Skip to main content


ROB logo BELSPO logo KUL logo

DELPHFI (DEep Learning Prediction and Hindsight of Flare Initiation) is a collaboration project between ROB and KU Leuven University, within the Brain-be 2.0 framework funded by the Belgian Federal Science Policy (BELSPO). It started on December 15, 2020 and has a duration of 4 years.

The DELPHFI project aims at:

  • Improving our understanding of the mechanisms that lead to solar flares, thanks to the interpretation of the results obtained by modern Machine Learning techniques (solar flares originate in the “active regions” of the Sun. These are sudden energy releases that emit extensive radiation in a very broad range of wavelengths from X-ray to radio ranges);
  • Demonstrating how modern Machine Learning techniques, based on the automatic extraction of active region features, can provide better flare forecasts than human operators or existing automatic methods;
  • Increasing the expertise of ROB in the development of Machine Learning flare predictions, and to set the ground for a new operational tool and its possible extension to other kinds of eruptive events.


View more details.