This research aims to look into the fundamental problem of formation of
nonlinear structures in earth’s magnetosphere through theoretical and
computational modelling along with the application of artificial neural networks.
We plan to employ the so-called Physics Informed Neural Network (PINN) with which we shall use Deep Operator and Multi-fidelity Network. The PINN is a relatively new technique that can be effectively employed for parameter discovery of a theoretical model. It can also be employed to model the necessary physics to be incorporated by feeding the relevant observational data.
Achievements
The project being undertaken by ISRO in collaboration with academia is directly or indirectly beneficial to the society.