Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products

In this study, we construct two kinds of data sets from distinct time periods, both comprising line-of-sight magnetograms and knowledge-informed features. We develop eight models for forecasting ≥M-class flares within 24 hr, including the image-based convolutional neural network (CNN), CNN-BiLSTM, C...

Full description

Saved in:
Bibliographic Details
Main Authors: Xuebao Li, Shunhuang Zhang, Yanfang Zheng, Ting Li, Rui Wang, Yingbo Liu, Hongwei Ye, Noraisyah Mohamed Shah, Pengchao Yan, Xuefeng Li, Xiaotian Wang, Yongshang Lv, Jinfang Wei, Honglei Jin, Changtian Xiang
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Astrophysical Journal Supplement Series
Subjects:
Online Access:https://doi.org/10.3847/1538-4365/ade687
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849254030196342784
author Xuebao Li
Shunhuang Zhang
Yanfang Zheng
Ting Li
Rui Wang
Yingbo Liu
Hongwei Ye
Noraisyah Mohamed Shah
Pengchao Yan
Xuefeng Li
Xiaotian Wang
Yongshang Lv
Jinfang Wei
Honglei Jin
Changtian Xiang
author_facet Xuebao Li
Shunhuang Zhang
Yanfang Zheng
Ting Li
Rui Wang
Yingbo Liu
Hongwei Ye
Noraisyah Mohamed Shah
Pengchao Yan
Xuefeng Li
Xiaotian Wang
Yongshang Lv
Jinfang Wei
Honglei Jin
Changtian Xiang
author_sort Xuebao Li
collection DOAJ
description In this study, we construct two kinds of data sets from distinct time periods, both comprising line-of-sight magnetograms and knowledge-informed features. We develop eight models for forecasting ≥M-class flares within 24 hr, including the image-based convolutional neural network (CNN), CNN-BiLSTM, CNN-BiLSTM-Attention, and Vision Transformer models, as well as the knowledge-informed neural network, BiLSTM, BiLSTM-Attention, and iTransformer models. We analyze the importance of knowledge-informed features by assessing categorical and probabilistic performance using the true skill statistic (TSS) and the Brier skill score (BSS), respectively. This is the first time the iTransformer has been applied to flare forecasting. Subsequently, we compare the forecasting performance of the eight models. Then, we investigate the generalization ability of the models across three different data products. Finally, we fairly compare the forecasting performance of iTransformer with that of the currently advanced NASA/CCMC models. The major results are as follows. (1) The R_VALUE feature consistently shows the best performance in both categorical and probabilistic forecasting for the knowledge-informed models. (2) The iTransformer yields the highest forecasting performance, with TSS and BSS scores of 0.768 ± 0.072 and 0.513 ± 0.063, respectively. The knowledge-informed deep learning models consistently outperform image-based models. (3) The three image-based models demonstrate good generalization performance in categorical forecasting on Space-weather Helioseismic and Magnetic Imager (HMI) Active Region Patches (SHARP), HMI, and Full-Disk Magnetograph (FMG), while the two knowledge-informed models exhibit excellent generalization performance on SHARP and HMI. This is the first time that FMG magnetograms and knowledge-informed features are used for flare forecasting. Additionally, the five models also demonstrate strong generalization ability on SHARP across different time periods. (4) The iTransformer exhibits superior forecasting performance compared to NASA/CCMC.
format Article
id doaj-art-a51e4bcd7bef46f2a77a971776a4333e
institution Kabale University
issn 0067-0049
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series The Astrophysical Journal Supplement Series
spelling doaj-art-a51e4bcd7bef46f2a77a971776a4333e2025-08-20T03:56:09ZengIOP PublishingThe Astrophysical Journal Supplement Series0067-00492025-01-0127924610.3847/1538-4365/ade687Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data ProductsXuebao Li0https://orcid.org/0000-0003-0397-4372Shunhuang Zhang1https://orcid.org/0009-0003-0325-1366Yanfang Zheng2https://orcid.org/0000-0003-0229-3989Ting Li3https://orcid.org/0000-0001-6655-1743Rui Wang4https://orcid.org/0000-0001-5205-1713Yingbo Liu5https://orcid.org/0000-0003-4031-4285Hongwei Ye6https://orcid.org/0009-0009-9722-5794Noraisyah Mohamed Shah7https://orcid.org/0000-0002-1320-8711Pengchao Yan8https://orcid.org/0000-0002-3667-3587Xuefeng Li9https://orcid.org/0009-0000-6353-5473Xiaotian Wang10https://orcid.org/0009-0004-1408-8055Yongshang Lv11https://orcid.org/0009-0004-9110-0425Jinfang Wei12https://orcid.org/0009-0001-3404-3186Honglei Jin13https://orcid.org/0009-0006-5294-3571Changtian Xiang14https://orcid.org/0009-0009-8072-3519School of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.com; State Key Laboratory of Space Weather, Chinese Academy of Sciences , Beijing 100190, People’s Republic of ChinaSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.com; State Key Laboratory of Space Weather, Chinese Academy of Sciences , Beijing 100190, People’s Republic of ChinaSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.com; State Key Laboratory of Space Weather, Chinese Academy of Sciences , Beijing 100190, People’s Republic of ChinaCAS Key Laboratory of Solar Activity, National Astronomical Observations, Chinese Academy of Sciences , Beijing 100101, People’s Republic of China; School of Astronomy and Space Science, University of Chinese Academy of Sciences , Beijing 100049, People’s Republic of China; National Space Science Center, Chinese Academy of Sciences , Beijing 100190, People’s Republic of ChinaState Key Laboratory of Space Weather, Chinese Academy of Sciences , Beijing 100190, People’s Republic of ChinaBig Data Research Institute of Yunnan Economy and Society, Yunnan University of Finance and Economics , Kunming 650221, People’s Republic of China ; liuyb@ynufe.edu.cnSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.com; State Key Laboratory of Space Weather, Chinese Academy of Sciences , Beijing 100190, People’s Republic of ChinaDepartment of Electrical Engineering, Faculty of Engineering, University of Malaya , MalaysiaSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comSchool of Computer Science, Jiangsu University of Science and Technology , Zhenjiang 212100, People’s Republic of China ; zyf062856@163.comIn this study, we construct two kinds of data sets from distinct time periods, both comprising line-of-sight magnetograms and knowledge-informed features. We develop eight models for forecasting ≥M-class flares within 24 hr, including the image-based convolutional neural network (CNN), CNN-BiLSTM, CNN-BiLSTM-Attention, and Vision Transformer models, as well as the knowledge-informed neural network, BiLSTM, BiLSTM-Attention, and iTransformer models. We analyze the importance of knowledge-informed features by assessing categorical and probabilistic performance using the true skill statistic (TSS) and the Brier skill score (BSS), respectively. This is the first time the iTransformer has been applied to flare forecasting. Subsequently, we compare the forecasting performance of the eight models. Then, we investigate the generalization ability of the models across three different data products. Finally, we fairly compare the forecasting performance of iTransformer with that of the currently advanced NASA/CCMC models. The major results are as follows. (1) The R_VALUE feature consistently shows the best performance in both categorical and probabilistic forecasting for the knowledge-informed models. (2) The iTransformer yields the highest forecasting performance, with TSS and BSS scores of 0.768 ± 0.072 and 0.513 ± 0.063, respectively. The knowledge-informed deep learning models consistently outperform image-based models. (3) The three image-based models demonstrate good generalization performance in categorical forecasting on Space-weather Helioseismic and Magnetic Imager (HMI) Active Region Patches (SHARP), HMI, and Full-Disk Magnetograph (FMG), while the two knowledge-informed models exhibit excellent generalization performance on SHARP and HMI. This is the first time that FMG magnetograms and knowledge-informed features are used for flare forecasting. Additionally, the five models also demonstrate strong generalization ability on SHARP across different time periods. (4) The iTransformer exhibits superior forecasting performance compared to NASA/CCMC.https://doi.org/10.3847/1538-4365/ade687Solar active regionsSolar active region magnetic fieldsNeural networksAstronomy image processingSolar flares
spellingShingle Xuebao Li
Shunhuang Zhang
Yanfang Zheng
Ting Li
Rui Wang
Yingbo Liu
Hongwei Ye
Noraisyah Mohamed Shah
Pengchao Yan
Xuefeng Li
Xiaotian Wang
Yongshang Lv
Jinfang Wei
Honglei Jin
Changtian Xiang
Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products
The Astrophysical Journal Supplement Series
Solar active regions
Solar active region magnetic fields
Neural networks
Astronomy image processing
Solar flares
title Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products
title_full Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products
title_fullStr Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products
title_full_unstemmed Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products
title_short Forecasting Major Flares Using Magnetograms and Knowledge-informed Features: A Comparative Study of Deep Learning Models with Generalization to Multiple Data Products
title_sort forecasting major flares using magnetograms and knowledge informed features a comparative study of deep learning models with generalization to multiple data products
topic Solar active regions
Solar active region magnetic fields
Neural networks
Astronomy image processing
Solar flares
url https://doi.org/10.3847/1538-4365/ade687
work_keys_str_mv AT xuebaoli forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT shunhuangzhang forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT yanfangzheng forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT tingli forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT ruiwang forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT yingboliu forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT hongweiye forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT noraisyahmohamedshah forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT pengchaoyan forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT xuefengli forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT xiaotianwang forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT yongshanglv forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT jinfangwei forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT hongleijin forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts
AT changtianxiang forecastingmajorflaresusingmagnetogramsandknowledgeinformedfeaturesacomparativestudyofdeeplearningmodelswithgeneralizationtomultipledataproducts