Abstract
Solar wind prediction is a critical aspect of space weather forecasting, and current research has primarily focused on feature extraction from historical wind speed or individual solar images. To enhance the quality of data and improve prediction accuracy, we propose a novel approach that leverages multi-modality, combining both temporal and spatial dimensions. Additionally, we utilize prior knowledge to guide model training, specifically in the image preprocessing and matrix multiplication stages, where prior knowledge constraints are applied. Our study introduces the spatio-temporal attention model (STA) for solar wind prediction, which comprises an image branch and a solar wind speed data branch. The image branch uses a shared-weight feature extraction network to extract features from EUV images, while the solar wind speed data branch models temporal dynamics with sequence networks. Furthermore, we incorporated an attention-based feature extraction module and a feature fusion module to enhance the model’s performance. Our experimental results demonstrate that the proposed STA model outperforms existing state-of-the-art models.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Brown, J.E., Svoboda, E.: Attention-based machine vision models and techniques for solar wind speed forecasting using solar EUV images. Remote Sens. 13(7), 1343 (2021)
Zhao, L., Wicks, R.T.: Negative correlation between the peak speed of the solar wind and the co-latitude of the corresponding solar source coronal hole. Astrophys. J. 830(1), 56 (2016)
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems, vol. 30, pp. 5998–6008 (2017)
Raju, S.S., Das, A.: Solar wind speed backcasting from solar EUV images using convolutional neural networks. Sol. Energy 220, 183–193 (2021). https://doi.org/10.1016/j.solener.2021.01.014
Tan, M., Le, Q.: EfficientNet: rethinking model scaling for convolutional neural networks. arXiv preprint: arXiv:1905.11946 (2019)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
Schrijver, C.J., et al.: Understanding space weather to shield society: a global road map for 2025–2050 commissioned by COSPAR and ILWS. Adv. Space Res. 55(12), 2745–2807 (2015)
St. Cyr, O.C., et al.: Coronal mass ejections and the solar wind. J. Geophys. Res.: Space Phys. 105(A12), 27421–27438 (2000)
Kilpua, E.K., Koskinen, H.E.J., Pulkkinen, T.I., Vourlidas, A.: Introduction to ICMEs and space weather. In: ICMEs and Space Weather: Causes, Characteristics, and Consequences, pp. 1–26. Springer, Cham (2017)
Upendran, L., Kwon, H.D., Kang, S.B., Park, H., Moon, Y.J.: Solar wind forecasting using deep learning techniques with solar EUV images. Astron. Astrophys. 642, A26 (2020)
Szegedy, C., et al.: Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–9 (2014)
Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 248–255. IEEE (2009)
Acknowledgements
This work was supported in part by the National Natural Science Foundation of China under Grant 62076179.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Cai, P., Yang, L., Sun, Y. (2023). Spatio-Temporal Attention Model with Prior Knowledge for Solar Wind Speed Prediction. In: Iliadis, L., Papaleonidas, A., Angelov, P., Jayne, C. (eds) Artificial Neural Networks and Machine Learning – ICANN 2023. ICANN 2023. Lecture Notes in Computer Science, vol 14262. Springer, Cham. https://doi.org/10.1007/978-3-031-44201-8_29
Download citation
DOI: https://doi.org/10.1007/978-3-031-44201-8_29
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44200-1
Online ISBN: 978-3-031-44201-8
eBook Packages: Computer ScienceComputer Science (R0)