A method for optimizing the spatial layout of LID facilities by coupling the SWMM model with the GDE3 algorithm

Abstract

The invention discloses a LID facility space optimization layout method coupling SWMM model and GDE3 algorithm, the method steps include: constructing SWMM model to simulate and evaluate hydrological effects of different LID facility space layouts, and obtaining the runoff reduction rate of the study area; constructing a cost calculator to obtain the total construction cost under different LID facility space layouts; establishing a multi-objective decision model for LID facility space layout, determining the decision variables of LID layout, taking the minimum total construction cost of LID facilities and the maximum runoff reduction rate of the study area as the objective function, and determining the boundary conditions; GDE3 algorithm generates decision variables through crossover mutation, inputs them into SWMM model and cost calculator, obtains the runoff reduction rate and total construction cost of the study area, and uses non-dominated sorting and congestion distance calculation to obtain the optimal solution set of decision variables in the multi-objective decision model for LID facility space layout. The invention can improve the search capability and obtain a higher runoff reduction rate.

Description

Technical Field

[0001] The present invention relates to the technical field of LID facility space optimization, and in particular to a LID facility space optimization layout method coupling a SWMM model and a GDE3 algorithm. Background Art

[0002] LID (Low Impact Development) facilities are an important part of sponge city construction, and there are many ways to implement them, including bioretention ponds, green roofs, rain barrels, and permeable pavement. At present, the layout method of LID facilities mainly relies on the simulation of hydrological models in practice. In the early stage, it mainly focused on the simulation and evaluation of the transformation effect of a single optimization goal (such as runoff control), ignoring the mutual constraints between construction costs and runoff reduction benefits; secondly, the spatial optimization layout of LID measures is a comprehensive project, and its transformation effect should be considered from multiple aspects to meet the multi-dimensional engineering objectives in reality. Therefore, multi-objective optimization algorithms, especially non-dominated sorting genetic algorithm (NSGA-II), are widely used in the optimization of LID measures spatial layout. However, due to its low crossover mutation efficiency and slow convergence speed, it is unable to generate a solution set with better distribution for some complex problems. Summary of the invention

[0003] In order to overcome the defects and shortcomings of the existing technology, the existing sponge city planning method that relies on manual spatial layout of sponge city LID facilities has the problems of low modeling efficiency and difficulty in exhausting potential optimization solutions. The present invention provides a LID facility spatial optimization layout method coupling SWMM model and GDE3 algorithm to achieve "low cost and high benefit" hydrological control effect, and provide technical support for the scientific planning and design of future sponge cities.

[0004] A second object of the present invention is to provide a LID facility space optimization layout system that couples the SWMM model and the GDE3 algorithm.

[0005] A third object of the present invention is to provide a computer-readable storage medium.

[0006] A fourth object of the present invention is to provide a computing device.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] The present invention provides a LID facility space optimization layout method coupling a SWMM model and a GDE3 algorithm, comprising the following steps:

[0009] A SWMM model was constructed to simulate and evaluate the hydrological effects of different LID facility spatial layouts to obtain the runoff reduction rate of the study area;

[0010] Build a cost calculator to obtain the total construction cost for different LID facility space layouts;

[0011] A multi-objective decision-making model for the spatial layout of LID facilities was established to determine the decision variables for LID layout, and the boundary conditions were determined with the minimum total construction cost of LID facilities and the maximum runoff reduction rate in the study area as the objective function.

[0012] Coupling SWMM model and GDE3 algorithm: The GDE3 algorithm generates decision variables in the multi-objective decision-making model through crossover mutation, and inputs them into the SWMM model and cost calculator to obtain the runoff reduction rate and total construction cost of the study area. The non-dominated sorting and congestion distance calculations are used to obtain the optimal solution set of decision variables in the multi-objective decision-making model of the spatial layout of LID facilities.

[0013] As a preferred technical solution, the SWMM model is constructed by obtaining basic data of the study area in advance, wherein the basic data include digital elevation model, satellite remote sensing image data, pipe network distribution and land use type data, dividing sub-catchment areas based on the basic data, determining sub-catchment areas and pipe network parameters, and determining the type and proportion of LID suitable for construction in the study area.

[0014] As a preferred technical solution, the construction cost calculator predetermines the total construction cost and maintenance cost of various LID facilities and calculates the annual discounted cost of various LID facilities, which is specifically expressed as:

[0015]

[0016] Among them, A LID and P LID Respectively represent the annual construction cost and the total construction cost, i represents the annual interest rate; n LID Indicates the life cycle of each LID facility.

[0017] As a preferred technical solution, the multi-objective decision model for the spatial layout of LID facilities, the decision variables select the construction ratio or area of ​​LID facilities in each sub-catchment area;

[0018] The objective function is to minimize the total construction cost of LID facilities and maximize the runoff reduction rate in the study area, which is specifically expressed as:

[0019] Maximize the runoff reduction rate in the study area:

[0020]

[0021] Among them, Runoff baseline is the outlet flow of the study area without LID facilities, Runoff LIDis the outlet flow of the study area under different LID layouts generated by the multi-objective optimization algorithm;

[0022] Minimum goal for total construction cost of LID facilities:

[0023]

[0024] Among them, A ij represents the area occupied by LID facility type j suitable for construction in subcatchment i, C j represents the total annual cost per unit LID facility type j, m represents the total number of subcatchments, and n represents the total number of LID facility types.

[0025] As a preferred technical solution, the boundary condition is expressed as:

[0026]

[0027] Wherein, subscript l and superscript n represent LID type and subcatchment name, respectively; It represents the maximum suitable buildable area of ​​LID type l in subcatchment n.

[0028] As a preferred technical solution, the specific steps of coupling the SWMM model and the GDE3 algorithm include:

[0029] Randomly initialize the population in the feasible solution space of decision variables;

[0030] Perform mutation and crossover operations on the parent individuals to generate another new population of offspring;

[0031] The new population individuals correspond to the area or proportion of each suitable LID type in each newly generated sub-catchment, and the overland flow width and imperviousness of each sub-catchment are recalculated;

[0032] The new population individuals and the changed overland flow width and imperviousness parameters were sequentially input into the SWMM model for hydrological simulation, the outflow of the study area outlet was obtained, and the runoff reduction rate was calculated. At the same time, the individuals were input into the cost calculator for calculation to obtain the converted annual total cost. The parent and child populations were quickly non-dominated sorted based on the objective function. The population size was kept at the initial size based on the crowding distance method, and the iteration was repeated to the set number of iterations.

[0033] As a preferred technical solution, the above-mentioned steps of performing mutation and crossover operations on the parent individuals to generate another new population of offspring include:

[0034] For each individual X in generation j i (j) Implement the mutation operation to obtain the corresponding mutant individual X m (j+1); for individual X r1(j) and the generated variant individuals X m (j+1) performs crossover operation to generate the test individual U(j+1):

[0035] The mutation operation is expressed as:

[0036] X m (j+1)=X r1 (j)+F*[X r2 (j)-X r3 (j)]

[0037] Where j represents the evolutionary generation; individuals r1, r2, r3∈{1,2,3...,NP} are different from each other and from individual m; X r1 (j) is the parent basis vector, [X r2 (j)-X r3 (j)] is the parent difference variable, and F is the scaling factor;

[0038] The crossover operation is expressed as:

[0039]

[0040] Among them, rand(0,1) is a uniformly distributed random number between [0, 1], CR is the crossover probability between [0, 1], j rand is a random quantity between {1,2,3...,NP}.

[0041] In order to achieve the above second purpose, the present invention adopts the following technical solutions:

[0042] A LID facility space optimization layout system coupling a SWMM model and a GDE3 algorithm, comprising: a SWMM model building module, a cost calculator building module, a multi-objective decision model building module, and a SWMM model and GDE3 algorithm coupling module;

[0043] The SWMM model building module is used to build a SWMM model, simulate and evaluate the hydrological effects of different LID facility spatial layouts, and obtain the runoff reduction rate of the study area;

[0044] The cost calculator building module is used to build a cost calculator to obtain the total construction cost under different LID facility space layouts;

[0045] The multi-objective decision model building module is used to establish a multi-objective decision model for the spatial layout of LID facilities, determine the decision variables for LID layout, and determine the boundary conditions with the minimum total construction cost of LID facilities and the maximum runoff reduction rate in the study area as the objective function;

[0046] The SWMM model and GDE3 algorithm coupling module is used to couple the SWMM model and the GDE3 algorithm: the GDE3 algorithm generates decision variables in the multi-objective decision-making model through crossover mutation, inputs them into the SWMM model and the cost calculator, obtains the runoff reduction rate and the total construction cost of the study area, and uses non-dominated sorting and congestion distance calculation to obtain the optimal solution set of decision variables in the multi-objective decision-making model of the spatial layout of LID facilities.

[0047] In order to achieve the third objective, the present invention adopts the following technical solutions:

[0048] A computer-readable storage medium stores a program, which, when executed by a processor, implements the LID facility spatial optimization layout method of coupling the SWMM model and the GDE3 algorithm as described above.

[0049] In order to achieve the fourth objective, the present invention adopts the following technical solutions:

[0050] A computing device includes a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the LID facility space optimization layout method coupling the SWMM model and the GDE3 algorithm as described above is implemented.

[0051] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0052] The present invention adopts a technical solution of coupling SWMM model and third-generation differential evolution algorithm (GDE3) to optimize the spatial layout of LID facilities, which solves the problems of resource waste and inefficiency caused by blind planning and design and single optimization target in LID facility construction, and achieves the technical effect of improving search capability by differential mutation crossover operation and obtaining higher runoff reduction rate at a lower construction cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of the LID facility space optimization layout method coupling the SWMM model and the GDE3 algorithm of the present invention;

[0054] Figure 2 The digital elevation model and pipe network distribution map of Changqian area in Tianhe District, Guangzhou City adopted by the present invention;

[0055] Figure 3 A schematic diagram of the steps of fast non-dominated sorting of parent and offspring populations of the present invention;

[0056] Figure 4 The figure is a schematic diagram comparing the running results of this embodiment with the running results of other optimization algorithms (NSGAII, SPEA2 and MOSFLA). DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] Example 1

[0059] like Figure 1 As shown, this embodiment provides a LID facility space optimization layout method coupling the SWMM model and the GDE3 algorithm, including the following steps:

[0060] S1. Construct a SWMM hydrological model to simulate and evaluate the hydrological effects of different LID facility spatial layouts to obtain the runoff reduction rate of the study area;

[0061] In this example, the Changqian area in the northwest of Tianhe District, Guangzhou City is taken as the research area, with a total area of ​​about 1.57 km 2 The underlying surface of this area is relatively complex, with mountainous areas in the north and highly urbanized areas with dense buildings in the south. Figure 2 As shown in step S1, the construction of the SWMM model requires the acquisition of basic data of the study area in advance. The basic data include digital elevation model, satellite remote sensing image data, pipe network distribution, and land use type data. Based on this, sub-catchment areas are divided, sub-catchment area and pipe network parameters are determined, and the type and proportion of LID suitable for construction in the study area are determined.

[0062] Based on the basic data provided by relevant departments, 198 pipelines, 198 wells and 2 outlets were extracted. According to DEM data, buildings, traffic road distribution and pipeline network direction, the study area was artificially divided into 86 sub-catchments, with the smallest sub-catchment area of ​​1110.95m 2 The maximum subcatchment area is 54130.5m 2 The generalized results of the study area can be found in Figure 2 . Refer to the SWMM manual, the values ​​of subcatchment empirical parameters and pipe network parameters are specifically shown in Table 1. In addition, the land use types in the study area are mainly building land and road land, and the LID types suitable for construction are green roofs, rain barrels and permeable pavement.

[0063] In this embodiment, the sub-catchment area and pipe network parameters are shown in Table 1 below:

[0064] Table 1 Subcatchment area and pipe network parameters

[0065] Subcatchment parameters Numeric Manning coefficient of impervious surface 0.011 Manning coefficient of permeable surface 0.24 Impervious surface depression storage 0.5 Permeable surface depression storage capacity 1 Maximum penetration rate 10mm / h Minimum penetration rate 1.25mm / h Attenuation coefficient 4 Pipe Manning coefficient N 0.01-0.013

[0066] S2. construct a cost calculator to obtain the total construction cost under different LID facility spatial layouts;

[0067] In step S2, the construction of the cost calculator requires predetermining the total construction cost and maintenance cost of various LID facilities, and calculating the annual discounted cost of various LID facilities according to formula (1).

[0068]

[0069] Among them, A LID and P LID Respectively represent annual construction cost and total construction cost; i represents annual interest rate (%); n LID Indicates the life cycle of each LID facility.

[0070] In this embodiment, the total construction cost and maintenance cost of the LID facility are shown in Table 2 below:

[0071] Table 2 Total construction and maintenance costs of LID facilities

[0072]

[0073] The unit of green roof and permeable pavement is RMB / m 2 ,The cost unit of rain barrel is yuan / piece.

[0074] S3. Establish a multi-objective decision-making model for the spatial layout of LID facilities, determine the decision variables for LID layout, take the minimum total construction cost of LID facilities and the maximum runoff reduction rate in the study area as the objective function, and determine the boundary conditions;

[0075] In step S3, the multi-objective decision-making model for the spatial layout of LID facilities includes: (i) the decision variable is the construction ratio / area of ​​the selected LID facilities in each sub-catchment area; (ii) the objective function includes two, the first is the runoff control objective - maximizing the runoff reduction rate in the study area (see formula (2)), and the other is the cost minimization objective (see formula (3)); (iii) the boundary conditions are shown in formula (4).

[0076]

[0077] Among them, Runoff baseline is the outlet flow of the study area under the basic scenario (i.e., without the deployment of LID facilities), Runoff LID is the outlet flow of the study area under different LID layouts generated by the multi-objective optimization algorithm; the other is the cost minimization objective:

[0078]

[0079] Among them, Aij represents the area occupied by LID facility type j suitable for construction in subcatchment i, C j represents the total annual cost per unit LID facility type j, m represents the total number of subcatchments, and n represents the total number of LID facility types.

[0080]

[0081] The subscript l and superscript n represent the LID type and subcatchment name, respectively; represents the maximum suitable construction area of ​​LID type l on subcatchment n, and the lower boundary condition of the decision variable is usually set to 0.

[0082] S4. Coupling SWMM model and GDE3 algorithm: The GDE3 algorithm generates decision variables in the multi-objective decision-making model through crossover mutation, and inputs them into the SWMM model and cost calculator to obtain the runoff reduction rate and total construction cost of the study area. Then, the non-dominated sorting and congestion distance calculation are used to obtain the optimal solution set of decision variables in the multi-objective decision-making model of the spatial layout of LID facilities.

[0083] In step S4, the specific process of coupling the SWMM model and the GDE3 algorithm is as follows:

[0084] S41. Randomly initialize the population in the feasible solution space of decision variables The population size is NP;

[0085] S42. Perform mutation and crossover operations on the parent individuals to generate another new population of offspring. Specifically, the following is performed: First, for each individual X in generation j, i (j) Implement the mutation operation to obtain the corresponding mutant individual X m (j+1); then, use formula (6) to calculate X r1 (j) and the variant individual X generated by formula (5) m (j+1) performs crossover operation to generate the test individual U(j+1):

[0086] X m (j+1)=X r1 (j)+F*[X r2 (j)-X r3 (j)] (5)

[0087] Where j represents the evolutionary generation; individuals r1, r2, r3∈{1,2,3...,NP} are different from each other and from individual m; X r1 (j) is the parent basis vector, [X r2 (j)-X r3 (j)] is the parent difference variable, and F is the scaling factor;

[0088]

[0089] Among them, rand(0,1) is a uniformly distributed random number between [0, 1]; CR is the crossover probability between [0, 1]; j rand is a random quantity between {1,2,3...,NP} to avoid the individuals after the crossover operation being consistent with the parent individuals.

[0090] S43. New population individuals (i.e., the area / proportion of each suitable LID type in each newly generated sub-catchment) will affect the overland flow width and imperviousness parameters of the sub-catchment in the SWMM model. It is necessary to recalculate the overland flow width and imperviousness of each sub-catchment according to formulas (7) and (8).

[0091]

[0092]

[0093] S44, input the new population individuals and the changed overland flow width and imperviousness parameters into the SWMM model for hydrological simulation, obtain the outflow of the study area outlet, and calculate the runoff reduction rate by formula (2), which is equivalent to the first objective function f1; at the same time, input the individuals into the cost calculator for calculation to obtain the converted annual total cost, which is equivalent to the second objective function f2, as shown in Figure 3 As shown in the figure, based on the first objective function (maximizing the runoff reduction rate in the study area) and the second objective function (minimizing the cost), the parent and child populations (the number is 2NP) are quickly non-dominated sorted, and then the method based on crowding distance is used to make the population size still NP.

[0094] Assume that the population after the parent and offspring are combined is P, and the individual is p i ,i∈{1,2,3,...,2NP}, each individual p i There are two objective functions f1(p) and f2(p). For any two individuals p1 and p2, when f1(p1)>f1(p2) and f2(p1)<f2(p2), p1 dominates p2; when f1(p1)>f1(p2) and f2(p1)>f2(p2) or f1(p1)<f1(p2) and f2(p1)<f2(p2), p1 and p2 do not dominate each other. Calculate the value of each individual p in P in this way. i The two parameters n p and S p , where n p is the number of individuals that dominate individual p in the population, S p is the set of individuals dominated by individual p in the population. Fast non-dominated sorting is a cyclic objective function value classification process, and the main steps are as follows:

[0095] ① Find all n in the population p =0 individuals and save them in the current set F1;

[0096] ②For each individual i in the current set F1, the set of individuals it controls is S i , traverse S i For each individual l in n l = 0, then save individual l in the set H

[0097] ③Remember the individuals obtained in F1 as the individuals in the first non-dominated layer, and take H as the current set, repeat the above operation until the entire population is classified

[0098] Sort the individuals in the entire layer from low to high according to the non-dominated sorting level, and put them into the preferred new parent population C in turn until a certain layer F j The population C size exceeds the population size limit NP; finally, according to F j The individual crowding distances in continue to fill the population C in descending order until the population size reaches NP. The crowding distance is calculated as follows:

[0099] ① The crowding distance of boundary individuals is assigned to infinity to ensure that they are retained during the evolution process and maintain the diversity of the target space population;

[0100] ② The intermediate individual is calculated using formula (9)

[0101]

[0102] Among them, f1 max and f1 min They represent the maximum and minimum values ​​of the objective function 1 of the non-dominated layer, respectively. and They represent the maximum and minimum values ​​of the non-dominated layer objective function 2, f1 i+1 and f1 i-1 They represent the values ​​of objective function 1 of the two individuals i+1 and i-1 adjacent to i on this layer, and Respectively represent the values ​​of objective function 2 of the two individuals i+1 and i-1 adjacent to i on this layer,

[0103] S45. Repeat steps S42 and S43 until the set number of iterations is reached.

[0104] In order to highlight the performance of GDE3 algorithm in LID space optimization layout, such as Figure 4As shown, this embodiment selects three commonly used multi-objective optimization algorithms NSGAII, SPEA2 and MOSFLA to couple with the SWMM model to optimize the layout of LID in space, and obtains the operation comparison results;

[0105] As shown in Table 3 below, the running time and annual total cost of the GDE3 algorithm in the LID space optimization layout are obtained:

[0106] Table 3 Running time and annual total cost of GDE3 algorithm in LID spatial optimization layout

[0107]

[0108]

[0109] In terms of running time, it can be seen from Table 3 that the running speed of the GDE3 algorithm ranks third, but the time difference between it and the fastest NSGAII algorithm is only 172s, which is almost negligible compared with the total time of 3675s, indicating that the running speed of the GDE3 algorithm is faster. Figure 4 It can be seen that the solution set generated by the GDE3 algorithm has the widest distribution range, indicating that the most LID optimization layout schemes are available for decision makers to choose; in addition, it can be seen from Table 3 that the annual total cost of the LID layout scheme generated by the GDE3 algorithm under the unit reduction rate is the smallest, which indicates that to achieve the same runoff reduction rate, the annual total cost of the LID spatial layout scheme generated by the GDE3 algorithm is the smallest. Taking the above factors into consideration, the present invention can efficiently and scientifically and rationally realize the spatial optimization layout of low-impact development facilities in sponge cities, and provide a reference for the planning, design and construction of future sponge cities.

[0110] Example 2

[0111] A LID facility space optimization layout system coupling a SWMM model and a GDE3 algorithm, comprising: a SWMM model building module, a cost calculator building module, a multi-objective decision model building module, and a SWMM model and GDE3 algorithm coupling module;

[0112] As a preferred technical solution, the SWMM model building module is used to build a SWMM model to simulate and evaluate the hydrological effects of different LID facility spatial layouts and obtain the runoff reduction rate of the study area;

[0113] As a preferred technical solution, the cost calculator building module is used to build a cost calculator to obtain the total construction cost under different LID facility space layouts;

[0114] As a preferred technical solution, the multi-objective decision model construction module is used to establish a multi-objective decision model for the spatial layout of LID facilities, determine the decision variables for LID layout, and determine the boundary conditions with the minimum total construction cost of LID facilities and the maximum runoff reduction rate in the study area as the objective function;

[0115] As a preferred technical solution, the SWMM model and GDE3 algorithm coupling module is used to couple the SWMM model and the GDE3 algorithm: the GDE3 algorithm generates decision variables in the multi-objective decision-making model through crossover mutation, and inputs them into the SWMM model and cost calculator to obtain the runoff reduction rate and total construction cost of the study area, and uses non-dominated sorting and congestion distance calculation to obtain the optimal solution set of decision variables in the multi-objective decision-making model of the spatial layout of LID facilities.

[0116] Example 3

[0117] A computer-readable storage medium stores a program. The storage medium may be a storage medium such as a ROM, RAM, a disk, or an optical disk. The storage medium stores one or more programs. When the program is executed by a processor, the LID facility space optimization layout method of coupling a SWMM model and a GDE3 algorithm of Example 1 is implemented.

[0118] Example 4

[0119] A computing device, which may be a desktop computer, a laptop computer, a smart phone, a PDA handheld terminal, a tablet computer or other terminal device with a display function, comprises a processor and a memory, wherein the memory stores one or more programs, and when the processor executes the programs stored in the memory, the LID facility space optimization layout method of coupling the SWMM model and the GDE3 algorithm of Example 1 is implemented.

[0120] The techniques described in this disclosure may be implemented by various means. For example, these techniques may be implemented in hardware, firmware, software, or a combination thereof. For a hardware implementation, the processing module may be implemented in one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, electronic devices, other electronic units designed to perform the functions described in this disclosure, or a combination thereof.

[0121] For firmware and / or software implementations, the techniques may be implemented with modules (e.g., processes, steps, flows, etc.) that perform the functions described herein. The firmware and / or software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.

[0122] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A LID facility spatial optimization layout method coupling SWMM model and GDE3 algorithm, characterized in that: The steps include: A SWMM model was constructed to simulate and evaluate the hydrological effects of different LID facility spatial layouts to obtain the runoff reduction rate of the study area; Build a cost calculator to obtain the total construction cost for different LID facility space layouts; A multi-objective decision-making model for the spatial layout of LID facilities was established to determine the decision variables for LID layout, and the boundary conditions were determined with the minimum total construction cost of LID facilities and the maximum runoff reduction rate in the study area as the objective function. Coupling the SWMM model and the GDE3 algorithm: The GDE3 algorithm generates decision variables in the multi-objective decision model through crossover mutation, and inputs them into the SWMM model and the cost calculator to obtain the runoff reduction rate and total construction cost of the study area. The non-dominated sorting and congestion distance calculations are used to obtain the optimal solution set of decision variables in the multi-objective decision model of the spatial layout of LID facilities. The specific steps of coupling the SWMM model and the GDE3 algorithm include: Randomly initialize the population in the feasible solution space of decision variables; Perform mutation and crossover operations on the parent individuals to generate another new population of offspring; The new population individuals correspond to the area or proportion of each suitable LID type in each newly generated sub-catchment, and the overland flow width and imperviousness of each sub-catchment are recalculated; The new population individuals and the changed overland flow width and impervious rate parameters were sequentially input into the SWMM model for hydrological simulation, the outflow of the study area outlet was obtained, and the runoff reduction rate was calculated. At the same time, the individuals were input into the cost calculator for calculation to obtain the converted annual total cost. The parent and child populations were quickly non-dominated sorted based on the objective function. The population size was kept at the initial size based on the crowding distance method, and the iteration was repeated to the set number of iterations. The parent generation individuals are mutated and crossovered to generate another new population of offspring, and the specific steps include: Randomly initialize the population in the feasible solution space of the decision variables The population size is NP; For each individual X in generation j i (j) Implement the mutation operation to obtain the corresponding mutant individual X m (j+1); for individual X r1 (j) and the generated variant individuals X m (j+1) performs crossover operation to generate the test individual U(j+1): The mutation operation is expressed as: X m (j+1)=X r1 (j)+F*[X r2 (j)-X r3 (j)] Where j represents the evolutionary generation; individuals r1, r2, r3∈{1,2,3...,NP} are different from each other and from individual m; X r1 (j) is the parent basis vector, [X r2 (j)-X r3 (j)] is the parent difference variable, and F is the scaling factor; The crossover operation is expressed as: Among them, rand(0,1) is a uniformly distributed random number between [0, 1], CR is the crossover probability between [0, 1], j rand is a random quantity between {1,2,3...,NP}.

2. The LID facility space optimization layout method of coupling the SWMM model and the GDE3 algorithm according to claim 1 is characterized in that: The SWMM model is constructed by obtaining basic data of the study area in advance, wherein the basic data include digital elevation model, satellite remote sensing image data, pipe network distribution and land use type data, dividing sub-catchment areas based on the basic data, determining sub-catchment areas and pipe network parameters, and determining the type and proportion of LID suitable for construction in the study area.

3. The LID facility space optimization layout method of coupling SWMM model and GDE3 algorithm according to claim 1 is characterized in that: The construction cost calculator predetermines the total construction cost and maintenance cost of various LID facilities and calculates the annual discounted cost of various LID facilities, which is specifically expressed as: Among them, A LID and P LID Respectively represent the annual construction cost and the total construction cost, i represents the annual interest rate; n LID Indicates the life cycle of each LID facility.

4. The LID facility space optimization layout method of coupling the SWMM model and the GDE3 algorithm according to claim 1 is characterized in that: The multi-objective decision model for the spatial layout of LID facilities, wherein the decision variables are selected from the construction ratio or area of ​​LID facilities in each sub-catchment area; The objective function is to minimize the total construction cost of LID facilities and maximize the runoff reduction rate in the study area, which is specifically expressed as: Maximize the runoff reduction rate in the study area: Among them, Runoff baseline is the outlet flow of the study area without LID facilities, Runoff LID is the outlet flow of the study area under different LID layouts generated by the multi-objective optimization algorithm; Minimum goal for total construction cost of LID facilities: Among them, A ij represents the area occupied by LID facility type j suitable for construction in subcatchment i, C j represents the total annual cost per unit LID facility type j, m represents the total number of subcatchments, and n represents the total number of LID facility types.

5. The LID facility space optimization layout method of coupling SWMM model and GDE3 algorithm according to claim 1 is characterized in that: The boundary condition is expressed as: Wherein, subscript l and superscript n represent LID type and subcatchment name, respectively; It represents the maximum suitable buildable area of ​​LID type l in subcatchment n.

6. A LID facility space optimization layout system coupling SWMM model and GDE3 algorithm, characterized in that: A method for spatial optimization layout of LID facilities for implementing the coupling SWMM model and GDE3 algorithm as described in any one of claims 1 to 5, the system comprising: a SWMM model building module, a cost calculator building module, a multi-objective decision model building module, a SWMM model and GDE3 algorithm coupling module; The SWMM model building module is used to build a SWMM model, simulate and evaluate the hydrological effects of different LID facility spatial layouts, and obtain the runoff reduction rate of the study area; The cost calculator building module is used to build a cost calculator to obtain the total construction cost under different LID facility space layouts; The multi-objective decision model building module is used to establish a multi-objective decision model for the spatial layout of LID facilities, determine the decision variables for LID layout, and determine the boundary conditions with the minimum total construction cost of LID facilities and the maximum runoff reduction rate in the study area as the objective function; The SWMM model and GDE3 algorithm coupling module is used to couple the SWMM model and the GDE3 algorithm: the GDE3 algorithm generates decision variables in the multi-objective decision-making model through crossover mutation, inputs them into the SWMM model and the cost calculator, obtains the runoff reduction rate and the total construction cost of the study area, and uses non-dominated sorting and congestion distance calculation to obtain the optimal solution set of decision variables in the multi-objective decision-making model of the spatial layout of LID facilities.

7. A computer-readable storage medium storing a program, characterized in that: When the program is executed by a processor, the LID facility spatial optimization layout method of coupling the SWMM model and the GDE3 algorithm as described in any one of claims 1 to 5 is implemented.

8. A computing device comprising a processor and a memory for storing a program executable by the processor, characterized in that: When the processor executes the program stored in the memory, the LID facility space optimization layout method of coupling the SWMM model and the GDE3 algorithm as described in any one of claims 1 to 5 is implemented.

Images