Goals

The aim of the project is to implement a Continuous Intelligence platform based on Machine Learning models for predicting LSTID a few hours in advance through the analysis and interpretation of a wide range of geomagnetic observations of the solar corona, interplanetary medium, magnetosphere, ionosphere, and atmosphere. Specifically, the platform will provide a real-time monitoring tool, which by exploiting the predictive capabilities of Machine Learning models, will implement an Early Warning service. In addition, the prediction models will be built using information captured from a large and heterogeneous set of time-dependent signals (e.g., time series) from various regions of geospace and using as a training label the European sector-related LSTID datasets, available to the partners (e.g., INGV) as a result of previous experience in projects dedicated to LSTID detection. These data provide a solid and reliable basis on which to build robust and accurate supervised Machine Learning models.

Partners