Automatic Contingency Grouping Using Partial Least Squares and Feed Forward Neural Network Technologies Applied 10 the Static Security Assessment Problem

The paper shows how a number of Feed Forward Back Propagation Neural Networks can be trained to predict power system bus voltages after a contingency. The approach is designed to use very few learning examples thus being suitable for on-line use. The method was applied to the 10-machine 39-bus New England Power System model.