Early Prediction of Transient Instabilities Based on Pre-Fault Phasor Measurements using Decision Tree-based Methods

Yıl 2019, , 6 - 14, 01.04.2019

Can Berk Saner Mert Kesici Mohammed Mahdı Yusuf Yaslan , V. M. İstemihan Genç

https://doi.org/10.19113/sdufenbed.474888

Öz

In
recent years, many blackouts occurred in power systems of different parts of
the world, affecting millions of people and causing great economic losses. Power
system stability, which has a critical role in the design and operation of
electrical power systems, maintains its importance today. Monitoring the
stability status of a power system in real time is regarded as a primary task
in preventing system blackouts. This allows of a sufficient amount of time to
take appropriate corrective control actions. In this study, the pre-fault
voltage magnitudes and angles taken from the phasor measurement units (PMU),
clearing time of the fault and topology information of the transmission line
that has been tripped for clearing the fault are used to predict the transient
instabilities by two different methods based on the decision trees. The success
and the effectiveness of the proposed machine learning models are shown as they
are applied to the 127-bus Western Systems Coordinating Council (WSCC) test
system.    

Anahtar Kelimeler

Transient instabilities, Classification, Prediction, Power systems, Machine learning

Güç Sistemlerinde Geçici Hal Kararsızlığının Arıza Öncesi Fazör Ölçümleri Kullanarak Karar Ağacı Tabanlı Kestirimi

Öz

Geçtiğimiz yıllarda, dünya çapında farklı güç sistemlerinde,
çok sayıda geniş çaplı enerji kesintileri meydana gelmiş, bu kesintiler
milyonlarca tüketicinin olumsuz etkilenmesine neden olmuş ve büyük miktarda
mali zararlara neden olmuştur. Elektrik güç sistemi tasarımı ve işletmesinde kritik
bir role sahip sistem kararlılığı, günümüzdeki önemini korumaktadır. Bir güç
sisteminin kararlılık durumunu gerçek zamanlı olarak izlemek, sistem
çökmelerini önlemede birincil öneme sahip bir görev olarak kabul edilmektedir. Şebekenin
kararlılık durumunun gerçek zamanlı olarak izlenmesi, geniş alan izleme, koruma
ve kontrol sistemlerinin verimliliği açısından önemli bir fonksiyondur. Bu
fonksiyon ile düzeltici kontrol eylemlerinin zamanında gerçekleştirilebilmesi
sağlanabilir. Bu çalışmada, güç sisteminde meydana gelebilecek arızalar
öncesinde fazör ölçüm birimlerinden alınan gerilimlere ait genlik ve açıların
yanı sıra, arızanın temizlenme süresi ve arızanın temizlenmesi için devreden
çıkarılan iletim hattına ait topoloji bilgileri de kullanılarak, geçici hal
kararsızlıklarının kestirimi, karar ağaçlarına dayalı iki farklı yöntem ile
gerçekleştirilmiştir. Önerilen makine öğrenmesi modellerinin başarımları ve
etkinlikleri 29 jeneratörlü 127 baralı WSCC (Batı Eyaletleri Koordinasyon
Kurulu) test sisteminde uygulanarak gösterilmiştir.    

Anahtar Kelimeler

Geçici hal kararlılığı, Sınıflandırma, Kestirim, Güç sistemleri, Makine öğrenmesi

Kaynakça

• [1] Mohammed S. M. Mahdi. 2018. Wide-area measurement-based early prediction and corrective control for transient stability in power systems. İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Doktora Tezi, 144s. İstanbul.
• [2] North American Synchrophasor Initiative (NASPI) Time Synchronization Task Force (2017). Time synchronization in the electric power system.
• [3] D. Karlsson and S. . Lindahl, "Wide area protection and emergency control," in Proc. IEEE Power Eng. Soc. General Meeting, 2004, vol. 1, p. 5.
• [4] Dahal, N., L. King, R., & Madani, V. (2012). Online dimension reduction of synchrophasor data. In Transmission and Distribution Conference and Exposition (T&D). Orlando, FL, USA: IEEE PES, 1-7
• [5] Anderson, P. M., Fouad, A. A. (2008). Power system control and stability. Sultan Bazar, India: IEEE India.
• [6] Kundur, P., Balu, N., Lauby, M. (2009). Power system stability and control. New York: McGraw-Hill.
• [7] M. A. Pai (1989). Energy Function Analysis for Power System Stability. Boston : Kluwer Academic Publishers
• [8] Ruiz Vega, D., Pavella, M. (2003). A comprehensive approach to transient stability control: part I-near optimal preventive control. IEEE Transactions On Power Systems, 18(4), 1446-1453.
• [9] Srinivasan, D., Chang, C., Liew, A., & Leong, K. (n.d.). Power system security assessment and enhancement using artificial neural network. Proceedings of EMPD 98. 1998 International Conference on Energy Management and Power Delivery
• [10] C. F. Kucuktezcan, V. M. I. Genc 2010. Dynamic security assessment of a power system based on probabilistic neural networks. IEEE PES Innovative Smart Grid Technologies Conf. Europe (ISGT Europe), 11-13 October, 1-6.
• [11] Lotufo, A., Lopes, M., Minussi, C. (2007). Sensitivity analysis by neural networks applied to power systems transient stability. Electric Power Systems Research, 77(7), 730-738.
• [12] I. B. Sulistiawati, M. Abdillah, A. Soeprijanto 2011. Neural network based transient stability model to analyze the security of Java-Bali 500 kV power system. International Conference on Electrical Engineering and Informatics (ICEEI), 1-6.
• [13] Ferreira, W., Silveira, M., Lotufo, A., & Minussi, C. (2006). Transient stability analysis of electric energy systems via a fuzzy ART-ARTMAP neural network. Electric Power Systems Research, 76(6-7), 466-475.
• [14] V. Vittal 2012. Application of phasor measurements for dynamic security assessment using decision trees. IEEE Power and Energy Society General Meeting, San Diego, USA, 2012.
• [15] Z. H. Rather, L. Chengxi, C. Zhe, C. L. Bak, & P. Thogersen 2013. Dynamic security assessment of Danish power system based on decision trees: Today and tomorrow. IEEE PowerTech, Grenoble, France, 2013.
• [16] Liu, C., Sun, K., Rather, Z., Chen, Z., Bak, C., Thogersen, P., & Lund, P. (2014). A Systematic Approach for Dynamic Security Assessment and the Corresponding Preventive Control Scheme Based on Decision Trees. IEEE Transactions On Power Systems, 29(2), 717-730.
• [17] L. Chengxi, Z. H. Rather, C. Zhe, C. L. Bak, and P. Thogersen 2013. Importance sampling based decision trees for security assessment and the corresponding preventive control schemes: The Danish case study, IEEE PowerTech Grenoble, France, 2013.
• [18] Genc, I., Diao, R., Vittal, V., Kolluri, S. and Mandal, S. (2010). Decision Tree-Based Preventive and Corrective Control Applications for Dynamic Security Enhancement in Power Systems. IEEE Transactions on Power Systems, 25(3), pp.1611-1619.
• [19] Moulin, L., daSilva, A., El-Sharkawi, M., & MarksII, R. (2004). Support Vector Machines for Transient Stability Analysis of Large-Scale Power Systems. IEEE Transactions On Power Systems, 19(2), 818-825.
• [20] Xu, Y., Dong, Z., Zhao, J., Zhang, P., & Wong, K. (2012). A Reliable Intelligent System for Real-Time Dynamic Security Assessment of Power Systems. IEEE Transactions On Power Systems, 27(3), 1253-1263.
• [21] Mahdi, M., & Genc, V. M. I. (2018). Post-fault prediction of transient instabilities using stacked sparse autoencoder. Electric Power Systems Research, 164, 243-252.
• [22] L. Breiman, J. Friedman, R. A. Olshen, & C. J. Stone 1984. Classification and Regression Trees. Belmont, CA: Wadsworth.
• [23] Shi, H. 2007. Best-first Decision Tree Learning. The University of Waikato, Yüksek Lisasns Tezi, Hamilton, New Zealand.
• [24] Meng, Q., Ke, G., Wang, T., Chen, W., Ye, Q., Ya, Z., & Liu, T. (2016). A Communication-Effcient Parallel Algorithm for Decision Tree. In 30th International Conference on Neural Information Processing Systems. Barcelona, Spain, 1271-1279.
• [25] Fan, D. 2008. Synchronized Measurements And Applications During Power System Dynamics. Virginia Polytechnic Institute and State University, Doktora Tezi, 182s, Virginia.
• [26] Beyranvand, P. , Genç, V. M. I. , Çataltepe, Z. 2018. Multilabel learning for the online transient stability assessment of electric power systems. Turkish J. Electr. Eng. Comput. Sci, 26(2018), 2661-2675.
• [27] DSATools. 2018. Dynamic Security Assessment Software Package. http://www.dsatools.com (Erişim Tarihi: 19.10.2018)
• [28] Louppe, G., Wehenkel, L., Sutera, A., Geurts, P. 2013. Understanding variable importances in forests of randomized trees. 6th International Conference on Neural Information Processing Systems. Nevada, 431-439.
• [29] Guyon, I., Weston, J., Barnhilll, S. 2002. Gene Selection for Cancer Classification using Support Vector Machines. Machine Learning, 46(2002), 389-422.