What started with an algorithm designed to capitalise on the value of information and data from wind and solar farms has evolved into an innovative platform for forecasting renewable energy production, particularly offshore wind power. Predico, developed by INESC TEC, is already being tested in an operational environment with Elia, a system operator from Belgium and Germany.
More than just an energy forecasting platform, Predico proposes a collaborative approach and a data market: it connects companies interested in renewable production forecasts with the suppliers of said forecasts. “Interested companies fund specific forecasting challenges, while suppliers submit their forecasts and are compensated based on forecasting errors and the relevance of their data to the platform’s collaborative forecasting process. This machine learning system can combine individual predictions and optimise accuracy – a key factor in more efficient power system management,” said Ricardo Andrade, a researcher at INESC TEC.
The system will help reduce the error margin in suppliers’ forecasts, facilitating smoother integration of wind power into the power grid and addressing the growing need for forecasting solutions in the renewable energy sector.
But let’s rewind to 2021 and the European project Smart4RES: Carla Gonçalves and Ricardo Bessa developed a prototype that has since evolved into a robust and accessible software, which will be available as open-source by the end of the year.
“We started by using real-time production data from a set of wind and solar farms to help other renewable electricity generation points reduce errors in forecasting their production. Now, we focus on providing a platform that will allow external access to this solution,” said Ricardo Bessa, INESC TEC Power and Energy coordinator.
Elia’s involvement – launching the challenge of developing a tailored solution for renewable energy production forecasting – focused on the offshore wind power sector, specifically addressing medium power forecasts and ramps. This collaboration was crucial for developing and demonstrating the platform in an industrial environment.
The Future… is it Predictable?
Predico’s innovation lies in its ability to generate detailed probabilistic forecasts, providing the market an unprecedented outlook of wind production and addressing the complex needs of grid operators.
“This is more than creating probabilistic forecasts for average power, which is already widely used in the industry. We focus on the variability and identification of ramps (including direction and magnitude),” explained Ricardo Andrade. Additionally, the system includes a component that identifies and quantifies the value of the data provided by each participant to the buyer. “This feature is particularly innovative in an information-flooded context, as it allows people to identify vital data. It promotes the creation of fair incentives for information sharing, fosters a healthy data and forecasting market, and facilitates collaboration chains that benefit participants and add real value to the buyer.”
INESC TEC aims to evolve this technology beyond the proof of concept by developing new features and robust business models that clearly demonstrate competitive advantages for industrial customers.
“Our goal is for Predico to become the reference in the renewable energy forecasting market. This platform has the potential to transform the energy sector, promoting a collaborative forecasting system that benefits all stakeholders in the value chain and contributes significantly to the sustainable energy transition,” said Ricardo Bessa.
The researchers mentioned in this news piece are associated with INESC TEC.
News, current topics, curiosities and so much more about INESC TEC and its community!
