Solar-Terrestrial Physics Solar-Terrestrial Physics Solar-Terrestrial Physics 2500-0535 56980 10.12737/stp-101202411 Results of current research Results of current research Results of current research Recognition of geomagnetic storms from time series of matrix observations with the muon hodoscope URAGAN using neural networks of deep learning Recognition of geomagnetic storms from time series of matrix observations with the muon hodoscope URAGAN using neural networks of deep learning Гетманов Виктор Григорьевич Getmanov Viktor Grigoryevich vgetm2015@yandex.ru

доктор технических наук;

doctor of technical sciences;

 https://orcid.org/0000-0002-4874-7475 Гвишиани Алексей Джерменович Gvishiani Alexei D. rjes@wdcb.ru

доктор физико-математических наук;

doctor of physical and mathematical sciences;

 https://orcid.org/0000-0002-6476-9471 Соловьёв Анатолий Александрович Soloviev Anatoly Aleksandrovich rjes@wdcb.ru

доктор физико-математических наук;

doctor of physical and mathematical sciences;

 Зайцев Константин Сергеевич Zajtsev Konstantin Sergeevich kszajtsev@mephi.ru

доктор технических наук;

doctor of technical sciences;

 Дунаев Максим Евгеньевич Dunaev Maksim Evgenyevich max.dunaev@mail.ru Ехлаков Эдуард Владимирович Ehlakov Eduard Vladimirovich fruha1980@gmail.com Геофизический центр РАН Москва Россия Geophysical Center of RAS Moscow Russian Federation Институт физики Земли им. О.Ю. Шмидта РАН Москва Россия Schmidt Institute of Physics of the Earth, RAS Moscow Russian Federation Геофизический центр Российской Академии Наук Москва Россия Geophysical Center of the Russian Academy of Sciences Moscow Russian Federation Институт физики Земли им. О.Ю. Шмидта Российской академии наук Москва Россия Schmidt Institute of the Physics of the Earth Russian Academy of Sciencies Moscow Russian Federation Геофизический центр РАН Москва Россия Geophysical Center RAS Moscow Russian Federation Институт физики Земли им. О.Ю. Шмидта РАН Москва Россия Schmidt Institute of Physics of the Earth, RAS Moscow Russian Federation Национальный исследовательский ядерный универси-тет МИФИ Москва Россия National Reasearch Nuclear University MEPHI Moscow Russian Federation Национальный исследовательский ядерный универси-тет МИФИ Москва Россия National Reasearch Nuclear University MEPHI Moscow Russian Federation Национальный исследовательский ядерный универси-тет МИФИ Москва Россия National Reasearch Nuclear University MEPHI Moscow Russian Federation 26 03 2024 26 03 2024 10 1 76 83 28 01 2023 16 02 2024 https://naukaru.ru/en/nauka/article/56980/view

We solve the problem of recognizing geomagnetic storms from matrix time series of 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 Yᴅ₀=–45 nT we obtain acceptable probabilities of correct and false recognitions, which amount to β=0.8212 and α=0.0047. We confirm the effectiveness and prospects of the proposed neural network approach.

We solve the problem of recognizing geomagnetic storms from matrix time series of 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 Yᴅ₀=–45 nT we obtain acceptable probabilities of correct and false recognitions, which amount to β=0.8212 and α=0.0047. We confirm the effectiveness and prospects of the proposed neural network approach.

 geomagnetic storms recognition neural networks probabilities of correct and false recognitions matrix observations muon hodoscope geomagnetic storms recognition neural networks probabilities of correct and false recognitions matrix observations muon hodoscope The work was supported by a project part of the Government Assignment from the Geophysical Center RAS and Sсhmidt Institute of Physics of the Earth RAS, approved by the Ministry of Science and Higher Education of the Russian Federation The work was supported by a project part of the Government Assignment from the Geophysical Center RAS and Schmidt Institute of Physics of the Earth RAS, approved by the Ministry of Science and Higher Education of the Russian Federation

Astapov I.I., Barbashina N.S., Borog V.V., Dmitrieva A.N, Shulzhenko I.A., Shutenko V.V., et al. Muon Diagnostics of Earth’s Magnetosphere and Atmosphere. Moscow, MEPhI, 2014. 132 p. (In Russian). Astapov I.I., Barbashina N.S., Borog V.V., Dmitrieva A.N, Shulzhenko I.A., Shutenko V.V., et al. Muon Diagnostics of Earth’s Magnetosphere and Atmosphere. Moscow, MEPhI, 2014. 132 p. (In Russian). Ba J.L., Kingma D.P. Adam: A method for stochastic optimization. 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings. 2015, pp. 1-15. Ba J.L., Kingma D.P. Adam: A method for stochastic optimization. 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings. 2015, pp. 1-15. Barkhatov N.A. Artificial Neural Networks in Solar-Terrestrial Physics. Nizhny Novgorod, Povolzh’e, 2010. 707 p. (In Russian). Barkhatov N.A. Artificial Neural Networks in Solar-Terrestrial Physics. Nizhny Novgorod, Povolzh’e, 2010. 707 p. (In Russian). Belov A.V., Gvishiani A.D., Getmanov V.G., Kovylyaeva A.A., Solovyev A.A., Chinkin V.E., et al. Identification of geomagnetic storms based on neural model estimates of Dst indices. J. Computer and Systems Sciences International. 2022, no. 1, pp. 56-66. (In Russian). Belov A.V., Gvishiani A.D., Getmanov V.G., Kovylyaeva A.A., Solovyev A.A., Chinkin V.E., et al. Identification of geomagnetic storms based on neural model estimates of Dst indices. J. Computer and Systems Sciences International. 2022, no. 1, pp. 56-66. (In Russian). Berngardt O.N. 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 Solar-Terr. Phys. 2022, vol. 8, iss. 4, pp. 63-72. Berngardt O.N. 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 Solar-Terr. Phys. 2022, vol. 8, iss. 4, pp. 63-72. Chen X., Yu R., Ullah S., Wu D., Zhiqiang Li Zh., Li Q., et al. A novel loss function of deep learning in wind speed forecasting. Energy. 2022, vol. 238, p. 121808. Chen X., Yu R., Ullah S., Wu D., Zhiqiang Li Zh., Li Q., et al. A novel loss function of deep learning in wind speed forecasting. Energy. 2022, vol. 238, p. 121808. Chinkin V.E., Astapov I.I., Gvishiani A.D., Getmanov V.G., Dmitrieva A.N., Dobrovolsky M.N., et al. Method for the identification of heliospheric anomalies based on the functions of the characteristic deviations for the observation matrices of the muon hodoscope. Physics of Atomic Nuclei. 2019, vol. 82, no. 6, pp. 924-928. Chinkin V.E., Astapov I.I., Gvishiani A.D., Getmanov V.G., Dmitrieva A.N., Dobrovolsky M.N., et al. Method for the identification of heliospheric anomalies based on the functions of the characteristic deviations for the observation matrices of the muon hodoscope. Physics of Atomic Nuclei. 2019, vol. 82, no. 6, pp. 924-928. Dolenko S.A., Orlov Yu.V., Persianinov I.G., Shugai Ju.S. Neural network algorithm for events forecasting and its application to space physics data. Lecture Notes in Computer Science. 2005, vol. 3697, pp. 527-532. Dolenko S.A., Orlov Yu.V., Persianinov I.G., Shugai Ju.S. Neural network algorithm for events forecasting and its application to space physics data. Lecture Notes in Computer Science. 2005, vol. 3697, pp. 527-532. Getmanov V.G., Chinkin V.E., Gvishiani A.D., Dobrovolsky M.N., Sidorov R.G., Soloviev A.A., et al. Application of indicator matrices for the recognition of local anisotropies of muon fluxes in time series of matrix observations of the URAGAN hodoscope. Pattern Recognition and Image Analysis: Adv. in Mathematical Theory and Applications. 2022a, vol. 32, no. 3, pp. 717-728. Getmanov V.G., Chinkin V.E., Gvishiani A.D., Dobrovolsky M.N., Sidorov R.G., Soloviev A.A., et al. Application of indicator matrices for the recognition of local anisotropies of muon fluxes in time series of matrix observations of the URAGAN hodoscope. Pattern Recognition and Image Analysis: Adv. in Mathematical Theory and Applications. 2022a, vol. 32, no. 3, pp. 717-728. Getmanov V.G., Chinkin V.E., Gvishiani A.D., Sidorov R.V., Gvishiani A.D., Dobrovolskii M.N., Solov’ev A.A., et al. Geomagnetic storm prediction based on the neural network digital processing of joint observations of the URAGAN muon hodoscope and neutron monitor stations. Geomangetism and Aeronomy. 2022b, vol. 62, iss. 4, pp. 388-398. Getmanov V.G., Chinkin V.E., Gvishiani A.D., Sidorov R.V., Gvishiani A.D., Dobrovolskii M.N., Solov’ev A.A., et al. Geomagnetic storm prediction based on the neural network digital processing of joint observations of the URAGAN muon hodoscope and neutron monitor stations. Geomangetism and Aeronomy. 2022b, vol. 62, iss. 4, pp. 388-398. Gruet M.A., Chandorkar M., Sicard A., Camporeale E. Multiplehour-ahead forecast of the Dst-index using a combination of long short-term memory neural network and Gaussian process. Space Weather. 2018, vol. 16, iss. 11, pp. 1882-1896. DOI: 10.1029/2018SW001898. Gruet M.A., Chandorkar M., Sicard A., Camporeale E. Multiplehour-ahead forecast of the Dst-index using a combination of long short-term memory neural network and Gaussian process. Space Weather. 2018, vol. 16, iss. 11, pp. 1882-1896. DOI: 10.1029/2018SW001898. Efitorov A.O., Myagkova I.N., Shirokii V.P., Dolenko S.A. The prediction of the Dst-index based on machine learning methods. Cosmic Res. 2018, vol. 56, iss. 6, pp. 434-441. Efitorov A.O., Myagkova I.N., Shirokii V.P., Dolenko S.A. The prediction of the Dst-index based on machine learning methods. Cosmic Res. 2018, vol. 56, iss. 6, pp. 434-441. Lundstredt H. Geomagnetic storm predictions from solar wind data with the use of dynamic neural networks. J. Geophys. Res. 1997, vol. 102, no. A7, pp.14,255-14,268. Lundstredt H. Geomagnetic storm predictions from solar wind data with the use of dynamic neural networks. J. Geophys. Res. 1997, vol. 102, no. A7, pp.14,255-14,268. Murzin V.S. Astrophysics of Cosmic Rays. Moscow, 2007. 488 p. Murzin V.S. Astrophysics of Cosmic Rays. Moscow, 2007. 488 p. Myagkova I.N., Shirokii V.R., Vladimirov R.D., Barinov O.G., Dolenko S.A. Prediction of the Dst-index using adaptive мethods. Russian Meteorology and Hydrology. Allerton Press Inc. 2021, vol. 46, no. 3, pp. 157-162. Myagkova I.N., Shirokii V.R., Vladimirov R.D., Barinov O.G., Dolenko S.A. Prediction of the Dst-index using adaptive methods. Russian Meteorology and Hydrology. Allerton Press Inc. 2021, vol. 46, no. 3, pp. 157-162. Pallochia G., Amota E., Consolini G., Marcucci M.F., Bertello I. Geomagnetics Dst-index forecast based on IMF data only. Ann. Geophys. 2006, vol. 24, pp. 989-999. Pallochia G., Amota E., Consolini G., Marcucci M.F., Bertello I. Geomagnetics Dst-index forecast based on IMF data only. Ann. Geophys. 2006, vol. 24, pp. 989-999. Stepanova M.V., Perez P. Autoprediction of Dst-index using neural network techniques and relationship to the auroral geomagnetics indices. Geofisica International. 2000, vol. 39, no. 1, pp. 143-146. Stepanova M.V., Perez P. Autoprediction of Dst-index using neural network techniques and relationship to the auroral geomagnetics indices. Geofisica International. 2000, vol. 39, no. 1, pp. 143-146. Suigiura M. Hourly values of equatorial Dst for the IGY. Ann. Int. Geophys. Year. Pergamon Press, Oxford. 1964, vol. 35, pp. 9-45. Suigiura M. Hourly values of equatorial Dst for the IGY. Ann. Int. Geophys. Year. Pergamon Press, Oxford. 1964, vol. 35, pp. 9-45. Yashin I.I., Astapov I.I., Barbashina N.S., Bogod V.V., Chernov D.V., Dmitieva A.N., et al. Real-time data of muon hodoscope URAGAN. Adv. Space Res. 2015, vol. 56, iss. 12, pp. 2693-2705. Yashin I.I., Astapov I.I., Barbashina N.S., Bogod V.V., Chernov D.V., Dmitieva A.N., et al. Real-time data of muon hodoscope URAGAN. Adv. Space Res. 2015, vol. 56, iss. 12, pp. 2693-2705. URL: https://wdc.kugi.kyoto-u.ac.jp (accessed December 23, 2023). URL: https://wdc.kugi.kyoto-u.ac.jp (accessed December 23, 2023). URL: http://www.nevod.mephi.ru/ (accessed March 13, 2024). URL: http://www.nevod.mephi.ru/ (accessed March 13, 2024). URL: https://it.mephi.ru/hpc/perfomance (accessed December 23, 2023). URL: https://it.mephi.ru/hpc/perfomance (accessed December 23, 2023). URL: https://arxiv.org/abs/1905.11946/ (accessed December 23, 2023). URL: https://arxiv.org/abs/1905.11946/ (accessed December 23, 2023). URL: https://arxiv.org/abs/1409.1556v6 (accessed December 23, 2023). URL: https://arxiv.org/abs/1409.1556v6 (accessed December 23, 2023). URL: https://arxiv.org/abs/1608.06993v5 (accessed December 23, 2023). URL: https://arxiv.org/abs/1608.06993v5 (accessed December 23, 2023). URL: https://arxiv.org/abs/1512.00567v3 (accessed March 13, 2024). URL: https://arxiv.org/abs/1512.00567v3 (accessed March 13, 2024). URL: https://arxiv.org/pdf/2107.07699.pdf (accessed December 23, 2023). URL: https://arxiv.org/pdf/2107.07699.pdf (accessed December 23, 2023). URL: https://arxiv.org/abs/1512.03385 (accessed December 23, 2023). URL: https://arxiv.org/abs/1512.03385 (accessed December 23, 2023). URL: https://arxiv.org/pdf/1412.6980.pdf (accessed December 23, 2023). URL: https://arxiv.org/pdf/1412.6980.pdf (accessed December 23, 2023). URL: https://pytorch.org/docs/stable/generated/torch.optim. Adam.html (accessed December 23, 2023). URL: https://pytorch.org/docs/stable/generated/torch.optim. Adam.html (accessed December 23, 2023). URL: https://learndatasci/glossary/binary-classification (accessed October 23, 2023). URL: https://learndatasci/glossary/binary-classification (accessed October 23, 2023). URL: https://www.learndatasci.com/glossary/binary-classification/ (accessed March 13, 2024). URL: https://www.learndatasci.com/glossary/binary-classification/ (accessed March 13, 2024). URL: https://helenkapatsa.ru/blogpost/otsienka-f1 (accessed March 13, 2024). URL: https://helenkapatsa.ru/blogpost/otsienka-f1 (accessed March 13, 2024). URL: https://www.izmiran.ru/ionosphere/weather/storm/ (accessed March 13, 2024). URL: https://www.izmiran.ru/ionosphere/weather/storm/ (accessed March 13, 2024).