Integrated Digital Solution for Real-Time Control of Renewable Energy Systems

Digital Twin for Precise System Analysis

A crucial first step is to create a Digital Twin of all system components. This digital image:

• simulates the dynamic behavior of the overall system
• forecasts the availability of renewable energies and energy demand
• monitors the technical condition of individual systems
• recognizes changes in the grid structure, for example after faults

The underlying concept (see illustration) is general and can be applied flexibly to electricity and heating networks.

Use of Machine Learning to Optimize Energy Management

In order to use the models efficiently during operation, we use methods for model order reduction and Machine Learning. This results in generic, online-capable models for grid behavior as well as for individual components such as generators, lines or loads. These models also form the basis for distributed predictive control (MPC), supplemented by forecasts for renewable energies and consumption loads.

Once the Digital Twin has been validated, the second step is to develop the digital core of the project: a hierarchical, distributed, predictive model controller (HDMPC). It is equipped with key features: plug-and-play capability for simple integration and scalability, self-healing to respond to faults and intelligent degradation management for sustainable system control.

This enables scalable digital solutions that can easily integrate existing and future networks. The entire energy system can thus be viewed as a network of energy cells in which the connections encompass both physical and digital areas.

Self-Healing Systems and Degradation Management for Sustainable Energy Systems

The self-healing function enables the system to predict possible faults, recognize changes in the physical topology and map this knowledge digitally. A self-healing controller can adapt its prediction model to reflect physical events and adapt to new, undesirable situations to ensure reliable operation of the energy system.

Predictive Maintenance: Efficient Maintenance Through Intelligent System Analysis

Another building block is Predictive Maintenance. It is based on condition data and supplements this with information on the ageing condition (degradation) of individual components. The system not only plans maintenance work more efficiently, but also takes the service life of the systems into account. »INTEREST« thus introduces an innovative concept of degradation management that integrates operation and maintenance more closely and increases the reliability of the overall system. The system architecture developed can also be transferred to gas grids and thus offers potential for cross-sector applications.

Goals of the Interest Project: Intelligent Control for Safe and Sustainable Energy Systems.

In the project, we are jointly pursuing the goal of developing a distributed, real-time-capable control system for energy systems with renewable sources and demonstrating it in practice. The focus is on the MPC framework, which enables intelligent control and monitoring.

To achieve this, the project has six specific development goals:

• Development of a distributed real-time MPC system for the intelligent control of renewable energy systems
• Integration of fault detection and Predictive Maintenance to stabilize grid operation and avoid unplanned outages
• Development of predictive control strategies for stable, reliable management of cross-sector grids
• Improving sector coupling by simultaneously integrating electricity, heating/cooling and hydrogen to increase efficiency and flexibility
• Creating an open, vendor-neutral data and communication framework for scalable, interoperable energy systems
• Promoting the international application of the solutions – as support for political decision-making processes and user groups in the EU and worldwide

Our Technological Focus at a Glance

• Digital Twins: creating dynamic models of system components to predict energy demand and generation
• Machine Learning: use of algorithms for pattern recognition and forecasting
• Blockchain Technology: secure and transparent transactions between participants
• IoT Systems (Internet of Things): Reliable status data through networked sensor technology – IoT systems are networked devices and sensors that provide real-time data on energy consumption, grid load or system failures – creating the basis for intelligent control of the energy system
• Visualization Tools: Display of load forecasts, CO₂ balances and maintenance planning

The result is a platform that not only combines technical excellence, but also accelerates the transition to networked, decentralized and sustainable energy systems.

Our Role as Fraunhofer ITWM: Developing a Digital Twin to Optimize Energy Systems

We at Fraunhofer ITWM are contributing our expertise in the control and monitoring of energy systems to develop a Digital Twin for advanced control solutions.

We provide models for power grid components that form the foundation for the Digital Twin. We also extend this twin with mechanisms that enable both fault detection and status prediction. The Digital Twin supports the development of a flexible MPC framework, leading to self-healing control strategies that significantly improve the reliability and efficiency of energy systems.

Our Project Partners

The project brings together an international consortium from research, industry and energy suppliers:

• SINTEF, Koordinator 
• Es geht!
• Fraunhofer Chalmers 
• OFFSET Energy
• SWW Wunsiedel GmbH 
• Universidad de Sevilla 
• Universidade Federal de Santa Catarina 
• University of Technology Sydney

Project Funding and Duration

The European Union finances and supports the project with funds from the European Union's Horizon Europe research and innovation program under grant agreement no. 101160594.

The project started in September 2024 and runs until August 2027.