在全球暖化的影響下，環保成為了未來發展的重要一環，不管是國家、企業或是個人，對於環保的要求與日俱增，為了減少環境的污染，將電力作為直接能源使用的物品越來越多，隨之一起發展的還有各種電力儲存的裝置，像是各式鋰離子電池與超級電容器等。因此，預測其壽命的需求也開始增加。本研究旨在探討鋰離子電池的健康度預測方法，目標為達到精準的電池健康管理。鋰離子電池的測試既昂貴又耗時，本文選用了公開的NASA數據庫中的實驗數據，先透過相關係數對數據進行篩選，再使用支持向量回歸進行機器學習模型的建立和訓練。此模型能夠根據電池的運作數據(如充放電曲線、溫度、電壓等)預測其健康狀況，以提前發現電池性能下降的跡象，並採取適當的維護措施。根據實驗結果，使用Spearman關聯係數所找出的四個特徵進行支持向量迴歸之訓練結果為最佳，平均相對誤差僅0.0142，相較於使用Pearson關聯係數、灰色關聯分析(Grey relational analysis, GRA)以及使用全部特徵值所得之平均相對誤差分別可改善11.15 %，172.88 %以及14.23 %。

With the impact of global warming, environmental protection has become an essential aspect of future development. The demand for environmental preservation is increasing among nations, businesses, and individuals. In order to reduce environmental pollution, the utilization of electrical appliances as direct energy sources has been growing. Consequently, various power storage devices, such as lithium-ion batteries and electrostatic double-layer capacitors, have also been developed. Therefore, the need for predicting their lifespan has increased. This study aims to explore a state of health (SOH) prediction method for lithium-ion batteries to achieve accurate battery management. Lithium-ion battery testing is both expensive and time-consuming. In this thesis, publicly available experimental data from the NASA database is selected. The data is initially filtered using correlation coefficients, and then a support vector regression is employed to establish and train a machine learning model. This model can predict the battery's SOH based on operating data such as charging-discharging curves and temperature. This enables the early detection of battery performance degradation and the implementation of appropriate maintenance measures. Experimental results indicate that using the four features identified by Spearman correlation coefficient for support vector regression yields the best training results, with an average relative error of only 0.0142. This represents an improvement of 11.15 % compared to using Pearson correlation coefficient, 172.88 % compared to grey relational analysis, and 14.23 % compared to using all features.

[1] 高柏科技股份有限公司(2022/9/5)。從溫度效應對電池包效率之影響了解提高電池效能20%之可行性。取自：https://www.materialsnet.com.tw/DocView.aspx?id=51334
[2] Joseph Lee Rodgers; W. Alan Nicewander,The American Statistician, Vol. 42, No. 1. (Feb., 1988), pp. 59-66. February 1988.
[3] Stigler, Stephen M. Francis Galton's Account of the Invention of Correlation. Statistical Science. 1989, 4 (2): 73-79. JSTOR 2245329. doi:10.1214/ss/117
7012580.
[4] Ji Wu, Junxiong Chen, Xiong Feng, Haitao Xiang, Qiao Zhu, State of health estimation of lithium-ion batteries using Autoencoders and Ensemble Learning, Journal of Energy, Volume 55, Part C, 2022, 105708, ISSN 2352-152X.
[5] Yajun Zhang, Yajie Liu, Jia Wang, Tao Zhang, State-of-health estimation for lithium-ion batteries by combining model-based incremental capacity analysis with support vector regression, Energy, Volume 239, Part B, 2022, 121986, ISSN 0360-5442.
[6] Qianglong Li, Dezhi Li, Kun Zhao, Licheng Wang, Kai Wang, State of health estimation of lithium-ion battery based on improved ant lion optimization and support vector regression,Journal of Energy Storage, Volume 50, 2022, 104215, ISSN 2352-152X.
[7] W. Xiong, Y. Mo and C. Yan, "Online State-of-Health Estimation for Second-Use Lithium-Ion Batteries Based on Weighted Least Squares Support Vector Machine," in IEEE Access, vol. 9, pp. 1870-1881, 2021, doi: 10.1109/ACCESS.2020.3026552
[8] Spearman C. The proof and measurement of association between two things. By C. Spearman, 1904. Am J Psychol. 1987 Fall-Winter;100(3-4):441-71. PMID: 3322052.
[9] Yuanwang Deng, Hejie Ying, Jiaqiang E, Hao Zhu, Kexiang Wei, Jingwei Chen, Feng Zhang, Gaoliang Liao, Feature parameter extraction and intelligent estimation of the State-of-Health of lithium-ion batteries, Energy, Volume 176, 2019, Pages 91-102, ISSN 0360-5442
[10] Wenjie Pan, Qi Chen, Maotao Zhu, Jie Tang, Jianling Wang, A data-driven fuzzy information granulation approach for battery state of health forecasting, Journal of Power Sources, Volume 475, 2020, 228716, ISSN 0378-7753
[11] J. Wu, L. Fang, J. Meng, M. Lin and G. Dong, "Optimized Multi-Source Fusion Based State of Health Estimation for Lithium-Ion Battery in Fast Charge Applications," in IEEE Transactions on Energy Conversion, vol. 37, no. 2, pp. 1489-1498, June 2022, doi: 10.1109/TEC.2021.3137423.
[12] Lei Cai, Jinhao Meng, Daniel-Ioan Stroe, Guangzhao Luo, Remus Teodorescu, An evolutionary framework for lithium-ion battery state of health estimation, Journal of Power Sources, Volume 412, 2019, Pages 615-622, ISSN 0378-7753,
[13] M. Xiang et al., "State-of-Health Prognosis for Lithium-Ion Batteries Considering the Limitations in Measurements via Maximal Information Entropy and Collective Sparse Variational Gaussian Process," in IEEE Access, vol. 8, pp. 188199-188217, 2020, doi: 10.1109/ACCESS.2020.3029276.
[14] Geetanjali Singh, Namita Mittal & Satyendra Singh Chouhan (2022), A Systematic Review of Deep Learning Approaches for Natural Language Processing in Battery Materials Domain, IETE Technical Review, 39:5, 1046-1057, DOI: 10.1080/02564602.2021.1984323
[15] B. Saha and K. Goebel. "Battery Data Set", NASA Prognostics Data Repository(2009-2010), NASA Ames Research Center, Moffett Field, CA. from: https://www.nasa.gov/content/prognostics-center-of-excellence-data-set-repository