Short term load estimation with artificial intelligence methods

Abstract

Electric load estimation is very important in efficient operation of power systems and future planning. Load estimation is based on estimating the future electrical load by examining past conditions. Daily or hourly electricity consumption data are generally used for short term load estimation. In this study, Turkey’s four-year, short-term load forecasting modeling was done using daily electricity consumption data. In this modeling, past electrical load values and temperature values are used as input. In order to increase the prediction accuracy, the days were categorized according to their characteristics and classified according to the seasons. MAPE values were calculated in the study. Among the models developed with ANN, the best MAPE value was obtained as 2.51% and the worst MAPE value was 4.48%. When the season criterion is added, a more successful MAPE value was obtained.