LARGE-SCALE DYNAMIC SECURITY SCREENING AND RANKING USING NEURAL NETWORKS

This paper reports on the findings of a recently completed Canadian El ectric Association (CEA) funded project [1] exploring the application of neural network to dynamic security contingency screening and rankin g. The idea is to use the information on the prevailing operating cond ition and directly provide contingency screening and ranking using a t rained neural network. To train the two neural networks for the large scale systems of B.C. Hydro and Hydro Quebec, in total 1691 derailed t ransient stability simulation were conducted, 1158 for B.C. Hydro syst em and 533 for the Hydro Quebec system. The simulation program was equ ipped with the Energy Margin Calculation Module (Second Kick) [4] to m easure the energy margin in each run. The first set of results showed poor performance for the neural networks in assessing the dynamic secu rity. However a number of corrective measures improved the results sig nificantly. These corrective measures included : a) the effectiveness of output, b) the number of outputs, c) the type of features (static v ersus dynamic), d) the number of features, e) system partitioning and f) the ratio of training samples to features. The final results obtain ed using the large scale systems of B.C. Hydro and Hydro Quebec demons trates a good potential for neural network in dynamic security assessm ent contingency screening and ranking.