電力系統面對難以避免的干擾，在加裝保護設備或因應改善之道前，先進行電力品質干擾事件辨識分類，便能依其類別找出故障發生的原因，進而改善系統供電品質。本論文提出一個辨識分類的架構，區分電力受干擾而引發不正常時的情形，如電壓突升、電壓突降及電壓閃爍等，利用動態時間扭曲法與粒子群最佳化演算法作為比對辨識的法則。
本論文提出兩個方法作電力品質干擾事件之辨識分類。一是利用動態時間扭曲法，首先在測試訊號進入擷取參數之前，以啟發式法則作適當分類，再利用快速傅立葉轉換或華爾斯轉換，得到波形的特徵參數，再以向量量化的觀念，將特徵參數作另一個變化來加快辨識速度，最後再利用動態時間扭曲法作為辨識法則，針對可能的類型去做比對辨識的工作，得到區分結果。本論文並對此一辨識系統容許雜訊干擾的能力作一探討分析。
二是利用粒子群最佳化演算法，測試訊號以粒子群最佳化演算法作為辨識分類法則，利用資料庫中的參考樣本去做比對辨識的工作，得到辨識分類結果。為了得到良好的辨識分類結果，本文針對粒子群最佳化演算法的特性，改變一些參數及資料庫內的參考樣本數多寡，作了一些的探討與分析，以得到更好的辨識分類結果。

Power system confronts power quality disturbances that are hard to avoid. Before assembling a protective device or finding a way of improvement, first the disturbance types need to be identified. Then, we can know what disturbances belong and why power system hitches. And we can get better power quality. This paper presents two approaches based on dynamic time warping and particle swarm optimization for classification of power quality disturbances.
The first approach is based on heuristic rules and dynamic time warping for classification of power quality disturbances. In the classification process, the Walsh transform and fast Fourier transform are first used to get feature parameters for the input signals. Then, the vector quantization is used to speed up the dynamic time warping operation. Moreover, in order to reduce the dynamic time warping computational cost and increase the classification accuracy, the heuristic rules are introduced. Finally the effectiveness of the proposed approach is demonstrated by disturbance classification.
The second approach is based on particle swarm optimization for classification of power quality disturbances. The test signal and reference pattern are matched with particle swarm optimization. It is changed that the characteristics of particle swarm optimization for get better results. It is concluded from the results that the method is very effective for classification power quality disturbances.

中文摘要 --------------------------------------------------------------- i
英文摘要 --------------------------------------------------------------- ii
致謝 --------------------------------------------------------------- iii
目錄 --------------------------------------------------------------- iv
表目錄 --------------------------------------------------------------- vii
圖目錄 --------------------------------------------------------------- viii
第一章 緒論--------------------------------------------------------- 1
1.1 研究背景與動機------------------------------------------ 1
1.2 研究方法--------------------------------------------------- 2
1.3 論文大綱--------------------------------------------------- 3
第二章 電力品質干擾事件的辨識分類------------------------ 5
2.1 前言--------------------------------------------------------- 5
2.2 文獻回顧--------------------------------------------------- 5
2.3 本論文辨識分類之架構--------------------------------- 7
2.3.1 訊號之獲取------------------------------------------------ 8
2.3.2 訊號特徵參數的擷取------------------------------------ 9
2.3.2.1 利用離散傅立葉轉換------------------------------------ 9
3.2.2.2 利用華爾斯轉換------------------------------------------ 10
2.3.3 向量量化--------------------------------------------------- 10
2.3.4 辨識分類法則--------------------------------------------- 12
2.4 採用的電力品質干擾事件訊號------------------------ 12
2.5 本章結論--------------------------------------------------- 19
第三章 利用啟發式法則與動態時間扭曲法於電力品質干擾事件之辨識分類--------------------------------------- 20
3.1 前言--------------------------------------------------------- 20
3.2 啟發式法則作測試訊號的分類------------------------ 22
3.3 動態時間扭曲法------------------------------------------ 24
3.4 動態時間扭曲法作樣本辨識分類比對--------------- 24
3.4.1 決定測試樣本與參考樣本之間的距離--------------- 25
3.4.2 決定測試樣本與參考樣本的最佳匹配--------------- 28
3.5 測試與結果之探討--------------------------------------- 29
3.5.1 資料庫與編碼簿大小之探討--------------------------- 31
3.5.2 快速傅立葉轉換與華爾斯轉換之特徵參數擷取之探討--------------------------------------------------------- 33
3.5.3 有無加入啟發式法則分類對辨識結果的影響之探討------------------------------------------------------------ 34
3.5.4 雜訊干擾對辨識分類結果的影響之探討------------ 36
3.6 本章結論--------------------------------------------------- 37
第四章 利用粒子群最佳化演算法於電力品質干擾事件之辨識分類--------------------------------------------------- 38
4.1 前言--------------------------------------------------------- 38
4.2 粒子群最佳化演算法------------------------------------ 38
4.3 以粒子群最佳化演算法作電力品質干擾事件之辨識分類------------------------------------------------------ 40
4.3.1 粒子參數組及能量函數的建立------------------------ 40
4.3.2 以粒子群最佳化演算法作電力品質干擾事件之辨識分類演算法的步驟------------------------------------ 43
4.4 測試與結果之探討--------------------------------------- 48
4.4.1 架構一------------------------------------------------------ 48
4.4.1.1 粒子數目的多寡對其辨識率影響之探討------------ 51
4.4.2.2 干擾事件參考樣本數的多寡對其辨識率影響之探討------------------------------------------------------------ 53
4.4.2 架構二------------------------------------------------------ 56
4.4.2.1 資料庫大小與粒子數目多寡之探討------------------ 57

4.4.2.2 快速傅立葉轉換與華爾斯轉換之特徵參數擷取之探討--------------------------------------------------------- 60
4.4.2.3 有無加入啟發式法則分類對辨識結果的影響之探討------------------------------------------------------------ 61
4.4.2.4 雜訊干擾對辨識結果的影響之探討------------------ 63
4.4.2.5 粒子群最佳化演算法與類神經網路之比較--------- 63
4.5 本章結論--------------------------------------------------- 66
第五章 結論與未來研究方向------------------------------------ 67
5.1 結論--------------------------------------------------------- 67
5.2 未來研究方向--------------------------------------------- 68
參考文獻 --------------------------------------------------------------- 69
作者簡介 --------------------------------------------------------------- 72

[1]胡忠興，工業配電系統之電力品質與抑制對策，電機月刊，第九卷第二期，1999。
[2]江榮城，電力品質實務（一），全華科技圖書股份有限公司，台北，2000。
[3]周翌生，電力系統諧波問題之探討，電機月刊，第六卷第十期，1999。
[4]陳建富，電力品質監測儀表系統設計，中原大學電機系碩士論文，1999。
[5]G. T. Heydt, P. S. Fjeld, C. C. Liu, D. Pierce, L. Tu, and G. Hensley, “Applications of the windowed FFT to electric power quality assessment,” IEEE Trans. on Power Delivery, Vol. 14, pp. 1411-1416, 1999.
[6]S. Santoso, W. M. Grady, E. J. Powers, J. Lamoree, and S. C. Bhatt, ” Characterization of distribution power quality events with Fourier and wavelet transforms,” IEEE Trans. on Power Delivery, Vol. 15, pp. 247-254, 2000.
[7]A. M. Youssef, T. K. Abdel-Galil, E. F. El-Saadany, and M. M. A. Salama, “Disturbance classification utilizing dynamic time warping classifier,” IEEE Trans. on Power Delivery, Vol. 19, pp. 272-278, 2004.
[8]P. K. Dash, S. Mishra, M. M. A. Salama, and A. C. Liew, “Classification of power system disturbances using a fuzzy expert system and a Fourier linear combiner,” IEEE Trans. on Power Delivery, Vol. 15, pp. 472–477, 2000.
[9]W. Min, P. Ochenkowski, and A. Mamishev, ” Classification of power quality disturbances using time-frequency ambiguity plane and neural networks,” Proceedings of the IEEE Power Engineering Society Summer Meeting, Vol. 2, pp. 1246–1251, 2001.
[10]J. H. Huang, M. Negnevitsky, and D. T. Nguyen, “A neural-fuzzy classifier for recognition of power quality disturbances,” IEEE Trans. on Power Delivery, Vol. 17, pp. 609-616, 2002.
[11]IEEE recommended practice for monitoring electric power quality, IEEE standard 1159-1995, New York, 1995.
[12]J. V. D. Vegte, Fundamentals of Digital Signal Processing, Upper Saddle River, NJ: Prentice Hall, 2002.
[13]N. Ahmed, and K. R. Rao, Orthogonal Transforms for Digital Signal Processing, New York: Springer-Verlag, 1975.
[14]A. Gersho, R. M. Gray, and R. Gallager, Vector Quantization and Signal Compression, Norwell, MA: Kluwer, 1991.
[15]L. Rabiner, and B. Juang, Fundamentals of Speech Recognition, Englewood Cliffs, NJ: Prentice-Hall, 1993.
[16]J. R. Deller, J. H. L. Hansen, and J. G. Proakis, Discrete-Time Processing of Speech Signals, Piscataway, NJ: IEEE Press Classic Reissue, 2000.
[17]B. Schuller, “Towards intuitive speech interaction by the integration of emotional aspects,” Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, Vol. 6, 2002.
[18]S. Naka, T. Genji, T. Yura, and Y. Fukuyama, “A hybrid particle swarm optimization for distribution state estimation,” IEEE Trans. on Power Systems, Vol. 18, No. 1, pp. 60-68, 2003.
[19]T. Tsukada, T. Tamura, S. Kitagawa, and Y. Fukuyama, “Optimal operational planning for cogeneration system using particle swarm optimization,” Proceedings of the IEEE on Swarm Intelligence Symposium, pp. 138-143, 2003.
[20]曾俊傑，一個智慧型指紋辨識系統的設計方法論，義守大學電機系碩士論文， 2001.
[21]X. H. Zhi, X. L. Xing, Q. X. Wang, L. H. Zhang, X. W. Yang, C. G. Zhou, and Y. C. Liang, “A discrete PSO method for generalized TSP problem,” Proceedings of 2004 International Conference on Machine Learning and Cybernetics Vol. 4, pp. 2378 – 2383, 2004.
[22]J. Kennedy, and R. Eberhart, “Particle swarm optimization,” Proceedings of the IEEE International Conference on Neural Networks, Vol. 4, pp. 1942-1948, 1995.
[23]R. Eberhart, and J. Kennedy, “A new optimizer using particle swarm theory,” Proceedings of the Sixth International Symposium on Micro Machine and Human Science, pp. 39-43, 1995.