Modeling the coronal magnetic ﬁeld of the Sun is an important objective in heliophysics. In this study, we illustrate how to use deep learning to estimate the parameter for a magnetic ﬁeld model. A linear force-free magnetic ﬁeld conﬁguration is employed to model, as an initial illustration, an active region by using two magnetic dipoles and to determine the associated linear force-free ﬁeld (LFFF) α parameter from a set of pseudocoronal loop images which serve as training and validation sets to existing deep learning algorithms. Our results show very high accuracy of determining the LFFF parameter α from pseudocoronal loop images.
Keywords?Sun; solar corona; solar activity; magnetic ﬁeld models; deep learning;
Authors: Bernard Benson, Zhuocheng Jiang, W. David Pan, G. Allen Gary and Qiang Hu
Publication Status: Accepted to CSCI-ISAI 2017, Las Vegas Conference
Last Modified: 2017-12-05 11:51