March 17th-20th – Trondheim, Norway (SSCI 2025)
Organized by Polytechnic of Porto (ISEP)
Polytechnic of Porto (ISEP): Luis Gomes, Zita Vale, Pedro Faria, João Soares
Registration
The registration of participants is mandatory and must be made online using the following form: link
Competition Outline
Forecasting of electric energy consumption can be a very difficult tasks when handling building-level data. However, an accurate forecast is needed to boost the potential of energy management systems. The need to forecast energy consumption grows as our reliance on renewable energy sources, such as solar and wind power, grows. This means that to meet consumer demand with renewable energy generation, energy management systems must operate based on accurate energy forecasting models for both short and long-term periods. Energy consumption forecasting techniques that can manage a variety of scenarios, including varying prediction timeframes, accessible data, data frequency, and even data quality, have been the subject of intense research. There is no one-size-fits-all approach, where certain situations call for different approaches. The goal of this competition is to compile and evaluate the most recent advances in energy consumption forecasting techniques.
Competition goals
Obtain accurate electric energy consumption forecast for a smart building based on historic data regarding overall and area electric energy consumption, weather information, external temperature, rooftop photovoltaic panels generation, other sensors data (e.g. movement, CO2). The competition targets both forecasting methods and data pre-processing methods aiming at providing a good environment for discussion and comparison of results of the different approaches used by the applicants.
Submissions
Each submission must provide an hourly 24-hour forecast of energy consumption, meaning that each submission is required to have 24 values. In total, the participants must submit results for 5 days.
The submission will be made by email until 22:59 (GMT) of each competition date. The email must be sent to: log@isep.ipp.pt, zav@isep.ipp.pt, pnf@isep.ipp.pt, and jan@isep.ipp.pt.
In your submissions, please add the subject as: Application for the 2025 Competition on Electric Energy Consumption Forecast
In each submission, the participants must attach the following:
- the results obtained for that day;
- the execution time;
- (optional) execution energy consumption;
- the code used to generate the results (equal in all submission days);
- a simple readme to describe how the code can be executed (equal in all submission days);
- the description of the forecast model used (equal in all submission days);
- declaration of authorship (equal in all submission days).
Competition timeline
Before competition:
- 01/12/2024: release of the v1.0 of the dataset, including one year of data from a smart building with readings taken every 5 minutes (https://zenodo.org/records/14275645);
- 30/12/2024: release of the v2.0 of the dataset, including 40 days of data from a smart building with readings taken every 5 minutes;
During competition:
- 05/01/2025, until 01:00 (GMT): release by email of the forecasted weather temperature for the day (06/01/2025) with 1-hour readings;
- 06/01/2025, until 22:59 (GMT): submission of forecasted data for the first day;
- 07/01/2025, until 01:00 (GMT): release by email of the forecasted weather temperature for the day (07/01/2025) with 1-hour readings, and the real consumption of the previous day (06/01/2025);
- 07/01/2025, until 22:59 (GMT): submission of forecasted data for the first day;
- 08/01/2025, until 01:00 (GMT): release by email of the forecasted weather temperature for the day (08/01/2025) with 1-hour readings, and the real consumption of the previous day (07/01/2025);
- 08/01/2025, until 22:59 (GMT): submission of forecasted data for the first day;
- 09/01/2025, until 01:00 (GMT): release by email of the forecasted weather temperature for the day (09/01/2025) with 1-hour readings, and the real consumption of the previous day (07/01/2025);
- 09/01/2025, until 22:59 (GMT): submission of forecasted data for the first day;
- 10/01/2025, until 01:00 (GMT): release by email of the forecasted weather temperature for the day (10/01/2025) with 1-hour readings, and the real consumption of the previous day (09/01/2025);
- 10/01/2025, until 22:59 (GMT): submission of forecasted data for the first day;
After competition:
- At SSCI conference: winner announcement.
Datasets will be published in Zenodo: https://zenodo.org/records/14275645
On the right side, it is visible the versions available.
Results evaluation
The results will be evaluated according to:
- Root mean square error (RMSE);
- Execution time;
- (optional) Energy consumption.
Organizers
