鋰離子電池是當今最主要的能源儲存技術之一，廣泛應用於消費電子產品、電動汽車和儲能系統。然而，隨著使用時間的延長，鋰離子電池的性能會逐漸下降，主要表現在電池的健康狀態(State of Health, SOH)和剩餘容量(State of Charge, SOC)的變化上。準確估測電池的SOH和SOC對於保障設備運行和提高用戶體驗至關重要。
本文比較了不同最佳化演算法(Optimization Algorithm)對類神經網路(Artificial Neural Network, ANN)模型的優化效果，旨在找到最適合的最佳化演算法。通過模擬發現，粒子群最佳化演算法(Particle Swarm Optimization, PSO)在優化類神經網路模型時得到了最低的誤差。因此，本文進一步使用PSO演算法對不同維度的類神經網路進行優化並進行比較。
第一組類神經網路(用於SOH估測)在使用PSO優化後達到了平均相對誤差(Mean Relative Error, MRE)為0.8929%。第二組類神經網路(用於SOC估測)在使用PSO優化後達到了MRE為1.8600%。
進一步的多維度搜索包括特徵值篩選、隱藏層大小、神經元數量和轉移函數的選擇。結果顯示，經過多維度搜索優化後，兩組類神經網路均取得了更佳的結果。第一組類神經網路(SOH)經過多維度搜索優化後的MRE為0.8690%，比單一維度優化減少了2.6767%的誤差。第二組類神經網路(SOC)經過多維度搜索優化後的MRE為1.7868%，比單一維度優化減少了3.9355%的誤差。
透過本文的模擬結果得知，PSO是一種有效且可靠的最佳化方法，不僅能顯著提升類神經網路在鋰離子電池SOH和SOC估測中的性能，還可以優化多維度類神經網路。這為鋰離子電池管理系統提供了一種高效、準確的解決方案，並為未來相關研究和應用提供了重要參考。

Lithium-ion batteries are one of the most prominent energy storage technologies today, widely used in consumer electronics, electric vehicles, and energy storage systems. However, as usage time increases, the performance of lithium-ion batteries gradually declines, mainly reflected in changes in the State of Health (SOH) and State of Charge (SOC). Accurately estimating the SOH and SOC of batteries is crucial for ensuring device operation and improving user experience.
This study compares the optimization effects of different optimization algorithms on artificial neural network (ANN) models, aiming to find the most suitable optimization algorithm. Simulations show that the Particle Swarm Optimization (PSO) algorithm achieves the lowest error when optimizing ANN models. Therefore, this paper further uses the PSO algorithm to optimize and compare neural networks of different dimensions.
The first group of neural networks (for SOH estimation) achieved a Mean Relative Error (MRE) of 0.8929% after using PSO optimization. The second group of neural networks (for SOC estimation) achieved an MRE of 1.8600% after using PSO optimization.
Further multidimensional searches included feature selection, hidden layer size, number of neurons, and transfer function selection. Results show that after multidimensional search optimization, both neural networks achieved better results. The first group of neural networks (SOH) had an MRE of 0.8690% after multidimensional search optimization, reducing the error by 2.6767% compared to single-dimensional optimization. The second group of neural networks (SOC) had an MRE of 1.7868% after multidimensional search optimization, reducing the error by 3.9355% compared to single-dimensional optimization.
The simulation results in this study indicate that PSO is an effective and reliable optimization method, significantly enhancing the performance of ANN in lithium-ion battery SOH and SOC estimation, and optimizing multidimensional neural networks. This provides an efficient and accurate solution for lithium-ion battery management systems and offers important references for future related research and applications.

[1] 鄭偉呈，「基於類神經網路之鋰離子電池健康狀態估測技術研究」，台灣科技大學電機工程系碩士學位論文，2021年。
[2] 鍾顏澤，「基於類神經網路之鋰離子電池動態交流阻抗估測剩餘容量之研究」，台灣科技大學電機工程系碩士學位論文，2022年。
[3] O. O. Solomon, W. Zheng, J. Chen, and Z. Qiao, “State of charge estimation of Lithium-ion battery using an improved fractional-order extended Kalman filter,” Journal of Energy Storage, vol. 49, p. 104007, May 2022.
[4] X. Zhang, J. Hou, Z. Wang, and Y. Jiang, “Joint SOH-SOC Estimation Model for Lithium-Ion Batteries Based on GWO-BP Neural Network,” Energies, vol. 16, no. 1, p. 132, Dec. 2022.
[5] Z. Bao, J. Jiang, C. Zhu, and M. Gao, “A New Hybrid Neural Network Method for State-of-Health Estimation of Lithium-Ion Battery,” Energies, vol. 15, no. 12, p. 4399, Jun. 2022.
[6] Gold Peak Industries Ltd. “Lithium Ion technical handbook,” http://large.stanford.edu/courses/2015/ph240/uang2/docs/li-handbook.pdf, 2003.
[7] H. C. Choi, Y. M. Jung, I. Noda, and S. B. Kim, “A Study of the Mechanism of the Electrochemical Reaction of Lithium with CoO by Two-Dimensional Soft X-ray Absorption Spectroscopy (2D XAS), 2D Raman, and 2D Heterospectral XAS−Raman Correlation Analysis,” the Journal of Physical Chemistry. B, vol. 107, no. 24, pp. 5806–5811, May 2003.
[8] L. Liu et al., “Layered ternary metal oxides: Performance degradation mechanisms as cathodes, and design strategies for high-performance batteries,” Progress in Materials Science/Progress in Materials Science, vol. 111, p. 100655, Jun. 2020.
[9] C. S. Moo, K. S. Ng, Y. P. Chen, and Y. C. Hsieh, “State-of-Charge Estimation with Open-Circuit-Voltage for Lead-Acid Batteries,” Apr. 2007.
[10] C. Duan, C. Wang, and M. Liu, “A fast approach for predicting battery open circuit voltage based on exponential recovery voltage,” Sep. 2017.
[11] D. Saji, P. S. Babu, and K. Ilango, “SoC Estimation of Lithium Ion Battery Using Combined Coulomb Counting and Fuzzy Logic Method,” May 2019.
[12] Texas Instruments, “TI電池電量監測基礎知識培訓-TI培訓電子書籍系列”， https://www.dianyuan.com/index.php?do=ti_lesson_list&tid=15, 2013.
[13] Texas Instruments, “Theory and Implementation of Impedance Track™ Battery Fuel-Gauging Algorithm in bq20zxx Product Family,” https://www.ti.com/lit/an/slua364b/slua364b.pdf, 2006.
[14] Texas Instruments, “Impedance Track™ Gas Gauge for Novices,” https://www.ti.com/lit/an/slua375/slua375.pdf, 2006.
[15] R. Xiong, L. Li, and J. Tian, “Towards a smarter battery management system: A critical review on battery state of health monitoring methods,” Journal of Power Sources, vol. 405, pp. 18–29, Nov. 2018.
[16] M. Berecibar, I. Gandiaga, I. Villarreal, N. Omar, J. Van Mierlo, and P. Van Den Bossche, “Critical review of state of health estimation methods of Li-ion batteries for real applications,” Renewable & Sustainable Energy Reviews, vol. 56, pp. 572–587, Apr. 2016, doi: 10.1016/j.rser.2015.11.042.
[17] X. Han, M. Ouyang, L. Lu, J. Li, Y. Zheng, and Z. Li, “A comparative study of commercial lithium ion battery cycle life in electrical vehicle: Aging mechanism identification,” Journal of Power Sources, vol. 251, pp. 38–54, Apr. 2014.
[18] M. Dubarry and B. Y. Liaw, “Identify capacity fading mechanism in a commercial LiFePO4 cell,” Journal of Power Sources, vol. 194, no. 1, pp. 541–549, Oct. 2009.
[19] M. Dubarry et al., “Identifying battery aging mechanisms in large format Li ion cells,” Journal of Power Sources, vol. 196, no. 7, pp. 3420–3425, Apr. 2011.
[20] B. Sood, M. Osterman, and M. Pecht, “Health monitoring of lithium-ion batteries,” Oct. 2013.
[21] J. Remmlinger, M. Buchholz, M. Meiler, P. Bernreuter, and K. Dietmayer, “State-of-health monitoring of lithium-ion batteries in electric vehicles by on-board internal resistance estimation,” Journal of Power Sources, vol. 196, no. 12, pp. 5357–5363, Jun. 2011.
[22] R. Xiong, F. Sun, X. Gong, and C. Gao, “A data-driven based adaptive state of charge estimator of lithium-ion polymer battery used in electric vehicles,” Applied Energy, vol. 113, pp. 1421–1433, Jan. 2014.
[23] A. Widodo, M.-C. Shim, W. Caesarendra, and B.-S. Yang, “Intelligent prognostics for battery health monitoring based on sample entropy,” Expert Systems With Applications, vol. 38, no. 9, pp. 11763–11769, Sep. 2011.
[24] D. Andre, C. Appel, T. Soczka-Guth, and D. U. Sauer, “Advanced mathematical methods of SOC and SOH estimation for lithium-ion batteries,” Journal of Power Sources.
[25] V. Klass, M. Behm, and G. Lindbergh, “A support vector machine-based state-of-health estimation method for lithium-ion batteries under electric vehicle operation,” Journal of Power Sources, vol. 270, pp. 262–272, Dec. 2014.
[26] R. Xiong, Y. Zhang, J. Wang, H. He, S. Peng, and M. Pecht, “Lithium-Ion Battery Health Prognosis Based on a Real Battery Management System Used in Electric Vehicles,” IEEE Transactions on Vehicular Technology, vol. 68, no. 5, pp. 4110–4121, May 2019.
[27] Y. Zhang, R. Xiong, H. He, and M. G. Pecht, “Lithium-Ion Battery Remaining Useful Life Prediction With Box–Cox Transformation and Monte Carlo Simulation,” IEEE Transactions on Industrial Electronics, vol. 66, no. 2, pp. 1585–1597, Feb. 2019.
[28] J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Networks, vol. 61, pp. 85–117, Jan. 2015.
[29] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, May 2015.
[30] A. A. Hussein, “Adaptive Artificial Neural Network-Based Models for Instantaneous Power Estimation Enhancement in Electric Vehicles’ Li-Ion Batteries,” IEEE Transactions on Industry Applications, vol. 55, no. 1, pp. 840–849, Jan. 2019.
[31] A. A. Hussein, “Capacity Fade Estimation in Electric Vehicle Li-Ion Batteries Using Artificial Neural Networks,” IEEE Transactions on Industry Applications, vol. 51, no. 3, pp. 2321–2330, May 2015.
[32] J. A. K. Suykens, B. L. R. De Moor, and J. Vandewalle, “Static and dynamic stabilizing neural controllers, applicable to transition between equilibrium points,” Neural Networks, vol. 7, no. 5, pp. 819–831, Jan. 1994.
[33] H. J. Escalante, M. Montes, and L. E. Sucar, “Multi-class particle swarm model selection for automatic image annotation,” Expert Systems With Applications, vol. 39, no. 12, pp. 11011–11021, Sep. 2012.
[34] M. Aghaabbasi, M. Ali, M. Jasiński, Z. Leonowicz, and T. Novák, “On Hyperparameter Optimization of Machine Learning Methods Using a Bayesian Optimization Algorithm to Predict Work Travel Mode Choice,” IEEE Access, vol. 11, pp. 19762–19774, Jan. 2023.
[35] J. H. Holland, Adaptation in Natural and Artificial Systems. MIT Press, 1992.
[36] 呂肯岳，「基於交流阻抗量測之線上鋰離子電池電荷狀態與健康狀態估測器研製，國立台灣科技大學電機工程系碩士學位論文，民國112年01月。