期刊
JOURNAL OF COMPUTATIONAL SCIENCE
卷 51, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.jocs.2021.101323
关键词
Numerical forecast; Data assimilation; Machine learning; Four-Dimensional Variational Assimilation; Multi-layer perceptron
资金
- State Key Laboratory of Aerodynamics, China, China Aerodynamics Research and Development Center [SKLA20180303]
- Natural Science Foundation of Shanghai, China [19ZR1417700]
- Transforming Systems through Partnership, Newton Fund [TSPC1086]
This paper introduces a fast data assimilation (FDA) method based on machine learning, which greatly reduces the time of the data assimilation process.
Data assimilation (DA) can provide the more accurate initial state for numerical forecasting models. But traditional DA algorithms has the problem of long calculation time. This paper proposes fast data assimilation (FDA) based on machine learning. For training model, FDA uses 4DVAR, iForest, ?, and also includes a modified model that does not require observations. This paper applies FDA in the Lorenz63 dynamical system. The experimental results show that the single analysis time of FDA is almost 524 times faster than 4DVAR. FDA greatly reduces the time of the DA process.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据