In 2018, Alexela received a grant from Archimedes Foundation for applied research in smart specialisation growth area with an intention to carry out, in cooperation with TalTech, a research on electricity consumption of its customers.
Alexela sells electricity to private and business customers since 2014 and over the years the company has grown into one of the largest electricity sellers in Estonia – Alexela customers consume over 0,5 TWh electricity per year.
Electricity sellers need to predict as accurately as possible hourly consumption of all their customers to ensure minimum purchase cost of electricity. The aim of the project was to find, in cooperation with TalTech, the best electricity consumption forecasting model. Different variables (temperature, cloudiness, seasonal factors, etc) as well as various prediction models, based on machine learning were tested. The goal was to find a methodology that would provide the lowest hourly forecast error.
As a result, a report was prepared that highlighted the best methods for achieving this result as outlined in academic literature, as well as the best model found by the TalTech research team and possible future improvements.
In order to carry out the project, Alexela received a maximum grant of 28,000 euros from Archimedes Foundation.