Decision Trees Applied to Spanish Power Systems Applications

Decision trees are useful data mining tools designed to face classification problems, dividing the output variable space by the evaluation of partition rules developed from explanatory variables. The output variable is classified by the explanatory variables rules that give the maximum quality to the decision tree. This paper provides a comprehensive review of the different forecasting techniques available pointing their advantages and disadvantages. In addition, it describes three different applications developed in the Spanish electricity market context faced with success employing decision trees, pointing the great versatility of decision trees and their adequacy to be applied to very diverse different real applications. Firstly, decision trees have been used to predict stochastic residual demand curves in the Spanish electricity market. Secondly, a methodology based on decision trees has been developed to estimate the daily load pattern of units, which have not been cleared in the daily energy market and can be connected to alleviate voltage constraints. Finally, decisions trees have also been applied to predict the values of reactors and capacitors of the Spanish power system in a short-term time scope (24 and 36 hours ahead)