ADABEL
Research Goal and objectives of the ADABEL project:
The ADABEL project lies at the frontier between Electrical Power Systems and Machine Learning, with the aim to support the Belgian Energy Transition throughout the development of advanced data analytics. Descriptive and/or predictive models are thus designed to extract and exploit at best the information contained in real-life power systems and market data in order to improve decision-making models for the Belgian electricity system.
Such data-driven decision-making models are then applied on three critical challenges within the Belgian energy transition:
-
Adequacy, where the research objective is to improve the models of cross-border exchanges within adequacy studies.
-
Balancing, where the research objective is to improve the cost-effectiveness of procedures for sizing, procuring and activating balancing reserves.
-
Voltage-related Security of Supply, where the research objective is to better represent the TSO/DSO interfaces within reactive power investment planning.
ADABEL
Advanced Data Analytics ensuring cost-efficient security of supply and balancing throughout the BELgian energy transition
Consortium, duration and funding:
The ADABEL project is a joint project between the University of Mons (Power Systems & Markets Research Group) and the Katholieke Universiteit Leuven (Energy Systems Integration & Modeling Research Group), and has been running from 1 February 2020 till 31 January 2024.
ADABEL has been funded by the Energy Transition Fund of the Belgian Federal government, organized by the FPS economy, S.M.E.s, Self-employed and Energy.
Context:
Our energy system is undergoing a major transition towards a more sustainable and less carbon-intensive system, characterized by an increased penetration of renewable energy sources, higher energy efficiency and reduced emissions of greenhouse gasses. In Belgium, the nuclear phase-out, which currently represents 40% of the installed capacity, fosters this integration of decentralized, intermittent generation (e.g., wind and solar power).
In this context, ensuring a secure, reliable and cost-efficient management of our electricity network is significantly more complex than in the past. The success of new management strategies, hence, of the energy transition, strongly relies on the knowledge of the system state, which must be enabled through a better observability of the transmission grids and generation capacity. There is therefore an increased interest in collecting, maintaining, and sharing data of the electricity system, not only in terms of energy, but also for relevant weather and market data. Massive datasets are in that way appearing in a community not used to deal with such an amount of heterogeneous data with a high temporal and geographical resolution, leading to a considerable need for new tools and expertise to fully leverage the underlying information.
Research Goal and objectives of the ADABEL project:
ADABEL organization:
The ADABEL project lies at the frontier between Electrical Power Systems and Machine Learning, with the aim to support the Belgian Energy Transition throughout the development of advanced data analytics. Descriptive and/or predictive models are thus designed to extract and exploit at best the information contained in real-life power systems and market data in order to improve decision-making models for the Belgian electricity system.
Such data-driven decision-making models are then applied on three critical challenges within the Belgian energy transition:
-
Adequacy, where the research objective is to improve the models of cross-border exchanges within adequacy studies.
-
Balancing, where the research objective is to improve the cost-effectiveness of procedures for sizing, procuring and activating balancing reserves.
-
Voltage-related Security of Supply, where the research objective is to better represent the TSO/DSO interfaces within reactive power investment planning.
As depicted in Fig. 1, the structure of ADABEL is divided into 4 WPs:
-
WP1 is a transversal WP, which supplies the oriented-application WPs 2-4 with power system and market data. Hence, contributions in energy forecasting, mainly based on innovations in Machine Learning and adapted to the context of electrical power systems, were investigated in order to exploit at best the information contained in the databases.
-
WP2 suggests gradual improvements of the representation of cross-border exchanges of the TSO methodology for adequacy studies.
-
WP3 proposes novel methodologies for data-driven probabilistic forecasts of system imbalances, better informing the procurement and activation of balancing reserves.
-
WP4 aims at understanding and predicting the reactive power behavior in transmission grids, in order to efficiently quantify the needs in reactive power control.
ADABEL organization:
As depicted in Fig. 1, the structure of ADABEL is divided into 4 WPs:
-
WP1 is a transversal WP, which supplies the oriented-application WPs 2-4 with power system and market data. Hence, contributions in energy forecasting, mainly based on innovations in Machine Learning and adapted to the context of electrical power systems, were investigated in order to exploit at best the information contained in the databases.
-
WP2 suggests gradual improvements of the representation of cross-border exchanges of the TSO methodology for adequacy studies.
-
WP3 proposes novel methodologies for data-driven probabilistic forecasts of system imbalances, better informing the procurement and activation of balancing reserves.
-
WP4 aims at understanding and predicting the reactive power behavior in transmission grids, in order to efficiently quantify the needs in reactive power control.
