Load forecasting using fuzzy logic, artificial neural network, and adaptive neuro-fuzzy inference system approaches: application to South-Western Morocco

Abstract

The demand for energy on a global scale is continuously rising due to the expansion of energy infrastructure and the increasing number of new appliances. To address this growing need, an efficient energy management system (EMS) has become indispensable. By implementing EMS, both residential and commercial buildings can significantly improve their energy efficiency and consumption. One crucial aspect of enabling EMS to operate efficiently is load forecasting. The accuracy of load forecasting depends on numerous factors. A reliable load forecast model should consider the region’s weather forecast, as it plays a crucial role in developing an accurate prediction. This study is about the medium-term load forecasting (MTLF) for the Province of Taroudant, Morocco, using historical monthly load and weather data for five years (2018 to 2022). To forecast consumed energy three methods are used namely artificial neural network (ANN), fuzzy logic (FL) and adaptive neuro-fuzzy inference system (ANFIS). This paper selects absolute percentage error (APE), mean absolute percentage error (MAPE), correlation coefficient (R) and root mean square error (RMSE) to compare and evaluate the prediction accuracy of models. It has been observed through results analysis that the ANFIS model produces very accurate forecasting prediction with MAPE of 4.75% while ANN and FL models give respectively MAPE of 7.36% and 8.42%.