Electric Power System Anomaly Detection Using Neural Networks
The aim of this work is to propose an approach to monitor and protect Electric Power System by learning normal system behaviour at substations level, and raising an alarm signal when an abnormal status is detected; the problem is addressed by the use of autoassociative neural networks, reading substation measures. Experimental results show that, through the proposed approach, neural networks can be used to learn parameters underlaying system behaviour, and their output processed to detecting anomalies due to hijacking of measures, changes in the power network topology (i.e. transmission lines breaking) and unexpected power demand trend.
KeywordsRoot Mean Square Error Transmission Line Measure Vector Electric Power System Load Demand
Unable to display preview. Download preview PDF.
- 1.Thompson, B., Marks, R., Choi, J., El-Sharkawi, M.A., Huang, M., Bunje, C.: Implicit Learning in Autoencoder Novelty Assessment. In: International Joint Conference on Neural Networks, 2002 IEEE World Congress on Computational Intelligence, May 12-17 (2002)Google Scholar
- 2.Haykin, S.: Neural Networks: A comprehensive Fundation, 2nd edn. Prentice-Hall, Englewood Cliffs (1998)Google Scholar
- 3.Markou, M., Singh, S.: Novelty Detection: A Review - part 2: Neural network based approaches. Signal Processing 83 (2003)Google Scholar
- 5.The Safeguard Project website, http://www.ist-safeguard.org
- 6.: Reliability Test System Task Force of the application of probability methods subcommittee, The IEEE Reliability Test System - 1996. IEEE Transaction on Power Systems 14(3) ( August 1999)Google Scholar