Urban blue-green space morphology optimization method based on multiple ecological benefit objectives

Abstract

Disclosed in the present invention is an urban blue-green space morphology optimization method based on multiple ecological benefit objectives. The method comprises the following steps: S1, setting decision variables for an urban blue-green space morphology optimization model; S2, setting initial parameters for the urban blue-green space morphology optimization model; S3, constructing multiple ecological benefit objective functions for urban blue-green spaces; S4, setting constraint conditions for the urban blue-green space morphology optimization model; S5, performing urban blue-green space morphology optimization on the basis of an NSGA-II algorithm to obtain Pareto frontier solutions corresponding to multiple ecological benefit objectives; and S6, using an entropy weight method to select an optimal urban blue-green space morphology scheme. The present invention can realize urban blue-green space morphology optimization on the basis of taking multiple ecological objectives into comprehensive consideration, thereby facilitating improvement to the quality of urban ecological environment.

Description

Urban blue-green space form optimization method based on multiple ecological benefit targets TECHNICAL FIELD

[0001] The present application belongs to the field of urban ecological environment management, and particularly relates to an urban blue-green space form optimization method based on multiple ecological benefit targets. BACKGROUND

[0002] Blue-green space is an important object of concern in urban development. The past urbanization process has fragmented the original ecological background, and blue-green space has been disturbed and invaded, with its ecological efficiency being severely weakened. In addition, there is a phenomenon of separation between "form" and "ecology" in the practice of urban blue-green space construction, which also limits the ecological benefits of urban blue-green space synergy. Blue-green space optimization has become an important research topic after China's cities have entered the "stock" development stage. Urban blue-green space has multiple ecological service functions, but the previous form optimization method only targets a single ecological benefit of urban blue-green space, and lacks consideration of comprehensive ecological benefits, which has a certain distance from urban construction practice application. Therefore, how to realize the optimization of urban blue-green space form on the basis of comprehensive adjustment of multiple ecological targets is a technical problem to be solved in the field.

[0003] SUMMARY

[0004] To solve the above problems, the present application discloses an urban blue-green space form optimization method based on multiple ecological benefit targets, which breaks through the limitations and deficiencies of previous researches which only optimize single ecological targets. By constructing an urban blue-green space form optimization model based on genetic algorithm, the urban blue-green space form optimization under the unified adjustment of multiple ecological benefit targets such as urban carbon increment, cooling, rain flood resilience and biodiversity is realized. The optimal urban blue-green space layout scheme can be quickly obtained by entropy weight method, the urban blue-green space form optimization is realized, and the collaborative development of urban blue-green space and the improvement of ecological environment quality are promoted.

[0005] To achieve the above purpose, the technical scheme of the present application is as follows:

[0006] An urban blue-green space form optimization method based on multiple ecological benefit targets, specifically comprising the following steps:

[0007] S1, setting the decision variables of the urban blue-green space form optimization model,

[0008] S2, setting the initial parameters of the urban blue-green space form optimization model,

[0009] S3, constructing the urban blue-green space multiple ecological benefit target function,

[0010] S4, setting the constraint conditions of the urban blue-green space form optimization model,

[0011] S5, based on the NSGA-II algorithm (Non-dominated Sorting Genetic Algorithm II, also known as Non-dominated Sorting Genetic Algorithm II) to optimize the urban blue-green space form, and obtain the Pareto frontier solution of multiple ecological benefit targets,

[0012] S6, the optimal urban blue-green space layout scheme is screened out by using the entropy weight method;

[0013] Preferably, the decision variables in S1 include land use type coding, area, perimeter, geographic location coordinates and land parcel conversion value, and the setting of the urban blue-green space form optimization model decision variables includes the following sub-steps:

[0014] Step 1, classifying land use of remote sensing image data;

[0015] Step 2, reclassifying land use classification data to extract urban blue-green space distribution;

[0016] Step 3, constructing a land use type coding mapping table;

[0017] Step 4, converting urban blue-green space area, boundary, type and other characteristic basic data into JSON (JavaScript Object Notation) data, and constructing decision variables together with land use type custom type coding;

[0018] Preferably, the initial parameters in S2 include population size, iteration number, crossover rate and mutation rate;

[0019] Preferably, the multiple ecological benefit targets in S3 include carbon sequestration target, cooling target, rainwater resilience target and biodiversity target; the method for constructing the carbon sequestration target function is: multiplying the land use type area and the carbon sequestration coefficient to obtain the carbon sequestration benefit, and the calculation formula of the carbon sequestration target is:

[0020] In the formula, K is the total number of land use types, C k is the carbon sequestration coefficient of the kth land use type, and A k is the total area of the kth land use type;

[0021] The method for constructing the cooling target function is: multiplying the cooling intensity coefficient of different land use types and the area of the corresponding land use type to obtain the cooling benefit, and the calculation formula of the cooling target is:

[0022] In the formula, K is the total number of land use types, I k is the cooling intensity coefficient of the kth land use type, and A kA k is the total area of the kth land use type;

[0023] The rainwater resilience target function is constructed, and the specific method is that the rainwater resilience benefit is calculated by using the product of the land use type area and the runoff control coefficient, and the calculation result also needs to meet the rainwater runoff rate specification standard of the urban built-up area, and the calculation formula of the rainwater resilience target is:

[0024] In the formula, K is the total number of land use types, W k is the rainwater elimination coefficient of the kth land use type, A k is the total area of the kth land use type;

[0025] The biodiversity target function is constructed, and the specific method is that the blue area and the green area are added and then divided by the total area, and the calculation formula of the biodiversity target is:

[0026] In the formula, A grass represents the total area of all green spaces; A water represents the total area of all blue spaces; A sum represents the total area of all category blocks; K is the total number of land use types, A k is the total area of the kth land use type.

[0027] Preferably, the constraint condition S4 comprises a space constraint, a region constraint and other constraints; the region constraint refers to an ecological land red line constraint, an immovable construction land constraint and a region constraint with ecological protection or cultural protection speciality; the space constraint refers to a land use type space transfer strategy; and the other constraints are a non-negative land use type patch area constraint and a land block type conversion constraint.

[0028] Preferably, the urban blue-green space form optimization based on the NSGA-Ⅱ algorithm S5 comprises the following sub-steps:

[0029] Step 1, input the decision variables and initial parameters obtained in S1 and S2;

[0030] Step 2, a set of randomly generated blue-green space layout schemes in Matlab (Matrix & Laboratory) software is calculated, and the ecological target benefit value corresponding to each blue-green space layout scheme is calculated;

[0031] Step 3, the blue-green space layout schemes calculated in the initialization are hierarchically sorted according to the fitness size on multiple target functions, and a Pareto front solution is obtained;

[0032] Step 4, a new scheme is generated by running the crossover and mutation operations. The crossover and mutation operations refer to generating a new blue-green space layout scheme by combining the information of any two blue-green space layout schemes randomly generated in step 2;

[0033] Step 5, the schemes obtained in steps 3 and 4 are combined;

[0034] Step 6, based on the scheme obtained in step 5, the fitness size is first calculated and a hierarchical sorting is performed to obtain a Pareto front solution, then an elite reservation strategy is used to screen the scheme, and finally a new scheme set is obtained;

[0035] Step 7, the operations of steps 4 to 6 are cycled until the maximum number of iterations is reached, and then the optimized scheme solution set is output, wherein the solution set is a Pareto front solution graph and decision variable data corresponding to each scheme.

[0036] Preferably, the optimal urban blue-green space layout scheme screened by the entropy weight method in S6 comprises the following sub-steps:

[0037] Step 1, the function values optimized based on the carbon increment target, the temperature reduction target, the rainwater target and the biodiversity target are normalized;

[0038] Step 2, for each target function, the information entropy in all solutions is calculated;

[0039] Step 3, based on the information entropy, the weight of each target is reversely calculated;

[0040] Step 4, the weight proportion of the four targets calculated by the entropy weight method is input into the optimization model, and the optimization model is re-run to obtain the decision variable data of the optimal scheme;

[0041] Step 5, based on the calculated decision variable data, the corresponding optimal urban blue-green space form scheme graph is generated by using the

JSON to feature

[0042] The present application has the following advantages:

[0043] The urban blue-green space form optimization method based on multiple ecological benefit targets disclosed by the present application breaks through the limitations and deficiencies of the previous research which only optimizes a single ecological target. By constructing an urban blue-green space form optimization model based on a genetic algorithm, the urban blue-green space form optimization under the unified adjustment of multiple ecological benefit targets such as urban carbon increment, temperature reduction, rainwater resilience and biodiversity is realized. The entropy weight method can quickly obtain the optimal urban blue-green space layout scheme, realize the optimization of urban blue-green space form, and promote the coordinated development of urban blue-green space and the improvement of ecological environment quality. BRIEF DESCRIPTION OF DRAWINGS

[0044] Fig. 1 is a structural flowchart of the present application;

[0045] Fig. 2 is a land use classification map in an embodiment of the present application;

[0046] Fig. 3 is a city blue-green space classification data map in an embodiment of the present application;

[0047] Fig. 4 is an ecological land red line constraint map in an embodiment of the present application;

[0048] Fig. 5 is a city construction land constraint map in an embodiment of the present application;

[0049] Fig. 6(a, b, c) is a Pareto relationship diagram between carbon increment, cooling, rainstorm resilience, and biodiversity in a city blue-green space form optimization model in an embodiment of the present application;

[0050] Fig. 7 is an optimal city blue-green space form scheme map in an embodiment of the present application. DETAILED DESCRIPTION

[0051] The present application will be further illustrated below in conjunction with the drawings and specific embodiments, and it should be understood that the following specific embodiments are only used to illustrate the present application and not to limit the scope of the present application.

[0052] Taking Nanjing City as an example, as shown in Fig. 1, the city blue-green space form optimization method based on multiple ecological benefit targets described in the present application has the following specific steps:

[0053] S1, setting the decision variable of the city blue-green space form optimization model,

[0054] wherein the decision variable includes land use type code, area, perimeter, geographic location coordinates, and land block convertible value;

[0055] Setting the decision variable of the blue-green space form optimization model includes the following sub-steps:

[0056] Step 1, land use classification of remote sensing image data; based on the ENVI (The Environment for Visualizing Images) software to establish a vector graphics tool, the natural administrative boundary of the study area is drawn by itself, and the administrative boundary of Nanjing central urban area is obtained by cutting. Based on the ENVI software, the resolution is improved to 15 m, and the land use classification of Nanjing central urban area is obtained as shown in Fig. 2.

[0057] Step 2, reclassification based on land use classification data to obtain a blue-green space distribution map; taking the land use classification data of Nanjing central urban area as the basis, the blue-green space distribution map of Nanjing central urban area is obtained by reclassification, as shown in Fig. 3.

[0058] Step 3: Construct a land use type coding mapping table, as shown in Table 1.

[0059] Table 1 Land Use Type Coding Mapping Table

[0060] Step 4: Convert the basic data on the area, boundaries, and types of blue-green spaces into JSON data, and construct decision variables together with the custom type code for land use types. Using the ArcGIS software platform, process the JSON data of land use types in the central urban area of ​​Nanjing, which will then serve as the input values ​​for the blue-green space morphology optimization model. A selection of data is shown in Table 2.

[0061] Table 2. Multi-attribute data of blue-green spaces in Nanjing's central urban area (partial)

[0062] S2 sets the initial parameters for the urban blue-green space morphology optimization model.

[0063] The initial parameters mentioned above include population size, number of iterations, crossover rate, and mutation rate;

[0064] In this embodiment, the parameters used in the application of the urban blue-green space morphology optimization model are shown in Table 3.

[0065] Table 3 Application parameters of the urban blue-green space morphology optimization model

[0066] S3, constructing a multi-ecological-benefit objective function for urban blue-green spaces.

[0067] The aforementioned multi-ecological benefit objectives include carbon sequestration targets, cooling targets, stormwater resilience targets, and biodiversity targets;

[0068] Specifically, in this embodiment, for the carbon sequestration target, the total carbon sequestration benefit is obtained by multiplying the land use type area and the carbon sequestration coefficient. The formula for maximizing the carbon sequestration target is as follows:

[0069] In the formula: C k C is the carbon sequestration coefficient for the k-th land use type; k Let be the total area of ​​the kth land use type.

[0070] The cooling effect is obtained by multiplying the cooling intensity coefficient of different land use types by the area of ​​the applied land type. The formula for maximizing the cooling target is as follows:

[0071] In the formula: C k Let A be the carbon sequestration coefficient for the k-th land use type; kTotal area of the kth land use type.

[0072] The total benefit of rainwater resilience is calculated by the product of the area of land use type and the runoff control coefficient, and the rainwater runoff control rate of urban built-up area is required to be not less than 75%, and the rainwater resilience target maximization formula is:

[0073] f SW ∈[0.75,1)

[0074] In the formula, K is the total number of land use types; W k is the rainwater resilience coefficient of the kth land use type, A k is the total area of the kth land use type.

[0075] In order to maximize the biodiversity target, the specific method is to add the blue area and the green area and then divide by the total area, and the biodiversity target maximization function formula is:

[0076] In the formula, A grass represents the total area of all green spaces, A water represents the total area of all blue spaces, and A sum represents the total area of all category blocks.

[0077] S4, establishing a constraint condition of the urban blue-green space form optimization model,

[0078] The constraint condition includes spatial constraints, regional constraints and other constraints;

[0079] Specifically, in the embodiment, the regional constraints include total area constraints, vegetation cover constraints, landscape diversity constraints, and urban water surface rate constraints.

[0080] The constraint formula of the total area is as follows:

[0081] In the formula, k is the count of land use types, A k is the total area of the kth land use type.

[0082] In order to ensure that the vegetation coverage in the optimization scenario is equal to or greater than the vegetation coverage of the current 2020 Nanjing blue-green space, the vegetation coverage coefficients of forest land and grassland are set to 1 and 0.49 respectively, and the vegetation cover constraint formula is as follows:

[0083] In the formula, A k is the area of the kth land use type in the optimization scenario, and S kis the area of the kth land use type in the study area under the non-optimized scenario, k=2 and k=3 represent the two land use types of forest land and grassland respectively.

[0084] In order to better realize the diversity constraint, the landscape diversity needs to be given a conditional constraint, that is, the comprehensive percentage of the areas of forest land, grassland, water body and wetland in the optimized scenario must be greater than or equal to the percentage before optimization. The constraint formula is as follows:

[0085] In the formula, A k is the area of the kth land use type in the optimized scenario, S k is the area of the kth land use type in the study area under the non-optimized scenario, k=2, 3, 4, 5 represent the four land use types of forest land, grassland, water body and wetland respectively.

[0086] The requirement of the urban water surface rate constraint involves that when urban planning and land use optimization are carried out, the proportion of the area of urban water body in the total area must not be lower than the level before adjustment after adjustment. It is converted into a mathematical condition, that is, the percentage of the area of water body in the optimized scenario must be greater than or equal to the percentage before optimization. The constraint formula of the urban water surface rate is as follows:

[0087] In the formula, A k is the area of the kth land use type in the optimized scenario, S k is the area of the kth land use type in the study area under the non-optimized scenario, k=4, 5 represent the two land use types of water body and wetland.

[0088] In the embodiment, the space constraint includes the ecological land red line constraint and the urban construction land constraint.

[0089] The ecological land red line constraint is shown in FIG. 4. The urban blue-green space within the red line range is protected, and only the area outside the red line range is optimized and controlled. Xuanwu Lake, Zijin Mountain, Mufu Mountain, Niushou Mountain and the Xixia residual massif in the northeast region are the main mountain protection areas; the wetland along the Yangtze River and the Qinhuai branch are the main water body protection areas; in addition, the Fuzimiao Qinhuai protection area is also an area that cannot be mobilized. The urban construction land constraint is shown in FIG. 5.

[0090] In the embodiment, the other constraint is the non-negative constraint of the area of the land use type patch and the land type conversion constraint; different land use types are the constituent units of the urban blue-green space, and the area of the patch cannot be negative, so the area of different land use types is always greater than 0. The expression formula of the non-negative constraint is as follows:

[0091] A k > 0

[0092] In the formula: k indicates the kth land use type; A k indicates the area of the kth land use type.

[0093] The construction land is the main non-blue-green space in the city, and it is not changed because of the little possibility of change. There is a small amount of cultivated land in the central urban area of Nanjing, which is mainly distributed on the edge of the central urban area. The cultivated land is a production base for providing food and is protected by policy. Therefore, the cultivated land in the study area is not transformed. The water body in the city is affected by natural factors such as topography and climate, and the evolution has a long spatiotemporal span, so the transformation between the water body and other land use types is limited. The feasibility of mutual conversion between different land use types is shown in Table 4.

[0094] Table 4: Conversion limitation table of land use types in the central urban area of Nanjing

[0095] Note: X1 cultivated land, X2 forest land, X3 grassland, X4 wetland, X5 water body, X6 construction land, X7 unused land

[0096] S5, based on the NSGA-Ⅱ algorithm, the morphology of the blue-green space is optimized to obtain the Pareto frontier solution of the four objectives;

[0097] The morphology of the urban blue-green space based on the NSGA-Ⅱ algorithm includes the following sub-steps:

[0098] Step 1, input the decision variables and initial parameters obtained in S1 and S2;

[0099] Step 2, a set of randomly generated blue-green space layout schemes in Matlab software is generated, and the ecological target benefit value corresponding to each blue-green space layout scheme is calculated;

[0100] Step 3, according to the fitness size of the initialized blue-green space layout scheme on multiple objective functions, hierarchical sorting is performed to obtain the Pareto frontier solution;

[0101] Step 4, run the crossover and mutation operation to generate a new scheme. The crossover and mutation operation refers to generating a new blue-green space layout scheme by combining the information of any two blue-green space layout schemes randomly generated in step 2;

[0102] Step 5, combine the schemes obtained by steps 3 and 4;

[0103] Step 6, based on the schemes obtained in step 5, first calculate the fitness size and perform hierarchical sorting to obtain the Pareto frontier solution, then use the elite retention strategy to screen the schemes, and finally obtain a new scheme set;

[0104] Step 7, repeat steps 4 to 6 until the maximum number of iterations is reached, and then output the optimized solution set, which is a Pareto frontier solution graph and the decision variable data corresponding to each solution.

[0105] In this embodiment, the blue-green space form optimization model of the central urban area of Nanjing is solved, and the solving is iterated 200 times, and the completion time is about 1 hour. The calculation result of the blue-green space form optimization model of the central urban area of Nanjing based on NSGA-Ⅱ is a series of Pareto frontier solutions, and each point represents a Pareto solution, that is, a blue-green space layout scheme. Four ecological benefit objectives are optimized, and the four objectives are arranged and combined, and three Pareto graphs are generated, as shown in FIG. 6, which respectively show the Pareto relationship among carbon increment, temperature reduction, rainwater resilience and biodiversity. FIG. 6-a shows the relationship among carbon increment, temperature reduction and rainwater resilience, FIG. 6-b shows the relationship among temperature reduction, rainwater resilience and biodiversity, and FIG. 6-c shows the relationship among biodiversity, carbon increment and rainwater resilience.

[0106] S6, screening the optimal scheme of the blue-green space layout scheme by using the entropy weight method;

[0107] Further, the screening of the optimal scheme of the urban blue-green space layout scheme by using the entropy weight method comprises the following sub-steps:

[0108] Step 1, normalizing the function values optimized based on the carbon increment target, the temperature reduction target, the rainwater target and the biodiversity target;

[0109] Step 2, calculating the information entropy of each target function in all solutions;

[0110] Step 3, calculating the weight of each target based on the information entropy. In this embodiment, the weights of the four targets are obtained as follows: the carbon increment target accounts for 43%, the temperature reduction target accounts for 20%, the rainwater resilience target accounts for 25%, and the biodiversity target accounts for 12%, and the weights of the four targets account for 100%;

[0111] Step 4, inputting the weight proportions of the four targets calculated based on the entropy weight method into the optimization model, and re-running the optimization model to obtain the decision variable data of the optimal scheme;

[0112] Step 5, based on the calculated decision variable data, using the

JSON to feature

[0113] It should be noted that the above content only illustrates the technical idea of the present application, and cannot limit the protection scope of the present application. For ordinary skilled in the art, without departing from the principle of the present application, a number of improvements and refinements can be made, which fall within the protection scope of the claims of the present application.

Claims

1. A method for optimizing urban blue-green space form based on multiple ecological benefit targets, characterized in that: The method comprises the following steps: S1, setting the decision variables of the urban blue-green space form optimization model; S2, setting the initial parameters of the urban blue-green space form optimization model; S3, constructing the multi-ecological benefit objective function of the urban blue-green space; S4, setting the constraint conditions of the urban blue-green space form optimization model; S5, optimizing the urban blue-green space form based on the NSGA-II algorithm to obtain the Pareto frontier solution of the multi-ecological benefit objective; S6, screening the optimal urban blue-green space form scheme by using the entropy weight method.

2. The method according to claim 1, wherein, The decision variables in S1 include land use type coding, area, perimeter, geographic location coordinates and land parcel conversion value, and the setting of the decision variables of the urban blue-green space form optimization model comprises the following sub-steps: Step 1, classifying the land use of remote sensing image data; Step 2, reclassifying the land use classification data to extract the urban blue-green space distribution; Step 3, constructing a land use type coding mapping table; Step 4, converting the characteristics of the urban blue-green space area, boundary and type into JSON data, and constructing the decision variables together with the self-defined type coding of the land use type.

3. The method of claim 1, wherein the method is characterized by: The initial parameters in S2 include population size, iteration number, crossover rate and mutation rate.

4. The method according to claim 1, wherein, The multi-ecological benefit target includes a carbon increment target, a temperature reduction target, a rain flood resilience target, and a biodiversity target. The carbon increment target function is constructed in the following manner: multiplying the land use type area and the carbon increment coefficient to obtain the carbon increment benefit, and the calculation formula is: wherein: K is the total number of land use types, C k is the carbon sequestration coefficient of the kth land use type; A k is the total area of the kth land use type; The construction cooling target function is built, and the specific method is as follows: the cooling intensity coefficient of different land types is multiplied by the area of the corresponding land type to obtain the cooling benefit, and the calculation formula is as follows: wherein: K is the total number of land use types, I k is the cooling intensity coefficient of the kth land use type, A k is the total area of the kth land use type; The rain flood resilience target function is constructed, and the specific method is as follows: the rain flood resilience benefit is calculated by using the product of the land use type area and the runoff control coefficient, and the calculation result also needs to meet the rainwater runoff rate specification standard of the urban built area, and the calculation formula is as follows: wherein: K is the total number of land use types, W k is the stormwater runoff coefficient for the kth land use type, A k is the total area of the kth land use type; The construction of the biodiversity objective function is specifically as follows: the blue area and the green area are added and then divided by the total area, and the calculation formula of the biodiversity objective is: where: A grass represents the total area of all green spaces; A water represents the total area of all blue spaces; A sum represents the total area of all category blocks; k is the total number of land use types, A k is the total area of the kth land use type.

5. The method of claim 1, wherein the method is characterized by: The constraint conditions in S4 include spatial constraints, regional constraints and other constraints; the regional constraints refer to ecological land red line constraints, non-variable construction land constraints and regional constraints with special ecological protection or cultural protection; the spatial constraints refer to land type space transfer strategies; and the other constraints are non-negative patch area constraints and land type conversion constraints.

6. The method of claim 1, wherein the method is characterized by: The urban blue-green space form optimization based on the NSGA-II algorithm in S5 comprises the following sub-steps: Step 1, inputting the decision variables and initial parameters obtained in S1 and S2; Step 2, randomly generating a set of blue-green space layout schemes in Matlab software, and calculating the ecological objective benefit value corresponding to each blue-green space layout scheme; Step 3, hierarchically sorting the initialization blue-green space layout schemes according to the fitness size on multiple objective functions to obtain the Pareto frontier solution; Step 4, generating a new scheme by running the crossover and mutation operation; the crossover and mutation operation refers to generating a new blue-green space layout scheme by combining the information of any two blue-green space layout schemes randomly generated in step 2; Step 5, merging the schemes obtained in steps 3 and 4; Step 6, based on the schemes obtained in step 5, first calculating the fitness size and hierarchically sorting to obtain the Pareto frontier solution, then screening the schemes by using the elite retention strategy, and finally obtaining a new scheme set; Step 7, performing the operation from step 4 to step 6 in a loop until the maximum iteration number is reached, and then outputting the optimization scheme solution set; the solution set is the Pareto frontier solution graph and the decision variable data corresponding to each scheme.

7. The method according to claim 1, wherein, The screening of the optimal urban blue-green space form scheme by using the entropy weight method in S6 comprises the following sub-steps: Step 1, normalizing the function values optimized based on the carbon increment target, temperature reduction target, rainwater target and biodiversity target; Step 2, calculating the entropy weight of each function value; Step 2, for each objective function, calculate its information entropy in all solutions; Step 3, based on the information entropy, calculate the weight of each objective in reverse; Step 4, input the weight proportion of the four objectives calculated based on the entropy weight method into the optimization model, re-run the optimization model, and get the decision variable data of the optimal scheme; Step 5, based on the calculated decision variable data, use the 【JSON to feature】 tool in ArcGIS software to generate the corresponding optimal urban blue-green space form scheme map.

Images