RECOGNITION OF GEOMAGNETIC STORMS FROM TIME SERIES OF MATRIX OBSERVATIONS WITH THE MUON HODOSCOPE URAGAN USING NEURAL NETWORKS OF DEEP LEARNING

被引：0

作者：

Getmanov, V. G. ^{[1
,2
]}

Zajtsev, K. S. ^{[3
]}

Gvishiani, A. D. ^{[1
,2
]}

Dunaev, M. E. ^{[3
]}

Soloviev, A. A. ^{[1
,2
]}

Ehlakov, E., V ^{[3
]}

机构：

[1] Geophys Ctr RAS, Moscow, Russia

[2] Schmidt Inst Phys Earth RAS, Moscow, Russia

[3] Natl Reasearch Nucl Univ MEPhI, Moscow, Russia

来源：

SOLAR-TERRESTRIAL PHYSICS | 2024年 / 10卷 / 01期

关键词：

geomagnetic storms; recognition; neural networks; probabilities of correct and false recognitions; matrix observations; muon hodoscope; PREDICTION;

D O I：

10.12737/stp-101202411

中图分类号：

P3 [地球物理学]; P59 [地球化学];

学科分类号：

0708 ; 070902 ;

摘要：

We solve the problem of recognizing geomagnetic storms from time series of matrix observations with the URAGAN muon hodoscope, using deep learning neural networks. A variant of the neural network software module is selected and its parameters are determined. Geomagnetic storms are recognized using binary classification procedures; a decision -making rule is formed. We estimate probabilities of correct and false recognitions. The recognition of geomagnetic storms is experimentally studied; for the assigned Dst threshold Y D 0 =-45 nT we obtain acceptable probabilities of cor rect and false recognitions, which amount to beta=0.8212 and alpha=0.0047. We confirm the effectiveness and prospects of the proposed neural network approach.

引用

页码：76 / 83

页数：8

共 33 条

[1]

[Anonymous], About us

[2]

Astapov I.I., 2014, Muon Diagnostics of Earth's Magnetosphere and Atmosphere

[3]

Ba J., 2015, P INT C LEARN REPR

[4]

Barkhatov N.A., 2010, ARTIFICIAL NEURAL NE

[5]

Belov A.V., 2022, J. Computer and Systems Sciences International, P56

[6] THE FIRST COMPARATIVE ANALYSIS OF METEOR ECHO AND SPORADIC SCATTERING IDENTIFIED BY A SELF-LEARNING NEURAL NETWORK IN EKB AND MAGW ISTP SB RAS RADAR DATA [J].

Berngardt, O. I. .

SOLAR-TERRESTRIAL PHYSICS, 2022, 8 (04) :63-72

[7] A novel loss function of deep learning in wind speed forecasting [J].

Chen, Xi ;

Yu, Ruyi ;

Ullah, Sajid ;

Wu, Dianming ;

Li, Zhiqiang ;

Li, Qingli ;

Qi, Honggang ;

Liu, Jihui ;

Liu, Min ;

Zhang, Yundong .

ENERGY, 2022, 238

[8] Method for the Identification of Heliospheric Anomalies Based on the Functions of the Characteristic Deviations for the Observation Matrices of the Muon Hodoscope [J].

Chinkin, V. E. ;

Astapov, I. I. ;

Gvishiani, A. D. ;

Getmanov, V. G. ;

Dmitrieva, A. N. ;

Dobrovolsky, M. N. ;

Kovylyaeva, A. A. ;

Sidorov, R. V. ;

Soloviev, A. A. ;

Yashin, I. I. .

PHYSICS OF ATOMIC NUCLEI, 2019, 82 (06) :924-928

[9]

Dolenko SA, 2005, LECT NOTES COMPUT SC, V3697, P527

[10] The Prediction of the Dst-Index Based on Machine Learning Methods [J].

Efitorov, A. O. ;

Myagkova, I. N. ;

Shirokii, V. R. ;

Dolenko, S. A. .

COSMIC RESEARCH, 2018, 56 (06) :434-441

← 1 2 3 4 →