The delay in generating synthetic data for time series – fundamental elements in energy forecasting scenarios – was one of the motivations behind GENESIS, a project that aims to provide local electricity markets with contextual synthetic data and reliable Artificial Intelligence (AI) models.
Technological limitations, lack of data, and the absence of accessible tools are among the challenges faced by Local Energy Markets (LEMs) – initiatives promoted by OMIE (the Iberian electricity market operator) that aim to empower consumers and integrate renewable energy sources, thus increasing the efficiency of the grid. All these factors hinder the creation of representative and explainable synthetic data, resulting in unreliable scenarios for these markets.
The GENESIS project seeks to overcome these obstacles, by combining contextual generative models with explainable AI to develop reliable, transparent, and accessible solutions – but above all, solutions validated in real-world energy applications. The project, bringing together INESC TEC’s expertise in power systems and human engineering systems, is divided into two development phases.
The first focuses on data collection – crucial to feed AI models -, including data on energy consumption, renewable production, market prices, and contextual factors. The second phase aims at developing generative models capable of creating realistic and contextual synthetic data. This step is crucial for building explainable AI models adapted to different user profiles.
According to Tiago Campelos Pinto, INESC TEC researcher, the GENESIS project aims to create “new scientific knowledge” combined with “innovative technological solutions.” “We could apply the results directly to consumer and producer training, decision-making support for regulators and grid operators, and the development of new commercial solutions for companies in the energy sector,” he highlighted.
Expected outcomes include the creation of “realistic contextual synthetic data that can be used to simulate complex energy scenarios without compromising user privacy.” Another aspect involves the implementation of “explainable AI models capable of tailoring explanations to the needs and profiles of different users” – all with the goal of “increasing transparency and trust in solutions”.
Concerning local energy markets, they will also benefit from a simulator that will make it possible to “test, compare, and validate different market models in a controlled environment,” with the active participation of experts and end users.
The GENESIS project, as an R&D initiative, is promoted under the Foundation for Science and Technology, also involving researchers from São Paulo State University – a partnership that strengthens “the project’s international dimension and adds expertise in optimisation, applied AI, and energy systems planning,” concluded the researcher.
The researcher mentioned in this news piece is associated with INESC TEC and the University of Trás-os-Montes e Alto Douro (UTAD).
News, current topics, curiosities and so much more about INESC TEC and its community!
