Generator load profiles estimation using artificial intelligence

Abstract

The security criteria of a power system require that branch power flows and bus voltages are within their limits, not only in normal operating conditions but also when any credible contingency occurs. In the Spanish electricity market, voltage constraints are solved by the system operator by connecting a set of off-line generators located in the areas where they occur. Thus, for a market participant it is necessary to predict approximately when its generating units are connected in order to prepare the annual budget andor decide the time and location of new plants. This paper proposes a methodology to forecast if a non-connected unit will be committed by the system operator in order to remove voltage violations. For that purpose different artificial intelligence techniques are combined: neural networks, decision trees and clustering techniques. The performance of the methodology is illustrated with a study case of a real unit operating in the Spanish market.