Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Rainfall nowcasting method and device based on deep learning

A nowcasting and deep learning technology, applied in the information field, can solve problems such as insufficient attention in strong echo prediction, difficulty in predicting radar echo time-space sequence, etc., and achieve the effect of enhancing prediction ability

Pending Publication Date: 2022-01-14
CHENGDU UNIV OF INFORMATION TECH +1
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the traditional model training does not pay enough attention to the strong echo prediction and solve the problem of difficult prediction of the radar echo time-space sequence

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
  • Rainfall nowcasting method and device based on deep learning
  • Rainfall nowcasting method and device based on deep learning
  • Rainfall nowcasting method and device based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] For the convenience of those skilled in the art to better understand the technical solution of the present invention, the present invention is described in detail as follows:

[0060] This application first uses the precipitation data collected by Beijing Radar from June to September 2015 to 2016, and the precipitation data collected from Haikou and Dongfang Radar from 2018 to September to December 2019 for data preprocessing, including combined reflectance generation and normalization processing , outlier filtering, clipping, etc. into standard data formats available for model training, and establish a radar echo data set. Afterwards, the ConvLSTM network structure, which combines the advantages of convolutional neural network and long-term short-term memory network, is combined with the Encoder-Forecast model to solve the problem of radar echo time-space sequence prediction. W-MSE and W-MAE are combined to obtain a weight loss function, which is used as a loss functio...

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 relates to the technical field of information, and provides a rainfall nowcasting method and device based on deep learning. The objective of the invention is to solve the problems of insufficient attention of traditional model training on strong echo prediction and difficulty in radar echo space-time sequence prediction. According to the main scheme, the method comprises the following steps: S1, preprocessing weather radar base data; S2, dividing the preprocessed radar echo data into a training set, a verification set and a test set for deep learning network training; S3, performing space-time coding prediction network model training by using the training set, the verification set and the test set, performing radar echo extrapolation by using the trained space-time coding prediction network model to obtain predicted echo data, and comparing the predicted echo data with actual observation data; and S4, finally, carrying out radar quantitative rainfall estimation on the predicted radar echoes, and comparing the radar echoes with ground real rainfall data to carry out rainfall forecast detection.

Description

technical field [0001] The invention relates to the field of information technology, and provides a method and device for nowcasting precipitation based on deep learning. Background technique [0002] At present, the research methods of precipitation nowcasting are mainly based on the radar echo extrapolation technology based on weather radar data and the numerical weather prediction (Numerical Weather Prediction, NWP) technology. The traditional radar echo extrapolation technology is based on the radar observations at several moments The radar echo data is used to infer the future moving position and intensity changes of the radar echo, so as to realize the tracking and forecasting of strong convective weather. Radar echo area tracking algorithm. The radar echo cell centroid tracking algorithm first identifies the storm cell, and then linearly infers the echo position at the next moment by fitting the path of the identified radar echo centroid, but this method relies on th...

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): G06V10/40G06V10/774G06K9/62G06N3/04G01S13/95G01S7/41G01W1/10
CPCG01S13/95G01S7/418G01W1/10G06N3/044G06F2218/08G06F18/214Y02A90/10
Inventor 苏德斌史磊唐田野杨雅婷孙晓光曹杨郭在华樊昌元
Owner CHENGDU UNIV OF INFORMATION TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products