Severe convective weather prediction method and system for improving three-dimensional generative adversarial neural network based on hybrid evolutionary algorithm

A hybrid evolution and weather forecasting technology, applied in biological neural network models, weather condition forecasting, neural learning methods, etc., can solve the problem of low forecasting accuracy and achieve the effect of improving forecasting accuracy

Active Publication Date: 2022-05-17
HENAN UNIVERSITY
View PDF9 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] Aiming at the problem of low prediction accuracy existing in existing nowcasting methods, the present invention proposes a strong convective weather prediction method and system based on a hybrid evolutionary algorithm to improve the three-dimensional confrontation generation neural network, through the extraction of the original Doppler radar The radar echo data trains the neural network, and at the same time predicts the radar echo data, and judges whether there is strong convective weather based on the radar echo data, which can effectively improve the accuracy of strong convective weather prediction

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
  • Severe convective weather prediction method and system for improving three-dimensional generative adversarial neural network based on hybrid evolutionary algorithm
  • Severe convective weather prediction method and system for improving three-dimensional generative adversarial neural network based on hybrid evolutionary algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0069] The present invention will be further explained below in conjunction with accompanying drawing and specific embodiment:

[0070] Such as figure 1 As shown, a strong convective weather prediction method based on a hybrid evolutionary algorithm to improve the three-dimensional confrontation generative neural network, including:

[0071] Step S101: Read the radar echo data from the original Doppler weather radar, store the altitude, distance and azimuth in the Doppler radar echo data as a dimension into a three-dimensional matrix respectively, and generate (three-dimensional confrontation generation neural network model) training data set, and divide the training data set into multiple sets of input data.

[0072] Step S102: Construct a hybrid evolutionary algorithm by using the genetic algorithm and the cross-entropy algorithm. Genetic algorithm is a kind of evolutionary algorithm, which can jump out of the trap of local optimal solution and has good global search perfo...

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 severe convective weather prediction method and system for improving a three-dimensional generative adversarial neural network based on a hybrid evolutionary algorithm, and the method comprises the steps: reading radar echo data from an original Doppler weather radar, generating a training data set, and dividing the training data set into a plurality of groups of input data; constructing a hybrid evolutionary algorithm by using a genetic algorithm and a cross entropy algorithm; establishing an improved three-dimensional confrontation generation neural network model based on a hybrid evolutionary algorithm; training the three-dimensional generative adversarial neural network model through the training data set to obtain a trained three-dimensional generative adversarial neural network model; the number N of radar echo data needing to be input is obtained according to the to-be-predicted time range, and the latest N pieces of preprocessed radar echo data are input into the trained three-dimensional confrontation generation neural network model for severe convective weather prediction. The severe convective weather prediction accuracy can be effectively improved.

Description

technical field [0001] The invention belongs to the technical field of short-term near-weather prediction, and in particular relates to a strong convective weather prediction method and system based on a hybrid evolution algorithm to improve a three-dimensional confrontation generation neural network. Background technique [0002] Severe convective weather is a very dangerous weather phenomenon in summer in my country. It may be accompanied by hail, lightning, strong wind and short-term heavy rainfall. Due to the small spatial scale and short duration of this kind of weather, it is difficult to effectively predict it with conventional observation instruments. Doppler weather radar, with its high spatial and temporal resolution, is currently the only practical instrument capable of routinely sampling the detailed three-dimensional structure of severe convective storms. Therefore, various nowcasting algorithms based on Doppler weather radar data have been developed to assist ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G01W1/10G01S7/41G06K9/00G06K9/62G06N3/04G06N3/08
CPCG01W1/10G01S7/417G06N3/088G06N3/045G06F2218/02G06F2218/08G06F18/214Y02A90/10
Inventor 张磊孟坤颖沈夏炯韩道军丁文珂
Owner HENAN UNIVERSITY
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