A method for measuring the distribution of urban atmospheric pollutants based on cross-scan

Abstract

The application discloses a method for measuring horizontal distribution of urban atmospheric pollutants based on cross scanning, comprising the following steps: setting multiple atmospheric composition hyperspectral scanners and performing grid processing on a target observation area; determining a sequence of observation azimuth angles of the scanners based on the grid area, changing the azimuth angle observation level atmospheric scattering spectrum alternately through any two adjacent scanners according to the sequence of observation azimuth angles, and obtaining the differential slant column density of atmospheric pollutants based on the horizontal atmospheric scattering spectrum; determining the weight of each grid according to the light path of each scanner passing through the grid, constructing an equation set based on the weight and the differential slant column density of atmospheric pollutants observed by each scanner, and performing iterative tomographic inversion on the equation set under the constraint condition to obtain the high-resolution horizontal concentration distribution information of each atmospheric pollutant, thereby improving the horizontal detection precision.

Description

Technical Field

[0001] The invention belongs to the technical field of optical measurement, and in particular relates to a method for measuring the horizontal distribution of urban air pollutants based on cross scanning. Background Art

[0002] With the rapid growth of the economy and the accelerating pace of urbanization, air pollution problems are constantly emerging. Today, air pollution incidents still exhibit significant regional uneven distribution and frequent regional transmission. Cities are home to numerous industrial and transportation air pollution sources. The diffusion, transmission, and photochemical reactions of primary emissions of pollutants in the atmosphere result in complex, regional air pollution at multiple scales, including point, surface, and domain. As air pollutant concentrations continue to decrease, further improving air quality becomes increasingly difficult, necessitating more refined monitoring to support more sophisticated pollution prevention and control efforts.

[0003] Currently, there are two main methods for detecting the horizontal distribution of atmospheric pollution components using ground-based remote sensing methods. The first method is mainly based on passive solar light sources and the onion peeling algorithm. The method is based on the different light absorption of different gas components in different bands. The effective optical path corresponding to instruments in different bands is calculated through relatively stable gases (such as O2, N2, etc.), and then the onion peeling algorithm is used to calculate the horizontal distribution of different trace gases in the atmosphere. However, its limitation is that the spatial resolution is limited, and the general resolution reaches several kilometers. At the same time, this method is greatly affected by the absorption of aerosols and trace gases in different bands.

[0004] The second method is mainly based on active light sources combined with echo signals and segmentation algorithms. The active light source of this method often uses a relatively stable and long-distance laser. The initial distance is determined by the signal-to-noise ratio threshold in the laser radar's echo signal, and then the effective optical path is analyzed through segmented intervals to obtain the horizontal distribution of different atmospheric components in different segmented intervals. However, its limitation is that the laser itself has a monitoring blind spot at a relatively close distance, and this method can monitor fewer aerosol and trace gas components.

[0005] Because both of the above two methods have certain technical defects, neither method can meet the requirements of accurate detection of actual pollution areas in detecting the horizontal distribution of atmospheric pollution components.

[0006] The atmospheric composition hyperspectral scanner is a new type of hyperspectral instrument for detecting the distribution of atmospheric trace gases and aerosols. It uses solar scattered light as a light source and collects solar scattered light spectra from the zenith and multiple off-axis directions. These spectra contain information such as the absorption, reflection and scattering of trace gases, aerosols and clouds.

[0007] Therefore, there is an urgent need for a hyperspectral scanner that combines atmospheric composition to solve the technical defects of the above two detection methods and achieve accurate measurement of the distribution of atmospheric pollutants. SUMMARY

[0008] In view of the above, the purpose of the present application is to provide a method for measuring the horizontal distribution of urban atmospheric pollutants based on cross-scan, which realizes cross-scan observation in the target observation area by multiple atmospheric composition hyperspectral scanners in dense observation azimuth angles, and further realizes more accurate horizontal monitoring of atmospheric pollution components by combining with tomographic imaging technology, especially obtaining the horizontal distribution of atmospheric pollution components with ultra-high spatial resolution.

[0009] To achieve the above-mentioned purpose of the application, the method for measuring the horizontal distribution of urban atmospheric pollutants based on cross-scan provided by the embodiment comprises the following steps:

[0010] Multiple atmospheric composition hyperspectral scanners are arranged at the boundary of the target observation area, and the distance between any two scanners is ensured to be within the detection range of the scanner;

[0011] The target observation area is subjected to grid processing, ensuring that the light path information of at least one scanner is covered in each grid, and the key observation area contained in the target observation area must cover the light path information of multiple scanners;

[0012] The observation azimuth angle sequence of the scanner is determined based on the grid area, the horizontal atmospheric scattering spectrum is observed by alternately changing the azimuth angle observation level of any two adjacent scanners according to the azimuth angle observation sequence, and the differential slant column density of atmospheric pollutants is obtained based on the horizontal atmospheric scattering spectrum;

[0013] The weight of each grid is determined according to the light path of each scanner through the grid, an equation set is constructed based on the weight and the differential slant column density of atmospheric pollutants observed by each scanner, and the horizontal concentration distribution information of each atmospheric pollutant is obtained by iterative tomographic inversion of the equation set under the constraint condition.

[0014] Preferably, the differential slant column density of atmospheric pollutants is obtained based on the horizontal atmospheric scattering spectrum, comprising:

[0015] The horizontal atmospheric scattering spectrum of the zenith is taken as the reference spectrum, and the least square method based on characteristic absorption is used to solve the horizontal atmospheric scattering spectrum at different azimuth angles to obtain the differential slant column density of atmospheric pollutants.

[0016] Preferably, the weight of each scanner is determined according to the light path of each scanner through the grid, comprising:

[0017] The intersection positions of the light path of the scanner through each grid boundary are calculated according to the spatial position and observation azimuth angle of the scanner, and then the light path length of each grid is calculated according to the distance between the two intersection positions on the same grid, which is taken as the weight of each grid for the differential slant column density observed by a single scanner.

[0018] Preferably, the relationship between the differential slant column density of atmospheric pollutants observed by the single scanner and the grid weight is:

[0019]

[0020] where SCD n,k represents the differential slant column density data of atmospheric pollutants obtained by the nth scanner under k azimuth observations, W ij represents the weight of the grid in the ith row and jth column, and c ij represents the atmospheric pollutant concentration to be solved in the grid in the ith row and jth column.

[0021] Preferably, the introduced constraint condition is:

[0022]

[0023] where represents the atmospheric pollutant concentration in the grid in the ith row and jth column solved, and represents the atmospheric pollutant concentration after the constraint.

[0024] Preferably, the equation group is constructed based on the weight and the differential slant column density of atmospheric pollutants observed by each scanner, including:

[0025] S = Wc

[0026] where S represents the differential slant column density matrix composed of SCD n,k , W represents the weight matrix composed of W ij , and c represents the atmospheric pollutant concentration matrix to be solved composed of c ij .

[0027] The atmospheric pollutant horizontal concentration in the target observation area is obtained by iterative tomographic inversion of the equation group under the constraint condition.

[0028]

[0029] where argmin represents the value of |S-Wc| 2 when |S-Wc| is minimized.

[0030] In the inversion process, the initial concentration of each grid needs to be assumed first, which is generally based on the annual or monthly average concentration of the nearby national control site or region. First, it is assumed that the concentration in the observation area is uniform, and the initial concentration c is set. After that, the predicted output of the model is calculated using the current parameter value:

[0031]

[0032] Error vector of the current calculated and the actual SCD:

[0033]

[0034] Based on the e, the optimal direction is indicated, so as to obtain the suitable gradient Δc of , so as to update the concentration of the next round of iteration:

[0035]

[0036] Finally, if the iterative process gradually converges, and the change of the final target function is less than the preset threshold ε (e<ε), the concentration of each grid of the atmospheric composition obtained by the tomographic inversion method is obtained

[0037] Preferably, the least square method is used to solve to obtain the concentration of all grids in the target observation area so as to determine the position of the high concentration, and analyze the emission source information.

[0038] Preferably, when the scanners are arranged at the boundary of the target observation area, the distance between the scanners is 5-10 kilometers.

[0039] Compared with the prior art, the present application has at least the following beneficial effects:

[0040] The horizontal atmospheric scattering spectrum of different azimuth angles is collected by the multiple atmospheric composition hyperspectral scanners at the boundary of the target observation area, the multi-azimuth collection full coverage of the target observation area is realized, the concentration data corresponding to each azimuth angle is decomposed into the sum of the product of the concentration data of all grids and the weight, and the equation set is constructed based on this, and the accurate horizontal concentration distribution of the atmospheric pollutants in the target observation area is obtained by iterative tomographic inversion under the constraint condition. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0042] Figure 1 is a method flowchart for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning;

[0043] Figure 2 and Figure 3 is a method technical principle diagram for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning; is a method technical principle diagram for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning;

[0044] Figure 4 is a case area map of a method for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning;

[0045] Figure 5 is a flow chart of weight function calculation in the method for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning;

[0046] Figure 6 is a flow chart of tomographic inversion in the method for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning. DETAILED DESCRIPTION

[0047] In order to make the objectives, technical solutions and advantages of the present application clearer, further detailed description will be made to the present application in combination with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application, and do not limit the protection scope of the present application.

[0048] As shown in Figure 1 , the embodiment provides a method for measuring the horizontal distribution of urban atmospheric pollutants based on cross scanning, which comprises the following steps:

[0049] S1, multiple atmospheric composition hyperspectral scanners are arranged at the boundary of the target observation area, and the distance between any two scanners is within the detection range of the scanner.

[0050] The multiple atmospheric composition hyperspectral scanners all have high-resolution bands from ultraviolet to visible light, are installed at the boundary position of the target observation area, the optimal distance between the scanners is 5-10 km, the scanners are set up at a site selected as far as possible in a region without obvious obstruction. The observation angle is adjusted by rotation, the range includes azimuth angle which can be adjusted to 360° panorama with an accuracy of 0.1°, the observation elevation angle is fixed as horizontal, and the distance between any two scanners is ensured to be within the detection range of the instrument as far as possible. Multiple azimuth angles are set in the region in the horizontal direction by rotating the observation range by a sector.

[0051] In the case, an industrial park in a certain city is selected as the target observation area, the area is 6 km x 7 km, and then two atmospheric composition hyperspectral scanners are installed at the corners of the 7 km boundary area of the target observation area, as shown in Figures 2-4 , wherein Sp1, Sp2 and Sp3 represent the positions of the scanners, k represents the observation azimuth angle of the equipment, and S represents the distance between any two scanners. kThe light path corresponding to the observation azimuth angle k is represented, a is the observation elevation angle, h is the area height, d is the area width, and c(x, y) represents any point in the polluted air mass. The target observation area of the two scanners avoids horizontal and higher position shielding objects as much as possible, and the distance between the instruments can be ensured to be less than the observation light path in sunny daytime. The two atmospheric composition hyperspectral scanners are set to collect horizontal solar scattering spectra by rotating the pitch angle and azimuth angle of the outer machine.

[0052] S2, the target observation area is gridded to ensure that the light path information of at least one scanner is covered in each grid, and the key observation area contained in the target observation area must cover the light path information of multiple scanners.

[0053] In the embodiment, the area is gridded based on the building distribution information of the target observation area, as shown in Figure 3 It must be ensured that the light path information of at least one scanner is covered in each grid, and it is necessary to cover the light path information of multiple scanners in the key observation area, because the area grid is covered by multiple light paths, which increases the reliability of the inversion result.

[0054] In the case, the target observation area is exemplarily divided into 1 km x 1 km grids, and of course the area can be more refined, so that one of A1~F (left) can observe the observation area with an azimuth angle interval of 5°~85° at every 5°, and the other instrument (right) can observe the observation area with an azimuth angle interval of -85°~-5° at every 5°. The observation paths of the two instruments have more overlaps, and the grids in the central key observation pollution area are basically covered by 5 light path information in each grid.

[0055] S3, the observation azimuth angle sequence of the scanner is determined based on the gridded area, the horizontal atmospheric scattering spectrum is observed by alternately changing the azimuth angle of any two adjacent scanners according to the azimuth angle observation sequence, and the differential slant column density of atmospheric pollutants is obtained based on the horizontal atmospheric scattering spectrum.

[0056] In the embodiment, the horizontal atmospheric scattering spectrum obtained by alternately changing the azimuth angle of the multiple scanners in one azimuth angle sequence, and taking the horizontal atmospheric scattering spectrum at zenith as the reference spectrum, the least square method based on characteristic absorption is used to inverse solve the horizontal atmospheric scattering spectrum at different azimuth angles to obtain the differential slant column density data SCD of different trace gases. The SCD data in the target observation area in a time period is regarded as a group of results, which is used for the next inversion to obtain the horizontal distribution of atmospheric pollutants in the area.

[0057] In the case, the differential slant column densities of different atmospheric pollutants at the corresponding azimuth angle can be obtained by inversion of the spectra collected by each 5° azimuth angle. If the collection time of a spectrum is 30 s, then it takes about 10 minutes to complete a set of collection. It is generally believed that the concentration of regional trace gases and other atmospheric pollutants will not change significantly within a short period of 10 minutes, and thus a set of SCD data SCD N,K where N is the number of instruments installed, and K is the number of set azimuth angles. In the case, there are a total of 2 instruments, so the total number of N is 2, and the number of azimuth angles for each instrument is 17, so the total number of K is 17. Here, a set of typical SCD results are selected as the reference values of the case:

[0058] Instrument 1 Figure 4 Left side resting) Instrument 1 Figure 4 Right side up) ± 5° azimuth 2.4931e+17 molec / cm 2 ]] 2.7270e+17 molec / cm 2 ]] ± 10° azimuth 2.6366e+17 molec / cm 2 ]] 2.8503e+17 molec / cm 2 ]] ± 15° azimuth 3.2498e+17 molec / cm 2 ]] 3.0162e+17 molec / cm 2 ]] ± 20° azimuth 3.6812e+17 molec / cm 2 ]] 3.3723e+17 molec / cm 2 ]] ± 25° azimuth 3.9323e+17 molec / cm 2 ]] 3.9926e+17 molec / cm 2 ]] ± 30° azimuth 4.2511e+17 molec / cm 2 ]] 4.3422e+17 molec / cm 2 ]] ± 35° azimuth 4.6326e+17 molec / cm 2 ]] 4.5968e+17 molec / cm 2 ]] ± 40° azimuth 5.0518e+17 molec / cm 2 ]] 5.1671e+17 molec / cm 2 ]] ± 45° azimuth 4.9146e+17 molec / cm 2 ]] 4.9986e+17 molec / cm 2 ]]> ± 50° azimuth 4.2104e+17 molec / cm 2 ]] 4.0271e+17 molec / cm 2 ]] ± 55° azimuth 3.6098e+17 molec / cm 2 ]]> 3.3903e+17 molec / cm 2 ]] ± 60° azimuth 3.2325e+17 molec / cm 2 ]] 3.1930e+17molec / cm 2 ]] ± 65° azimuth 2.8678e+17 molec / cm 2 ]] 3.0831e+17 molec / cm 2 ]] ± 70° azimuth 2.672e+17 molec / cm 2 ]] 3.0233e+17 molec / cm 2 ]] ± 75° azimuth 2.4720e+17 molec / cm 2 ]] 2.7245e+17 molec / cm 2 ]]> ± 80° azimuth 2.0468e+17 molec / cm 2 ]] 2.3873e+17 molec / cm 2 ]] ± 85° azimuth 2.006e+17 molec / cm 2 ]] 2.3376e+17 molec / cm 2 ]]

[0059] S4, determining the weight of each grid according to the situation of the light path of each scanner passing through the grid, constructing an equation set based on the weight and the differential slant column densities of atmospheric pollutants observed by each scanner, and performing iterative tomographic inversion on the equation set under the constraint condition to obtain the horizontal concentration distribution information of each atmospheric pollutant.

[0060] Based on each set of observed SCD data, considering that the concentration of a small distance through a pollution air mass has the greatest contribution to the SCD at the azimuth angle k on the observation direction corresponding to the azimuth angle k, the weight of each grid is introduced. Specifically, as shown in Figure 5 the intersection positions of the light path of the scanner passing through each grid boundary are calculated according to the spatial position of the scanner and the observation azimuth angle, and then the light path length of each grid is calculated according to the distance between the two intersection positions on the same grid, which is taken as the weight of each grid to the differential slant column density observed by a single scanner.

[0061] In the case, the target observation area of the two scanners is divided into a 1 km x 1 km area. Thus, the target observation area is divided into 42 sub-areas, and the path information of each grid for each observation SCD n,kContribution weight. The estimation of weight mainly relies on judging the distance that the observation light path passes through each grid. In this case, the distance is calculated at a 40° azimuth angle of the instrument on the left. It can be found that the 40° azimuth angle passes through a total of 12 grids, F1, F2, E2, E3, D3, D4, C4, C5, B5, B6, A6, and A7. This can be regarded as the light path that passes through the most grids. The parallel plane formula can be used to calculate the weights of the 12 grids as 1.3054, 0.2503, 1.0551, 0.5006, 0.8048, 0.7509, 0.5545, 1.0013, 0.3041, 1.2516, 0.0538, and 1.3054, respectively.

[0062] In this embodiment, it is assumed that the horizontal atmospheric pollutant concentration in each area is constant, and considering that the pollutant gas concentration at different distances has different contributions to the SCD, a weight matrix W is introduced to convert the pollutant gas SCD in different directions into multiple equations related to the grid concentration and the observation optical path (that is, the weight) according to the grid:

[0063]

[0064] Among them, SCD n,k W represents the differential oblique column concentration data of atmospheric pollutants obtained by the nth scanner under observation at k azimuth angles, ij represents the weight of the grid in row i and column j, c ij represents the concentration of atmospheric pollutants to be solved in the grid of row i and column j, S ij Represents the differential oblique column concentration data for the i-th row and j-th column grid.

[0065] In the case, Figure 4 For example, the differential oblique column concentration data SCD obtained by the scanner at each azimuth angle n,k It can be expressed as multiple equations, and the following equations are given as examples: SCD 1,5 °=1.0038·C F1 +1.0038·C F2 +1.0038·C F3 +1.0038·C F4

[0066] +1.0038·C F5 +1.0038·C F6 +1.0038·C F6 SCD 1,10 °=1.0154·C F1 +1.0154·C F2 +1.0154·C F3 +1.0154·C F4

[0067] +1.0154 °C F5 +0.6816 °C F6 +0.3338 °C E6 +1.0154 °C E7 SCD 1,15 = 1.0353 °C F1 +1.0353 °C F2 +1.0353 °C F3 +0.7579 °C F4

[0068] +0.2774 °C E4 +1.0353 °C E6

[0069] SCD 1,20° = 1.0642 °C F1 +1.0642 °C F2 +0.7954 °C F3 +0.2687 °C E4

[0070] +1.0642 °C E5 +0.5267 °C E6 +0.5375 °C D6 +1.0642 °C D7 SCD 1,25° = 1.1034 °C F1 +1.1034 °C F2 +0.1594 °C F3 +0.9439 °C E3

[0071] +1.1034 °C E4 +0.3189 °C E5 +0.7845 °C D5 +1.1034 °C D6

[0072] +0.4783 °C D7 +0.6250 °C C7

[0073] SCD 1,30° = 1.1034 °C F1 +1.1034 °C F2 +0.1594 °C F3 +0.9439 °C E3

[0074] +1.1034 °C E4+0.3189 °C E5 +0.7845 °C D5 +1.1034 °C D6

[0075] +0.4783 °C D7 +0.6250 °C C7

[0076] SCD 1,35° = 1.1547 °C F1 +0.8453 °C F2 +0.3094 °C E2 +1.1547 °C E3

[0077] +0.5359 °C E4 +0.6188 °C D4 +1.1547 °C D5 +0.2265 °C D6

[0078] +0.9282 °C C6 +1.0718 °C C7 +0.0829 °C B7 SCD 1,40° = 1.2208 °C F1 +0.5227 °C F2 +0.6981 °C E2 +1.0453 °C E3

[0079] +0.1754 °C D3 +1.2208 °C D4 +0.3481 °C D5 +0.8735 °C C5

[0080] +0.8699 °C C6 +0.3509 °C B6 +1.2208 °C B7

[0081] SCD 1,45° = 1.4141 °C F1 +1.4141 °C E2 +1.4141 °C D3 +1.4141 °C C4

[0082] +1.4141 °C B5 +1.4141 °C A6

[0083] Wherein, A, B, C, D, E, F represent different grids respectively.

[0084] In the embodiment, the equation group is constructed based on the weight and the differential slant column concentration of the atmospheric pollutants observed by each scanner:

[0085] S = Wc

[0086] Wherein, S represents the differential slant column concentration matrix composed of SCD n,k , W represents the weight matrix composed of W ij , and c represents the atmospheric pollutant concentration matrix to be solved composed of c i,j .

[0087] In order to obtain the gas concentration information of different grids, the concentration c 2 in the target grid is determined by obtaining |S-Wc| i,j . At the same time, in order to avoid the rapid change of the concentration between spatial grid points, the constraint condition is established, the Tikhonov regularization term is introduced, and combined with the non-negative restriction condition of the atmospheric pollution component concentration, there is:

[0088]

[0089] Wherein, represents the atmospheric pollutant concentration of the i-th row and j-th column grid to be solved, represents the atmospheric pollutant concentration after constraint.

[0090] Under this constraint condition, based on the tomographic inversion method, the concentration c of the target grid can be calculated.

[0091]

[0092] In the whole inversion process, two constraint conditions are established for the grid boundary condition and the atmospheric pollution component:

[0093] ① The atmospheric pollutant concentration is mainly concentrated in the six key observation area grids of C3, C4, C5, D3, D4 and D5, so the concentration of the key pollution source in these six areas is greater than the surrounding concentration: c i,j >c i±1,j±1

[0094] ② The concentration of all grids is non-negative:

[0095] Therefore:

[0096] For example, Figure 6As shown, by iterative operation, the minimization step is performed step by step, and the concentration value is output when the convergence condition is reached. In the case, based on the above inversion and technology, the concentration result is obtained by iteration 10000 times, and the concentration result is shown in the following table (the gray background is the key observation area):

[0097] Concentration (pg / m 3 )]]> 1 2 3 4 5 6 7 A 14.45 63.65 24.49 8.88 47.37 1.29 38.48 B 35.44 47.87 34.12 42.81 59.34 31.65 28.23 C 24.59 52.84 52.77 24.15 76.20 21.52 30.52 D 44.69 50.72 76.04 73.62 76.38 45.84 22.09 E 24.92 49.04 37.48 24.11 49.91 48.50 36.91 F 8.55 12.56 8.33 5.43 28.57 26.83 21.65

[0098] According to the results, it can be seen that there are 4 places in the key observation area that exceed 70 μg / m 3 The NO2 key high value area (indicated in bold) gradually decreases to the surrounding area, and the NO2 concentration in the most marginal zone is basically lower than 40 μg / m 3 Therefore, it can be basically determined that there is a location of pollutant emission source in the key observation area.

[0099] The method of the present application repeatedly scans the pollution source area by multiple atmospheric composition hyperspectral scanning analyzers, and can obtain the fine grid distribution of the horizontal concentration of atmospheric pollutants in the region based on chromatographic inversion.

[0100] The specific embodiments described above have described the technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the most preferred embodiment of the present application and is not intended to limit the present application. Any modification, supplement and equivalent replacement within the principle range of the present application should be included in the protection scope of the present application.

Claims

1. A method for measuring the distribution of urban air pollutants based on cross-scanning, characterized in that: The following steps are involved: Multiple atmospheric composition hyperspectral scanners are set up at the boundaries of the target observation area, and the distance between any two scanners is ensured to be within the scanner detection range; Grid the target observation area to ensure that each grid covers the optical path information of at least one scanner, and the key observation areas included in the target observation area must cover the optical path information of multiple scanners; The observation azimuth angle sequence of the scanner is determined based on the grid area. The horizontal atmospheric scattering spectrum is observed by alternating the azimuth angles of any two adjacent scanners according to the azimuth angle observation sequence, and the differential oblique column concentration of atmospheric pollutants is obtained based on the horizontal atmospheric scattering spectrum. The weight of each grid is determined based on the light path of each scanner passing through the grid. A set of equations is constructed based on the weights and the differential oblique column concentrations of atmospheric pollutants observed by each scanner. Under the constraints, the set of equations is iteratively inverted to obtain the horizontal concentration distribution information of each atmospheric pollutant. The equations are constructed based on the weights and the differential oblique column concentrations of atmospheric pollutants observed by each scanner, including: S=Wc Among them, S represents SCD n,k The differential oblique column concentration matrix composed of W ij The weight matrix composed of c ij The atmospheric pollutant concentration matrix to be solved is composed of: Under the constraint conditions, the equations are iterated and inverted to obtain the atmospheric pollutant concentration in the target observation area. Here, argmin means making |S-Wc| 2 When the minimum value is reached The value of .

2. The method for measuring the horizontal distribution of urban air pollutants based on cross scanning according to claim 1 is characterized in that: The differential oblique column concentration of atmospheric pollutants is obtained based on the horizontal atmospheric scattering spectrum, including: The horizontal atmospheric scattering spectrum at the zenith is used as the reference spectrum, and the least square method based on characteristic absorption is used to solve the horizontal atmospheric scattering spectra at different azimuth angles to obtain the differential oblique column concentrations of atmospheric pollutants.

3. The method for measuring the horizontal distribution of urban air pollutants based on cross scanning according to claim 1 is characterized in that: The weight of each scanner is determined based on the light path of each scanner passing through the grid, including: The intersection position of the scanner's light path passing through each grid boundary is calculated based on the scanner's spatial position and observation azimuth. Then, the light path length of each grid is calculated based on the distance between the two intersection positions on the same grid. The light path length is used as the weight of the differential oblique column concentration of each grid observed by a single scanner.

4. The method for measuring the horizontal distribution of urban air pollutants based on cross scanning according to claim 1 is characterized in that: The relationship between the differential oblique column concentration of atmospheric pollutants observed by a single scanner and the grid weight is: Among them, SCD n,k W represents the differential oblique column concentration data of atmospheric pollutants obtained by the nth scanner under observation at k azimuth angles, ij represents the weight of the grid in row i and column j, c ij Represents the concentration of atmospheric pollutants to be solved for the grid in row i and column j.

5. The method for measuring the level distribution of urban air pollutants based on cross scanning according to claim 1 is characterized in that: The constraints introduced are: in, represents the concentration of air pollutants in the grid of row i and column j to be solved, Represents the concentration of air pollutants after constraint.

6. The method for measuring the level distribution of urban air pollutants based on cross scanning according to claim 1, characterized in that: Solve using the least squares method Get the concentration of all grids in the target observation area This will help determine locations with high concentrations and analyze emission source information.

7. The method for measuring the level distribution of urban air pollutants based on cross scanning according to claim 1, characterized in that: When the scanners are set at the boundaries of the target observation area, the scanner spacing is 5 to 10 kilometers.

Images