Abstract
In this paper, we propose models to forecast the demand of energy to allow Senelec to be able to maintain the balance between supply and demand of electricity in Senegal. We first analyze the consumption behavior of the different categories of customers accordingto Senelec’s 2020 billing database. After we present two prediction models of the energy demand of Senelec’s MediumVoltage customers: a SARIMA model based on classical statistical modeling of time series and an LSTM neural network model based on artificial intelligence. The out-of-sample predictive performance of the LSTM model is globally better than that of the SARIMA model according to the indicators (MAE and RMSE (RMSE: Roote Mean Squart Error)) we used to compare them. For the LSTM model, we have a Mean Absolute Error (MAE (MAE: Mean Absolute Error)) of 0.148597 on the test data while, for the SARIMA model, we have a Mean Absolute Error (MAE) of 0.166997. These two models have been integrated into a web application to offer a simple and user-friendly interface for the prediction of energy demand in Senegal.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
SMES, Senegal Energy Management Strategy (SMES) (2021)
Haida, T., Muto, S.: Regression based peak load forecasting using a transformation technical. IEEE Trans. Power Syst. 9, 1788–1794 (1994)
Hobbs, B., Jitprapaikulsarn, S., Konda, S., Chankong, V., Loparo, K., Maratukulam, D.: Analysis of the value for unit commitment of improved load forecasts. IEEE Trans. Power Syst. 14, 1342–1348 (1999)
Soares, L.J.: Medeiros, MC modeling and forecasting short-term electricity load : a comparison of methods with an application to Brazilian data. Int. J. Forecast. 24, 630–644 (2008)
Bellahsen, A.: Artificial intelligence for the optimization of electrical energy in a smart grid. PhD Thesis, Polytechnique Montreal (2020)
Talbi, R.: Le preceptron multicouches approximateur universel. Ph.D. thesis, Alger (2005)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
Werbos, P.J.: Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. Doctoral Dissertation, Harvard University, Cambridge (1974)
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back-propagating errors. Nature 323, 533–536 (1986)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Drame, M., Seck, D.A.N., Ndiaye, B.S. (2023). Analysis and Forecast of Energy Demand in Senegal with a SARIMA Model and an LSTM Neural Network. In: Younas, M., Awan, I., Benbernou, S., Petcu, D. (eds) The 4th Joint International Conference on Deep Learning, Big Data and Blockchain (DBB 2023). Deep-BDB 2023. Lecture Notes in Networks and Systems, vol 768. Springer, Cham. https://doi.org/10.1007/978-3-031-42317-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-031-42317-8_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-42316-1
Online ISBN: 978-3-031-42317-8
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)