Resource allocation method of non-dominated sorting genetic algorithm containing heuristic information

Abstract

The invention relates to the field of space resource allocation of old urban relocation optimization, in particular to a resource allocation method of a non-dominated sorting genetic algorithm containing heuristic information. Comprising the steps of obtaining a dwelling quality comprehensive score of a current plot; selecting candidate immigration land parcels; calculating the compensation amount of each candidate immigration plot; and constructing a multi-target relocation optimization model including a compensation amount, introducing heuristic information to correct decision variables, solving the multi-target relocation model by using a non-dominated sorting genetic algorithm, and calculating a final immigration land parcel. According to the method, the candidate immigration land parcels and the relocation compensation strategy based on the land parcel comprehensive score difference are designed, and the relocation fairness and acceptability are guaranteed. Meanwhile, by introducing a non-dominated sorting genetic algorithm containing heuristic information, the multi-target intelligent relocation optimization model is solved, and a high-quality solution can be effectively searched under the condition that the number of dimensions of solution and the search range are remarkably reduced.

Description

Technical Field

[0001] This invention relates to the field of spatial resource allocation optimization for relocation of old urban areas, and particularly to a resource allocation method using a non-dominated sorting genetic algorithm containing heuristic information. Background Technology

[0002] With the continuous advancement of urbanization in my country, a large number of old urban blocks urgently need to be relocated and transformed through spatial resource allocation in order to meet the dual needs of modern urban function upgrading and development.

[0003] As a flexible relocation strategy, relocation and replacement aims to dynamically match relocation and resettlement by guiding residents to plots within the urban area with similar structures and living quality. However, in practical applications, this strategy involves multiple objectives, multiple constraints, and a high-dimensional decision space, typically manifesting as a complex multi-objective combinatorial optimization problem. Its optimization objectives usually include maximizing the total area of ​​vacated courtyards, maximizing the number of adjacent vacated courtyards, and minimizing the number of relocated residents, in order to improve the developer's economic benefits.

[0004] Traditional methods for relocating and replacing residents in old urban areas mostly rely on empirical judgment and lack a systematically designed and reasonable candidate relocation site selection mechanism and renovation compensation plan, which can easily lead to low relocation acceptance or unreasonable compensation costs. Furthermore, the relocation problem is essentially a multi-objective, high-dimensional combinatorial optimization problem. Existing algorithms suffer from high computational costs and frequently get trapped in local optima, making them unsuitable for practical applications. Therefore, there is an urgent need for a reasonable and efficient multi-objective intelligent optimization relocation method to improve decision-making quality and computational efficiency. Summary of the Invention

[0005] This invention discloses a resource allocation method for a non-dominated sorting genetic algorithm containing heuristic information, the specific steps of which are as follows:

[0006] S1. Obtain the comprehensive residential quality score for the current plot of land;

[0007] S2. Based on the comprehensive score of the current residential quality of the plot, select candidate relocation plots;

[0008] S3. Calculate the compensation amount for each candidate relocation plot according to the preset compensation plan;

[0009] S4. Construct a multi-objective relocation optimization model that includes compensation amount, introduce heuristic information to correct decision variables, use a non-dominated sorting genetic algorithm to solve the multi-objective relocation model, and calculate the final relocation site.

[0010] Furthermore, the comprehensive residential quality score of the current plot is obtained in S1, and the specific steps are as follows:

[0011] S11. Assign the current plot shape H and orientation D, and calculate the area percentage P;

[0012] S12. Based on the hierarchical progressive entropy method and TOPSIS, construct the evaluation factors for the current residential quality of the land parcel;

[0013] S13. Based on the analytic hierarchy process, construct an evaluation of the current residential quality of the land parcel.

[0014] Furthermore, in S11, the current plot shape H and orientation D are assigned, and the area percentage P is calculated. The specific steps are as follows:

[0015] S111. Based on the overall outline features of the courtyard, plot k is divided into different morphological categories and assigned discrete scores to form variable shape scores H.

[0016] S112. Considering sunlight and spatial comfort, a score is assigned based on the orientation of plot k.

[0017] S113, According to formula P k =S k / A k Calculate the area percentage.

[0018] Furthermore, in S12, the evaluation factors for the residential quality of the current plot are constructed, and the specific steps are as follows:

[0019] S121. Quality evaluation factors include orientation D, plot area S, layout L, living comfort C, and psychological price M. Among them, the evaluation of layout L considers plot area S, plot area ratio P, and courtyard shape H; living comfort C considers orientation D and layout L; psychological price M considers plot area S and living comfort C. The three are calculated using the entropy method and TOPSIS.

[0020] S122. Determining the attribute entropy value E using the entropy method. j and attribute weight w j ;

[0021] S123, TOPSIS calculates the specific values ​​of each factor.

[0022] Furthermore, in S13, a residential quality assessment of the current plot is constructed, with the following specific steps:

[0023] S131. The five evaluation factors, namely plot area S, orientation D, living comfort score C, plot layout score L, and psychological price score M, are used as secondary indicators to obtain the comprehensive score Z of the plot.

[0024] S132. Rank the importance of the evaluation factors, construct a pairwise comparison judgment matrix among the 5 factors, and calculate the weight coefficient ω;

[0025] S133. Calculate the consistency ratio CR. If the test result CR < 0.1, the consistency test is passed.

[0026] S134. Calculate the current residential quality evaluation of the land parcel. The specific formula is as follows:

[0027] Z k =ω S .S k +ω D .D k +ω C .C k +ω L .L k +ω M .M k .

[0028] Further, candidate relocation sites are selected in S2, and the specific steps are as follows:

[0029] For each relocation plot k out Its overall score is For candidate relocation site k in Its overall score is Designed candidate relocation site k in The following conditions must be met:

[0030]

[0031] In the formula, a is the lower limit parameter and b is the upper limit parameter.

[0032] Furthermore, in S3, the compensation amount for each candidate relocation plot is calculated using the following method:

[0033] Area compensation plan:

[0034]

[0035] Lighting compensation solution:

[0036]

[0037] Repair and compensation plan:

[0038]

[0039] Where, r 3,max The maximum repair compensation is determined by the developer.

[0040] Furthermore, a multi-objective relocation optimization model, including compensation amounts, is constructed in S4. The specific steps are as follows:

[0041] S411, the plot of land that needs to be relocated The corresponding candidate relocation sites are listed as decision variables. in, A collection of residential plots. Let be the set of vacant land parcels. To simplify the expression of variables, let

[0042] S412, Objective Function 1: Maximize the total area of ​​the complete courtyard vacated after relocation. Among them, A k Let K be the area of ​​the courtyard to which plot k belongs;

[0043] S413, Objective Function 2: Maximize the number of complete adjacent courtyards freed up after relocation. Where {L near} indicates the grouping of adjacent courtyards after relocation;

[0044] S414, Objective Function 3: Minimize the number of relocated households

[0045] S415. Minimize the overall objective function: min F = min(-α1f1 - α2f2 + α3f3), where α i The objective function weights are the weights of the i-th objective function.

[0046] S416, Constraint 1: Each resident may relocate a maximum of times;

[0047] S417, Constraint 2: Each plot of land can accommodate a maximum number of households;

[0048] S418, Constraint 3: Area of ​​the relocation plot l Not exceeding the original plot k out area The preset multiple;

[0049] S419, Constraint 4: The total relocation cost shall not exceed ξ max .

[0050] Furthermore, heuristic information is introduced in S4 to modify the decision variables. The specific steps are as follows:

[0051] S421. Define the three indicators in the heuristic information: the proportion of residential courtyard area P. k Adjacent courtyards ratings s near,k Empty courtyard area A e,k ;

[0052] S422. Based on the heuristic information, through

[0053]

[0054] Conditional modification of relocation plot kout , making in, The set of residential plots after correction, with th1 and th2 being the set thresholds;

[0055] pass

[0056]

[0057] Conditional modification of candidate relocation site k in , making in, This is the set of vacant residential plots after correction, and th3 is the set threshold.

[0058] Furthermore, in S4, a non-dominated sorting genetic algorithm is used to solve the multi-objective relocation model. The specific steps are as follows:

[0059] S431. Initialize the population, set the population size, and the number of generations;

[0060] S432. Divide the population into levels and assign virtual fitness. If all individuals have completed the leveling, proceed to step S434; otherwise, proceed to step S433.

[0061] S433. Identify the current non-dominated set of individuals F i Specify the virtual fitness function to regulate the density distribution of the solution, and execute step S432;

[0062] S434, Population reproduction, performing crossover and mutation operations on individuals;

[0063] S435. Repeat steps S432 to S434 above until gen ≥ gen. max If the value is less than the tolerance of the objective function, the evolution is terminated, and the final solution set is output.

[0064] Due to the adoption of the above technical solutions, the present invention has the following beneficial effects:

[0065] This invention designs candidate relocation sites and relocation compensation strategies based on the differences in comprehensive site scores, ensuring the fairness and acceptability of relocation. Simultaneously, by introducing a non-dominated sorting genetic algorithm with heuristic information, a multi-objective intelligent relocation optimization model is solved, effectively searching for high-quality solutions while significantly reducing the dimensionality and search range.

[0066] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0067] The accompanying drawings of this invention are described below.

[0068] Figure 1 This is a comprehensive scoring chart for the residential quality of the land parcel.

[0069] Figure 2 Assign a map to the shape of the courtyard belonging to the plot.

[0070] Figure 3 A diagram illustrating the modification of decision variables based on heuristic information.

[0071] Figure 4 This is a histogram showing the proportion of residential courtyard area.

[0072] Figure 5 This is a histogram of the area of ​​the empty courtyard.

[0073] Figure 6 The flowchart shows the process of using a non-dominated sorting genetic algorithm to solve for the optimal relocation decision.

[0074] Figure 7 Simulation diagram for solving the optimal relocation decision using a non-dominated sorting genetic algorithm.

[0075] Figure 8 The diagram shows the solution results for the optimal relocation decision.

[0076] Figure 9 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0077] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0078] A resource allocation method for a non-dominated sorting genetic algorithm with heuristic information, such as Figure 9 As shown, the specific steps are as follows:

[0079] S1. Obtain the overall residential quality score for the current plot of land, such as... Figure 1 As shown.

[0080] S11. Data preprocessing: Assigning scores to the shape H and orientation D of the land parcels, and calculating the area percentage P of each parcel. Further steps include:

[0081] S111, dataset source: Problem B of the 15th MathorCup Mathematical Applications Challenge;

[0082] S112, Calculate the plot shape assignment H k The scoring criteria are illustrated in the diagram below. Figure 2 As shown: The scoring criteria are as follows: 1 point for extremely irregular courtyard shape; 2 points for partially regular courtyard; 3 points for overall regular courtyard.

[0083] S113, Calculate the orientation score of the land parcel D k Based on the orientation of plot K, the scores are as follows: South / North 3 points; East 2 points; West 1 point;

[0084] S114, Calculate the area percentage P k =S k / A k , of which S k Let A be the area of ​​plot k. k Let K be the area of ​​the courtyard to which plot k belongs.

[0085] S12. Using the hierarchical progressive entropy method combined with the TOPSIS method, establish evaluation factors for the residential quality of land parcels, further including:

[0086] S121. Select the evaluation factors for the residential quality of the plot, which consist of five variables: orientation (D), plot area (S), layout (L), residential comfort (C), and psychological price point (M). Further, these include:

[0087] S1211, Evaluation Factor 1 (Orientation D) can be expressed as:

[0088]

[0089] S1212, Evaluation Factor 2, Plot Area S;

[0090] S1213, Evaluation Factor 3, Layout L:L k =TOPSIS(ω1S) k ,ω2P k ,ω3H k );

[0091] S1214, Evaluation Factor 4, Living Comfort C: C k =TOPSIS(ω4D) k ,ω5L k );

[0092] S1215, Evaluation Factor 5, Psychological Price Level M: M k =TOPSIS(ω6C k ,ω7S k ).

[0093] S122. Determining attribute weights using the entropy method:

[0094] S123, TOPSIS Measurement Factors: Calculate and rank relative proximity. This yields the factor values.

[0095] S13. Using the analytic hierarchy process (AHP), weights are determined for each evaluation factor to construct a residential quality evaluation system for the land parcel, which further includes:

[0096] S131. Evaluation Factor Setting: Five factors are selected as secondary evaluation indicators: plot area S, orientation D, living comfort score C, plot layout score L, and psychological price score M. An evaluation system is constructed to calculate the comprehensive score Z of the living quality of the plot.

[0097] S132. Judgment Matrix Construction and Weight Coefficient Calculation: Based on empirical rules, the importance of each evaluation factor is ranked, and the priority is determined as follows: Plot Area S > Orientation D > Residential Comfort C = Plot Layout L = Psychological Price M. Based on this, a pairwise comparison judgment matrix is ​​constructed between the five factors. The weight coefficient ω of each evaluation factor is calculated using the eigenvalue method, as shown in the table below:

[0098] Z S D C L M ω S 1 1 / 1.2 1.1 1.1 1.1 0.2307 D 1.2 1 1.2 1.2 1.2 0.2036 C 1 / 1.1 1 / 1.2 1 1 1 0.1886 L 1 / 1.1 1 / 1.2 1 1 1 0.1886 M 1 / 1.1 1 / 1.2 1 1 1 0.1886

[0099] S133. Consistency test: Calculate the consistency ratio CR = 0.0002. The test result CR < 0.1, the consistency test is passed, the logic is reasonable, and the analysis can continue.

[0100] S134. Based on the obtained weighting coefficients, a linear weighting method is used to comprehensively score the land parcels and calculate the score Z. k =0.2307S k +0.2036D k +0.1886C k +0.1886L k +0.1886M k As shown in the table below:

[0101]

[0102] S2. Based on the comprehensive score of the current residential quality of the plot, select candidate relocation plots.

[0103] S21. Selection of Candidate Relocation Sites: When designing the relocation compensation plan for residents, the rationality and feasibility of the relocation are assessed based on the comprehensive score of each site. Specifically, for each relocation site k... out Its overall score is For candidate relocation site k in Its overall score is Designed candidate relocation site k in The following conditions must be met;

[0104]

[0105] S22. Candidate Relocation Site Selection Analysis: This scoring tolerance range ensures that the difference in residential scores before and after relocation does not exceed 10%, i.e.

[0106] S3. Calculate the compensation amount for each candidate relocation plot according to the preset compensation plan.

[0107] S31, Area Compensation Plan is as follows:

[0108] S32, Lighting compensation scheme is as follows:

[0109] S33, Repair and Compensation Plan Where r 3,max =20, the unit is 10,000 yuan, that is, the upper limit of the budget for the renovation of the relocation site is set at 200,000 yuan.

[0110] S34. The obtained relocation compensation plan is shown in the table below:

[0111]

[0112] S4. Construct a multi-objective relocation optimization model that includes compensation amount, introduce heuristic information to correct decision variables, use a non-dominated sorting genetic algorithm to solve the multi-objective relocation model, and calculate the final relocation site.

[0113] S41. Construct a multi-objective relocation optimization model and introduce heuristic information to correct decision variables.

[0114] S411, the plot of land that needs to be relocated The corresponding candidate relocation sites are listed as decision variables. in, A collection of residential plots. Let be the set of vacant land parcels. To simplify the expression of variables, let

[0115] S412. Establish objective function 1: Maximize the total area of ​​the complete courtyard vacated after relocation. in, This indicates the area of ​​the courtyard belonging to the relocation plot, in square meters (m²). 2 ;

[0116] S413. Establish objective function 2: Maximize the number of complete adjacent courtyards freed up after relocation. in, Indicates the courtyard number to which the relocation plot belongs, {L near} indicates the grouping of adjacent courtyards after relocation;

[0117] S414. Establish objective function 3: Minimize the number of relocated households. To minimize the costs associated with relocation and the uncertainty of whether the relocation will be successful;

[0118] S415. Establish the overall objective function: Minimize the overall objective function minF = min(-f1 / β1 - 1.5f2 / β2 + 0.2f3 / β3), where β i Let β1 = 43609, β2 = 107, and β3 = 113 be the maximum values ​​of the i-th objective function.

[0119] S416. Establish constraint 1: Each resident can relocate at most once: For each resident, their relocation status must satisfy p. i ∈{0,1}, where p i =1 indicates that the resident has already moved. Cannot move out or be moved in; p i =0 indicates that there has been no relocation. You can move out;

[0120] S417. Establish constraint 2: Each plot can accommodate at most one household: For each vacant plot, at most one household can be accommodated. For plot k, this can be represented as...

[0121] S418. Establish constraint 3: Area of ​​the relocation plot. Not exceeding the original plot k out area 1.3 times, that is The unit is m 2 ;

[0122] S419. Establish constraint 4: The developer can provide a maximum of ξ max =The relocation cost is 20 million yuan, therefore, the relocation plan should be formulated to ensure that the relocation cost is less than the minimum planned cost. Further includes:

[0123] S4191, Communication costs are expressed as the communication and management costs incurred in coordinating with each household.

[0124] S4192, The area loss is given by T1 = (9 + 8 / 12)·365, where T1 is the number of days and θ is the rent. k Note: The rent for the east and west wing rooms is 8 x 10. -4 10,000 yuan / m 2 / day, the north and south wing rooms are 15×10 -4 10,000 yuan / m 2 / day. The rent for a completely vacant courtyard is 30 x 10 due to overall efficiency. -4 10,000 yuan / m 2 / day. If multiple vacant courtyards are adjacent to each other, the rental value is 1.2 times the original rent;

[0125] S4193, The time loss is represented by T2 = 4 / 12·365, where the unit is days.

[0126] S42. For the multi-objective relocation optimization model, heuristic information is introduced to modify the decision variables.

[0127] S421. Establish three metrics in the heuristic information, further including:

[0128] S4211, Percentage of residential courtyard area (P) k =S k / A k ;

[0129] S4212, Adjacent courtyard ratings s near,k =g(yard) i γ close );

[0130] S4213, Area A of the empty courtyard e,k =A k ,(if L k (is empty);

[0131] S422. Introduce heuristic information to control the solution space and modify decision variables, such as... Figure 3 As shown, further includes:

[0132] S4221, Reduce the solution dimension (threshold th1 = 0.4, set the standard as follows) Figure 4 As shown: via Conditions were modified to allow the relocation of the land parcels to

[0133] S4222. Reduce the search range, set the threshold th3 = 600 as follows: Figure 5 As shown: via Conditional modification of candidate relocation sites makes

[0134] S43. Solve the multi-objective relocation optimization model using a non-dominated sorting genetic algorithm. For example... Figure 6 As shown.

[0135] S431. Initialize the population, set the population size, and set the number of generations to gen = 0;

[0136] S432. Divide the population into levels and assign virtual fitness. If all individuals have completed the leveling, proceed to step S434; otherwise, proceed to step S433.

[0137] S433. Identify the current non-dominated set of individuals F i Specify the virtual fitness function to regulate the density distribution of the solution, and execute step S432;

[0138] S434. During population reproduction, a tournament selection operator is used to construct the breeding pool; a single-point crossover mechanism is used for crossover operations; and a uniform mutation operator is used for mutation operations.

[0139] S435. Repeat steps S432 to S434 above. In each generation, the parent generation and the offspring generation are merged into a joint population R. t =P t ∪Q t Then, by using non-dominated sorting, the individuals at the top are selected to form a new generation population P. t+1 And let gen = gen + 1. The iteration process continues until gen ≥ gen. max If the value is less than the tolerance of the objective function, the termination condition is met, and the final solution set is output.

[0140] S436. The specific parameters for the non-dominated sorting genetic algorithm are shown in the table below:

[0141]

[0142] S437. Solving a multi-objective relocation optimization model using a non-dominated sorting genetic algorithm with heuristic information. The fitness variation with iteration number during the simulation of the solution process is as follows: Figure 7 As shown, the number of relocated households is 23, and the solution is visualized as follows. Figure 8 As shown in the table below:

[0143]

[0144]

[0145] In this embodiment, heuristic information is introduced to control the solution space in the traditional non-dominated sorting genetic algorithm, modifying decision variables and reducing the solution dimension and search range, as shown in the table below:

[0146]

[0147] In this embodiment, heuristic information is introduced to control the solution space in the traditional non-dominated sorting genetic algorithm. The performance before and after the introduction of heuristic information is compared in the table below to verify whether the solution performance remains essentially unchanged.

[0148]

[0149] In this embodiment, the quality and effectiveness of the relocation decision are verified based on the relocation decision obtained by the non-dominated sorting genetic algorithm containing heuristic information. The performance comparison before and after the relocation is shown in the table below:

[0150]

[0151] In this embodiment, candidate relocation sites and renovation compensation schemes that take into account both developer profits and resident acceptance are constructed, which can improve the feasibility and acceptability of the relocation plan. Furthermore, based on the traditional non-dominated sorting genetic algorithm, heuristic information is introduced to modify decision variables, realizing efficient solution of multi-objective relocation optimization model. Compared with the traditional non-dominated sorting genetic algorithm, the solution dimension and search range are significantly reduced, while maintaining the solution performance basically unchanged. At the same time, the optimal relocation decision obtained by this method achieves significant improvements in multiple indicators such as complete courtyard area, adjacent courtyard area, number of complete courtyards, number of adjacent courtyards, and rental income compared to no relocation decision.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A resource allocation method for a non-dominated sorting genetic algorithm containing heuristic information, characterized in that, The specific steps are as follows: S1. Obtain the comprehensive residential quality score for the current plot of land; S2. Based on the comprehensive score of the current residential quality of the plot, select candidate relocation plots; S3. Calculate the compensation amount for each candidate relocation plot according to the preset compensation plan; S4. Construct a multi-objective relocation optimization model that includes compensation amount, introduce heuristic information to correct decision variables, use a non-dominated sorting genetic algorithm to solve the multi-objective relocation model, and calculate the final relocation site.

2. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 1, characterized in that, The specific steps to obtain the comprehensive residential quality score of the current plot in S1 are as follows: S11. Assign the current plot shape H and orientation D, and calculate the area percentage P; S12. Based on the hierarchical progressive entropy method and TOPSIS, construct the evaluation factors for the current residential quality of the land parcel; S13. Based on the analytic hierarchy process, construct an evaluation of the current residential quality of the land parcel.

3. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 2, characterized in that, In S11, the current plot shape H and orientation D are assigned, and the area percentage P is calculated. The specific steps are as follows: S111. Based on the overall outline features of the courtyard, plot k is divided into different morphological categories and assigned discrete scores to form variable shape scores H. S112. Considering sunlight and spatial comfort, a score is assigned based on the orientation of plot k. S113, According to formula P k =S k / A k Calculate the area percentage.

4. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 2, characterized in that, The specific steps for constructing the residential quality evaluation factors for the current land parcel in S12 are as follows: S121. Quality evaluation factors include orientation D, plot area S, layout L, living comfort C, and psychological price M. Among them, the evaluation of layout L considers plot area S, plot area ratio P, and courtyard shape H; living comfort C considers orientation D and layout L; psychological price M considers plot area S and living comfort C. The three are calculated using the entropy method and TOPSIS. S122. Determining the attribute entropy value E using the entropy method. j and attribute weight w j ; S123, TOPSIS calculates the specific values ​​of each factor.

5. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 2, characterized in that, The specific steps for constructing a residential quality assessment for the current land parcel in S13 are as follows: S131. The five evaluation factors, namely plot area S, orientation D, living comfort score C, plot layout score L, and psychological price score M, are used as secondary indicators to obtain the comprehensive score Z of the plot. S132. Rank the importance of the evaluation factors, construct a pairwise comparison judgment matrix among the 5 factors, and calculate the weight coefficient ω; S133. Calculate the consistency ratio CR. If the test result CR < 0.1, the consistency test is passed. S134. Calculate the current residential quality evaluation of the land parcel. The specific formula is as follows: Z k =ω S .S k +oh D .D k +oh C .C k +oh L .L k +oh M .M k 。 6. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 1, characterized in that, The specific steps for selecting candidate relocation sites in S2 are as follows: For each relocation plot k out Its overall score is For candidate relocation site k in Its overall score is Designed candidate relocation site k in The following conditions must be met: In the formula, a is the lower limit parameter and b is the upper limit parameter.

7. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 1, characterized in that, The compensation amount for each candidate relocation plot is calculated in S3 using the following method: Area compensation plan: Lighting compensation solution: Repair and compensation plan: Where, r 3,max The maximum repair compensation is determined by the developer.

8. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 1, characterized in that, The following are the specific steps for constructing a multi-objective relocation optimization model in S4, which includes compensation amounts: S411, the land parcels that need to be relocated The corresponding candidate relocation sites are listed as decision variables. Where S represents the set of residential land parcels. Let be the set of vacant land parcels. To simplify the expression of variables, let S412, Objective Function 1: Maximize the total area of ​​the complete courtyard vacated after relocation. Among them, A k Let K be the area of ​​the courtyard to which plot k belongs; S413, Objective Function 2: Maximize the number of complete adjacent courtyards freed up after relocation. Where {L near } indicates the grouping of adjacent courtyards after relocation; S414, Objective Function 3: Minimize the number of relocated households S415. Minimize the overall objective function: min F = min(-α1f1 - α2f2 + α3f3), where α i The objective function weights are the weights of the i-th objective function. S416, Constraint 1: Each resident may relocate a maximum of times; S417, Constraint 2: Each plot of land can accommodate a maximum number of households; S418, Constraint 3: Area of ​​the relocation plot l Not exceeding the original plot k out area The preset multiple; S419, Constraint 4: The total relocation cost shall not exceed ξ max .

9. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 8, characterized in that, In S4, heuristic information is introduced to modify the decision variables. The specific steps are as follows: S421. Define the three indicators in the heuristic information: the proportion of residential courtyard area P. k Adjacent courtyards ratings s near,k Empty courtyard area A e,k ; S422. Based on the heuristic information, through Conditional modification of relocation plot k out , making in, The set of residential plots after correction, with th1 and th2 being the set thresholds; pass Conditional modification of candidate relocation site k in , making in, This is the set of vacant residential plots after correction, and th3 is the set threshold.

10. The resource allocation method of the non-dominated sorting genetic algorithm containing heuristic information as described in claim 9, characterized in that, S4 uses a non-dominated sorting genetic algorithm to solve the multi-objective relocation model. The specific steps are as follows: S431. Initialize the population, set the population size, and the number of generations; S432. Divide the population into levels and assign virtual fitness. If all individuals have completed the leveling, proceed to step S434; otherwise, proceed to step S433. S433. Identify the current non-dominated set of individuals F i Specify the virtual fitness function to regulate the density distribution of the solution, and execute step S432; S434. Population reproduction, performing crossover and mutation operations on individuals; S435. Repeat steps S432 to S434 above until gen ≥ gen. max If the value is less than the tolerance of the objective function, the evolution is terminated, and the final solution set is output.

Images