Digital twins in energy systems
Digital twin technology is increasingly used to create digital representations of physical systems that can be used for monitoring, analysis, and optimisation. In the context of energy systems, digital twins provide virtual representations of infrastructure, processes, environmental conditions, and operational activities.
By combining sensor data, digital models, and analytical tools, digital twins allow operators and researchers to observe system behaviour in real time and to analyse how changes in operation or design may affect performance. Digital twins can therefore support optimisation, predictive maintenance, and improved operational decision making.
However, creating effective digital twins requires more than a single model or software tool. It requires an integrated software infrastructure capable of connecting sensors, operational systems, data platforms, simulation environments, and intelligent decision support systems.
The role of agentic AI in digital twin infrastructures
Recent advances in artificial intelligence have introduced new possibilities for integrating intelligent agents into digital twin environments. Agentic AI systems based on large language models and other AI technologies can interpret information, coordinate workflows, interact with software tools, and assist human operators in analysing complex systems.
When integrated with digital twin platforms, agentic AI systems can help interpret operational data, coordinate optimisation processes, and support decision making in complex infrastructures. Such systems can also interact with simulation environments and digital twins to explore alternative operational strategies or system configurations.
However, integrating agentic AI into digital twin infrastructures introduces new challenges related to system architecture, data integration, reliability, and evaluation. Understanding how these technologies should be combined in practical software infrastructures is therefore an important research challenge.
Investigating digital twin software stacks
The Digital Twin and Agentic AI Infrastructure Lab focuses on investigating the software stack required to implement digital twin infrastructures in complex energy related systems. This includes technologies for data integration, digital modelling, simulation, artificial intelligence, and system orchestration.
A digital twin infrastructure typically involves multiple layers, including data acquisition from sensors and operational systems, data platforms for storing and managing information, digital models and simulation environments for analysing system behaviour, and intelligent applications that support optimisation and decision making.
The lab investigates how these components can be integrated into coherent software architectures that support reliable and scalable digital twin applications.
Experimental environments for digital twin research
The lab provides experimental environments where digital twin technologies and agentic AI systems can be developed and evaluated in realistic settings. These environments allow researchers to connect data sources, simulation tools, digital twin models, and intelligent agents in order to study how these technologies interact.
Through these experiments, the lab investigates questions related to interoperability, software architecture, data management, and the integration of artificial intelligence into operational systems.
By providing a controlled experimental infrastructure, the lab enables systematic evaluation of technologies and architectures for future digital twin based systems.
Supporting intelligent energy infrastructures
The Digital Twin and Agentic AI Infrastructure Lab contributes to the development of intelligent infrastructures that support the green transition of the energy sector. Digital twins and intelligent software systems can improve the monitoring, optimisation, and management of complex infrastructures such as energy networks, industrial systems, and urban energy environments.
By investigating how digital twins and agentic AI systems can be integrated into operational infrastructures, the lab contributes to the development of more adaptive, efficient, and resilient digital energy systems.