Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network
Abstract In this study, we developed a novel deep learning model to predict the disturbance storm time (Dst) index 1–4 hr ahead. We also employed the Monte Carlo (MC) dropout technique to estimate the uncertainty and provide the prediction interval by introducing a recalibration factor. The proposed...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-12-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2024SW004002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832575427009314816 |
|---|---|
| author | Ruyao Wang Jianhui Wang Tuo Liang Huixiong Zhang |
| author_facet | Ruyao Wang Jianhui Wang Tuo Liang Huixiong Zhang |
| author_sort | Ruyao Wang |
| collection | DOAJ |
| description | Abstract In this study, we developed a novel deep learning model to predict the disturbance storm time (Dst) index 1–4 hr ahead. We also employed the Monte Carlo (MC) dropout technique to estimate the uncertainty and provide the prediction interval by introducing a recalibration factor. The proposed model achieves excellent scalability by extracting representative embeddings from the Dst index time series through an encoder‐decoder framework and integrating these with external solar wind parameters via a prediction network. We utilized magnetic storm data from 1998 to 2018 to evaluate the performance of the prediction model. Experimental results indicate that the proposed model reduces root mean square errors (RMSE) and improves the coefficient of determination (R2) compared to existing methods. Quantitative uncertainty analysis demonstrates that the prediction interval is reliable in most cases. The percent distribution across test storms by comparing the uncertainty‐based prediction of the minimum peak with the mean prediction indicates the probabilistic forecast model is competitive. |
| format | Article |
| id | doaj-art-96991c15b1e34f2687caca8485ffb327 |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-96991c15b1e34f2687caca8485ffb3272025-02-01T08:10:33ZengWileySpace Weather1542-73902024-12-012212n/an/a10.1029/2024SW004002Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning NetworkRuyao Wang0Jianhui Wang1Tuo Liang2Huixiong Zhang3School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu ChinaSchool of Information and Software Engineering University of Electronic Science and Technology of China Chengdu ChinaSchool of Finance Institute of Chinese Financial Studies Southwestern University of Finance and Economics Chengdu ChinaSchool of Life Science and Technology University of Electronic Science and Technology of China Chengdu ChinaAbstract In this study, we developed a novel deep learning model to predict the disturbance storm time (Dst) index 1–4 hr ahead. We also employed the Monte Carlo (MC) dropout technique to estimate the uncertainty and provide the prediction interval by introducing a recalibration factor. The proposed model achieves excellent scalability by extracting representative embeddings from the Dst index time series through an encoder‐decoder framework and integrating these with external solar wind parameters via a prediction network. We utilized magnetic storm data from 1998 to 2018 to evaluate the performance of the prediction model. Experimental results indicate that the proposed model reduces root mean square errors (RMSE) and improves the coefficient of determination (R2) compared to existing methods. Quantitative uncertainty analysis demonstrates that the prediction interval is reliable in most cases. The percent distribution across test storms by comparing the uncertainty‐based prediction of the minimum peak with the mean prediction indicates the probabilistic forecast model is competitive.https://doi.org/10.1029/2024SW004002AILSTMgeomagnetic stormuncertainty estimationdeep learningneural network |
| spellingShingle | Ruyao Wang Jianhui Wang Tuo Liang Huixiong Zhang Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network Space Weather AI LSTM geomagnetic storm uncertainty estimation deep learning neural network |
| title | Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network |
| title_full | Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network |
| title_fullStr | Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network |
| title_full_unstemmed | Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network |
| title_short | Short‐Term Prediction of the Dst Index and Estimation of Efficient Uncertainty Using a Hybrid Deep Learning Network |
| title_sort | short term prediction of the dst index and estimation of efficient uncertainty using a hybrid deep learning network |
| topic | AI LSTM geomagnetic storm uncertainty estimation deep learning neural network |
| url | https://doi.org/10.1029/2024SW004002 |
| work_keys_str_mv | AT ruyaowang shorttermpredictionofthedstindexandestimationofefficientuncertaintyusingahybriddeeplearningnetwork AT jianhuiwang shorttermpredictionofthedstindexandestimationofefficientuncertaintyusingahybriddeeplearningnetwork AT tuoliang shorttermpredictionofthedstindexandestimationofefficientuncertaintyusingahybriddeeplearningnetwork AT huixiongzhang shorttermpredictionofthedstindexandestimationofefficientuncertaintyusingahybriddeeplearningnetwork |