DETERMINATION OF MUONS MULTIPLICITY IN DECOR EVENTS USING DEEP MACHINE LEARNING METHODS
https://doi.org/10.56304/S2079562925010154
EDN: LQFITA
Abstract
The DECOR coordinate-track detector is designed for registration of charged cosmic ray particles at large zenith angles. At the moment, analyses of the installation measurements are performed manually, which affects the performance. Application of deep machine learning methods allows to automate the processing process and increase the sample of processed data. The artificial neural network (ANN) architectures described in the paper have shown high accuracy in counting the multiplicity of muons in the data of the DECOR facility. Estimates of ANN performance on events with different muon multiplicity are given: for the number of particles 5–6 the accuracy was 1 track, and for more than 100 particles – 7 tracks.
About the Authors
E. A. Miroshnichenko
National Research Nuclear University MEPhI (Moscow Engineering Physics Institute)
Russian Federation
Russian Federation
V. S. Vorobev
National Research Nuclear University MEPhI (Moscow Engineering Physics Institute)
Russian Federation
Russian Federation
References
1. Barbashina N.S. et al. // Instrum. Exp. Tech. 2000. V. 43 (6). P. 743–746.
2. Yashin I.I. et al. // J. Instrum. 2021. V. 16. P. T08014.
3. Bogdanov A.G. et al. // Phys. At. Nucl. 2010. V. 73. P. 1852–1869.
4. Bogdanov A.G. et al. // Astropart. Phys. 2018. V. 98. P. 13–20.
5. Kang D., Arteaga-Velázquez J.C. Bertaina M., et al. // Proc. Sci. 2023. V. 444. P. 307. https://doi.org/10.22323/1.444.0307
6. Petrukhin A.A. // Phys.-Usp. 2015. V. 58 (5). P. 486–494.
7. Kindin V.V. et al. // Instrum. Exp. Tech. 2018. V. 61 (5). P. 649–657.
8. Нейронная сеть. Большая российская энциклопедия [в 35 т.] гл. ред. Осипов Ю.С. 2017. Москва: Большая российская энциклопедия.
9. Droz D. et al. // J. Instrum. 2021. V. 16. P. P07036.
10. Aurisano A. et al. // J. Instrum. 2016. V. 11. P. P09001.
11. Martinez J. Arjona et al. // Eur. Phys. J. Plus. 2019. V. 134. P. 333.
12. Воробьев В.С. и др. // Ядерн. физ. инжинир. 2021. Т. 12 (1). С. 26−31. [Vorob’ev V.S. et al. // Phys. At. Nucl. 2021. V. 84 (9). P. 1567–1571].
13. Минский М., Пейперт С. Персептроны. (Пер. с . англ.). 1971. Москва: Мир.
14. Горбачевская Е.Н. // Вестн. Волжского унив. им. В. Н. Татищева. 2012. № 2. С. 128.
15. Adam. https://pytorch.org/docs/stable/generated/torch.optim.Adam.html.
16. MSELoss. https://pytorch.org/docs/stable/generated/torch.nn.MSELoss.html.
17. Neural networks, fundamental principles of operation, diversity and topology. https://habr.com.
Review
For citations:
Miroshnichenko E.A., Vorobev V.S. DETERMINATION OF MUONS MULTIPLICITY IN DECOR EVENTS USING DEEP MACHINE LEARNING METHODS. Nuclear Physics and Engineering. 2025;16(5):617-622. (In Russ.) https://doi.org/10.56304/S2079562925010154. EDN: LQFITA
Views:
17
JATS XML
Email this article 