太陽光電發電系統中，由於線路老化或其它原因，可能會發生包括串聯電弧故障
(SAF)和並聯電弧故障(PAF)在內的電弧故障。如果沒有及時發現和處理，持續電弧故
障帶來的高溫可能會導致火災等重大安全事故。此外，因爲串聯電弧故障不會有劇烈
電流變化，較難檢測，因此可靠的電弧故障檢測演算法開發於實務應用是十分必要的。
為了保護光伏電站的安全，本文首先提出了一種基於優化變分模態分解(OVMD)
和支持向量機(SVM)的智慧型檢測演算法。該演算法利用優化變分模態分解從電流信
號中提取故障資訊，然後利用自適應特徵篩選(AFS)對信號各頻段的統計資訊進行篩
選。將與分類強相關的特徵作為輸入，藉助粒子群演算法（PSO）優化後的支援向量
機進行分類。該智慧型演算法不僅能準確識別發生在不同位置的串聯電弧故障，還能
識別發生在不同位置的並聯電弧故障。在出現動態遮蔭、逆變器啟動、受風干擾的串
聯電弧故障等情況下，均能保持良好的診斷結果。此外，本文還以不同地區的單串伏
發電系統和多串太陽能光伏發電系統為例，驗證了該演算法的通用性。實驗結果表明，
在所有檢測條件下，檢測準確率均在 98.21%以上。
II
基於擴展實用性及降低執行時間的需求，本文進一步提出了一種基於經驗模態分
解(EMD)和門控循環單元神經網路(GRU-NN)的較短執行時間之電弧故障智慧檢測演
算法。該演算法利用經驗模態分解從電流信號中提取故障資訊，然後根據模態順序對
經驗模態分解中各模態的統計指標進行排序。此外，利用門控循環單元神經網捕捉不
同模態間的特徵和變化趨勢，實現電弧故障檢測。實驗結果表明，在所有檢測條件下，
檢測準確率均在 98.72%以上。此外，該方法還提出了一種線上更新策略，更進一步
提高所提出演算法的適應性。結合該線上更新策略，可以快速修正模型，即使在不同
的光伏電站也能夠保證電弧故障識別的準確性。線上更新的性能也將通過在分別位於
臺灣和中國大陸的光伏電站的實驗來進行驗證。

In a solar photovoltaic (PV) power generation system, arc faults including series arc fault (SAF) and parallel arc fault (PAF) may occur due to aging of joints or other reasons. It may lead to a major safety accident such as fire if the high temperature caused by the continuous arc fault is not identified and solved in time. What’s worse, the SAF without drastic current change is difficult to detect. Thus, the development of reliable arc fault detection algorithms is essential in practical applications.
In order to protect the safety of PV power stations, an intelligent detection algorithm based on the optimized variational mode decomposition (OVMD) and the support vector machine (SVM) is first investigated in this thesis. The proposed algorithm uses the VMD to extract the fault information from current signals, and then screens the statistical information of the signals in each frequency band by the proposed adaptive feature screening (AFS). The features to be strongly correlated with classification are taken as inputs into the SVM optimized by the particle swarm optimization (PSO) for classification eventually. This intelligent framework not only can accurately identify the SAF occurring at different locations, but also identify the PAF. Moreover, it also can maintain good diagnosing results under the occurrence of dynamic shading, inverter startup, and SAF under wind blowing. In addition, single series PV string and solar PV power generation systems in different countries are also used to examine the universal ability of the proposed algorithm. As for experimental results, the detection accuracy is more than 98.21% under all examined conditions.
Due to the requirement of practicality and the reduction of execution time, an intelligent arc-fault detection algorithm with short execution time based on the empirical mode decomposition (EMD) and the gate recurrent unit neural network (GRU-NN) is further investigated in this thesis. The proposed algorithm uses the EMD to extract the fault information from current signals, and then sequences the statistical indexes of each mode from the EMD according to modal orders. Moreover, the GRU-NN is used to capture the features and variation trends among different modes, and realize the arc-fault detection. As for experimental results, the detection accuracy is over 98.72% under all examined conditions. In addition, an online updating method is also proposed in this thesis to ensure the adaptability of the proposed algorithm. Combined with this online-updating method, the proposed scheme could quickly modify the model and ensure the accuracy of the arc fault identification, even for different PV stations. The performance of online updating ability will be also verified by experiments in Taiwan and China mainland PV stations.

References
[1] I. Colak, H. Wilkening, G. Fulli, J. Vasiljevska, F. Issi, and O. Kaplan, “Analysing the efficient use of energy in a small smart grid system,” in 2012 Int. Conf. Renew. Energ. Res. Appl., NGS, JPN, 2012, pp. 1-4.
[2] A. Amiri, H. Ssamet, and T. Ghanbari, “Recurrence plots based method for detecting series arc faults in photovoltaic systems,” IEEE Trans. Ind. Electron., vol. 69, no. 6, pp.6308-6315, June 2022.
[3] Y. Wang, X. Lin, and M. Pedram, “A near-optimal model-based control algorithm for households equipped with residential photovoltaic power generation and energy storage systems,” IEEE Trans. Sustain. Energy, vol. 7, no. 1, pp. 77-86, Jan. 2016.
[4] N. M. Haegel, H. Atwater JR., T. Barbes, A. Burrell, Y. M. Chiang, et al., ‘‘Terawatt-scale photovoltaics: Transform global energy,’’ Science, vol. 364, no. 6443, pp. 836-838, May 2019.
[5] REN21, “Renewables 2021 global status report: Market and industry trends,” Paris: REN21, 2021.
[6] S. Lu, B. T. Phung, D. Z S. Lu, B. T. Phung, and D. Zhang, “A comprehensive review on DC arc faults and their diagnosis methods in photovoltaic systems,” Renew. Sustain. Energy Rev., vol. 89, pp. 88-98, June 2018.
[7] Commercial Rooftop Solar Safety, Clean Energy Associates (CEA), Accessed: June 26, 2022. [Online]. Available: https://www.cea3.com/commercial-rooftop-solar-safety.
[8] 自來水園區太陽能板起火，暴露運維與保險需求。 Accessed: June 25, 2022. [Online]. Available: https://blog.xuite.net/shaoweiwu088/project/439056586-105.08.08.
[9] 台南雞舍大火,屋頂裝設太陽能板增加灌救難度。 Accessed: June 25, 2022. [Online]. Available: https://news.pts.org.tw/article/546914.
[10] 嘉義縣太陽能光電爆炸，電死2千頭種母豬、肉豬。 Accessed: June 25, 2022. [Online]. Available: https://udn.com/news/story/7320/6396268?utm_source =udnplus.
[11] M. K. Alam, F. Khan, J. Johnson, and J. Flicker, “A comprehensive review of catastrophic faults in PV arrays: types, detection, and mitigation techniques,” IEEE J. Photovolt., vol. 5, no. 3, pp. 982-997, Feb. 2015.
[12] S. R. Madeti and S.N. Singh, “A comprehensive study on different types of faults and detection techniques for solar photovoltaic system,” Solar Energ., vol. 158, pp. 161-185, Aug. 2017.
[13] UL 1699B-Outline of Investigation for Photovoltaic (PV) DC Arc-Fault Circuit Protection, Underwriters Laboratories, Northbrook, IL, USA, Jan. 2013.
[14] National Electrical Code, Article 690—Solar Photovoltaic Systems, 2011.
[15] C. He, L. Mu, and Y. Wang, “The detection of parallel arc fault in photovoltaic systems based on a mixed criterion,” IEEE J. Photovolt., vol. 7, no. 6, pp. 1717-1724, Nov. 2017.
[16] C. Strobl and P. Meckler, “Arc faults in photovoltaic systems,” in Proc. 56th IEEE Conf. Elect. Contacts, SC, USA, 2010, pp. 1-7.
[17] S. McCalmont, “Low cost arc fault detection and protection for PV systems: January 30, 2012 - September 30, 2013,” Report of National Renewable Energy Laboratory (NREL), Oct. 2013. [Online] Available: www.nrel.gov/publications.
[18] S. Lu, B.T. Phung, and D. Zhang, “A comprehensive review on DC arc faults and their diagnosis methods in photovoltaic systems,” Renew. Sust. Energ. Rev., vol. 89, pp.88-89, Mar. 2018.
[19] H. Haeberlin and M. Real, “Arc detector for remote detection of dangerous arcs on the DC side of PV plants,” in 22nd European Photovolt. Solar Energ. Conf., MI, IT, Sept. 2007.
[20] Q. Xiong, S. Ji, L. Zhu, L. Zhong, and Y. Liu, “A novel DC arc fault detection method based on electromagnetic radiation signal,” IEEE Trans. Plasma Sci., vol. 45, no. 3, pp. 472-478, Mar. 2017.
[21] N. L. Georgijevic, M. V. Jankovic, S. Srdic, and Z. Radakovic, “The detection of series arc fault in photovoltaic systems based on the arc current entropy,” IEEE Trans. Power Electron., vol. 31, no. 8, pp. 5917-5930, Aug. 2016.
[22] Y. Liu, F. Guo, Z. Ren, P. Wang, T. N. Nguyen, J. Zheng, and X. Zhang, “Feature analysis in time-domain and fault diagnosis of series arc fault,” in 2017 IEEE Conf. Elect. Contacts, CO, USA, 2017, pp. 306-311.
[23] M. Ahmadi, H. Samet, and T. Ghanbari, “A new method for detecting series arc fault in photovoltaic systems based on the blind-source separation,” IEEE Trans. Ind. Electron., vol. 67, no. 6, pp. 5041-5049, June 2020.
[24] M. Ahmadi, H. Samet, and T. Ghanbari, “Series arc fault detection in photovoltaic systems based on signal-to-noise ratio characteristics using cross-correlation function,” IEEE Trans. Ind. Electron., vol. 16, no. 5, pp. 3198-3209, May 2020.
[25] K. Xia, Z. He, Y. Yuan, Y. Wang, and P. Xu, “An arc fault detection system for the household photovoltaic inverter according to the DC bus currents,” in 2015 18th Int. Conf. Elect. Mach. Syst., PYX, TH, 2015, pp. 1687-1690.
[26] S. Chae, J. Park, and S. Oh, “Series DC arc fault detection algorithm for DC microgrids using relative magnitude comparison,” IEEE Trans. Emerg. Sel. Topics Power Electron., vol. 4, no. 4, pp. 1270-1278, Dec. 2016.
[27] S. Chen, X. Li, and J. Xiong, “Series arc fault identification for photovoltaic system based on time-domain and time-frequency-domain analysis,” IEEE J. Photovolt., vol. 7, no. 4, pp. 1105-1114, July 2017.
[28] S. Liu, L. Dong, X. Liao, X. Cao, X. Wang, and B. Wang, “Application of the variational mode decomposition-based time and time–frequency domain analysis on series dc arc fault detection of photovoltaic arrays,” IEEE Access, vol. 7, pp. 126177-126190, Sept. 2019.
[29] W. Fenz, S. Thumfart, R. Yatchak, H. Roitner, and B. Hofer, “Detection of arc faults in PV systems using compressed sensing,” IEEE J. Photovolt., vol. 10, no. 2, pp. 676-684, Jan. 2020.
[30] S. Chen, X. Li, Y. Meng, and Z. Xie, “Wavelet-based protection strategy for series arc faults interfered by multicomponent noise signals in grid-connected photovoltaic systems,” Solar Energy, vol. 183, pp. 327-336, Mar. 2019.
[31] K. Xia, S. He, Y. Tan, Q. Jiang, J. Xu, and W. Yu, “Wavelet packet and support vector machine analysis of series dc arc fault detection in photovoltaic system,” 2018 Electr. Eng., JPN, vol. 14, pp. 192-200, Feb. 2019.
[32] W. Miao, Q. Xu, K. H. Lam, P. W. T. Pong, and H. V. Poor, “DC arc-fault detection based on empirical mode decomposition of arc signatures and support vector machine,” IEEE Sens. J., vol. 21, no. 5, pp. 7024-7033, Mar. 2021.
[33] Z. Yin, L. Wang, B. Zhang, L. Meng, and Y. Zhang, “An integrated DC series arc fault detection method for different operating conditions,” IEEE Trans. Ind. Electron., vol. 68, no. 12, pp. 12720-12729, Dec. 2020.
[34] W. Gao and R. J. Wai, “Series arc fault detection of grid-connected PV system via SVD denoising and IEWT-TWSVM,” IEEE J. Photovolt., vol. 11, no. 6, pp. 1493-1510, Nov. 2021.
[35] S. Lu, T. Sirojan, B. T. Phung, D. Zhang, and E. Ambikairajah, “DA-DCGAN: An effective methodology for dc series arc fault diagnosis in photovoltaic systems,” IEEE Access, vol. 7, pp. 45831-45840, Apr. 2019.
[36] H. L. Dang, S. Kwak, and S. Choi, “Different domains based machine and deep learning diagnosis for DC series arc failure,” IEEE Access, vol. 9, pp. 166249-166261, Dec. 2021.
[37] X. Li, M.Liu, and S. Wang, “Research on the EEMD algorithm of penetration acceleration signal processing based on independent component analysis,” in 2010 3rd Int. Congr. Image and Signal Process., TYN, CN, 2010, pp. 4135-4138.
[38] Z. Tan, J. Chen, Q. Kang, M. Zhou, A. Abusorrah, and K. Sedraoui, “Dynamic embedding projection-gated convolutional neural networks for text classification,” IEEE Trans. Neural Netw. Learn. Syst., vol. 33, no. 3, pp. 973-982, Mar. 2022.
[39] M. Khodayar, G. Liu, J. Wang, O. Kaynak, and M. E. Khodayar, “Spatiotemporal behind-the-meter load and PV power forecasting via deep graph dictionary learning,” IEEE Trans. Neural Netw. Learn. Syst., vol. 32, no. 10, pp. 4713-4727, Dec. 2020.
[40] V. Le, X. Yao, C. Miller, and B. Tsao, “Series DC arc fault detection based on ensemble machine learning,” IEEE Trans. Power Electron., vol. 35, no. 8, pp. 7826–7839, Aug. 2020.
[41] K. Dragomiretskiy and D. Zosso, “Variational mode decomposition,” IEEE Trans. Signal Process., vol. 62, no. 3, pp. 531-544, Feb. 2014.
[42] L. Wang, H. Qiu, P. Yang, and L. Mu, “Arc fault detection algorithm based on variational mode decomposition and improved multi-scale fuzzy entropy,” Energies, vol. 14, no. 14, Article 4137, July 2021.
[43] T. Ma, E. Tian, Z. Liu, S. Liu, T. Guo, T. Wang, and L. Fu, “Detection of DC series arc fault based on VMD and ELM,” in J. Phys.: Conf. Ser., vol. 1486, pp. 62037-62042, Apr. 2021.
[44] M. Ju and L. Wang, “Arc fault modeling and simulation in DC system based on Habedank model,” in 2016 Progn. Syst. Health Manage. Conf., CN, USA, 2016, pp. 1-4.
[45] S. Boyd, “Multitone signals with low crest factor,” IEEE Trans. Circuits Syst., vol. 33, no. 10, pp. 1018-1022, Oct. 1986.
[46] P. W. Kang, Y. Guo, and Y. Gao, “Intrinsic mode function determination of faulty rolling element bearing based on kurtosis,” in 2015 IEEE Int. Conf. Inf. Automat., LJG, CN, 2015, pp. 1536-1540.
[47] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, pp. 273-297, Mar. 1995.
[48] J. Bergstra and Y. Bengio, “Random search for hyper-parameter optimization,” J. Mach. Learn. Res., vol. 13, pp. 281-305, Dec. 2012.
[49] Z. Beheshti and S. M. Shamsuddin, “A review of population-based meta-heuristic algorithms,” Int. J. Adv. Soft Comput. Appl., vol. 5, no. 1, pp. 1-35, Mar. 2013.
[50] J. Kennedy and R. Eberhart, “Particle swarm optimization,” in Proc. IEEE Int. Conf. Neural Netw., WA, AUS, 1995, pp. 1942-1948.
[51] N. E. Huang et al., “The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis,” Proc. Roy. Soc. London. A, Math., Phys. Eng. Sci., vol. 454, no. 1971, pp. 903–995, Mar. 1998.
[52] K. Cho, B. van Merrienboer, D. Bahdanau, and Y. Bengio, “On the properties of neural machine translation: Encoder–decoder approaches,” Oct. 2014, arXiv:1409.1259. [Online] Available: http://arxiv.org/abs/1409.1259
[53] D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” Dec. 2014, arXiv:1412.6980. [Online] Available: https://arxiv.org/abs/1412.6980
[54] W. Lu, B. Liang, Y. Cheng, D. Meng, J. Yang, and T. Zhang, “Deep model based domain adaptation for fault diagnosis,” IEEE Trans. Ind. Electron., vol. 64, no. 3, pp. 2296-2305, Mar. 2017.
[55] L. Guo, M. Li, S. Xu, F. Yang, and J. Zhang, “Investigation of Adam for low-frequency electromagnetic problems,” in Int. Conf. Numer. Electromagn. Multiphys. Model. Optim., 2020, pp. 1293-1298.
[56] S. Ruder, “An overview of gradient descent optimization algorithms,” Sept. 2016, arXiv:1609.04747. [Online] Available: https://arxiv.org/abs/1609.04747
[57] N. Qian, “On the momentum term in gradient descent learning algorithms,” Neural Networks, vol. 12, no. 1, pp. 145-151, Jan. 1999.
[58] G. Huang, Q. Zhu, and C. Siew, “Extreme learning machine: theory and applications,” Neurocomputing, vol. 70, pp. 489-501, May 2006.
[59] X. R. Cai and R. J. Wai, “Intelligent DC arc-fault detection of solar PV power generation system via optimized VMD-based signal processing and PSO-SVM classifier,” IEEE J.Photovolt., vol. 12, pp. 1058-1077, Jul. 2022.
[60] L. Kang, S. Hong, S. Lee, and J. Ahn, “Apparatus for detecting arc,” TW Patent I431557, Apr. 16, 2011.
[61] T. Charles, B. Bruce, H. Gerlomon, K. Robert, and O. Wurt, “Arc detection using electromagnetic sensing,” TW Patent 222355, Apr. 11, 1994.
[62] G. Zhang, X. Zhang, H. Liu, T. Zhang, H. Ji, and Y. Sun, “Series fault arc detection method and its special device,” CN Patent 103915818, July 9, 2014.
[63] H.Yang, X. Fan, Y. Zheng, Z. Zhang, and J. Xue, “Photovoltaic DC arc fault identification method and device based on random forest algorithm,” CN Patent 114169398, Mar. 11, 2022.
[64] H. Hu, B.Chen, Y. Liang, and B. Yan, “A invention relates to a photovoltaic element level DC arc fault detection and fast shutdown device and method,” CN Patent 114567253, May 31, 2022.