Abstract:
COVID-19 has been an unprecedented situation that disrupted the stability of many socioeconomic
dynamics all over the world and in particular on the energy sector in West-Africa.
This study seeks to provide technical based analysis to inform policy decision-making and
experts in the sector on how the pandemic impacted the countries in the region and actions
to reduce vulnerability. This study analyses the impact of COVID-19 on Electricity Supply-
Demand in Benin, Togo, Côte d’Ivoire, Senegal and Niger using both quantitative and
qualitative data. The latter is based on semi-structured interviews targeting power utilities in
these countries. A comparative assessment is conducted between the observed consumption
and forecasted in 2020. Advanced forecasting methods with machine learning algorithms
are explored including ARIMA, Prophet, ETS, TBATS, NNAR, GLMNET, Random Forest
and hybrid ones which are regressed with climatic factors (temperature, humidity and
solar radiation) and calendar effect (working days). The best models after the performance
evaluation are the NNAR and GLMNET which show good measure compared to others.
The assessment shows globally that despite the pandemic the demand has risen above
forecast averaging 3.28%. Three distinct periods can be discerned from the time series:
a pre-COVID where the demand rose in all countries, and slowed down as the pandemic
intensified (in COVID period) and the post COVID period where the consumption rose
up back as a result of the release of restriction measures (economic recovery). From one
country to another, the recovery time can be longer or shorter.
Description:
A Publication submitted to the West African Science Service Centre on Climate Change and Adapted Land Use and the Université Abdou Moumouni, Niger in partial fulfillment of the requirements for the degree of Master of Science Degree in Climate Change and Energy
