Seminar: Machine Learning Algorithms for Geomagnetically Induced Currents
Wednesday, April 28, 2021 - 3:00pm to 4:00pm
Speaker: Amy Keesee, UNH
Abstract: Geomagnetically induced currents (GIC) can occur during geomagnetically active intervals. GICs can drive power outages and damage power grid components while also affecting pipelines and train systems. Developing the ability to predict local GICs is important to protecting infrastructure and limiting the impact of geomagnetic storms on public safety and the economy. While GIC data is not readily available, variations in the magnetic field, dB/dt, measured by ground magnetometers can be used as a proxy for GICs. Supported by a NSF EPSCoR Track II award, the MAGICIAN team is a collaboration between UNH and the University of Alaska Fairbanks to study this issue. The availability of ground magnetometer data from SuperMAG enables the application of machine learning techniques to model perturbations in the magnetic field. We are developing a set of neural networks to predict dB/dt intervals that lead to favorable conditions for GICs. Our initial work is to study the connection of solar wind and interplanetary magnetic field properties using OmniWeb data with the occurrence of the magnetic field perturbations. We also plan to incorporate data from magnetosphere and ionosphere missions to improve the models and the resulting predictions.
This online seminar will take place on Zoom and will require a password to join. If you wish to watch the seminar and have not received a password via email, please contact Robbin McPherson at email@example.com.