Data Mining Classifiers for Static Security Evaluation in Power System

This paper addresses the application of data mining approach on Static Security Evaluation (SSE) of deregulated power system. The process of building binary class classifiers is divided into two components: (i) comparison the methods, and (ii) selection of the best classifier. Preliminary results of using eleven algorithms of Decision Tree’s classifiers (DTC) for SSA are presented. A comprehensive comparison of the proposed classifiers for the purpose of SSA classification is discussed. A set of training cases generated on the IEEE 30 and 300-bus system were used to train and test the classifiers that discriminates the system security. The results show that DT’s classifiers are capable of system security classification. Finally, empirical results indicate that C4.5 tree can be used to design a SSAC that is lightweight, efficient and effective for real time