Metodologia de controle preventivo baseado em árvore de decisão para a melhoria da segurança estática e dinâmica de sistemas elétricos de potência

Abstract

This work aims to present a set of computational tools to support the real time operation and preventive control to enhance the static and dynamic security of power systems. The data mining technique known as decision tree was utilized to determine power system operating state as well as to provide operating guidelines for the preventing control actions necessary to avoid continuous decline of bus voltages and transient instability problems. Preliminary tests were carried out using operation historical data collected by SCADA/SAGE host located at Eletrobrás Eletronorte's regional control center. The obtained results validated the set of computational tools and also demonstrated the prospective application in real time operating environment.