Basin integrated ecological restoration technology configuration method

Abstract

The invention discloses a drainage basin integrated ecological restoration technology configuration method. The method comprises the following steps: 1) identifying drainage basin ecological problems and collecting related single ecological restoration technology information; 2) assigning scores to the ecological problems and the single ecological restoration technology in each evaluation dimension; 3) taking the single ecological restoration technology subjected to score assignment as a variable, calculating a contour coefficient after clustering analysis to determine an optimal classification number, and performing specific classification again through clustering analysis to divide the technology into technology clusters; 4) determining specific parameters of a fitness function of the genetic algorithm according to the assignment conditions of the ecological problem and the single ecological restoration technology; and 5) matching the ecological problem with the technical cluster through a genetic algorithm to realize the construction of an integrated ecological restoration technical system.

Description

Technical Field

[0001] This invention relates to the field of ecological restoration technology, and more specifically to a method for configuring integrated watershed ecological restoration technology. Background Technology

[0002] Human activities are exacerbating ecological problems, necessitating an improvement in ecological civilization. Rapid economic and social development, urban expansion, overdevelopment, and pollution have led to ecological degradation, hindering sustainable development. Therefore, ecological issues have become a major focus, with proposals to eliminate ecological stresses, optimize spatial layout, and enhance ecosystem service functions through integrated protection and restoration. However, traditional ecological restoration technologies are limited to localized problems, lack a systematic approach, and are ill-suited to addressing the challenges of complex ecosystems.

[0003] The theory and practice of integrated protection and restoration require addressing the spatial, systematic, and dynamic characteristics of watershed ecosystems. Spatiality manifests in the spatial heterogeneity of resource distribution; systematicity requires coordinating the interactions of elements such as mountains, water, forests, fields, lakes, grasslands, and sand; and dynamicity necessitates consideration of spatiotemporal scale changes. While existing research has proposed a "watershed-function-geomorphology" framework and a "goal-cost-benefit" synergistic optimization model, it suffers from three major shortcomings: a lack of theoretical research (focusing on single ecological restoration while neglecting the "integrated" aspect and insufficient interdisciplinary integration); an incomplete technical system (traditional technologies lack customized solutions and have poor long-term effectiveness); and unclear implementation paths (restoration measures are loosely connected, lacking cross-departmental collaboration, and technical combinations are difficult to adapt to regional characteristics). Therefore, existing protection measures are fragmented, and restoration effects are not sustainable, necessitating the support of a systematic technical system.

[0004] Therefore, how to achieve systematic, long-term, and effective integrated ecological restoration is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides a method for configuring integrated watershed ecological restoration technologies to solve the problem of optimizing technology combinations in watershed-scale ecological restoration, and to achieve holistic protection and restoration of the ecosystem.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A method for configuring integrated watershed ecological restoration technology includes the following steps:

[0008] Step 1: Collect information on watershed ecological problems and corresponding geographic information, as well as information on individual ecological restoration technologies. Construct an ecological problem-spatial unit correlation matrix based on the ecological problems and corresponding geographic information.

[0009] Step 2: Assign scores to ecological problems and individual ecological restoration technologies across multiple evaluation dimensions using the established scoring rules. Calculate the final scores for ecological problems and individual ecological restoration technologies across multiple evaluation dimensions using the analytic hierarchy process (AHP) based on the initial scores obtained from the scoring.

[0010] Step 3: Based on the final score of each individual ecological restoration technology, Ward hierarchical clustering is used to generate technology cluster classification results;

[0011] Step 4: Construct an ecological problem demand matrix based on the final score of the ecological problem, and select the core ecological problem demand matrix from the ecological problem demand matrix. Calculate the technology cluster capability matrix based on the final score corresponding to the technology cluster classification results.

[0012] Step 5: Based on the ecological problem demand matrix, the ecological problem core demand matrix, and the technology cluster capability matrix, a genetic algorithm is used to match and optimize the ecological problems and technology clusters to obtain the optimal ecological problem-technology cluster combination. The ecological problem-spatial unit correlation matrix is ​​then used to spatially locate the ecological problems in the optimal ecological problem-technology cluster combination to generate an integrated ecological restoration technology solution.

[0013] Preferably, the process of calculating the final score using the Analytic Hierarchy Process (AHP) is as follows:

[0014] Step 21: Construct a judgment matrix based on the collected ecological problems or individual ecological restoration technologies and evaluation dimensions;

[0015] Step 22: Divide each element in the judgment matrix by the sum of the elements in its column to obtain the normalized matrix;

[0016] Step 23: Calculate the average value of each row of elements in the normalized matrix to obtain the weight vector matrix; such as (W=[w1,w2,w3,w4,w5], where w1 to w5 represent the weight vector matrices of the five evaluation dimensions respectively;

[0017] Step 24: Based on the weight vector matrix, the initial scores for each ecological problem or individual ecological restoration technology across all evaluation dimensions are weighted and summed to obtain the final score. The initial score S_initial=[s1,s2,s3,s4,s5] is weighted and summed with the weight vector matrix W to calculate the comprehensive final score: S_final=w1 s1+w2 s2+w3 s3+w4 s4+w5 s5.

[0018] Preferably, the process of constructing the judgment matrix in step 21 is as follows:

[0019] Step 211: Construct a hierarchical structure model based on ecological problems, individual ecological restoration technologies, and evaluation dimensions, including a target layer, a criterion layer, and a solution layer. The target layer includes the final decision-making target, the criterion layer includes the evaluation dimensions, and the solution layer includes ecological problems or individual ecological restoration technologies.

[0020] Step 212: Based on the established hierarchical structure model, determine all evaluation dimensions contained in the criteria layer, and compare each evaluation dimension of the criteria layer with all other evaluation dimensions for the target layer.

[0021] Step 213: Based on the comparison results, assign scores according to relevant literature on the watershed and form a judgment matrix.

[0022] Preferably, step 23 also includes a consistency check on the weight vector matrix. If the weight vector matrix passes the check, it proceeds to step 24. If the check fails, it returns to step 21 to readjust the judgment matrix.

[0023] Preferably, the evaluation dimensions include forest degradation, soil erosion, pollution load, intensity of human activities, and impact on ecosystem service functions; forest degradation: evaluating problems such as uneven forest cover and simple structure; soil erosion: evaluating the severity of soil erosion and loss; pollution load: evaluating the degree of pollution, mainly non-point source pollution; intensity of human activities: evaluating the degree of disturbance and damage to the ecosystem caused by human activities such as mining and engineering construction; impact on ecosystem service functions: evaluating the degree of degradation of ecosystem service functions.

[0024] Preferably, the specific process of step 3 is as follows:

[0025] Step 31: Standardize the final scores of individual ecological restoration technologies;

[0026] Step 32: Set the range of cluster numbers, use Ward hierarchical clustering to perform clustering based on the final score after standardization for different cluster numbers, and calculate the silhouette coefficient for different cluster numbers based on the clustering results;

[0027] Step 33: The number of clusters corresponding to the maximum silhouette coefficient is taken as the optimal number of classifications;

[0028] Step 34: Based on the optimal number of classifications, Ward hierarchical clustering is used to cluster individual ecological restoration technologies, generating multiple technology clusters and obtaining the technology cluster classification results; using the determined optimal number of classifications, the final cluster labels of the data points are obtained, thus achieving clustering.

[0029] Preferably, the Ward hierarchical clustering method inputs the final score of each individual ecological restoration technology after standardization into the linkage function, calculates the distance between each individual ecological restoration technology, and constructs a cluster tree based on the distance. The core objective of the Ward method is to minimize the increase in total variance caused by each cluster merging, and it tends to generate clusters with relatively uniform size and regular shape using a bottom-up agglomeration strategy.

[0030] Preferably, step 34 further includes calculating the ecological function consistency index for the technology cluster, using the ecological function consistency index to test the complementarity and synergy of individual ecological restoration technologies within the cluster, and when the ecological function consistency index is greater than the set consistency threshold, the test is passed and the technology cluster classification result is output; otherwise, the test fails and the parameters of the Ward hierarchical clustering method are adjusted.

[0031] Preferably, the calculation process for the ecological function consistency index is as follows:

[0032] Step 341: Calculate the score variance of each individual ecological restoration technology based on its final score across multiple evaluation dimensions within each technology cluster;

[0033] Step 342: Normalize the reciprocal of the score variance to obtain the ecological function consistency index. The ecological function consistency index is used to measure the consistency of dominant ecological functions among technologies within the same technology cluster. The higher the index, the more consistent the dominant functions of the technologies within the cluster and the better the technological synergy.

[0034] Preferably, in step 4, the final score of the ecological problem is used as the demand vector to construct an ecological problem demand matrix. This demand matrix directly reflects the differences in the "symptoms" of different ecological problems, laying the foundation for subsequent accurate matching. From the ecological problem demand matrix, the two highest-scoring items in the demand vector corresponding to each ecological problem are selected to construct a core ecological problem demand matrix. Based on the final score corresponding to each individual ecological restoration technology within each technology cluster, the average score of each individual ecological restoration technology in each evaluation dimension is calculated, constructing an M... The technology cluster capability matrix is ​​6, where M represents the number of technology clusters.

[0035] Preferably, the matching and optimization process using a genetic algorithm includes:

[0036] Step 51: Create individuals and populations. Use the DEAP library to define individuals as technology clusters and ecological problems, and the population size is 100.

[0037] Step 52: Calculate the overall fitness of each individual based on the ecological problem demand matrix, the ecological problem core demand matrix, and the technology cluster capability matrix;

[0038] Step 53: Set the crossover probability and mutation probability, and optimize the ecological problem-technology cluster combination through selection, crossover and mutation operations;

[0039] Step 54: Iterate and optimize until the maximum number of iterations is reached, and output the optimal ecological problem-technology cluster combination.

[0040] Preferably, the specific process of step 52 is as follows:

[0041] Step 521: Calculate the capability vector for each individual based on the technology cluster capability matrix. The expression is:

[0042] tech_cluster_capability=capability_matrix[individual[i]];

[0043] Where tech_cluster_capability represents the capability vector, capability_matrix represents the technology cluster capability matrix, and individual[i] represents the i-th individual;

[0044] Step 522: Calculate the coverage based on the capability vector and the ecological problem demand matrix. The coverage represents the degree of fit between the capabilities of the technology cluster and the ecological problem demand, expressed as:

[0045] coverage=calculate_coverage(problem_demand,tech_cluster_capability);

[0046] Where, calculate_coverage represents the coverage function; problem_demand represents the ecological problem demand matrix;

[0047] Step 523: Calculate the adaptation based on the capability vector and the core ecological problem requirements matrix. Adaptation represents the degree of matching between the capabilities of the technology cluster and the problem requirements of the ecological issue. The expression is:

[0048] adaptation=calculate_adaptation(problem_demand,tech_cluster_capability,core_demand_weights);

[0049] Where, calculate_adaptation represents the fitness function; core_demand_weights represents the core demand matrix for ecological issues;

[0050] Step 524: Calculate the cost based on the capability vector. The cost represents the implementation cost of the technology cluster, and its expression is:

[0051] cost=calculate_cost(tech_cluster_capability);

[0052] Where, calculate_cost represents the cost function;

[0053] Step 525: Calculate the overall fitness based on coverage, fitness, and cost. The expression is:

[0054] fitness=w1 total_coverage+w2 total_adaptation-w3 total_cost-penalty;

[0055] penalty=penalty_factor (num_clusters-len(set(individual)))

[0056] Where w1, w2, and w3 represent weights, total_coverage represents the total coverage of all individual ecological restoration technologies within a technology cluster, total_adaptation represents the total adaptability of all individual ecological restoration technologies within a technology cluster, total_cost represents the total cost of all individual ecological restoration technologies within a technology cluster, penalty represents the penalty term, penalty_factor represents the penalty factor, num_clusters represents the number of technology clusters that the ecological problem is expected to match, and len(set(individual)) represents the number of technology clusters that the ecological problem is actually matched.

[0057] Preferably, the number of clusters ranges from [2, 10].

[0058] Preferably, the crossover probability is set to 0.7, the mutation probability to 0.2, and the maximum number of iterations to 50.

[0059] Preferably, the method also includes visualization processing, displaying the results of technology cluster classification and the optimal ecological problem-technology cluster combination. The technology cluster classification results are displayed as a tree diagram of ecological restoration technology clusters, with the name of each ecological restoration technology as a label, and the classification structure and distance sorting of the technology clusters. The optimal ecological problem-technology cluster combination is displayed as a matching relationship network diagram, using the networkx library to generate a directed graph, with the ecological problem as the starting point and the technology cluster as the ending point, to draw the matching relationship network diagram.

[0060] Preferably, a Python environment, R language, or MATLAB platform can be used to execute Ward hierarchical clustering and genetic algorithms.

[0061] As can be seen from the above technical solution, compared with the prior art, this invention discloses a method for configuring integrated watershed ecological restoration technology, realizing the construction of a full-chain integrated ecological restoration technology system from problem identification and technology classification to optimized configuration, effectively improving the systematicness, long-term effectiveness, and feasibility of watershed ecological restoration. Specifically, it includes the following beneficial effects:

[0062] (1) Systematic optimization configuration: By constructing an ecological problem-spatial unit correlation matrix, combined with the analytic hierarchy process (AHP) and multi-dimensional scoring, a systematic identification and quantitative evaluation of watershed ecological problems and restoration technologies was achieved, overcoming the problems of fragmented technology combinations and lack of integrity in traditional methods.

[0063] (2) Intelligent clustering and matching: The silhouette coefficient is used to optimize the number of clusters. The Ward hierarchical clustering method is combined to divide individual technologies into functionally coordinated technology clusters. The genetic algorithm is used to achieve efficient matching between ecological problems and technology clusters, which significantly improves the accuracy and adaptability of technology configuration.

[0064] (3) Multi-objective collaborative optimization: Introduce multi-objective functions such as coverage, fitness and cost into the genetic algorithm, and combine them with a penalty mechanism to ensure that the selected technology combination achieves the optimal balance in terms of functional matching, cost control and implementation feasibility.

[0065] (4) Enhanced visualization and interpretability: The classification and matching results of technology clusters are displayed intuitively through tree diagrams and matching relationship network diagrams, which facilitates decision-makers' understanding and implementation, and improves the operability and promotion value of the solution. Attached Figure Description

[0066] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0067] Figure 1 A schematic diagram of the configuration method for an integrated watershed ecological restoration technology provided by the present invention;

[0068] Figure 2 This is a schematic diagram illustrating the scoring of a single ecological restoration technology within the technology cluster provided by this invention;

[0069] Figure 3This is a schematic diagram of a technology cluster tree in an embodiment provided by the present invention;

[0070] Figure 4 This is a schematic diagram of the matching result structure in the embodiments provided by the present invention;

[0071] Figure 5 This is a schematic diagram of the coverage matrix provided in the embodiments of the present invention;

[0072] Figure 6 This is a schematic diagram of the fitness matrix provided in the embodiments of the present invention;

[0073] Figure 7 This is a schematic diagram of the cost matrix provided in the embodiments of the present invention. Detailed Implementation

[0074] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] This invention discloses a method for configuring integrated watershed ecological restoration technology, such as... Figure 1 As shown, it includes the following steps:

[0076] S1: Collect information on watershed ecological problems and corresponding geographic information, as well as information on individual ecological restoration technologies. Construct an ecological problem-spatial unit correlation matrix based on the ecological problems and corresponding geographic information. The ecological problem-spatial unit correlation matrix facilitates the spatial positioning of technology configuration.

[0077] S2: Using the set scoring rules, ecological problems and individual ecological restoration technologies are scored across multiple evaluation dimensions. Based on the initial scores obtained from the scoring, the analytic hierarchy process (AHP) is used to calculate the final scores of ecological problems and individual ecological restoration technologies across multiple evaluation dimensions.

[0078] S3: Based on the final score of each individual ecological restoration technology, Ward hierarchical clustering is used to generate technology cluster classification results;

[0079] S4: Construct an ecological problem demand matrix based on the final score of the ecological problem, and select the core ecological problem demand matrix from the ecological problem demand matrix. Calculate the technology cluster capability matrix based on the final score corresponding to the technology cluster classification results.

[0080] S5: Based on the ecological problem demand matrix, the ecological problem core demand matrix, and the technology cluster capability matrix, a genetic algorithm is used to match and optimize ecological problems and technology clusters to obtain the optimal ecological problem-technology cluster combination. Then, the ecological problem in the optimal ecological problem-technology cluster combination is spatially located using the ecological problem-spatial unit correlation matrix to generate an integrated ecological restoration technology solution.

[0081] Furthermore, the process of calculating the final score using the Analytic Hierarchy Process (AHP) is as follows:

[0082] S21: Construct a judgment matrix based on the collected ecological problems or individual ecological restoration technologies and evaluation dimensions;

[0083] S22: Divide each element in the judgment matrix by the sum of the elements in its column to obtain the normalized matrix;

[0084] S23: Calculate the average value of each row of elements in the normalized matrix to obtain the weight vector matrix;

[0085] S24: Based on the weight vector matrix, the initial scores of each ecological problem or individual ecological restoration technology on each evaluation dimension are weighted and summed to obtain the final score.

[0086] Furthermore, the process of constructing the judgment matrix in step 21 is as follows:

[0087] Step 211: Construct a hierarchical structure model based on ecological problems, individual ecological restoration technologies, and evaluation dimensions. This model includes a target layer, a criterion layer, and a scheme layer. The target layer includes the final decision-making objective, namely the evaluation of watershed ecological restoration. The criterion layer includes evaluation dimensions (such as forest degradation, soil erosion, etc.), which are the evaluation criteria that need to be considered to achieve the final decision-making objective. The scheme layer includes ecological problems or individual ecological restoration technologies, which are the specific objects to be evaluated.

[0088] Step 212: Based on the established hierarchical structure model, determine all evaluation dimensions contained in the criteria layer, and compare each evaluation dimension of the criteria layer with all other evaluation dimensions for the target layer.

[0089] Step 213: Based on the comparison results, assign scores according to relevant literature on the watershed and form a judgment matrix.

[0090] Furthermore, S23 also includes a consistency check on the weight vector matrix. If the check passes, the weight vector matrix proceeds to S24; if the check fails, it returns to S21 to readjust the judgment matrix.

[0091] Furthermore, the evaluation dimensions include forest degradation, soil erosion, pollution load, intensity of human activities, and impact on ecosystem service functions; forest degradation: evaluating problems such as uneven forest cover and monoculture; soil erosion: evaluating the severity of soil erosion and loss; pollution load: evaluating the degree of pollution, mainly non-point source pollution; intensity of human activities: evaluating the degree of disturbance and damage to the ecosystem caused by human activities such as mining and engineering construction; impact on ecosystem service functions: evaluating the degree of degradation of ecosystem service functions.

[0092] Furthermore, the specific process of S3 is as follows:

[0093] S31: Standardize the final score of each individual ecological restoration technology;

[0094] S32: Set the range of cluster numbers, use Ward hierarchical clustering to cluster according to the final score after standardization based on different cluster numbers, and calculate the silhouette coefficient for different cluster numbers based on the clustering results; iteratively evaluate the clustering quality of different cluster numbers (2 to 10), and use the silhouette coefficient to represent the clustering quality. The higher the silhouette coefficient, the more compact the clusters are and the better the separation between clusters.

[0095]

[0096] Where a is the average distance within a cluster, and b is the distance to the nearest cluster; the distance is calculated based on the final score.

[0097] S33: The number of clusters corresponding to the maximum silhouette coefficient is taken as the optimal number of classifications; find the number of clusters corresponding to the maximum value from the silhouette coefficient list, which is the better number of clusters for the current data, and take it as the optimal number of classifications.

[0098] S34: Based on the optimal number of classifications, Ward hierarchical clustering is used to cluster individual ecological restoration technologies, generating multiple technology clusters and obtaining technology cluster classification results; using the determined optimal number of classifications, the final cluster labels of data points are obtained, thus achieving clustering.

[0099] Furthermore, the Ward hierarchical clustering method inputs the standardized final score of each individual ecological restoration technology into the linkage function to calculate the distance between each individual ecological restoration technology. Based on the distance, clustering is performed to construct a cluster tree and obtain the technology cluster classification results. The core objective of the Ward method is to minimize the increase in total variance caused by each cluster merging, and it tends to generate clusters with relatively uniform size and regular shape, using a bottom-up agglomeration strategy.

[0100] Furthermore, S34 also includes calculating the ecological function consistency index for the technology cluster, using the ecological function consistency index to test the complementarity and synergy of individual ecological restoration technologies within the cluster. When the ecological function consistency index is greater than the set consistency threshold, the test passes and the technology cluster classification result is output; otherwise, the test fails and the parameters of the Ward hierarchical clustering method are adjusted.

[0101] Furthermore, the calculation process for the ecological function consistency index is as follows:

[0102] S341: Calculate the score variance of each individual ecological restoration technology based on the final scores of each individual ecological restoration technology in each technology cluster across multiple evaluation dimensions;

[0103] S342: Normalize the reciprocal of the score variance to obtain the ecological function consistency index. The ecological function consistency index is used to measure the consistency of the dominant ecological functions of various technologies within the same technology cluster. The higher the index, the more consistent the dominant functions of the technologies within the cluster and the better the technological synergy.

[0104] Furthermore, in S4, the final scores of ecological issues are used as demand vectors to construct an ecological issue demand matrix. This demand matrix directly reflects the differences in the "symptoms" of different ecological issues, laying the foundation for subsequent accurate matching. For example, the demand vector for "low and uneven forest coverage" is [0.9, 0.7, 0.2, 0.6, 0.8], but its core needs are ecosystem service improvement and soil and water conservation, with a lower weight for the pollution load dimension. The two highest-scoring items in the demand vectors corresponding to each ecological issue are selected from the ecological issue demand matrix to construct a core ecological issue demand matrix. For example, [0.9, 0.8] is selected from the demand vector for "low and uneven forest coverage". Based on the final scores of individual ecological restoration technologies within each technology cluster, the average score of each individual ecological restoration technology across various evaluation dimensions is calculated to construct an M... The technology cluster capability matrix is ​​6, where M represents the number of technology clusters.

[0105] Furthermore, the process of matching and optimization using genetic algorithms includes:

[0106] S51: Create individuals and populations, using the DEAP library to define individuals as technology clusters and ecological issues, with a population size of 100;

[0107] S52: Calculate the overall fitness of each individual based on the ecological problem demand matrix, the ecological problem core demand matrix, and the technology cluster capability matrix;

[0108] S53: Set crossover and mutation probabilities, and optimize the combination of ecological problems and technology clusters through selection, crossover, and mutation operations;

[0109] S54: Iterate and optimize until the maximum number of iterations is reached, and output the optimal ecological problem-technology cluster combination.

[0110] Furthermore, the specific process of S52 is as follows:

[0111] S521: Calculate the capability vector for each individual based on the technology cluster capability matrix, expressed as follows:

[0112] tech_cluster_capability=capability_matrix[individual[i]];

[0113] Where tech_cluster_capability represents the capability vector, capability_matrix represents the technology cluster capability matrix, and individual[i] represents the i-th individual;

[0114] S522: Calculate coverage based on the capability vector. Coverage represents the degree of alignment between the capabilities of a technology cluster and the needs of the ecosystem problem. The expression is:

[0115] coverage=calculate_coverage(problem_demand,tech_cluster_capability);

[0116] Where, calculate_coverage represents the coverage function; problem_demand represents the ecological problem demand matrix;

[0117] S523: Calculate the adaptation based on the capability vector and the ecological problem demand matrix. Adaptation represents the degree of matching between the capabilities of the technology cluster and the problem demands of the ecological issues. The expression is:

[0118] adaptation=calculate_adaptation(problem_demand,tech_cluster_capability,core_demand_weights);

[0119] Where, calculate_adaptation represents the fitness function; tech_cluster_capability represents the capability vector; and core_demand_weights represents the core demand matrix for ecological issues;

[0120] S524: Calculate the cost based on the capability vector. The cost represents the implementation cost of the technology cluster, expressed as:

[0121] cost=calculate_cost(tech_cluster_capability);

[0122] Where, calculate_cost represents the cost function;

[0123] S525: Calculate overall fitness based on coverage, fitness, and cost, using the following expression:

[0124] fitness=w1 total_coverage+w2 total_adaptation-w3 total_cost-penalty;

[0125] penalty=penalty_factor (num_clusters-len(set(individual)))

[0126] Where w1, w2, and w3 represent weights, total_coverage represents the total coverage of all individual ecological restoration technologies within a technology cluster, total_adaptation represents the total adaptability of all individual ecological restoration technologies within a technology cluster, total_cost represents the total cost of all individual ecological restoration technologies within a technology cluster, penalty represents the penalty term, penalty_factor represents the penalty factor, num_clusters represents the number of technology clusters that the ecological problem is expected to match, and len(set(individual)) represents the number of technology clusters that the ecological problem is actually matched.

[0127] Furthermore, the set cluster number range is [2, 10].

[0128] Furthermore, the crossover probability is set to 0.7, the mutation probability to 0.2, and the maximum number of iterations to 50.

[0129] Furthermore, it also includes visualization processing, displaying the results of technology cluster classification and the optimal ecological problem-technology cluster combination. The technology cluster classification results are displayed as a tree diagram of ecological restoration technology clusters, with the name of each ecological restoration technology as a label, and the classification structure and distance sorting of the technology clusters. The optimal ecological problem-technology cluster combination is displayed as a matching relationship network diagram, using the networkx library to generate a directed graph, with the ecological problem as the starting point and the technology cluster as the ending point, to draw the matching relationship network diagram.

[0130] Furthermore, Python, R, or MATLAB can be used to execute Ward's hierarchical clustering and genetic algorithms. Python utilizes the Scikit-learn library for clustering, MATLAB uses its built-in clustering algorithm library, and R uses the cluster library.

[0131] On the other hand, in a specific embodiment, taking a certain watershed as an example, the technical configuration is implemented using a Python 3.8 environment. The main dependent libraries include: pandas (data processing), scikit-learn (cluster analysis), DEAP (genetic algorithm), and networkx (visualization), specifically including the following steps.

[0132] 1. Ecological problem identification and technical information collection

[0133] Using ArcGIS 10.7, we identified core ecological problems in the watershed and established a list of problems, determining the corresponding geographic information for each ecological problem. We collected relevant information on individual ecological restoration technologies and formed a technical attribute table. Based on the ecological problems and corresponding geographic information, we constructed an ecological problem-spatial unit association matrix to facilitate the spatial positioning of technical configurations.

[0134] 2. Scoring and evaluation;

[0135] The identified ecological problems and individual ecological restoration technologies were scored across various evaluation dimensions, as shown in Table 1 below.

[0136] Table 1. Scoring Table for Each Evaluation Dimension of Ecological Issues and Individual Ecological Restoration Technologies

[0137]

[0138] 3. Cluster analysis generates technology clusters;

[0139] Data standardization: StandardScaler() is used to normalize the scoring data of individual ecological restoration technologies; individual ecological restoration technologies and their corresponding scoring data within a technology cluster are as follows: Figure 2 As shown;

[0140] Determining the optimal number of clusters: Set the maximum number of clusters to 10, calculate the silhouette coefficients for different numbers of clusters based on the normalized assigned data, and take the number of clusters corresponding to the maximum silhouette coefficient as the optimal number of clusters;

[0141] Hierarchical clustering execution: Based on the optimal number of clusters, Ward's hierarchical clustering method is used to generate clusters, and a dendrogram is output, such as... Figure 3 As shown.

[0142] 4. Calculation of fitness function parameters;

[0143] Construct an ecological problem demand matrix based on the scoring data of ecological problems;

[0144] Calculate the technology cluster capability matrix based on the scoring data corresponding to the technology cluster classification results;

[0145] Calculate the capability vector based on the technology cluster capability matrix, and then use the capability vector to calculate the coverage matrix, such as... Figure 5 As shown, the expression is:

[0146] coverage=calculate_coverage(problem_demand,tech_cluster_capability)

[0147] The adaptation matrix is ​​calculated based on the capability vector and the ecological problem demand matrix, such as... Figure 6 As shown;

[0148] Calculate the cost matrix based on the capability vector, such as Figure 7 As shown, the expression is: Cost[i,j]=100-(ability score) 48) The greater the capability, the higher the cost;

[0149] 5. Genetic algorithm matching techniques and ecological problems: searching for the optimal matching combination;

[0150] Fitness function design:

[0151] ;

[0152] ;

[0153] ;

[0154] penalty=0.1 (num_clusters-len(set(individual))) penalizes unused clusters;

[0155] fitness=(0.5 total_coverage + 0.5 total_adaptation-0.01 total_cost-penalty,)

[0156] Genetic operations: two-point crossover (probability 0.7), uniform mutation (probability 0.2), iterate for 50 generations to output the optimal solution; fitness represents the overall fitness.

[0157] Visualization output: The networkx utility is used to draw a directed graph to display the structure diagram, such as... Figure 4 As shown.

[0158] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0159] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for configuring integrated watershed ecological restoration technology, characterized in that, Includes the following steps: Step 1: Collect information on watershed ecological problems and corresponding geographic information, as well as information on individual ecological restoration technologies. Construct an ecological problem-spatial unit correlation matrix based on the ecological problems and corresponding geographic information. Step 2: Assign scores to ecological problems and individual ecological restoration technologies across multiple evaluation dimensions using the established scoring rules. Calculate the final scores for ecological problems and individual ecological restoration technologies across multiple evaluation dimensions using the analytic hierarchy process (AHP) based on the initial scores obtained from the scoring. Step 3: Based on the final score of each individual ecological restoration technology, Ward hierarchical clustering is used to generate technology cluster classification results; Step 4: Construct an ecological problem demand matrix based on the final score of the ecological problem, and select the core ecological problem demand matrix from the ecological problem demand matrix. Calculate the technology cluster capability matrix based on the final score corresponding to the technology cluster classification results. Step 5: Based on the ecological problem demand matrix, the ecological problem core demand matrix, and the technology cluster capability matrix, a genetic algorithm is used to match and optimize the ecological problems and technology clusters to obtain the optimal ecological problem-technology cluster combination. The ecological problem-spatial unit correlation matrix is ​​then used to spatially locate the ecological problems in the optimal ecological problem-technology cluster combination to generate an integrated ecological restoration technology solution.

2. The method for configuring integrated watershed ecological restoration technology according to claim 1, characterized in that, The process of calculating the final score using the analytic hierarchy process is as follows: Step 21: Construct a judgment matrix based on the collected ecological problems or individual ecological restoration technologies and evaluation dimensions; Step 22: Divide each element in the judgment matrix by the sum of the elements in its column to obtain the normalized matrix; Step 23: Calculate the average value of each row of elements in the normalized matrix to obtain the weight vector matrix; Step 24: Use the weight vector matrix to sum the initial scores of each ecological problem or individual ecological restoration technology across various evaluation dimensions to obtain the final score.

3. The method for configuring integrated watershed ecological restoration technology according to claim 2, characterized in that, Step 23 also includes a consistency check on the weight vector matrix. If the weight vector matrix passes the check, it proceeds to step 24. If the check fails, it returns to step 21 to readjust the judgment matrix.

4. The method for configuring integrated watershed ecological restoration technology according to claim 1, characterized in that, The specific process of step 3 is as follows: Step 31: Standardize the final scores of individual ecological restoration technologies; Step 32: Set the range of cluster numbers, use Ward hierarchical clustering to perform clustering based on the final score after standardization for different cluster numbers, and calculate the silhouette coefficient for different cluster numbers based on the clustering results; Step 33: The number of clusters corresponding to the maximum silhouette coefficient is taken as the optimal number of classifications; Step 34: Based on the optimal number of classifications, Ward hierarchical clustering is used to cluster individual ecological restoration technologies, generating multiple technology clusters and obtaining the technology cluster classification results.

5. The method for configuring integrated watershed ecological restoration technology according to claim 4, characterized in that, Step 34 also includes calculating the ecological function consistency index for the technology cluster, and using the ecological function consistency index to test the complementarity and synergy of individual ecological restoration technologies within the technology cluster. When the ecological function consistency index is greater than the set consistency threshold, the test is passed and the technology cluster classification result is output; otherwise, the test fails and the parameters of the Ward hierarchical clustering method are adjusted.

6. The method for configuring integrated watershed ecological restoration technology according to claim 5, characterized in that, The calculation process for the ecological function consistency index is as follows: Step 341: Calculate the score variance of each individual ecological restoration technology based on the final scores of each individual ecological restoration technology in each technology cluster across multiple evaluation dimensions; Step 342: Normalize the inverse of the score variance to obtain the ecological function consistency index.

7. The method for configuring integrated watershed ecological restoration technology according to claim 4, characterized in that, In step 4, the final score of the ecological problem is used as the demand vector to construct the ecological problem demand matrix; the two highest-scoring items in the demand vector corresponding to each ecological problem are selected from the ecological problem demand matrix to construct the core ecological problem demand matrix; based on the final score of each individual ecological restoration technology within each technology cluster, the average score of each individual ecological restoration technology in each evaluation dimension is calculated to construct the technology cluster capability matrix.

8. The method for configuring integrated watershed ecological restoration technology according to claim 4, characterized in that, The process of matching and optimization using genetic algorithms includes: Step 51: Create individuals and populations, with individuals representing technology clusters and ecological issues; Step 52: Calculate the overall fitness of each individual based on the ecological problem demand matrix, the ecological problem core demand matrix, and the technology cluster capability matrix; Step 53: Set the crossover and mutation probabilities, and optimize the ecological problem-technology cluster combination through selection, crossover, and mutation operations; Step 54: Iterate and optimize until the maximum number of iterations is reached, and output the optimal ecological problem-technology cluster combination.

9. The method for configuring integrated watershed ecological restoration technology according to claim 8, characterized in that, The specific process of step 52 is as follows: Step 521: Calculate the capability vector for each individual based on the technology cluster capability matrix. The expression is: Step 522: Calculate the coverage rate based on the capability vector and the ecological problem demand matrix; Step 523: Calculate adaptability based on the capability vector and the core requirements matrix of ecological issues; Step 524: Calculate the cost based on the capability vector; Step 525: Calculate the overall fitness based on coverage, fitness, and cost.

10. The method for configuring integrated watershed ecological restoration technology according to claim 1, characterized in that, It also includes visualization processing, displaying the results of technology cluster classification and the optimal ecological problem-technology cluster combination; among them, the technology cluster classification results are displayed as a tree diagram of ecological restoration technology clusters, and the optimal ecological problem-technology cluster combination is displayed as a matching relationship network diagram.

Images