Neural network prediction of geomagnetic activity: A method using local Hölder exponents

Abstract

Local scaling and singularity properties of solar wind and geomagnetic time series were analysed using Hölder exponents α. It was shown that in analysed cases due to the multifractality of fluctuations, α changes from point to point. We argued there exists a peculiar interplay between regularity/irregularity and amplitude characteristics of fluctuations which could be exploited for the improvement of predictions of geomagnetic activity. To this end, a layered back-propagation artificial neural network model with feed-back connection was used for the study of the solar wind magnetosphere coupling and prediction of the geomagnetic Dst index. The solar wind input was taken from the principal component analysis of the interplanetary magnetic field, proton density and bulk velocity. Superior network performance was achieved in cases when the information on local Hölder exponents was added to the input layer.