Demand-side Management can contribute to the EU energy and climate objectives, by optimizing the energy consumption and minimising associated emissions. It can further help to increase power system efficiency and at the same time maximize economic benefits for energy providers and users by capitalizing on market opportunities such as energy arbitrage strategies. DSM schemes provide relatively simple and cost-effective ways to better account for intermittence and flexibility of VRE generation and use.

Islands are especially interesting as they provide a closed and well-defined development and testing ground for DSM solutions. Testing DSM on islanded systems will provide valuable insights into its overall value, benefits, challenges and applicability to larger power systems, highlighting both similarities and key differences.

This programme will support European targets towards net-zero emissions and address Challenge 1 in the Strategic Research and Innovation Agenda (SRIA) of the CETPartnership and the following 9 objectives:

• [O1] Demonstrate how to increase VRE contribution and utilisation in power systems.
• [O2] Provide evidence of DSM benefits for system operators.
• [O3] Provide evidence of DSM benefits for DSM flexibility providers, energy generators and users.
• [O4] Contribute to a widespread implementation of DSM techniques for islanded power systems.
• [O5] Promote transnational collaboration between system operators and DSM providers.
• [O6] Assess the replicability and scalability for similar systems.
• [O7] Promote gender equality and diversity in the energy sector.
• [O8] Build a transnational joint programming platform focusing on future smart energy and flexibility services.
• [O9] Create an innovation ecosystem that supports capacity building for smart energy solutions.

Project ambitions

Islands are an excellent development and testing ground for DSM solutions, because they provide a closed and well-defined environment. However, implementing DSM requires a framework that is adaptive to a dynamic environment, responsive to consumer preferences, and fully automated. Without these elements, the system may fail to deliver optimal results, and consumers could experience a phenomenon known as response fatigue, ultimately reducing their willingness to provide flexibility. Therefore, to effectively implement DSM, three main challenges need to be tackled:

• Developing scalable and easy to adapt solutions for fully automatic control and scheduling of demand-side resources.
• Recognizing accurate approaches for load forecasting.
• Identifying the right consumers who are eager to participate and considering their preferences.

Beyond designing a DSM system and its control algorithms, further studies are needed to assess how consumer responses impact real-time balancing, frequency control of the island grid and key technical parameters of the power system. This can be achieved by modelling the power grids in an appropriate power system analysis software and using simulation tools.

High-level methodology

We will investigate the individual DSM potential for different islands, and we will create a novel AI-based DSM control solution (see figure 1 below). To achieve this, machine learning techniques will be used along with novel interoperability protocols.