Unlock instant, AI-driven research and patent intelligence for your innovation.

GNSS-R sea wind inversion method based on multi-dimensional feature mining neural network

A neural network and multi-dimensional feature technology, applied in the research fields of atmospheric science and computer science, can solve the problems of limited inversion accuracy and single characteristic parameters, and achieve the effect of easy transplantation and improvement of overall inversion accuracy

Pending Publication Date: 2022-06-21
NAT SPACE SCI CENT CAS
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method usually chooses a single characteristic parameter, which makes the constructed function model relatively simple, which in turn leads to limited inversion accuracy.

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
  • GNSS-R sea wind inversion method based on multi-dimensional feature mining neural network
  • GNSS-R sea wind inversion method based on multi-dimensional feature mining neural network
  • GNSS-R sea wind inversion method based on multi-dimensional feature mining neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] Embodiment 1 of the present invention proposes a GNSS-R sea wind inversion method based on a multi-dimensional feature mining neural network.

[0057] The original data comes from the L1 band data of the Cyclone Global Navigation Satellite System (CYGNSS), which was launched by NASA at the end of 2016 and is a GNSS-R constellation operational application system consisting of 8 microsatellites in synchronous orbit. , the eight satellites of the satellite system can work simultaneously and provide high spatial and temporal resolution data with latitude coverage between 38°N and 38°S. The true value of wind speed was selected from the reanalysis data of the European Centre for Medium-Range Weather Forecasts (ECMWF).

[0058] like figure 1 shown, the specific steps are as follows:

[0059] The first step, original data collection and feature value selection: select the observations from the CYGNSS L1 band as the feature input, the true wind speed value is from ECWMF, and ...

Embodiment 2

[0065] Embodiment 2 of the present invention proposes a GNSS-R sea wind inversion system based on multi-dimensional feature mining neural network, which is implemented based on the method of Embodiment 1. The system includes: a wind speed inversion model, a preprocessing module, and an inversion output module, which,

[0066] The preprocessing module is used for selecting different types of characteristic parameters from the collected GNSS-R data of the GNSS reflection signal, and performing preprocessing and format conversion;

[0067] The inversion output module is used for inputting the format-converted feature parameters into a pre-established and trained wind speed inversion model to obtain the inversion wind speed value;

[0068] The wind speed inversion model realizes wind speed inversion by mining data information between different types of characteristic parameters and extracting the correlation of data time.

[0069] The invention designs and builds a hybrid neural ...

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 GNSS-R sea wind inversion method based on a multi-dimensional feature mining neural network, and the method comprises the steps: selecting different types of feature parameters from collected global navigation satellite system reflected signal remote sensing GNSS-R data, and carrying out the preprocessing and format conversion; inputting the characteristic parameters after format conversion into a pre-established and trained wind speed inversion model to obtain an inverted wind speed value; the wind speed inversion model realizes wind speed inversion by mining data information among different types of characteristic parameters and extracting correlation of data time. According to the method, the building, training and inversion of the model can be realized on a single working machine, the overall inversion precision of the model is high, the model is easy to transplant, cross-platform application is convenient to realize, the complex nonlinear relationship between a plurality of features and the sea surface wind speed is established, and a new feature reference is provided for GNSS-R sea surface wind speed inversion.

Description

technical field [0001] The invention relates to the research fields of atmospheric science and computer science, in particular to a GNSS-R sea wind inversion method based on a multi-dimensional feature mining neural network. Background technique [0002] Ocean surface wind speed is an important physical parameter of ocean state, which has an important impact on global and local climate. Forecasting, ensuring the safety of marine fisheries and military activities are also inseparable from the monitoring and research of wind fields on the sea surface. [0003] Global Navigation Satellite System Reflectometry (GNSS-R for short) technology is a relatively new technology that regards navigation satellites as emission sources, and receives and processes the reflected signals of navigation satellites through the onboard receiver to obtain the corresponding physical feature information. Remote sensing technology. Compared with traditional wind measurement methods, it has the advan...

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): G06F17/10G06N3/04G06N3/08
CPCG06F17/10G06N3/084G06N3/045
Inventor 白伟华刘小煦孙越强杜起飞刘黎军李伟王先毅蔡跃荣夏俊明孟祥广柳聪亮谭广远尹聪胡鹏黄飞雄王冬伟刘成吴春俊李福乔颢程双双刘艳
Owner NAT SPACE SCI CENT CAS