Predicting geomagnetic storms from solar-wind data using time-delay neural networks

We have used time-delay feed-forward neural networks to compute the geomagnetic-activity index <i>D<sub>st</sub></i> one hour ahead from a temporal sequence of solar-wind data. The input data include solar-wind density <i>n</i>, velocity <i>V</i> an...

Full description

Bibliographic Details
Main Authors: H. Gleisner, H. Lundstedt, P. Wintoft
Format: Article
Language:English
Published: Copernicus Publications 1996-07-01
Series:Annales Geophysicae
Online Access:https://www.ann-geophys.net/14/679/1996/angeo-14-679-1996.pdf