Model error correction in data assimilation by integrating neural networks
Model error correction in data assimilation by integrating neural networks
复制标题
通过集成神经网络进行数据同化中的模型误差校正
DOI：
--
复制
发表时间：
2019
影响因子：
13.6
通讯作者：
Yike Guo
中科院分区：
文献类型：
--
作者：
Jiangcheng Zhu;Shuang Hu;Rossella Arcucci;Chao Xu;Jihong Zhu;Yike Guo
在本文中，我们提出了一种新的方法，结合神经网络（NN）的数据同化（DA）。针对结构模式的不确定性，我们提出了一个框架集成神经网络与物理模式的DA算法，以改善同化过程和预测结果。随着观测数据的更新，NN被迭代训练。这里使用的主要DA模型是卡尔曼滤波器和变分方法。通过算例和灵敏度分析验证了该算法的有效性。
In this paper, we suggest a new methodology which combines Neural Networks (NN) into Data Assimilation (DA). Focusing on the structural model uncertainty, we propose a framework for integration NN with the physical models by DA algorithms, to improve both the assimilation process and the forecasting results. The NNs are iteratively trained as observational data is updated. The main DA models used here are the Kalman filter and the variational approaches. The effectiveness of the proposed algorithm is validated by examples and by a sensitivity study.