臺灣博碩士論文加值系統

現今電池電荷狀態估算法有開路電壓法、庫倫積分法、擴展式卡爾曼濾波器及類神經網路法等等，但由於鋰電池特性在其80%至20%充放電期間電壓幅度較小，不適合以電壓估算法進行估測，又因庫倫積分法有累積誤差及初始值誤差等缺點，為解決上述問題，本文採用類神經網路，估算電池的電荷狀態。
本論文以磷酸鋰鐵電池作為實驗的電池，並將易於觀測的電壓、電流及溫度作為輸入參數，分別使用長短期記憶、雙向長短期記憶和門控循環單位三種神經網路，估算出電荷狀態，並使用不同的溫度進行混合脈波能力特性測驗及動態壓力測試，實驗在Simulink進行動態即時的電池電荷狀態估算。結果表明 ，門控循環單位模型對於複雜的電荷狀態估算的準確率會降低，長短期記憶及雙向長短期記憶模型則對異常的雜訊較敏感。

Nowadays, there are open circuit voltage method, Coulomb integral method, extended Kalman filter and artificial neural network method to estimate the SOC of battery, due to the characteristics of lithium battery, the voltage range is small during the 80% to 20% charge and discharge period. It is not suitable to use the voltage estimation method for estimation, and because the Coulomb integral method has shortcomings such as cumulative errors and initial value errors, in order to solve the above problems, this paper use artificial neural network to estimate the SOC of the battery.
In this thesis, the LiFePO4 battery is used as the experimental battery. The easily observed voltage, current and temperature are used as input parameters, and the SOC is estimated by LSTM, Bi-LSTM and GRU artificial neural network respectively with different temperatures in Hybrid pulse power characterization test and dynamic stress test. The experiment performs dynamic and real-time battery SOC estimation in Simulink. The results show that the accuracy of the GRU model for complex SOC estimation will decrease, while the LSTM and Bi-LSTM models are more sensitive to abnormal noise.

摘要.....................................i
Abstract................................ii
誌謝.....................................iii
目錄.....................................iv
表目錄...................................vi
圖目錄...................................vii
第一章 緒論.............................1
1.1、 研究背景與動機....................1
1.2、 文獻回顧.........................1
1.3、 論文架構與貢獻....................2
第二章 電池及相關技術介紹................3
2.1、 電池等效模型......................3
2.2、 電池電荷狀態......................4
第三章 深度學習網路及應用................6
3.1、 前言.............................6
3.2、 神經網路.........................6
3.3、 激活函數.........................6
3.4、 損失函數.........................7
3.5、 優化器...........................8
3.6、 Epoch、Batch Size及Iteration.....9
3.7、 RNN..............................10
3.8、 LSTM.............................10
3.9、 Bi-LSTM..........................11
3.10、 GRU..............................12
第四章 研究內容與方法....................14
4.1、 電池特性..........................14
4.2、 電池實驗設備......................15
4.3、 實驗流程..........................16
4.4、 電池實驗測試......................16
4.5、 資料庫建立及資料預處理.............22
4.6、 神經網路模型建置...................25
4.7、 移動平均法(Moving Average)........26
第五章 實驗結果..........................27
5.1、 Simulink..........................27
5.2、 變溫度1C電流HPPC實驗...............28
5.3、 變溫度DST實驗......................33
第六章 結論與未來展望.....................38
6.1、 結論..............................38
6.2、 未來展望...........................38
References................................39
Extended Abstract.........................41

[1]Hardik Keshan, Jesse Thornburg, and Taha Selim Ustun, “Comparison of lead-acid and lithium ion batteries for stationary storage in off-grid energy systems”, IET Clean Energy and Technology Conference, pp.1-7, 2016.
[2]M. A. Hannan, M. M. Hoque, A. Hussain, Y. Yusof and P. J. Ker, "State-of-the-Art and Energy Management System of Lithium-Ion Batteries in Electric Vehicle Applications: Issues and Recommendations," in IEEE Access, vol. 6, pp. 19362-19378, 2018.
[3]P. A. Topan, M. N. Ramadan, G. Fathoni, A. I. Cahyadi and O. Wahyunggoro, "State of Charge (SOC) and State of Health (SOH) estimation on lithium polymer battery via Kalman filter," 2016 2nd International Conference on Science and Technology-Computer (ICST), Yogyakarta, 2016, pp. 93-96.
[4]D. V. Do, C. Forgez, K. El Kadri Benkara and G. Friedrich, "Impedance Observer for a Li-Ion Battery Using Kalman Filter," in IEEE Transactions on Vehicular Technology, vol. 58, no. 8, pp. 3930-3937, Oct. 2009.
[5]Z. Xia and J. A. Abu Qahouq, "State-of-Charge Balancing of Lithium-Ion Batteries With State-of-Health Awareness Capability," in IEEE Transactions on Industry Applications, vol. 57, no. 1, pp. 673-684, Jan.-Feb. 2021
[6]E. Chemali, P.J. Kollmeyer, M. Preindl, A. Emadi, State-of-charge estimation of Li- ion batteries using deep neural networks: a machine learning approach, J. Power Sources 400 (2018) 242–25
[7]H. Chaoui and C. C. Ibe-Ekeocha, “State of charge and state of health estimation for lithium batteries using recurrent neural networks,” IEEE Transactions
[8]S. J. Kim, S. H. Kim, H. M. Lee, S. H. Lim, G. -Y. Kwon and Y. -J. Shin, "State of Health Estimation of Li-Ion Batteries Using Multi-Input LSTM with Optimal Sequence Length," 2020 IEEE 29th International Symposium on Industrial Electronics (ISIE), 2020, pp. 1336-1341, doi: 10.1109/ISIE45063.2020.9152544.
[9]E. Chemali, P. J. Kollmeyer, M. Preindl, R. Ahmed, and A. Emadi, “Long short-term memory networks for accurate state-of-charge estimation of li-ion batteries,” IEEE Transactions on Industrial Electronics, vol. 65, no. 8, pp. 6730–6739, 2017.
[10]N. Wang, Q. Yang, J. Xing, H. Jia and W. Chen, "Thermal Fault Diagnosis Method of Lithium Battery Based on LSTM and Battery Physical Model," 2022 41st Chinese Control Conference (CCC), Hefei, China, 2022, pp. 4147-4152, doi: 10.23919/CCC55666.2022.9902341.
[11]F. Huo and C. -H. Chen, "The State of Charge Estimation Based on GRU for Lithium-ion Batteries," 2022 IEEE 11th Global Conference on Consumer Electronics (GCCE), Osaka, Japan, 2022, pp. 220-221, doi: 10.1109/GCCE56475.2022.10014096.
[12]K. Jia, Z. Gao, M. Li and H. Chai, "An Optimization Algorithm with Hausdorff-like Derivative in GRU for SOC Estimation of Lithium-ion Battery," 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS), Xiangtan, China, 2023, pp. 66-71, doi: 10.1109/DDCLS58216.2023.10166443.
[13] P. R. Teja and R. Shanmughasundaram, "Analysis on SoC estimation of Lithium ion battery using EKF and LSTM," 2021 IEEE Madras Section Conference (MASCON), Chennai, India, 2021, pp. 1-6, doi: 10.1109/MASCON51689.2021.9563396.
[14]A. A. Hussein, "Capacity Fade Estimation in Electric Vehicle Li-Ion Batteries Using Artificial Neural Networks," in IEEE Transactions on Industry Applications, vol. 51, no. 3, pp. 2321-2330, May-June 2015, doi: 10.1109/TIA.2014.2365152.
[15]蕭文長，”基於擴展式卡爾曼濾波器於鋰電池殘電量估算”，國立虎尾科技大學電機工程系碩士論文，一月，2023