Prediction method of ionospheric vertical total electron content based on two-dimensional time series

Abstract

The present invention discloses a method for forecasting an ionospheric vertical total electron content based on time-series two-dimensionalization. The method comprises four steps: step S1, by means of known ionospheric vertical total electron content data, analyzing variation characteristics of the ionospheric vertical total electron content over time; step S2, establishing a time-series two-dimensionalization plane, determining a weight value between an ionospheric vertical total electron content to be forecast and the known ionospheric vertical total electron content, and according to the weight value, performing weighting processing on the known ionospheric vertical total electron content data that is to be used as an input layer; step S3, establishing a neural network model; and step S4, by using the neural network model, forecasting the ionospheric vertical total electron content. According to the invention, by performing time-series two-dimensionalization processing on the ionospheric vertical total electron content and by setting different weight values, a good forecast of ionospheric vertical total electron content can be achieved.

Description

technical field [0001] The invention relates to a method for forecasting the vertical total electron content of the ionosphere, in particular to a method for forecasting the vertical total electron content of the ionosphere based on time series two-dimensionalization. Background technique [0002] Affected by solar ultraviolet rays, ray radiation and high-energy particles, there are a large number of free electrons in the 60-2000km atmospheric region, forming the earth's ionosphere. The ionosphere has an important impact on the measurement accuracy of satellite positioning, navigation, timing, and remote sensing and telemetry. Therefore, the monitoring, forecasting and current forecasting system of ionospheric parameters has been widely valued by scholars at home and abroad, and has also been included in China's space physics research strategy. one. Ionospheric research is used to ensure the reliable operation of systems such as radio communications, radio and television, a...

AI Technical Summary

Problems solved by technology

[0004] The ionospheric forecast empirical models mainly include Klobuchar model, Bent model, IRI model, ICED model and FAIM model, among which the ARIMA model and Klobuchar model are the most commonly used, but their forecast accuracy is not high, and it is unreasonable for night forecast

Images