Research Article
On the Machine Learning Models to Predict Town-Scale Energy Consumption in Burkina Faso
@INPROCEEDINGS{10.1007/978-3-031-35883-8_5, author={Baowendsomme Armel Yameogo and Tounwendyam Fr\^{e}d\^{e}ric Ouedraogo and Constantin Zongo}, title={On the Machine Learning Models to Predict Town-Scale Energy Consumption in Burkina Faso}, proceedings={Emerging Technologies for Developing Countries. 5th EAI International Conference, AFRICATEK 2022, Bloemfontein, South Africa, December 5-7, 2022, Proceedings}, proceedings_a={AFRICATEK}, year={2023}, month={7}, keywords={Predictive model Machine learning Energy consumption}, doi={10.1007/978-3-031-35883-8_5} }- Baowendsomme Armel Yameogo
Tounwendyam Frédéric Ouedraogo
Constantin Zongo
Year: 2023
On the Machine Learning Models to Predict Town-Scale Energy Consumption in Burkina Faso
AFRICATEK
Springer
DOI: 10.1007/978-3-031-35883-8_5
Abstract
The lack of forecast of the electricity consumption in Burkina Faso is a great concern. This leads to selective power cut due to insufficient production. This situation leads to selective power cut, which has a negative impact on economic activities. In this paper, we present a model to forecast household energy consumption in the Middle West region of Burkina Faso. We propose three predictive models of energy consumption that we evaluate to select the best one in our context. The proposed models are the Decision Tree Regressor, the Random Forest Regressor, and the Neural Network. We assessed them through three measures, namely the Root Mean Squared Error, the Mean Absolute Error and the R2-Score. We found that the neural networks has more satisfactory performance. The type of meter, the power of the meter, and the month of the year are factors that influence household energy consumption. Our contribution is an important step in the planning of energy production to meet the growing consumption of households.
