中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 新疆天文台
  3. 研究单元未命名
Spectuner-D1: Spectral Line Fitting of Interstellar Molecules Using Deep Reinforcement Learning

文献类型：期刊论文

作者Qiu, Yisheng6; Zhang, Tianwei6; Liu, Tie5; Zhu, Fengyao6; Meng, Dezhao3,4,5; Chen, Huaxi6; Moller, Thomas2; Schilke, Peter2; Quan, Donghui1,6
刊名ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES
出版日期2026-03-01
卷号283期号:1页码:16
ISSN号0067-0049
DOI10.3847/1538-4365/ae3742
通讯作者Qiu, Yisheng()
英文摘要Spectral lines from interstellar molecules provide crucial insights into the physical and chemical conditions of the interstellar medium. Traditional spectral line analysis relies heavily on manual intervention, which becomes impractical when handling the massive datasets produced by modern facilities such as the Atacama Large Millimeter/submillimeter Array (ALMA). To address this challenge, we introduce a novel deep reinforcement learning framework to automate spectral line fitting. Using observational data from ALMA, we train a neural network that maps both molecular spectroscopic data and observed spectra to physical parameters such as excitation temperature and column density. The neural network predictions can serve as initial estimates and be further refined using a local optimizer. Our method achieves fitting results consistent with global optimization with multiple runs, while reducing the number of forward modeling runs by an order of magnitude. We apply our method to pixel-level fitting for an observation of the G327.3-0.6 hot core and validate our results using xclass. We perform the fitting for typical complex organic molecules of hot cores, including CH3OH, CH3OCHO, CH3OCH3, C2H5CN, and C2H3CN. For a region of 100 x 100 pixels covering 5 GHz bandwidth, the fitting process requires 4.9-41.9 minutes using a desktop with 16 cores and one consumer-grade GPU card.
WOS关键词COLOGNE DATABASE ; I. ; PACKAGE ; ASTROPY ; MODEL ; IDENTIFICATION ; SUBMILLIMETER ; SPECTROSCOPY ; MILLIMETER ; SOFTWARE
资助项目National Natural Science Foundation of China[12373026] ; National Natural Science Foundation of China[2012.1.00712] ; National Natural Science Foundation of China[2023R01008] ; National Natural Science Foundation of China[2015.1.01193] ; National Natural Science Foundation of China[2024SSYS0012] ; National Natural Science Foundation of China[2016.1.00168] ; National Natural Science Foundation of China[12403030] ; National Natural Science Foundation of China[2017.1.00108] ; National Natural Science Foundation of China[2023TQ0330] ; Leading Innovation and Entrepreneurship Team of Zhejiang Province of China[2012.1.00712] ; Key R&D Program of Zhejiang, China[2023R01008] ; Young Scientists Fund of the National Natural Science Foundation of China[2015.1.01193] ; China Postdoctoral Science Foundation[2024SSYS0012] ; [2018.1.01496] ; [2019.1.00685] ; [2021.1.01164] ; [2022.1.01354] ; [2024.1.00653]
WOS研究方向Astronomy & Astrophysics
语种英语
WOS记录号WOS:001684934000001
出版者IOP Publishing Ltd
资助机构National Natural Science Foundation of China ; Leading Innovation and Entrepreneurship Team of Zhejiang Province of China ; Key R&D Program of Zhejiang, China ; Young Scientists Fund of the National Natural Science Foundation of China ; China Postdoctoral Science Foundation
源URL[http://ir.xao.ac.cn/handle/45760611-7/8555]  
专题研究单元未命名
通讯作者Qiu, Yisheng
作者单位1.Xian Jiaotong Liverpool Univ, Dept Phys, 111 Renai Rd, Suzhou 215123, Peoples R China
2.Univ Cologne, Phys Inst 1, Zulpicher Str 77, D-50937 Cologne, Germany
3.Univ Chinese Acad Sci, Beijing 100080, Peoples R China
4.Chinese Acad Sci, Xinjiang Astron Observ, Urumqi 830011, Peoples R China
5.Chinese Acad Sci, Shanghai Astron Observ, 80 Nandan Rd, Shanghai 200030, Peoples R China
6.Zhejiang Lab, Res Ctr Astron Comp, Hangzhou 311121, Peoples R China
推荐引用方式
GB/T 7714		 Qiu, Yisheng,Zhang, Tianwei,Liu, Tie,et al. Spectuner-D1: Spectral Line Fitting of Interstellar Molecules Using Deep Reinforcement Learning[J]. ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES,2026,283(1):16.
APA Qiu, Yisheng.,Zhang, Tianwei.,Liu, Tie.,Zhu, Fengyao.,Meng, Dezhao.,...&Quan, Donghui.(2026).Spectuner-D1: Spectral Line Fitting of Interstellar Molecules Using Deep Reinforcement Learning.ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES,283(1),16.
MLA Qiu, Yisheng,et al."Spectuner-D1: Spectral Line Fitting of Interstellar Molecules Using Deep Reinforcement Learning".ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES 283.1(2026):16.

入库方式： OAI收割

来源：新疆天文台

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。