of charge estimation of Li-ion battery based on an improved unscented Kalman filter approach, J. Applied Mathematical Modelling 70 (2019) 532–544.
[30] L. Zhao, Z. Liu, G. Ji, Lithium-ion battery state of charge estimation with model parameters adaptation using H∞ extended Kalman filter, J. Control Engineering Practice 81 (2018) 114–128.
[31] A. Hentunen, T. Lehmuspelto, J. Suomela, Time-domain parameter extraction method for the ́venin-equivalent circuit battery models, IEEE Trans. Energy Conversion 29 (3) (2014) 558–566.
[32] Y. Zhang, C. Wang, Cycle-life characterization of automotive lithium-ion batteries with LiNiO2 cathode, J. Electrochem. Soc. 156 (2009) A527.
[33] G. Nagasubramanian, E. Roth, D. Ingersoll, Electrical and electrochemical performance characteristics of small commercial Li-ion cells, in: Battery Conference on Applications and Advances, 1999, pp. 91–96.
[34] F. Berthier, J. Diard, R. Michel, Distinguishability of equivalent circuits containing CPEs: part I. Theoretical part, J. Electroanal. Chem. 510 (1–2) (2001) 1–11.
[35] E. Locorotondc, L. Pugi, L. Berziet, et al., Modeling and simulation of constant phase element for battery electrochemical impedance spectroscopy, in: 2019 IEEE 5th International forum on Research and Technology for Society and Industry (RTSI), 2019, pp. 225–230.
[36] Y. Cheng, M. Chuang, Y. Liu, et al., "A particle swarm optimization based power dispatch algorithm with roulette wheel re-distribution mechanism for equality constraint," Renewable Energy, vol. 88, pp. 58-72, 2016.