A method for dividing urban river prevention and control units

Abstract

The present application relates to a kind of urban river prevention and control unit division method, by in prevention and control unit division index system basis, to water unit grid division, prevention and control unit hierarchical processing, prevention and control unit boundary refinement and adjustment, prevention and control unit priority analysis, using index system construction, water unit grid division, clustering analysis, analytic hierarchy process, fuzzy evaluation method, for "single control element", "unit demarcation one-size-fits-all" such as prevention and control unit division problem, proposed a kind of prevention and control unit based on water ecology and water environment characteristics as spatial basis urban river prevention and control unit division method, and can deduce water pollution control and water ecological restoration scheme from management goal, not only facilitate pollution reduction performance evaluation, and more practical operability, realized the grade ordering of prevention and control unit, provide strong theoretical basis for carrying out environmental governance.

Description

TECHNICAL FIELD

[0001] The application relates to a city river prevention and control unit division method and belongs to the technical field of water ecological environment protection. BACKGROUND

[0002] Key indicators for water ecological environment management and control are water quality and total pollutant amount, and division of city prevention and control units is an important means of fine management of a river basin. A river basin is a complete closed unit of water system movement, and the principle of 'classification, zoning, grading and staging' should be followed in river basin water pollution treatment and water ecological restoration. As the smallest spatial carrier for water ecological environment management of a river basin, a prevention and control unit can realize the organic combination of river basin management and administrative management and the effective connection of regional pollution discharge and river basin water ecological environment quality, and is one of important means for river basin water pollution treatment and water ecological restoration.

[0003] Common methods for division of a river basin prevention and control unit mainly include a division method based on a hydrological unit, a division method based on administrative region planning and a division method based on a water ecological system. The division method based on a hydrological unit is to divide a river basin prevention and control unit according to the catchment characteristics of a river basin and to construct a hydrological index library, but only the hydrological index is considered, the control elements are too single, and the increasing water environment and water ecological protection requirements cannot be met. The division method based on administrative region planning is mainly based on administrative region boundaries, combined with water body basin characteristics, water system characteristics and other elements to divide a prevention and control unit, but this method only considers the administrative region boundaries, water system characteristics and other index elements, and lacks comprehensive consideration of water quality and water ecology, so that the prevention and control unit divided is more biased towards administrative management and it is difficult to ensure accurate management of each small river basin. The division method based on a water ecological system is to maintain the biodiversity of a water ecological system and curb the degradation of a river basin water ecological system, and the division of a prevention and control unit is carried out according to water ecological index parameters on the basis of water body ecological characteristic identification and ecological function determination. Although the division method based on a water ecological system is a relatively advanced division method at present, it still has some shortcomings. On the one hand, only the water ecological function and characteristics are considered in the division process, and the similarities and differences between regions are ignored, and different habitat characteristics determine the ecological restoration measures implemented in the region to some extent. On the other hand, the boundary conditions and scale size of the water ecological division are difficult to control. If the management angle is excessively emphasized and the water ecological division boundary is modified to be as close to or coincided with the administrative boundary as possible, the integrity of the water ecological division will be damaged and the spatial characteristic difference cannot be fully reflected. SUMMARY

[0004] In order to solve the above technical problems, the application provides a city river prevention and control unit division method, and the specific technical scheme comprises the following steps:

[0005] Step 1: Obtain the prevention and control unit division index system;

[0006] Step 2: Water body unit grid division, establish a basic database according to the prevention and control unit division index system, the basic database includes: hydrological unit division map, administrative division map, electronic water system map, hydrological and water quality monitoring site distribution map, the establishment method of the basic database is as follows:

[0007] Step 21: Collection and vectorization of maps, using Arc GIS platform to align the administrative division map, hydrological and water quality monitoring site map, give appropriate geographic coordinate system, and vectorize it;

[0008] Step 22: Sub-basin (hydrological unit) division, the resolution of digital elevation model used in sub-basin (hydrological unit) division is 30m x 30m. Use the watershed division sub-module of the SWAT plug-in of the Arc GIS platform to divide the sub-basin; there are two advantages of using the SWAT plug-in to divide the sub-basin: ① the model supports adding electronic water system map and manually adding and deleting node number; under the same conditions, the generated simulated water system is closer to the actual situation and is convenient for controlling the number of sub-basins; ② the generated sub-basins can be used for hydrological and water quality simulation, non-point source and water environment capacity quantitative calculation of the SWAT model, which is convenient for docking with the "total amount control of prevention and control unit" technology.

[0009] Step 23: Drawing of environmental pressure map, using kernel density analysis (Kernel Density Estimation) method, using kernel function to fit each point or line into a smooth conical surface and calculate its density in the surrounding neighborhood; this patent uses GIS-based kernel density analysis technology, i.e. using spatial weighted difference method, to analyze various environmental pressures (such as domestic sewage, industrial wastewater and land use, etc.) in the watershed according to the spatial distribution and emission intensity of pollution sources, and obtain the spatial distribution density map.

[0010] Step 3: Hierarchical processing of prevention and control unit, according to the principles of prevention and control unit division, based on Arc GIS platform, project transform the water ecological zoning map, sub-basin division map, administrative division map, water system map, hydrological and water quality monitoring site distribution map and surface water environmental function regionalization map obtained in step 2 to convert them into a unified projection type, then use Arc GIS platform to perform spatial overlay analysis on the unified projection multiple thematic maps;

[0011] Step 4: Prevention and control unit boundary refinement and adjustment. On the basis of the water quality improvement target in the prevention and control unit, the protection of water ecological system function in the unit also needs to be considered. Therefore, after obtaining the preliminary prevention and control unit region by analyzing various basic geographic information data on the Arc GIS platform, the relevant water environment and water ecological element index data are obtained through field investigation and sampling monitoring, the K-means clustering method is applied to classify the sampling points in the basin, and the prevention and control unit boundary is further adjusted and refined;

[0012] Step 5: Prevention and control unit priority analysis. The fuzzy analytic hierarchy process (FAHP) is proposed by combining the analytic hierarchy process and the fuzzy evaluation method. The fuzzy analytic hierarchy process (FAHP) can better integrate the advantages of the analytic hierarchy process and the fuzzy evaluation method, and more objectively evaluate the fuzzy comprehensive evaluation of water ecological value. The specific steps of the fuzzy analytic hierarchy process (FAHP) are as follows:

[0013] Step 51: Check the consistency of the fuzzy analytic hierarchy process (FAHP) decision by consistency ratio (CR). The test formula is as follows:

[0014]

[0015] In formula 5: RI is a random consistency index, which depends on the matrix dimension (n);

[0016] CI is the consistency index, which can be expressed as:

[0017]

[0018] In formula 6: λ max is the main eigenvalue.

[0019] If CR≤0.1, it is considered that the subjective evaluation of the decision is consistent;

[0020] Step 52: The parameters of each prevention and control unit are different, and the standardization calculation is carried out to obtain the corresponding weight W’ i，j :

[0021]

[0022] In formula 7: W’ i，j is the standardized value of the i-th parameter of the j-th prevention and control unit, W’ i，max and W’ i，min are the original upper and lower boundaries of the i-th parameter, and W i，j is the original value of the i-th parameter of the j-th prevention and control unit;

[0023] Step 53: Calculate the entropy value H i of the prevention and control unit parameter:

[0024]

[0025]

[0026]

[0027] In formula 8-10: P i，j is the i-th parameter weight of the j-th prevention and control unit;

[0028] Step 54: Evaluate the weight x of the prevention and control unit parameter i :

[0029]

[0030] In formula 11: x i is the weight of a single factor, satisfying and 0 < x i < 1. It can be seen that the index weight and the importance of the evaluation index increase with the decrease of the entropy value;

[0031] Step 55: The priority (F j ) of the prevention and control unit is calculated by adding the product of the normalized parameter value and the corresponding weight, and the formula is as follows:

[0032]

[0033] In formula 12: x i represents the weight of each evaluation parameter; W i，j is the normalized value of the i-th parameter of the j-th prevention and control unit;

[0034] Step 6: Based on the FAHP method, the management and control priority of each sampling point is obtained, and then the average value of the sampling point priority in several sections is taken according to the distribution of the sampling points in each section, to obtain the priority score of the corresponding management and control section.

[0035] Further, the specific method of kernel density analysis (Kernel Density Estimation) in step 23 is:

[0036] Suppose there are industrial wastewater discharge outlets x1, x2,..., x n , then the kernel density estimation at x is:

[0037]

[0038] In formula 1: x is the estimated discharge outlet position, is the kernel function, h is the search radius, and x-x i represents the distance from the estimation point to the sample x iIn the nuclear density analysis, the points falling into the search area have different weights, the points or lines close to the search center are given greater weights, and vice versa, the points or lines far from the search center are given smaller weights, as the distance from the search radius increases, the generated density map is smoother and the generalization degree is higher, and vice versa, the information in the generated density map is more detailed.

[0039] Further, if the watershed spatial scale is large in step 3, it is difficult to directly determine the control unit, then the control unit is processed in stages, and the specific steps are as follows:

[0040] Step 31: Determine the control point of the first-level control unit, preferentially consider the intersection (or approximate intersection) of the three of water system and administrative district boundary as the control point of the first-level control unit, and as much as possible meet the division principle of the control unit, so as to facilitate the delineation of the first-level control unit;

[0041] Step 32: Delineate the first-level control unit, take the first-level control point as the starting point, and use the method combining qualitative analysis and expert judgment to find the boundary of the water environment function regionalization and the administrative district overlapping or close to overlapping on the upstream of the control point, and use the capture function (Snap) of ArcGIS to delineate the closed polygon in the new layer;

[0042] Step 34: Repeat step 32 until all first-level control units are divided;

[0043] Step 35: Delineate the second-level control unit, the second-level control unit is a subunit in the first-level control unit, on the basis of the first-level control unit, divide the second-level control unit, mainly divide the first-level control unit by the administrative district boundary, and secondarily divide the first-level control unit by the sub-watershed and the water environment function regionalization, determine the control point of the second-level control unit, and use the capture function (Snap) of ArcGIS to delineate the boundary of the second-level control unit;

[0044] Step 36: Superimpose the above special maps and control units again, according to the order from the tributary to the main stream and from the upstream to the downstream, correct the control unit in detail, and finally obtain the division scheme of the first-level and second-level control units of the watershed.

[0045] Further, in step 35, the special water environment function area and the key pollution area are preferentially controlled, and they are taken as separate control units. The special water environment function area includes nature reserves and mining areas.

[0046] Further, in step 36, if the watershed scale is small, the number of sub-watershed areas and administrative districts inside the watershed is small, then the first-level control unit does not need to be divided, and the township administrative district special map is used as the control unit division.

[0047] Further, the step 4 applies the K-means clustering method to classify each sampling point of the basin, and couples it with the gridization of the basin water body unit based on the environmental stress response, to comprehensively analyze the natural catchment law, ecological service function and ecological environment sensitivity, finely divide the basin, and obtain the corresponding water body prevention and control unit, and the specific process is as follows:

[0048] Step 41: Basic data acquisition, according to the actual situation of the basin, comprehensively considering the historical data of water quality monitoring section, and complying with the representative principle of data, on-site sampling is carried out on the monitoring section, and the specific indexes include pH, dissolved oxygen, nitrite nitrogen, nitrate nitrogen, ammonia nitrogen, active phosphate, turbidity, total nitrogen, total phosphorus, chemical oxygen demand, temperature, salinity, water depth, flow rate and flow; the bottom detection indexes include total carbon, total nitrogen and total phosphorus; the water biota status survey is carried out on the phytoplankton, zooplankton, benthic animals and macrophytes in the basin, and three sections of left, middle and right are set according to each section; the water biota detection indexes include phytoplankton (dominant species and density), zooplankton (dominant species, density and biomass), macro benthic animals (dominant species, density and biomass) and macrophytes (dominant species, density and biomass);

[0049] Step 42: Cluster analysis, according to the analysis of multiple characteristics of the basic data in step 41, the statistical quantity of similar indexes is found, and on this basis, the statistical quantity is taken as the standard for dividing types, the sample objects with greater similarity are placed in the same class, and the sample objects with greater similarity are placed in another class, and the relationship between the classes is aggregated according to the distance, the objects with close relationship are clustered in the same class, and the objects with distant relationship are clustered in different classes, wherein, the K-means clustering is a classical algorithm in cluster analysis, and the main clustering idea is to represent the clustering of the objects closest to the K points in space as the center, and the algorithm of K-means clustering analysis is as follows: (1) initialization: randomly selecting initial clustering centers,

[0050] (2) loop: ① condensing the remaining records to the class center according to the nearest principle, ② calculating the class center of each initial classification, ③ reclassifying with the calculated center position, ④ repeating the cycle until the class center position converges.

[0051] Further, the algorithm of K-means clustering analysis in step 42 is as follows:

[0052] Suppose the data set is X = {x1, x2,..., x i ,..., x n}, the number of clusters is K, and the clustering center is C = {c1, c2,..., c i ,..., c k},

[0053] a. randomly select K samples from the data set X as initial cluster centers;

[0054] b. for each sample xi (i = 1, 2,..., n) in the data set, calculate its distance from the cluster center cj (j = 1, 2,..., k), and the distance calculation formula is:

[0055]

[0056] In the formula: m is the dimension of the sample;

[0057] According to the calculated distance of each sample to the cluster center, find the minimum distance and divide the sample into the corresponding cluster;

[0058] The cluster center is re-calculated and updated according to formula one, and the result of the objective function is calculated according to formula two, and the calculation formula is as follows:

[0059]

[0060]

[0061] The cluster center and the objective function are judged, and the algorithm is ended if the requirement is reached, otherwise step b is continued.

[0062] The beneficial effects of the present application are:

[0063] The complex watershed water environment and water ecological problems are divided into each prevention and control unit, so that specific watershed water ecological and water environment management measures and policies can be effectively implemented and implemented. The water quality improvement target in each prevention and control unit is not only clear, but also the protection of the unit water ecological system function, so as to protect the implementation of the present application to realize the improvement of the quality of the watershed water quality and water ecology. Different dimensions are comprehensively divided into each prevention and control unit with different characteristics, and the different characteristics of the prevention and control unit are correspondingly matched, and the appropriate technology is followed to optimize the combination, so as to solve the specific problems of the prevention and control unit. The prevention and control unit is the smallest spatial carrier of watershed water ecological and water environment management, and through the present application, the organic combination of watershed management and administrative management can be realized, and the effective connection of regional pollution and watershed water environment ecological quality can be realized, which is an important new method for fine watershed water pollution control and water ecological restoration.

[0064] The present application realizes the grade sorting of the prevention and control unit by the prevention and control unit division index system, the water unit grid division, the hierarchical processing of the prevention and control unit, the prevention and control unit boundary refinement and adjustment, and the prevention and control unit priority analysis, which provides a strong theoretical basis for the development of environmental governance.

[0065] The application adopts an index system construction, water body unit grid division, cluster analysis, analytic hierarchy process and fuzzy evaluation method, and in view of the problems of single control element and unit division, proposes a city river control unit division method taking the water ecological and water environmental characteristics as the spatial basis, and can deduce the fine water pollution control and water ecological restoration scheme from the management target, which is not only convenient for pollution reduction performance evaluation, but also has practical operability. DETAILED DESCRIPTION

[0066] The city river control unit division method of the application comprises the following steps.

[0067] Step 1: Obtain the control unit division index system. The control unit division index system is directly obtained in the patent, and the following describes how the control unit division index system is obtained in actual work, and the obtaining process is as follows: combined with the screening principle of the control unit division index, and considering the related zoning results (such as water resources zoning, main function regionalization and ecological function regionalization), the index system for control unit division is constructed from five aspects of natural environment background, water environment pressure, water environment state, water ecological condition and administrative management index. The natural environment background index mainly reflects the characteristics of the natural environment of the basin, such as water resources condition, hydrological characteristics, topographic characteristics, etc.; the water environment pressure index mainly reflects various factors affecting the water environment quality, for example, population pressure, water and electricity development, domestic pollution, industrial pollution, geological disasters, livestock breeding, river bank hardening, etc.; the water environment state index mainly reflects the water environment quality situation, such as the change of regional water quality index and the water quality standard condition, etc.; the water ecological condition index mainly reflects the water ecological health condition, including the water ecological service function condition, water ecological community structure, ecological diversity, etc. There are complex interconnections among various indexes, and the natural environment background has important supply, support and assimilation functions for the water environment pressure, water environment state and water ecological condition; the water environment pressure is an important driving force for the change of water environment state and water ecological condition; and the natural environment background, water environment state and water ecological condition jointly determine the ecological environment and sustainability of the basin. The following types of indexes need to be considered in the control unit division of the application:

[0068] (1) Natural environment background index. Including the basic characteristics of the basin (basin range, water system distribution, etc.), water quality historical change trend, etc.

[0069] (2) Water environment pressure index. Land use condition of the basin, pollution source distribution condition (industrial, agricultural and domestic pollution sources).

[0070] (3) Water environment state index. Basin water function regionalization, eutrophication zoning condition, water quality (substrate) index change condition, etc.

[0071] (4) Water ecological condition index. Water ecological function regionalization, ecological diversity, water ecological service function, etc. Among them, the water ecological service function includes the biomass, density, dominant species, etc. of water plankton, phytoplankton, macrophyte and benthic animal.

[0072] (5) Administrative management index. Administrative division boundary.

[0073] The established prevention and control unit division index system is shown in Table 1. According to the constructed index system library, reasonable selection and adjustment are made in combination with the local actual situation.

[0074] Table 1 Prevention and control unit division index system

[0075]

[0076]

[0077] Step 2: Water unit grid division. Establishment of basic database and pre-processing. The basic database is established according to the prevention and control unit division index system, mainly including: hydrological unit division map, administrative division map, electronic water system map, hydrological and water quality monitoring site distribution map, etc. The database is established as follows:

[0078] Step 21: Collection and vectorization of maps. The administrative division map, hydrological and water quality monitoring site map, etc. obtained in the early stage are aligned using the Arc GIS platform, appropriate geographic coordinate system is given, and vectorization is performed.

[0079] Step 22: Division of sub-basin (hydrological unit). The digital elevation model resolution used for division of basin (hydrological unit) is 30m x 30m. The SWAT (plug-in basin division sub-module of Arc GIS platform is used for division of sub-basin. There are two advantages of using SWAT plug-in to divide sub-basin: ① The model supports adding electronic water system map and manually adding and deleting node number; under the same conditions, the generated simulated water system is closer to the actual situation and is convenient for controlling the number of sub-basins; ② The generated sub-basins can be used for hydrological and water quality simulation, non-point source and water environment capacity quantitative calculation of SWAT model, which is convenient for docking with the "total amount control of prevention and control unit".

[0080] Step 23: Drawing of the environmental pressure map. The Kernel Density Estimation method is used to fit each point or line into a smooth conical surface and calculate its density in the surrounding neighborhood. In this study, the GIS-based Kernel Density Estimation technique is used, which is a spatially weighted interpolation method, to analyze various environmental pressures (such as domestic sewage, industrial wastewater, and land use) in the watershed based on the spatial distribution and emission intensity of pollution sources to obtain their spatial distribution density map. The specific analysis process and main formula are as follows. Taking industrial wastewater as an example, suppose there are industrial wastewater discharge outlets x1, x2,..., x n The kernel density estimation at x is:

[0081]

[0082] In the formula, x is the estimated discharge location, is the kernel function, h is the search radius, and x-xi represents the distance from the estimation point to the sample xi. In kernel density analysis, points falling within the search area have different weights. Points or lines close to the search center are given greater weights, while points or lines far from the search center are given smaller weights. As the distance from the search radius increases, the generated density map becomes smoother and more generalized. Conversely, the generated density map contains more detailed information.

[0083] Step 3: Hierarchical processing of the control unit. According to the principles of control unit division, based on the ArcGIS platform, the water ecological zoning map, sub-basin division map, administrative division map, water system map, hydrological and water quality monitoring station distribution map, and surface water environmental function zoning map obtained in the above steps are projected and transformed to convert them into a unified projection type. Then, using the ArcGIS platform, the above-mentioned multiple thematic maps with unified projection are subjected to spatial overlay analysis. If the spatial scale of the watershed is large, it is difficult to directly determine the control unit, and it is appropriate to perform hierarchical processing on the control unit. The specific steps are as follows:

[0084] Step 31: Determination of control points for the first-level control unit. Priority is given to the intersection (or approximate intersection) of the water system and the administrative boundary as the control point of the first-level control unit. The division principles of the control unit are considered as much as possible to facilitate the delineation of the first-level control unit.

[0085] Step 32: Delineation of the first-level control unit. Taking the first-level control point as the starting point, a combination of qualitative analysis and expert judgment is used to find the overlapping or nearly overlapping boundaries of the water environmental function zoning and administrative division upstream of the control point. The capture function (Snap) of ArcGIS is used to delineate a closed polygon in a newly created layer; this operation is repeated until all first-level control units are divided.

[0086] Step 34: delineate the 2nd level prevention and control unit. The 2nd level prevention and control unit is a sub-unit within the 1st level prevention and control unit. Based on the 1st level prevention and control unit, the 2nd level prevention and control unit is divided. The administrative boundary is used as the main method to divide the 1st level prevention and control unit, and the sub-basin and water environment function zone are used as the auxiliary method. The control points of the 2nd level prevention and control unit are determined, and the Snap function of Arc GIS is used to delineate the boundary of the 2nd level prevention and control unit. In addition, special water environment function zones (such as nature reserves and mining areas) and key pollution areas are given priority control, and are recommended to be treated as separate prevention and control units.

[0087] Step 35: superimpose the above thematic maps and prevention and control units again, and make detailed corrections to the prevention and control units in the order of tributaries to the main stream, from upstream to downstream, to finally obtain the division scheme of the 1st and 2nd level prevention and control units in the basin.

[0088] If the basin scale is small and the number of sub-basins and administrative areas within the basin is small, it is not necessary to divide the 1st level prevention and control unit. If necessary, the township administrative area thematic map can be used as a reference for prevention and control unit division, and the rest of the methods are the same as above.

[0089] Step 4: prevention and control unit boundary refinement and adjustment. The prevention and control unit of the water ecological environment pollution control in the basin needs to consider the protection of the water ecological system function within the unit in addition to the clear water quality improvement target within the unit. Therefore, after obtaining the preliminary division of the prevention and control unit area by analyzing various basic geographic information data on the Arc GIS platform, relevant water environment and water ecological element index data are obtained through field investigation and sampling monitoring. The K-means clustering method is applied to classify the sampling points in the basin, and the prevention and control unit boundary is further adjusted and refined.

[0090] Step 41: obtain basic data. According to the actual situation of the basin, the historical data of the water quality monitoring section are comprehensively considered, and the data representative principle is followed to conduct on-site sampling of the monitoring section. The specific indicators include pH, dissolved oxygen, nitrite nitrogen, nitrate nitrogen, ammonia nitrogen, active phosphate, total chromium, turbidity, total nitrogen, total phosphorus, chemical oxygen demand, temperature, salinity, water depth, flow rate and flow volume; the bottom sediment detection indicators include total carbon, total nitrogen and total phosphorus. The current situation of aquatic organisms in the basin is investigated, including phytoplankton, zooplankton, benthic animals and macrophytes. Three sections are set up for each section, i.e. left, middle and right. The aquatic organism detection indicators include: phytoplankton (dominant species and density), zooplankton (dominant species, density and biomass), macrobenthos (dominant species, density and biomass) and macrophytes (dominant species, density and biomass).

[0091] Step 42: Cluster analysis. According to the multiple characteristics of the base data in step 41, statistical quantities of similar indicators are found, and on this basis, the statistical quantities are taken as the standard for dividing types, similar sample objects are placed in the same class, and other sample objects with greater similarity are placed in another class. The objects with close relationships are clustered in the same class, and the objects with relatively distant relationships are clustered in different classes. Among them, K-means clustering is a classic algorithm in cluster analysis. The main clustering idea is to represent K points in space as the center of clustering, and to classify the objects closest to them. The algorithm of K-means clustering analysis can be summarized as follows: (1) initialization: randomly select initial cluster centers. (2) loop: ① according to the nearest principle, the remaining records are condensed to the class center. ② calculate the class center of each initial classification. ③ re-cluster with the calculated center position. Repeat the loop until the class center position converges. The specific steps are as follows:

[0092] Suppose the data set is X = {x1, x2,..., xn}, the number of clusters is K, and the cluster centers are C = {c1, c2,..., ck},..., cn}. i ,..., xn}, the number of clusters is K, and the cluster centers are C = {c1, c2,..., ck},..., cn}. n i ,..., xn}, the number of clusters is K, and the cluster centers are C = {c1, c2,..., ck},..., cn}. k

[0093] a. Randomly select K samples from the data set X as the initial cluster centers;

[0094] b. For each sample xi (i = 1, 2,..., n) in the data set, calculate its distance from the cluster center cj (j = 1, 2,..., k). The distance calculation formula is:

[0095]

[0096] In the formula, m is the dimension of the sample.

[0097] According to the calculated distance of each sample to the cluster center, find the minimum distance and divide the sample into the corresponding cluster;

[0098] Recalculate and update the cluster center according to formula one, and calculate the result of the objective function according to formula two. The calculation formula is as follows:

[0099]

[0100]

[0101] Judge the cluster center and the objective function. If the requirement is met, the algorithm ends, otherwise continue to step b.

[0102] ​​Overall, the K-means clustering method is applied to classify each sampling point in the basin, which is coupled with the grid-based watershed water unit based on environmental stress response. The natural catchment law, ecological service function and ecological environment sensitivity are comprehensively analyzed to finely divide the basin, and the corresponding water body prevention and control unit can be obtained.

[0103] Step 5: Priority analysis of prevention and control unit. The present application combines the analytic hierarchy process and fuzzy evaluation method to propose a fuzzy analytic hierarchy process (FAHP), which can better integrate the advantages of the above two methods and more objectively evaluate the fuzzy comprehensive evaluation of water ecological value. The calculation method and specific steps are as follows:

[0104] Step 51: To judge the consistency of fuzzy analytic hierarchy process (FAHP), the consistency ratio (CR) is used for testing:

[0105]

[0106] In the formula: RI is a random consistency index, which depends on the matrix dimension (n);

[0107] CI is the consistency index, which can be expressed as:

[0108]

[0109] In the formula: λ max is the main eigenvalue.

[0110] If CR≤0.1, it is considered that the subjective evaluation of the decision is consistent;

[0111] Step 52: The parameters of each prevention and control unit considered in the present application are different, and the standardization calculation is carried out to obtain the corresponding weight W' i，j :

[0112]

[0113] In the formula: W' i，j is the standardized value of the i-th parameter of the j-th prevention and control unit, W' i，max and W' i，min are the original upper and lower boundaries of the i-th parameter, respectively, and W i，j is the original value of the i-th parameter of the j-th prevention and control unit.

[0114] Step 53: Calculate the entropy value H i of the prevention and control unit parameter:

[0115]

[0116]

[0117]

[0118] In the formula, P i，j is the i-th parameter weight of the j-th prevention and control unit;

[0119] Step 54: evaluate the weight x of the prevention and control unit parameter i :

[0120]

[0121] In the formula, x i is the weight of a single factor, satisfying and 0 < x i < 1. It can be seen that the index weight and the importance of the evaluation index increase with the decrease of the entropy value;

[0122] Step 55: the priority (F j ) of the prevention and control unit is calculated by adding the product of the normalized parameter value and the corresponding weight, and the formula is as follows:

[0123]

[0124] In the formula, x i represents the weight of each evaluation parameter; W i，j is the normalized value of the i-th parameter of the j-th prevention and control unit.

[0125] Based on the FAHP method, the management and control priority of each sampling point is obtained, and then the average value of the sampling point priority in a plurality of sections is obtained according to the distribution of the sampling points in the sections, so as to obtain the priority score of the corresponding management and control section.

[0126] The prevention and control unit based on the water ecological and water environmental characteristics and targets is mainly based on the water ecological region and the water function region, and according to the differences and similarities of the environmental elements, the water ecological system characteristics and the ecological service functions in different regions, the basin and the water body are divided into different spatial units, so as to realize the goal of water ecological and water environmental protection. It has been proved that taking the prevention and control unit based on the water ecological and water environmental characteristics as the spatial basis, and deducing the water pollution control and water ecological restoration scheme from the management target, not only facilitates the pollution reduction performance evaluation, but also has better practical operability.

[0127] Based on the above ideal embodiments according to the present application, through the above description, relevant personnel can certainly make various changes and modifications without deviating from the technical idea of the present application.

Claims

1. A method for dividing urban river control units, characterized in that: Includes the following steps: Step 1: Obtain the indicator system for dividing prevention and control units; Step 2: Water body unit grid division. A basic database is established based on the prevention and control unit division index system. The basic database includes: hydrological unit division map, administrative division map, electronic water system map, and distribution map of hydrological and water quality monitoring stations. The method for establishing the basic database is as follows: Step 21: Map collection and vectorization. Use the ArcGIS platform to register the administrative division map, hydrological and water quality monitoring station map, assign a suitable geographic coordinate system, and vectorize them. Step 22: Sub-basin (hydrological unit) division. The sub-basin division is performed using the watershed division sub-module of the SWAT plugin in the ArcGIS platform. Step 23: Drawing the environmental pressure map. Using the kernel density estimation method, the kernel function is used to fit each point or line to a smooth conical surface and calculate its density in the surrounding neighborhood. Step 3: Hierarchical processing of prevention and control units. Based on the ArcGIS platform, the water ecological zoning map, sub-basin division map, administrative division map, water system map, hydrological and water quality monitoring station distribution map and surface water environment functional zoning map obtained in Step 2 are projected and transformed into a unified projection type. Then, multiple thematic maps with unified projection are spatially overlaid and analyzed using the ArcGIS platform. Step 4: Refinement and adjustment of prevention and control unit boundaries. After analyzing various basic geographic information data using the ArcGIS platform and obtaining the initially delineated prevention and control unit areas, relevant water environment and water ecological element index data are obtained through field investigation and sampling monitoring. The K-means clustering method is applied to classify the sampling points in the watershed and further refine the boundaries of the prevention and control units. Step 5: Priority analysis of prevention and control units. Combining the Analytic Hierarchy Process (AHP) and fuzzy evaluation method, a fuzzy AHP (FAHP) is proposed. FAHP integrates the advantages of both AHP and fuzzy evaluation method, providing a more objective fuzzy comprehensive evaluation of water ecological value. The specific steps of FAHP are as follows: Step 51: Test the decision consistency of fuzzy hierarchical analysis (FAHP) using the consistency ratio (CR). The test formula is as follows: In Equation 5: RI is the random consistency index, which depends on the matrix dimension (n); CI stands for Consistency Index, and is expressed as: In Equation 6: λ max Principal eigenvalues; If CR≤0.1, then the subjective evaluation of the decision is considered to be consistent; Step 52: Since the parameter ranges for each prevention and control unit are different, standardize the calculations to obtain the corresponding weights W'. i，j : In Equation 7: W' i，j W' is the standardized value of the i-th parameter of the j-th control unit. i，max and W' i，min W represents the original upper and lower boundaries of the i-th parameter. i，j This represents the original value of the i-th parameter in the j-th control unit; Step 53: Calculate the entropy value H of the control unit parameter. i : In Equation 8-10: P i，j It is the weight of the i-th parameter of the j-th prevention and control unit; Step 54: Evaluate the weights x of the control unit parameters i : In Equation 11: x i For the weight of a single factor, satisfying and 0 < x i <1 indicates that the weight of the indicator and the importance of the evaluation indicator increase as the entropy value decreases; Step 55: Priority of control units (F) j The result is obtained by adding the products of the normalized parameter values ​​and their corresponding weights, as shown in the following formula: In Equation 12: x i W' represents the weight of each evaluation parameter. i，j This is the standardized value of the i-th parameter for the j-th control unit; Step 6: Based on the FAHP method, obtain the control priority of each sampling point, and then take the average of the priorities of the sampling points in several sections according to the distribution of the sampling points in each section to obtain the priority score of the corresponding control section.

2. The method for dividing urban river control units according to claim 1, characterized in that: The specific method of kernel density estimation in step 23 is as follows: Assume there are industrial wastewater discharge outlets x1, x2, ..., x n Then the kernel density estimate at x is: In Equation 1: x is the location of the emission outlet to be estimated. Here, h is the kernel function, and xx is the search radius. i Indicates the estimated point to sample x i In kernel density analysis, points falling within the search area have different weights. Points or lines closer to the search center are given greater weight, while those farther away have less weight. As the distance from the search radius increases, the generated density map becomes smoother and more generalized; conversely, the generated density map contains more detailed information.

3. The method for dividing urban river control units according to claim 1, characterized in that: If the watershed has a large spatial scale in step 3, making it difficult to directly determine its control units, then the control units are classified. The specific steps are as follows: Step 31: Determine the control points of the Level 1 control unit. Priority should be given to the intersection of the water system boundary, the watershed boundary, and the administrative region boundary as the control points of the Level 1 control unit. The division principle of the control unit should be met as much as possible to facilitate the depiction of the Level 1 control unit. Step 32: Depict Level 1 control units. Starting from Level 1 control points, use a combination of qualitative analysis and expert judgment to find the boundaries of overlapping or nearly overlapping water environment functional zones and administrative regions upstream of the control points. Use ArcGIS's snap function to depict closed polygons in a new layer. Step 34: Repeat step 32 until all Level 1 prevention and control units have been divided; Step 35: Depict the Level 2 control units. The Level 2 control units are sub-units within the Level 1 control units. Based on the Level 1 control units, the Level 2 control units are divided, mainly by administrative boundaries, and secondarily by sub-basins and water environment functional zoning. The control points of the Level 2 control units are determined, and the boundaries of the Level 2 control units are depicted using ArcGIS's snap function. Step 36: Overlay the above thematic maps and control units again, and make detailed corrections to the control units in the order from tributaries to main streams and from upstream to downstream, to finally obtain the division scheme of the first and second level control units of the watershed. The thematic maps include: water ecological zoning map, sub-basin division map, administrative division map, water system map, hydrological basin scope map, water quality monitoring station distribution map, and surface water environment functional zoning map.

4. The method for dividing urban river control units according to claim 3, characterized in that: In step 35, special water environment functional areas and key pollution areas are given priority for control and are treated as separate prevention and control units. Special water environment functional areas include nature reserves and mining areas.

5. The method for dividing urban river control units according to claim 3, characterized in that: If the watershed scale is small and the number of sub-watersheds and administrative regions within the watershed is small in step 36, then there is no need to divide it into Level 1 prevention and control units. Instead, a thematic map of township administrative regions can be used as the unit for dividing the prevention and control.

6. The method for dividing urban river control units according to claim 1, characterized in that: Step 4 applies the K-means clustering method to classify the sampling points in the watershed, and couples them with the watershed water body unit grid based on environmental pressure response. By comprehensively analyzing the watershed's natural catchment patterns, ecosystem service functions, and ecological sensitivity, the watershed is finely divided to obtain corresponding water body control units. The specific process is as follows: Step 41: Basic data acquisition. Based on the actual situation of the watershed, and taking into account the historical data of the water quality monitoring sections, and adhering to the principle of data representativeness, on-site sampling is carried out at the monitoring sections. Specific indicators include pH, dissolved oxygen, nitrite nitrogen, nitrate nitrogen, ammonia nitrogen, reactive phosphate, total chromium, turbidity, total nitrogen, total phosphorus, chemical oxygen demand, temperature, salinity, water depth, flow velocity and flow rate. The sediment testing indicators include: total carbon, total nitrogen, and total phosphorus; a survey of the current status of aquatic organisms, including phytoplankton, zooplankton, benthic animals, and macroaquatic plants, was conducted in the watershed, with three sections (left, middle, and right) set up for each section; the aquatic organism testing indicators include: phytoplankton, zooplankton, macrobenthic animals, and macroaquatic plants; Step 42: Cluster analysis. Based on the basic data in Step 41, multiple features are analyzed to find the statistical measure of similarity indicators. Based on this, the statistical measure is used as the standard for classifying the types. Sample objects with high similarity are placed in the same class, and other sample objects with high similarity are placed in another class. Aggregation is performed between classes according to distance. Closely related objects are clustered in the same class, and those with far-reaching relationships are clustered in different categories. Among them, K-means clustering is a classic algorithm in cluster analysis. The main clustering idea is to cluster around K points in space and classify the objects closest to them. The algorithm of K-means cluster analysis is (1) Initialization: Randomly select the initial cluster center. (2) Cyclic: ① Aggregate the remaining records to the class center according to the principle of proximity, ② Calculate the class center of each initial classification, ③ Re-cluster using the calculated center positions, ④ Repeat this process until the class center positions converge.

7. The method for dividing urban river control units according to claim 6, characterized in that: The algorithm for K-means clustering analysis in step 42 is as follows: Assume the dataset is X = {x1, x2, ..., x...} i , ..., x n The number of clusters is K, and the cluster centers are C = {c1, c2, ..., c3}. i c k }, a. Randomly select K samples from dataset X as initial cluster centers; b. For each sample xi (i = 1, 2, ..., n) in the dataset, calculate its distance to the cluster vertical cj (j = 1, 2, ..., k). The distance calculation formula is: In Equation 2: m is the dimension of the sample; Based on the calculated distances from each sample to the cluster center, find the sample with the minimum distance and assign it to the corresponding cluster. The cluster centers are recalculated and updated according to Equation 3, and the objective function is calculated according to Equation 4. The calculation formulas are as follows: The algorithm judges the cluster centers and objective function. If the requirements are met, the algorithm ends; otherwise, it continues to step b.

Images