Automatic siting for air quality monitoring stations

Patent metadata

Field	Value
Publication number	US-10725008-B2
Application number	US-201715494759-A
Country	US
Kind code	B2
Filing date	Apr 24, 2017
Priority date	Apr 24, 2017
Publication date	Jul 28, 2020
Grant date	Jul 28, 2020

Abstract

A system, a computer readable storage medium, and a method for automatically siting for air quality monitoring stations includes collecting from air quality monitoring stations air pollution concentration data, collecting from meteorological stations meteorological data, and interpolating, by the at least one or more processors, the air pollution concentration data to provide interpolated air pollution concentration data. The method and system can further cluster the interpolated air pollution concentration data and the meteorological data to provide clustered data and select a candidate site for an air monitoring station using the clustered data. The method and system can further evaluate a relationship of the air pollution concentration data with the meteorological data by weighting the air pollution data with the meteorological data.

First claim

What is claimed is:

1. A computer implemented method, comprising:
providing a set of air quality monitoring stations individually placed and spaced apart at respective first set of locations in a defined geographic area to measure and monitor air pollution concentration data at the respective first set of locations, the set of air quality monitoring stations being communicatively coupled with a computer system;
providing a set of meteorological stations individually placed and spaced apart at respective second set of locations in the defined geographic area to measure and monitor meteorological data at the respective second set of locations, the set of meteorological stations being communicatively coupled with the computer system;
collecting from the set of air quality monitoring stations, by at least one or more processors of at least one computing device of the computing system, air pollution concentration data at the respective first set of locations;
collecting from meteorological stations, by the at least one or more processors, meteorological data at the respective second set of locations;
interpolating in time and space, by the at least one or more processors, at least the air pollution concentration data collected at the respective first set of locations to provide interpolated air pollution concentration data for more locations than the respective first set of locations in the defined geographic area;
clustering, by the at least one or more processors, the interpolated air pollution concentration data and the meteorological data to provide clustered data;
selecting, by the at least one or more processors, a candidate site for siting an air monitoring station using the clustered data by
identifying a plurality of candidate sites and, for each time slice in a plurality of different time slices, a respective concentration of an air pollutant at each candidate site in the plurality of candidate sites in the defined geographic area respective first set of locations in a defined geographic area;
comparing, for each time slice in the plurality of different time slices, a concentration of the air pollutant at each candidate site to an average of all concentrations of the air pollutant at the plurality of candidate sites;
identifying, for each time slice in the plurality of different time slices, a center of meteorological fields in the defined geographic area;
identifying trajectory of centers of meteorological fields in the defined geographic area, based on combining time slices of clustered data of meteorological data;

applying parameters and constraints to an automatic site finder after the automatic site finder receives all the candidate sites; and
selecting a candidate site from the plurality of candidate sites based on an atmospheric transmission pattern of the air pollutant that is identified with the identified trajectory of centers of meteorological fields and the concentration of the air pollutant at each candidate site in the plurality of candidate sites in the plurality of different time slices; and

siting an air quality monitoring station at the selected candidate site.

2. The computer implemented method of claim 1 , wherein the interpolating comprises dividing the air pollution concentration data by time and applying time sliced data to different geographic areas.

3. The computer implemented method of claim 1 , wherein the method further comprises evaluating a relationship of the air pollution concentration data with the meteorological data by weighting the air pollution concentration data with the meteorological data.

4. The computer implemented method of claim 1 , wherein the method further comprises weighting the meteorological data by season of the year and the air pollution data by type of pollutant.

5. The computer implemented method of claim 1 , wherein the method further comprises using meteorological data to find a center of meteorological fields of one time slice, combining a different time slice to find a trajectory of a meteorological condition in a certain region, matching the trajectory with the air pollution concentration data and trajectory clustering, and generating at least one of the candidate sites.

6. The computer implemented method of claim 1 , wherein the method further comprises automatically generating a center of a field of pressure and a scope of the field of pressure.

7. The computer implemented method of claim 1 , wherein the method further comprises automatically generating a center of a field of pressure and a scope of the field of pressure by matching isobars of pressure and extreme points of pressure, determining the Hausdoff distance of isobars and extreme points and determining the Sobel gradient of the extreme points and a nearest isobar.

8. The computer implemented method of claim 1 , wherein the parameters and constraints comprises calculating a weight of a candidate site based on a reshaping and clustering result, and iteratively calculating an object function to select a final output of the candidate site or candidate sites.

9. The computer implemented method of claim 1 , wherein the parameters comprises one or more of a number of stations, a range of evaluation stations by percentage, range of reference stations by percentage, a range of highly polluting stations by percentage, a range of transform stations by percentage, or a minimum distance between every two stations.

Assignees

• IBM

Inventors

• Dong Jin
• Rui Xiao Guang
• Yao Jian
• Yin Wen Jun
• Zhang Chao
• Zhang Xin
• Zhao yu xin

Classifications

• G01N2015/0046

  in gas, e.g. smoke · CPC title

• G01N2035/00881

  network configurations · CPC title

• G01N35/00871

  Communications between instruments or with remote terminals · CPC title

• G01N15/06

  Investigating concentration of particle suspensions (by weighing G01N5/00; investigating sedimentation of particle suspensions G01N15/04; investigating individual particles G01N15/10) · CPC title

• G01N33/0075 Primary

  for multiple spatially distributed sensors, e.g. for environmental monitoring · CPC title