Deep learning-based SuperDARN radar flow map short-term forecasting method

Abstract

The invention discloses a deep learning-based SuperDARN radar flow map short-term forecasting method. The method comprises the following steps of 1, obtaining SuperDARN radar observation data and OMNI satellite observation data and carrying out preprocessing; 2, matching and aligning the OMNI satellite observation data and the SuperDARN radar observation data to form a complete data set; 3, performing feature selection by using a Pearson's correlation coefficient method and standardizing the feature selection; 4, inputting into a deep learning model built by a Tensorflow module in Python for training and parameter debugging to obtain a forecasting model; and 5, applying the forecasting model to test data. Based on the development of deep learning and big data correlation theory technology, a neural network correlation method is adopted, a SuperDARN radar is utilized to obtain massive data with rich information, the relationship between the data is expressed by a neural network with strong fitting ability, short-term forecasting of the high-latitude ionospheric convection image is realized, and the requirement of space weather forecast is met.

Description

technical field [0001] The invention belongs to the technical field of space weather forecasting, in particular to a method for short-term forecasting of SuperDARN radar convection graphs based on deep learning. Background technique [0002] Plasma convection in the high-latitude ionosphere is an important phenomenon of space weather and an important parameter in the study of the ionosphere, which contains a series of important information about the energy transfer of the solar wind to the magnetosphere-ionosphere system. Many studies on ionospheric convection have focused on analyzing the dependence of ionospheric plasma convection on near-Earth space parameters, such as interplanetary magnetic field components, solar wind velocity, solar wind dynamic pressure, and geomagnetic activity index. Studies have shown that the structure of the convective map is also significantly different depending on the direction and magnitude of the interplanetary magnetic field IMF. When the ...

AI Technical Summary

Problems solved by technology

When the IMF is heading north, the coupling is weaker and the circulation is also weaker, but the convective image is more complex, often presenting a distorted double-vortex structure or even a multi-vortex structure

Images