We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
High‐Precision and Fast Prediction of Regional Wind Fields in Near Space Using Neural‐Network Approximation of Operators.
- Authors
Chen, Biao; Sheng, Zheng; He, Yang
- Abstract
Fine modeling and fast prediction of regional wind field in the middle and upper atmosphere has always been a difficult problem. We designed a neural operator method to solve this problem. We combine the idea of data assimilation with deep learning method to design a regional wind field operator suitable for near space. The annual Root mean square error of the zonal wind and meridional wind of the operator model at the height of 30 km are 0.903 and 0.881, respectively, which is three times that of ConvLSTM. Moreover, we validate the sparse spatio‐temporal modeling method of regional wind field operator at 20/30/40/50 km altitude. The result shows that the model is mesh‐free, and can get high‐precision modeling of different spatio‐temporal resolutions, multiple regions and arbitrary positions at one time, which lays an foundation for fine regional modeling and rapid utilization of near space. Plain Language Summary: The complex variation mechanism of regional wind fields in near space leads to the difficulty of high‐precision modeling and fast prediction, which seriously affects the design and flight of near space vehicles. In this study, a regional wind field neural operator method has been proposed, which can achieve the fine modeling of the regional wind field in the middle and upper atmosphere. The new method is highly flexible, and can get high‐precision modeling and rapid prediction in different spatial‐temporal resolutions, multiple regions and arbitrary positions. Key Points: The neural operator is first used to study high‐precision spatio‐temporal modeling and rapid prediction of regional wind fields in near spaceThe Root mean square error accuracy of regional wind field operator model is three times that of ConvLSTMThe novel method is suitable for sparse spatio‐temporal modeling at any location with different data resolutions
- Subjects
STANDARD deviations; MERIDIONAL winds; UPPER atmosphere; ZONAL winds; MIDDLE atmosphere
- Publication
Geophysical Research Letters, 2023, Vol 50, Issue 22, p1
- ISSN
0094-8276
- Publication type
Article
- DOI
10.1029/2023GL106115