進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-0602201823413400
論文名稱(中文) 應用類神經網路與支援向量迴歸於磷酸鋰鐵電池剩餘電量之估測
論文名稱(英文) Residual State-of-Charge Estimation for LiFePO4 Battery Using Artificial Neural Network and Support Vector Regression
校院名稱 成功大學
系所名稱(中) 系統及船舶機電工程學系
系所名稱(英) Department of Systems and Naval Mechatronic Engineering
學年度 106
學期 1
出版年 107
研究生(中文) 顧哲綸
研究生(英文) Zhe-Lun Gu
學號 P16054214
學位類別 碩士
語文別 中文
論文頁數 67頁
口試委員 口試委員-王醴
口試委員-白富升
口試委員-陳良瑞
指導教授-李建興
中文關鍵字 磷酸鋰鐵電池  剩餘電量  類神經網路  支援向量迴歸  波伊卡特效應 
英文關鍵字 LiFePO4 Battery  State of Charge  Artificial Neural Network  Support Vector Regression  Peukert’s Effect 
學科別分類
中文摘要 本論文使用類神經網路及支援向量迴歸兩種方法進行磷酸鋰鐵電池剩餘電量估測,因電池於放電中存在波伊卡特效應,在估測電池放電電量與實際剩餘電量間會產生誤差,故需先對電池進行波伊卡特效應的校正。於校正前後,放電電量與實際剩餘電量的平均估測誤差從原本的2.12%降為0.49%,由此證實透過波伊卡特效應的校正能得到更準確剩餘電量的估測。另外,亦針對不同放電時間與放電率探討類神經網路及支援向量迴歸於放電電量與剩餘電量間的均方根誤差;在0.5C與1.0C脈衝放電與定電流充放電的條件下,使用支援向量迴歸法所得剩餘電量與估測電量的均方根誤差都比類神經網路估測所得的均方根誤差要來的小。最後,本論文比較上述兩方法於電量估測的準確度、估測效率與穩健性,由結論得知,支援向量迴歸法比類神經網路法有較高的可靠性。
英文摘要 In this research, we use two methods: Artificial Neural Network (ANN) and Support Vector Regression (SVR) to estimate the residual state-of-charge (SOC) of LiFePO4 Battery. Owing to existing Peukert’s effect when battery discharging, it may cause the error between the estimated discharging value and actual residual value. Therefore, the Peukert’s effect of the LiFePO4 battery should be calibrated before estimating the SOC. Before and after calibration, the average estimated error of the estimated discharging value and actual residual value is from original 2.12% to 0.49%, confirming that through the calibration, the more accurate experimental value can be obtained. In addition, we discuss the root mean square error (RMSE) of ANN and SVR according to the different discharging rate and discharging time. Under 0.5C and 1.0C and constant current constant voltage discharging conditions, the RMSE between the residual SOC and estimated SOC of the SVR was smaller than the RMSE of the ANN method. Finally, this research compares accuracy, efficiency and robustness of the above two methods for SOC estimation. From the conclusion, it is proved the SVR has higher reliability than the ANN.
論文目次 摘要 i
Abstract ii
誌謝 vi
目錄 vii
表目錄 x
圖目錄 xi
符號說明 xiv
第一章 緒論 1
1.1 前言 1
1.2 研究動機與目的 1
1.3 文獻回顧 3
1.4 本論文之貢獻 5
1.5 本論文之架構 5
第二章 鋰離子電池等效電路模型 7
2.1 前言 7
2.2 電池等效電路模型 8
2.2.1 等效電路模型分類 9
2.2.2 等效電路模型建立與驗證 10
第三章 研究方法 21
3.1 前言 21
3.2 類神經網路法 21
3.3 支援向量迴歸 29
3.3.1 支援向量迴歸原理 29
3.3.2 支援向量迴歸參數介紹 30
第四章 實驗與模擬結果 33
4.1 前言 33
4.2 波伊卡特效應 33
4.3 磷酸鋰鐵電池放電實驗 37
4.3.1 實驗架構 39
4.3.2 實驗結果 39
4.4 模擬結果 42
4.4.1 類神經網路架構 43
4.4.2 支援向量迴歸參數之選用 52
4.4.3 模擬結果比較 57
4.5 電量估測要素討論 59
4.5.1 準確度 59
4.5.2 估算效率 60
4.5.3 穩健性 61
第五章 結論與未來研究方向 63
5.1 結論 63
5.2 未來研究方向 63
參考文獻 65

參考文獻 [1] X. Chen, W. Shen, T. T. Vo, Z. Cao, and A. Kapoor, “An overview of lithium-ion batteries for electric vehicles,” 10th International Power & Energy Conference, December 12-14, 2012, Ho Chi Minh City, Vietnam.
[2] V. Pop, H. J. Bergverd, D. Danilov, P. P. L. Regtien, and P. H. L. Notten, Battery management systems: accurate state-of-charge indication for battery-powered applications, Phillips Research Book Series, Springer Verlag, Eindhoven, 2008.
[3] W. Y. Chang, “The state of charge estimating methods for battery: A review”, Hindawi Publishing Corporation, ISRN Applied Mathematics, May 12, 2013.
[4] J. Meng, M. Ricco, G. Luo, M. Swierczynski, D. I. Stroe, A. I. Stroe, and R. Teodorescu, “An Overview and Comparison of Online Implementable SOC Estimation Methods for Lithium-ion Battery,” IEEE Transactions on Industry Applications, accepted.
[5] J. Chen, Q. Ouyang, C. Xu and H. Su, “Neural Network-Based State of Charge Observer Design for Lithium-Ion Batteries,” IEEE Transactions on Control Systems Technology, vol. 26, no. 1, pp. 313-320, January, 2018.
[6] S. Ioannou, K. Dalamagkidis, E. K. Stefanakos, K. P. Valavanis and P. H. Wiley, “Runtime, capacity and discharge current relationship for lead acid and lithium batteries,” 24th Mediterranean Conference on Control and Automation, June, 2016, Athens, Greece.
[7] G. Wu, C. Li, D. Jiao, Y. Liu, C. Hao, Y. Zhang, H. Yu and M. Zhang, “State of charge estimation for Li-Ion battery based on an improved Peukert's equation with temperature correction factor,” IEEE Vehicle Power and Propulsion Conference, pp. 1-4, October17-20, 2016, Hangzhou, China.
[8] 如何選擇演算法 (Retrieval date: December 2017):
https://docs.microsoft.com/zh-tw/azure/machine-learning/studio/algorithm-choice
[9] M. Charkhgard and M. Farrokhi, “State-of-charge estimation for lithium-ion batteries using neural networks and EKF,” IEEE Transactions on Industrial Electronics, vol. 57, no. 12, pp. 4178-4187, December, 2010.
[10] F. Liu, T. Liu and Y. Fu, “An improved SoC estimation algorithm based on artificial neural network,” 2015 International Symposium on Computational Intelligence and Design, December 12-13, 2015, Hangzhou, China.
[11] D. Stathakis, “How many hidden layers and nodes?” International Journal of Remote Sensing, vol. 30, no. 8, pp. 2133-2147, April 20, 2009.
[12] J. C. Álvarez Antón, P. J. García Nieto, C. Blanco Viejo and J. A. Vilán Vilán, “Support vector machines used to estimate the battery state of charge,” IEEE Transactions on Power Electronics, vol. 28, no. 12, pp. 5919-5926, December, 2013.
[13] Electropaedia, Battery and Energy Technologies (Retrieval date: April 2017):http://www.mpoweruk.com/lithium_failures.htm#lifetime
[14] 阮彙權,“鋰離子電池參數估算方法之研究”,國立清華大學電機工程學系,碩士論文,民國99年7月。
[15] M. Chen, and G. A. R. Mora, “Accurate electrical battery model capable of predicting runtime and I–V performance,” IEEE Transactions on Energy Conversion, vol. 21, no. 2, pp. 504-511, June, 2006.
[16] 劉峰其,“非線性鋰電池之充放電模型”,國立中央大學電機工程學系,碩士論文,民國99年6月。
[17] C. Weng, J. Sun, and H. Peng, “An open-circuit-voltage model of Lithium-ion batteries for effective incremental capacity analysis,” Proceedings of the ASME Dynamic Systems and Control Conference, October 21-23, 2013, Palo Alto, California, USA.
[18] M. Y. Ke, Y. H. Chiu, and C. Y. Wu, “Battery modelling and SOC estimation of a LiFePO4 battery,” 2016 International Symposium on Computer, Consumer and Control, pp. 208-211, July 4-6, 2016.
[19] T. Kim and W. Qiao, “A hybrid battery model capable of capturing dynamic circuit characteristics and nonlinear capacity effects,” IEEE Transactions on Energy Conversion, vol. 26, no. 4, pp. 1172-1180, December, 2011.
[20] S. A. Sharkh, and D. Doerffel, “Rapid test and non-linear model characterisation of solid-state Lithium-ion batteries,” Journal of Power Sources, vol. 130, no. 1-2, pp. 266-274, May 2004.
[21] K. C. Bae, S. C. Choi, J. H. Kim, C. Y. Won, and Y. C. Jung, “LiFePO4 dynamic battery modeling for battery simulator,” IEEE International Conference on Industrial Technology, September, 2014, Busan, South Korea.
[22] 羅華強,“類神經網路-MATLAB的應用”,清蔚科技股份有限公司,民國99年9月。
[23] 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, May/June, 2015.
[24] C. T. Lin and C. S. Lee, Neural Fuzzy Systems: A neuro-fuzzy synergism to intelligent systems, Upper Saddle River, NJ, USA, Prentice-Hall Inc., 1996.
[25] 林大仁,“運用支撐向量迴歸與倒傳遞類神經網路建構產品造形設計之預測模式”,國立成功大學工業設計研究所,碩士論文,民國96年6月。
[26] I. N. Haq, R. H. Saputra, F. Edison, D. Kurniadi, E. Leksono and B. Yuliarto, “State of charge (SoC) estimation of LiFePO4 battery module using support vector regression,” Proceedings of the Joint International Conference on Electric Vehicular Technology and Industrial, Mechanical, Electrical and Chemical Engineering (ICEVT & IMECE), pp. 16-21, 2015, Surakarta.
[27] D. Doerffel and S. A. Sharkh, “A critical review of using the Peukert equation for determining the remaining capacity of lead-acid and lithium-ion batteries,” Journal of Power Sources, pp. 395-400, June 23, 2005.
[28] J. Lee, Y. Kim, and H. Cha, “A new battery parameter identification considering current, SOC and Peukert’s effect for hybrid electric vehicles,” IEEE Energy Conversion Congress and Exposition, November, 2011, Phoenix, AZ, USA.
[29] LIBSVM: a library for support vector machines (Retrieval date: September 2017): http://www.csie.ntu.edu.tw /~ cjlin/libsvm/.
[30] 損失函數-經驗風險最小化-結構風險最小化 (Retrieval date: October 2017):http://www.voidcn.com/article/p-rzyzscpf-ben.html.
[31] 黃珮瑜,“支持向量機於颱洪時期流量預報之研究”,國立臺灣大學土木工程研究所,碩士論文,民國98年7月。
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2023-02-01起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2023-02-01起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw