Prediction and Analysis of Sunspot Activity Based on Multivariable LSTM Network

Liang Bo; Lin Yuqi; Dai Wei; Feng Song; Yang Yunfei

Astronomical Research and Technology > 2020 > 17(3): 322-333.

Liang Bo, Lin Yuqi, Dai Wei, Feng Song, Yang Yunfei. Prediction and Analysis of Sunspot Activity Based on Multivariable LSTM Network[J]. Astronomical Research and Technology, 2020, 17(3): 322-333.

Citation:

Liang Bo, Lin Yuqi, Dai Wei, Feng Song, Yang Yunfei. Prediction and Analysis of Sunspot Activity Based on Multivariable LSTM Network[J]. Astronomical Research and Technology, 2020, 17(3): 322-333.

Citation:

Liang Bo, Lin Yuqi, Dai Wei, Feng Song, Yang Yunfei. Prediction and Analysis of Sunspot Activity Based on Multivariable LSTM Network[J]. Astronomical Research and Technology, 2020, 17(3): 322-333.

PDF (5383 KB)

Prediction and Analysis of Sunspot Activity Based on Multivariable LSTM Network

Liang Bo^1,2,
Lin Yuqi^1,2,
Dai Wei^1,2,
Feng Song^1,2, ,,
Yang Yunfei^1,2

1. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China;
2. Key Laboratory of Computer Technology Appication of Yunnan Province, Kunming University of Science and Technology, Kunming 650500, China

More Information

Received Date: November 26, 2019
Revised Date: January 16, 2020
Available Online: November 20, 2023

Graphical Abstract

Abstract

Abstract

By adding the length of sunspot cycle, the multivariable input data of Long Short-Term Memory (LSTM) is constructed to predict the change of sunspot number in the next ten years with multiple time steps. According to the length of training data, the data set is divided into two groups of time series segments, namely, 11-slice and 6-slice. The prediction effect of univariate and multivariable on single-time step and multi-time step is compared. Finally, the main results are as follows:(1) Compared with the method of 6-slice, the method of 11-slice has lower Root Mean Squared Error (RMSE). (2) The optimal multi-step prediction is better than the single-step. (3) The accuracy of the starting point, ending point, and the maximum amplitude position of the cycle on the legend proves that prediction effect of the multivariate multi-step method is better than that of single-step method.
- Sunspot,
- LSTM,
- Multivariate,
- Time step,
- RMSE

FullText(HTML)

References (20)

References

[1]	苏同卫, 李可军. 太阳活动对中国中纬度地区8级大地震的可能触发[J]. 天文研究与技术——国家天文台台刊, 2007, 4(2):195-198.
[2]	沈玫, 吕达仁. 太阳活动与天气和气候变化的关系的评述[J]. 云南天文台台刊, 1989(增刊1):242-254.
[3]	DU Z L, LI R, WANG H N.The predictive power of ohl's precusor method[J]. The Astronomical Journal, 2009, 138(6):1998-2001.
[4]	SCHATTEN K H, SOFIA S. Forecast of an exceptionally large even-numbered solar cycle[J]. Geophysical Research Letters, 2013, 14(6):632-635.
[5]	SARP V, KILCIK A, YURCHYSHYN V, et al. Prediction of solar cycle 25:a non-linear approach[J]. Monthly Notices of the Royal Astronomical Society, 2018, 481(3):2981-2985.
[6]	KILCIK A, ANDERSON C N K, ROZELOT J P, et al. Nonlinear prediction of solar cycle 24[J]. The Astrophysical Journal, 2008, 693(2):1173-1177.
[7]	AGUIRRE L A, LETELLIER C, MAQUET J. Forecasting the time series of sunspot numbers[J]. Solar Physics, 2008, 249(1):103-120.
[8]	MARIS G, ONCICA A. Solar cycle 24 forecasts[J]. Sun & Geosphere, 2006, 1(1):8-11.
[9]	KITIASHVILI I, KOSOVICHEV A G. Application of data assimilation method for predicting solar cycles[J]. The Astrophysical Journal, 2008, 688(1):L49-L52.
[10]	KITIASHVILI I N, KOSOVICHEV A G. Nonlinear dynamical modeling of solar cycles using dynamo formulation with turbulent magnetic helicity[J]. Geophysical & Astrophysical Fluid Dynamics, 2009, 103(1):53-68.
[11]	PETROVAY K. Solar cycle prediction[J]. Living Reviews in Solar Physics, 2010, 7(1):1-59.
[12]	SARP V, KILCIK A, YURCHYSHYN V, et al. Prediction of solar cycle 25:a non-linear approach[J]. Monthly Notices of the Royal Astronomical Society, 2018, 481(3):2981-2985.
[13]	BHOWMIK P, NANDY D. Prediction of the strength and timing of sunspot cycle 25 reveal decadal-scale space environmental conditions[J]. Nature Communications, 2018, 9(1):1-10.
[14]	PESNELL W D, SCHATTEN K H. An early prediction of the amplitude of Solar Cycle 25[J]. Solar Physics, 2018, 293(7):112-121.
[15]	PALA Z, ATICI R. Forecasting sunspot time series using deep learning methods[J]. Solar Physics, 2019, 294(5):50-63.
[16]	ATTIA A F, ISMAIL H A, BASURAH H M. A neuro-fuzzy modeling for prediction of solar cycles 24 and 25[J]. Astrophysics and Space Science, 2013, 344(1):5-11.
[17]	DANI T, SULISTIANI S. Prediction of maximum amplitude of solar cycle 25 using machine learning[C]//Proceedings of the 10th Southeast Asia Astronomy Network. 2019.
[18]	HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780.
[19]	GALILI T. R-bloggers:time series deep learning:forecasting sunspots with kerasstatefullstm in r[EB/OL]. (2018-04-17)[2019-11-27]. https://www.r-bloggers.com/time-series-deep-learning-forecasting-sunspots-with-keras-stateful-lstm-in-r/.
[20]	DANCHO M, KEYDANA S. RStudio AI Blog:predicting sunspot frequency with keras[EB/OL]. (2018-06-25)[2019-11-27]. https://blogs.rstudio.com/tensorflow/posts/2018-06-25-sunspots-lstm/.

Articles Related

[1]	Jie Cao, Tingting Xu, Linhua Deng, Xueliang Zhou, Xinhua Zhao, Nanbin Xiang, Fuyu Li, Miao Wan, Weihong Zhou. Hemispheric prediction of solar cycles 25 and 26 from multivariate sunspot time-series data via Informer models [J]. Astronomical Techniques and Instruments, 2025, 2(1): 16-26. DOI: 10.61977/ati2024047
[2]	Zhang Yu, Qu Lili, Dong Shaowu, Zhang Jihai, Bai Shanshan, Wang Yiheng. Data Analysis and Research Based on Time-frequency Data Platform [J]. Astronomical Techniques and Instruments, 2023, 20(5): 471-477. DOI: 10.14005/j.cnki.issn1672-7673.20221102.001
[3]	Gao Qingpeng, Li Chunxiao, Li Rongwang, Duan Jianfeng, Li Yuqiang. Magpie Bridge Satellite Laser Ranging Time Window and Distance Probability Analysis [J]. Astronomical Research and Technology, 2019, 16(4): 422-430.
[4]	Li Linsen. Evolutional Time Scale of Mass and Radius of T Tauri and Its Influence on the Rotation under the Action of the Ejected Material and Gravitational Contraction in the Step of the Slow Gravitational Contraction [J]. Astronomical Research and Technology, 2017, 14(2): 150-156.
[5]	Qiu Guangliang, Zhang Xiong, Qian Wenfeng. Analysis of Integration Time and Photometric Error of the CCD Photometry for the BL Lac Object ON231 [J]. Astronomical Research and Technology, 2013, 10(1): 1-6.
[6]	Zheng Lianhui, Jin Zhenyu, Xiang Yongyuan. Analysis of Effects of Errors of the Absolute Difference Algorithm on Day-time Seeing Measurement [J]. Astronomical Research and Technology, 2012, 9(2): 157-161.
[7]	Kang Silin, Li Yuqiang. Error Analysis of GPS Positioning [J]. Astronomical Research and Technology, 2010, 7(3): 222-230.
[8]	XIE Rui-xiang, HAO Long-fei, YAN Yi-hua, XU Chun, LIU Yu-ying, SHI Shuo-biao, LI Wei-hua. Radio Time Profiles and Magnetic Configurations of Two 2-step Triggering Solar Flares [J]. Astronomical Research and Technology, 2006, 3(3): 247-257.
[9]	Liu W, Hu J. Y.. CCD Time-Resolved Photometry of Four Newly Discovered CVs [J]. Publications of the Yunnan Observatory, 1999, 0(S1): 386-390.
[10]	Mei Qing GAO, Nan KONG, Jing Tian XIANG, Gui Qing ZHANG. THE PREDICTION OF GEOMAGNETIC ACTIVITY MAXIMUM AND PEAK TIME IN 22ND SOLAR CYCLE [J]. Publications of the Yunnan Observatory, 1989, 0(S1): 317-322.