Forecasting method of solar eruption events based on machine learning technology forecasting model

Abstract

The invention discloses a solar burst event forecasting method based on machine learning technology forecasting model. The method includes: A. a step of depicting solar active region magnetic field characteristics; and B. a step of establishing a solar burst event forecasting model. By adoption of the method provided by the invention, solar active region photosphere vector magnetic field measurement can be utilized to establish the solar burst event comprehensive forecasting model through a machine learning technology, thereby realizing comprehensive forecasting of solar flares and coronal mass ejection which are related with the solar active region.

Description

technical field

[0001] The present invention relates to solar activity forecasting technology, in particular to a solar eruption event forecasting method based on machine learning technology forecasting model. Background technique

[0002] The sun is the source of space weather. Vigorous eruptions originating from the sun produce a large number of high-energy particles and strong electromagnetic radiation. When a large number of high-energy particles reach near the earth's orbit, it may affect the safety of satellites in geosynchronous orbit and astronauts, and when they reach near the earth's upper atmosphere, they may cause geomagnetic storms; when electromagnetic radiation reaches near the earth, it may damage the earth's ionosphere, thereby affecting radio communication. Therefore, the prediction of solar eruption events has important practical value.

[0003] Solar flares and coronal mass ejections are two of the most violent solar eruptive events. In the prediction...

Images