Predicting Solar Eruptive Events Using Artificial Neural Networks
Abstract
Solar eruptive events, such as coronal mass ejections (CMEs), are the major sources of space weather events, causing disruptions and damages to our technological infrastructure, and endangering humans in space. Prediction of such eruptive events is highly challenging mainly because of lack of a single physics-based model adequately representing the complex solar dynamo and the magnetized, turbulent plasma. Methods of artificial intelligence (AI) can develop algorithms to identify patterns in long-stretches of data and make effective predictions with reasonable accuracy. Making use of deep neural networks (DNNs), we make an attempt to predict a few events in the SHARP catalog. Presented here are the results and accuracies of our prediction.
- Publication:
-
Solar Heliospheric and INterplanetary Environment (SHINE 2018)
- Pub Date:
- July 2018
- Bibcode:
- 2018shin.confE.157P