期刊名称：Lecture Notes in Engineering and Computer Science
印刷版ISSN：2078-0958
电子版ISSN：2078-0966
出版年度：2018
卷号：2235&2236
页码：273-277
出版社：Newswood and International Association of Engineers
摘要：Predicting solar storms from real-time satellites data is
extremely important for the protection of various aviation, power and
communication infrastructures. There is therefore much current
interest in creating systems which can make accurate solar flare
predictions. This research investigates whether we can process
Geostationary Operational Environmental Satellite (GOES) data, from
pre-flare phases, to provide useful predictions for flares by convert
GOES X-ray flux 1-minute data from 2011 to 2016 to Gramian
Angular Fields (GAF) images. Then the GAF images are used as input
to Deep Learning Neural Network platform. GOES data and deep
learning technologies are not used heavily for flares prediction and in
this paper, the potential and challenges of developing new deeplearning
based space weather technology, are investigated.
关键词：Deep learning; Convolutional neural
networks; Solar flares; Flare prediction; GOES
