Industrial Domains

Energy

The Energy industrial domain is driving the transition to sustainable, resilient energy systems by using AI to address challenges like RES variability and forecasting uncertainties. AI enhances decision-making in real-time energy system operations, improving grid stability and integrating decentralized resources like renewable energy, energy storage, and EVs.
AI also aids in the evolution of electricity markets by automating decision-making in local energy communities, such as peer-to-peer trading. It supports system resilience by integrating diverse data sources, helping to mitigate risks like climate change and cyber-attacks. Additionally, AI optimizes electricity demand patterns through smart devices, enabling innovations like energy poverty forecasting and efficiency recommendations, reducing costs and increasing grid flexibility.

Sub-Domains
Use Cases
Projects
Project: Generic Framework for Systemic and Local Explainability of Dynamic Systems

Third parties involved

This project leverages transformer-based models with explainable AI to analyze sleep dynamics and physiological signals related to schizophrenia relapse using wearable devices. Expected outcomes include pre-trained models for sleep analysis, new explainability techniques, enhanced libraries for signal analysis with deep learning, and improved relapse prediction to advance AI-driven healthcare in Europe.

Project: Physics-aware neural networks and edge processing for low voltage systems (PANNEL)

Third parties involved: Stefanos Petridis, Petros Iliadis, Angelos Skembris, Dimitrios Rakopoulos – Sustenegro 
This project develops AI/ML algorithms for Low Voltage Distribution Networks (LVDNs) to deliver a granular view of observability, topology, and phase imbalances, while leveraging edge processing on metering devices for on-device pre-processing that reduces data volume and computational load. At its core is a physics-aware neural network that performs full-network state estimation and topology inference. This is complemented by: (i) a bad-data and fault detection module based on support vector machines (SVMs), and (ii) a technical and non-technical loss evaluation module powered by a deep neural network (DNN)—together enabling accurate, scalable, and efficient grid intelligence at the edge.

Project: Advancing Power Grid Inspection with SCENE-Net 1.5

Third parties involved: Sajjad Fattaheian Dehkordi – Aalto University
 
The project aims to advance power grid inspection by introducing SCENE-Net 1.5, an enhanced version of the existing SCENE-Net. Featuring a classifier head for multiclass segmentation, the model targets improved accuracy by detecting all elements in the 3D scene. By releasing a power-grid point-cloud dataset, the project intends to disseminate this problem to the research community. Overall, the proposal promotes safe and fast power grid inspection methods.

Project: SEASHINE – Safe intelligent agent to optimize ship energy management

Third parties involved: Udayanto Dwi Atmojo, Erald Shahinas, Carl Akira King, Valeriy Vyatkin – Aalto University

The project aims to define and standardize green AI metrics across the entire AI lifecycle. It will review existing metrics in sustainable software engineering, propose standardized calculations, and introduce new metrics where current ones are inadequate. These metrics will be evaluated and validated. The project will also outline the data requirements and functionalities for a benchmarking tool to minimize AI’s environmental impact. Expected outcomes include a robust open-source green AI metrics framework and its application to assess AI architectures for energy efficiency and performance. This will provide a foundation for European policymakers to develop regulations and eco-labels that promote environmentally responsible AI development and deployment.

Projects: AI-based Edge Control of Electric Vehicles in Low-Voltage Grids Considering Operational Issues in Green Energy Systems with High Penetration of Renewable Energies

Third parties involved: Sajjad Fattaheian Dehkordi – Aalto University

The project addresses the growing integration of renewable energy sources (RESs) into low-voltage grids and rising demand, which can lead to operational issues such as congestion. These challenges stem from overgeneration/overconsumption by RESs/loads and the traditional “fit-and-forget” design of such systems. To mitigate these issues, the project will leverage flexible resources—particularly electric vehicles (EVs). Specifically, the project will employ artificial intelligence (AI) techniques to model and predict EV behavior and to influence charging through control signals, enabling utilities to manage operational constraints like congestion in future green energy systems. By doing so, the project will facilitate the transition toward green, smart energy systems. The project is currently framed around congestion management but remains adaptable to closely related topics if required.