dorsal/arxiv
View SchemaNeural network prediction of geomagnetic activity: a method using local H\"{o}lder exponents
| Authors | Z. Vörös, D. Jankovičová |
|---|---|
| Categories | |
| ArXiv ID | physics/0411062 |
| URL | https://arxiv.org/abs/physics/0411062 |
| DOI | 10.5194/npg-9-425-2002 |
| Journal | Nonlinear Processes in Geophysics, 9, 2002, 425-433 |
Abstract
Local scaling and singularity properties of solar wind and geomagnetic time series were analysed using H\"{o}lder exponents $\alpha$. It was shown that in analysed cases due to multifractality of fluctuations $\alpha$ 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 improvement of predictions of geomagnetic activity. To this end layered backpropagation artificial neural network model with feedback connection was used for the study of the solar wind - magnetosphere coupling and prediction of geomagnetic $D_{st}$ index. The solar wind input was taken from principal component analysis of interplanetary magnetic field, proton density and bulk velocity. Superior network performance was achieved in cases when the information on local H\"{o}lder exponents was added to the input layer.
{
"annotation_id": "cdc33ad4-79c3-4583-98ab-2dc2f84bfdbf",
"date_created": "2026-03-02T18:00:53.241000Z",
"date_modified": "2026-03-02T18:00:53.241000Z",
"file_hash": "8011f466d12e4136ee702838527c7312b1f0ea06d10350fa7011ee90c0bf2728",
"private": false,
"record": {
"abstract": "Local scaling and singularity properties of solar wind and geomagnetic time\nseries were analysed using H\\\"{o}lder exponents $\\alpha$. It was shown that in\nanalysed cases due to multifractality of fluctuations $\\alpha$ changes from\npoint to point. We argued there exists a peculiar interplay between regularity\n/ irregularity and amplitude characteristics of fluctuations which could be\nexploited for improvement of predictions of geomagnetic activity. To this end\nlayered backpropagation artificial neural network model with feedback\nconnection was used for the study of the solar wind - magnetosphere coupling\nand prediction of geomagnetic $D_{st}$ index. The solar wind input was taken\nfrom principal component analysis of interplanetary magnetic field, proton\ndensity and bulk velocity. Superior network performance was achieved in cases\nwhen the information on local H\\\"{o}lder exponents was added to the input\nlayer.",
"arxiv_id": "physics/0411062",
"authors": [
"Z. V\u00f6r\u00f6s",
"D. Jankovi\u010dov\u00e1"
],
"categories": [
"physics.comp-ph",
"physics.space-ph"
],
"doi": "10.5194/npg-9-425-2002",
"journal_ref": "Nonlinear Processes in Geophysics, 9, 2002, 425-433",
"title": "Neural network prediction of geomagnetic activity: a method using local H\\\"{o}lder exponents",
"url": "https://arxiv.org/abs/physics/0411062"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "0fb01e43-c135-4a38-95ba-096b1a9cce89",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}