- PII
- S3034502225080064-1
- DOI
- 10.7868/S3034502225080064
- Publication type
- Article
- Status
- Published
- Authors
- Volume/ Edition
- Volume 65 / Issue number 8
- Pages
- 1195-1205
- Abstract
- In this paper, we present and discuss the results of using the machine learning methods to compile a catalog of solar flares observed with the Siberian Radioheliograph (SRH). The high sensitivity of the instrument, as well as the use of time profiles of the sum of correlation coefficients of antenna pairs (correlation plots) for event searching allowed us to include in the catalog the weak events, which are poorly distinguished in the time profiles of the emission flux. We proposed and tested a technique for selecting candidate events that allows the onset, maximum, and end of a solar flare to be determined by analyzing the derivative of the time profile given by a numerical function. Since the purpose of the catalog was to select wideband events, we introduced a criterion that allows automatic event selection based on simultaneous responses at several frequencies. Support Vector Machine (SVM) method was used in the test mode to assert the solar origin of the events and specify the quality of observational data. The volume of observational data obtained by the SRH in the second half of 2023 and in 2024 provides extensive material for both training and testing the model. The method was applied to the time profiles obtained in the 9–10 GHz band to divide them into “flare”, “background”, and “artifact”.
- Keywords
- солнечные вспышки микроволновое излучение методы машинного обучения каталоги и классификация
- Date of publication
- 17.06.2025
- Year of publication
- 2025
- Number of purchasers
- 0
- Views
- 26
References
- 1. Алтынцев А.Т., Лесовой С.В., Глоба М.В. и др. Многоволновый Сибирский Радиогелиограф // Солнечно-земная физика. Т. 6. № 2. С. 37–50. 2020.
- 2. Лесовой С.В., Кобец В.С. Корреляционные кривые Сибирского Радиогелиографа // Солнечно-земная физика. Т. 3. № 1. С. 17–21. 2017.
- 3. Мотык И.Д., Кашапова Л.К., Рожкова Д.В. Средние временные профили микроволнового излучения солнечных вспышек: морфология и применение // Астрономический журнал. Т. 102. № 8. С. 728–742. 2025.
- 4. Cortes C., Vapnik V. Support-Vector Networks // Machine Learning. V. 20. P. 273–297. 1995.
- 5. Hao N., Flagg L., Jayawardhana R. Detecting and Classifying Flares in High-resolution Solar Spectra with Supervised Machine Learning // The Astrophysical Journal. V. 973. P. 7. 2024.
- 6. Lysenko A.L., Ulanov M.V., Kuznetsov A.A. et al. KW-Sun: The Konus-Wind Solar Flare Database in Hard X-Ray and Soft Gamma-RayRanges // The Astrophysical Journal Supplement Series. V. 262. P. 32. 2022.
- 7. Rifkin R., Klautau A. In Defense of One-Vs-All Classification // Journal of Machine Learning Research. V. 5. P. 101–141. 2004.
- 8. Widodo A., Yang B.-S. Support vector machine in machine condition monitoring and fault diagnosis // Mechanical Systems and Signal Processing. V. 21. № 6. P. 2560–2574. 2007.
