<article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" article-type="research-article" dtd-version="1.2" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">Geomagnetism and Aeronomy</journal-id><journal-title-group><journal-title>Geomagnetism and Aeronomy</journal-title></journal-title-group><issn publication-format="print">0016-7940</issn><issn publication-format="electronic">3034-5022</issn><publisher><publisher-name>Russian Academy of Science</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.31857/S0016794022600612</article-id><title-group><article-title>Predicting the Functional Dependence of the Sunspot Number
in the Solar Activity Cycle Based on Elman Artificial Neural Network</article-title><trans-title-group xml:lang="ru"><trans-title>Прогнозирование функциональной зависимости числа солнечных пятен в цикле солнечной активности на основе ИНС Элмана</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid"></contrib-id><name-alternatives><name xml:lang="en"><surname>Krasheninnikov</surname><given-names>I. V.</given-names></name><name xml:lang="ru"><surname>Крашенинников</surname><given-names>И. В. </given-names></name></name-alternatives><email>krasheninnikov_i_v_noemail@ras.ru</email><xref ref-type="aff" rid="aff-1"></xref><xref ref-type="aff" rid="aff-2"></xref></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid"></contrib-id><name-alternatives><name xml:lang="en"><surname>Chumakov</surname><given-names>S. O.</given-names></name><name xml:lang="ru"><surname>Чумаков</surname><given-names>С. О. </given-names></name></name-alternatives><email>chumakov_s_o_noemail@ras.ru</email><xref ref-type="aff" rid="aff-3"></xref></contrib></contrib-group><aff-alternatives id="aff-1"><aff><institution xml:lang="ru">Институт земного магнетизма, ионосферы и распространения радиоволн им. Н.В. Пушкова РАН (ИЗМИРАН)</institution><institution xml:lang="en">Pushkov Institute of Terrestrial Magnetism, the Ionosphere, and Radio Wave Propagation, Russian Academy of Sciences (IZMIRAN)</institution></aff></aff-alternatives><aff-alternatives id="aff-2"><aff><institution xml:lang="ru"></institution><institution xml:lang="en"></institution></aff></aff-alternatives><aff-alternatives id="aff-3"><aff><institution xml:lang="ru">Институт земного магнетизма, ионосферы и распространения радиоволн им. Н.В. Пушкова РАН (ИЗМИРАН)</institution><institution xml:lang="en">Pushkov Institute of Terrestrial Magnetism, the Ionosphere, and Radio Wave Propagation, Russian Academy of Sciences (IZMIRAN)</institution></aff></aff-alternatives><pub-date date-type="pub" iso-8601-date="2023-02-01" publication-format="electronic"><day>01</day><month>02</month><year>2023</year></pub-date><volume>63</volume><issue>2</issue><fpage>247</fpage><lpage>256</lpage><abstract xml:lang="en"><p>The possibility of predicting the function of the time dependence of the sunspot number (SSN) in the solar activity cycle is analyzed based on the application of the Elman artificial neural network platform to the historical series of observational data. A method for normalizing the initial data for preliminary training of the ANN algorithm is proposed, in which a sequence of virtual idealized cycles is constructed using scaled duration coefficients and the amplitude of solar cycles. The correctness of the method is analyzed in a numerical experiment based on modeling the time series of sunspots. The intervals of changing the adaptable parameters in the ANN operation are estimated and a mathematical criterion for choosing a solution is proposed. The significant asymmetry of its ascending and descending branches is a characteristic property of the constructed functional dependence of the sunspot number cycle. A forecast of the time course for the current 25th cycle of solar activity is presented and its correctness is discussed in comparison with other forecast results and the available data of solar activity status monitoring</p></abstract><trans-abstract xml:lang="ru"><p>Анализируется возможность прогнозирования функции временнóй зависимости числа солнечных пятен (SSN) в цикле солнечной активности на основе применения платформы искусственной нейронной сети Элмана к историческому ряду данных наблюдений. Предложен метод нормализации исходных данных для предварительного обучения алгоритма ИНС, в котором строится последовательность виртуальных идеализированных циклов, используя масштабируемые коэффициенты по длительности и значения максимумов в солнечных циклах. Корректность метода анализируется в численном эксперименте, основанном на моделировании временнóго ряда солнечных пятен. Оценены интервалы изменения адаптируемых параметров в работе ИНС и предложен математический критерий для выбора решения. Характерным свойством построенной функциональной зависимости в цикле числа солнечных пятен является значительная асимметрия ее восходящей и спадающей ветвей. Представлен прогноз временнóго хода на текущий 25-й цикл солнечной активности и обсуждается его корректность в сравнении с другими результатами прогнозирования и имеющимися данными обсерваторских наблюдений.</p></trans-abstract></article-meta></front><body></body><back><ref-list><ref id="B1"><label>B1</label><citation-alternatives><mixed-citation xml:lang="ru">– Бархатов Н.А., Королёв А.В., Пономарев С.М., Сахаров С.Ю. Долгосрочное прогнозирование индексов солнечной активности методом искусственных нейронных сетей // Изв. Вузов. Радиофизика. Т. XLIV. № 9. С. 806–814. 2001.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B2"><label>B2</label><citation-alternatives><mixed-citation xml:lang="ru">– Головко В.А. Нейронные сети: обучение, организация и применение // Минск: ИПРЖР. 255 с. 2001.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B3"><label>B3</label><citation-alternatives><mixed-citation xml:lang="ru">– Головко В.А., Краснопрошин В.В. Нейросетевые технологии обработки данных // Минск: БГУ. 263 с. 2017.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B4"><label>B4</label><citation-alternatives><mixed-citation xml:lang="ru">– Крашенинников И.В., Чумаков С.О. Метод ИНС в задаче долгосрочного прогнозирования индексов солнечной активности // Физика плазмы в солнечной системе. 16-я ежегодная конференция. М.: ИКИ РАН. С. 264. 2021.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B5"><label>B5</label><citation-alternatives><mixed-citation xml:lang="ru">– Benson B., Pan W.D., Prasad A., Gary G.A., Hu Q. Forecasting Solar Cycle 25 Using Deep Neural Networks // Sol. Phys. V. 295(65). 2020. https://doi.org/10.1007/s11207-020-01634-y</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B6"><label>B6</label><citation-alternatives><mixed-citation xml:lang="ru">– Bothmer V., Daglis I.A. Space weather: physics and effects. Springer, Dordrecht. 476 p. 2007.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B7"><label>B7</label><citation-alternatives><mixed-citation xml:lang="ru">– Elman J.L. Finding structure in time // Cogn. Sci. V. 14. P. 179–211. 1990. https://doi.org/10.1207/s15516709cog1402_1</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B8"><label>B8</label><citation-alternatives><mixed-citation xml:lang="ru">– Fessant F., Bengio S., Collobert D. On the prediction of solar activity using different neural network models // Ann. Geophys. V. 14(1). P. 20–26. 1996.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B9"><label>B9</label><citation-alternatives><mixed-citation xml:lang="ru">– Hathaway D.H., Wilson R.M., Reichmann E.J. The shape of the sunspot cycle // Sol. Phys. V. 151. P. 177–190. 1994. https://doi.org/10.1007/BF00654090</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B10"><label>B10</label><citation-alternatives><mixed-citation xml:lang="ru">– Hathaway D.H., Wilson R.M. Geomagnetic activity indicates large amplitude for sunspot cycle 24 // Geophys. Res. Lett. 33(L18101). 2006. https://doi.org/10.1029/2006GL027053</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B11"><label>B11</label><citation-alternatives><mixed-citation xml:lang="ru">– Hathaway D.H. The Solar Cycle. // Living Rev. Sol. Phys. 2015. 12:4. https://doi/org/https://doi.org/10.1007/lrsp-2015-4</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B12"><label>B12</label><citation-alternatives><mixed-citation xml:lang="ru">– Macpherson K. Neural network computation techniques applied to solar activity prediction // Adv. Space Res. V. 13. № 9. P. 375–450. 1993. https://doi.org/10.1016/0273-1177(93)90518-G</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B13"><label>B13</label><citation-alternatives><mixed-citation xml:lang="ru">– Nandy D., Martens P.C.H., Obridko V., Dash S., Georgieva K. Solar evolution and extrema: current state of understanding of long-term solar variability and its planetary impacts // Space Sci. Rev. V. 217. № 3. 2021. https://doi.org/10.1007/s11214-021-00799-7</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B14"><label>B14</label><citation-alternatives><mixed-citation xml:lang="ru">– Bondar T.N., Rotanova N.M., Obridko V.N. Stochastic autoregression modeling and forecasting of the Wolf-number time series // Astron. Rep. V. 39. P. 115–122. 1995.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B15"><label>B15</label><citation-alternatives><mixed-citation xml:lang="ru">– Pala Z., Atici R. Forecasting Sunspot Time Series Using Deep Learning Methods // Sol. Phys. V. 294(50). 2019. https://doi.org/10.1007/s11207-019-1434-62019</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B16"><label>B16</label><citation-alternatives><mixed-citation xml:lang="ru">– Pesnell W.D. Solar cycle predictions (invited review) // Sol. Phys. V. 281. № 1. P. 507–532. 2012. https://doi.org/10.1007/s11207-012-9997-5</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B17"><label>B17</label><citation-alternatives><mixed-citation xml:lang="ru">– Podladchikova T., Van der Linden, Kalman R.A Filter Technique for Improving Medium-Term Predictions of the Sunspot Number // Sol. Phys. V. 277. P. 397–416. 2012. https://doi.org/10.1007/s11207-011-9899-y</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B18"><label>B18</label><citation-alternatives><mixed-citation xml:lang="ru">– Sarp V., Kilcik A., Yurchyshyn V., Rozelot J.P., Ozguc A. Prediction of solar cycle 25: a non-linear approach // MNRA-S. V. 481. P. 2981–2985. 2018. https://doi.org/10.1093/mnras/sty2470</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B19"><label>B19</label><citation-alternatives><mixed-citation xml:lang="ru">– Sello S. Solar cycle forecasting: a nonlinear dynamics approach // A &amp;amp; A. V. 377. P. 312–320. 2001.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B20"><label>B20</label><citation-alternatives><mixed-citation xml:lang="ru">– Thompson R.J. A Technique for Predicting the Amplitude of the Solar Cycle // Sol. Phys. V. 148. №. 2. P. 383–388. 1993.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B21"><label>B21</label><citation-alternatives><mixed-citation xml:lang="ru">– Wang Y.M., Sheeley N.R. Understanding the geomagnetic precursor of the solar cycle // Astrophys. J. V. 694. P. L11–L15. 2009.</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref><ref id="B22"><label>B22</label><citation-alternatives><mixed-citation xml:lang="ru">– Willamo T., Hackman T., Lehtinen J.J. et al. Shapes of stellar activity cycles // A &amp;amp; A. V. 638. A69. 2020. https://doi.org/10.1051/0004-6361/202037666</mixed-citation><mixed-citation xml:lang="en"></mixed-citation></citation-alternatives></ref></ref-list></back></article>