ABSTRACT
Global energy consumption is growing due to multiple reasons, such as the COVID-19 pandemic. In order to improve the efficiency of energy consumption and thus contribute to the protection of the environment, governments are implementing new energy efficiency policies. Prediction of energy consumption is one of the most important objectives in this regard. Forecasting algorithms based on machine learning approaches have proven to be a robust solution to provide predictions based on energy consumption data. In this paper, we present a comparative study of different forecasting approaches on an energy consumption dataset collected from a Paraguayan electricity distribution provider. In the analysis, historical windows, W, of {7, 14, 28, 84} days and a prediction horizon, h of one day were used. Models were evaluated using the coefficient of determination (R-2), the mean absolute error(MAE), the root mean squared error (RMSE), and the mean absolute percentage error (MAPE). The results achieved show that, among the techniques studied, Artificial Neural Networks are the best strategy to capture the complexity of the data. Furthermore, the performance of linear regression is outstanding, taking into account its simplicity.