面向 SPHERE-3 实验的卷积神经网络在广泛空气喷雾分离中的应用
天体物理仪器与方法
2025-10-16 v2
摘要
正在开发新的 SPHERE-3 望远镜，用于研究 5--1000 PeV 能量范围内的宇宙射线光谱和质量组成。使用反射 Cherenkov 光方法注册广泛空气喷雾，需要精准的触发系统以准确分离由星光和大气荧光反射来自夜空的背景事件。本文介绍了卷积神经网络用于对来自探测器 Monte Carlo 模拟图像分类的成果。探测器响应模拟包括穿透光学系统的光子追踪、硅光子二极管运算以及电子学响应和数位化过程。结果与 SPHERE-2 触发系统性能进行了比较。
引用
@article{arxiv.2410.01781,
title = {Application of convolutional neural networks for extensive air shower separation in the SPHERE-3 experiment},
author = {E. L. Entina and D. A. Podgrudkov and C. G. Azra and E. A. Bonvech and O. V. Cherkesova and D. V. Chernov and V. I. Galkin and V. A. Ivanov and T. A. Kolodkin and N. O. Ovcharenko and T. M. Roganova and M. D. Ziva},
journal= {arXiv preprint arXiv:2410.01781},
year = {2025}
}
备注
7 pages, 5 figures Published as a conference paper at DLCP2024, June 19-21, 2024, Moscow, Russia. https://dlcp2024.sinp.msu.ru Submitted to Moscow University Physics Bulletin