A Decision-Making Method and System for Ecological Protection of Rock Slopes Based on Multi-Objective Optimization

By employing a multi-objective optimization method, and combining geological, hydrological, and vegetation parameters for quantitative assessment and multi-dimensional coupling adaptability analysis, the problem of insufficient quantitative basis in traditional ecological protection schemes for rock slopes is solved. This achieves synergistic optimization among cost, construction period, safety, and ecological benefits, and provides a scientific basis for decision-making.

CN123116790APending Publication Date: 2026-11-17RES INST OF HIGHWAY MINIST OF TRANSPORT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611230614.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-14
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Traditional ecological protection schemes for rock slopes lack systematic quantitative assessment basis, making it difficult to accurately match the different characteristics of geological structure, hydrological seepage and vegetation growth conditions of different slope sections. Furthermore, they fail to comprehensively balance engineering economics and ecological benefits, and human experience is insufficient to find the optimal configuration in a vast scheme space.

Method used

A multi-objective optimization method was adopted. By acquiring geological structure, hydrological seepage and vegetation growth parameters of the slope section, a multi-dimensional coupling adaptability quantitative assessment was carried out. A multi-objective optimization objective vector function was constructed, and the Pareto optimal solution set was obtained by using a non-dominated sorting genetic algorithm. Finally, the optimal ecological protection decision scheme was screened by the entropy weight method-TOPSIS model.

Benefits of technology

It enables refined and quantitative assessment of rock slopes, allowing for coordinated balancing of cost, schedule, safety, and ecology, providing a scientific basis for engineering decision-making, and taking into account both economic efficiency and ecological benefits.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116790A_ABST
    Figure CN123116790A_ABST
Patent Text Reader

Abstract

This application relates to a decision-making method and system for ecological protection of rock slopes based on multi-objective optimization. The method includes: acquiring geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters for each slope segment in a rock slope; performing multi-dimensional slope stability analysis on the above parameters to obtain slope stability characteristic data; acquiring a library of candidate protection measures, and combining the above parameters and slope stability characteristic data to conduct a multi-dimensional coupling adaptability quantitative assessment of protection measures for each slope segment, obtaining adaptability data including the applicability value and expected protection effect value of each measure in each segment; constructing a multi-objective optimization objective vector function based on the adaptability data and performing optimization to obtain a Pareto optimal solution set for the protection scheme; filtering and ranking the Pareto optimal solution set of the protection scheme, and outputting the optimal ecological protection decision scheme. This method can achieve accurate assessment of slope zoning and multi-objective collaborative optimization, realizing collaborative optimal decision-making.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of slope protection decision-making, and in particular relates to a method and system for ecological protection decision-making of rock slopes based on multi-objective optimization. Background Technology

[0002] With the rapid development of infrastructure construction in sectors such as transportation, water conservancy, and mining, ecological protection technology for rock slopes is receiving increasing attention. Unlike traditional rigid protection methods such as masonry and shotcrete, ecological protection technology emphasizes restoring slope vegetation and improving the surrounding ecological environment while ensuring slope stability. Currently, the design of ecological protection schemes for rock slopes typically involves engineering designers selecting and combining various protection methods, substrate types, and vegetation configurations based on on-site geological surveys and engineering experience.

[0003] In traditional techniques, the determination of slope protection schemes mainly relies on the engineer's personal experience and analogy with existing engineering cases. The feasibility verification of the scheme often needs to be carried out through numerical calculations or field tests after the design is completed. This approach can meet basic engineering safety requirements to a certain extent.

[0004] However, traditional methods have the following problems. First, the selection of protective measures lacks a systematic quantitative assessment basis, making it difficult to accurately match the differentiated characteristics of different slope sections in terms of geological structure, hydrological seepage, and vegetation growth conditions. Second, traditional methods usually only use the slope safety factor as a single evaluation index, failing to incorporate ecological protection costs, construction period, and vegetation restoration effects into a unified analytical framework for comprehensive consideration. Third, when there are many slope sections and a large number of possible combinations of measures, relying on manual experience makes it difficult to efficiently find the optimal configuration scheme that balances engineering economy and ecological benefits within a vast space of options. Therefore, there is an urgent need for a decision-making method for ecological protection of rock slopes that can systematically integrate multi-dimensional slope characteristic data, scientifically quantify the suitability of measures, and achieve multi-objective synergistic optimization. Summary of the Invention

[0005] Therefore, it is necessary to provide a decision-making method and system for ecological protection of rock slopes based on multi-objective optimization to address the above-mentioned technical problems.

[0006] Firstly, this application provides a decision-making method for ecological protection of rock slopes based on multi-objective optimization, including:

[0007] S1. Obtain the geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section in the rock slope.

[0008] S2. Multidimensional analysis of slope stability was performed on geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters to obtain slope stability characteristic data.

[0009] S3. Obtain a library of candidate protection measures, and based on the library, combine geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters and slope stability characteristic data to conduct a multi-dimensional coupling adaptability quantitative evaluation of protection measures for each slope section, and obtain protection measure adaptability data; the protection measure adaptability data includes the applicability value and expected protection effect value of each candidate protection measure in each slope section.

[0010] S4. Based on the adaptability data of protective measures, a multi-objective optimization objective vector function is constructed;

[0011] S5. Optimize the multi-objective optimization objective vector function to obtain the Pareto optimal solution set of the protection scheme;

[0012] S6. Based on the Pareto optimal solution set of the protection schemes, each candidate optimal protection scheme is screened and ranked to obtain the optimal ecological protection decision scheme; the optimal ecological protection decision scheme includes the configuration of target protection measures and the configuration of target protection parameters.

[0013] In one embodiment, S2 includes:

[0014] S21. Normalize the geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section to obtain a standardized slope parameter vector.

[0015] S22. Based on the standardized slope parameter vector, calculate the stability level membership degree of each slope section to obtain the membership degree matrix of each slope section at each stability level.

[0016] S23. Based on the membership matrix, the initial stability level of each slope segment is obtained, the structural plane attitude data in the geological structural parameters is extracted, and the spatial combination relationship parameters of the structural plane and the slope are calculated based on the structural plane attitude data. Based on the spatial combination relationship parameters, the dominant instability mode of each slope segment is obtained.

[0017] S24. Based on the membership matrix, initial stability level, and dominant instability mode, slope stability characteristic data are obtained.

[0018] In one embodiment, S3 includes:

[0019] S31. Based on the mechanism of action of each candidate protection measure in the candidate protection measure library, construct a set of response functions of each candidate protection measure to geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters and slope stability characteristic data, and obtain the multidimensional response feature vector of each candidate protection measure.

[0020] S32. Based on the multidimensional response feature vector and the safety factor of each slope segment in the slope stability feature data, calculate the measure-slope coupling coordination degree of each protection candidate measure in each slope segment, and obtain the coupling coordination degree matrix.

[0021] S33. Based on the coupling coordination degree matrix, calculate the applicability value of each candidate protection measure in each slope section to obtain the applicability matrix;

[0022] S34. Based on the applicability matrix and the slope stabilization mechanism parameters of each candidate protection measure in the candidate protection measure library, the expected protection effect value of each candidate protection measure in each slope section is calculated through the expected protection effect function to obtain the expected protection effect matrix.

[0023] S35. Based on the applicability matrix and the expected protection effect matrix, the applicability value and expected protection effect value of each candidate protection measure in each slope section are integrated to obtain the protection measure suitability data.

[0024] In one embodiment, S4 includes:

[0025] S41. Based on the applicability values ​​of each candidate protection measure in each slope section from the protection measure adaptability data, obtain the decision variable vector for each slope section.

[0026] S42. Based on the unit area construction cost and unit area construction period of each candidate protection measure in the candidate protection measure library, and combined with the area parameters and decision variable vectors of each slope section, construct the objective function for minimizing the total ecological protection cost and the objective function for minimizing the total construction period; wherein, the expressions for the total ecological protection cost and the total construction period are:

[0027] ;

[0028] ;

[0029] in, Represents the global decision variable vector. This represents the total cost of ecological protection. Indicates the total construction period. Indicates the total number of segments. This indicates the total number of candidate protective measures. Indicates the first The area of ​​each slope section Indicates the first The unit area construction cost of the various protective candidate measures No. Construction period per unit area for each of the proposed protective measures;

[0030] S43. Based on the expected protection effect values ​​in the adaptability data of protection measures, extract the expected increment of the safety factor and the ecological restoration potential value of each candidate protection measure in each slope section, and construct the objective function for maximizing the overall slope safety factor and the objective function for maximizing the expected total vegetation coverage; wherein, the expressions of the objective function for maximizing the overall slope safety factor and the objective function for maximizing the expected total vegetation coverage are:

[0031] ;

[0032] ;

[0033] in, Represents the global decision variable vector. This represents the objective function for maximizing the overall safety factor of the slope. This represents the objective function for maximizing the expected total vegetation cover. Indicates the total number of segments. This indicates the total number of candidate protective measures. Indicates the first Candidate protective measures in the first The expected increase in the safety factor of each section Indicates the potential value for ecological restoration. Choose a variable for binary, when the first Select the first section The value is 1 if a candidate protective measure type is selected, otherwise the value is 0.

[0034] S44. Based on the applicability values ​​in the adaptability data of protective measures, construct the feasibility constraints for decision variables of each slope section;

[0035] S45. Based on the objective functions of minimizing the total cost of ecological protection, minimizing the total construction period, maximizing the overall safety factor of the slope, and maximizing the expected total vegetation coverage, and combined with feasibility constraints, a multi-objective optimization objective vector function is constructed.

[0036] In one embodiment, S5 includes:

[0037] S51. Based on the decision variable vectors and feasibility constraints of each slope section, the population is initialized using a random generation method to obtain an initial population containing multiple individuals; where each individual represents a global protection scheme decision variable vector.

[0038] S52. Based on the multi-objective optimization objective vector function, calculate the objective vector value corresponding to each individual in the initial population to obtain the multi-objective fitness vector of each individual.

[0039] S53. Based on the multi-objective fitness vector of each individual, perform non-dominated sorting on each individual to obtain the non-dominated level number of each individual; and calculate the crowding distance value of each individual based on the target vector value of each individual within the same non-dominated level.

[0040] S54. Based on the non-dominated hierarchy number and crowding distance value of each individual, a tournament selection strategy is adopted to select parent individuals from the current population, and simulated binary crossover and polynomial mutation operations are performed on the selected parent individuals to generate the offspring population.

[0041] S55. Merge the current population and the offspring population into a mixed population. Re-execute the non-dominated sorting and crowding distance calculation on the mixed population, and select a new generation of population through the elite retention strategy. Repeat the iteration until the termination condition is met to obtain the Pareto optimal solution set of the protection scheme. The Pareto optimal solution set of the protection scheme includes multiple candidate optimal protection schemes. Each candidate optimal protection scheme includes the protection measure type and protection parameters of each slope section.

[0042] In one embodiment, S6 includes:

[0043] S61. Based on the multi-objective fitness vectors of each candidate scheme in the Pareto optimal solution set of the protection scheme, construct the decision matrix of the candidate scheme, and perform range normalization on the decision matrix to obtain the standardized decision matrix.

[0044] S62. Based on the standardized decision matrix, calculate the objective weights of each objective dimension, and combine them with the preset subjective weight vector of the decision-maker to calculate the comprehensive weight vector of each objective dimension.

[0045] S63. Based on the standardized decision matrix and the comprehensive weight vector, calculate the first weighted Euclidean distance between each candidate solution and the positive ideal solution, and calculate the second weighted Euclidean distance between each candidate solution and the negative ideal solution. Based on the first weighted Euclidean distance and the second weighted Euclidean distance, calculate the relative closeness of each candidate solution.

[0046] S64. Based on the relative similarity of each candidate scheme, select the candidate scheme with the highest relative similarity as the optimal ecological protection decision scheme.

[0047] S65. Based on the global decision variable vector corresponding to the optimal ecological protection decision scheme, the protection measure type and protection parameters of each slope section are mapped from the protection candidate measure library to obtain the target protection measure configuration and target protection parameter configuration.

[0048] Secondly, this application also provides a decision-making system for ecological protection of rock slopes based on multi-objective optimization, including:

[0049] The slope section parameter acquisition module is used to acquire geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section in a rock slope.

[0050] The slope stability multidimensional analysis module is used to perform multidimensional analysis of slope stability based on geological structure parameters, hydrological seepage parameters and vegetation growth constraint parameters, and obtain slope stability characteristic data.

[0051] The protective measure adaptability quantitative assessment module is used to obtain a library of protective candidate measures. Based on the library, and combined with geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters, and slope stability characteristic data, it performs a multi-dimensional coupled adaptability quantitative assessment of protective measures for each slope section, obtaining protective measure adaptability data. The protective measure adaptability data includes the applicability value and expected protective effect value of each protective candidate measure in each slope section.

[0052] The multi-objective optimization objective function construction module is used to construct a multi-objective optimization objective vector function based on the adaptability data of protective measures;

[0053] The multi-objective optimization solution module is used to optimize the objective vector function of the multi-objective optimization to obtain the Pareto optimal solution set of the protection scheme;

[0054] The optimal ecological protection scheme decision module is used to screen and rank each candidate optimal protection scheme based on the Pareto optimal solution set of protection schemes to obtain the optimal ecological protection decision scheme; the optimal ecological protection decision scheme includes the configuration of target protection measures and the configuration of target protection parameters.

[0055] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0056] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0057] The aforementioned multi-objective optimization-based decision-making method and system for ecological protection of rock slopes first acquires geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of slope sections and performs multi-dimensional stability analysis. This comprehensively quantifies the stability characteristics and instability modes of each slope section, providing a refined section-level data foundation for measure selection. Based on a database of candidate protection measures and the aforementioned multi-dimensional parameters, a multi-dimensional coupling adaptability quantitative assessment is performed. This accurately calculates the applicability and expected protection effect of each measure in each section, thus addressing the lack of quantitative basis for measure selection. Finally, based on the adaptability data, a comprehensive ecological protection system is constructed. The four-dimensional objective vector function of cost, construction period, slope safety factor, and expected vegetation coverage is used to solve multi-objective optimization problems. This allows for synergistic trade-offs among cost, construction period, safety, and ecology, resulting in a set of non-dominant Pareto optimal solutions. This addresses the limitations of single-index evaluation and the inefficiency of human experience in finding optimal solutions. By filtering and ranking the Pareto optimal solutions and outputting the corresponding measure types and protection parameters, the optimization results can be directly mapped to executable design configurations for each slope section. This provides a scientific basis for engineering decisions while balancing engineering economics and ecological benefits. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart illustrating a decision-making method for ecological protection of rock slopes based on multi-objective optimization in one embodiment.

[0060] Figure 2 This is a schematic diagram of the structure of a decision-making system for ecological protection of rock slopes based on multi-objective optimization in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] In one embodiment, reference Figure 1The document presents a flowchart illustrating a multi-objective optimization-based decision-making method for ecological protection of rock slopes provided in this application. This embodiment uses the application of this method to a decision-making terminal (hereinafter referred to as the terminal) as an example. It is understood that this method can also be applied to a server, or to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0063] S1. Obtain the geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section in the rock slope.

[0064] For example, the decision-making terminal divides the rock slope to be treated into several slope segments along the slope direction and height, based on the boundaries of geological structural units and the location of slope morphology changes. For each slope segment, the decision-making terminal collects geological structural parameters, hydrological seepage parameters, and vegetation growth constraint parameters. Geological structural parameters may include the occurrence elements of the rock strata constituting the slope, the degree of development of structural surfaces, the roughness coefficient of structural surfaces, the uniaxial compressive strength of the rock, the internal friction angle of the rock, and the cohesion. Hydrological seepage parameters may include the saturated permeability coefficient of the rock and soil mass, the natural water content, the pore water pressure distribution value, and the relative elevation of the groundwater level. Vegetation growth constraint parameters may include the thickness of the topsoil layer, the pH value of the topsoil layer, the organic matter content of the topsoil layer, the cation exchange capacity of the topsoil layer, and the sunshine duration and average annual precipitation of the slope. The decision-making terminal organizes the geological structural parameters, hydrological seepage parameters, and vegetation growth constraint parameters according to the slope segment index to form an original parameter vector. For the first... For each slope section, the original parameter vector can be represented as: Each component of this vector corresponds to the measured values ​​of the aforementioned geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters. Physical quantities of different dimensions retain their original units and dimensions during this stage; for example, the unit for geological strength parameters is MPa, the unit for permeability coefficient is m / day, and the value for pH is a logarithmic scale value. The decision-making terminal can store the acquired original parameters for all sections in a local database.

[0065] S2. Multidimensional analysis of slope stability was conducted on geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters to obtain slope stability characteristic data.

[0066] For example, after acquiring the parameters, the decision-making terminal performs multidimensional analysis on the original parameter vector of each slope segment. The multidimensional analysis includes standardization, stability level membership calculation, and instability mode discrimination. Specifically, the decision-making terminal can normalize the range of each component in the original parameter vector, mapping parameters of different dimensions to the same numerical range. The decision-making terminal can calculate the degree to which the segment belongs to each stability level using a multidimensional cloud model, obtaining a set of membership values. Simultaneously, the decision-making terminal can extract the structural plane attitude elements from the geological structural parameters, calculate the spatial angular relationship between the structural plane and the slope surface, and determine the most likely failure mode of the segment during instability based on this angular relationship. After the multidimensional analysis is completed, the decision-making terminal can merge the membership value sequence, the initial stability level label obtained from the discrimination, the category code of the dominant instability mode, and the safety factor calculated by the limit equilibrium method into a single stability feature vector. The decision-making terminal can perform the same operation on all slope segments to obtain a set of stability feature data for the entire slope.

[0067] S3. Obtain a library of candidate protection measures, and based on the library, combine geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters and slope stability characteristic data to conduct a multi-dimensional coupling adaptability quantitative assessment of protection measures for each slope section, and obtain protection measure adaptability data; the protection measure adaptability data includes the applicability value and expected protection effect value of each candidate protection measure in each slope section.

[0068] For example, after obtaining slope stability characteristic data, the decision terminal retrieves a pool of candidate protection measures from local storage or an external interface. This pool is a structured database containing various optional ecological protection measures, such as thick-layer substrate spraying, topsoil-net spraying, vegetation trenching, and vegetation concrete. For each measure, the pool records its mechanism of action, applicable geological conditions, slope stabilization mechanism parameters, unit area construction cost, and unit area construction period. Using slope sections as the basic evaluation unit, the decision terminal integrates the geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters, and stability characteristic data obtained in step S24 for each section, performing a multi-dimensional coupling adaptability quantitative evaluation of each measure in the pool. This multi-dimensional coupling adaptability quantitative evaluation can include four stages: response feature extraction, coupling coordination calculation, applicability calculation, and expected protection effect calculation. After the assessment is completed, the decision-making terminal can generate a suitability data pair for each slope segment and each combination of protection measures. This suitability data pair can include an applicability value and an expected protection effect value. The suitability data pairs for all segments and all measures constitute the protection measure suitability data.

[0069] S4. Based on the adaptability data of protective measures, a multi-objective optimization objective vector function is constructed.

[0070] For example, after obtaining the adaptability data of protective measures, the decision-making terminal constructs a multi-objective optimization objective vector function. This function uses a global decision variable vector as independent variables and four objective function values ​​as dependent variables. The decision-making terminal can use mixed-integer encoding for the type of protective measures, substrate type, and vegetation configuration scheme for each slope section, transforming discrete selection schemes into binary decision variables. Based on the unit area cost and unit area construction period in the protective candidate measure library, the decision-making terminal can construct cost minimization and construction period minimization objective functions. Simultaneously, the decision-making terminal can extract the expected increment of the safety factor and the ecological restoration potential value from the expected protective effect value, constructing a safety factor maximization objective function and an ecological coverage maximization objective function. Based on the applicability value, the decision-making terminal can construct feasibility constraints for the decision variables. Finally, the decision-making terminal integrates the four objective functions into a four-dimensional objective vector function. The output of this function is a four-dimensional vector, with each component corresponding to the total cost, total construction period, negative safety factor value, and negative ecological restoration potential value, respectively. This objective vector function serves as the standard input for subsequent optimization solution modules.

[0071] S5. Optimize the objective vector function of the multi-objective optimization to obtain the Pareto optimal solution set of the protection scheme.

[0072] Optionally, after constructing the multi-objective optimization objective vector function, the decision terminal can employ a non-dominated sorting genetic algorithm as the solver. The decision terminal performs optimization on the multi-objective optimization objective vector function. The non-dominated sorting genetic algorithm first randomly generates a set of initial solutions in the decision variable space, each solution corresponding to a complete global protection scheme. Then, the decision terminal evaluates the performance of each solution across the four objective dimensions and divides individuals in the population into different levels based on non-dominated relationships. The decision terminal calculates the crowding distance between individuals within the same level to maintain the diversity of the solution set. The decision terminal selects dominant individuals as parents based on level and crowding distance using a tournament selection mechanism, and generates offspring through simulated binary crossover and polynomial mutation. The decision terminal merges the parents and offspring and recalculates the non-dominated sorting and crowding distance, selecting the next generation of the population through an elite retention strategy. This process is repeated until a preset termination condition is met. When the algorithm terminates, all individuals in the final population at the first non-dominated level are output as the Pareto optimal solution set for the protection scheme. Each solution in the Pareto optimal solution set of the protection scheme represents a candidate scheme that reaches the Pareto optimal boundary among the four objectives. No single solution can improve the other objectives without worsening at least one objective.

[0073] S6. Based on the Pareto optimal solution set of the protection schemes, each candidate optimal protection scheme is screened and ranked to obtain the optimal ecological protection decision scheme; the optimal ecological protection decision scheme includes the configuration of target protection measures and the configuration of target protection parameters.

[0074] For example, after obtaining the Pareto optimal solution set, the decision-making terminal performs scheme screening and ranking operations to determine the final single optimal solution. Multiple schemes in the Pareto optimal solution set exhibit non-dominant trade-offs among the four objectives, requiring a multi-attribute decision-making method to select the scheme with the best overall performance. The decision-making terminal can employ a multi-attribute decision-making process combining entropy weighting and TOPSIS methods. The multi-attribute decision-making process first constructs a decision matrix for all schemes in the Pareto solution set according to the four objective dimensions and performs range normalization to eliminate the influence of dimensions. The decision-making terminal calculates the objective weights of each objective from the normalized data using entropy weighting and receives the subjective preference weights input by the engineering decision-maker, linearly combining the objective and subjective weights to obtain the comprehensive weight. The decision-making terminal determines the positive and negative ideal solutions, calculates the weighted Euclidean distance between each scheme and the positive and negative ideal solutions, and further calculates the relative proximity. The decision-making terminal can sort the schemes in descending order based on the relative proximity and select the scheme with the highest proximity as the optimal solution. The decision terminal can map the binary decision variables of the optimal solution back to the specific measure type, substrate type and plant configuration scheme, and output the complete target protection measure configuration and target protection parameter configuration.

[0075] The aforementioned multi-objective optimization-based decision-making method for ecological protection of rock slopes first obtains geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters for each section of the rock slope. Then, multi-dimensional stability analysis is performed using a multi-dimensional cloud model and instability mode discrimination, comprehensively quantifying the stability level membership and dominant instability modes of each section, thus solving the problem of inaccurate slope condition characterization in traditional methods. Next, based on a library of candidate protection measures, and combined with the aforementioned multi-dimensional parameters, a multi-dimensional coupling adaptability quantitative evaluation of each measure in each section is conducted using a coupling coordination degree model and grey relational analysis. This yields quantified applicability values ​​and expected protection effect values, thus addressing the lack of systematic quantitative basis for measure selection. Finally, based on the applicability... The decision variables are constructed by encoding mixed integer values. A multi-objective optimization objective vector function is constructed with four optimization objectives: total ecological protection cost, total construction period, overall slope safety factor, and expected total vegetation coverage. The Pareto optimal solution set is obtained by solving the Pareto optimal solution set through a non-dominated sorting genetic algorithm. This approach can simultaneously consider engineering economics and ecological benefits to achieve multi-objective collaborative optimization, thereby solving the problem that traditional human experience is difficult to optimize in a large solution space. Finally, based on the Pareto optimal solution set, the entropy weight method-TOPSIS combined evaluation model is used for screening and sorting. This allows the ecological protection decision scheme with the best comprehensive benefits to be selected from multiple non-dominated candidate schemes, thus providing a scientific and quantifiable decision basis for the ecological protection design of rock slopes.

[0076] In an optional embodiment, S2 includes:

[0077] S21. Normalize the geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section to obtain a standardized slope parameter vector.

[0078] Optionally, the decision terminal can read the first... The original parameter vector of each slope section The original parameter vector contains One portion, The value is the total number of categories of the collected parameters. For the first... Each slope segment has a component. The decision terminal calculates the maximum and minimum values ​​for each component, with the maximum value denoted as... The minimum value is denoted as The decision-making terminal can [address the issue of] the [other aspects]. The first section Each component To perform range normalization calculations, the normalization formula can be: .in, Indicates the first The first slope section The normalized values ​​of the parameters range from [0,1]. The physical basis of range normalization is to eliminate the interference of dimensional differences between different physical quantities on subsequent comprehensive analysis. For example, it maps the megapascal dimension of geological intensity, the meter-day dimension of permeability coefficient, and the logarithmic dimension of pH to a unitless relative proportion. The decision terminal can perform the above calculations on all parameter components of each segment sequentially to obtain the first value. Standardized parameter vectors of each segment The standardized parameter vector preserves the relative magnitudes of each parameter across all segments, providing input data with a consistent scale for subsequent membership calculations.

[0079] S22. Based on the standardized slope parameter vector, calculate the membership degree of the stability level of each slope section, and obtain the membership degree matrix of each slope section at each stability level.

[0080] Optionally, after completing the normalization calculation, the decision terminal calculates the stability level membership degree of each slope segment using a multi-dimensional cloud model. The stability level set can be predefined as... ,in, Indicates the total number of grades. Indicates the first The stability level. For the first... The standardized parameter vector for each slope section is: The decision-making terminal for the first For each stability level, obtain two cloud parameters for that level across each parameter dimension, i.e., the expected value. and entropy Expected value Characterizing the first The stability level is at the [number]th [level]. Standard center value of the term parameter, entropy value Characterizing this level in The tolerable float width on the parameter. The decision terminal calculates the... The first section belongs to the first Membership of each stability level The formula for calculating membership degree can be: .in, For the first The segment is relative to the first The multidimensional membership degree of the first level, whose value is closer to 1, indicates that the parameter distribution of that segment is more closely related to the first level. The more closely they match the standard centers at each level; The first one obtained in step S21 Section 1 The normalized value of the parameter; For the first The level in the Expected value on the parameter; For the first The level in the The entropy value of the parameter. The decision terminal... From 1 to and From 1 to Calculate the membership matrix by performing calculations on all combinations. The matrix is OK A matrix of columns, the matrix of columns Line number The column element is The physical basis of the multidimensional cloud model lies in the ambiguity of the grade boundaries in slope stability evaluation. The multidimensional cloud model can be used to quantitatively describe this ambiguity using two parameters: expectation and entropy.

[0081] S23. Based on the membership matrix, the initial stability level of each slope segment is obtained, the structural plane attitude data in the geological structural parameters is extracted, and the spatial combination relationship parameters of the structural plane and the slope are calculated based on the structural plane attitude data. Based on the spatial combination relationship parameters, the dominant instability mode of each slope segment is obtained.

[0082] Optionally, the decision-making terminal calculates the membership matrix. Then, the initial stability level of each slope segment is determined according to the principle of maximum membership. For the first... Each segment, the decision terminal scans that segment across all... Membership at each level arrive The level label corresponding to the maximum value is selected as the initial stability level label for this section. Its calculation formula is . Values ​​range from 1 to Positive integers between [a certain value] and [a certain value]. Simultaneously, the decision-making terminal extracts the [a certain value] from the geological structure parameters. Structural surface attitude data for each segment can include the dip angle of the structural surface. Inclination angle and the slope angle and tilt angle These angles are all in degrees. The decision terminal calculates the absolute value of the angle between the structural surface dip and the slope dip. And compare the dip angles of the structural surfaces. With slope angle The size relationship. When Less than the preset angle threshold and Less than When the dominant instability mode of the segment is determined to be a bedding-sliding instability mode, the decision terminal determines it to be a wedge-shaped instability mode when two or more structural planes can form a closed wedge. When neither of these conditions is met, the decision terminal determines it to be a global shear instability mode. The decision terminal encodes the determination result as an integer value. The data is stored, with the layered sliding type, wedge-shaped instability type, and overall shear type corresponding to the code values ​​1, 2, and 3, respectively.

[0083] S24. Based on the membership matrix, initial stability level, and dominant instability mode, slope stability characteristic data are obtained.

[0084] Optionally, after completing the membership calculation and instability mode identification, the decision-making terminal integrates the above multidimensional analysis results. For the first... For each slope segment, the decision-making terminal constructs a stability feature vector for that segment. This vector includes the initial stability level label. Membership values ​​of this section across all stability levels arrive , dominant instability mode coding value and safety factor Safety factor The calculation method involves using the pore water pressure distribution value from the hydrological seepage parameters at the decision-making terminal, and employing the limit equilibrium analysis method. The physical meaning of this safety factor is the ratio of the slope's anti-sliding force to its sliding force. Stability eigenvector The expression can be The decision-making terminal repeats the above construction operation for all slope sections, forming a slope stability feature data set by combining the stability feature vectors of all sections. ,in, This represents the total number of slope sections. The slope stability characteristic dataset includes quantitative information on the stability level of each section, qualitative information on instability modes, and quantitative information on safety margins. These information together constitute the complete input data for one side of the slope in the subsequent suitability assessment of measures.

[0085] In an optional embodiment, S3 includes:

[0086] S31. Based on the mechanism of action of each candidate protection measure in the candidate protection measure library, construct a set of response functions of each candidate protection measure to geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters and slope stability characteristic data, and obtain the multidimensional response feature vector of each candidate protection measure.

[0087] Optionally, when initiating the adaptability quantitative assessment, the decision-making terminal establishes a response function for each protective measure in the protective measure candidate library. The protective measure candidate library contains a total of... This measure, using Indicates the first This measure. For the first This measure involves the decision-making terminal determining its response to input parameters based on its mechanism of action. Input parameter vector. By the Geological structure parameter vector of each section Hydrological seepage parameter vector Vegetation growth constraint parameter vector and stability eigenvectors Combining, that is The decision terminal will be the first Such measures in the The response feature vector of each segment is denoted as... It is 3D column vector, This represents the total number of response dimensions. The th element of this response feature vector... Each component Indicates the first Such measures in the Input parameters on each response dimension The response values ​​are determined by different measures. Different measures exhibit different response function forms. For example, the response of thick substrate spraying to slope roughness shows a positive correlation, while the response to slope gradient shows a rapid decay in response value after the gradient exceeds a certain value. The decision terminal establishes a complete set of response functions for each measure. Each response feature vector maps the numerical values ​​in the input parameter space to multidimensional feature values ​​in the response space. Response feature vectors can be used to reflect the differentiated applicability of different measures under the same slope conditions.

[0088] S32. Based on the multidimensional response feature vector and the safety factor of each slope segment in the slope stability feature data, calculate the measure-slope coupling coordination degree of each protection candidate measure in each slope segment, and obtain the coupling coordination degree matrix.

[0089] Optionally, after obtaining the multidimensional response feature vectors of each measure, the decision-making terminal calculates the coupling coordination degree between each measure and each slope segment. For the first... Such measures and the first In each slope section, the decision-making terminal extracts the measure in the [number]th [section]. The response feature vector of each segment And extract the first one obtained in step S24 Stability feature vector of each segment The decision terminal first calculates the Euclidean norm of the response feature vector. The Euclidean norm of the response eigenvector can be used to characterize the first... Such measures in the The overall response intensity of each segment. The Euclidean norm of the stability eigenvector is calculated at the decision terminal. The Euclidean norm of the stability eigenvector can be used to characterize the first... The overall stability state of each segment. The decision terminal calculates the coupling degree. ,in, For the first This measure and the first The coupling degree of each section ranges from 0 to 1. A coupling degree closer to 1 indicates a stronger correlation between the response measures and the slope condition. The decision-making terminal calculates the coupling coordination degree. ,in, For the first This measure and the first The degree of coupling coordination of each segment; and These are the weighting coefficients. , The contribution weight of the response intensity of the characterization measures in the comprehensive coordination index. The contribution weights of slope stability status to the comprehensive coordination index are both dimensionless. (Decision-making terminal...) From 1 to and From 1 to All combinations of calculations This constitutes the coupling coordination degree matrix. The coupling coordination degree matrix is OK A matrix of columns. Coupling coordination degree can be used to measure the tightness of coupling between measures and slopes, as well as the level of synergistic development between the two, avoiding the one-sidedness of evaluation by a single indicator.

[0090] S33. Based on the coupling coordination degree matrix, calculate the applicability value of each candidate protection measure in each slope section to obtain the applicability matrix.

[0091] Optionally, the decision-making terminal obtains the coupling coordination degree matrix. Then, grey relational analysis can be used to calculate the applicability value of each measure in each slope section. For the first... For each slope section, the decision-making terminal uses the coupling coordination degree matrix. The Extract all from column Coupling and coordination value of the measures arrive And find the maximum value in that column and denote it as . ,Right now . Indicates the first The coordination degree value corresponding to the measure with the highest coupling coordination degree in each segment. The decision terminal for the first segment... Such measures in the For each segment, calculate its grey relational coefficient with respect to the ideal value. ,in, The grey relational coefficient; This represents the minimum absolute difference across all sections and all measures. This represents the maximum absolute difference across all sections and all measures. The discrimination coefficient is a positive number between 0 and 1, typically set to 0.5. The decision-making terminal normalizes the grey relational coefficient to obtain the applicability value. ,in, For the first Such measures in the The applicability value of each segment satisfies the condition that the sum of the applicability values ​​of all measures in the same segment equals 1. The decision-making terminal can calculate the applicability values ​​for all measures and all segments, forming an applicability matrix. Grey relational analysis determines relative merit by measuring the geometric similarity between each measure and a virtual ideal measure. It is suitable for multi-factor decision-making scenarios with limited samples and inconsistent indicator dimensions.

[0092] S34. Based on the applicability matrix and the slope stabilization mechanism parameters of each candidate protection measure in the candidate protection measure library, the expected protection effect value of each candidate protection measure in each slope section is calculated through the expected protection effect function to obtain the expected protection effect matrix.

[0093] Optionally, after calculating the applicability value, the decision-making terminal further calculates the expected protective effect value of each measure in each slope section. The expected protective effect value includes the expected increment of the safety factor and the ecological restoration potential value. The decision-making terminal can extract the first protective measure from the protective candidate measure library. The slope stabilization mechanism parameters for this measure can include anchorage force parameters, root-soil composite shear strength increment parameters, and slope constraint force parameters. The decision terminal substitutes these slope stabilization mechanism parameters into the limit equilibrium method equations to calculate the first... Such measures in the The increase in the safety factor of each section after implementation relative to the unprotected state is denoted as: The decision-making terminal can extract the first [measure] from the pool of candidate protection measures. The vegetation growth suitability parameters for this measure may include substrate thickness, water retention capacity, and nutrient supply rate. The decision terminal can compare these vegetation growth suitability parameters with those obtained in step S1. The vegetation growth constraint parameters of each segment are matched and calculated to obtain the first... Such measures in the Ecological restoration potential value of each section Its value ranges from 0 to 1. The decision terminal constructs a function to determine the expected protection effect. ,in, For the first Such measures in the The expected protective effect value for each section; This represents the weighting factor of the safety factor increment in the expected protection effect. The value range of is [0,1]. The value of this parameter can be used to reflect the relative emphasis that decision-makers place on project safety and ecological restoration. The decision-making terminal can calculate the expected protection effect value for all measures and all sections, forming an expected protection effect matrix. The expected protection effect function can be used to unify and quantify the two non-substitutable dimensions of slope stabilization and ecological restoration into a single comparable indicator.

[0094] S35. Based on the applicability matrix and the expected protection effect matrix, the applicability value and expected protection effect value of each candidate protection measure in each slope section are integrated to obtain the protection measure suitability data.

[0095] Optionally, the decision-making terminal obtains the applicability matrix. and expected protection effect matrix Next, the two matrices are integrated element-wise. For the first... Such measures in the Each segment, the decision-making terminal from the applicability matrix Extract the applicability value for that position. From the expected protection effect matrix Extract the expected protection value for that location. Combine the two into a binary pair . That is, the first Candidate protective measures in the first Complete adaptability data for each slope section. Decision terminal for all From 1 to and From 1 to The combined execution of the integration operation yields a set of protection measure adaptability data. In the protective measure suitability dataset, the applicability value... It can be used to reflect the first This measure, compared to other measures, is in the first The relative applicability of each section and the expected protective effect value It can be used to reflect the absolute protective effect level of the measure in that section. The two dimensions describe the matching relationship between the measure and the slope from different perspectives. The applicability value can also be used to screen the set of feasible candidate measures for each slope section, and the expected protective effect value can also be used to construct the safety factor increment and ecological restoration benefits in the optimization objective function.

[0096] In an optional embodiment, S4 includes:

[0097] S41. Based on the applicability values ​​of each candidate protection measure in each slope section from the protection measure adaptability data, obtain the decision variable vector for each slope section.

[0098] Optionally, the decision-making terminal bases its decisions on the applicability values ​​in the protective measure adaptability data. Determine the first The set of protective measures that can be selected for each segment. For the first segment... In each segment, the decision-making terminal can define three types of binary selection variables: protection measure type selection variable, substrate type selection variable, and plant configuration scheme selection variable. The protection measure type selection variable can be... ,in, Traverse 1 to , The total number of candidate protective measures, The value can be 0 or 1, with a value of 1 indicating the first... Select the first section Types of protective measures. The substrate type selection variable can be... ,in Traverse 1 to , This represents the total number of candidate substrate types. The value can be 0 or 1. The variable for selecting the plant configuration scheme can be... ,in, Traverse 1 to , The total number of candidate plant configuration schemes. The value can be either 0 or 1. The three types of selection variables satisfy a mutual exclusion constraint, meaning that for each segment, the sum of the selection variables for protection measure type equals 1, the sum of the selection variables for substrate type equals 1, and the sum of the selection variables for plant configuration scheme equals 1. The decision terminal can then... All binary variables of each segment are arranged sequentially to form the decision variable vector for that segment. All components of the decision variable vector are binary integers. The decision terminal can further concatenate the decision variable vectors of all segments into a global decision variable vector. Mixed-integer encoding can be used to transform engineering selection problems into a standard combinatorial optimization variable form, enabling subsequent evolutionary algorithms to effectively search the discrete selection space.

[0099] S42. Based on the unit area construction cost and unit area construction period of each protective candidate measure in the protective candidate measure library, and combined with the area parameters and decision variable vectors of each slope section, construct the objective function for minimizing the total ecological protection cost and the objective function for minimizing the total construction period.

[0100] Optionally, after completing the coding of decision variables, the decision terminal constructs a cost objective function and a schedule objective function. The decision terminal then extracts the first [missing information] from the pool of candidate protection measures. Construction cost per unit area of ​​various protective candidate measures The unit is yuan per square meter, and the construction period per unit area is... The unit is per square meter per day. The decision terminal obtains the data stored in step S1. Slope area of ​​each slope section The unit is square meters. Select the first section The cost contribution of this measure is The contribution to the construction period is Due to decision variables A value of 1 indicates that the measure is selected, and a value of 0 indicates that it is not selected. Therefore, the first... The actual cost of each segment can be expressed as: The actual construction period can be expressed as The decision-making terminal can sum the costs of all sections to obtain the objective function for the total cost of ecological protection. ,in, Represents the global decision variable vector The corresponding total ecological protection cost is expressed in yuan. The decision-making terminal can sum the construction periods for all sections to obtain the objective function for the total construction period. ,in, Represents the global decision variable vector The corresponding total construction period is in days. Both objective functions are minimization functions. The total cost objective can be used to constrain the feasibility of the scheme from an economic perspective, while the total construction period objective can be used to constrain the efficiency of the scheme from a construction organization perspective.

[0101] S43. Based on the expected protection effect values ​​in the adaptability data of protection measures, extract the expected increment of the safety factor and the ecological restoration potential value of each candidate protection measure in each slope section, and construct the objective function for maximizing the overall slope safety factor and the objective function for maximizing the expected total vegetation coverage; wherein, the expressions of the objective function for maximizing the overall slope safety factor and the objective function for maximizing the expected total vegetation coverage are:

[0102] ;

[0103] ;

[0104] in, Represents the global decision variable vector. This represents the objective function for maximizing the overall safety factor of the slope. This represents the objective function for maximizing the expected total vegetation cover. Indicates the total number of segments. This indicates the total number of candidate protective measures. Indicates the first Candidate protective measures in the first The expected increase in the safety factor of each section Indicates the potential value for ecological restoration. Choose a variable for binary, when the first Select the first section The value is 1 if the candidate protective measure type is selected, otherwise the value is 0.

[0105] Optionally, after establishing cost and schedule targets, the decision-making terminal can further establish safety factor and ecological coverage targets. The decision-making terminal can derive these targets from the expected protection effectiveness values ​​derived from the protective measure adaptability data. Departure, according to The definition is used to extract the expected increment of the safety factor. and ecological restoration potential value . No. Select the first section When this measure is implemented, its contribution to the overall safety factor of the slope is: The contribution to the total vegetation coverage is The decision-making terminal can sum up the contributions of all sections to obtain the objective function for the overall slope safety factor. ,in, Represents the global decision variable vector The corresponding overall slope safety factor. The decision terminal simultaneously constructs an objective function for the expected total vegetation cover. ,in, Represents the global decision variable vector The corresponding expected total vegetation cover. Since optimization algorithms typically take the form of minimization problems, the decision-making terminal can, in actual optimization, consider the... and Taking the negative value as a component of the objective vector transforms the maximization problem into an equivalent minimization problem.

[0106] S44. Based on the applicability values ​​in the adaptability data of protective measures, construct feasibility constraints for decision variables of each slope section.

[0107] Optionally, the decision-making terminal constructs feasibility constraints on the decision variables while constructing the objective function. The decision-making terminal uses the applicability value calculated in step S33. Constrain the values ​​of decision variables. Applicability value. It can be used to reflect the first Such measures in the The relative applicability of each segment is represented by a value between 0 and 1. The decision-making terminal presets an applicability threshold. The value ranges from 0 to 1. The decision-making terminal specifies the... Variables for selecting the type of protective measures for each section The value of must satisfy ,in, For indicator functions, when When the indicator function takes a value of 1, When this condition is met, the indicator function takes a value of 0. The physical meaning of this constraint is that the applicability value is below a threshold. The measures were forcibly excluded from the first The feasible set of choices for each segment, where the decision variable for that segment is fixed at 0 in the corresponding dimension. This constraint operation can significantly reduce the search space of the decision variables, limiting the optimization search scope to combinations of measures that are actually feasible in engineering. The decision terminal can also incorporate this constraint, along with the mutual exclusion constraint in step S41, into the constraint set of the optimization problem for use in the population initialization stage in step S51 and the correction stage after the mutation operation in step S54.

[0108] S45. Based on the objective functions of minimizing the total cost of ecological protection, minimizing the total construction period, maximizing the overall safety factor of the slope, and maximizing the expected total vegetation coverage, and combined with feasibility constraints, a multi-objective optimization objective vector function is constructed.

[0109] Optionally, after defining all sub-objective functions and constraints, the decision terminal integrates them into a unified multi-objective optimization objective vector function. The decision terminal can define a four-dimensional objective vector. ,in Total cost target, in yuan; The total project duration is in days. The target is a negative safety factor; The objective is to achieve a negative ecological coverage rate. The first two objectives are kept in minimization form, while the latter two objectives are negativeized and transformed into a minimization form, thus unifying the four objectives mathematically into a minimization problem. Decision variables. The feasible region is defined by the mutual exclusion constraint in step S41 and the applicability threshold constraint in step S44. The complete multi-objective optimization problem can be formulated as follows: The constraints can be , , ,as well as For all and all All of these hold true. The objective vector function can integrate four dimensions of engineering decision indicators into a single mathematical framework, providing standardized input for multi-objective evolutionary optimization.

[0110] In an optional embodiment, S5 includes:

[0111] S51. Based on the decision variable vectors and feasibility constraints of each slope section, the population is initialized using a random generation method to obtain an initial population containing multiple individuals; where each individual represents a global protection scheme decision variable vector.

[0112] Optionally, the decision terminal performs population initialization before initiating the optimization solution. The decision terminal sets the population size to be... , It is a positive integer. For the th in the population... For each individual, the decision-making terminal needs to generate a complete global decision variable vector for it. The generation method is random, but it must satisfy the feasibility constraints defined in step S44. For the... For each slope section, the decision-making terminal obtains all data for that section. Applicability value of various protective measures arrive And filter out those that meet the requirements The measures are considered as the set of feasible measures for that section. The decision-making terminal randomly selects a measure from the set of feasible measures with equal probability, and assigns the corresponding binary selection variable to that measure. Set it to 1, and set the binary selection variables for the remaining measures to 0. Simultaneously, the decision terminal... One candidate substrate type is randomly selected with equal probability from the candidate substrate types, and the corresponding substrate is then... Set to 1, and the rest to 0. The decision terminal starts from... One of the candidate plant configuration schemes is randomly selected with equal probability, and the corresponding... Set to 1 for all others, and 0 for the rest. The decision terminal processes all... The above random selection operation is performed on each segment to assemble the first segment. The vector of global decision variables for each individual The decision-making terminal repeats the above process. Next, generate Individuals constitute the initial population. The initial population refers to a uniformly random sample within a decision space that satisfies all engineering feasibility constraints, providing diverse starting search points for subsequent evolutionary iterations.

[0113] S52. Based on the multi-objective optimization objective vector function, calculate the objective vector value corresponding to each individual in the initial population to obtain the multi-objective fitness vector of each individual.

[0114] Optionally, after completing population initialization, the decision-making terminal evaluates the multi-objective fitness of each individual. The decision-making terminal can then assess the multi-objective fitness of the first individual. The vector of global decision variables for each individual Substitute the multi-objective optimization objective vector function constructed in step S45 The calculations are performed within this framework. The calculation process can be followed as follows: Execute item by item. Cost target. Calculate according to the formula in step S42 to obtain the first... The total cost of each option is expressed in yuan. Project timeline target. Similarly, according to the formula in step S42, the result is obtained. The total construction period for each plan is expressed in days. The safety factor is a negative target. The negative value can be calculated according to the formula in step S43 and then the result can be taken. Negative ecological coverage target. Similarly, the negative sign can be taken after calculating according to the formula in step S43. The decision terminal can combine the four calculation results into a four-dimensional column vector. ,in, arrive These correspond to the four target values ​​mentioned above. This four-dimensional column vector is the [number]th [value]. The multi-objective fitness vector of each individual. The decision terminal can evaluate all individuals in the population. The fitness vector of each individual is calculated to obtain the fitness matrix of the entire population. The fitness matrix can also be used for non-dominated sorting operations.

[0115] S53. Based on the multi-objective fitness vector of each individual, perform non-dominated sorting on each individual to obtain the non-dominated level number of each individual; and calculate the crowding distance value of each individual based on the objective vector value of each individual within the same non-dominated level.

[0116] Optionally, after obtaining the population fitness matrix, the decision-making terminal performs a non-dominated ranking operation. The decision-making terminal uses a ranking method based on Pareto dominance. For any two individuals in the population... and ,individual Dominant Individual If and only if the individual The target value in all four target dimensions is no greater than the individual value. The target value, and is strictly smaller than the individual value in at least one dimension. The target value is determined by the decision terminal comparing the pairwise dominance relationships of all individuals in the population. Individuals not dominated by any other individual are classified as the first non-dominated level, denoted as level 1. These individuals are then removed from the population. The comparison and classification process is repeated for the remaining individuals to obtain the second non-dominated level, denoted as level 2. This process is repeated until all individuals are assigned a level number. , Values ​​are positive integers, and are dimensionless. The decision-making terminal simultaneously calculates the crowding distance for individuals within the same level. For the... For each target dimension, the decision-making terminal sorts individuals within the same level according to their target values ​​in ascending order. The difference between the target values ​​of two adjacent individuals after sorting is divided by the difference between the maximum and minimum values ​​of the entire population in that dimension to obtain the normalized distance for that dimension. The normalized distances of the four dimensions are then summed to obtain the crowding distance value for that individual. The formula for calculating the crowding distance value can be: .in, and For the same level, the first edge The objective function values ​​of each individual adjacent to the target direction. and For the current population in the 1st The maximum and minimum values ​​in each target direction. It is a very small positive number with a dimensionless unit, used to prevent the denominator from being zero. Crowding distance is the degree of sparsity of the solutions around an individual in the target space. Individuals with a large crowding distance are more conducive to maintaining a uniform distribution of the Pareto front.

[0117] S54. Based on the non-dominated hierarchy number and crowding distance value of each individual, a tournament selection strategy is used to select parent individuals from the current population, and simulated binary crossover and polynomial mutation operations are performed on the selected parent individuals to generate the offspring population.

[0118] Optionally, after completing the non-dominated sorting and crowding distance calculation, the decision-making terminal performs selection, crossover, and mutation operations to generate the offspring population. The decision-making terminal employs a binary tournament selection strategy. This strategy is repeated. Each time, two distinct individuals are randomly selected from the current population, and their non-dominant level numbers are compared. The individual with the smaller tier number is selected as the winner. If two individuals have the same tier number, their crowding distance is compared. The individual with the largest crowding distance is selected as the winner. Each tournament produces one parent individual. The tournament produced a total of A parent individual. The decision-making terminal then... Each parent individual is paired up, with two adjacent parent individuals forming a pair. For each pair of parent individuals, the decision-making terminal uses a preset crossover probability. Perform a simulated binary crossover operation. The simulated binary crossover generates two offspring individuals based on the weighted average of the parent individuals' components. The distribution exponent for the crossover operation is... , It is a positive real number. It can be used to control the dispersion of offspring relative to their parents. The decision terminal assigns a preset mutation probability to each component of the newly generated offspring individuals. Perform a polynomial mutation operation. A polynomial mutation adds a random perturbation to the original component values, and the distribution exponent of the mutation operation is... , It is a positive real number. It can be used to control the distribution pattern of variable-length crossover. All offspring individuals generated after crossover and mutation operations constitute the offspring population. ,in, Indicates the index of the current algebra. The population size was the same as that of the parent population, both being [missing information]. Individuals. After the mutation operation is completed, the decision terminal can correct individuals in the offspring population that violate the constraints of step S44, forcibly adjusting the variable components that violate the constraints to the nearest feasible value.

[0119] S55. Merge the current population and the offspring population into a mixed population. Re-execute the non-dominated sorting and crowding distance calculation on the mixed population, and select a new generation of population through the elite retention strategy. Repeat the iteration until the termination condition is met to obtain the Pareto optimal solution set of the protection scheme. The Pareto optimal solution set of the protection scheme includes multiple candidate optimal protection schemes. Each candidate optimal protection scheme includes the protection measure type and protection parameters of each slope section.

[0120] Optionally, after generating the offspring population, the decision terminal performs an elite retention operation. The decision terminal will then... The current population of this generation and offspring population Merged into a mixed population , The total number of individuals is Decision-making terminal for mixed populations All individuals in the mixed population undergo a re-execution of the non-dominated sorting and crowding distance calculation operations in step S53, resulting in a new non-dominated hierarchy number and crowding distance value for each individual. The decision terminal then selects individuals layer by layer according to the non-dominated hierarchy number in ascending order to fill the new generation of the population. When the number of individuals in a certain layer exceeds When there is remaining capacity, the decision-making terminal sorts the individuals in the layer from largest to smallest based on their crowding distance, and selects several individuals with the largest crowding distance to fill the remaining space. Until the limit is reached. This operation ensures that individuals with smaller non-dominant hierarchies and greater crowding distances are given priority for the next generation, thus preserving superior genes within the population. Decision terminal completes. After construction, check the current algebra. Has the preset maximum number of iterations been reached? , For positive integers, the unit is s. If not reached, the decision terminal will... Return to step S54 for the current population to continue the evolutionary operation for the next generation. If the desired outcome is achieved, the decision terminal terminates the iteration and returns to the final population. Extract all individuals with a non-dominated hierarchy number of 1 to form the Pareto optimal solution set of the protection scheme. Each individual in the Pareto optimal solution set of this protection scheme is a candidate optimal protection scheme. Each candidate optimal protection scheme includes the values ​​of the protection measure type selection variable, the substrate type selection variable, and the vegetation configuration scheme selection variable for each slope section.

[0121] In an optional embodiment, S6 includes:

[0122] S61. Based on the multi-objective fitness vectors of each candidate scheme in the Pareto optimal solution set of the protection scheme, construct the decision matrix of the candidate scheme, and perform range normalization on the decision matrix to obtain the standardized decision matrix.

[0123] Optionally, the first step in the solution selection phase for the decision-making terminal is to construct a standardized decision matrix. The Pareto optimal solution set for the protection schemes. The CCP One candidate solution, It is a positive integer. For the th There are 10 candidate solutions, and the decision-making terminal knows their multi-objective fitness vectors. ,in, Total cost, in yuan; The total construction period is in days. The safety factor is negative. A negative value for ecological coverage. The decision-making terminal can display all... The objective vectors of the proposed solutions are arranged in rows to form a decision matrix. ,in, For scheme indexing, Indexing for target dimensions. The decision terminal indexes each target dimension. Statistics of all The minimum value of each solution in this dimension and maximum value Then for the first... The first scheme is in the Target values ​​for each dimension Calculate its range normalized value Normalized value The value range is [0,1]. Range normalization is used to map target values ​​of different dimensions and orders of magnitude to the same scale, ensuring that the contribution weights of each dimension in subsequent distance calculations are not affected by the size of the dimension. The decision terminal can process all... The normalization calculations for each option and all four dimensions were completed to obtain the standardized decision matrix. .

[0124] S62. Based on the standardized decision matrix, calculate the objective weights of each objective dimension, and combine them with the preset subjective weight vector of the decision-maker to calculate the comprehensive weight vector of each objective dimension.

[0125] Optionally, after obtaining the standardized decision matrix, the decision terminal uses the entropy weight method to calculate the objective weights of each objective dimension. For the first... The decision-making terminal can calculate the first objective dimension. The feature proportion of each scheme in this dimension The calculation formula is: ,in, It is a very small positive number, with a dimensionless unit, used to avoid a denominator of zero. Characteristic weight The unit is dimensionless, and all schemes satisfy the condition in the th... The sum of the feature weights in each dimension equals 1. The decision terminal then calculates the weight of the [missing dimension]. Entropy value of each dimension ,in, The value of entropy ranges from 0 to 1. The smaller the value, the higher the value. The more dispersed the numerical distribution of an objective across different options, the greater the information content of that objective in the decision-making process. The decision-making terminal converts entropy values ​​into difference coefficients. Then calculate the objective weights. Objective weighting The objective weights of the four dimensions are satisfied if the sum of their objective weights equals 1. The decision-making terminal receives the subjective weight vector preset by the decision-maker. Each component is a dimensionless positive number, and their sum equals 1. The decision-making terminal can linearly combine the objective weights and subjective weights to obtain the comprehensive weight. ,in, The contribution coefficient for objective weighting, with a value ranging from 0 to 1; For the first The overall weights of the four dimensions are combined, and the sum of the overall weights of the four dimensions equals 1. The overall weights include both the discrimination information of the data itself and the decision-maker's engineering preference information.

[0126] S63. Based on the standardized decision matrix and the comprehensive weight vector, calculate the first weighted Euclidean distance between each candidate solution and the positive ideal solution, and calculate the second weighted Euclidean distance between each candidate solution and the negative ideal solution. Based on the first weighted Euclidean distance and the second weighted Euclidean distance, calculate the relative closeness of each candidate solution.

[0127] Optionally, after determining the comprehensive weights, the decision-making terminal can use the TOPSIS method to calculate the relative closeness of each scheme. The decision-making terminal first determines the ideal solution. and negative ideal solution Since all four objectives in the standardized decision matrix exhibit the characteristic of minimizing, the positive ideal solution takes the minimum value of each dimension, i.e. The negative ideal solution takes the maximum value of each dimension, that is... For the first For each candidate solution, the decision terminal calculates its weighted Euclidean distance to the positive ideal solution. ,in, The comprehensive weights calculated in step S62; For the first The distance between the first solution and the ideal optimal solution in the weighted geometric space. The decision terminal simultaneously calculates the distance between the second solution and the ideal optimal solution. The weighted Euclidean distance between each solution and the negative ideal solution ,in, For the first The distance between each alternative and the ideal worst alternative in the weighted geometric space. The decision terminal then calculates the distance between each alternative and the ideal worst alternative. The relative similarity of the two schemes ,in, The value range is from 0 to 1. Relative closeness The closer the value is to 1, the closer the solution is to the positive ideal solution and the further away it is from the negative ideal solution in a geometric sense, and the better it is in the sense of multi-objective integration.

[0128] S64. Based on the relative similarity of each candidate scheme, select the candidate scheme with the highest relative similarity as the optimal ecological protection decision scheme.

[0129] Optionally, the decision-making terminal calculates all Relative closeness of the candidate solutions arrive Next, the optimal solution selection process is executed. The decision-making terminal can then perform this... The relative proximity values ​​are arranged in descending order to obtain the sorted sequence. After sorting, the decision-making terminal selects the largest value in the sequence. The corresponding candidate scheme is taken as the optimal ecological protection decision scheme, and the global decision variable vector corresponding to this scheme is denoted as . The relative closeness score can be used to comprehensively reflect the overall degree of similarity between each candidate solution and the ideal optimal solution in four dimensions: total cost, total construction period, safety factor, and ecological coverage. Under the decision-maker's pre-set preference system, the solution with the highest relative closeness score is the compromise solution with the best overall project benefits. The decision-making terminal can... As the final output of this decision, it is input into S65 to perform scheme parsing and parameter mapping operations.

[0130] S65. Based on the global decision variable vector corresponding to the optimal ecological protection decision scheme, the protection measure type and protection parameters of each slope section are mapped from the protection candidate measure library to obtain the target protection measure configuration and target protection parameter configuration.

[0131] Optionally, the decision-making terminal determines the global decision variable vector corresponding to the optimal solution. Next, the scheme analysis operation is performed. The decision terminal can then... Extract the first Three types of decision variables for each slope section. For the variable regarding the choice of protective measure type... arrive The decision terminal identifies the dimensions with a value of 1. This dimension index Indicates the first The selected protection measure type for each section. Variables are selected for substrate type. arrive The decision terminal identifies the dimensions with a value of 1. This dimension index Indicates the first The selected substrate type for each section. Variables for selecting plant configuration schemes. arrive The decision terminal identifies the dimensions with a value of 1. This dimension index Indicates the first The selected plant configuration schemes for each section. The decision terminal can then... Retrieve the first from the protective candidate measure database. The specific names and construction process parameters of each measure. According to... Retrieve the corresponding substrate formulation parameters, which may include substrate thickness, substrate porosity, and substrate nutrient content. According to... Retrieve the corresponding plant configuration parameters, which may include plant species combination, sowing density, and expected root depth. The decision terminal will then... The combination of measure names, substrate parameters, and plant configuration parameters for each section constitutes the protection plan configuration for that section. The decision-making terminal processes all... The above analysis operation is performed sequentially on each slope section. After summarizing the configurations of all sections, the complete target protection measure configuration and target protection parameter configuration are obtained, and this configuration is used as the final output.

[0132] The aforementioned multi-objective optimization-based decision-making method for ecological protection of rock slopes first obtains geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters for each slope section. Based on these multi-source heterogeneous parameters, a multi-dimensional analysis of slope stability is conducted, which can quantitatively assess the stability level membership degree and dominant instability mode of each section, thus solving the technical problem of traditional methods lacking quantitative characterization of slope instability uncertainty. On this basis, a multi-dimensional response function of each measure to geological and hydrological conditions and stability characteristics is constructed using a database of candidate protection measures. The applicability value and expected protection effect value of each measure in each section are calculated using a coupled coordination degree model and grey relational analysis, providing a quantitative basis for measure selection and solving the technical problem of lacking systematic quantitative evaluation in manual experience-based selection. Furthermore, feasibility constraints are constructed based on the applicability value, with four optimization objectives: minimizing the total ecological protection cost, minimizing the total construction period, maximizing the overall slope safety factor, and maximizing the expected total vegetation coverage. This method constructs a multi-objective optimization objective vector function and uses a non-dominated sorting genetic algorithm to solve for the Pareto optimal solution set. This comprehensively considers the synergistic optimization of engineering economics and ecological benefits, thus solving the technical problem that traditional single-objective optimization cannot effectively balance cost, schedule, safety, and ecology. Then, based on the Pareto solution set, TOPSIS multi-attribute decision screening is performed. The relative proximity of each scheme is calculated using entropy weighting and subjective preference, and the optimal scheme is selected. This allows for the selection of the most suitable comprehensive optimal scheme from multiple non-dominated candidate schemes, thus solving the technical problem that the Pareto solution set lacks engineering decision-making direction. Finally, based on the decision variable vector of the optimal scheme, the specific measure types and protection parameters for each section are mapped from the measure library. This directly transforms the optimization results into an executable engineering design scheme, thus providing a complete closed-loop decision-making chain for the ecological protection of rock slopes, from data collection, stability analysis, measure adaptation, multi-objective optimization to scheme decision.

[0133] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0134] Based on the same inventive concept, this application also provides a multi-objective optimization-based ecological protection decision-making system for rock slopes, used to implement the aforementioned multi-objective optimization-based ecological protection decision-making method for rock slopes. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the multi-objective optimization-based ecological protection decision-making system for rock slopes provided below can be found in the limitations of the multi-objective optimization-based ecological protection decision-making method for rock slopes described above, and will not be repeated here.

[0135] In one exemplary embodiment, such as Figure 2 As shown, a schematic diagram of a multi-objective optimization-based ecological protection decision-making system 10 for rock slopes is provided, including:

[0136] The slope section parameter acquisition module 11 is used to acquire the geological structure parameters, hydrological seepage parameters and vegetation growth constraint parameters of each slope section in the rock slope.

[0137] The slope stability multidimensional analysis module 12 is used to perform multidimensional analysis of slope stability based on geological structure parameters, hydrological seepage parameters and vegetation growth constraint parameters, and obtain slope stability characteristic data.

[0138] The protective measure adaptability quantitative assessment module 13 is used to obtain a library of protective candidate measures, and based on the library, combined with geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters and slope stability characteristic data, to conduct a multi-dimensional coupled adaptability quantitative assessment of protective measures for each slope section, and obtain protective measure adaptability data; the protective measure adaptability data includes the applicability value and expected protective effect value of each protective candidate measure in each slope section;

[0139] The multi-objective optimization objective function construction module 14 is used to construct a multi-objective optimization objective vector function based on the adaptability data of protective measures;

[0140] The multi-objective optimization solution module 15 is used to optimize the multi-objective optimization objective vector function to obtain the Pareto optimal solution set of the protection scheme;

[0141] The optimal ecological protection scheme decision module 16 is used to screen and rank each candidate optimal protection scheme based on the Pareto optimal solution set of protection schemes to obtain the optimal ecological protection decision scheme; the optimal ecological protection decision scheme includes the configuration of target protection measures and the configuration of target protection parameters.

[0142] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the rock slope ecological protection decision-making method based on multi-objective optimization as described above.

[0143] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0144] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0145] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A decision-making method for ecological protection of rock slopes based on multi-objective optimization, characterized in that, The method includes: S1. Obtain the geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section in the rock slope. S2. Perform multidimensional analysis of slope stability on the geological structure parameters, hydrological seepage parameters and vegetation growth constraint parameters to obtain slope stability characteristic data; S3. Obtain a library of candidate protection measures, and based on the library, combined with the geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters, and slope stability characteristic data, conduct a multi-dimensional coupling adaptability quantitative evaluation of protection measures for each slope section to obtain protection measure adaptability data; the protection measure adaptability data includes the applicability value and expected protection effect value of each candidate protection measure in each slope section; S4. Based on the adaptability data of the protection measures, a multi-objective optimization objective vector function is constructed; S5. Optimize the multi-objective optimization objective vector function to obtain the Pareto optimal solution set of the protection scheme; S6. Based on the Pareto optimal solution set of the protection scheme, each of the candidate optimal protection schemes is screened and sorted to obtain the optimal ecological protection decision scheme; the optimal ecological protection decision scheme includes the configuration of target protection measures and the configuration of target protection parameters.

2. The method according to claim 1, characterized in that, S2 includes: S21. Normalize the geological structure parameters, hydrological seepage parameters and vegetation growth constraint parameters of each slope section to obtain a standardized slope parameter vector. S22. Based on the standardized slope parameter vector, calculate the stability level membership degree of each slope segment to obtain the membership degree matrix of each slope segment at each stability level. S23. Based on the membership matrix, the initial stability level of each slope segment is obtained, the structural plane attitude data in the geological structural parameters is extracted, and the spatial combination relationship parameters of the structural plane and the slope are calculated based on the structural plane attitude data. Based on the spatial combination relationship parameters, the dominant instability mode of each slope segment is obtained. S24. Based on the membership matrix, the initial stability level, and the dominant instability mode, obtain the slope stability characteristic data.

3. The method according to claim 2, characterized in that, S3 includes: S31. Based on the mechanism of action of each candidate protection measure in the candidate protection measure library, construct a set of response functions of each candidate protection measure to the geological structure parameters, the hydrological seepage parameters, the vegetation growth constraint parameters and the slope stability characteristic data, and obtain a multidimensional response feature vector of each candidate protection measure; S32. Based on the multidimensional response feature vector and the safety factor of each slope segment in the slope stability feature data, calculate the measure-slope coupling coordination degree of each of the protection candidate measures in each of the slope segments to obtain the coupling coordination degree matrix. S33. Based on the coupling coordination degree matrix, calculate the applicability value of each of the candidate protection measures in each of the slope sections to obtain the applicability matrix; S34. Based on the applicability matrix and the slope stabilization mechanism parameters of each candidate protection measure in the candidate protection measure library, calculate the expected protection effect value of each candidate protection measure in each slope section through the expected protection effect function to obtain the expected protection effect matrix. S35. Based on the applicability matrix and the expected protection effect matrix, the applicability value and the expected protection effect value of each of the candidate protection measures in each of the slope sections are integrated to obtain protection measure suitability data.

4. The method according to claim 1, characterized in that, S4 includes: S41. Based on the applicability value of each candidate protection measure in each slope section in the protection measure adaptability data, obtain the decision variable vector for each slope section; S42. Based on the unit area construction cost and unit area construction period of each protective candidate measure in the protective candidate measure library, and combined with the area parameters of each slope section and the decision variable vector, construct an objective function to minimize the total ecological protection cost and an objective function to minimize the total construction period; wherein, the expressions for the total ecological protection cost and the total construction period are: ; ; in, Represents the global decision variable vector. This represents the total cost of ecological protection. Indicates the total construction period. Indicates the total number of segments. This indicates the total number of candidate protective measures. Indicates the first The area of ​​each slope section Indicates the first The unit area construction cost of the various protective candidate measures No. Construction period per unit area for each of the proposed protective measures; S43. Based on the expected protection effect value in the adaptability data of the protection measures, extract the expected increment of the safety factor and the ecological restoration potential value of each candidate protection measure in each slope section, and construct the objective function for maximizing the overall slope safety factor and the objective function for maximizing the expected total vegetation coverage; wherein, the expressions of the objective function for maximizing the overall slope safety factor and the objective function for maximizing the expected total vegetation coverage are: ; ; in, Represents the global decision variable vector. This represents the objective function for maximizing the overall safety factor of the slope. This represents the objective function for maximizing the expected total vegetation cover. Indicates the total number of segments. This indicates the total number of candidate protective measures. Indicates the first Candidate protective measures in the first The expected increase in the safety factor of each section Indicates the potential value for ecological restoration. Choose a variable for binary, when the first Select the first section The value is 1 if a candidate protective measure type is selected, otherwise the value is 0. S44. Based on the applicability values ​​in the adaptability data of the protection measures, construct the feasibility constraints for the decision variables of each slope section; S45. Based on the objective function of minimizing the total ecological protection cost, the objective function of minimizing the total construction period, the objective function of maximizing the overall slope safety factor, and the objective function of maximizing the expected total vegetation coverage, and in combination with the feasibility constraints, the multi-objective optimization objective vector function is constructed.

5. The method according to claim 4, characterized in that, S5 includes: S51. Based on the decision variable vectors of each slope section and the feasibility constraints, the population is initialized using a random generation method to obtain an initial population containing multiple individuals; where each individual represents a global protection scheme decision variable vector. S52. Based on the multi-objective optimization objective vector function, calculate the objective vector value corresponding to each individual in the initial population to obtain the multi-objective fitness vector of each individual; S53. Based on the multi-objective fitness vector of each individual, perform non-dominated sorting on each individual to obtain the non-dominated level number of each individual; and calculate the crowding distance value of each individual based on the target vector value of each individual within the same non-dominated level. S54. Based on the non-dominated hierarchy number and crowding distance value of each individual, a tournament selection strategy is used to select parent individuals from the current population, and simulated binary crossover and polynomial mutation operations are performed on the selected parent individuals to generate the offspring population. S55. Merge the current population and the offspring population into a mixed population. Re-execute non-dominated sorting and crowding distance calculation on the mixed population, and select a new generation population through an elite retention strategy. Repeat the iteration until the termination condition is met to obtain the Pareto optimal solution set of the protection scheme. The Pareto optimal solution set of the protection scheme includes multiple candidate optimal protection schemes. Each candidate optimal protection scheme includes the protection measure type and protection parameters of each slope section.

6. The method according to claim 5, characterized in that, S6 includes: S61. Based on the multi-objective fitness vectors of each candidate scheme in the Pareto optimal solution set of the protection scheme, construct the candidate scheme decision matrix, and perform range normalization on the decision matrix to obtain the standardized decision matrix. S62. Based on the standardized decision matrix, calculate the objective weights of each target dimension, and combine them with the preset subjective weight vector of the decision-maker to calculate the comprehensive weight vector of each target dimension. S63. Based on the standardized decision matrix and the comprehensive weight vector, calculate the first weighted Euclidean distance between each candidate solution and the positive ideal solution, and calculate the second weighted Euclidean distance between each candidate solution and the negative ideal solution, and calculate the relative closeness of each candidate solution based on the first weighted Euclidean distance and the second weighted Euclidean distance. S64. Based on the relative proximity of each candidate scheme, select the candidate scheme with the highest relative proximity as the optimal ecological protection decision scheme; S65. Based on the global decision variable vector corresponding to the optimal ecological protection decision scheme, map the protection measure type and protection parameters of each slope section from the protection candidate measure library to obtain the target protection measure configuration and the target protection parameter configuration.

7. A decision-making system for ecological protection of rock slopes based on multi-objective optimization, characterized in that, The system includes: The slope section parameter acquisition module is used to acquire geological structure parameters, hydrological seepage parameters, and vegetation growth constraint parameters of each slope section in a rock slope. The slope stability multidimensional analysis module is used to perform multidimensional analysis of slope stability on the geological structure parameters, hydrological seepage parameters and vegetation growth constraint parameters to obtain slope stability characteristic data. The protective measure adaptability quantitative assessment module is used to acquire a library of protective candidate measures, and based on the library, combined with the geological structure parameters, hydrological seepage parameters, vegetation growth constraint parameters, and slope stability characteristic data, to conduct a multi-dimensional coupled adaptability quantitative assessment of protective measures for each slope section, thereby obtaining protective measure adaptability data; the protective measure adaptability data includes the applicability value and expected protective effect value of each protective candidate measure in each slope section; A multi-objective optimization objective function construction module is used to construct a multi-objective optimization objective vector function based on the adaptability data of the protective measures; The multi-objective optimization solution module is used to optimize and solve the multi-objective optimization objective vector function to obtain the Pareto optimal solution set of the protection scheme; The optimal ecological protection scheme decision module is used to screen and sort the candidate optimal protection schemes based on the Pareto optimal solution set of the protection schemes to obtain the optimal ecological protection decision scheme; the optimal ecological protection decision scheme includes the configuration of target protection measures and the configuration of target protection parameters.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.