The use of artificial neural networks (ANN) in solving power system operation and planning problems includes many areas: security assessment, contingency selection, network observability for state estimation, predicting critical clearing times, load forecasting, unit commitment. The interest in ANN is their ability to learn complicated and changing scenarios and thus they are very robust. Our work has been in exploring their use in classifying contingencies and identifying the critical ones.
1. "On the Design of Neural Networks for Classifying Power System Contingencies" R. Fischl, M. Kam, J-C Chow, H. Yan, 10th Power System Computation Conference, Graz, Austria, Aug. 19-24, 1990.
2. "On the Design of Neural Networks for Classifying Power System Contingencies" M. Kam, R. Fischl, J-C Chow, H. Yan, Proc. of 1990 Inter. Circuit and System Symposium, New Orleans, LA, May 1990.
3. "Screening Power System Contingencies using Back Propagation Trained Multi-Perceptrons," M. Kam, J-C Chow, R. Fischl, and S. Ricciardi, Proc. of 1989 IEEE Int. Symp. on Circuits and Systems , Portland, OR, pp. 1835-8, May 1989.
