Pollutant source analysis method

Abstract

The present invention discloses a pollutant source analysis method, including: taking the area to be analyzed as a receptor, determining several target emission sources of the receptor, and obtaining fingerprint spectra of several target emission sources; obtaining receptor monitoring data; eliminating unknown source components in the receptor monitoring data; and obtaining pollutant source analysis results according to the fingerprint spectrum of the target emission source of the receptor, the receptor monitoring data with unknown source components eliminated, and a pre-constructed chemical mass balance model. By establishing a pollutant emission source list and determining the receptor target emission source in the list, the pertinence and positioning accuracy of source tracing analysis are improved, and the boundary cross-contamination problem of the enterprise production site is considered. At the same time, a reasonable and feasible pollutant source analysis process and a method for determining collinear sources and unknown source components are designed, which solves the collinearity problem to a certain extent and improves the accuracy of positioning the source of pollutants and the precision of quantitative analysis.

Description

Technical Field

[0001] The present invention relates to the field of air pollution prevention and control, and in particular to a method for analyzing the sources of pollutants. Background Art

[0002] Understanding the emission characteristics of atmospheric pollutants in key areas and qualitatively and quantitatively clarifying the sources and contributions of characteristic pollutants are important conditions for scientifically and rationally establishing a long-term mechanism for the prevention and control of atmospheric pollution.

[0003] The common pollutant source analysis methods currently only consider the pollution status of the receptor location, and do not thoroughly investigate the surrounding emission sources that may have an impact on the receptor location. In addition, although the commonly used factor analysis as a receptor source analysis technology has made a component analysis, the analyzed component factors only have mathematical significance, and their physical significance is difficult to express.

[0004] Therefore, when analyzing the sources of pollutants in the prior art, there are problems such as unclear sources of pollutants in cross-contamination areas at enterprise boundaries, inaccurate positioning, and inaccurate analysis results. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a method for analyzing the source of pollutants to solve the above technical problems, at least partially solve the above problems.

[0006] The pollutant source analysis method provided by an embodiment of the present invention includes: determining several target emission sources of a receptor and obtaining fingerprint spectra of several of the target emission sources; obtaining receptor monitoring data; eliminating unknown source components in the receptor monitoring data; and obtaining pollutant source analysis results based on the fingerprint spectra of the target emission sources of the receptor, the receptor monitoring data with the unknown source components eliminated, and a pre-constructed chemical mass balance model.

[0007] Optionally, the determining of several target emission sources of the receptor and obtaining fingerprint spectra of several target emission sources includes: establishing a list of pollutant emission sources around the receptor, and determining the fingerprint spectra of the pollutant emission sources on the list of pollutant emission sources; and analyzing several pollutant emission sources in the list of pollutant emission sources that can cause pollution to the receptor, that is, several target emission sources, and determining the fingerprint spectra of several target emission sources.

[0008] Optionally, establishing a list of pollutant emission sources around the receptor includes: taking one or more devices as an emission source within the production site of the enterprise to which the receptor belongs; and taking one enterprise as an emission source outside the production site of the enterprise to which the receptor belongs.

[0009] Optionally, the pollutant emission characteristics of the multiple devices serving as one emission source are the same and / or the multiple devices are located adjacent to each other without obvious boundaries.

[0010] Optionally, before eliminating the unknown source components in the receptor monitoring data, the pollutant source analysis method also includes: determining whether there are collinear sources in the target emission sources; when there are collinear sources in the target emission sources, obtaining the principal components of the receptor monitoring data, and determining whether there are unknown source components in the principal components of the receptor monitoring data; if there are unknown source components in the principal components of the receptor monitoring data, eliminating the unknown source components in the receptor monitoring data.

[0011] Optionally, the determining whether there are collinear sources in the target emission sources includes: obtaining a first correlation coefficient between fingerprint spectra of every two target emission sources among a plurality of target emission sources of the receptor; if the first correlation coefficient between the fingerprint spectra of any two target emission sources is less than a first predetermined value, then it is determined that there are no collinear sources in the target emission sources; otherwise, it is determined that there are collinear sources in the target emission sources.

[0012] Optionally, the obtaining of the first correlation coefficient between the fingerprint spectra of every two of the target emission sources among the plurality of target emission sources at the receptor position includes: the first correlation coefficient is obtained using the following formula: Among them, ρ XY represents the first correlation coefficient between emission source X and emission source Y, X and Y represent the fingerprint spectra of emission source X and emission source Y respectively, Cov(X, Y) represents the covariance of the fingerprint spectra of emission source X and emission source Y, D(X) represents the variance of the fingerprint spectrum of emission source X, and D(Y) represents the variance of the fingerprint spectrum of emission source Y.

[0013] Optionally, obtaining the principal components of the receptor monitoring data and determining whether there are unknown source components in the principal components of the receptor monitoring data includes: obtaining the principal components of the receptor monitoring data using a principal component analysis method or a positive definite matrix factor analysis method; and obtaining a second correlation coefficient between each principal component of the receptor monitoring data and a fingerprint spectrum of each of the target emission sources and a third correlation coefficient between each principal component of the receptor monitoring data and the combined fingerprint spectrum of the target emission sources; if the second correlation coefficient of a principal component of the receptor monitoring data is less than a second predetermined value and the third correlation coefficient is less than a third predetermined value, then the principal component of the receptor monitoring data is determined to be an unknown source component; otherwise, it is determined that there are no unknown source components in the principal components of the receptor monitoring data.

[0014] Optionally, when the unknown source component exists in the main component of the receptor monitoring data, eliminating the unknown source component in the receptor monitoring data includes: determining whether the variance contribution rate of the unknown source component to the receptor monitoring data is greater than a fourth predetermined value; if the variance contribution rate is greater than the fourth predetermined value, eliminating the unknown source component in the receptor monitoring data.

[0015] On the other hand, the present invention provides a machine-readable storage medium having stored thereon instructions for enabling a machine to execute the above-mentioned pollutant source analysis method.

[0016] The present invention improves the pertinence and positioning accuracy of source tracing analysis by establishing a pollutant emission source list and determining the receptor target emission source in the list. It also specifically targets the boundary cross-contamination area of ​​the enterprise production site and takes the enterprise as a unit, and includes the pollution sources outside the enterprise production site in the pollutant emission source list in the area to be analyzed, thereby improving the accuracy of the analysis. At the same time, a reasonable and feasible pollutant source analysis process and a method for determining collinear sources and unknown source components are designed, which solves the collinearity problem to a certain extent and improves the accuracy of quantitative analysis of pollutant sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of a pollutant source analysis method provided by an embodiment of the present invention;

[0018] Figure 2 A flowchart for determining whether there are collinear sources in the target emission sources of a receptor and whether there are unknown source components in the receptor monitoring data provided by an embodiment of the present invention; and

[0019] Figure 3 It is a flowchart of a specific application example of an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0021] The embodiments of the present invention are mainly used for analyzing the sources of atmospheric pollutants in refining and chemical enterprises, wherein the emission source list refers to the device as the basic unit within the scope of the enterprise production site, and the enterprise as the basic unit outside the enterprise production site. According to the emission characteristics such as the main types and concentration ratios of pollutants, two or more devices with adjacent positions and no obvious boundaries can be regarded as one emission source, or two or more devices with the same pollutant emission characteristics can be regarded as the same emission source, or devices with the same pollutant emission characteristics and adjacent positions without obvious boundaries can be regarded as the same emission source, and the pollutant emission source list is listed according to this principle. Among them, the same pollutant emission characteristics means that the types of pollutant emissions of the two devices and the concentration ratios of each type of pollutants are basically the same.

[0022] The fingerprint spectrum of the pollution source in the present invention refers to a spectrum that can represent its emission characteristics after long-term monitoring of the pollution source, using certain analysis and processing methods. Its chemical information and emission information can be reflected in the fingerprint spectrum, which has the characteristics of fuzziness and integrity, and can be analyzed as a whole. The fingerprint spectrum of each emission source in the emission source list in the area to be analyzed is determined according to the type and concentration ratio of the pollutant emission source.

[0023] The target emission source refers to an emission source that may cause pollution to a receptor. In the present invention, the target emission source is one or more pollutant emission sources that may cause pollution to a receptor determined from a pollutant emission source list.

[0024] The following receptor monitoring data is an array representing the pollutant concentration value, and the fingerprint spectrum and unknown source components are arrays representing the pollutant concentration ratio or proportion, and their expression form is the same, for example: (methane: 25PPB; benzene: 19PPB; toluene: 42PPB) (PPB is the concentration unit).

[0025] Figure 1 is a flow chart of the pollutant source analysis method provided by an embodiment of the present invention, such as Figure 1 As shown, the following steps may be included:

[0026] S101. Obtain several target emission sources of the receptor and their fingerprint spectra.

[0027] The embodiment of the present invention first takes the device as the basic unit within the scope of the enterprise production site, and takes the enterprise as the basic unit outside the enterprise production site. According to the emission characteristics such as the main types and concentration ratios of pollutants, two or more devices that are adjacent to each other and do not have obvious boundaries can be regarded as one emission source, or two or more devices with the same pollutant emission characteristics can be regarded as the same emission source, or devices with the same pollutant emission characteristics and adjacent positions without obvious boundaries can be regarded as the same emission source. A list of pollutant emission sources is listed based on this principle. Among them, the same pollutant emission characteristics means that the types of pollutants emitted by the two devices and the concentration ratios of each type of pollutants are basically the same. Then, according to the types and concentration ratios of pollutants emitted by each emission source, the fingerprint spectrum of each emission source is determined, and finally one or more pollutant emission sources that may cause pollution to the receptor are determined from the list of pollutant emission sources around the receptor as the target emission source of the receptor. The fingerprint spectrum of the determined pollutant emission source is the fingerprint spectrum of the receptor target emission source.

[0028] Among them, the methods for determining the target emission sources can adopt characteristic pollutant comparison method, diffusion simulation method, etc.

[0029] S102. Obtain receptor monitoring data.

[0030] The pollutants at the receptor position are monitored. In the embodiment of the present invention, no less than 50 sets of receptor monitoring data are collected for analysis. The receptor monitoring data refers to the data on the types and concentrations of pollutants monitored at the receptor position. For example, if one set of data is collected every minute, it takes at least 50 minutes to collect 50 sets of receptor monitoring data.

[0031] S103, eliminating unknown source components in the receptor monitoring data.

[0032] Unknown source components will have an impact on the analysis of pollutant sources, especially those that have a greater impact on the receptors. They may cause large deviations and lead to inaccurate analysis of pollutant sources. Therefore, it is necessary to eliminate the unknown source components in the receptor monitoring data in order to obtain more accurate analysis of pollutant sources.

[0033] Before removing the unknown source components in the receptor monitoring data, the pollutant source analysis method of the embodiment of the present invention also includes determining whether there are collinear sources in the target emission sources of the receptor and whether there are unknown source components in the receptor monitoring data. The specific steps and implementation methods will be described in Figure 2 Detailed description is given in , and will not be repeated here.

[0034] S104. Obtain pollutant source analysis results.

[0035] In the embodiment of the present invention, the fingerprint spectrum of the target emission source and the receptor monitoring data after eliminating the unknown source components are brought into the chemical mass balance model to obtain the pollutant source analysis results.

[0036] The chemical mass balance model is a mathematical model of multivariate statistical analysis that uses the chemical element balance method to identify the source of pollutants. The mass of the pollutant (including the chemical components contained) should be the sum of the emissions of each emission source, that is, the amount of pollutants emitted by the emission source to the receptor is balanced with the amount of pollutants present in the receptor environment. Based on this idea, a linear combination model of chemical components between the emission source and the receptor is established, and various types of emission sources of pollutants and their relative contribution rates are obtained through optimization solutions such as the least squares method. In the embodiment of the present invention, the fingerprint spectrum of the target emission source is regarded as the unit amount of pollutant emissions, and the receptor monitoring data is regarded as the amount present in the environment to obtain the pollutant source analysis results.

[0037] Figure 2 It is a flowchart of determining whether there is a collinear source in the target emission source at the receptor position and whether there is an unknown source component in the receptor monitoring data provided by an embodiment of the present invention, such as Figure 2 As shown, before removing the unknown source components in the receptor monitoring data, the pollutant source analysis method may further include the following steps:

[0038] S201. Determine whether there is a collinear source in the target emission source.

[0039] Obtain a first correlation coefficient between the fingerprint spectra of every two of the target emission sources of the receptor; if the first correlation coefficient between the fingerprint spectra of any two of the target emission sources is less than a first predetermined value, it is determined that there is no collinear source in the target emission source, and at this time, execute step S104; otherwise, it is determined that there is a collinear source in the target emission source, and execute step S202.

[0040] Among them, the first correlation coefficient is obtained using the following formula:

[0041]

[0042] Among them, ρ XY represents the first correlation coefficient between emission source X and emission source Y;

[0043] X and Y represent the fingerprint spectra of emission sources X and Y, respectively;

[0044] Cov(X, Y) represents the covariance of the fingerprint spectra of emission sources X and Y;

[0045] D(X) represents the variance of the fingerprint spectrum of emission source X;

[0046] D(Y) represents the variance of the fingerprint spectrum of emission source Y.

[0047] Among them, the first correlation coefficient is a value ranging from -1 to 1. If the first correlation coefficient is equal to 1, X and Y are exactly the same. In the embodiment of the present invention, the first correlation coefficient is set to 0.8-0.9, and the first correlation coefficient is preferably 0.85. For example, during a pollutant source analysis, the first correlation coefficient is set to 0.85. After analysis and calculation, it is found that the first correlation coefficient of two target emission sources of a certain receptor is greater than 0.85, then it is determined that there is a collinear source in the target emission source at the receptor position.

[0048] S202. When there are collinear sources in the target emission sources, obtain the principal components of the receptor monitoring data.

[0049] According to the determination result of step S201, if there is a collinear source in the target emission source, it is necessary to further analyze the receptor monitoring data. First, the principal components of the receptor monitoring data need to be obtained. As described in step S102, generally, no less than 50 groups of monitoring data need to be collected. First, the obviously abnormal monitoring data are eliminated, and then the principal component analysis is performed on multiple groups of receptor monitoring data. The principal components of the receptor monitoring data are obtained according to the following principles: the principal components with larger variance are preferentially screened, the number of principal components is as small as possible, and the sum of the variance contribution rates of the principal components is not less than 90%. The sum of the variance contribution rates of the principal components can also be set to 85%-95% according to the actual situation.

[0050] Here, the positive definite matrix factor analysis method can also be used to obtain the principal components of the receptor monitoring data.

[0051] S203, determining whether there is an unknown source component in the principal component of the receptor monitoring data.

[0052] The judgment method may include: obtaining a second correlation coefficient between each principal component of the receptor monitoring data and each fingerprint spectrum of the target emission source and a third correlation coefficient between each principal component of the receptor monitoring data and the combined fingerprint spectrum of the target emission source; if the second correlation coefficient of a principal component is less than a second predetermined value and the third correlation coefficient is less than a third predetermined value, then the principal component of the receptor monitoring data is determined to be an unknown source component, and at this time, step S204 is executed; otherwise, it is determined that there is no unknown source component in the principal component of the receptor monitoring data, and step S104 is executed.

[0053] Among them, the second predetermined value can be 0.5-0.7, the third predetermined value can be 0.3-0.5, and the combined fingerprint spectrum is obtained by superimposing the fingerprint spectra of generally no more than 3 target emission sources. The number of target emission sources included in the combined fingerprint spectrum can be set according to actual conditions. For example, the combined fingerprint spectrum can also be set to be obtained by superimposing the fingerprint spectra of no more than 5 target emission sources.

[0054] For example: Set the second preset value to 0.6, the third preset value to 0.4, and define M target emission source fingerprint spectra X i , where i∈[1,M] and is an integer, the following conditions must be met to determine that the principal component P of the receptor monitoring data is an unknown source component: any X i Both And for any combination of fingerprints Both in The target emission source is X i Correlation coefficient with the principal component P of the receptor monitoring data, is the correlation coefficient between the combined fingerprint spectrum X0 of the target emission source and the principal component P of the receptor monitoring data, k i Equal to 0 or 1 and generally not more than 3 k i Equal to 1.

[0055] S204: If there are unknown source components in the principal components of the receptor monitoring data, the unknown source components in the receptor monitoring data are removed.

[0056] Preferably, eliminating the unknown source components in the receptor monitoring data includes: determining whether the variance contribution rate of the unknown source components to the receptor monitoring data is greater than a fourth predetermined value; if the variance contribution rate is greater than the fourth predetermined value, executing step S104 after eliminating the unknown source components in the receptor monitoring data; if the variance contribution rate is not greater than the fourth predetermined value, executing step S104 directly.

[0057] Wherein, the fourth predetermined value may be set to 5%-10%. For example, the fourth predetermined value is set to 8%. When the sum of the variance contribution rates of all unknown source components to the receptor monitoring data is calculated to be greater than 8%, the unknown source components in the receptor monitoring data are removed and then step S104 is executed. Otherwise, step S104 may be directly executed.

[0058] Eliminating the unknown source components in the receptor monitoring data refers to eliminating the portion of the unknown source components in the receptor monitoring data and obtaining new receptor monitoring data. The portion of the unknown source components is the component of each group of receptor monitoring data on the unknown source component vector.

[0059] Example: Define N groups of receptor monitoring data A j, where j∈[1,N] is an integer, define A j The component on an unknown source component P is ΔP j , then the new N group receptor monitoring data is A j -ΔP j .

[0060] The present invention is described below by using application examples of the embodiments of the present invention. Figure 3 is a flowchart of a specific application example of an embodiment of the present invention, such as Figure 3 As shown, the following steps may be included:

[0061] S301. Establish a list of pollutant emission sources around the receptors.

[0062] S302: Determine multiple target emission sources and their fingerprint spectra of the receptors on the pollutant emission source list.

[0063] S303, obtaining multiple groups of receptor monitoring data.

[0064] S304, determine whether there are collinear sources among the multiple target emission sources; if there are collinear sources, execute step S305, otherwise execute step S309.

[0065] S305, obtaining the principal components of the receptor monitoring data, performing principal component analysis on the receptor monitoring data, and screening multiple principal components. The screening requirements include giving priority to screening principal components with larger variances, the fewer the number of principal components, the better, and the sum of the principal component variance contribution rates being no less than 90%.

[0066] S306. Determine whether there is an unknown source component among the multiple principal components. If so, execute step S307; otherwise, execute step S309.

[0067] S307, determine whether the sum of the variance contribution rates of the unknown source components is greater than a fourth predetermined value. In this application example, the fourth predetermined value is set to 8%; if so, execute S308, otherwise execute step S309.

[0068] S308, eliminating the portion of the receptor monitoring data occupied by the unknown source component, and obtaining new receptor monitoring data; the portion of the unknown source component is the component of each group of receptor monitoring data on the unknown source component vector.

[0069] S309. The obtained receptor monitoring data and the target emission source fingerprint spectrum are brought into the chemical mass balance model to obtain the pollutant source analysis results.

[0070] The specific implementation details of the application example are the same as above Figure 1 and Figure 2 The description is not repeated here.

[0071] The present invention improves the pertinence of source tracing analysis by establishing a pollutant emission source list and determining the receptor target emission source in the list, and specifically targets the boundary cross-contamination area of ​​the enterprise production site, and includes the pollution sources outside the enterprise production site in the pollutant emission source list in the area to be analyzed, thereby improving the accuracy of the analysis. At the same time, a reasonable and feasible pollutant source analysis process and a method for determining collinear sources and unknown source components are designed, which solves the collinearity problem to a certain extent and improves the accuracy of quantitative analysis of pollutant sources.

[0072] The above-mentioned pollutant source analysis method can be applied to the analysis of atmospheric pollutant sources in refining enterprises, and can also be applied to the analysis of atmospheric pollutant sources in other enterprises with more atmospheric pollutants. The receptor in the present invention refers to the polluted area to be analyzed, that is, the area polluted by the surrounding emission sources (including the equipment of the enterprise where the receptor is located and the enterprises outside the production range of the receptor), and the pollutant source analysis needs to be carried out.

[0073] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited thereto. Within the technical concept of the present invention, the technical scheme of the present invention can be subjected to a variety of simple modifications, for example, the principal component analysis can be changed to be replaced by a positive definite matrix factor analysis, the correlation coefficient calculation can be replaced by a divergence coefficient calculation, including the combination of various specific technical features in any suitable manner, such as adjusting the index of the determination of the collinearity source and the unknown source component. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations. However, these simple modifications and combinations should also be regarded as the contents disclosed by the present invention, and all belong to the protection scope of the present invention. The various specific technical features described in the above-mentioned specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the embodiments of the present invention will not further describe various possible combinations.

[0074] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a single-chip microcomputer, a chip or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0075] In addition, various implementation modes of the embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed by the embodiments of the present invention.

Claims

1. A method for analyzing the source of pollutants, characterized in that: The pollutant source analysis method comprises: Taking the area to be analyzed as a receptor, determining several target emission sources of the receptor, and obtaining fingerprint spectra of several target emission sources; Access to receptor monitoring data; Eliminating unknown source components in the receptor monitoring data; and The pollutant source analysis results are obtained based on the fingerprint spectrum of the target emission source of the receptor, the receptor monitoring data excluding the unknown source components, and the pre-built chemical mass balance model. Wherein, before removing the unknown source components in the receptor monitoring data, the pollutant source analysis method further includes: Determine whether there is a collinear source among the target emission sources; When there are collinear sources in the target emission sources, the principal components of the receptor monitoring data are obtained, and it is determined whether there are unknown source components in the principal components of the receptor monitoring data. If there are unknown source components in the principal components of the receptor monitoring data, the unknown source components in the receptor monitoring data are removed. The obtaining of the principal components of the receptor monitoring data and determining whether there are unknown source components in the principal components of the receptor monitoring data include: Obtaining the principal components of the receptor monitoring data using a principal component analysis method or a positive definite matrix factor analysis method; and Obtaining a second correlation coefficient between each principal component of the receptor monitoring data and a fingerprint spectrum of each target emission source and a third correlation coefficient between each principal component of the receptor monitoring data and a combined fingerprint spectrum of the target emission source, If the second correlation coefficient of a principal component of the receptor monitoring data is less than a second predetermined value and the third correlation coefficient is less than a third predetermined value, the principal component of the receptor monitoring data is determined to be an unknown source component; otherwise, it is determined that there is no unknown source component in the principal component of the receptor monitoring data. Among them, when the unknown source component exists in the main component of the receptor monitoring data, the elimination of the unknown source component in the receptor monitoring data includes: judging whether the variance contribution rate of the unknown source component to the receptor monitoring data is greater than a fourth predetermined value; if the variance contribution rate is greater than the fourth predetermined value, the unknown source component in the receptor monitoring data is eliminated.

2. The method for pollutant source analysis according to claim 1, characterized in that: The step of determining a plurality of target emission sources of the receptor and obtaining fingerprint spectra of the plurality of target emission sources comprises: Establishing a pollutant emission source list around the receptor, and determining the fingerprint spectrum of the pollutant emission source on the pollutant emission source list; and A number of the pollutant emission sources in the pollutant emission source list that can cause pollution to the receptor are analyzed, namely, a number of target emission sources, and fingerprint spectra of the number of the target emission sources are determined.

3. The method for pollutant source analysis according to claim 2, characterized in that: The establishment of a list of pollutant emission sources around the receptor includes: taking one or more devices as an emission source within the production site of the enterprise to which the receptor belongs; and taking one enterprise as an emission source outside the production site of the enterprise to which the receptor belongs.

4. The method for pollutant source analysis according to claim 3, characterized in that: The pollutant emission characteristics of the multiple devices serving as one emission source are the same and / or the multiple devices are located adjacently without obvious boundaries.

5. The method for pollutant source analysis according to claim 1, characterized in that: The determining whether there is a collinear source in the target emission source comprises: Obtaining a first correlation coefficient between fingerprint spectra of every two of the target emission sources of the receptor, If the first correlation coefficient between the fingerprint spectra of any two of the target emission sources is less than a first predetermined value, it is determined that there is no collinear source among the target emission sources; Otherwise, it is determined that there are collinear sources in the target emission sources.

6. The method for pollutant source analysis according to claim 5, characterized in that: The first correlation coefficient between the fingerprint spectra of each two of the target emission sources among the plurality of target emission sources of the receptor is obtained, comprising: The first correlation coefficient is obtained by using the following formula: Among them, ρ XY represents the first correlation coefficient between emission source X and emission source Y; X and Y represent the fingerprint spectra of emission sources X and Y, respectively; Cov(X,Y) represents the covariance of the fingerprint spectra of emission sources X and Y; D(X) represents the variance of the fingerprint spectrum of emission source X; D(Y) represents the variance of the fingerprint spectrum of emission source Y.

7. A machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the pollutant source analysis method described in any one of claims 1 to 6 of the present application.

Images