Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation

Since the variety of their light curve morphologies, the vast majority of the known heartbeat stars (HBSs) have been discovered by manual inspection. Machine learning, which has already been successfully applied to the classification of variable stars based on light curves, offers another possibilit...

Full description

Saved in:
Bibliographic Details
Main Authors: Min-Yu Li, Sheng-Bang Qian, Li-Ying Zhu, Wen-Ping Liao, Lin-Feng Chang, Er-Gang Zhao, Xiang-Dong Shi, Fu-Xing Li, Qi-Bin Sun, Ping Li
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Astronomical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-3881/aded86
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849406069344829440
author Min-Yu Li
Sheng-Bang Qian
Li-Ying Zhu
Wen-Ping Liao
Lin-Feng Chang
Er-Gang Zhao
Xiang-Dong Shi
Fu-Xing Li
Qi-Bin Sun
Ping Li
author_facet Min-Yu Li
Sheng-Bang Qian
Li-Ying Zhu
Wen-Ping Liao
Lin-Feng Chang
Er-Gang Zhao
Xiang-Dong Shi
Fu-Xing Li
Qi-Bin Sun
Ping Li
author_sort Min-Yu Li
collection DOAJ
description Since the variety of their light curve morphologies, the vast majority of the known heartbeat stars (HBSs) have been discovered by manual inspection. Machine learning, which has already been successfully applied to the classification of variable stars based on light curves, offers another possibility for the automatic detection of HBSs. We propose a novel feature extraction approach for HBSs. First, the orbital frequencies are calculated automatically according to the Fourier spectra of the light curves. Then, the amplitudes of the first 100 harmonics are extracted. Finally, these harmonics are normalized as feature vectors of the light curve. A training data set of synthetic light curves is constructed using ELLC, and their features are fed into recurrent neural networks (RNNs) for supervised learning, with the expected output being the eccentricity of these light curves. The performance of the RNNs is evaluated using a test data set of synthetic light curves, achieving 95% accuracy. When applied to known HBSs from the Optical Gravitational Lensing Experiment, Kepler, and Transiting Exoplanet Survey Satellite surveys, the networks achieve an average accuracy of 86%. This method successfully identifies four new HBSs within the eclipsing binary catalog of Kirk et al. The use of orbital harmonics as features for HBSs proves to be a practical approach that significantly reduces the computational cost of neural networks. RNNs show excellent performance in recognizing this type of time series data. This method not only allows efficient identification of HBSs but can also be extended to recognize other types of periodic variable stars.
format Article
id doaj-art-5e7af91ab68048b8b7612ffa15bb72c5
institution Kabale University
issn 1538-3881
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series The Astronomical Journal
spelling doaj-art-5e7af91ab68048b8b7612ffa15bb72c52025-08-20T03:36:31ZengIOP PublishingThe Astronomical Journal1538-38812025-01-01170316110.3847/1538-3881/aded86Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and ValidationMin-Yu Li0https://orcid.org/0000-0002-8564-8193Sheng-Bang Qian1https://orcid.org/0000-0002-5995-0794Li-Ying Zhu2https://orcid.org/0000-0002-0796-7009Wen-Ping Liao3https://orcid.org/0000-0001-9346-9876Lin-Feng Chang4https://orcid.org/0000-0002-8421-4561Er-Gang Zhao5Xiang-Dong Shi6https://orcid.org/0000-0002-5038-5952Fu-Xing Li7https://orcid.org/0000-0002-0285-6051Qi-Bin Sun8https://orcid.org/0000-0003-0516-404XPing Li9https://orcid.org/0009-0004-0289-2732Yunnan Observatories, Chinese Academy of Sciences , Kunming 650216, People’s Republic of China ; zhuly@ynao.ac.cnDepartment of Astronomy, School of Physics and Astronomy, Key Laboratory of Astroparticle Physics of Yunnan Province, Yunnan University , Kunming 650091, People’s Republic of China ; qiansb@ynu.edu.cnYunnan Observatories, Chinese Academy of Sciences , Kunming 650216, People’s Republic of China ; zhuly@ynao.ac.cn; University of Chinese Academy of Sciences , No.1 Yanqihu East Road, Huairou District, Beijing 101408, People’s Republic of ChinaYunnan Observatories, Chinese Academy of Sciences , Kunming 650216, People’s Republic of China ; zhuly@ynao.ac.cn; University of Chinese Academy of Sciences , No.1 Yanqihu East Road, Huairou District, Beijing 101408, People’s Republic of ChinaDepartment of Astronomy, School of Physics and Astronomy, Key Laboratory of Astroparticle Physics of Yunnan Province, Yunnan University , Kunming 650091, People’s Republic of China ; qiansb@ynu.edu.cnYunnan Observatories, Chinese Academy of Sciences , Kunming 650216, People’s Republic of China ; zhuly@ynao.ac.cnYunnan Observatories, Chinese Academy of Sciences , Kunming 650216, People’s Republic of China ; zhuly@ynao.ac.cnDepartment of Astronomy, School of Physics and Astronomy, Key Laboratory of Astroparticle Physics of Yunnan Province, Yunnan University , Kunming 650091, People’s Republic of China ; qiansb@ynu.edu.cnDepartment of Astronomy, School of Physics and Astronomy, Key Laboratory of Astroparticle Physics of Yunnan Province, Yunnan University , Kunming 650091, People’s Republic of China ; qiansb@ynu.edu.cnYunnan Observatories, Chinese Academy of Sciences , Kunming 650216, People’s Republic of China ; zhuly@ynao.ac.cn; University of Chinese Academy of Sciences , No.1 Yanqihu East Road, Huairou District, Beijing 101408, People’s Republic of ChinaSince the variety of their light curve morphologies, the vast majority of the known heartbeat stars (HBSs) have been discovered by manual inspection. Machine learning, which has already been successfully applied to the classification of variable stars based on light curves, offers another possibility for the automatic detection of HBSs. We propose a novel feature extraction approach for HBSs. First, the orbital frequencies are calculated automatically according to the Fourier spectra of the light curves. Then, the amplitudes of the first 100 harmonics are extracted. Finally, these harmonics are normalized as feature vectors of the light curve. A training data set of synthetic light curves is constructed using ELLC, and their features are fed into recurrent neural networks (RNNs) for supervised learning, with the expected output being the eccentricity of these light curves. The performance of the RNNs is evaluated using a test data set of synthetic light curves, achieving 95% accuracy. When applied to known HBSs from the Optical Gravitational Lensing Experiment, Kepler, and Transiting Exoplanet Survey Satellite surveys, the networks achieve an average accuracy of 86%. This method successfully identifies four new HBSs within the eclipsing binary catalog of Kirk et al. The use of orbital harmonics as features for HBSs proves to be a practical approach that significantly reduces the computational cost of neural networks. RNNs show excellent performance in recognizing this type of time series data. This method not only allows efficient identification of HBSs but can also be extended to recognize other types of periodic variable stars.https://doi.org/10.3847/1538-3881/aded86Binary starsElliptical orbitsStellar oscillationsNeural networks
spellingShingle Min-Yu Li
Sheng-Bang Qian
Li-Ying Zhu
Wen-Ping Liao
Lin-Feng Chang
Er-Gang Zhao
Xiang-Dong Shi
Fu-Xing Li
Qi-Bin Sun
Ping Li
Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
The Astronomical Journal
Binary stars
Elliptical orbits
Stellar oscillations
Neural networks
title Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
title_full Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
title_fullStr Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
title_full_unstemmed Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
title_short Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation
title_sort heartbeat stars recognition based on recurrent neural networks method and validation
topic Binary stars
Elliptical orbits
Stellar oscillations
Neural networks
url https://doi.org/10.3847/1538-3881/aded86
work_keys_str_mv AT minyuli heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT shengbangqian heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT liyingzhu heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT wenpingliao heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT linfengchang heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT ergangzhao heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT xiangdongshi heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT fuxingli heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT qibinsun heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation
AT pingli heartbeatstarsrecognitionbasedonrecurrentneuralnetworksmethodandvalidation