Asistencia al diagnóstico posfalla mediante análisis de componente principales de cantidades incrementales de voltaje para ingenieros de protecciones
Assisting protection engineers in post-event diagnosis through principal component analysis of incremental voltage quantities
DOI:
https://doi.org/10.15446/sicel.v12.127079Palabras clave:
extracción de características, Diagnóstico de Fallas, cortocircuitos, operaciones de maniobras (es)pattern extraction, faults identification, short-circuits, switching operations (en)
Descargas
La interpretación de señales posfalla constituye una responsabilidad fundamental de los ingenieros de protección y del personal de operación, especialmente cuando el origen de una perturbación no resulta inmediatamente evidente. Tradicionalmente, esta tarea se basa en la interpretación manual de oscilogramas y diagramas fasoriales basados en archivos COMTRADE, los cuales, bajo condiciones complejas o límite, pueden resultar ambiguos y difíciles de decodificar. Una revisión exhaustiva de la literatura indica que se han propuesto numerosos enfoques para el análisis posfalla, con especial énfasis en técnicas basadas en la Transformada de Fourier y la Transformada Wavelet. Sin embargo, la efectividad de estas herramientas matemáticas es altamente sensible a la elección de los niveles de descomposición y de las funciones base, tales como determinadas wavelets madre.
En este trabajo se presenta una metodología visual novedosa para el análisis fuera de línea de eventos posfalla, basada en cantidades incrementales de voltaje y análisis de componentes principales. La misma ha sido específicamente diseñada para asistir a especialistas humanos mediante la proyección de datos transitorios de alta dimensionalidad en espacios de baja dimensión, donde los patrones se tornan visualmente distinguibles. Esta reducción dimensional permite a los ingenieros identificar y diferenciar cortocircuitos de maniobras operativas, así como también reconocer características relevantes de la falla, tales como la resistencia de falla, el ángulo de inicio y las fases involucradas.
La metodología fue validada utilizando el Sistema Argentino de Interconexión a 500 kV, simulado mediante el Alternative Transients Program. De este modo, se propone una herramienta diagnóstica complementaria que mejora los métodos tradicionales al proporcionar patrones visuales interpretables. En consecuencia, se reduce la carga cognitiva del personal técnico en campo y se posibilita una toma de decisiones más rápida e informada.
Post-fault signal interpretation is a key responsibility of protection engineers and operations personnel, especially when the origin of a disturbance is not immediately evident. Traditionally, this task relies on the manual interpretation of COMTRADE-based oscillograms and phasor diagrams, which under complex or borderline conditions may be ambiguous and difficult to decode. A comprehensive review of the literature indicates that numerous approaches have been proposed for post-fault analysis, with particular emphasis on techniques based on the Fourier Transform and the Wavelet Transform. However, the effectiveness of these mathematical tools is highly sensitive to the choice of decomposition levels and basis functions, such as specific mother wavelets.
This paper presents a novel visual methodology for offline post-event analysis, based on voltage incremental quantities and principal component analysis. It is specifically designed to assist human experts by projecting high-dimensional transient data into low-dimensional spaces where patterns become visually distinguishable. This dimensionality reduction allows engineers to identify and differentiate short-circuits from switching operation events, while also recognizing key fault characteristics such as fault resistance, inception angle, and involved phases.
The methodology was validated using the Argentine Interconnection System at 500 kV simulated using the Alternative Tran-sients Program. Thus, a complementary diagnostic tool that enhances traditional methods by providing interpretable visual patterns is proposed. In doing so, it reduces the cognitive load on field engineers and enables faster, more informed decision-making.
Referencias
[1] Jared Bestebreur, Eric McCollum, John Town and Andy Gould, “Correlating Protective Relay Reports for System-Wide,” Post-Event Analysis, 2018 IEEE Rural Electric Power Conference.
[2] Manish Kumar Sharma, Prateem Pan, DK Mohanta and RC Jha, “A Window Based Threshold Technique for Improved Post Event Analysis using SPMs”, IEEE International Conference on Technologies for Smart-City Energy Security and Power (ICSESP-2018), March 28- 30, 2018, Bhubaneswar, India.
[3] Brett M. Cockerham and John C. Town, “Understanding the Limitations of Replaying Relay-Created COMTRADE Event Files Through Microprocessor-Based Relays,” Schweitzer Engineering Laboratories, Inc., 20th Annual Georgia Tech Fault and Dis-turbance Analysis Conference, May 2017.
[4] IEEE Std C37.111-2013, “Measuring relays and protection equipment – Part 24: Common format for transient data ex-change (COMTRADE) for power systems”.
[5] IEEE Std C37.118-2005.
[6] Yujie Yin, Saman Alaeddini, and Yong Fu, “Automatic Fault Analysis and Visualization of Digital Substation Event,” 2020 IEEE.
[7] SR Samantaray, I. Kamwa, Geza Joos, “Decision Tree Based Fault Detection and Classification in Distance Relaying”, Inter-national Journal of Engineering Intelligent Systems for Electrical Engineering and Communications, September 2011.
[8] Moez Ben Hessine, Sahbi Marrouchi and Souad Chebbi, “A Fault Classification Scheme with High Robustness for Transmis-sion Lines using Fuzzy-Logic system”, 2017 International Con-ference on Advanced Systems and Electric Technologies, 2017 IEEE.
[9] Bancha Sreewirote and Atthapol Ngaopitakkul, “Classification of Fault Type on Loop-Configuration Transmission System us-ing Support Vector Machine”, 2017 6th IIAI International Con-gress on Advanced Applied Informatics, 2017 IEEE.
[10] Nandhini K. and Prajith CA, “Review on Fault Detection and Classification in Transmission Line using Machine Learning Methods”, 2023 International Conference on Control, Com-munication and Computing (ICCC) 19-21 May 2023, 2023 IEEE.
[11] Ajay Yadav and Shalini Goad, “Fault Detection on Transmission Lines Using Artificial Neural Network”, IJSART - Volume 7 Issue 9 – SEPTEMBER 2021.
[12] Yuanqian Maa, Xianyong Xiaob and Yang Wangb, “Identifying the root cause of power system disturbances based on wave-form templates”, Electric Power Systems Research, Volume 180, March 2020, 106107.
[13] Tarik Hubana, Mirza Saric and Samir Avdaković, “Approach for identification and classification of HIFs in medium voltage distri-bution networks”, IET Gener. Transm. Distrib., 2018, Vol. 12 Iss. 5, pp. 1145-1152.
[14] Mahfuz Ali Shuvra and Alberto Del Rosso, “Root Cause Identi-fication of Power System Faults using Waveform Analytics”, IEEE, 2018.
[15] Daniel Sabin and Jon A. Bickel, Analytics enabled by waveform análisis, Schneider Electric White Paper, 998-21168644_GMA.
[16] Aaron J. Wilson, Donald R. Reising, Robert W. Hay, Ray C. Johnson, Abdelrahman A. Karrar and T. Daniel Loveless, “Au-tomated Identification of Electrical Disturbance Waveforms within an Operational Smart Power Grid”, IEEE Transactions on Smart Grid, 2020.
[17] Daniel Sabin and Jon A. Bickel, Analytics enabled by waveform análisis, Schneider Electric White Paper, 998-21168644_GMA.
[18] Aaron J. Wilson, Donald R. Reising, Robert W. Hay, Ray C. Johnson, Abdelrahman A. Karrar and T. Daniel Loveless, “Au-tomated Identification of Electrical Disturbance Waveforms within an Operational Smart Power Grid”, IEEE Transactions on Smart Grid, 2020.
[19] Mahfuz Ali Shuvra and Alberto Del Rosso, “Root Cause Identi-fication of Power System Faults using Waveform Analytics”, 2018 IEEE.
[20] Hassan Haes Alhelou, Mohamad Esmail Hamedani-Golshan, Takawira Cuthbert Njenda and Pierluigi Siano, “A Survey on Power System Blackout and Cascading Events: Research Moti-vations and Challenges”, Energies 2019, 12, 682; doi:10.3390/en12040682, 20 February 2019.
[21] U.S.-Canada Power System Outage Task Force. (2004). “Final Report: Causes of the August 14, 2003 Blackout”.
[22] “Blackout in Italy: The Causes and Consequences” (2004). Italian Electricity Transmission System Operator (Terna).
[23] California Independent System Operator (CAISO). “2011 California Transmission Outage Report”.
[24] Edmund O. Schweitzer, Bogdan Kasztenny, Armando Guzmán, Veselin Skendzic, and Mangapathirao V. Mynam, Schweitzer En-gineering Laboratories, Inc., “Speed of Line Protection – Can We Break Free of Phasor Limitations?”, 68th Annual Confer-ence for Protective Relay Engineers, IEEE, 2015.
[25] Edmund O. Schweitzer and Bogdan Kasztenny, Schweitzer Engineering Laboratories, Inc., “Distance Protection: Why Have We Started with a Circle, Does It Matter, and What Else Is Out There?”, 71st Annual Conference for Protective Relay Engi-neers, IEEE, 2018.
[26] Edmund o. Schweitzer, Bogdan Kasztenny, Mangapathirao V. Mynam, Normann Fischer and Armando Guzman, “Solving Line Protection Challenges with Transient-Based Relays”, Protection, Automation and Control Magazine, PAC World, March 2023, Volume 63.
[27] Bogdan Kasztenny, Armando Guzmán, Normann Fischer, Mangapathirao V. Mynam, and Douglas Taylor, Schweitzer En-gineering Laboratories, Inc., “Practical Setting Considerations for Protective Relays That Use Incremental Quantities and Travel-ing Waves”, 2016.
[28] I. Jollife. “Principal Component Analysis, Second Edition”, Springer-Verlag, New York, 2002.
[29] John Morales, Flavio Quizhpi, Hector Villarroel-Gutiérrez, Edu-ardo Muñoz, Eduardo Orduña and Ernesto Vasquez, “A Novel Methodology for the Determination of Impulse Response Coef-ficients Applied to Transmission Line Protection Relays”, IEEE Access, 2020.
[30] Gerardo Moreno, John Morales and Eduardo A. Orduña, “Pla-taforma de simulación multipropósito aplicada a sistemas eléc-tricos de potencia”, XVI Simposio Iberoamericano Sobre Pro-tección de Sistemas Eléctricos de Potencia (2024) 1–7, Univer-sidad Autónoma de Nuevo León, México.
Cómo citar
APA
ACM
ACS
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Descargar cita
Licencia
Derechos de autor 2026 José Gerardo Moreno Bustos Moreno Bustos, John Morales, Eduardo A. Orduña

Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.