Wharf atmospheric pollutant monitoring method and system based on pollutant diffusion model

Abstract

The invention relates to the technical field of atmospheric dust pollution prevention and environmental protection, and discloses a wharf atmospheric pollutant monitoring method and system based on a pollutant diffusion model. The method comprises the steps of dividing meteorological types based on meteorological data and determining the occurrence probability of the meteorological types; pollutant types are determined according to port operation characteristics, and grid division is carried out; simulating pollutant diffusion by adopting a diffusion model and calculating the evaluation concentration of each grid; establishing a coordinate system to determine potential layout point coordinates; quantizing the influence of external factors by using a fuzzy evaluation technology, generating a fuzzy evaluation matrix and preliminarily screening grids; performing condensation type clustering analysis on the screening result; and finally, determining an optimal layout point by combining a clustering result and a monitoring requirement. According to the method, the diffusion model, the fuzzy evaluation and the clustering analysis technology are fused, the defects that a traditional method is high in subjectivity and insufficient in scientificity are overcome, pollutant space distribution characteristics can be accurately reflected, and effective technical support is provided for precise treatment of port atmospheric pollution.

Description

Technical Field

[0001] This invention relates to the field of environmental protection technology for the prevention and control of atmospheric dust pollution, and in particular to a method and system for monitoring atmospheric pollutants at wharves based on a pollutant diffusion model. Background Technology

[0002] my country has a large number of general-purpose terminals, which are widely distributed and handle a large volume of cargo. The loading and unloading of dust-generating goods such as coal and ore produces significant amounts of pollutants, including NOx and PM. Therefore, accurately monitoring pollutant emissions within ports has become an urgent task.

[0003] Currently, the deployment of monitoring points at general-purpose terminals relies heavily on traditional experience or conventional standards, which is highly subjective and leads to significant uncertainty and randomness in point placement, making it difficult to guarantee the reliability, scientific validity, and representativeness of the monitoring network. Furthermore, the diffusion of pollutants within terminals is influenced by numerous factors at the microscale, including meteorological conditions, machinery, vehicles, and structures. Traditional methods cannot effectively address these complex constraints, accurately reflect the spatial distribution of pollutants, or provide scientific support for precise pollution control. Therefore, there is an urgent need to develop a monitoring point deployment method that comprehensively considers the influence of multiple factors, including monitoring resources, safe operations, and meteorological environment. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for monitoring air pollutants at a wharf based on a pollutant diffusion model, comprising the following steps: S1. Based on meteorological monitoring data of the port area, different meteorological types are classified according to atmospheric stability, and the occurrence probability of each meteorological type is determined. S2. Determine the pollutant type based on the port operation type and energy type, and divide the port area into grids; S3. Select a local pollutant diffusion model, combine meteorological type and pollutant type to simulate pollutant diffusion, and calculate the pollutant evaluation concentration for each grid. S4. Establish a coordinate system with the geometric center of the port area as the origin, determine the coordinates of the center point of each grid, and use the center point of each grid as the potential deployment point for pollution monitoring. S5. Use fuzzy evaluation technology to quantify the influence of external factors on potential deployment points, generate a fuzzy evaluation matrix, and filter the grid based on the fuzzy evaluation matrix to obtain preliminary screening results; S6. Agglomerative cluster analysis was performed on the preliminary screening results to obtain the clustering results; specifically including: S61. Initialize each grid in the initial screening results as a separate category; S62. Calculate the pairwise Euclidean distance between all categories. The calculation formula is as follows: ; Where x and y are the two categories whose distances are being calculated, x k and y k These are the normalized evaluation values ​​of category x and category y on the k-th indicator, respectively; S63. Set a distance threshold, merge two categories whose Euclidean distance is less than the distance threshold into a new category, and recalculate the Euclidean distance between the new category and all other categories; S64. Repeat S62~S63 until all grid classifications no longer change, and obtain the clustering results; S7. Based on the clustering results and the monitoring requirements of the port area, determine the final deployment points for pollution monitoring.

[0005] Furthermore, in S3, pollutant diffusion is simulated by combining meteorological type and pollutant type, and the pollutant assessment concentration for each grid is calculated, including: Combining the meteorological types and atmospheric pollutant emission types in the port area, a pollutant emission diffusion simulation was conducted to obtain the diffusion characteristics of pollutants of different emission types under different meteorological types. The evaluation concentration of pollutants in different grids was calculated using the following formula: in, Let i be the pollutant evaluation concentration in grid i, where i is the grid number. j As for the type of pollutant, w As a weather type, J Where W represents the total number of pollutant types and W represents the total number of meteorological types. To determine the type of pollutant j Weighting coefficients for potential monitoring and , Weather type w The probability of occurrence, For grid i in weather type w Types of pollutants j average concentration, Weather type w The wind speed below, They are respectively weather types w The lateral and vertical diffusion coefficients below, This indicates the emission intensity of pollutant j.

[0006] Furthermore, in S5, fuzzy evaluation technology is used to quantify the influence of external factors on potential deployment points, generating a fuzzy evaluation matrix. Based on the fuzzy evaluation matrix, the grid is screened to obtain preliminary screening results, including: S51. Set n evaluation indicators for each grid, and construct a judgment matrix for the grid based on the evaluation indicators. , where a ikLet m be the original evaluation value of the k-th indicator in the i-th grid, where m is the number of grids and n is the number of evaluation indicators. S52. Normalize the judgment matrix A to obtain the normalized matrix B; S53. Calculate the entropy weight of the evaluation index based on the normalized matrix B, and construct the weight index matrix Z based on the entropy weight and the normalized matrix B. S54. Set up a comment set, use the membership function to calculate the membership degree of the elements in the weight index matrix Z to the comment set, and then construct the fuzzy evaluation matrix R for each grid. S55. Perform fuzzy synthesis operation on the fuzzy evaluation matrix R of each grid to obtain the fuzzy evaluation subset of each grid, and normalize and comprehensively score the fuzzy evaluation subset. Based on the comprehensive score, obtain the preliminary screening results.

[0007] Furthermore, the normalized matrix B in S52 is: ; in, b ik It is the normalized evaluation value of the i-th grid at the k-th evaluation index, a max and a min These are the maximum and minimum values ​​among all the original evaluation values ​​of the grid under the k-th evaluation index, respectively.

[0008] Furthermore, in S53, the entropy weights of the evaluation indicators are calculated based on the normalized matrix B, and a weighted indicator matrix Z is constructed based on the entropy weights and the normalized matrix B, including: The entropy value of the k-th index is: ; Among them, H k f is the entropy value of the k-th index. ik It represents the ratio of the normalized evaluation value of the k-th evaluation index of the i-th grid to the sum of the normalized evaluation values ​​of the k-th evaluation index of all grids, and when f ik When =0, ;f ik The calculation formula is: ; The entropy weight of the k-th index is: ; Among them, W k Let be the entropy weight of the k-th evaluation index, and satisfy . ; Construct the weight index matrix Z using the obtained entropy weights and normalization matrix B: ; Among them, z ikz represents the evaluation weight of the i-th grid on the k-th evaluation metric. ik The calculation formula is: ; W is the normalized evaluation value of the i-th grid at the k-th evaluation index. k Let be the entropy weight of the k-th evaluation index.

[0009] Furthermore, in S55, a fuzzy synthesis operation is performed on the fuzzy evaluation matrix R of each grid. The calculation formula for the fuzzy synthesis operation is as follows: ; Among them, V i Let A be the fuzzy subset of the i-th grid, A be the judgment matrix, and R be the fuzzy subset of the i-th grid. i Let be the fuzzy evaluation matrix for the i-th grid.

[0010] The present invention also provides a port air pollutant monitoring system based on a pollutant diffusion model, used to execute the port air pollutant monitoring method based on a pollutant diffusion model described in any of the above claims. The system includes the following modules: The meteorological data processing module is used to classify meteorological data from the port area into different meteorological types based on atmospheric stability and determine the probability of occurrence of each meteorological type. The gridding and pollution source analysis module, connected to the meteorological data processing module, is used to determine the type of pollutants based on the port operation type and energy type, and to divide the port area into grids. The diffusion simulation calculation module, connected to the gridding and pollution source analysis module, is used to select a local pollutant diffusion model, combine meteorological type and pollutant type to simulate pollutant diffusion, and calculate the pollutant evaluation concentration for each grid. The coordinate generation module, connected to the diffusion simulation calculation module, is used to establish a coordinate system with the geometric center of the port area as the origin, determine the coordinates of the center point of each grid, and use the center point of each grid as the potential deployment point for pollution monitoring. The fuzzy evaluation and screening module, connected to the coordinate generation module, is used to quantify the influence of external factors on potential deployment points using fuzzy evaluation technology, generate a fuzzy evaluation matrix, and screen the grid based on the fuzzy evaluation matrix to obtain preliminary screening results. The clustering analysis optimization module, connected to the fuzzy evaluation and screening module, is used to perform agglomerative clustering analysis on the preliminary screening results to obtain the clustering results. The monitoring site deployment decision module, connected to the cluster analysis optimization module, is used to determine the final deployment sites for pollution monitoring by combining the clustering results and the monitoring requirements of the port area.

[0011] The embodiments of the present invention have the following technical effects: This invention effectively addresses the significant uncertainties and randomness inherent in traditional general-purpose port air monitoring site selection, which relies heavily on manual experience, is highly subjective, and lacks scientific rigor. It integrates local pollutant diffusion models, fuzzy evaluation theory, and cluster analysis techniques. First, a diffusion model simulates the spatial distribution patterns of pollutants under different meteorological conditions, providing an objective data foundation for monitoring site selection. Then, fuzzy evaluation technology quantifies the complex interactions of multiple heterogeneous factors such as environmentally sensitive points, monitoring resources, and safety risks, reducing uncertainties in the site selection process. Cluster analysis is used to spatially optimize the initially selected sites, resolving the potential for unreasonable layouts caused by single mathematical evaluation methods. The resulting site selection scheme accurately reflects the spatiotemporal distribution characteristics of pollutants within the port area, providing a reliable data foundation for the effective control and management of port air pollution and promoting the green and sustainable development of ports. Attached Figure Description

[0012] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a method for monitoring air pollutants at a dock based on a pollutant diffusion model, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a port air pollutant monitoring system based on a pollutant diffusion model provided in an embodiment of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0015] This invention proposes a method for monitoring air pollutants at docks based on a pollutant diffusion model. Figure 1 This is a flowchart of a port air pollutant monitoring method based on a pollutant diffusion model provided in an embodiment of the present invention. See also... Figure 1 Specifically, it includes: S1. Based on meteorological monitoring data of the port area, different meteorological types are classified according to atmospheric stability, and the occurrence probability of each meteorological type is determined.

[0016] In some embodiments, guided by the "Technical Guidelines for Environmental Impact Assessment - Atmospheric Environment", meteorological monitoring data of the port area, specifically including data such as surface wind speed, daytime solar radiation, or nighttime cloud cover, are divided into six types (A, B, C, D, E, F) according to atmospheric stability, and the occurrence probability of each type is calculated. Among them, A (w=1) is strongly unstable with an occurrence probability of 2.5%, B (w=2) is unstable with an occurrence probability of 8.8%, C (w=3) is weakly unstable with an occurrence probability of 12.3%, D (w=4) is neutral with an occurrence probability of 65.1%, E (w=5) is relatively stable with an occurrence probability of 7.2%, and F (w=6) is stable with an occurrence probability of 4.1%.

[0017] S2. Determine the pollutant type based on the port operation type and energy type, and divide the port area into grids.

[0018] In some embodiments, the port's operations primarily handle coal, iron ore, and building materials, and the fuel used for mobile machinery operations is mainly diesel. Based on this, the main pollutant types are identified as follows: j=1 represents PM, j=2 represents NOx, j=3 represents SO2, j=4 represents CO, and j=5 represents VOCs. Based on these pollutant types and the spatial distribution of major operational processes and structures within the port area, the port area is divided into grids, using the land boundary of the wharf as the reference, covering all berths, storage yards, main roads, and fixed operational facilities. Geographic Information System (GIS) software is used to create regular grids for the port area. Following the recommendations of the "Technical Guidelines for Environmental Impact Assessment (Atmospheric Environment)," and considering the port scale and pollution source distribution, a grid spacing of 100 meters × 100 meters is selected, dividing the port area into 15 (columns) × 8 (rows) = 120 grids. This scale effectively balances simulation accuracy and computational efficiency. Each grid is uniquely numbered using a "row number-column number" rule.

[0019] S3. Select a local pollutant diffusion model, combine meteorological type and pollutant type to simulate pollutant diffusion, and calculate the pollutant evaluation concentration for each grid.

[0020] Specifically, by combining the meteorological type and atmospheric pollutant emission type of the port area, a pollutant emission diffusion simulation is conducted in the port area to obtain the diffusion characteristics of pollutants of different emission types under different meteorological types, and the evaluation concentration of pollutants in different grids is calculated. The calculation formula is as follows: in, Let i be the pollutant evaluation concentration in grid i, where i is the grid number. j As for the type of pollutant, w As a weather type, JWhere W represents the total number of pollutant types and W represents the total number of meteorological types. To determine the type of pollutant j Weighting coefficients for potential monitoring and , Weather type w The probability of occurrence, For grid i in weather type w Types of pollutants j average concentration, Weather type w The wind speed below, They are respectively weather types w The lateral and vertical diffusion coefficients below, This indicates the emission intensity of pollutant j.

[0021] In some embodiments, a local pollutant dispersion model is employed, using the aforementioned six meteorological types (w=1~6) and their occurrence probabilities, five pollutant types (j=1~5), and 120 grid divisions as input parameters. Based on these parameters, the local pollutant model is run to simulate each pollutant j under each combination of meteorological type w, requiring a total of 30 simulations (5 pollutants × 6 meteorological types). After each simulation, the model outputs the average concentration of pollutant j at the center of each grid i under each meteorological type w. .

[0022] After obtaining the average concentration value for all grids, the pollutant assessment concentration for each grid is calculated based on this average concentration value using the provided formula. The formula is as follows: Among them, the monitoring weight coefficient Different coefficient values ​​are assigned to pollutants based on their priority in terms of environmental and health impacts. For example: ψ j=1 =0.35, ψ j=2 =0.2, ψ j=3 =0.25, ψ j=4 =0.1, ψ j=5 =0.1, and the sum of all monitoring weight coefficients is 1; the occurrence probability of each meteorological type is directly adopted from the annual occurrence probability of each stability obtained above; for each grid, the pollutant assessment concentration of that grid is obtained by summing all 30 weighted concentration values. The above calculation process is automatically executed for all 120 grids through programming scripts, and finally the pollutant assessment concentration of each grid is obtained.

[0023] S4. Establish a coordinate system with the geometric center of the port area as the origin, determine the coordinates of the center point of each grid, and use the center point of each grid as the potential deployment point for pollution monitoring.

[0024] In some embodiments, the boundary polygon map of the port area is loaded into the software, and the geometric center of the polygon is automatically calculated based on the software's computational geometry capabilities. This point is then set as the origin (0,0) of the local Cartesian coordinate system. With geographic north as the positive Y-axis and east as the positive X-axis, this coordinate system is a Cartesian coordinate system, and the unit is meters.

[0025] Based on the aforementioned coordinate system, the coordinates of the center point of each grid are calculated. Since the grid is a regular rectangle, the coordinates of the center point of each grid can be quickly calculated using its row and column numbers relative to the origin. Then, the coordinates of the center point of each grid are used as potential deployment points for pollutant monitoring. This embodiment successfully transforms the abstract grid into potential monitoring points with precise geographic coordinates and assigns key pollution concentration attributes to each point. It achieves the transformation from theoretical simulation to actual spatial planning, providing clear candidate locations for the final scientific deployment of monitoring equipment.

[0026] S5. Use fuzzy evaluation technology to quantify the influence of external factors on potential deployment points, generate a fuzzy evaluation matrix, and filter the grid based on the fuzzy evaluation matrix to obtain preliminary screening results.

[0027] Specifically, S5 includes: S51. Set n evaluation indicators for each grid, and construct a judgment matrix for the grid based on the evaluation indicators. , where a ik Let m be the original evaluation value of the k-th indicator in the i-th grid, m be the number of grids, and n be the number of evaluation indicators.

[0028] In some embodiments, four evaluation indicators are set for each grid: k=1 pollutant exposure level, k=2 distance from environmentally sensitive points, k=3 power supply convenience, and k=4 operational safety risk. For the 120 grids, the original evaluation value 'a' of the above four indicators is calculated for each grid. ik All the obtained data are organized into a judgment matrix A, which is a 120-row × 4-column matrix, where the rows represent the 1st to 120th grids, the columns represent the 1st to 4th evaluation indicators, and the matrix element a ik It is the original evaluation value of the i-th grid on the k-th indicator.

[0029] The above embodiments digitize and integrate the advantages and disadvantages of all potential deployment points across different dimensions (pollution level, representativeness, feasibility, and safety) into a single judgment matrix A. This judgment matrix A serves as the foundation and data source for subsequent normalization, entropy weight calculation, and fuzzy evaluation. It transforms the abstract influence of multiple factors into concrete data that can be processed by computers and analyzed by mathematical models, thus laying the foundation for the entire fuzzy evaluation technology.

[0030] S52. Normalize the judgment matrix A to obtain the normalized matrix B.

[0031] Specifically, the normalized matrix B mentioned above is: ; in, b ik It is the normalized evaluation value of the i-th grid at the k-th evaluation index, a max and a min These are the maximum and minimum values ​​among all the original evaluation values ​​of the grid under the k-th evaluation index, respectively.

[0032] In some embodiments, firstly, for each grid, the maximum value 'a' of each evaluation metric in that grid is calculated. max and minimum value a min Then, based on the maximum and minimum values, use the formula Calculate the normalized evaluation value b for each indicator in each grid. ik This yields the normalized matrix B.

[0033] S53. Calculate the entropy weight of the evaluation index based on the normalized matrix B, and construct the weight index matrix Z based on the entropy weight and the normalized matrix B.

[0034] Specifically, S53 includes: The entropy value of the k-th index is: ; Among them, H k f is the entropy value of the k-th index. ik It represents the ratio of the normalized evaluation value of the k-th evaluation index of the i-th grid to the sum of the normalized evaluation values ​​of the k-th evaluation index of all grids, and when f ik When =0, ;f ik The calculation formula is: ; The entropy weight of the k-th index is: ; Among them, W k Let be the entropy weight of the k-th evaluation index, and satisfy . ; Construct the weight index matrix Z using the obtained entropy weights and normalization matrix B: ; Among them, z ik z represents the evaluation weight of the i-th grid on the k-th evaluation metric. ik The calculation formula is: ; W is the normalized evaluation value of the i-th grid at the k-th evaluation index. k Let be the entropy weight of the k-th evaluation index.

[0035] In some embodiments, for the normalized matrix B obtained above, the sum of the normalized evaluation values ​​of all grids for each evaluation index (each column) is calculated; then for each element b in matrix B... ik Calculate its specific gravity Based on the above proportions, calculate the entropy value of each indicator k: ; Where m = 120 (total number of grid cells), and ln is the natural logarithm. The entropy values ​​of the four indicators are calculated, and then the entropy weight of each indicator is calculated based on these entropy values: ; Where n=4 (total number of indicators). The entropy weights are combined with a normalized matrix B based on the four indicators. The elements z in the weighted indicator matrix Z are calculated. ik That is, the evaluation weight of the i-th grid on the k-th evaluation index, calculated by the following formula: Finally, the weight index matrix Z is obtained.

[0036] By combining the normalized data B with the entropy weight W, the weight index matrix Z is obtained. This matrix not only eliminates the influence of dimensions but also incorporates the relative importance of each index, thus preparing for the next step of constructing a fuzzy evaluation matrix.

[0037] S54. Set up a comment set, use the membership function to calculate the membership degree of the elements in the weight index matrix Z to the comment set, and then construct the fuzzy evaluation matrix R for each grid. In some embodiments, a rating set consisting of five rating levels is defined: let the rating set be... That is, E = {Excellent, Good, Average, Poor, Poor}, where e represents the rating level for each evaluation indicator k. p Define a membership function for an isosceles triangle. This function is determined by three parameters: the starting point a, the vertex b, and the ending point c. For positive indicators (the larger the value, the better, such as k=3, k=4): a high value corresponds to the rating "excellent"; for negative indicators (the smaller the value, the better, such as k=1, k=2): a low value corresponds to the rating "excellent".

[0038] Using the triangular membership function, the membership degree of all indicators for each grid to the comment set is calculated. Based on the obtained membership degrees, a fuzzy evaluation matrix R is constructed for each grid. This matrix is ​​a 4x5 matrix, where the rows represent the 1st to 4th evaluation indicators, the columns represent the 1st to 5th comment levels, and the matrix elements r kp It is the weighted value z of the k-th indicator. ikThe membership degree of the p-th rating level.

[0039] S55. Perform fuzzy synthesis operation on the fuzzy evaluation matrix R of each grid to obtain the fuzzy evaluation subset of each grid, and normalize and comprehensively score the fuzzy evaluation subset. Based on the comprehensive score, obtain the preliminary screening results.

[0040] Specifically, the above involves performing a fuzzy synthesis operation on the fuzzy evaluation matrix R for each grid. The calculation formula for the fuzzy synthesis operation is as follows: ; Among them, V i Let A be the fuzzy subset of the i-th grid, A be the judgment matrix, and R be the fuzzy subset of the i-th grid. i Let be the fuzzy evaluation matrix for the i-th grid.

[0041] In some embodiments, the calculation formula of fuzzy synthesis operation is used. Calculate the fuzzy evaluation subset V for each grid. i To make the comprehensive evaluation results easier to interpret, the fuzzy evaluation subset V is usually... i Perform normalization so that the sum of its elements is 1.

[0042] Assign a score (e.g., on a percentage scale) to each rating level, then calculate a weighted average score as the overall score S for that grid. i The evaluation rating score vector is set as F=[f1, f2, f3, f4, f5]=[95, 80, 65, 40, 20]; where Excellent = 95 points, Good = 80 points, Average = 65 points, Poor = 40 points, and Fairly Poor = 20 points. Based on the normalized fuzzy evaluation subset and the evaluation rating score vector, the comprehensive score for each grid is calculated. After calculating the comprehensive score for all 120 grids, they are sorted according to the score. According to actual needs, such as planning to deploy 10 monitoring points, the 10 grids with the highest comprehensive scores are selected as the initial screening results.

[0043] Through fuzzy synthesis and comprehensive scoring in this embodiment, the complex multi-index fuzzy evaluation information of each grid is aggregated into a single, comparable comprehensive score. Based on this score, the preliminary site selection scheme with the best comprehensive conditions can be scientifically and objectively screened from a large number of potential deployment points, providing a data foundation for subsequent cluster optimization. Points with high scores are usually locations with a balanced development of pollution concentration, representativeness, feasibility, and safety.

[0044] S6. Agglomerative cluster analysis was used on the preliminary screening results to obtain the clustering results.

[0045] Specifically, the preliminary screening results were analyzed using agglomerative clustering, including: S61. Initialize each grid in the initial screening results as a separate category; S62. Calculate the pairwise Euclidean distance between all categories. The calculation formula is as follows: ; Where x and y are the two categories whose distances are being calculated, x k and y k These are the normalized evaluation values ​​of category x and category y on the k-th indicator, respectively; S63. Set a distance threshold, merge two categories whose Euclidean distance is less than the distance threshold into a new category, and recalculate the Euclidean distance between the new category and all other categories; S64. Repeat S62~S63 until all grid classifications no longer change, and obtain the clustering results.

[0046] In some embodiments, agglomerative clustering analysis is used for the initial screening results. First, each of the 10 selected grid points (P1 ~ P10) is initialized as a separate category. At this point, there are 10 categories: {P1}, {P2}, ..., {P10}. Then, the Euclidean distance between all pairs of categories is calculated, and a distance threshold L = 0.5 is set to find two categories whose distance is less than the threshold L.

[0047] Suppose the distance between {P1} and {P6} is 0.15, which is less than the threshold of 0.5; the distance between {P2} and {P9} is 0.4, which is less than the threshold of 0.5; and the distance between {P8} and {P6} is 0.3, which is less than the threshold of 0.5. Merge them into three new classes: C1={P1,P6,P8}, C2={P2,P9}.

[0048] The categories now become: C1, C2, {P3}, {P4}, {P5}, {P7}, {P10}. Recalculate the distances: Calculate the distances between the new categories C1 and C2 and all other categories. The average linking method is typically used, where the distance between two categories is defined as the average distance between all point pairs in the two categories.

[0049] Repeat the above steps until the distance between all categories is no less than the threshold L, and the classification result is stable. Assume that after several iterations, the 10 points are aggregated into 3 categories. Category A is {P1, P6, P8}, characterized by low pollution concentration, proximity to sensitive points, convenient power access, and low safety risk; Category B is {P2, P5, P9}, characterized by relatively balanced indicators; Category C is {P3, P4, P7, P10}, characterized by high pollutant concentration but poor power access and safety.

[0050] S7. Based on the clustering results and the monitoring requirements of the port area, determine the final deployment points for pollution monitoring.

[0051] Based on the above clustering results and the monitoring requirements of the port area, the most representative points from each category were selected as the final deployment points. The monitoring requirements should meet the following conditions: the monitoring points should be representative of the characteristics of different types of areas within the port area; areas with severe pollution should be given priority monitoring; and points with convenient power supply and low safety risks should be given priority.

[0052] The decision-making process for determining the deployment points: Sites in Category A are suitable as environmental background reference points or sensitive monitoring points. Site P6, with the highest relative pollutant concentration in this category, was selected for deployment to monitor the potential impact of port activities on clean areas. Sites in Category B have moderate indicators. Site P2, with the highest overall score in this category, was selected for deployment, representing the general operational conditions of the port area. Sites in Category C are the key monitoring areas. Site P7, with the best relative convenience of power supply and safety risk indicators in this category, was selected for deployment to monitor the core pollution area while ensuring the safe and stable operation of the equipment. The final determined pollution monitoring deployment points are: P6, P2, and P7.

[0053] Figure 2 This is a schematic diagram of the structure of a port air pollutant monitoring system based on a pollutant diffusion model provided in an embodiment of the present invention. This system is used to execute the port air pollutant monitoring method based on a pollutant diffusion model described in the above embodiment, such as... Figure 2 As shown, the system includes the following modules: The meteorological data processing module is used to classify meteorological data from the port area into different meteorological types based on atmospheric stability and determine the probability of occurrence of each meteorological type. The gridding and pollution source analysis module, connected to the meteorological data processing module, is used to determine the type of pollutants based on the port operation type and energy type, and to divide the port area into grids. The diffusion simulation calculation module, connected to the gridding and pollution source analysis module, is used to select a local pollutant diffusion model, combine meteorological type and pollutant type to simulate pollutant diffusion, and calculate the pollutant evaluation concentration for each grid. The coordinate generation module, connected to the diffusion simulation calculation module, is used to establish a coordinate system with the geometric center of the port area as the origin, determine the coordinates of the center point of each grid, and use the center point of each grid as the potential deployment point for pollution monitoring. The fuzzy evaluation and screening module, connected to the coordinate generation module, is used to quantify the influence of external factors on potential deployment points using fuzzy evaluation technology, generate a fuzzy evaluation matrix, and screen the grid based on the fuzzy evaluation matrix to obtain preliminary screening results. The clustering analysis optimization module, connected to the fuzzy evaluation and screening module, is used to perform agglomerative clustering analysis on the preliminary screening results to obtain the clustering results. The monitoring site deployment decision module, connected to the cluster analysis optimization module, is used to determine the final deployment sites for pollution monitoring by combining the clustering results and the monitoring requirements of the port area.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring atmospheric pollutants at a port based on a pollutant dispersion model, characterized in that, The method comprises the following steps: S1, based on the meteorological monitoring data of the port area, dividing into different weather types according to atmospheric stability, and determining the occurrence probability of each weather type; S2, according to the port operation type and energy type to determine the pollutant type, and grid division of the port area; S3, selecting a local pollutant diffusion model, combining the weather type and the pollutant type to simulate pollutant diffusion, and calculating the pollutant evaluation concentration of each grid; S4, establishing a coordinate system with the geometric center of the port area as the origin, determining the coordinates of the center points of each grid, and taking the center points of each grid as potential layout points for pollution monitoring; S5, using fuzzy evaluation technology to quantify the influence of external factors on the potential layout points, generating a fuzzy evaluation matrix, and screening grids based on the fuzzy evaluation matrix to obtain a preliminary screening result; S6, using condensed clustering analysis on the preliminary screening result to obtain a clustering result; Specifically comprising: S61, initializing each grid in the preliminary screening result as a separate category; S62, calculating the Euclidean distance between all categories, and the calculation formula is as follows: ； wherein x, y are two categories of which the distance is calculated, x k and y k are the normalized evaluation values of category x and category y at the kth index, respectively; S63, setting a distance threshold, merging two categories with a Euclidean distance less than the distance threshold into a new category, and recalculating the Euclidean distance between the new category and all other categories; S64, repeating S62-S63 until all grid classifications no longer change, and obtaining a clustering result; S7, combining the clustering result and the monitoring requirements of the port area to determine the final layout point for pollution monitoring. 2.The method of claim 1, wherein, In S3, combining the weather type and the pollutant type to simulate pollutant diffusion and calculate the pollutant evaluation concentration of each grid, comprising: In combination with the weather type and the atmospheric pollutant emission type of the port area, the port area pollutant emission diffusion simulation is carried out, the diffusion characteristics of different emission types of pollutants under different weather types are obtained, the evaluation concentration of pollutants in different grids is calculated, and the calculation formula is as follows: wherein, is the pollutant evaluation concentration for grid i, i being the grid number, j is the pollutant type, w is the weather type, J is the total number of pollutant types, W is the total number of weather types, is the weight coefficient for potential monitoring of the pollutant type j and , is the occurrence probability of the weather type w , is the average concentration of the pollutant type w in grid i under the weather type j , is the wind speed under the weather type w , are the lateral and vertical dispersion coefficients, respectively, under the weather type w , denotes the emission intensity of the pollutant j. 3.The wharf atmospheric pollutant monitoring method based on a pollutant diffusion model according to claim 1, characterized in that, In S5, using fuzzy evaluation technology to quantify the influence of external factors on the potential layout points, generating a fuzzy evaluation matrix, and screening grids based on the fuzzy evaluation matrix to obtain a preliminary screening result, comprising: S51, set n evaluation indexes for each grid, and construct a judgment matrix of the grid based on the evaluation indexes where a ik is the original evaluation value of the ith grid for the kth index, m is the number of grids, and n is the number of evaluation indexes S52, normalizing the judgment matrix A to obtain a normalized matrix B; S53, calculating the entropy weight of the evaluation index based on the normalized matrix B, and constructing a weight index matrix Z based on the entropy weight and the normalized matrix B; S54, setting a comment set, calculating the membership degree of the elements in the weight index matrix Z to the comment set by using a membership function, and then constructing a fuzzy evaluation matrix R for each grid; S55, performing fuzzy synthesis operation on the fuzzy evaluation matrix R of each grid to obtain a fuzzy evaluation subset of each grid, and performing normalization and comprehensive scoring on the fuzzy evaluation subset to obtain a preliminary screening result according to the comprehensive score. 4.The method of claim 3, wherein, In S52, the normalized matrix B is: ； wherein, , b ik is the normalized evaluation value of the i-th grid at the k-th evaluation index original evaluation value, a max and a min are the maximum value and the minimum value, respectively, among all grid original evaluation values under the k-th evaluation index. 5.The method of claim 3, wherein, In S53, the entropy weight of the evaluation index is calculated based on the normalized matrix B, and the weight index matrix Z is constructed based on the entropy weight and the normalized matrix B, comprising: The entropy value of the kth index is: ； wherein H k is the entropy value of the kth index, f ik is the ratio of the normalized evaluation value of the kth evaluation index of the ith grid to the sum of the normalized evaluation values of the kth evaluation index of all grids, and when f ik = 0, ; f ik The calculation formula is: ; The entropy weight of the kth index is: ； wherein W k is the entropy weight of the kth evaluation index, and satisfies ; The weight index matrix Z is constructed by using the obtained entropy weight and the normalized matrix B: ； wherein z ik represents the evaluation weight of the i-th grid on the k-th evaluation index, z ik The calculation formula is: ; is the normalized evaluation value of the i-th grid on the k-th evaluation index original evaluation value, W k is the entropy weight of the k-th evaluation index. 6.The method of claim 3, wherein, In the S55, a fuzzy synthesis operation is performed on the fuzzy evaluation matrix R of each grid, including: The calculation formula of the fuzzy synthesis operation is: ； Where, V i is the fuzzy evaluation matrix of the i-th grid. A is the judgment matrix, R i is the fuzzy evaluation matrix of the i-th grid.

7. A port atmospheric pollutant monitoring system based on pollutant dispersion model for implementing the port atmospheric pollutant monitoring method based on pollutant dispersion model according to any one of claims 1-6, characterized in that, The system comprises the following modules: The meteorological data processing module is configured to divide the port area into different meteorological types according to atmospheric stability based on meteorological monitoring data of the port area, and determine the occurrence probability of each meteorological type; The grid division and pollution source analysis module is connected to the meteorological data processing module, and is configured to determine the type of pollutants according to the type of port operation and the type of energy, and divide the port area into grids; The diffusion simulation calculation module is connected to the grid division and pollution source analysis module, and is configured to select a local pollutant diffusion model, simulate the diffusion of pollutants in combination with the meteorological type and the type of pollutants, and calculate the evaluation concentration of pollutants in each grid; The coordinate generation module is connected to the diffusion simulation calculation module, and is configured to establish a coordinate system with the geometric center of the port area as the origin, determine the coordinates of the center points of the grids, and take the center points of the grids as potential layout points for pollution monitoring; The fuzzy evaluation screening module is connected to the coordinate generation module, and is configured to use fuzzy evaluation technology to quantify the influence of external factors on the potential layout points, generate a fuzzy evaluation matrix, and screen the grids based on the fuzzy evaluation matrix to obtain a preliminary screening result; The clustering analysis optimization module is connected to the fuzzy evaluation screening module, and is configured to use agglomerative clustering analysis on the preliminary screening result to obtain a clustering result; The monitoring point decision module is connected to the clustering analysis optimization module, and is configured to determine the final layout points for pollution monitoring in combination with the clustering result and the monitoring requirements of the port area.

Images