R&D Nester participated in IEEE ODS competition on building energy consumption forecasting, with its AI.Forecasting tool, reaching 3^rd place among 42 participants.

This competition is organized by the IEEE Power & Energy Society Technical Committee on Analytic Methods for Power Systems (IEEE PES AMPS) / Intelligent Systems Subcommittee (ISS). The main objective of this IEEE competition is bringing together the most recent advances in building energy consumption forecasting methods. 

Energy consumption forecasting is crucial in current and future power and energy systems. With the increasing penetration of renewable energy sources, with high associated uncertainty due to the dependence on natural conditions, such as wind speed or solar intensity, the need to balance the fluctuation of generation with the flexibility from the consumer side increases considerably. In this way, significant work has been done on the development of energy consumption forecasting methods, able to deal with different forecasting circumstances, e.g., the prediction time horizon, the available data, the frequency of data, or even the quality of data measurements. The main conclusion is that different methods are more suitable for different prediction circumstances, and it is not clear that a certain method can outperform all others in all situations.

Hence, R&D Nester developed the AI.Forecasting tool, an architecture of AI and machine learning methods that enables the automatic learning of energy forecasting models specific for each application. The developed ensemble solution is a combination of different ML methods that perform different non-linear transformations to the same data. The individual methods composing the ensemble are Support vector machines, Extreme gradient boosting and Deep learning using long short-term memory networks. This tool was developed under the scope of the GIFT project, an innovation project funded by the European Commission that aims at decarbonizing the European islands through the application of multiple innovative solutions, such as a virtual power system, energy management systems for harbors, factories and homes, better prediction of supply and demand, visualisation through a GIS platform, as well as innovative storage systems, allowing synergy between electrical, heating and transportation networks.

The competition was held during a full business week (14-18 June, 2021). Each day of this week participants were asked to provide their consumption forecasts for the following day.

Immediately after the deadline for submission, the real data from the respective day were provided, so that it could be used to generate the next output.

In order to build and refine the forecasting models, a full year of historical data was provided by mid-April.

Two weeks before the competition week, the following 40 days of historical data was provided - referring to the period immediately before the days to be forecasted.

Finally, during the competition week, the data that refers to the last day was provided on a daily basis.

Final results of this competition were revealed on July 27^th, during the IEEE Power & Energy Society General Meeting that was held in Washington, DC, were Ângelo Casaleiro, represented R&D Nester in this competition, presenting this AI.Forecasting Tool.

R&D Nester's final presentation was very well received, generating interest, several comments and questions from the audience and other participants. This generated a very interesting final debate on the methodology of this tool developed by R&D Nester under the scope of the GIFT project, an innovation project funded by the European Commission that aims at decarbonizing the European islands through the application of multiple innovative solutions, such as a virtual power system, energy management systems for harbors, factories and homes, better prediction of supply and demand, visualization through a GIS platform, as well as innovative storage systems, allowing synergy between electrical, heating and transportation networks.

Improved forecasting is key to contribute to an efficient and reliable energy system in the energy transition process.

R&D Nester - Creating a Smart Energy Future!

For more information:

IEEE ODS Competition website

GIFT Project

2021 IEEE PES GM Conference website