Revue d'économie et de statistique appliquée
Volume 22, Numéro 2, Pages 52-67
2025-12-09

Hourly Electricity Consumption Forecasting Using Advanced Machine Learning: A Case Study Of Algeria



Authors : Kouira Aimene .

Abstract

This study aims to conduct a comprehensive investigation into the forecasting of hourly electricity demand in Algeria through the implementation of a hybrid modeling framework that integrates the Autoregressive Integrated Moving Average (ARIMA) model with a Long Short-Term Memory (LSTM) neural network. Because industrial growth, weather variability, and human activity all have a significant impact on energy consumption patterns, Algeria presents a particularly difficult situation for projecting power demand. To address these complexities, the proposed hybrid ARIMA–LSTM model integrates the linear forecasting strengths of ARIMA with the nonlinear learning capabilities of LSTM, enabling it to capture both seasonal patterns and intricate temporal dependencies in the data. The study utilizes hourly electricity consumption data from January 2022 to December 2022, provided by the Regulatory Commission of Electricity and Gas (CREG). The dataset includes total national demand—encompassing industrial, commercial, and residential sectors—since sector-disaggregated data are not available. Ninety percent of the data were used for model training and ten percent for validation and testing. Performance evaluation, based on metrics such as Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE), reveals that the hybrid model consistently outperforms standalone ARIMA and LSTM approaches. The results highlight the potential of hybrid forecasting techniques to improve the accuracy and reliability of short-term load predictions, providing valuable insights for energy management, grid optimization, and sustainable power planning in Algeria.

Keywords

AI, Smart energy, Machine learning, Deep learning, ARIMA, LSTM, Hybrid Models, Electricity Forecasting, Algeria.

Les articles similaires

Electricity Consumption Forecasting In Algeria Using Arima And Long Short-term Memory Neural Network

Sahed Abdelkader .  Kahoui Hacene . 
pages 78-88.

Forecasting Electricity Consumption In Algeria Using Artificial Neural Networks

Chekouri Sidi Mohammed .  Sahed Abdelkader . 
pages 247-261.

International Digital Competitiveness And Economic Growth In Emerging Economies: A Comparative Empirical Study Using Advanced Machine Learning And Modern Panel Data Techniques

Hadjadj Mustapha .  Melouah Mohammed . 
pages 332-346.

Random Forest Modeling For Forecasting Economic Growth: A Machine Learning Approach In The Case Of Algeria (1964-2022)

Messaoudi Malika . 
pages 391-410.

Using Machine Learning Algorithms In Home Energy Management Systems For Energy Cost Forecasting

Abderraouf Bouakkaz .  Adel Lahsasna . 
pages 205-215.

Formats de citation