Short term load forecasting for power exchange between Brasil and Paraguay

Abstract

This work presents a case study of short term load forecasting to assist in the power exchange real time dispatch operation between Brazil and Paraguay at Itaipu Dam. A classical method with statistical approach, Seasonal Autoregressive Moving Average, is compared with an artificial intelligence method based on Artificial Neural Networks. The methods are tested using a time series representing the average hourly power exchange. The results were compared with the current forecast methods used to define the daily program of operation of Itaipu using the Mean Absolute Percentage Error method. The results of the analysis showed that the model based on the Seasonal Autoregressive Moving Average present a lower error index among the methods tested.