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

Method for acquiring near-surface ozone concentration based on ultraviolet and visible hyperspectrum

An ozone concentration and hyperspectral technology, applied in the field of ozone profile and near-surface ozone concentration inversion, can solve the problems of high sensitivity, low precision of satellite inversion of bottom ozone, etc., and achieve the effect of reducing inversion error

Active Publication Date: 2021-12-07
MINISTRY OF ECOLOGY & ENVIRONMENT CENT FOR SATELLITE APPL ON ECOLOGY ENVIRONMENT
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this processing method is difficult to avoid the heterogeneity among different payloads of the same satellite, and it is also difficult to deal with the high sensitivity of ozone visible light absorption (VIS) Chappuis band to near-surface ozone absorption. The method of line inversion is still insufficient, and urgently needs to be adjusted and improved
[0005] The EMI load of the GF-5 satellite covers both ultraviolet and visible hyperspectral bands. Therefore, in view of the low accuracy of the current satellite inversion of bottom ozone, the inventors carried out a multi-spectral fusion method and a machine learning inversion method for near-surface ozone concentration. fusion in order to design a method to obtain near-surface ozone concentration based on ultraviolet and visible hyperspectral that can solve the above problems

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 acquiring near-surface ozone concentration based on ultraviolet and visible hyperspectrum
  • Method for acquiring near-surface ozone concentration based on ultraviolet and visible hyperspectrum
  • Method for acquiring near-surface ozone concentration based on ultraviolet and visible hyperspectrum

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0158] The ozone profile in the middle and lower layers of the troposphere was retrieved for the area where Hangzhou City is located. The corresponding ultraviolet band spectrum and visible band spectrum of this area are as follows figure 2 As shown in , the inversion method includes the following steps:

[0159] Step 1, establish the correction coefficient calculation formula for spectral correction, namely

[0160] Step 2, retrieve the GF-5 ultraviolet hyperspectral data and GF-5 visible hyperspectral data in this area, and make sensitivity and error constraints on the spectral data, specifically including the following sub-steps:

[0161] Sub-step a, filter out the pixels with too large absorbing aerosol index, that is, filter out the pixels with AAI value greater than 1; where AAI is calculated by the following formula:

[0162] AAI=-100{Log10[(Iλ 1 / Iλ 2 )meas]-Log10[(Iλ 1 / Iλ 2 )calc]}.

[0163] Sub-step b: use the reflectance at 380nm to estimate the effective ...

Embodiment 2

[0181] Train the inversion model:

[0182] Step 1, retrieve training samples from satellite data, the training samples specifically include aod055 data, ERA5 feature data, TROPOMI Level2 feature data, time feature data and near-surface ozone concentration data obtained by ground base stations. Wherein, the tropospheric low-level ozone profile in the TROPOMI Level2 characteristic data utilizes the method in embodiment 1 to obtain;

[0183] Step 2, set up a Wide&Deep neural network, the input data of the Wide neural network part in the Wide&Deep neural network includes basic feature data and combined features

[0184] The input data of the Deep multilayer neural network part in the Wide&Deep neural network includes basic feature data;

[0185] Among them, the basic feature data is converted into features through the following formula (5), and then the combined features are screened out through the gbdt model;

[0186]

[0187] The transfer function in the Wide neural networ...

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 acquiring near-surface ozone concentration based on ultraviolet and visible hyperspectrums for EMI-load of Gaofen-5 satellites, which overcomes the defects of weak sensitivity of single hyperspectrum inversion ozone profile and heterogeneity among different sensors, and under an optimal estimation framework, a priori error covariance matrix and an observation noise covariance matrix are used for limitation, a truth value is gradually approached in an iteration mode, a high-precision troposphere middle and low layer ozone profile is further obtained, on the basis, a Wide&Deep combined model is constructed and trained, the model obtained through training can obtain memory and generalization capacity at the same time, and inversion precision is improved.

Description

technical field [0001] The invention relates to the technical field of ozone profile and near-surface ozone concentration inversion, in particular to a method for obtaining near-surface ozone concentration based on ultraviolet and visible hyperspectrum. Background technique [0002] The concentration of near-surface ozone pollution in my country is increasing year by year. Monitoring ozone on a regional and global scale is a critical social issue. Due to the scarcity of ground monitoring stations, especially in rural and remote areas, the spatial coverage is insufficient to accurately assess the overall pollution status of the region. Compared with ground monitoring sites, satellite remote sensing can provide ozone concentration information on a large regional scale. At present, most remote sensing monitoring applications in environmental assessment are directly using ozone column concentration to evaluate urban ozone pollution, which cannot accurately locate the man-made po...

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): G06N3/04G06N3/08G06N20/00G01N21/33
CPCG06N3/08G06N20/00G01N21/33G06N3/048G06N3/045
Inventor 马鹏飞陈翠红张玉环王玉周春艳张连华杨晓钰赵少华王中挺朱海涛游代安王飞
Owner MINISTRY OF ECOLOGY & ENVIRONMENT CENT FOR SATELLITE APPL ON ECOLOGY ENVIRONMENT
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com