Decision Trees-Aided Self-Organized Maps for Corrective Dynamic Security

Difficulties in expanding the generation and transmission system force modern power systems to operate often close to their stability limits, in order to meet the continuously growing demand. An effective way to face power system contingencies that can lead to instability is load shedding. This paper proposes a machine learning framework for the evaluation of load shedding for corrective dynamic security of the system. The proposed method employs a self-organized map with decision trees nested in some of its nodes in order to classify the load profiles of a power system. The method is applied on a realistic model of the Hellenic Power System and its added value is shown by comparing results with the ones obtained from the application of simple self-organized maps and simple decision trees.