The project at a glance
-
Start date:01 Nov 2023
-
Duration in months:24
-
Funding:Luxembourg National Research Fund (FNR)
-
Principal Investigator(s):Gilbert FRIDGENThomas Ernsdorf (external)
About
Electricity consumption forecasting is a crucial aspect of the energy sector. Currently, this is done using standardized load profiles and aggregated data. However, these traditional methods can lead to inaccuracies that can result in significant economic costs for energy suppliers. The use of smart residential meters in Luxembourg can help solve this problem by enabling more accurate forecasts through bottom-up forecasting. These smart meters can measure energy consumption every 15 minutes, allowing for more precise predictions. The DELPHI (Data-driven Electricity Load Prediction for Households and Small Industries) project emerges from a collaboration between the University of Luxembourg and Enovos, Luxembourg’s leading energy supplier. By leveraging advanced artificial intelligence techniques like machine learning and deep learning, along with the 28 million smart meter readings per day from Enovos, the project aims to create better predictions using one of Luxembourg’s High-Performance Computers (HPC). These models will help Enovos reduce prediction errors and make their operations smoother, leading to greater energy supply security and reducing potential economic discrepancies. In addition, the project will help Enovos become more competitive by accelerating its digital and green transformation process. The project is funded by the FNR-HPC bridges call and the Ministry of Economy (MECO).
Organisation and Partners
- Digital Financial Services and Cross-Organisational Digital Transformations Research Group (FINATRAX)
- Interdisciplinary Centre for Security, Reliability and Trust (SnT)
- Enovos
- Encevo
Project team
-
Gilbert FRIDGEN
-
Thomas Ernsdorf
Enovos
-
Joaquin DELGADO FERNANDEZ
-
Sergio POTENCIANO MENCI
-
Quoc Viet NGUYEN
-
Silvana Belegu
Student assistant
-
Hans-Peter Schmitz
Enovos
-
Jürgen Konter
Enovos
-
Holger Baten
Enovos
-
Paul Schmitz
Enovos
-
Thomas Kessels
Enovos
-
Alessio Magitteri
Enovos
-
Kim Schauss
Enovos
-
Alessio Magitteri
Enovos
-
Benoit Fautsch
Encevo
Keywords
- Artificial Intelligence
- Load forecasting
- Large-scale models
- Digitalisation