论文信息 - NEURAL NETWORK APPLICATIONS TO REAL-TIME AND OFF-LINE FAULT ANALYSIS

NEURAL NETWORK APPLICATIONS TO REAL-TIME AND OFF-LINE FAULT ANALYSIS

This paper is concerned with application of Neural Networks (NNs) to fault analysis for both the realtime applications such as protective relaying of transmission lines and the off-line applications such as postmortem study of fault events recorded with Digital Fault Recorders (DFRs). A supervised learning NN of the same type is utilized for both applications. It has been demonstrated that the NN approach reaches performance of the existing techniques in both application areas and yet shows some additional benefits.

Mladen Kezunovic | I. Rikalo | D. J. Sobajic

[1] Dejan J. Sobajic,et al. Artificial Neural-Net Based Dynamic Security Assessment for Electric Power Systems , 1989, IEEE Power Engineering Review.

[2] D. J. Sobajic,et al. On-line monitoring and diagnosis of powersystem operating conditions using artificial neural networks , 1989, IEEE International Symposium on Circuits and Systems,.

[3] R. D. Christie,et al. High impedance fault detection in low voltage networks , 1993 .

[4] G. Swift,et al. Detection of high impedance arcing faults using a multi-layer perceptron , 1992 .

[5] Mladen Kezunovic,et al. An expert system for transmission substation event analysis , 1993 .

[6] Oscar L. Chacón,et al. Neural network approach to fault detection in electric power systems , 1996, Proceedings of International Conference on Neural Networks (ICNN'96).