Edge AI: Deploying AI flexibility algorithm in Substations

With the increased number of distributed energy resources DERs onto the grid, energy operators need a predictive system based on consumption and production patterns to help them avoid congestion and overvoltage events in the grid. In this article, we cover a specific project we are running under the i-nergy programme, an EU-funded initiative aiming to support and develop new AI-based Energy Services.

Smart Grid
Written by:
Miren Zabaleta

Edge Computing to bring intelligence onto Medium and Low Voltage network

DSOs need to accommodate increasing numbers of distributed energy resources (DERs) on the distribution grid. The pressure is high in the Low Voltage Grid as it is much less digitized and it puts on hold the resilience of the grid.

Besides, Medium and Low Voltage network, with thousands of electrical Transformation Centers widely distributed geographically  have a need for real-time information that the Cloud cannot provide due to latency issues, cost and scalability. SCADA technologies highly oriented towards automation and with proprietary data structures, are neither very flexible and nor accessible, and do not cover the needs in an optimal way either.

The solution is halfway between the Cloud and SCADAs: Edge Computing.

"Edge Computing facilitates the real time use of information from IoT devices to provide greater flexibility, effectiveness and efficiency to Low Voltage Grid.”  Barbara

Barbara and Cuerva join forces to bring AI Flexibility algorithms to the Edge

The goal is to have a much clear view on how the grid will behave in the future as well as to address techniques that can solve the problems of congestion and overvoltage events.

Cuerva will develop and implement the AI algorithm for the demand and generation prediction in a Low Voltage network, through an Edge node located in their substation in Granada. Barbara will provide the Edge Platform to deploy Artificial Intelligence in the Edge and will enable MLOps techniques to facilitate the algorithm retraining in the Edge.

The system as a whole, thanks to these capabilities, will enable agents of the Energy Value Chain to create new business models and optimise their Operation and Planning processes in the Transformation Centers.

Technology must be an enabling vector for this goal and this project aims to achieve solutions that will allow operators to have a more resilient, enhanced, and secure grid in the context of rising number of distributed energy resources (DERs) on the distribution grid.

Barbara´s Edge Technology offers a series of unique features to DSOs

● Multi-author containerized algorithms that will allow new and novel flexibility business models to be tested and to co-exist cybersecurely in Edge Nodes. Cuerva will be able to deploy its own algorithms and algorithms developed by third parties.

● To certify a maximum cyber-security level, Edge Nodes will be IEC 62443 compliant.

● A Control Room to allow Edge Node firmware and their algorithms to be remotely and cyber-securely monitored, managed, and updated over-the air enabling limitless scalability.

● Edge Nodes will allow assets to be manufacturer-independent avoiding vendor lock-in.

A Horizon Europe programme

DSOs and TSOs around Europe will benefit from these technologies by having forward information of their users' behaviour allowing them to take grid operation decisions with more consciousness and security.

More precisely it will allow them to:

● Predict the patterns in their networks, allowing them to be one-step farther than the network issues, by using AI algorithms.

● Solve contingencies in their networks in an efficient and fast way, thanks to real-time data offered by Edge-Nodes.

● Cluster the Flexibility Service Providers (FSP) in their networks, one of the most important solutions to the problems caused in the network of the future.

● Create a data-infrastructure between DSOs, TSOs and consumers by using a common platform to Exchange data, information and algorithms.

● Include this intelligence into Barbara’s Edge Node, enabling the DSO to monitor its grid in real-time with great timing capabilities.

● Optimise the orchestration of distributed IA and allow the Machine Learning algorithms to be executed on the Edge with MLOps activities.

About i-nergy Open Call 2  

I-nergy is an EU funded initiative, aiming to support and develop new AI-based Energy Services. Barbara has been selectect together with Cuerva to research and implement AI Flexibility Algorithms that will contribute to European AI-on-demand (AIOD )platform.

The programme includes mentoring services and a maximum amount of financial support for each third party of up to 100.000 EUR.

The AI4Europe is one of the projects, funded under the Horizon Europe programme, that is responsible for the management, development and facilitation of the AI-on-Demand Platform (AIoD).

More information: https://i-nergy.eu/-open-call-2-evaluation-report