Arbor configuration method and device in green space based on multi-objective optimization

Abstract

The embodiment of the present application provides a method and device for arbor configuration in a green space based on multi-objective optimization. The method includes: repeatedly executing the following steps to obtain multiple arbor configuration schemes for the green space: generating an arbor combination, the arbor combination includes individuals of different ages of each type of arbor among multiple arbor types; calculating the sum of carbon sinks, the sum of maintenance costs, and the sum of vertical projection areas of all individuals in the arbor combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the sum of vertical projection areas of the arbor meet the preset constraints, the arbor combination is determined as an arbor configuration scheme, with carbon sinks, maintenance costs, and vertical projection areas of the arbor as optimization targets, and multi-objective optimization of multiple arbor configuration schemes is performed according to the three target values ​​to obtain an optimized arbor configuration scheme, so that the optimized arbor configuration scheme is more in line with the layout requirements of the green space in terms of carbon sinks, maintenance costs, and vertical projection areas of the arbor.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of ecological environment design, and in particular to a method and device for configuring trees in a green space based on multi-objective optimization. Background Art

[0002] Since the adoption of the United Nations Framework Convention on Climate Change in 1992, the world has paid more attention to environmental issues such as controlling the emission of greenhouse gases such as carbon dioxide (CO2) and responding to global warming. The carbon (C) sequestration capacity of forest land is of great significance to mitigating global warming, and their carbon absorption can effectively and economically offset carbon emissions.

[0003] Existing studies on the assessment of forest carbon sinks mainly focus on the distribution of afforestation area, usually only focusing on the carbon storage target, and failing to fully consider the balance and coordination among multiple objectives in the afforestation process.

[0004] At present, multi-objective optimization algorithms have been widely used in complex decision-making problems, but there is a lack of research on their application in afforestation layout. How to use multi-objective optimization algorithms to generate efficient and economically reasonable tree configuration plans while meeting multi-objective requirements such as carbon sequestration, cost, and area is the core problem that existing technologies can solve. Summary of the invention

[0005] In view of this, an embodiment of the present application provides a method and device for configuring trees in a green space based on multi-objective optimization, so as to at least solve or alleviate the above-mentioned problems.

[0006] According to one aspect of an embodiment of the present application, a method for configuring trees in a green space based on multi-objective optimization is provided, the method comprising:

[0007] Repeat the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each type of tree in different ages; calculate the total carbon sink amount, the total maintenance cost and the total vertical projection area of ​​all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the total vertical projection area of ​​the trees meet the preset constraints, determine the tree combination as the tree configuration scheme, and the tree configuration scheme indicates the type, quantity and age of the trees configured for the green space;

[0008] Taking carbon sink amount, maintenance cost and vertical projection area of ​​trees as optimization targets, multi-objective optimization of the plurality of tree configuration schemes is performed according to the three target values ​​to obtain an optimized tree configuration scheme.

[0009] According to another aspect of an embodiment of the present application, a device for arbor configuration in a green space based on multi-objective optimization is provided, the device comprising:

[0010] A configuration module is used to repeatedly perform the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each tree in a plurality of trees at different ages; calculate the sum of carbon sinks, the sum of maintenance costs and the sum of vertical projection areas of all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the sum of vertical projection areas of the trees meet preset constraints, determine the tree combination as the tree configuration scheme, and the tree configuration scheme indicates the type, quantity and age of the trees configured for the green space;

[0011] The optimization module is used to take carbon sink amount, maintenance cost and vertical projection area of ​​trees as optimization targets, perform multi-objective optimization of the plurality of tree configuration schemes according to the three target values, and obtain an optimized tree configuration scheme.

[0012] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other through the communication bus; the memory is used to store at least one executable instruction, and when the executable instruction is executed by the processor, the method for configuring trees in a green space based on multi-objective optimization as provided in one aspect above is implemented.

[0013] According to another aspect of an embodiment of the present application, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for configuring trees in a green space based on multi-objective optimization as described in one aspect above is implemented.

[0014] According to another aspect of an embodiment of the present application, a computer program product is provided, the computer program product comprising computer instructions, the computer instructions instructing a computing device to execute the method for configuring trees in green space based on multi-objective optimization as described in one aspect above.

[0015] According to the tree configuration method in green space based on multi-objective optimization provided in the embodiment of the present application, multiple tree combinations are generated, each tree combination is composed of individuals of multiple trees of different ages, and then the three target values ​​of the sum of carbon sequestration amounts, the sum of maintenance costs and the sum of vertical projection areas of the trees of all individuals in each tree combination during the entire life cycle of the green space are calculated, and a combination that meets the constraints is selected from the multiple tree combinations as the tree configuration scheme, and then the carbon sequestration amount, maintenance cost and vertical projection area of ​​the trees are used as optimization targets, and multi-objective optimization of multiple tree configuration schemes is performed according to the corresponding three target values. Through multi-objective optimization, a tree configuration scheme that is more suitable for the green space is obtained, so that the tree configuration on the green space can achieve a better carbon sequestration effect. At the same time, the builders' preference for the vertical projection area of ​​the trees in the green space and the cost preference for tree planting and maintenance are also taken into account. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 is a schematic diagram of an exemplary system used in one embodiment of the present application;

[0018] Figure 2 It is a flow chart of a method for configuring trees in a green space based on multi-objective optimization according to an embodiment of the present application;

[0019] Figure 3 is a flow chart of a method for configuring trees in a green space based on multi-objective optimization according to another embodiment of the present application;

[0020] Figure 4 is a schematic diagram of a device for arbor configuration in a green space based on multi-objective optimization according to an embodiment of the present application;

[0021] Figure 5 It is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] The present application is described below based on embodiments, but the present application is not limited to these embodiments. In the detailed description of the present application below, some specific details are described in detail. It is possible for a person skilled in the art to fully understand the present application without the description of these details. In order to avoid confusing the essence of the present application, known methods, processes, and flows are not described in detail. In addition, the drawings are not necessarily drawn to scale.

[0023] First, some nouns or terms that appear in the process of describing the embodiments of the present application are subject to the following explanations.

[0024] Carbon sink refers to the ability of plants to absorb carbon dioxide from the atmosphere through photosynthesis and fix it in vegetation and soil. The amount of carbon sink refers to the change in carbon storage in forest carbon pools over a certain period of time. Ecosystems such as forests, grasslands, and agriculture are all important carbon sinks, which play an important role in global climate change by absorbing and storing carbon dioxide.

[0025] Maintenance cost refers to the comprehensive cost consumed during a certain maintenance cycle of a plant, including the cost of seedlings consumed in the initial planting, planting labor costs and other planting costs, transportation costs, as well as maintenance costs such as labor maintenance costs consumed in the later maintenance, and the cost of purchasing nutrients and medicines required by the plants.

[0026] The vertical projection area of ​​trees refers to the vertical projection area of ​​the tree crown on the ground of green space. The crown width of the tree crown can be used as the vertical projection area of ​​the tree.

[0027] The entire life cycle of green space refers to the entire process from green space planning, design, construction, operation, maintenance to demolition and disposal.

[0028] Figure 1 An exemplary computer system is shown, which includes an electronic device 101 and a server 102 .

[0029] The electronic device 101 is installed and runs an application program, and the application program is applicable to the method for configuring trees in green space based on multi-objective optimization provided in the embodiment of the present application. For example, during the execution of the method for configuring trees in green space based on multi-objective optimization, initialization parameters such as population size, crossover probability, mutation probability, and maximum number of iterations are set through the application program.

[0030] The electronic device 101 is connected to the server 102 via a wireless network or a wired network. The server 102 includes at least one of a single server, multiple servers, a cloud computing platform, and a virtualization center. The server 102 is used to support the running of applications on the electronic device, that is, to provide background services for the applications running on the electronic device 101. For example, a plant database is constructed on the server 102, and the plant database records the relevant data of various plants, such as the type and age of the plant, and the carbon sequestration amount, crown width, unit price of the plant at different ages, the planting cost and maintenance cost of a single plant, etc.

[0031] Optionally, the server 102 undertakes the main computing work, and the electronic device 101 undertakes the secondary computing work; or, the server 102 undertakes the secondary computing work, and the electronic device 101 undertakes the main computing work; or, a distributed computing architecture is used between the server 102 and the electronic device 101 for collaborative computing. In the embodiments provided in this application, the electronic device undertakes the main computing work as an example for explanation.

[0032] Based on the above system, an embodiment of the present application provides a method for configuring trees in a green space based on multi-objective optimization. The method can be executed by an electronic device. The method is described in detail through multiple embodiments below.

[0033] Figure 2 FIG. 1 is a flow chart of a method for configuring trees in a green space based on multi-objective optimization according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0034] Step 210, repeatedly perform the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each type of tree in a plurality of trees at different ages; calculate the total carbon sink amount, the total maintenance cost and the total vertical projection area of ​​all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the total vertical projection area of ​​the trees meet the preset constraints, determine the tree combination as a tree configuration scheme, and the tree configuration scheme indicates the type, quantity, and age of the trees configured for the green space.

[0035] The electronic device randomly generates a tree combination, which includes multiple trees, and each tree includes individuals of different ages. The life cycles of multiple trees in the tree combination are the same, for example, all types of trees in the tree combination have a life cycle of 50 years; or the life cycles of multiple trees in the tree combination are different, for example, the life cycle of some types of trees in the tree combination is 30 years, and the life cycle of other types of trees is 50 years.

[0036] Tree species include evergreen trees and deciduous trees; evergreen trees also include flowering trees, colorful leaf trees and fruit trees, and deciduous trees also include flowering trees, colorful leaf trees and fruit trees. Each type of tree in the tree combination has its own number of individuals at different ages.

[0037] After generating the tree combination, obtain the carbon sink, maintenance cost and vertical projection area of ​​each tree at different ages from the pre-built plant database; based on the obtained data, calculate the total carbon sink, maintenance cost and vertical projection area of ​​all the trees in the tree combination in the whole life cycle of the green space, for example, calculate the total carbon sink, maintenance cost and vertical projection area of ​​all the trees in the tree combination in the 50-year life cycle of the green space, and obtain the three target values ​​for the three optimization goals of carbon sink, maintenance cost and vertical projection area of ​​the trees. The whole life cycle of the green space can be 20 years, 30 years, 50 years, etc., which can be set by the builder.

[0038] Optionally, the vertical projection area of ​​an individual tree can be expressed by the crown width of the individual tree.

[0039] If the area-related parameters including the sum of the vertical projection areas of the trees meet the preset constraints, then the tree combination can be used as a tree configuration scheme for the green space, used to indicate the type, quantity, and age of the trees configured for the green space.

[0040] Repeat the above process to obtain multiple tree configuration schemes.

[0041] Step 220, taking carbon sink amount, maintenance cost and vertical projection area of ​​trees as optimization targets, multi-target optimization of multiple tree configuration schemes is performed according to the three target values ​​to obtain an optimized tree configuration scheme.

[0042] To summarize, the method for configuring trees in green spaces based on multi-objective optimization provided in this embodiment generates multiple tree combinations, each tree combination is composed of individuals of multiple trees of different ages, and then calculates the three target values ​​of the total carbon sequestration amount, the total maintenance cost and the total vertical projection area of ​​the trees of all individuals in each tree combination during the entire life cycle of the green space, and selects a combination that meets the constraints from the multiple tree combinations as the tree configuration scheme, and then uses the carbon sequestration amount, maintenance cost and the vertical projection area of ​​the trees as optimization targets, and performs multi-objective optimization of multiple tree configuration schemes according to the corresponding three target values. Through multi-objective optimization, a tree configuration scheme that is more suitable for the green space is obtained, so that the tree configuration on the green space can achieve a better carbon sequestration effect. At the same time, it also takes into account the builders' preference for the vertical projection area of ​​the trees in the green space and the cost preference for tree planting and maintenance.

[0043] above Figure 2The tree configuration method in green space based on multi-objective optimization shown in FIG. 1 can be implemented by genetic algorithms such as non-dominated sorting genetic algorithms (NSGA) II and NSGA III. Taking NSGA III as an example, Figure 3 , the steps include:

[0044] Step 310, repeatedly perform the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each type of tree in a plurality of trees at different ages; calculate the total carbon sink amount, the total maintenance cost and the total vertical projection area of ​​all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the total vertical projection area of ​​the trees meet the preset constraints, the tree combination is determined as the tree configuration scheme.

[0045] The electronic device determines initialization parameters, which include a preset population size N, a crossover probability, a mutation probability, a maximum number of iterations, a number of reference points, and constraints.

[0046] Population size N: It is determined based on factors such as the problem size and computing resources. It is usually a fixed positive integer, such as a positive integer greater than or equal to 50 and less than or equal to 300.

[0047] Crossover probability P c : Generally set between 0 and 1.

[0048] Mutation probability P m : Generally set between 0 and 1.

[0049] Maximum number of iterations Max: Determined based on the convergence speed of the problem and the expected solution accuracy.

[0050] According to the data recorded in the plant database, the selectable tree species and their related attributes, such as carbon sequestration, age, crown width, maintenance cost, etc., are determined, and then the above steps of generating tree configuration schemes are repeated to obtain N tree configuration schemes. Exemplarily, among the trees aged 0 to 50 years, multiple types of trees of different ages are randomly selected; according to the selected trees and their ages, combined with the carbon sequestration fitting curve and the carbon sequestration, maintenance cost and vertical projection area of ​​each tree at different ages recorded in the plant database, the three target values ​​of the sum of carbon sequestration, maintenance cost and vertical projection area of ​​all individuals of the multiple trees in the tree combination during the entire life cycle of the green space are calculated; check whether the tree combination meets the constraints, such as canopy density and builder preference, etc. If the constraints are not met, the tree combination is regenerated until the constraints are met; if the constraints are met, the tree combination is determined as the tree configuration scheme as an individual in the initial population until N tree configuration schemes are generated.

[0051] Exemplarily, the objective function F1 for calculating the total carbon sink amount is:

[0052] F1= ;

[0053] ;

[0054] Z is the number of tree species in the tree combination, is the carbon sink of the vth tree, T1 is the life cycle of the tree, T2 is the whole life cycle of the green space, t is the growth age of the tree, is the number of trees of age t in the vth tree species, f v (t) is the carbon sequestration curve of the vth tree species during its life cycle, It is the total cumulative carbon sink of the v-th type of tree from age t to T2+t, that is, the total cumulative carbon sink of the v-th type of tree at age t during the entire life cycle of the green space.

[0055] The objective function F2 for calculating the total maintenance cost is:

[0056] F2= ;

[0057] ;

[0058] is the maintenance cost of the vth tree and, is the price of a single plant of the vth type of tree, The labor cost for planting a single tree of the vth type of tree is: is the transportation cost of a single plant of the vth tree, is the average annual maintenance cost of a single tree of the vth type, is the number of trees of the vth type.

[0059] Or, the maintenance cost of the vth tree and The calculation formula can also be:

[0060] ;

[0061] is the maintenance cost of a single tree of the vth type of tree during its entire life cycle in the green space.

[0062] The objective function F3 for calculating the sum of the vertical projection areas of trees is:

[0063] F3= ;

[0064] ;

[0065] is the sum of the projection areas of the vth plant (sum of the vertical projection areas of the trees), is the crown width of the v-th tree at age t1, is the crown width of the vth type of tree at age t2, t1 and t2 are the different growth ages of the tree (also the age of the tree when it was planted).

[0066] Optionally, constraints include:

[0067] 1) According to the technical regulations for forest carbon sequestration and sink enhancement, the first constraint condition is set:

[0068] 0.3 degree of depression 0.8, the closure degree is the ratio of the sum of the vertical projection areas of trees (the total planting area of ​​all types of trees in the tree combination) to the total green area of ​​green space. " means less than or equal to;

[0069] The total vertical projection area of ​​trees Total green area.

[0070] 2) According to the builder's preference, set the planting area ratio of each type of tree as the second constraint:

[0071] Planting area of ​​evergreen and deciduous trees Total green space area;

[0072] 0 Evergreen tree planting area 0.3× total vertical projection area of ​​trees;

[0073] Planting area of ​​flowering trees, colorful leaf trees and fruit trees Planting area of ​​evergreen trees and deciduous trees;

[0074] Among them, the planting area or planting ratio of flowering trees, foliage trees and fruit-viewing trees are set by the builders respectively. The planting ratio refers to the ratio of the planting area of ​​flowering trees, the planting area of ​​foliage trees and the planting area of ​​fruit-viewing trees.

[0075] In the algorithm, the tree configuration scheme is represented in a coded form. Therefore, after obtaining the tree configuration scheme, the relevant parameters of each tree are represented by a real vector, and based on the real vector corresponding to each tree, each tree configuration scheme in the multiple tree configuration schemes is encoded to obtain the individual code corresponding to each tree configuration scheme. Assume that there are a types of trees and b types of parameters in the tree configuration scheme to be encoded, and the e-th parameter of the c-th tree is represented as D ce (c=1, 2…, a; e=1, 2,…, b), the individual code P of the tree configuration scheme can be expressed as:

[0076] ;

[0077] Among them, D ce The actual value of the parameter.

[0078] Step 320, forming an initial population using multiple tree configuration schemes.

[0079] Step 330, for the population, loop through the following steps until the loop termination condition is met: take the population as the parent population, perform crossover and mutation operations on the parent population to generate a child population; merge the parent population and the child population to obtain a merged population; perform fast non-dominated sorting on the merged population to determine the non-dominated level corresponding to each tree configuration scheme; for each tree configuration scheme, determine the closest reference point to the tree configuration scheme according to its corresponding three target values, and determine the distance between the tree configuration scheme and the closest reference point as the reference distance of the tree configuration scheme, wherein the reference point consists of the reference value of the optimization target; select at least part of the tree configuration schemes from the merged population according to the non-dominated level, the closest reference point and the reference distance to obtain the next generation population.

[0080] The implementation of each part of this step is as follows:

[0081] Take the population as the parent population, perform crossover and mutation operations on the parent population to generate a child population; merge the parent population and the child population to obtain a merged population.

[0082] Assume The generation population set is , the number is N, after crossover and mutation operations, the offspring population is generated , the number is also N, merge the offspring population and the parent population to get the merged population .

[0083] Crossover operation: Generate new offspring individuals by exchanging some genes of two individuals in the parent population. The crossover operation includes multi-point crossover, which is to select multiple crossover points in the parent individuals and exchange gene fragments, so that the good genes of different parent individuals (i.e., individuals in the parent population) are recombined in the offspring individuals (i.e., individuals in the offspring population), which may produce better offspring individuals.

[0084] Exemplarily, the implementation of multi-point crossover is as follows: randomly select a pair of parent individuals, and then randomly generate a number Sr, whose value range is [0,1], that is, greater than or equal to 0 and less than or equal to 1; if Sr <P c (crossover probability), then perform a multi-point crossover operation; if Sr P c , no crossover operation is performed, and the two parent individuals are directly copied to the offspring population (that is, the offspring individuals are the same as the parent individuals). " means greater than or equal to.

[0085] Assume that the number of crossover points is O. In the gene sequence (real vector) length range [1, g], g = (a × b), randomly select O crossover points, denoted as C1, C2, … C O ; Two parents exchange corresponding fragments to generate offspring individuals. For example: when O=2, the crossover points are C1, C2, then for the parents P1, P2, the generated offspring P_1, P_2 are as follows:

[0086] ;

[0087] .

[0088] The mutation operation includes uniform mutation, which is to change the gene value of an individual (the number of trees of a certain age, the carbon sequestration of trees of a certain age, etc.) according to a certain mutation probability P. m , for the gene selected for mutation, a new value within its value range is randomly generated to replace the original value. This can make individuals produce more uniform changes in the real number space and increase the diversity of the population.

[0089] For example, uniform mutation is achieved by selecting individuals with low non-dominated ranks and crowding degrees for mutation, which has a probability of changing their performance. For the gene position (parameter) of the selected individual, a number Sr is randomly generated, and its value range is [0,1]. If Sr <P m(Mutation probability), then perform mutation operation. For the gene position (parameter) that needs to be mutated , let its value range be [ , ], then the value after mutation = +Sr( - ), where Sr is a newly generated random number in the interval [0,1].

[0090] Perform fast non-dominated sorting on the merged population to determine the non-dominated level corresponding to each tree configuration scheme.

[0091] In the embodiment of the present application, the dominance relationship of NSGA III is: if all optimization targets (carbon sink, maintenance cost, vertical projection area of ​​trees) of tree configuration scheme A are not worse than those of tree configuration scheme B, and at least one optimization target is stronger than that of tree configuration scheme B, then tree configuration scheme A is said to dominate tree configuration scheme B, or tree configuration scheme B is said to be dominated by tree configuration scheme A. The electronic device calculates the dominance relationship and lays the foundation for hierarchical sorting.

[0092] Hierarchical sorting is to divide the individuals in the population into different levels according to the non-dominated relationship, and the individuals in each level are mutually non-dominated. The electronic device hierarchically sorts the tree configuration schemes according to the dominance relationship between the tree configuration schemes. The non-dominated sorting of the tree configuration schemes in the population is:

[0093] (1) Initialization. For each individual P in the population, initialize the dominating set S P (stores individuals dominated by P) and the number of dominated n P (Record the number of individuals that dominate P.) An individual in the population is a tree configuration plan.

[0094] (2) Traverse each pair of individuals P and Q in the population and compare their dominance relationship: if P dominates Q, then add Q to P’s dominance set S P If Q dominates P, then the number of P dominated by n P Add 1.

[0095] (3) Place the individuals with a dominance number of 0 into the first non-dominated group f_1.

[0096] (4) For the current non-dominated group f_µ(µ 1) For each individual P in the , traverse its dominating set S P For each individual Q, the non-dominated number n of Q Q Subtract 1. If n Q becomes 0, and Q is placed in the next non-dominated level population f_(µ+1).

[0097] (5) Repeat step (4) until all individuals are assigned to the corresponding non-dominated level.

[0098] For each tree configuration scheme, the nearest reference point is associated with the tree configuration scheme according to the three corresponding target values, and the distance between the nearest reference point and the tree configuration scheme is determined as the reference distance of the tree configuration scheme, wherein the reference point is composed of the reference value of the optimization target.

[0099] The reference points are generated based on the value range of the objective function. A set of appropriate reference points is generated through the objective function to provide effective guidance and support for subsequent association and selection operations, helping the algorithm to search for high-quality non-dominated solutions in the multi-objective space.

[0100] Optionally, the reference points are generated by:

[0101] For each optimization target, find the maximum and minimum values ​​of the optimization target from the pre-built plant database, multiply the maximum and minimum values ​​by the population size N, and obtain the maximum and minimum values ​​of the value range of the optimization target; divide the value range of each optimization target into N / 2 parts at equal intervals to obtain N / 2 sub-value ranges; calculate the coordinates of the u-th reference point , where the coordinates of the u-th reference point on the j-th objective function are The calculation of is as follows:

[0102] ;

[0103] in, is the maximum value of the j-th optimization target in the u-th sub-value range, is the minimum value of the j-th optimization objective in the u-th sub-value range, the value of u is greater than zero and less than or equal to N / 2, and j is a positive integer less than or equal to 3.

[0104] After calculating the complete reference point coordinates After that, the reference points are normalized. For the normalized coordinates of the u-th reference point in the j-th objective function :

[0105] ;

[0106] in, is the maximum value of the jth optimization objective in its value range.

[0107] Finally, the validity of the reference points is checked to see if the generated reference points meet the constraints of the preset reference points. If not, the coordinate values ​​can be adjusted or discarded and regenerated, and N / 2 reference points are finally generated.

[0108] The association between the tree configuration scheme and the reference point includes: for each tree configuration scheme, based on the three target values ​​of the tree configuration scheme, generating a target vector of the tree configuration scheme; from multiple reference points, determining the reference point with the smallest Euclidean distance to the target vector, obtaining the nearest reference point of the tree configuration scheme, and determining the Euclidean distance between the reference point and the nearest reference point as the reference distance of the tree configuration scheme.

[0109] For example, to calculate the distance from individual A to the reference point As an example, based on the three target values ​​of individual A (i.e., tree configuration scheme A), the target vector , reference point The coordinates of , then individual A to the reference point The Euclidean distance The calculation formula is:

[0110] ;

[0111] According to the calculated Euclidean distance from the individual to the reference point, each individual is associated with the reference point closest to it. That is, for individual A, compare its distance to all reference points (u=1,2,…,N / 2), find the reference point with the smallest distance , the two are related.

[0112] · According to the non-dominated level, the nearest reference point and the reference distance, at least part of the tree configuration scheme is selected from the merged population to obtain the next generation population.

[0113] The electronic device selects N tree configuration schemes from the first L non-dominated levels according to the non-dominated level, the nearest reference point and the reference distance, where L is an integer greater than 1.

[0114] Optionally, n tree configuration schemes are selected from the first L-1 non-dominated levels, where L is an integer greater than 1 and n is a positive integer; if n is less than the population size N, the reference point with the smallest number of microhabitats is selected as the alternative reference point from the reference points associated with the n tree configuration schemes, where the number of microhabitats of the reference point is the number of schemes associated with the reference point; based on the alternative reference points, (Nn) tree configuration schemes are selected from the Lth non-dominated level.

[0115] Optionally, according to the alternative reference point, (Nn) tree configuration schemes are selected from the Lth non-dominated level, including: if the number of microhabitats of the alternative reference point is zero, selecting the tree configuration scheme that is closest to the alternative reference point from the Lth non-dominated level; if the number of microhabitats of the alternative reference point is not zero, selecting the tree configuration scheme associated with the alternative reference point from the Lth non-dominated level; until (Nn) tree configuration schemes are selected from the Lth non-dominated level.

[0116] The selection operation is to Select N individuals from +1 generation population set .

[0117] First, the parent population and progeny population Merge to obtain a merged population, which is a new population of size 2N (In the first round of selection, since there is no offspring population, the selection operation is performed directly on the parent population.) Perform fast non-dominated sorting to divide individuals into groups of different non-dominated levels f_1, f_2,…. Starting from the first non-dominated level group f_1, select individuals to enter the next generation population in sequence until the number of individuals in the next generation population reaches N. If the number of individuals in the Lth non-dominated level group f_L is greater than N, select the first Nn (assuming n have been selected) individuals to enter the next generation population.

[0118] If the number of individuals in f_1 is less than N, all individuals in f_1 are selected to join the population , continue to select from the second non-dominated level population f_1 until N.

[0119] Assume that this is the level non-dominated level, if = N, let = ;like >N, let , and select K=Nn individuals from f_L to join ,n= , also known as microhabitat retention strategy.

[0120] Niche Retention Strategies:

[0121] (1) Based on the number of microhabitats at the reference point (The associated individuals are selected from the current ),choose The smallest one is used as an alternative reference point , If they are equal, pick one at random.

[0122] (2) From Select the non-dominated set of level The associated individuals form a collection .like Empty, indicating that the reference point is If the level is not associated with any individual, delete the reference point; if it is not empty, further observe the number of microhabitats :like , indicating the reference point No contact with the previous -1 level individuals, in order to increase population diversity, from the collection Select and reference point The closest individual, join ;like , from the collection Randomly select an individual to join .

[0123] (3) Every time a new individual enters , k=k+1 (k is initially 0).

[0124] (4) Repeat until k=K. Finally, we get a new population .

[0125] Step 340, determining the tree configuration scheme in the next generation population obtained last time as the optimized tree configuration scheme.

[0126] After iterating Max times, the population obtained from the last iteration is The middle tree configuration plan is determined to be the optimized tree configuration plan.

[0127] To sum up, the method for configuring trees in green spaces based on multi-objective optimization provided in this embodiment adopts the NSGAIII algorithm to perform multi-objective optimization on the tree configuration scheme of the green space, considering the three objectives of carbon sequestration, maintenance cost and vertical projection area of ​​the trees, and generating an efficient and economically reasonable tree configuration scheme on the premise of meeting multi-objective requirements, so that the obtained tree configuration scheme can obtain better carbon sequestration effect on the green space, and also takes into account the builders' preference for the vertical projection area of ​​the trees in the green space and the cost preference for tree planting and maintenance.

[0128] In some other possible implementation methods, the total carbon sink-cost preference (cost type) or total carbon sink-area preference (benefit type) analysis can be conducted according to the builders' preference needs to determine the tree configuration plan that better matches the above preference needs.

[0129] Cost analysis:

[0130] From the optimal solution set obtained by iteration (including the optimized tree configuration scheme), the total carbon sink and total maintenance cost of each scheme are calculated according to the plant database. With the total carbon sink as the ordinate and the total maintenance cost as the abscissa, a scatter plot is drawn to analyze the distribution of these points and the changing trend of the total maintenance cost as the total carbon sink increases.

[0131] Benefit analysis:

[0132] For all the schemes that meet the constraints, the total carbon sink and the total vertical projection area of ​​the trees are calculated according to the plant database. With the total carbon sink as the ordinate and the total vertical projection area of ​​the trees as the abscissa, a scatter plot is drawn to analyze the distribution of these points and the changing trend of the total vertical projection area of ​​the trees as the total carbon sink increases.

[0133] According to the changing trend, the tree configuration plan with the best cost or the best area distribution can be determined from the optimized tree configuration plans.

[0134] In some other possible implementations, in order to select the best solution, the entropy weight method can also be used to calculate the weight of each optimization target, and the user can select the optimal tree configuration solution according to the weight ratio of different optimization targets.

[0135] For example, for each optimized tree configuration scheme, the three target values ​​of the three optimization targets of carbon sequestration, maintenance cost and vertical projection area of ​​the tree are standardized to obtain three standardized target values, among which the jth target value in the i-th optimized tree configuration scheme is The standardized formula is:

[0136] ;

[0137] Calculate the jth standardized target value in the i-th optimized tree configuration scheme The proportion in the corresponding optimization target , the formula is as follows:

[0138] = ;

[0139] Based on the weight of the jth optimization objective , calculate the entropy value of the jth optimization objective , the formula is as follows:

[0140] ;

[0141] Based on the entropy value of the jth optimization target , calculate the weight of the jth optimization objective , the formula is as follows:

[0142] ;

[0143] For each optimized tree configuration scheme, the weighted values ​​of the three target values ​​are calculated according to the three weights corresponding to the three optimization objectives to obtain the evaluation score of the optimized tree configuration scheme; according to the evaluation score, the optimal tree configuration scheme is determined from the optimized tree configuration schemes; wherein, is the minimum value of the jth optimization objective, is the maximum value of the jth optimization objective.

[0144] In some other possible implementations, the builder can also give preference weights, such as setting carbon sink weight W1, cost weight W2, and area weight W3, to satisfy W1+ W2+ W3=1. For each optimized tree configuration scheme, the weighted value of the three target values ​​is calculated according to the weights preset by the builder to obtain the evaluation score s for the optimized tree configuration scheme. i , the formula is as follows:

[0145] s i = W1 + W2 + W3 ;

[0146] According to the evaluation score i , determine the best tree configuration scheme from the optimized tree configuration schemes. For example, the score s i The larger it is, the better the optimized tree configuration plan is.

[0147] In some other possible implementations, trees with survival advantages in green spaces are selected for configuration according to the environment of the green space. Species distribution information and bioclimatic variables can be obtained according to the climate conditions of the dry and wet zones set by the builders; dominant tree species that adapt to the environmental conditions of the green space can be determined based on the species distribution information and bioclimatic variables; and then tree combinations are generated based on the dominant tree species. In other words, dominant tree species are selected from the plant database, and tree configuration plans are generated based on the dominant tree species.

[0148] For example, the server records the dry and wet divisions of various places: arid, semi-arid, semi-humid and humid areas, the climate conditions of arid, semi-arid, semi-humid and humid areas, species distribution information (longitude and latitude), and bioclimatic variables (including annual average temperature, average daily variation, isotherms, temperature seasonality, maximum temperature of the warmest month, minimum temperature of the coldest month, annual temperature range, average temperature of the wet season, average temperature of the driest season, average temperature of the hottest quarter, average temperature of the coldest season, annual precipitation, precipitation of the wettest month, precipitation of the driest month, precipitation seasonality, precipitation in the wet season, precipitation in the driest area, precipitation in the hottest season, precipitation in the coldest season, etc.).

[0149] Builders can select the location of the green space and the dry and wet zones to which the green space belongs on the electronic device. According to the location of the green space and the dry and wet zones to which it belongs, they can obtain information such as species distribution information and bioclimatic variables; then, based on the above series of information, the dominant tree species of the green space can be analyzed. For example, according to the dry and wet zones, the climate conditions of the dry and wet zones, the species distribution information (latitude and longitude), and the bioclimatic variables, the Maximum Entropy (MaxEnt) model is used to infer the ecological needs of the species through the known distribution data of the species and related environmental variables, and then the calculation results are projected into different spaces and times to predict the potential distribution area of ​​the species. MaxEnt has a small sample demand and excellent prediction ability. The MaxEnt model restricts the predicted distribution through a series of constraints to make it consistent with the known data and select the dominant tree species suitable for its environmental conditions. The dominant tree species are determined based on the potential feasibility established under local environmental conditions, in order to minimize unexpected failures while maximizing carbon sequestration benefits.

[0150] Secondly, the i-Tree ecological model is used to estimate the carbon storage of various tree types at several ages. After entering the i-tree ecological model, the geographical location of the tree is selected. The i-tree ecological model uses the tree growth model and biomass equation based on the Latin name of the tree, the diameter at breast height, to obtain the carbon storage of the tree at its current age. Based on the carbon storage of several ages within the tree's growth period T1, a regression analysis is performed on several data points to fit the carbon storage curve of the tree in its T1-year life cycle.

[0151] Before executing the tree configuration method in green space based on multi-objective optimization, construct a plant database, objective function, and carbon sequestration curve. If it is a tree, measure and collect the diameter at breast height, crown width data of dominant tree species and trees of several ages, as well as the tree planting unit price, and subsequently maintain the average unit price every year; the diameter of the trunk 1.3m above the ground is the diameter at breast height, and the average width of the tree in the north-south and east-west directions is the crown width. Fit the regression curves of the diameter at breast height, crown width and age of various types of trees based on several data points to simulate the relationship between the diameter at breast height, crown width and age of trees.

[0152] A plant database for the method provided in the embodiment of the present application is established to store data such as carbon sequestration, age, crown width, unit price of seedlings, construction cost, and maintenance cost of garden trees of various ages.

[0153] The first-level data indicators of the design database are divided into three first-level indicators according to different functions, namely basic characteristics, ornamental characteristics, and economic costs. Basic characteristics include tree age, breast diameter, crown width, and carbon sequestration; ornamental characteristics include flowering trees, colorful leaf trees, and fruit trees; economic costs include seedling unit price, planting construction costs, and later maintenance costs.

[0154] The carbon sequestration is selected as an objective function of the green space effect; considering that different green areas have different landscape features, the tree planting area is selected as another objective function; considering the cost requirements of the builders, the maintenance cost of the trees is selected as another objective function. Finally, three objective functions are constructed to achieve multi-objective optimization.

[0155] Corresponding to the above method embodiment, Figure 4 A schematic diagram of a tree configuration device in a green space based on multi-objective optimization is shown. Figure 4 As shown, the tree configuration device in a green space based on multi-objective optimization includes:

[0156] The configuration module 410 is used to repeatedly perform the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each tree in different ages among multiple trees; calculate the total carbon sink amount, the total maintenance cost and the total vertical projection area of ​​all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the total vertical projection area of ​​the trees meet the preset constraint conditions, determine the tree combination as a tree configuration scheme, and the tree configuration scheme indicates the type, quantity and age of the trees configured for the green space;

[0157] The optimization module 420 is used to take the carbon sink amount, maintenance cost and vertical projection area of ​​the trees as optimization targets, perform multi-objective optimization of multiple tree configuration schemes according to the three target values, and obtain an optimized tree configuration scheme.

[0158] In another possible implementation, the optimization module 420 takes carbon sink, maintenance cost and vertical projection area of ​​trees as optimization targets, performs multi-objective optimization of multiple tree configuration schemes according to the three target values, and obtains an optimized tree configuration scheme, including:

[0159] The initial population is formed with multiple tree configuration schemes;

[0160] For the population, the following steps are executed cyclically until the loop termination condition is met: the population is used as the parent population, and the parent population is subjected to crossover and mutation operations to generate a child population; the parent population and the child population are merged to obtain a merged population; the merged population is subjected to fast non-dominated sorting to determine the non-dominated level corresponding to each tree configuration scheme; for each tree configuration scheme, the nearest reference point of the tree configuration scheme association distance is determined according to the three corresponding target values, and the distance between the tree configuration scheme and the nearest reference point is determined as the reference distance of the tree configuration scheme, wherein the reference point is composed of the reference value of the optimization target; according to the non-dominated level, the nearest reference point and the reference distance, at least part of the tree configuration schemes are selected from the merged population to obtain the next generation population;

[0161] The tree configuration scheme in the next generation population obtained for the last time is determined as the optimized tree configuration scheme.

[0162] In another possible implementation, the optimization module 420 associates the nearest reference point with each tree configuration scheme according to the three corresponding target values, and determines the distance between the nearest reference point and the nearest reference point as the reference distance of the tree configuration scheme, including:

[0163] For each tree configuration scheme, a target vector of the tree configuration scheme is generated based on the three target values ​​of the tree configuration scheme; from multiple reference points, the reference point with the smallest Euclidean distance to the target vector is determined to obtain the nearest reference point of the tree configuration scheme, and the Euclidean distance between it and the nearest reference point is determined as the reference distance of the tree configuration scheme.

[0164] In another possible implementation, the reference point is generated as follows:

[0165] For each optimization target, the maximum value and the minimum value of the optimization target are found from the pre-built plant database, and the maximum value and the minimum value are respectively multiplied by the population size N to obtain the maximum value and the minimum value of the value range of the optimization target;

[0166] Divide the value range of each optimization target into N / 2 parts at equal intervals to obtain N / 2 sub-value ranges;

[0167] Calculate the coordinates of the u-th reference point , where the coordinates of the u-th reference point on the j-th objective function are The calculation of is as follows:

[0168] ;

[0169] in, is the maximum value of the j-th optimization target in the u-th sub-value range, is the minimum value of the j-th optimization objective in the u-th sub-value range, the value of u is greater than zero and less than or equal to N / 2, and j is a positive integer less than or equal to 3.

[0170] In another possible implementation, the optimization module 420 selects at least part of the tree configuration schemes from the merged population according to the non-dominated level, the nearest reference point, and the reference distance, including:

[0171] Select n tree configurations from the first L-1 non-dominated levels, where L is an integer greater than 1 and n is a positive integer;

[0172] If n is less than the population size N, the reference point with the smallest number of microhabitats is selected as the candidate reference point from the reference points associated with the n tree configuration schemes, where the number of microhabitats in the reference point is the number of schemes associated with the reference point;

[0173] According to the alternative reference points, (Nn) tree configuration solutions are selected from the Lth non-dominated level.

[0174] In another possible implementation, the optimization module 420 selects (Nn) tree configuration schemes from the Lth non-dominated level according to the candidate reference points, including:

[0175] If the number of microhabitats of the alternative reference point is zero, select the tree configuration scheme with the closest distance to the alternative reference point from the Lth non-dominated level; if the number of microhabitats of the alternative reference point is not zero, select the tree configuration scheme associated with the alternative reference point from the Lth non-dominated level;

[0176] Until (Nn) tree configuration schemes are selected from the Lth non-dominated level.

[0177] In another possible implementation, after obtaining the optimized tree configuration scheme, the optimization module 420 also standardizes the three target values ​​of the three optimization targets of carbon sequestration, maintenance cost and vertical projection area of ​​the tree for each optimized tree configuration scheme to obtain three standardized target values;

[0178] Calculate the jth standardized target value in the i-th optimized tree configuration scheme The proportion in the corresponding optimization target , the formula is as follows:

[0179] = ;

[0180] Based on the weight of the jth optimization objective , calculate the entropy value of the jth optimization objective , the formula is as follows:

[0181] ;

[0182] Based on the entropy value of the jth optimization target , calculate the weight of the jth optimization objective , the formula is as follows:

[0183] ;

[0184] For each optimized tree configuration scheme, the weighted values ​​of the three target values ​​are calculated according to the three weights corresponding to the three optimization objectives, and the evaluation score of the optimized tree configuration scheme is obtained;

[0185] According to the evaluation scores, the best tree configuration scheme is determined from the optimized tree configuration schemes;

[0186] Where N is the number of optimized tree configuration schemes.

[0187] In another possible implementation, after obtaining the optimized tree configuration scheme, the optimization module 420 further calculates the weighted values ​​of the three target values ​​for each optimized tree configuration scheme according to the weights preset by the builder to obtain the evaluation score of the optimized tree configuration scheme;

[0188] According to the evaluation scores, the optimal tree configuration plan is determined from the optimized tree configuration plans.

[0189] In another possible implementation, the constraints include:

[0190] 0.3 degree of depression 0.8, the closure degree is the ratio of the sum of the vertical projection areas of trees to the total green area of ​​green land;

[0191] Planting area of ​​evergreen and deciduous trees Total green area; 0 Evergreen tree planting area 0.3× total vertical projection area of ​​trees; planting area of ​​flowering trees, colorful leaf trees and fruit trees The planting area of ​​evergreen trees and deciduous trees, among which the planting area or planting proportion of flowering trees, colorful leaf trees and fruit-viewing trees shall be set by the builders respectively.

[0192] It should be noted that the device for configuring trees in green space based on multi-objective optimization in this embodiment is used to implement the corresponding method for configuring trees in green space based on multi-objective optimization in the aforementioned method embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0193] Figure 5 is a schematic block diagram of an electronic device provided in an embodiment of the present application. The specific embodiment of the present application does not limit the specific implementation of the electronic device. For example, the electronic device may be a server or a terminal. Figure 5 As shown, the electronic device may include: a processor (processor) 502, a communication interface (Communications Interface) 504, a memory (memory) 506, and a communication bus 508. Among them:

[0194] The processor 502 , the communication interface 504 , and the memory 506 communicate with each other via a communication bus 508 .

[0195] The communication interface 504 is used to communicate with other electronic devices or servers.

[0196] The processor 502 is used to execute the program 510, and specifically can execute the relevant steps in any of the aforementioned embodiments of the tree configuration method in the green space based on multi-objective optimization.

[0197] Specifically, the program 510 may include program codes, which include computer operation instructions.

[0198] The processor 502 may be a CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0199] RISC-V is an open source instruction set architecture based on the Reduced Instruction Set (RISC) principle. It can be applied to various aspects such as single-chip microcomputers and FPGA chips. Specifically, it can be applied to the fields of IoT security, industrial control, mobile phones, personal computers, etc., and because it is designed with small, fast, and low-power realities in mind, it is particularly suitable for modern computing devices such as warehouse-scale cloud computers, high-end mobile phones, and tiny embedded systems. With the rise of artificial intelligence Internet of Things (AIoT), the RISC-V instruction set architecture has also received more and more attention and support, and is expected to become the next generation of widely used CPU architecture.

[0200] The computer operation instructions in the embodiment of the present application may be computer operation instructions based on the RISC-V instruction set architecture, and correspondingly, the processor 502 may be designed based on the RISC-V instruction set. Specifically, the processor chip in the electronic device provided in the embodiment of the present application may be a chip designed using the RISC-V instruction set, and the chip may execute executable code based on the configured instructions, thereby realizing the tree configuration method in the green space based on multi-objective optimization in the above embodiment.

[0201] The memory 506 is used to store the program 510. The memory 506 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0202] The program 510 can be specifically used to enable the processor 502 to execute the method for configuring trees in a green space based on multi-objective optimization in any of the aforementioned embodiments.

[0203] The specific implementation of each step in program 510 can refer to the corresponding steps and corresponding descriptions in any of the above-mentioned embodiments of the method for configuring trees in green space based on multi-objective optimization, and will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described equipment and modules can refer to the corresponding process description in the above-mentioned method embodiment, and will not be repeated here.

[0204] The present application also provides a computer-readable storage medium storing instructions for causing a machine to execute the method for arbor configuration in a green space based on multi-objective optimization as described herein. Specifically, a system or device equipped with a storage medium may be provided, on which a software program code for implementing the functions of any of the above-mentioned embodiments is stored, and a computer (or CPU or MPU) of the system or device reads out and executes the program code stored in the storage medium.

[0205] In this case, the program code read from the storage medium itself can implement the function of any one of the above embodiments, so the program code and the storage medium storing the program code constitute part of the present application.

[0206] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0207] An embodiment of the present application also provides a computer program product, including computer instructions, which instruct a computing device to perform any corresponding operation in the above-mentioned multiple method embodiments.

[0208] It should be pointed out that, according to the needs of implementation, the various components / steps described in the embodiments of the present application can be split into more components / steps, or two or more components / steps or partial operations of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of the present application.

[0209] The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or implemented as a computer code originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded through a network and to be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a special-purpose processor or programmable or special-purpose hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown here, the execution of the code converts the general-purpose computer into a special-purpose computer for executing the method shown here.

[0210] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present application.

[0211] The above implementation methods are only used to illustrate the embodiments of the present application, and are not limitations on the embodiments of the present application. Ordinary technicians in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions also belong to the scope of the embodiments of the present application. The scope of patent protection of the embodiments of the present application should be limited by the claims.

Claims

1. A method for arbor configuration in green space based on multi-objective optimization, characterized in that: The method comprises: Repeat the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each type of tree in different ages; calculate the total carbon sink amount, the total maintenance cost and the total vertical projection area of ​​all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the total vertical projection area of ​​the trees meet the preset constraints, determine the tree combination as the tree configuration scheme, and the tree configuration scheme indicates the type, quantity and age of the trees configured for the green space; forming an initial population using the multiple tree configuration schemes; For the population, the following steps are executed in a loop until the loop termination condition is met: taking the population as the parent population, performing crossover operation and mutation operation on the parent population to generate a child population; merging the parent population and the child population to obtain a merged population; performing fast non-dominated sorting on the merged population to determine the non-dominated level corresponding to each tree configuration scheme; for each tree configuration scheme, according to its corresponding three target values, a reference point with the closest associated distance to the tree configuration scheme is selected, and its distance from the nearest reference point is determined as the reference distance of the tree configuration scheme, wherein the reference point is composed of the reference value of the optimization target, and the reference point is generated in the following manner: for each of the optimization targets, the maximum value and the minimum value of the optimization target are searched from a pre-constructed plant database, and the maximum value and the minimum value are respectively multiplied by the population size N to obtain the maximum value and the minimum value of the value range of the optimization target; the value range of each of the optimization targets is divided into N / 2 parts at equal intervals to obtain N / 2 sub-value ranges; the coordinates of the u-th reference point are calculated , where the coordinates of the u-th reference point on the j-th objective function are The calculation of is as follows: ; The coordinates of the u-th reference point are normalized to obtain normalized coordinates, wherein the normalized coordinates of the u-th reference point in the j-th objective function are : ; Until N / 2 reference points are generated; among them, is the maximum value of the j-th optimization target in the u-th sub-value range, is the minimum value of the j-th optimization target in the u-th sub-value range, is the maximum value of the jth optimization objective in the value range, the value of u is greater than zero and less than or equal to N / 2, and j is a positive integer less than or equal to 3; adopting a niche retention strategy, according to the non-dominated level, the nearest reference point and the reference distance, selecting at least part of the tree configuration scheme from the merged population to obtain the next generation population; The tree configuration scheme in the next generation population obtained for the last time is determined as the optimized tree configuration scheme.

2. The method according to claim 1, characterized in that For each of the tree configuration schemes, associating the tree configuration scheme with a reference point closest to the tree configuration scheme according to the three target values ​​corresponding to the tree configuration scheme, and determining the distance between the tree configuration scheme and the nearest reference point as the reference distance of the tree configuration scheme, including: For each of the tree configuration schemes, a target vector of the tree configuration scheme is generated based on the three target values ​​of the tree configuration scheme; from multiple reference points, a reference point with the smallest Euclidean distance to the target vector is determined to obtain the nearest reference point of the tree configuration scheme, and the Euclidean distance between the reference point and the nearest reference point is determined as the reference distance of the tree configuration scheme.

3. The method according to claim 1, characterized in that The selecting at least part of the tree configuration schemes from the merged population according to the non-dominated level, the nearest reference point and the reference distance comprises: Select n tree configurations from the first L-1 non-dominated levels, where L is an integer greater than 1 and n is a positive integer; If n is smaller than the population size N, the reference point with the smallest number of microhabitats is selected as the candidate reference point from the reference points associated with the n tree configuration schemes, wherein the number of microhabitats of the reference point is the number of schemes associated with the reference point; According to the candidate reference points, (Nn) tree configuration solutions are selected from the Lth non-dominated level.

4. The method according to claim 3, characterized in that: The step of selecting (Nn) tree configuration schemes from the Lth non-dominated level according to the candidate reference points includes: If the number of microhabitats of the candidate reference point is zero, select the tree configuration scheme with the shortest distance to the candidate reference point from the Lth non-dominated level; if the number of microhabitats of the candidate reference point is not zero, select the tree configuration scheme associated with the candidate reference point from the Lth non-dominated level; Until the (Nn) tree configuration schemes are selected from the Lth non-dominated level.

5. The method according to claim 1, characterized in that After determining the optimized tree configuration scheme, the method further includes: For each of the optimized tree configuration schemes, standardize the three target values ​​of the three optimization targets of the carbon sequestration amount, the maintenance cost and the vertical projection area of ​​the tree to obtain three standardized target values; Calculate the jth standardized target value in the i-th optimized tree configuration scheme The proportion in the corresponding optimization target , the formula is as follows: = ； Based on the weight of the jth optimization objective , calculate the entropy value of the jth optimization objective , the formula is as follows: ； Based on the entropy value of the jth optimization objective , calculate the weight of the jth optimization objective , the formula is as follows: ； For each of the optimized tree configuration schemes, according to the three weights corresponding to the three optimization objectives, the weighted values ​​of the three objective values ​​are calculated to obtain an evaluation score for the optimized tree configuration scheme; Determining the best tree configuration scheme from the optimized tree configuration schemes according to the evaluation scores; Wherein, N is the number of the optimized tree configuration schemes.

6. The method according to claim 1, characterized in that After determining the optimized tree configuration scheme, the method further includes: For each of the optimized tree configuration schemes, the weighted values ​​of the three target values ​​are calculated according to the weights preset by the builder to obtain an evaluation score for the optimized tree configuration scheme; According to the evaluation scores, the best tree arrangement scheme is determined from the optimized tree arrangement schemes.

7. The method according to claim 1, characterized in that The constraints include: 0.3 degree of depression 0.8, the density is the ratio of the sum of the vertical projection areas of the trees to the total green area of ​​the green space; The planting area of ​​evergreen trees and deciduous trees is ≤ the total green area; 0 Evergreen tree planting area 0.3× total vertical projection area of ​​trees; planting area of ​​flowering trees, colorful leaf trees and fruit trees The planting areas of the evergreen trees and the deciduous trees, among which, the planting areas or planting proportions of the flowering trees, the colorful leaf trees and the fruit-viewing trees are set by the builders respectively.

8. A device for arbor configuration in green space based on multi-objective optimization, characterized in that: The device comprises: A configuration module is used to repeatedly perform the following steps to obtain multiple tree configuration schemes for the green space: generate a tree combination, the tree combination includes individuals of each tree in a plurality of trees at different ages; calculate the sum of carbon sinks, the sum of maintenance costs and the sum of vertical projection areas of all individuals in the tree combination during the entire life cycle of the green space to obtain three target values; if the area-related parameters including the sum of vertical projection areas of the trees meet preset constraints, determine the tree combination as the tree configuration scheme, and the tree configuration scheme indicates the type, quantity and age of the trees configured for the green space; An optimization module, used for forming an initial population using the plurality of tree configuration schemes; The optimization module is used to loop through the following steps for the population until a loop termination condition is met: taking the population as the parent population, performing crossover and mutation operations on the parent population to generate a child population; merging the parent population and the child population to obtain a merged population; performing fast non-dominated sorting on the merged population to determine the non-dominated level corresponding to each tree configuration scheme; for each tree configuration scheme, according to its corresponding three target values, a reference point with the closest associated distance to the tree configuration scheme is selected, and its distance from the nearest reference point is determined as the reference distance of the tree configuration scheme, wherein the reference point is composed of reference values ​​of the optimization target, and the reference point is generated in the following manner: for each of the optimization targets, searching for the maximum and minimum values ​​of the optimization target from a pre-constructed plant database, multiplying the maximum and minimum values ​​by the population size N respectively, to obtain the maximum and minimum values ​​of the value range of the optimization target; dividing the value range of each of the optimization targets into N / 2 parts at equal intervals to obtain N / 2 sub-value ranges; calculating the coordinates of the u-th reference point , where the coordinates of the u-th reference point on the j-th objective function are The calculation of is as follows: ; The coordinates of the u-th reference point are normalized to obtain normalized coordinates, wherein the normalized coordinates of the u-th reference point in the j-th objective function are : ; Until N / 2 reference points are generated; among them, is the maximum value of the j-th optimization target in the u-th sub-value range, is the minimum value of the j-th optimization target in the u-th sub-value range, is the maximum value of the jth optimization objective in the value range, the value of u is greater than zero and less than or equal to N / 2, and j is a positive integer less than or equal to 3; adopting a niche retention strategy, according to the non-dominated level, the nearest reference point and the reference distance, selecting at least part of the tree configuration scheme from the merged population to obtain the next generation population; The optimization module is used to determine the tree configuration scheme in the next generation population obtained last time as the optimized tree configuration scheme.

Images