通过增强型卷积神经网络提升液态氦中中性双β衰变探测灵敏度
仪器与探测器
2026-03-26 v1 高能物理 - 实验
数据分析、统计与概率
摘要
双相液态氦时间投影室（TPC）是一种 pioneering 检测器技术，用于搜索暗物质。除了在暗物质直接探测方面的优势之外，天然氦-136 目标也允许其搜索中性双β衰变（）过程，这一过程会违反味部数守恒，表明中微子是马约拉纳粒子。然而，这类 搜索受限于来自探测器材料的伽马射线背景。本文设计了一个增强型卷积神经网络（A-CNN）模型，从探测器数据中提取额外的事件拓扑信息。使用来自 XENONnT 的仿真和校准数据，我们的模型在保持 90% 信号接受率的同时实现了超过 60% 的背景抑制。该抑制能力提高了 XENONnT 对 Xe 搜索的预测灵敏度约 40%。A-CNN 在未来液态氦观测站（如 XLZD）的数据分析中实现，进一步提升了 Xe 的灵敏度。
引用
@article{arxiv.2603.23549,
title = {Enhancing Neutrinoless Double-Beta Decay Sensitivity of Liquid-Xenon Time Projection Chamber with Augmented Convolutional Neural Network},
author = {E. Aprile and J. Aalbers and K. Abe and M. Adrover and S. Ahmed Maouloud and L. Althueser and B. Andrieu and E. Angelino and D. Antón Martin and S. R. Armbruster and F. Arneodo and L. Baudis and M. Bazyk and L. Bellagamba and R. Biondi and A. Bismark and K. Boese and R. M. Braun and G. Bruni and G. Bruno and R. Budnik and C. Cai and C. Capelli and J. M. R. Cardoso and A. P. Cimental Chávez and A. P. Colijn and J. Conrad and J. J. Cuenca-García and V. D'Andrea and L. C. Daniel Garcia and M. P. Decowski and A. Deisting and C. Di Donato and P. Di Gangi and S. Diglio and K. Eitel and S. el Morabit and R. Elleboro and A. Elykov and A. D. Ferella and C. Ferrari and H. Fischer and T. Flehmke and M. Flierman and R. Frankel and D. Fuchs and W. Fulgione and C. Fuselli and R. Gaior and F. Gao and R. Giacomobono and F. Girard and R. Glade-Beucke and L. Grandi and J. Grigat and H. Guan and M. Guida and P. Gyorgy and R. Hammann and A. Higuera and C. Hils and L. Hoetzsch and N. F. Hood and M. Iacovacci and Y. Itow and J. Jakob and F. Joerg and Y. Kaminaga and M. Kara and S. Kazama and P. Kharbanda and M. Kobayashi and D. Koke and K. Kooshkjalali and A. Kopec and H. Landsman and R. F. Lang and L. Levinson and A. Li and I. Li and S. Li and S. Liang and Z. Liang and Y. -T. Lin and S. Lindemann and M. Lindner and K. Liu and M. Liu and F. Lombardi and J. A. M. Lopes and G. M. Lucchetti and T. Luce and Y. Ma and C. Macolino and J. Mahlstedt and F. Marignetti and T. Marrodán Undagoitia and K. Martens and J. Masbou and S. Mastroianni and V. Mazza and A. Melchiorre and J. Merz and M. Messina and A. Michel and K. Miuchi and A. Molinario and S. Moriyama and K. Morå and M. Murra and J. Müller and K. Ni and C. T. Oba Ishikawa and U. Oberlack and S. Ouahada and B. Paetsch and Y. Pan and Q. Pellegrini and R. Peres and J. Pienaar and M. Pierre and G. Plante and T. R. Pollmann and A. Prajapati and L. Principe and J. Qin and D. Ramírez García and A. Ravindran and A. Razeto and R. Singh and L. Sanchez and J. M. F. dos Santos and I. Sarnoff and G. Sartorelli and J. Schreiner and P. Schulte and H. Schulze Eißing and M. Schumann and L. Scotto Lavina and M. Selvi and F. Semeria and F. N. Semler and P. Shagin and S. Shi and H. Simgen and Z. Song and A. Stevens and C. Szyszka and A. Takeda and Y. Takeuchi and P. -L. Tan and D. Thers and G. Trinchero and C. D. Tunnell and K. Valerius and S. Vecchi and S. Vetter and G. Volta and C. Weinheimer and M. Weiss and D. Wenz and C. Wittweg and V. H. S. Wu and Y. Xing and D. Xu and Z. Xu and M. Yamashita and J. Yang and L. Yang and J. Ye and M. Yoshida and L. Yuan and G. Zavattini and Y. Zhao and M. Zhong and T. Zhu},
journal= {arXiv preprint arXiv:2603.23549},
year = {2026}
}