http://arxiv.org/abs/1801.04435
We attempt to forecast the Sun’s sunspot butterfly diagram using neural networks as a prediction method. We use this approach to forecast in both latitude (space) and time, using a full spatial-temporal series of the sunspot diagram from 1874 to 2015 that trains the neural network. The analysis of the results show that it is indeed possible to reconstruct the overall shape and amplitude of the spatial-temporal pattern of sunspots using these feed-forward neural networks. However, we conclude that more data and/or improved neural network techniques are probably necessary for this approach to have real predictive power.
E. Covas, N. Peixinho and J. Fernandes
Tue, 16 Jan 18
49/79
Comments: N/A