A pm2.5 inversion method and monitoring area segmentation method
A monitoring area and inversion technology, which is applied in the monitoring of particulate matter in the air and in the field of PM2.5 inversion, can solve the problems of increasing the missing rate of satellite AOD and low accuracy of AOD inversion, so as to improve the spatial coverage and better anti-noise Ability to ensure stability
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 2
[0073] Example 2 The PM2.5 inversion method is provided, on the basis of the monitoring area segmentation method of the first example one, the monitoring area is divided into a number sub-area, and then the PM2.5 inversion is performed according to the following steps:
[0074] S410. Establish a random forest regression model to each sub-region, and the weather dynamic indicators and satellite AODs are entered into the random forest network. After the random forest network, the random forest regression model is optimally model, in the optimal model, inversion to find each sub-area PM2.5 concentration satellite estimation value In the meteorological dynamic indicator, for example, a surface temperature, a surface pressure, a wind speed, and a relative humidity and the boundary layer height.
[0075] The AOD is an aerosol optical thickness, and the integral of the aerosol extinction coefficient in the vertical direction, quantitatively describes the physical quantity of the gas sol...
Embodiment 3
[0086] Based on the second embodiment, the following steps further include:
[0087] S460, establish PM2.5 concentration satellite estimation value And space interpolation Fit function.
[0088] The corresponding site data of the PM2.5 concentration estimation value of the corresponding ground site data is calculated according to the fit function.
[0089] Satellite observations exist data missing, and interpolation interpolation Since the site is sparse and distributed uneven, the interpolation accuracy is also fluctuated in space, and the results of the two can be combined, and the error can be reduced to a certain extent, and the spatial coverage of the inversion results can be increased. The specific implementation process is divided into two steps, first establishing the fitting functions between the two, and there is no satellite observation, but there is space interpolation The pixels are rebuilt for satellite inversion results to fill the missing data.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


