Machine learning for electricity consumption forecasting in energy communities

Details

Supervisors

Legay, Axel

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] en sciences informatiques, à finalité spécialisée

Abstract

The climatic consequences linked to the exploitation of fossil fuels invite us to rethink our mode of energy production and turn to renewable energies. An energy community is a group of people, who collectively manage their production, consumption and eventually their storage of electricity. This type of renewable energy project could develop in the years to come as EU Member States are required to follow the recommendations of the "Clean Energy for All Europeans" package adopted in 2019. WeSmart is a Belgian company based in Brussels that sets up and manages energy communities. Thanks to smart meters, they can collect data from their users and develop monitoring tools to help members of energy communities to consume more efficiently. The data collected also enables them to develop a user optimisation and recommendation service. The idea is to provide community members with personalised advice on how to maximise their consumption of community-generated electricity. To achieve this, it may be useful to be able to forecast the community’s electricity production and each member’s consumption in the near future.