22. Edge Computing

Edge computing is evolving as new technologies such as artificial intelligence and machine learning bring new data analysis capabilities to the table, and emerging business models such as IoT as a service enable solution providers to deliver innovative offerings in new ways. With an average increase of 16% per year, spending on edge computing totalled $40 billion in 2022 in Europe and should reach $64 billion in 2025. Service providers, editors and manufacturers are sharpening their offers to meet the growing demand from companies.

Opportunities for DSOs

• Enhanced Grid Visibility and Control at the Grid Edge: The enhancement of computing capabilities enables the rapid execution of complex distributed analyses, especially in primary and secondary substations. This allows DSOs to move from reactive to proactive network management, for example by using local data to predict transformer overloads or proactively manage voltage fluctuations caused by high solar PV penetration.
• Improved Network Resilience and Reliability: This provides flexibility to the network and minimizes reaction times by enabling local decision-making and automation of grid operations (e.g., FLISR – Fault Location, Isolation, and Service Restoration). Directly translates to improved SAIDI/SAIFI reliability metrics, a key performance indicator for DSOs, by significantly reducing outage durations..
• Foundation for a Self-Healing, Autonomous Grid: Facilitates the autonomous operation of the medium and low-voltage network, [reducing the need for manual intervention and creating a pathway towards a more resilient, self-optimising electrical system.
• Efficient Data Management and Reduced Telecoms Costs: Optimizes data storage by processing raw data locally and only transmitting valuable insights or alarms back to central systems. This drastically reduces the cost and complexity of scaling communications networks to thousands of secondary substations.

Challenges for DSOs

• Lifecycle Management and Scalability: Equipment Lifecycle. Deploying, managing, patching, and updating the software and firmware on thousands or tens of thousands of distributed edge devices presents a significant operational challenge. DSOs must consider the total cost of ownership, including remote management and end-of-life replacement.
• Interoperability and vendor Lock-In: Increased likelihood of failure and difficulty in settling responsibilities. This is attributed to the integration of currently independent devices, resulting from the integration of various use cases. Without a commitment to open standards, DSOs risk being locked into proprietary, single-vendor ecosystems, which limits innovation and increases long-term costs. This makes industry initiatives like the E4S Alliance critical.
• Cybersecurity at the edge: While edge computing can enhance security in some ways, each intelligent device also represents a new potential entry point for cyber-attacks. Securing a massively expanded attack surface requires a robust, end-to-end security architecture from the device to the control room.
• Data governance and algorithm ownership: As critical functions like fault detection are run on edge devices, clear governance is needed. DSOs must address questions of who owns the data, who is responsible for the algorithms, and how to ensure their performance and reliability over time, especially when using third-party solutions.
• Skills and organisational change: Successfully leveraging edge computing requires converging traditional Operational Technology (OT) and Information Technology (IT) skills. DSOs will need to invest in training and potentially restructure teams to manage these new, software-defined grid assets effectively.

E.DSO considerations

• DSOs should conduct analyses and implement demonstrators to identify relevant use cases for edge computing.
• The technical benefits of edge computing solutions are evident for secondary substations. It is necessary to be monetised.
• DSOs should also recognise the contribution of such solutions at primary substations and throughout the network.
• Active participation in ongoing industry discussions is crucial for DSOs. These discussions should encompass governance and standards, including aspects such as cybersecurity, software solutions, and the definition of future standards.
• Participation is vital to advocate for the specific requirements of DSOs.
• DSOs need to develop a comprehensive security strategy for edge computing.

Potential use cases

Enhanced Grid Visibility & Operations

• LV Grid Visibility: Turns the low-voltage network from a “black box” into a fully visible asset, enabling proactive voltage management (e.g., for solar PV) and rapid fault location.
• Intelligent Data Gateway: Aggregates and pre-processes local data from smart meters and sensors, reducing the load on central systems and cutting communication costs.
• Predictive Analytics: Uses local AI/ML to forecast transformer load and capacity issues, allowing for pre-emptive action before an overload occurs.

Optimised Asset & Energy Management

• Condition-Based Asset Management: Continuously monitors transformer health (temperature, load, gases) to prevent catastrophic failures, reduce maintenance costs, and extend asset life.
• Local Flexibility Orchestration: Manages local Distributed Energy Resources (DERs) like batteries and EV chargers as a “non-wires alternative” to defer or avoid costly network upgrades.
• AI-Powered Inspections: Uses cameras and AI to automatically detect equipment hotspots, physical damage, or vegetation encroachment, reducing the need for manual site visits.

Ongoing projects

• Special mention deserves the E4S Alliance (Edge for Smart Secondary Substation), where a standards-based, open, interoperable, and secure architecture is being defined to enhance the automation, scalability, security, and manageability of secondary substations worldwide. Involved DSOs : i-DE, E-Redes, Enedis, UFD, Helen, Unareti/A2A (more info).
• i-DE and Iberdrola Group:
  • E4S (acting as DSO) – pending demonstration pilot during the first half of 2026.
  • VIRTGRID (acting as external DSO validator).
  • IA4TES (acting as promoter), testing Artificial Intelligence (AI) use cases and software applicable to the edge.
  • SEC2GRID (acting as external DSO validator), testing cybersecurity in distributed environments and virtualisation of cyber use cases.
  • Virtual Data Concentrator (internal development pilot at i-DE).
  • DPP solution (2023). RFI analysis and evaluation of state-of-the-art edge computing solutions to adapt to the DSO environment.
  • MiDE4S (2021). Demonstrative piloting of Minsait solution in i-DE’s environment.
• UFD and Naturgy Group:
  • E4S (role DSO) – pending demonstration pilot during the first half of 2026
  • VIRTGRID (acting as external validator DSO)
• Stedin:
  • Piloting AI on the edge with high frequency or real time data collection (from sensors, microphones and thermal cameras) in substations primarily for asset management and prediction. Visualisation of aggregated load via anonymized smart meter data, providing insights into grid performance.
  • Piloting modular architectures on edge computers to test security risks and flexibility benefits.
• HEDNO:
  • Participates in HEDGE-IoT and ODEON HE projects.
• ENEDIS:
  • In addition to its participation in E4S as DSO, Enedis is working on Edge Computing through academic partnerships and has been financing PhD works: “Analysis of new IT architectures for managing distribution networks implementing centralized and distributed smart functions” (PhD, INPG 2025-2027)

Last update: 30 September 2025