Demand-Supply Forecasting based on Deep Learning for Electricity Balance in Cameroon

Authors

  • Wulfran Fendzi Mbasso Laboratory of Technology and Applied Sciences, University Institute of Technology of Douala, University of Douala, Douala, Cameroon.
  • Reagan Jean Jacques Molu Laboratory of Technology and Applied Sciences, University Institute of Technology of Douala, University of Douala, Douala, Cameroon.
  • Serge Raoul Dzonde Naoussi Laboratory of Technology and Applied Sciences, University Institute of Technology of Douala, University of Douala, Douala, Cameroon.
  • Saatong Kenfack Laboratory of Technology and Applied Sciences, University Institute of Technology of Douala, University of Douala, Douala, Cameroon.

DOI:

https://doi.org/10.32479/ijeep.13092

Abstract

Electricity is becoming an important commodity in Cameroon. Within the years, its consumption and production have led to many studies. Hence, having an idea on its progression is one of research’ concerns. Thus, this paper aims to develop a model for forecasting electricity production and consumption in Cameroon based on Long Short-Term Memory (LSTM). Indeed, the LSTM approach, showing a good ability to grab the long-term dependencies between time steps of electricity production and consumption, allows a good prediction in 2030 of 7178GWh for consumption with 0.067 RMSE and 0.2965% MAPE and 8686GWh for production with 0.1631 RMSE and 0.4291%MAPE. Hence, the proposed model is more reliable, what makes possible to monitor the growth in electricity supply and demand, falling to the study of balance in Cameroon.

Keywords:

Forecasting, Long Short-term Memory, Electricity Production and Consumption

Downloads

Download data is not yet available.

Downloads

Published

2022-07-19

How to Cite

Mbasso, W. F., Molu, R. J. J., Naoussi, S. R. D., & Kenfack, S. (2022). Demand-Supply Forecasting based on Deep Learning for Electricity Balance in Cameroon. International Journal of Energy Economics and Policy, 12(4), 99–103. https://doi.org/10.32479/ijeep.13092

Issue

Section

Articles