Numerical weather forecast mixed data assimilation method based on triple multi-layer perceptrons

Abstract

The invention discloses a numerical weather forecast mixed data assimilation method based on triple multi-layer perceptrons, and the method comprises the following steps: building a first perceptron model based on a three-dimensional variational data assimilation method, and carrying out the training of the first perceptron model; establishing a second perceptron model based on an ensemble Kalmanfilter data assimilation method, and training the second perceptron model; establishing a third perceptron model of hybrid assimilation; training the third perceptron model; and calculating by using the trained first perceptron model, second perceptron model and third perceptron model to obtain analysis field data. According to the method, the physical law of atmospheric motion and the time characteristic of atmospheric state variables are utilized; a traditional data assimilation method is simulated, optimized and mixed, the flow dependence of the atmospheric variables is fully considered, time characteristics contained in atmospheric data are introduced accordingly, and the assimilation performance of the method is effectively improved compared with a traditional method.

Description

technical field [0001] The invention belongs to the technical field of numerical weather forecasting, and in particular relates to a numerical weather forecasting mixed data assimilation method based on triple multilayer perceptrons. Background technique [0002] Vihelm Bjerknes (1904) defined numerical weather prediction (NWP) as an initial boundary value problem. That is, the future state of the atmosphere depends on its detailed initial state, boundary conditions, and predictive models. Data assimilation (DA) can be described as a process that combines all available data resources in observations and short-term forecasts (that is, background fields or initial guesses) to obtain an analysis field, hoping to obtain a probability density of the actual atmospheric state best estimate of . That is, the goal of data assimilation is to produce the most accurate initial conditions for numerical weather prediction. [0003] In the past three decades, two mainstream data assimil...

AI Technical Summary

Problems solved by technology

The main reason is that researchers only pay attention to the statistical characteristics of the data, while ignoring the physical properties of the atmospheric state itself

Images