Open Call 1 Winners
Open Call 1 Winners – Building the Next Generation of Energy Data Space Solutions
Twelve SMEs and startups from across Europe have joined the HEDGE-IoT ecosystem to develop innovative data-driven services and technologies that strengthen the future of intelligent, interoperable and resilient energy systems.
Meet the 12 Innovators Selected in HEDGE-IoT Open Call 1
GUIDE-NILM
GUIDE‑NILM: Hybrid FHMM‑TCN Edge Microservice for Device‑Level Visibility
Bulgaria
GUIDE-NILM will develop a practical, pilot-ready Non-Intrusive Load Monitoring solution for the Greek Pilot site, enabling the identification of active household appliances from whole-home energy data without requiring additional hardware. It will focus on a limited set of high-value appliance categories, such as refrigeration, washing machines, dishwashers, HVAC or heat pumps, and EV charging where relevant.
Using proven open-source NILM methods and lightweight, reproducible data pipelines, the project will benchmark baseline models, fine-tune them with available pilot data, and deliver a minimal viable service that provides appliance on/off events and daily energy share estimates. The solution will be designed for lean integration with existing HEDGE-IoT enablers and validated through back-testing, calibration, and pilot deployment.
The work will include requirements consolidation, dataset audit, model evaluation, prototype implementation, pilot testing, and final reporting. Expected outcomes include a validated NILM prototype, API documentation, operational guidance, KPI results, and clear identification of performance limits by appliance type.
SEA-URCHIN
SEmantic mapper As a service for Universal and Reliable sCHema Interoperability in the eNergy domain
Poland
SEA-URCHIN will deliver a lightweight semantic mapping service that helps transform heterogeneous energy data sources into standardized, interoperable formats aligned with the HEDGE-IoT ecosystem. The solution addresses the gap between raw data acquisition and semantic interoperability by enabling flexible, reusable mappings to SAREF-based ontologies.
The service will use source-agnostic YAML/YARRRML mappings, supported by AI-assisted configuration, to make data transformation easier for technical users and reduce integration effort. Built on a streamlined version of the Rusalka platform, SEA-URCHIN will support efficient ingestion of diverse data sources into RDF knowledge graphs and enable integration with RDF-native systems.
Expected outcomes include reusable mappings from representative energy data schemas to SAREF, validation shapes, and a containerized edge-cloud deployable service. The solution will be validated with representative data samples and integrated into the HEDGE-IoT framework, supporting future adoption across energy, smart buildings, smart cities, manufacturing, and related domains.
HEDGE-ExpertAI
Context-Aware AI Discovery and Recommendation Assistant for the HEDGE-IoT App Store
Türkiye
HEDGE-ExpertAI will introduce a context-aware AI assistant to make the HEDGE-IoT App Store easier to search, understand, and use, especially for non-expert users. The assistant will support natural-language queries and provide relevant application recommendations with concise, source-grounded explanations.
The solution will combine catalogue indexing, hybrid search, lightweight AI reasoning, and standards-aligned metadata to improve the discovery of applications within the HEDGE-IoT ecosystem. An automated ingestion process will help keep the app catalogue updated as new applications are added.
The project will deliver an integrated plug-in and documented backend service, designed for deployment in the App Store environment. Expected outcomes include a validated AI-assisted discovery tool, reusable technical components, documentation, and an open-source implementation that can support broader use across HEDGE-IoT nodes and related digital marketplaces.
GRIDXAI
Explainable AI Fault Prediction for Enhanced Grid Observability and Resilience
FINLAND
GRIDXAI will use the IntelliView platform to deliver an explainable AI solution for improving grid observability and resilience through predictive fault analysis in digital substations. The solution will analyse existing disturbance recordings to detect early warning signs of faults and outages, estimate likely root causes, and support fault localisation without requiring additional sensors or hardware changes.
The project combines edge-based near-real-time analysis with cloud-based model management, benchmarking, and adaptive improvement. IntelliView uses advanced electrical and statistical features together with physics-informed AI modules to identify pre-fault patterns across different fault types while supporting reliable and transparent operation.
A key outcome will be an operator-facing decision-support tool that presents clear, noise-filtered explanations, relevant contributing factors, and comparable historical cases to support trust and pre-emptive action. GRIDXAI will validate the IntelliView platform in a realistic grid environment and deliver a replicable, interoperable module suitable for wider deployment within the HEDGE-IoT ecosystem.
ECAP
Edge-Cloud Adaptive Control for Portable Intelligent Systems
Greece
ECAP will develop an edge–cloud adaptive control framework for intelligent, energy-efficient management of IoT and energy systems within the HEDGE-IoT ecosystem. The project will demonstrate how self-learning control methods can support portable, resilient, and responsive energy services across hybrid edge and cloud environments.
The solution will combine cloud-based training and optimisation with low-latency edge execution, enabling control algorithms to learn from changing operating conditions and adapt their behaviour in real time. By using the Portuguese Pilot’s existing infrastructure, datasets, orchestration tools, and semantic interoperability capabilities, ECAP will support efficient deployment of distributed AI for energy optimisation, renewable integration, comfort management, and system responsiveness.
Expected outcomes include a validated and replicable control framework, improved edge–cloud orchestration patterns, and reusable components that strengthen AI portability and interoperability across HEDGE-IoT. The project will contribute to more adaptive, data-driven energy services while supporting future deployment in diverse smart energy and IoT environments.
FLAME
Federated Learning for Adaptive Modeling and Edge Forecasting in Energy Communities
Greece
FLAME will develop a scalable, privacy-preserving forecasting framework for energy-related time series prediction in real-world Energy Communities. Building on the FedMaestro federated learning orchestrator, the project will extend decentralised AI capabilities to the energy domain, supporting short-term forecasting of energy consumption and renewable generation at the edge.
The solution will use federated learning to improve forecasting while keeping data local and supporting data sovereignty. It will integrate with the HEDGE-IoT ecosystem through edge orchestration, semantic interoperability, and compliant data exchange mechanisms, enabling forecasting services to operate across households, flexible assets, and environmental data sources.
Expected outcomes include a validated and reproducible federated learning pipeline, benchmarking results against conventional forecasting approaches, and reusable open-source components for future Energy Community deployments. FLAME will support privacy-preserving edge intelligence, decentralised energy optimisation, and future applications such as flexibility services, predictive maintenance, and peer-to-peer energy trading.
HERMES
Hybrid Enhanced Robust Models for Edge-based System Forecasting
FRANCE
HERMES will develop an edge-native AI forecasting platform for improving distributed grid operation and DER optimisation within the HEDGE-IoT ecosystem. Designed for deployment in challenging energy environments, the solution will support reliable short-term forecasting even when data is noisy, incomplete, or affected by unstable connectivity.
The platform combines local data cleaning and gap-filling with adaptive machine learning models that can generate forecasts directly at the edge. This approach reduces reliance on cloud infrastructure while enabling faster, more resilient predictions of grid and energy behaviour to support congestion management and distributed optimisation.
HERMES will be delivered as a reusable, containerised microservice and validated within the HEDGE-IoT testbed. Expected outcomes include improved forecasting robustness, stronger edge autonomy, interoperability with relevant energy protocols and data models, and a replicable solution for future deployment in islands, rural microgrids, and other distributed energy settings.
PROOF-OF-FLEX
Tokenized Settlement and Auditing Layer for Cross-Border Flexibility Markets
ITALY
Proof-of-Flex will develop a blockchain-based settlement layer for verified energy flexibility events within the HEDGE-IoT ecosystem. The solution will use smart contracts, semantic data exchange, and data-sovereignty mechanisms to create a transparent and machine-auditable way to account for flexibility actions across multiple stakeholders.
The system will introduce technical accounting tokens that represent verified flexibility events without functioning as financial or speculative assets. These records will be linked to notarised proofs and anonymised data exchanged through secure data connectors, supporting privacy-preserving verification and auditability.
Expected outcomes include a validated laboratory prototype combining smart contracts, token-based accounting logic, and an off-chain settlement engine. Proof-of-Flex will support greater trust, interoperability, and transparency in flexibility markets, while providing a replicable settlement module for future HEDGE-IoT pilots and data-space ecosystems.
MICROGRID
Modular Integrated Campus Resource Optimisation and Geospatial Representation for Intelligent Distribution
THE NETHERLANDS
MICROGRID will deliver a digital-twin solution for campus-level energy management within the HEDGE-IoT ecosystem. The project is designed to help campus energy managers better understand and manage complex energy systems by providing a shared, decision-ready view of energy consumption, distribution, and system behaviour across the site.
The solution will combine a web-based 3D interface, integration with live HEDGE-IoT data streams, and basic simulation capabilities. Users will be able to explore the campus energy system visually, monitor real-time conditions, and assess what-if scenarios such as changes in demand or the addition of new distributed energy resources. The platform will also visualise alerts, anomalies, and operational insights in an intuitive geospatial format.
Expected outcomes include a working demonstrator integrated with HEDGE-IoT data streams, offering a user-friendly and replicable interface for improving transparency, decision-making, and optimisation in distributed campus energy systems. MICROGRID will contribute to the digitalisation of local energy management by combining real-time data, semantic interoperability, and interactive visualisation.
FlexCells.IT
FlexCells.IT – “DSO Perimeters” engine for VPP pre-qualification
ITALY
FlexCells.IT will deliver a containerised microservice for automatically defining stable electrical cells, or DSO perimeters, in distribution grids. The solution will help DSOs and aggregators identify reliable grid areas for pooling flexibility and supporting Virtual Power Plant pre-qualification.
The system will combine grid topology data, operational time series, forecasting, scenario simulation, and optimisation to generate connected perimeters that remain technically feasible under different operating conditions. Each candidate perimeter will be assessed through an interpretable stability score, helping users understand how robust it is across seasonal variation, switching conditions, and stress events.
FlexCells.IT will provide results through standard APIs and a lightweight web map, with deployment designed for the HEDGE-IoT Marketplace and compatible grid environments. Expected outcomes include reduced manual engineering effort, more reliable flexibility pre-qualification, improved operational transparency, and a reusable solution that can be replicated across different grids by adapting local data connectors and engineering parameters.
MIDAS
SPAIN
MIDAS Connector will integrate a lightweight, edge-capable data space connector into the HEDGE-IoT ecosystem to demonstrate scalable, standards-based, and technology-agnostic data exchange. The project will show that HEDGE-IoT can support heterogeneous connector implementations while maintaining interoperability across energy data-sharing scenarios.
The solution will validate secure and compliant data exchange across representative energy dataset types, working alongside existing HEDGE-IoT connectors. Its lightweight design is intended to lower infrastructure barriers and enable participation by smaller, resource-constrained actors, supporting a more distributed and inclusive energy data ecosystem.
Expected outcomes include a validated connector integration, demonstrated interoperability with HEDGE-IoT components, and a reusable standards-based solution that can be replicated beyond the energy sector. MIDAS Connector may also support future deployment in other IoT-intensive domains, such as smart cities, industrial IoT, and broader data space ecosystems.
AURORA
Slovenia
AURORA will develop a satellite-informed AI forecasting framework to support solar power prediction and DER optimisation within the HEDGE-IoT ecosystem. The solution will combine satellite imagery, solar generation data, and weather forecasts to produce short-term solar power predictions that can feed scheduling and optimisation tools.
The project will demonstrate how improved solar forecasting can support more efficient operation of distributed energy resources, including battery charge and discharge scheduling. By providing more reliable visibility of expected renewable generation, AURORA will help reduce operational uncertainty, improve local balancing, and support better use of storage and other flexible assets.
Expected outcomes include a validated forecasting framework, integration with DER optimisation workflows, and reusable forecasting outputs that can support multiple HEDGE-IoT use cases. The approach is designed to be scalable across different locations and applications, contributing to renewable energy integration, grid efficiency, and AI-enabled energy services.
HEDGE-IoT Open Call 2
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore.