New approach to applying neural network in nonlinear dynamic model
New approach to applying neural network in nonlinear dynamic model
复制标题
DOI：
10.1016/j.apm.2007.09.006
复制
发表时间：
2008-12
影响因子：
5
通讯作者：
F. P. Harter;H. Velho
中科院分区：
文献类型：
--
作者：
F. P. Harter;H. Velho
在这项工作中，径向基函数神经网络（RBF-NN）被应用于模拟数据同化场景中的扩展卡尔曼滤波器（EKF）。这里研究的动力学模型基于一维浅水方程DYNAMO-1D。与数值天气预报的操作原始方程模型相比，该代码很简单。虽然简单，但 DYNAMO-1D 可以丰富地表示一些大气运动，例如罗斯贝和重力波。文献表明，EKF 跟踪非线性模型的能力取决于观测值和模型误差的频率和精度。在某些情况下，仅四阶矩 EKF 效果很好，但应用于高维状态空间时会很笨重。人工神经网络 (ANN) 是解决此计算复杂性问题的替代解决方案，一旦使用高阶卡尔曼滤波器离线训练 ANN，即使该卡尔曼滤波器具有较高的计算成本（这在 ANN 训练阶段不是问题）。这项工作取得的成果鼓励我们将该技术应用于运营模型。然而，目前还不可能保证高维问题的收敛性。
In this work, radial basis function neural network (RBF-NN) is applied to emulate an extended Kalman filter (EKF) in a data assimilation scenario. The dynamical model studied here is based on the one-dimensional shallow water equation DYNAMO-1D. This code is simple when compared with an operational primitive equation models for numerical weather prediction. Although simple, the DYNAMO-1D is rich for representing some atmospheric motions, such as Rossby and gravity waves. It has been shown in the literature that the ability of the EKF to track nonlinear models depends on the frequency and accuracy of the observations and model errors. In some cases, just fourth-order moment EKF works well, but will be unwieldy when applied to high-dimensional state space. Artificial Neural Network (ANN) is an alternative solution for this computational complexity problem, once the ANN is trained offline with a high order Kalman filter, even though this Kalman filter has high computational cost (which is not a problem during ANN training phase). The results achieved in this work encourage us to apply this technique on operational model. However, it is not yet possible to assure convergence in high dimensional problems.