Reasoning on Critical Clearing Time with the Rules Extracted from a Multilayer Perceptron Artificial Neural Network

In the past two decades, artificial neural networks (ANN) have been employed to determine the critical clearing time (CCT). This issue mainly concerns the prediction rather than the explanation. ANNs were considered to be black boxes, unlike decision trees which rules could be deduced from. This paper describes the procedures for reasoning CCT by means of rules extracted from a multilayer perceptron (MLP) artificial neural network. Those extracted rules are presented in ´IF-THEN´ statement. Case studies of two power systems shall justify the proposal.