DETECTION OF SOLAR FLARES FROM IONOSPHERIC DATA USING DEEP LEARNING

АУТОР(И) / AUTHOR(S): Martin Sarnovský , Peter Butka , Peter Bednár , Aleksandra Nina , Vladimir Srećković , Luka Č. Popović , Aleksandra Kolarski , Filip Arnaut

Download Full Pdf   

DOI:  10.46793/Geoscira25.107S

САЖЕТАК / ABSTRACT:

This work addresses the detection of solar flares based on data from low-frequency radio wave signal amplitudes. The work is a cooperation between the Technical University of Košice and the Institute of Physics at the University of Belgrade. The primary goal is to explore the provided data and to look for patterns and connections in the ionosphere’s behavior during normal conditions and solar flares. To achieve this goal, an Autoencoder neural network is implemented in Python. Data exploration utilizes various types of data visualizations, including box plots, violin plots, and histograms. The data used included VLF signal amplitudes received in Belgrade and information on solar flares from NASA GOES satellites. Working with an extremely unbalanced dataset, low class distinctiveness, and noisy data presented significant challenges. Despite these difficulties, the final Autoencoder model, trained on non-flare data and using two key features derived from signal statistics, was able to capture 65% of real flare occurrences, although it produced many false positives. The results of this research work demonstrate that subtle differences in the signals related to flares exist and provide a basis for future efforts to improve detection.

КЉУЧНЕ РЕЧИ / KEYWORDS:

Solar flares; Machine learning; Deep learning; Neural networks

ПРОЈЕКАТ / ACKNOWLEDGEMENT:

This work was supported by the Slovak APVV agency Slovak-Serbian project SK-SRB-23-0029, and the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Serbian-Slovak project 337-00-3/2024-05/11). The authors acknowledge funding provided by the Institute of Physics Belgrade and the Astronomical Observatory (the contract 451-03-136/2025-03/200002) through the grants by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia.

ЛИТЕРАТУРА / REFERENCES:

• Azevedo, A., and Santos, M. F. (2008). KDD, SEMMA and CRISP-DM: a parallel overview. IADIS European Conference Data Mining, Amsterdam, Netherlands, 182.
• Berahmand, K., Daneshfar, F., Salehi, E. S. et al. (2024). Autoencoders and their applications in machine learning: a survey. Artif. Intell. Rev. 57, 28.
• Mishra, C., and Gupta, D. L. (2017). Deep machine learning and neural networks: An overview. IAES Int. J. Artif. Intell., 6(2), 66.
• Nina, A., Čadež, V. M., Popović, L. Č. et al. (2017). Diagnostics of plasma in the ionospheric D-region: detection and study of different ionospheric disturbance types. Eur. Phys. J. D., 71, 189.
• Pavlík, V. (2024). Spracovanie ionosférických dát z rádiových meraní pomocou hlbokého učenia. Technical University of Kosice, Kosice, Slovakia
• Šulić, D. M., Srećković, V. A., and Mihajlov, A. A. (2016). A study of VLF signals variations associated with the changes of ionization level in the d-region in consequence of solar conditions. Adv. Space Res., 57(4), 102943.