Ai-based Techniques To Predict Electricity Consumption In Algeria Using Rnn And Lstm Approaches

Résumé: Artificial Intelligence (AI) has emerged as a game-changer in various industries, and the energy sector is no exception. By leveraging machine learning (ML) and Deep Learning forecasting models, the energy industry is experiencing significant advancements in efficiency, sustainability, and reliability. This article delves into the application of a machine-learning model employing a deep-learning forecasting technique in the realm of energy and its transformative impact on the sector. The forecasting of daily Electricity power consumption improves the quality, reliability, and stability of the power system. This study aims to develop an LSTM technique for daily forecasting of electricity power consumption. Furthermore, RNN is used to daily perform predictions of electricity power consumption, we will use daily data on Algeria's electricity demand that was collected between January 2000 and December 2022. We run a comparative analysis for all these techniques. We found that the Deep learning technique (RNN) has better forecasting accuracy than other developed techniques in terms of better-performing goodness-of-fit metrics.

Publié dans la revue: Revue d'économie et de statistique appliquée