Machine Learning Demonstrator – Energy Systems
A machine learning (ML) Demonstrator will be developed to assist optimising power flow and ensuring resource adequacy across European electricity grids. The ML application aims for enhancing the speed, interactivity and accuracy of energy-system optimization processes, equipping users from the energy industry, civil society and policy makers.
In a nutshell
A machine learning (ML) Demonstrator will be developed to assist optimising power flow and ensuring resource adequacy across European electricity grids. The ML application aims for enhancing the speed, interactivity and accuracy of energy-system optimisation processes, equipping users from the energy industry, civil society and policy makers with tools and methods to ensure a safe and clean supply of energy at any place and at any time in Europe in accordance with the European climate and energy goals.
Technical Overview
Challenge
The European Union’s (EU) renewable energy targets (at least a 42.5% share of renewables in the total energy consumption by 2030), the planned electrification of heat and transportation, and foreseen impacts of climate change and extreme weather events on the energy infrastructure pose substantial challenges to planning, implementing and operating a stable and secured energy system in Europe.
Energy system modelling requires as accurate climate and weather information as possible, to better assess the implications of climate change and its uncertainty on energy grid planning and operation. To bridge this gap, DestinE’s use case Energy Systems (2022-2024, see here and here) provided a proof-of-concept of successful integration of climate and weather information into energy system modelling under a single workflow, while applying machine learning (ML) for predictions of cost-optimal power dispatch in critical situations for the (European) power system while using less resources.
DestinE Solution
Building on the results and the gained knowledge of the use case, DestinE’s ML Demonstrator for Energy Systems aims for validating and improving the integration process of climate and weather information towards developing more accurate and cost-efficient models of energy systems under conditions of uncertainty, to improve systems’ resilience.
The Demonstrator will use meteorological information in a semi-operational power system modeling workflow (following ENTSO-E’s Annual European Adequacy Assessment [ERAA]), replacing the common but complex and less-efficient linear-optimal-power-flow (LOPF) approach. For this purpose, the ML application draws on data from ERAA, time series from DestnE’s Climate Digital Twins (DT), alternative PECD scenarios and other open data sets.
The development of the ML Demonstrator will be defined and tested by a community of climate and energy modellers through dedicated user engagement activities. Key messages and milestones will be communicated on a regular basis throughout the project’s lifetime.
Impact
- Converting climate and environmental data from DestinE’s DT into actionable insights to show how ML can improve decision making processes in the energy sector.
- Demonstrating the transformative potential ML has for energy system management, thus laying the groundwork for future use and services.
- Demonstrating the potential improvement of grid efficiency, reducing potential costs and enhance system stability.
Contributions
Providers
