Method for forecasting solar flare outbreak based on 3D convolutional neural network

A convolutional neural network and solar flare technology, applied in the fields of astronomy and image processing, can solve the problem of inability to make full use of the time dimension information of solar observation data, and achieve the effect of avoiding picture distortion, reducing complexity and improving accuracy.

Active Publication Date: 2021-03-26
KUNMING UNIV OF SCI & TECH
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the above-mentioned solar flare prediction models based on convolutional neural networks use traditional 2D convolutional neural networks, and their convolution kerne

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method for forecasting solar flare outbreak based on 3D convolutional neural network
  • Method for forecasting solar flare outbreak based on 3D convolutional neural network
  • Method for forecasting solar flare outbreak based on 3D convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Embodiment 1: a kind of method that forecasts the outbreak of solar flare based on 3D convolutional neural network, comprises: step 1, constructs observation data cube, and is divided into training set and test set; Step 2, adopts training set to 3D volume The convolutional neural network model is trained to obtain a trained 3D convolutional neural network model; Step 3, the test set is input into the trained 3D convolutional neural network model to obtain prediction results and evaluate the results.

[0038] Further, the step 1 may be set to include: a step of obtaining raw observation data and solar activity record data of the solar active region; a step of preprocessing the observation data; and a step of constructing and classifying a data cube.

[0039] Further, the step 1 can be specifically set as:

[0040] S1.1. Obtain the original observation data of the solar active region, that is, the full-helix longitudinal magnetic map provided by SDO / HMI (Solar Dynamics O...

Embodiment 2

[0056] Embodiment 2: For a method of forecasting the outbreak of solar flares based on a 3D convolutional neural network, the following experimental steps are provided:

[0057] The 3D convolutional neural network model described in the present invention uses such as figure 1 The main feature of the 3D convolution technology shown is that it can perform convolution calculations in a three-dimensional space including the time dimension. 3D convolution technology compared to figure 2 The traditional convolution technique shown adds a time dimension, so it can effectively extract information including the time dimension from continuous solar observation data, and effectively improve the amount of information learned by the 3D convolutional neural network model, so as to improve the efficiency of the sun. The purpose of flare forecast accuracy.

[0058] The invention uses the data cube constructed by continuous observation data within 48 hours of each solar active area as input...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a method for forecasting solar flare outbreak based on a 3D convolutional neural network. The method comprises the following steps: 1, constructing an observation data cube byadopting continuous observation data; 2, training the 3D convolutional neural network model by adopting the training set; and 3, inputting the test set into the trained 3D convolutional neural networkmodel for prediction. According to the method, a series of solar activity area images which are continuous in time are positioned, tracked, intercepted and corrected, so that the problem of image distortion subsequently input into a model can be avoided, and a result is more realistic; and the processed data is matched with a specific 3D convolutional neural network of the method, and time sequence information contained in solar continuous observation data can be fully extracted, so that evolution process information of an activity area is captured, and the accuracy of solar flare forecastingof the model is effectively improved; and meanwhile, due to the fact that continuous sun observation data is used, the complexity of observation data collection and arrangement work is effectively reduced.

Description

technical field [0001] The invention relates to a method for forecasting the outbreak of solar flares based on a 3D convolutional neural network, which belongs to the fields of astronomical technology and image processing. Background technique [0002] With the development of human science and technology, especially the continuous deepening of space exploration and development activities, the impact of space weather on human activities continues to deepen. The solar flare outbreak is an important manifestation of the harsh conditions of space weather. High-intensity solar flares will pose a threat to radio communications, satellite navigation, and even space flight and the safety of astronauts. Therefore, it is of great significance to accurately predict the outbreak of solar flares to avoid or reduce the harm of solar flares to human beings. [0003] Solar flare is a violent eruption phenomenon of the sun, and the research on the prediction of solar flare explosion has alw...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G01W1/10G06N3/04G06N3/08
CPCG01W1/10G06N3/08G06N3/045Y02A90/10
Inventor 冯松丁维奇
Owner KUNMING UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap