Prizes worth two million euros have been awarded to three European prediction specialists for developing the most accurate predictions of electricity flow through a grid. The three winners of the Big Data Technologies Horizon Prize received their awards at a ceremony on 12 November in Austria.

Big Data Technologies Horizon Prize winners

The first prize of €1.2 million went to Professor José Vilar from Spain, while Belgians Sofie Verrewaere and Yann-Aël Le Borgne came in joint second place and won €400,000 each. The Big Data Technologies Prize challenge was open to individuals, groups and organisations from countries taking part in the EU’s research and innovation programme, Horizon 2020. This is the first time that the prize has been won by three individuals rather than organisations or groups.

Commissioner for Digital Economy and Society Mariya Gabriel said: “I am delighted to see that these prestigious prizes have been won by three great specialists. It is a clear sign of the depth of potential we have in European research - talent that needs to be nurtured and motivated. The wide range of possible applications of these winning submissions could bring tangible benefits to all European citizens.”

Carlos Moedas, Commissioner for Research, Science and Innovation, said: Energy is one of the crucial sectors that are being transformed by digital technologies. This Prize is a good example of how we support a positive transformation through the EU's research and innovation programme, Horizon 2020.  For the future, we have designed our next programme, Horizon Europe, to put even more emphasis on the merger of the physical and digital worlds across sectors such as energy, transport and health.

The challenge for the applicants was to create software that could predict the likely flow of electricity through a grid taking into account a number of factors including the weather and the generation source (i.e. wind turbines, solar cells, etc). Using a large quantity of data from electricity grids combined with additional data such as weather conditions, applicants had to develop software that could predict the flow of energy through the grid over a six-hour period. The software modules were tested for accuracy and speed, with seven out of 24 applications managing to complete the task successfully. The winners were selected based on combined rank of accuracy and speed, with greater weight being given to accuracy.

Real-life application

The decision to focus on energy grids for this particular prize was driven by a clear market need. Because today's energy is no longer produced at a few, stable plants, but at millions of dispersed and more unpredictable sites (wind turbines, solar cells, etc.), it is harder to ensure that electricity supply matches the demand at all times. Energy grids are complex and interconnected, with electricity traded across borders to meet demand peaks and get rid of surplus energy. This complexity means that huge amounts of data are produced at the energy generation sites (for example by turbines), in the grid (by power lines) and at the place where the energy is consumed. Being able to make accurate, short-term (i.e. the next hour) predictions about power grid traffic is therefore vital to reduce the risks of blackouts or waste of energy.

The aim is that the winning solutions will be taken up by the energy sector and eventually lead to smarter, more economic and more reliable power grids. They also have potential applications in other fields such as biology and healthcare, where reliable predictions are used to help diagnose and cure diseases as well as to allocate resources where they are most needed.

For more information