Arrangement method and system for cultivated land protection gathering area

Abstract

The invention provides a cultivated land protection gathering area layout method and system, and the method comprises the steps: obtaining evaluation data of a target area under a cultivated land gathering multi-index evaluation system, carrying out the spatial gathering potential evaluation based on the evaluation data, and obtaining a cultivated land gathering potential evaluation result; screening based on the cultivated land gathering potential evaluation result to obtain a preliminary cultivated land gathering potential area; analyzing under a coupling coordination degree model based on the initial cultivated land gathering potential area to obtain a coupling coordination degree, and performing screening and space filling under a preset coordination degree threshold value based on the coupling coordination degree to obtain an optimal space of the cultivated land gathering key area; local autocorrelation analysis is carried out based on the cultivated land gathering key area optimal space, and the gathering type of each space unit in the cultivated land gathering key area optimal space is obtained; and obtaining a minimum cultivated land demand area, and performing space optimization processing based on the aggregation type to obtain a cultivated land protection aggregation area layout result. According to the invention, the accuracy of the layout result of the cultivated land protection gathering area can be improved.

Description

Technical Field

[0001] This invention belongs to the field of natural resource management technology, and in particular relates to a method and system for the layout of farmland protection clusters. Background Technology

[0002] Against the backdrop of rapid urbanization, the fragmentation of arable land is becoming increasingly serious, which greatly restricts the large-scale development of modern agriculture. Therefore, it is necessary to optimize the delineation and layout of arable land protection clusters in order to achieve efficient utilization of arable land resources.

[0003] Existing farmland protection cluster layout schemes use the natural quality of farmland itself as the evaluation target and delineate the layout under a linear weighted model. However, the layout results obtained by this scheme may include high-quality plots located in the core areas of urban expansion, leading to functional conflicts within the delineated areas and reducing the accuracy of the farmland protection cluster layout results. Furthermore, existing farmland protection cluster layout schemes achieve regional connectivity through simple buffer analysis and patch merging, but they cannot distinguish whether a high-quality plot is surrounded by equally high-quality areas or by low-quality areas, ignoring spatial correlation. Therefore, the farmland protection clusters they design may simply be a collection of spatially adjacent discrete patches, resulting in low accuracy of the farmland protection cluster layout results. Summary of the Invention

[0004] The present invention aims to provide a method and system for the layout of farmland protection clusters to solve the above-mentioned technical problems and improve the accuracy of the layout results of farmland protection clusters.

[0005] To address the aforementioned technical problems, this invention provides a method for the layout of farmland protection clusters, comprising the following steps:

[0006] The evaluation data of the target area under the preset multi-index evaluation system for farmland agglomeration is obtained, and the spatial agglomeration potential is evaluated based on the evaluation data to obtain the farmland agglomeration potential evaluation result.

[0007] Based on the evaluation results of the farmland agglomeration potential, a preliminary farmland agglomeration potential area is obtained through screening.

[0008] Based on the preliminary potential areas for farmland agglomeration, analysis and processing are performed under the coupling coordination degree model to obtain the coupling coordination degree. Based on the coupling coordination degree, screening and spatial filling are performed under a preset coordination degree threshold to obtain the preferred space for key areas of farmland agglomeration.

[0009] Based on the optimized space of the key areas of farmland agglomeration, local autocorrelation analysis is performed to obtain the agglomeration type of each spatial unit in the optimized space of the key areas of farmland agglomeration;

[0010] Obtain the minimum arable land requirement area corresponding to the target area, and perform spatial optimization processing under the search algorithm based on the cluster type and the minimum arable land requirement area to obtain the layout result of the arable land protection cluster area corresponding to the target area.

[0011] In the aforementioned scheme, the potential areas for farmland agglomeration are analyzed and processed under a coupling coordination degree model to obtain the coupling coordination degree, which quantifies the synergistic constraints among multiple indicators. Subsequently, this scheme uses a coordination degree threshold for screening and fills in the voids in concentrated plots, meeting the needs of contiguous farmland protection and accurately identifying functionally coordinated areas, thereby improving the accuracy of the farmland protection agglomeration area layout results. Based on the functional coordination of each area in the optimized space of the key farmland agglomeration area, this scheme performs local autocorrelation analysis on the optimized space of the key farmland agglomeration area to clarify the spatial agglomeration patterns of each spatial unit in the optimized space of the key farmland agglomeration area. Under the constraint of the minimum farmland demand area, further spatial optimization is performed using a search algorithm, which can output the layout results of farmland protection agglomeration areas with high spatial agglomeration, improving the accuracy of the farmland protection agglomeration area layout results.

[0012] Further, before acquiring evaluation data of the target area under the preset multi-indicator evaluation system for cultivated land agglomeration, and before conducting spatial agglomeration potential evaluation based on the evaluation data to obtain the cultivated land agglomeration potential evaluation result, the construction of the multi-indicator evaluation system for cultivated land agglomeration includes: selecting multiple indicators according to preset evaluation dimensions; the evaluation dimensions include resource endowment matching, service function matching, and regional coordinated development matching; obtaining the quantitative indicator scores corresponding to each quantitative indicator using a segmented method based on the quantitative indicators among the indicators; obtaining the qualitative indicator scores corresponding to each qualitative indicator using a derived Delphi method based on the qualitative indicators among the indicators; performing normalization processing on the quantitative indicator scores and the qualitative indicator scores to obtain the indicator scores corresponding to each indicator; obtaining the weights corresponding to each indicator using the CRITIC method based on the indicator scores; and constructing the multi-indicator evaluation system for cultivated land agglomeration based on the indicators, the weights, and the indicator scores.

[0013] In the aforementioned scheme, resource endowment matching characterizes the degree of arable land agglomeration potential under the influence of natural resource background and human intervention; service function matching characterizes the production, living, and ecological benefits and value in the process of arable land agglomeration protection and utilization; and regional coordinated development matching characterizes the degree of coordination among production, living, and ecology in the process of arable land agglomeration protection and utilization. This scheme selects indicators from three pre-set evaluation dimensions—resource endowment matching, service function matching, and regional coordinated development matching—ensuring the comprehensiveness of the selected indicators and overcoming the limitations of single-dimensional indicators.

[0014] Furthermore, this scheme employs a segmented method and a derived Delphi method to determine the scores for quantitative and qualitative indicators respectively, ensuring the reliability of the indicator scores. Normalization of all indicator scores eliminates the impact of differences in the dimensions of different indicators. The CRITIC method is used to calculate the weights of each indicator, fully exploring the information contained in the similarities and conflicts between indicator data; that is, the greater the information content of an indicator, the greater its weight, and the greater its influence on the evaluation results. Therefore, this scheme provides a reliable quantitative basis for the subsequent spatial agglomeration potential evaluation process, improves the reliability of the cultivated land agglomeration potential evaluation results, and thus improves the accuracy of the layout results of cultivated land protection agglomeration areas.

[0015] Further, the step of acquiring evaluation data of the target area under a preset multi-indicator evaluation system for cultivated land agglomeration, and evaluating spatial agglomeration potential based on the evaluation data to obtain cultivated land agglomeration potential evaluation results, includes: performing grade classification processing based on the indicators, weights, and indicator scores corresponding to each evaluation dimension in the multi-indicator evaluation system for cultivated land agglomeration, to obtain a cultivated land agglomeration potential evaluation model; acquiring evaluation data of the target area under the preset multi-indicator evaluation system for cultivated land agglomeration; the evaluation data includes evaluation data of each regional unit in the target area corresponding to each evaluation dimension in the multi-indicator evaluation system for cultivated land agglomeration; and obtaining cultivated land agglomeration potential evaluation results corresponding to each regional unit under the cultivated land agglomeration potential evaluation model based on the evaluation data.

[0016] In the above scheme, the classification is based on indicators, weights, and indicator scores to ensure the consistency of the evaluation process in the constructed farmland agglomeration potential evaluation model. Based on this farmland agglomeration potential evaluation model, this scheme inputs the evaluation data of the target area under the preset multi-indicator evaluation system for farmland agglomeration into the farmland agglomeration potential evaluation model for processing. The resulting farmland agglomeration potential evaluation results have high reliability, and the evaluation results corresponding to different regional units have good comparability.

[0017] Further, the step of analyzing and processing the preliminary arable land agglomeration potential area under the coupling coordination degree model to obtain the coupling coordination degree, and then performing screening and spatial filling under a preset coordination degree threshold based on the coupling coordination degree to obtain the preferred space of key arable land agglomeration areas, includes: analyzing and processing the preliminary arable land agglomeration potential area under the coupling coordination degree model to obtain the coupling coordination degree corresponding to each potential area unit in the preliminary arable land agglomeration potential area; screening the preliminary arable land agglomeration potential area and the coupling coordination degree under a preset first coordination degree threshold to obtain the preferred space of initial key arable land agglomeration areas; and performing spatial filling under a preset second coordination degree threshold based on the preferred space of initial key arable land agglomeration areas to obtain the preferred space of key arable land agglomeration areas.

[0018] In the above scheme, by performing coupling coordination degree analysis on each potential area unit, high-quality units with high internal coordination can be identified. Furthermore, this scheme, through a screening process using a preset first coordination degree threshold, can effectively eliminate discrete or functionally conflicting areas with low coupling coordination degree, ensuring that the initial preferred space consists of core areas with good coordination and high stability. Moreover, this scheme introduces a preset second coordination degree threshold to fill in the initial preferred space, effectively connecting adjacent high-quality units and filling internal gaps without significantly reducing the overall coordination level. This ultimately forms a preferred space for key farmland agglomeration areas with good contiguousness, high internal coordination, and a more complete and stable spatial structure, thereby improving the accuracy of the subsequent farmland protection agglomeration area layout results.

[0019] Furthermore, the step of performing local autocorrelation analysis based on the selected space of the key cultivated land agglomeration area to obtain the agglomeration type of each spatial unit in the selected space of the key cultivated land agglomeration area includes: performing local autocorrelation analysis based on the selected space of the key cultivated land agglomeration area to obtain the LISA cluster map corresponding to each evaluation dimension in the preset multi-index evaluation system for cultivated land agglomeration; and obtaining the cluster type of each spatial unit in the selected space of the key cultivated land agglomeration area in each evaluation dimension based on the LISA cluster map.

[0020] The above-mentioned scheme, through local autocorrelation analysis of the selected space in the key areas of farmland agglomeration, can accurately reveal the spatial correlation between each spatial unit and its neighboring spatial units on the preset evaluation dimensions. Furthermore, based on the LISA clustering map corresponding to each evaluation dimension, this scheme obtains the clustering type of each spatial unit in each evaluation dimension, which can provide a reliable quantitative basis for the subsequent spatial optimization process to characterize the attributes of the spatial unit itself and its relationship with the surrounding spatial units, thereby improving the accuracy of the subsequent farmland protection agglomeration layout results.

[0021] Further, the step of obtaining the minimum arable land requirement area corresponding to the target area, and performing spatial optimization processing under a search algorithm based on the clustering type and the minimum arable land requirement area to obtain the layout result of the arable land protection cluster area corresponding to the target area, includes: performing quantitative overlay processing based on the clustering type to obtain the comprehensive score corresponding to each spatial unit in the preferred space of the key arable land cluster area; based on the comprehensive score and the preferred space of the key arable land cluster area, obtaining a preliminary set of arable land protection cluster areas and a set of map patches under a preset comprehensive score threshold; and based on the set of map patches and the preliminary set of arable land protection cluster areas... The process involves: 1) determining the clustering distances of each patch in the patch set; where the clustering distance is the minimum distance between the patch and the initial farmland protection cluster; 2) sorting the patches according to the comprehensive score and the clustering distance to obtain a patch queue; 3) obtaining the minimum farmland requirement area corresponding to the target area, and 4) performing spatial iterative optimization steps based on the patch queue, the clustering distance, and the minimum farmland requirement area to obtain farmland protection clusters; and 5) obtaining the farmland protection cluster layout results corresponding to the target area based on the farmland protection clusters.

[0022] The aforementioned scheme quantifies and superimposes the clustering types corresponding to each evaluation dimension of each spatial unit to obtain a comprehensive score, integrating multi-dimensional spatial correlation information into a unified quantitative score. Based on a preset comprehensive score threshold, this scheme can accurately identify core areas with better comprehensive conditions in the selected space of key cultivated land clustering areas as preliminary cultivated land protection clustering areas, while also clarifying the set of map patches to be optimized. Furthermore, this scheme generates a map patch queue based on the clustering distance and comprehensive score of each map patch. The resulting map patch queue ensures that subsequent spatial iterative optimization steps can prioritize the processing of map patches with better comprehensive conditions. Moreover, this scheme performs spatial iterative optimization based on this map patch queue, and under the premise of meeting the minimum cultivated land area requirement constraint, it can orderly merge adjacent high-quality map patches into the core area, thereby outputting the target area cultivated land protection clustering area layout result with complete spatial structure, compact and reasonable layout, and area meeting the standard, thus improving the accuracy of the cultivated land protection clustering area layout result.

[0023] Further, the step of obtaining the minimum arable land requirement area corresponding to the target area and performing a spatial iterative optimization step based on the map patch queue, the clustering distance, and the minimum arable land requirement area to obtain the arable land protection cluster area includes: obtaining the state of the map patch queue and the cumulative area of ​​the preliminary arable land protection cluster area set; if the state is empty or the cumulative area reaches the minimum arable land requirement area, then the preliminary arable land protection cluster area set is taken as the arable land protection cluster area, and the spatial iterative optimization step is terminated; otherwise, the current map patch is taken from the head of the map patch queue; if the clustering distance corresponding to the current map patch is not less than a preset first distance threshold, then the current map patch is removed from the map patch queue, and the spatial iterative optimization step is re-executed; otherwise, it is determined that the current map patch is related to any of the preliminary arable land protection cluster areas. The system checks whether the distance between the initial farmland protection clusters meets the preset search distance. If the distance between the current patch and any of the initial farmland protection clusters meets the preset search distance, the current patch is merged into the initial farmland protection cluster, removed from the patch queue, and the spatial iterative optimization step is re-executed. Otherwise, it checks whether there are other patches within a radius of the current patch and the preset search distance. If there are other patches within a radius of the current patch and the preset search distance, the current patch is removed from the patch queue and inserted at the end of the patch queue, and the spatial iterative optimization step is re-executed. Otherwise, the current patch is removed from the patch queue, and the spatial iterative optimization step is re-executed.

[0024] The aforementioned scheme continuously monitors the status and cumulative area of ​​the land parcel queue and compares it with the minimum required arable land area, ensuring that the final arable land protection clusters meet the standards in terms of quantity or area. Furthermore, by comparing the current land parcel's clustering distance with a preset first distance threshold, this scheme can quickly filter and eliminate spatially dispersed land parcels with low merging value, significantly improving the efficiency of spatial optimization. Moreover, by determining whether the distance between the current land parcel and the cluster meets the preset search distance, this scheme can identify and merge high-quality land parcels that are spatially adjacent and meet the merging conditions, effectively promoting the reasonable expansion of the arable land protection cluster scale. Furthermore, by determining whether other land parcels exist within the preset search distance range and adjusting the current land parcel's position in the queue accordingly, this scheme avoids local optima in complex target areas and ensures that the final output is a highly concentrated, compact, and area-compliant arable land protection cluster.

[0025] Furthermore, the step of obtaining the minimum arable land demand area corresponding to the target area and performing a spatial iterative optimization step based on the map patch queue, the clustering distance, and the minimum arable land demand area to generate the minimum arable land demand area in the arable land protection cluster area includes: obtaining historical population data and performing prediction processing under a logistic model based on the historical population data to obtain a population change curve; and performing scale calculation processing based on the population change curve to obtain the minimum arable land demand area. The above scheme, by obtaining historical population data and using a logistic model for prediction processing, can accurately depict the dynamic trends and long-term patterns of population development, providing a reliable basis for subsequent scale calculations. Moreover, this scheme, based on the population change curve for scale calculation processing, ensures that the minimum arable land demand area is a quantitative result dynamically correlated with the population changes in the target area, thereby ensuring that the protection scale of the final arable land protection cluster area layout result can match future needs, thus improving the accuracy of the arable land protection cluster area layout result.

[0026] This invention also provides a farmland protection cluster layout system, comprising: a clustering potential evaluation result acquisition module, used to acquire evaluation data of a target area under a preset multi-index evaluation system for farmland clustering, and based on the evaluation data, perform spatial clustering potential evaluation to obtain farmland clustering potential evaluation results; a preliminary screening processing module, used to perform screening processing based on the farmland clustering potential evaluation results obtained by the clustering potential evaluation result acquisition module to obtain preliminary farmland clustering potential areas; and a preferred spatial acquisition module, used to perform analysis and processing based on the preliminary farmland clustering potential areas obtained by the preliminary screening processing module under a coupling coordination degree model to obtain coupling coordination degree, and based on... The coupling coordination degree is used to filter and fill spaces under a preset coordination degree threshold to obtain the preferred space of key areas for farmland agglomeration; the local autocorrelation analysis module is used to perform local autocorrelation analysis on the preferred space of key areas for farmland agglomeration obtained by the preferred space acquisition module to obtain the agglomeration type of each spatial unit in the preferred space of key areas for farmland agglomeration; the farmland protection agglomeration area layout result acquisition module is used to obtain the minimum farmland demand area corresponding to the target area, and based on the agglomeration type obtained by the local autocorrelation analysis module and the minimum farmland demand area, perform spatial optimization processing under the search algorithm to obtain the layout result of farmland protection agglomeration area corresponding to the target area.

[0027] Further, before acquiring evaluation data of the target area under the preset multi-indicator evaluation system for cultivated land agglomeration, and before conducting spatial agglomeration potential evaluation based on the evaluation data to obtain the cultivated land agglomeration potential evaluation result, the construction of the multi-indicator evaluation system for cultivated land agglomeration includes: selecting multiple indicators according to preset evaluation dimensions; the evaluation dimensions include resource endowment matching, service function matching, and regional coordinated development matching; obtaining the quantitative indicator scores corresponding to each quantitative indicator using a segmented method based on the quantitative indicators among the indicators; obtaining the qualitative indicator scores corresponding to each qualitative indicator using a derived Delphi method based on the qualitative indicators among the indicators; performing normalization processing on the quantitative indicator scores and the qualitative indicator scores to obtain the indicator scores corresponding to each indicator; obtaining the weights corresponding to each indicator using the CRITIC method based on the indicator scores; and constructing the multi-indicator evaluation system for cultivated land agglomeration based on the indicators, the weights, and the indicator scores.

[0028] The aforementioned scheme analyzes and processes the preliminary potential areas for farmland agglomeration under a coupling coordination degree model, which can obtain the coupling coordination degree of quantifying the synergistic constraints among multiple indicators. Subsequently, this scheme filters based on the coordination degree threshold and fills in the void areas of concentrated plots, which can meet the needs of contiguous farmland protection and accurately identify functionally coordinated areas, thus improving the accuracy of the obtained farmland protection agglomeration area layout results. Furthermore, this scheme performs local autocorrelation analysis based on the optimized space of key farmland agglomeration areas, which can clarify the spatial agglomeration patterns of each spatial unit in the optimized space of key farmland agglomeration areas. Under the constraint of minimum farmland demand area, this scheme further utilizes a search algorithm for spatial optimization, which can output spatially highly agglomerated farmland protection agglomeration area layout results, improving the accuracy of the farmland protection agglomeration area layout results. Attached Figure Description

[0029] Figure 1 A flowchart illustrating the steps of a method for layout of farmland protection clusters provided in an embodiment of the present invention;

[0030] Figure 2 A population change curve diagram for a farmland protection cluster layout method provided in an embodiment of the present invention;

[0031] Figure 3 This is a schematic diagram of a farmland protection cluster layout system provided in an embodiment of the present invention. Detailed Implementation

[0032] 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.

[0033] Please see Figure 1 This embodiment provides a method for the layout of farmland protection clusters, including the following steps:

[0034] Step S1: Obtain evaluation data of the target area under the preset multi-index evaluation system for farmland agglomeration, and based on the evaluation data, conduct spatial agglomeration potential evaluation to obtain farmland agglomeration potential evaluation results;

[0035] Step S2: Based on the evaluation results of the farmland agglomeration potential, a screening process is performed to obtain preliminary farmland agglomeration potential areas;

[0036] Step S3: Based on the preliminary potential area of ​​cultivated land agglomeration, analyze and process it under the coupling coordination degree model to obtain the coupling coordination degree, and based on the coupling coordination degree, perform screening and spatial filling under the preset coordination degree threshold to obtain the preferred space of key areas of cultivated land agglomeration.

[0037] Step S4: Based on the optimized space of the key areas of farmland agglomeration, perform local autocorrelation analysis to obtain the agglomeration type of each spatial unit in the optimized space of the key areas of farmland agglomeration;

[0038] Step S5: Obtain the minimum arable land requirement area corresponding to the target area, and perform spatial optimization processing under the search algorithm based on the cluster type and the minimum arable land requirement area to obtain the layout result of the arable land protection cluster area corresponding to the target area.

[0039] In the above embodiments, the preliminary potential areas for farmland agglomeration are analyzed and processed under the coupling coordination degree model to obtain the coupling coordination degree, which quantifies the synergistic constraints among multiple indicators. Subsequently, this embodiment filters based on the coordination degree threshold and fills in the void areas of concentrated plots, which can meet the needs of contiguous farmland protection and accurately identify functionally coordinated areas, thereby improving the accuracy of the layout results of farmland protection agglomeration areas. Based on the functional coordination of each area in the optimized space of the key farmland agglomeration area, this embodiment performs local autocorrelation analysis based on the optimized space of the key farmland agglomeration area, which can clarify the spatial agglomeration law of each spatial unit in the optimized space of the key farmland agglomeration area. Under the constraint of the minimum farmland demand area, further spatial optimization is carried out using a search algorithm, which can output the layout results of farmland protection agglomeration areas with high spatial agglomeration, thereby improving the accuracy of the layout results of farmland protection agglomeration areas.

[0040] Further, before acquiring evaluation data of the target area under the preset multi-indicator evaluation system for cultivated land agglomeration, and before conducting spatial agglomeration potential evaluation based on the evaluation data to obtain the cultivated land agglomeration potential evaluation result, the construction of the multi-indicator evaluation system for cultivated land agglomeration includes: selecting multiple indicators according to preset evaluation dimensions; the evaluation dimensions include resource endowment matching, service function matching, and regional coordinated development matching; obtaining the quantitative indicator scores corresponding to each quantitative indicator using a segmented method based on the quantitative indicators among the indicators; obtaining the qualitative indicator scores corresponding to each qualitative indicator using a derived Delphi method based on the qualitative indicators among the indicators; performing normalization processing on the quantitative indicator scores and the qualitative indicator scores to obtain the indicator scores corresponding to each indicator; obtaining the weights corresponding to each indicator using the CRITIC method based on the indicator scores; and constructing the multi-indicator evaluation system for cultivated land agglomeration based on the indicators, the weights, and the indicator scores.

[0041] In the above embodiments, resource endowment matching characterizes the degree of arable land agglomeration potential under the influence of natural resource background and human activity intervention; service function matching characterizes the production, living, and ecological benefits and value in the process of arable land agglomeration protection and utilization; and regional coordinated development matching characterizes the degree of coordination among production, living, and ecology in the process of arable land agglomeration protection and utilization. This embodiment selects indicators from three preset evaluation dimensions—resource endowment matching, service function matching, and regional coordinated development matching—ensuring the comprehensiveness of the selected indicators and overcoming the limitations of single-dimensional indicators.

[0042] Furthermore, this embodiment employs a segmented method and a derived Delphi method to determine the scores for quantitative and qualitative indicators, respectively, ensuring the reliability of the indicator scores. Normalization of all indicator scores eliminates the impact of differences in the dimensions of different indicators. The use of the CRITIC method to calculate the weights of each indicator fully leverages the information contained in the similarities and conflicts between the indicator data; that is, the greater the information content of an indicator, the greater its weight, and the greater its influence on the evaluation results. Therefore, this embodiment provides a reliable quantitative basis for the subsequent spatial agglomeration potential evaluation process, improves the reliability of the farmland agglomeration potential evaluation results, and thus improves the accuracy of the farmland protection agglomeration area layout results.

[0043] Further, the step of acquiring evaluation data of the target area under a preset multi-indicator evaluation system for cultivated land agglomeration, and evaluating spatial agglomeration potential based on the evaluation data to obtain cultivated land agglomeration potential evaluation results, includes: performing grade classification processing based on the indicators, weights, and indicator scores corresponding to each evaluation dimension in the multi-indicator evaluation system for cultivated land agglomeration, to obtain a cultivated land agglomeration potential evaluation model; acquiring evaluation data of the target area under the preset multi-indicator evaluation system for cultivated land agglomeration; the evaluation data includes evaluation data of each regional unit in the target area corresponding to each evaluation dimension in the multi-indicator evaluation system for cultivated land agglomeration; and obtaining cultivated land agglomeration potential evaluation results corresponding to each regional unit under the cultivated land agglomeration potential evaluation model based on the evaluation data.

[0044] In the above embodiments, the grading process is based on indicators, weights, and indicator scores to ensure the consistency of the evaluation process of the constructed farmland agglomeration potential evaluation model. Based on this farmland agglomeration potential evaluation model, this embodiment inputs the evaluation data of the target area under a preset multi-indicator evaluation system for farmland agglomeration into the farmland agglomeration potential evaluation model for processing. The resulting farmland agglomeration potential evaluation results have high reliability, and the evaluation results corresponding to different regional units have good comparability.

[0045] In one embodiment, the evaluation results of the farmland agglomeration potential corresponding to each regional unit can be divided into four categories: high value area, medium value area, low value area, and no obvious value area. The agglomeration potential level lookup table corresponding to the farmland agglomeration potential evaluation model of this embodiment is shown in Table 1 below. In Table 1, a, b, and c represent the levels of resource endowment matching, service function matching, and regional coordinated development matching, respectively.

[0046] Table 1 Agglomeration Potential Level Inquiry Table

[0047]

[0048] Further, the step of analyzing and processing the preliminary arable land agglomeration potential area under the coupling coordination degree model to obtain the coupling coordination degree, and then performing screening and spatial filling under a preset coordination degree threshold based on the coupling coordination degree to obtain the preferred space of key arable land agglomeration areas, includes: analyzing and processing the preliminary arable land agglomeration potential area under the coupling coordination degree model to obtain the coupling coordination degree corresponding to each potential area unit in the preliminary arable land agglomeration potential area; screening the preliminary arable land agglomeration potential area and the coupling coordination degree under a preset first coordination degree threshold to obtain the preferred space of initial key arable land agglomeration areas; and performing spatial filling under a preset second coordination degree threshold based on the preferred space of initial key arable land agglomeration areas to obtain the preferred space of key arable land agglomeration areas.

[0049] In the above embodiments, by performing coupling coordination degree analysis on each potential regional unit, high-quality units with high internal coordination can be identified. Furthermore, this embodiment, through a screening process using a preset first coordination degree threshold, can effectively eliminate discrete or functionally conflicting regions with low coupling coordination degree, ensuring that the initial preferred space consists of core regions with good coordination and high stability. Moreover, this embodiment introduces a preset second coordination degree threshold to fill the initial preferred space, effectively connecting adjacent high-quality units and filling internal gaps without significantly reducing the overall coordination level. This ultimately forms a preferred space for key farmland agglomeration areas with good contiguousness, high internal coordination, and a more complete and stable spatial structure, thereby improving the accuracy of the subsequent farmland protection agglomeration area layout results.

[0050] Furthermore, the step of performing local autocorrelation analysis based on the selected space of the key cultivated land agglomeration area to obtain the agglomeration type of each spatial unit in the selected space of the key cultivated land agglomeration area includes: performing local autocorrelation analysis based on the selected space of the key cultivated land agglomeration area to obtain the LISA cluster map corresponding to each evaluation dimension in the preset multi-index evaluation system for cultivated land agglomeration; and obtaining the cluster type of each spatial unit in the selected space of the key cultivated land agglomeration area in each evaluation dimension based on the LISA cluster map.

[0051] The above embodiments, through local autocorrelation analysis of the selected space in the key areas of farmland agglomeration, can accurately reveal the spatial correlation between each spatial unit and its neighboring spatial units on the preset evaluation dimensions. Furthermore, based on the LISA clustering map corresponding to each evaluation dimension, this embodiment obtains the clustering type of each spatial unit in each evaluation dimension, which can provide a reliable quantitative basis for the subsequent spatial optimization process to characterize the attributes of the spatial unit itself and its relationship with the surrounding spatial units, thereby improving the accuracy of the subsequent farmland protection agglomeration area layout results.

[0052] In one embodiment, the clustering type is divided into five cases: HH type, HL type, LH type, LL type and NS type. Among them, HH type means that high-quality cells are also surrounded by high-quality cells, and the degree of clustering and quality are both high; HL type means that high-quality cells are surrounded by low-quality cells, and the high-quality cells are scattered; LH type means that low-quality cells are surrounded by high-quality cells, and the low-quality cells are scattered; LL type means that low-quality cells are also surrounded by low-quality cells, and the degree of clustering is high but the quality is low; NS type means that there is insufficient evidence to support that the region has spatial correlation.

[0053] Further, the step of obtaining the minimum arable land requirement area corresponding to the target area, and performing spatial optimization processing under a search algorithm based on the clustering type and the minimum arable land requirement area to obtain the layout result of the arable land protection cluster area corresponding to the target area, includes: performing quantitative overlay processing based on the clustering type to obtain the comprehensive score corresponding to each spatial unit in the preferred space of the key arable land cluster area; based on the comprehensive score and the preferred space of the key arable land cluster area, obtaining a preliminary set of arable land protection cluster areas and a set of map patches under a preset comprehensive score threshold; and based on the set of map patches and the preliminary set of arable land protection cluster areas... The process involves: 1) determining the clustering distances of each patch in the patch set; where the clustering distance is the minimum distance between the patch and the initial farmland protection cluster; 2) sorting the patches according to the comprehensive score and the clustering distance to obtain a patch queue; 3) obtaining the minimum farmland requirement area corresponding to the target area, and 4) performing spatial iterative optimization steps based on the patch queue, the clustering distance, and the minimum farmland requirement area to obtain farmland protection clusters; and 5) obtaining the farmland protection cluster layout results corresponding to the target area based on the farmland protection clusters.

[0054] The above embodiments obtain a comprehensive score by quantifying and superimposing the clustering types corresponding to each evaluation dimension of each spatial unit, thus integrating multi-dimensional spatial correlation information into a unified quantitative score. Based on a preset comprehensive score threshold, this embodiment can accurately identify core areas with better comprehensive conditions within the preferred space of key cultivated land clustering areas as preliminary cultivated land protection clustering areas, while also clarifying the set of map patches to be optimized. Furthermore, this embodiment generates a map patch queue based on the clustering distance and comprehensive score of each map patch. The resulting map patch queue ensures that subsequent spatial iterative optimization steps prioritize processing map patches with better comprehensive conditions. Moreover, this embodiment performs spatial iterative optimization based on this map patch queue, which, while meeting the minimum cultivated land area requirement constraint, orderly merges adjacent high-quality map patches into the core area, thereby outputting a cultivated land protection clustering area layout result with a complete spatial structure, compact and reasonable layout, and meeting the area standard, thus improving the accuracy of the cultivated land protection clustering area layout result.

[0055] In one embodiment, HH type is assigned a value of 4, HL type is assigned a value of 3, LH type is assigned a value of 2, LL type is assigned a value of 1, and NS type is assigned a value of 0; and in this embodiment, areas with a comprehensive score of 4 or higher are identified as preliminary farmland protection clusters. In another embodiment, the map patches are arranged in descending order of their clustering distance to the cluster and in ascending order of their comprehensive score to obtain a map patch queue.

[0056] Further, the step of obtaining the minimum arable land requirement area corresponding to the target area and performing a spatial iterative optimization step based on the map patch queue, the clustering distance, and the minimum arable land requirement area to obtain the arable land protection cluster area includes: obtaining the state of the map patch queue and the cumulative area of ​​the preliminary arable land protection cluster area set; if the state is empty or the cumulative area reaches the minimum arable land requirement area, then the preliminary arable land protection cluster area set is taken as the arable land protection cluster area, and the spatial iterative optimization step is terminated; otherwise, the current map patch is taken from the head of the map patch queue; if the clustering distance corresponding to the current map patch is not less than a preset first distance threshold, then the current map patch is removed from the map patch queue, and the spatial iterative optimization step is re-executed; otherwise, it is determined that the current map patch is related to any of the preliminary arable land protection cluster areas. The system checks whether the distance between the initial farmland protection clusters meets the preset search distance. If the distance between the current patch and any of the initial farmland protection clusters meets the preset search distance, the current patch is merged into the initial farmland protection cluster, removed from the patch queue, and the spatial iterative optimization step is re-executed. Otherwise, it checks whether there are other patches within a radius of the current patch and the preset search distance. If there are other patches within a radius of the current patch and the preset search distance, the current patch is removed from the patch queue and inserted at the end of the patch queue, and the spatial iterative optimization step is re-executed. Otherwise, the current patch is removed from the patch queue, and the spatial iterative optimization step is re-executed.

[0057] The above embodiments, by continuously monitoring the status and cumulative area of ​​the map patch queue and comparing it with the minimum arable land requirement area, can ensure that the final arable land protection clusters meet the standards in terms of quantity or area. Furthermore, by comparing the current map patch's clustering distance with a preset first distance threshold, this embodiment can quickly filter and eliminate spatially dispersed map patches with low merging value, significantly improving the efficiency of spatial optimization. Moreover, by determining whether the distance between the current map patch and the cluster meets the preset search distance, this embodiment can identify and merge high-quality map patches that are spatially adjacent and meet the merging conditions, effectively promoting the reasonable expansion of the arable land protection cluster scale. Furthermore, by determining whether other map patches exist within the preset search distance range and adjusting the current map patch's position in the queue accordingly, this embodiment can avoid local optima when laying out in target areas with complex spatial scenarios, and can ensure that the final output is a highly concentrated, compact, and area-compliant arable land protection cluster.

[0058] Further, the step of obtaining the minimum arable land requirement area corresponding to the target area and performing a spatial iterative optimization step based on the map patch queue, the clustering distance, and the minimum arable land requirement area to generate the minimum arable land requirement area in the arable land protection cluster includes: obtaining historical population data and performing prediction processing under the logistic model based on the historical population data to obtain a population change curve; and performing scale calculation processing based on the population change curve to obtain the minimum arable land requirement area.

[0059] The above embodiments, by acquiring historical population data and using a logistic model for predictive processing, can accurately depict the dynamic trends and long-term patterns of population development, providing a reliable basis for subsequent scale calculations. Furthermore, this embodiment, based on population change curves for scale calculation, ensures that the minimum arable land requirement area is a quantitative result dynamically correlated with population changes in the target area. This guarantees that the final layout of arable land protection clusters in the target area matches future needs, thereby improving the accuracy of the arable land protection cluster layout results.

[0060] In one embodiment, the formula for calculating population changes over a certain period using the Logistic model is as follows:

[0061]

[0062] Where x(t) represents the population in year t, x0 represents the initial population at t=0, r represents the growth rate, and x m This represents population carrying capacity; and in this embodiment, based on the population change curve, the calculation formula for the minimum arable land area required to meet the food demand of the population in the planning target year is as follows:

[0063]

[0064] Where S represents the minimum required arable land area, in hectares (hm²). 2 β represents the food self-sufficiency rate (%); N represents the population (in people); M represents the per capita food demand (in kg / person); and P represents the grain yield per unit area (in kg / hm²). 2 E represents the percentage of total sown area for grain crops; K represents the multiple cropping index for grain crops, in percentages.

[0065] In one embodiment, the target area is Zhuhai City. In this embodiment, a total of 14 natural resource endowment indicators are selected from four aspects: site conditions, hydrological conditions, transportation location, and farming stability; and a total of 7 multiple service function indicators of cultivated land are selected from three aspects: production function, economic support, and ecological quality; and a total of 5 indicators of association and difference between regions are selected from two aspects: main utilization methods and coincidence degree of planning development. The selection of indicators in this embodiment covers various land use situations, fully considering the multiple service functions of cultivated land and the regional differences within the region.

[0066] It should be noted that in the above embodiment, the ArcGIS software is used, and the selected indicator data is uniformly projected using the CGCS2000 coordinate system.

[0067] In one embodiment, the normal distribution segmentation method is adopted for the indicator data approximately following the normal distribution among the quantitative indicators. Taking the ecological land coverage rate indicator as an example in this embodiment, the mean and standard deviation of this indicator are calculated. Subsequently, five grades are divided at equal intervals into the score interval [0 - 100], and the original score interval is calculated based on the divided grades and score interval. In this embodiment, the score intervals corresponding to the ecological land coverage rate in the ranges (+∞, 0.8], [0.6, 0.8), [0.4, 0.6), [0.2, 0.4), [0, 0.2) are (80 - 100], (60 - 80], (40 - 60], (20 - 40], [0 - 20] respectively.

[0068] In one embodiment, the quantile segmentation method is adopted for the data that does not approximately follow the normal distribution among the quantitative indicators. Taking the cultivated land kernel density indicator as an example in this embodiment, the indicator data is evenly divided into 5 intervals. Among them, the calculation formula for the position L of the i-th quantile Q i is: where n is the total amount of data, and i = 1, 2, 3, 4; if the position L is an integer, the quantile is the average of the L-th data and the (L + 1)-th data, that is, the quantile If the position L is not a decimal, its integer part k and decimal part f are obtained based on the position L, where k + f = l, and 0 < f < 0. In this embodiment, the quantile is calculated by linear interpolation as Q i = x k + f × (x k+1 - x k ). Based on the quantiles, intervals are divided and assigned values in this embodiment. Through calculation, it can be obtained that in this embodiment, the score intervals corresponding to the cultivated land kernel density in the ranges (+∞, 42], [36, 42), [30, 36), [24, 30), [0 - 24) are (80 - 100], (60 - 80], (40 - 60], (20 - 40], [0 - 20] respectively.

[0069] In one embodiment, the derived Delphi method is used to determine the scores of qualitative indicators. Taking land use patterns as an example, this embodiment involves multiple rounds of consultation with 10 professional technicians. In the first round, the indicators are assigned scores, statistically processed, and their mean and variance are calculated. The results are then fed back to the experts. The experts conduct a second round of scoring based on the overall trend and dispersion of their opinions reflected in the mean and variance of each indicator's assigned scores. After the second round of scoring, statistical processing is also performed to calculate the mean and variance, and the variances of the first and second rounds are X-rayed. 2 Significance test, this embodiment uses Statistical values ​​are used to compare the uniformity of variances corresponding to the two rounds of assignment, when the X of each factor... 2 The values ​​are all less than When verifying the values, expert consultation ends here. The formula used in this embodiment is as follows:

[0070]

[0071] in, The value is the average, where n is the number of experts consulted using the Delphi method, and n = 10; x i S represents the score of the i-th expert. 2 Let X be the variance. 2 This is a statistical test quantity constructed from two rounds of variance. In this embodiment, the calculation shows that the scoring intervals for paddy fields, irrigated land, dry land, non-arable land and construction land in the land use index are (80-100], (60-80], (40-60], (20-40], and [0-20], respectively.

[0072] In the above embodiments, since each indicator includes both positive and negative indicators, the data needs to be normalized. The specific formula is as follows:

[0073]

[0074] Where n represents n samples to be evaluated, m represents m evaluation indicators, and in this embodiment m = 26, max(x j ) represents the maximum value of the j-th indicator, min(x) j Let be the minimum value among the j-th indicators. Then, in this embodiment, the CRITIC method is used to calculate the weights, as shown in the following formula:

[0075]

[0076] In this embodiment, the standard deviation σ is used. j This indicates the comparative strength of the j-th indicator; The average value for each indicator; x represents indicator i, y represents indicator j, The mean of indicator i, The representative indicator is the mean of index j; r ij The correlation coefficient between evaluation indicators i and j; C j W represents the information content of the j-th evaluation indicator. j Let be the weight of the j-th indicator. In this embodiment, the selected indicators, indicator scores, and weights are shown in Table 2 below.

[0077] Table 2. Indicators, Indicator Scores, and Weights Corresponding to the Evaluation System of Main Controlling Factors of Cultivated Land Agglomeration

[0078]

[0079]

[0080] In one embodiment, to fully reflect the self-correlation of the multi-functional service value of arable land, the resource endowment matching, service function matching, and regional coordinated development matching in the evaluation system of the main controlling factors of arable land agglomeration are respectively set as the X-axis, Y-axis, and Z-axis. The nodes are assigned values ​​from 1 to 5 according to their distance from the model origin; the larger the value, the larger the index value of that dimension. Thus, this embodiment further forms a 5×5×5 three-dimensional fifth-order Rubik's Cube model, which, based on different combinations of units, is divided into four categories: high-value area, medium-value area, low-value area, and no obvious agglomeration potential, with corresponding distribution areas accounting for 16.68%, 26.56%, 28.75%, and 28.01% of the target area area, respectively.

[0081] In the above embodiments, the median value area and the high value area are selected as the initial potential areas for arable land agglomeration.

[0082] In one embodiment, the population change curve of Zhuhai from 2024 to 2035 is predicted using 2023 as the base year. The resulting population change curve is shown below. Figure 2 As shown; based on the population change curve, the minimum arable land requirement area for Zhuhai City is calculated to be 59.06 km². 2 This serves as the spatial constraint for delineating the city's farmland protection clusters.

[0083] It should be noted that in the above embodiments, the three indicator dimensions corresponding to the preferred space of the key areas of cultivated land agglomeration—resource endowment matching, service function matching, and regional coordinated development matching—all passed the significance test, with a P value of 0.001. Therefore, Moran's I value in this embodiment is significant at the 99.90% confidence level. Thus, the cultivated land agglomeration potential of the preferred space of the key areas of cultivated land agglomeration all show high spatial agglomeration in spatial distribution.

[0084] In one embodiment, local autocorrelation analysis is performed on the three dimensions of the selected key areas of cultivated land agglomeration to generate corresponding statistical tables of local spatial autocorrelation types and LISA cluster maps for each dimension. In this embodiment, the LISA cluster maps of each dimension of the selected key areas of cultivated land agglomeration correspond to the data in Table 3 below, and can intuitively reflect the spatial distribution of local spatial autocorrelation.

[0085] Table 3. Statistical Table of Local Spatial Autocorrelation Types of Key Spatial Indicators for Farmland Agglomeration in Key Areas for Selection

[0086]

[0087] In one embodiment, the scores of each region in three dimensions are superimposed, wherein: HH type is assigned 4, HL type is assigned 3, LH type is assigned 2, LL type is assigned 1, and NS type is assigned 0; regions with a comprehensive score of 4 or above are identified as preliminary farmland protection clusters, and the map patches are arranged in descending order of distance to the cluster and ascending order of score to obtain an icon queue. In this embodiment, the first distance threshold is set to 1000m. The first patch A is taken from the queue. If the distance between patch A and the nearest pre-identified farmland protection cluster is not less than the first distance threshold, it is discarded; otherwise, the process proceeds to the next step. The search distance is set to 10m. The distances from patch A to all pre-identified farmland protection clusters are calculated. When this distance is greater than the search distance, it is determined whether there are other patches within the search distance of patch A. If not, the next patch is taken from the queue for iteration. If so, the number of iterations in the icon queue is obtained, and it is determined whether it is greater than the maximum number of iterations. If it is greater, the iteration ends; otherwise, patch A is inserted at the end of the queue. When the distance is not greater than the search distance, it is determined whether there are other patches within the search distance of patch A. If not, patch A is merged into the nearest farmland protection cluster.

[0088] The above embodiment recursively traverses all map patches within the selected key areas of arable land clusters until the cumulative area of ​​the identified arable land protection clusters exceeds the minimum arable land requirement area defined for the arable land protection clusters, or the map patch queue is empty. At this point, the optimization of the arable land protection clusters is complete. Furthermore, this embodiment ultimately identified a total of 59.84 km² of arable land protection clusters in Zhuhai City. 2 Furthermore, the layout results of the farmland protection cluster areas are mainly distributed in the advantageous areas of Nanping Town in Xiangzhou District, the northwest of Jishan Mountain, Lianzhou Town and Baijiao Town in Doumen District, and Pingsha Town and Hongqi Town in Jinwan District.

[0089] Please see Figure 3The present invention also provides a farmland protection cluster layout system, comprising: a clustering potential evaluation result acquisition module, used to acquire evaluation data of a target area under a preset multi-index evaluation system for farmland clustering, and based on the evaluation data, perform spatial clustering potential evaluation to obtain farmland clustering potential evaluation results; a preliminary screening processing module, used to perform screening processing based on the farmland clustering potential evaluation results obtained by the clustering potential evaluation result acquisition module to obtain preliminary farmland clustering potential areas; and a preferred spatial acquisition module, used to perform analysis and processing based on the preliminary farmland clustering potential areas obtained by the preliminary screening processing module under a coupling coordination degree model to obtain coupling coordination degree, and based on... Based on the coupling coordination degree, filtering and spatial filling are performed under a preset coordination degree threshold to obtain the preferred space of key areas for farmland agglomeration; the local autocorrelation analysis module is used to perform local autocorrelation analysis on the preferred space of key areas for farmland agglomeration obtained by the preferred space acquisition module to obtain the agglomeration type of each spatial unit in the preferred space of key areas for farmland agglomeration; the farmland protection agglomeration area layout result acquisition module is used to obtain the minimum farmland demand area corresponding to the target area, and based on the agglomeration type obtained by the local autocorrelation analysis module and the minimum farmland demand area, spatial optimization processing is performed under the search algorithm to obtain the layout result of farmland protection agglomeration area corresponding to the target area.

[0090] Further, before acquiring evaluation data of the target area under the preset multi-indicator evaluation system for cultivated land agglomeration, and before conducting spatial agglomeration potential evaluation based on the evaluation data to obtain the cultivated land agglomeration potential evaluation result, the construction of the multi-indicator evaluation system for cultivated land agglomeration includes: selecting multiple indicators according to preset evaluation dimensions; the evaluation dimensions include resource endowment matching, service function matching, and regional coordinated development matching; obtaining the quantitative indicator scores corresponding to each quantitative indicator using a segmented method based on the quantitative indicators among the indicators; obtaining the qualitative indicator scores corresponding to each qualitative indicator using a derived Delphi method based on the qualitative indicators among the indicators; performing normalization processing on the quantitative indicator scores and the qualitative indicator scores to obtain the indicator scores corresponding to each indicator; obtaining the weights corresponding to each indicator using the CRITIC method based on the indicator scores; and constructing the multi-indicator evaluation system for cultivated land agglomeration based on the indicators, the weights, and the indicator scores.

[0091] The above embodiments analyze and process the preliminary potential areas for farmland agglomeration under the coupling coordination degree model, which can obtain the coupling coordination degree of the synergistic constraint relationship between multiple indicators. Subsequently, this embodiment screens based on the coordination degree threshold and fills in the void areas of concentrated plots, which can meet the needs of contiguous farmland protection and accurately identify functionally coordinated areas, thus improving the accuracy of the obtained farmland protection agglomeration area layout results. Furthermore, this embodiment performs local autocorrelation analysis based on the optimized space of key farmland agglomeration areas, which can clarify the spatial agglomeration pattern of each spatial unit in the optimized space of key farmland agglomeration areas. Under the constraint of minimum farmland demand area, this embodiment further uses a search algorithm for spatial optimization, which can output the layout results of farmland protection agglomeration areas with high spatial agglomeration, thus improving the accuracy of the farmland protection agglomeration area layout results.

[0092] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for layouting farmland protection clusters, characterized in that, Includes the following steps: The evaluation data of the target area under the preset multi-index evaluation system for farmland agglomeration is obtained, and the spatial agglomeration potential is evaluated based on the evaluation data to obtain the farmland agglomeration potential evaluation result. Based on the evaluation results of the farmland agglomeration potential, a preliminary farmland agglomeration potential area is obtained through screening. Based on the preliminary potential areas for farmland agglomeration, analysis and processing are performed under the coupling coordination degree model to obtain the coupling coordination degree. Based on the coupling coordination degree, screening and spatial filling are performed under a preset coordination degree threshold to obtain the preferred space for key areas of farmland agglomeration. Based on the optimized space of the key areas of farmland agglomeration, local autocorrelation analysis is performed to obtain the agglomeration type of each spatial unit in the optimized space of the key areas of farmland agglomeration; Obtain the minimum arable land requirement area corresponding to the target area, and perform spatial optimization processing under the search algorithm based on the cluster type and the minimum arable land requirement area to obtain the layout result of the arable land protection cluster area corresponding to the target area.

2. The method for layout of farmland protection clusters according to claim 1, characterized in that, Before acquiring evaluation data of the target area under a preset multi-indicator evaluation system for farmland agglomeration, and conducting spatial agglomeration potential evaluation based on the evaluation data to obtain the farmland agglomeration potential evaluation result, the construction of the multi-indicator evaluation system for farmland agglomeration includes: Based on preset evaluation dimensions, multiple indicators are selected; the evaluation dimensions include resource endowment matching, service function matching, and regional coordinated development matching. Based on the quantitative indicators in the above indicators, the quantitative indicator scores corresponding to each quantitative indicator are obtained under the segmentation method; based on the qualitative indicators in the above indicators, the qualitative indicator scores corresponding to each qualitative indicator are obtained under the derived Delphi method. The quantitative and qualitative index scores are normalized to obtain the corresponding index scores for each index. Based on the scores of the indicators, the weights corresponding to each indicator are obtained under the CRITIC method; Based on the aforementioned indicators, weights, and indicator scores, a multi-indicator evaluation system for farmland agglomeration is constructed.

3. The method for layout of farmland protection clusters according to claim 2, characterized in that, The process of acquiring evaluation data of the target area under a preset multi-indicator evaluation system for farmland agglomeration, and conducting spatial agglomeration potential evaluation based on the evaluation data to obtain farmland agglomeration potential evaluation results, includes: Based on the indicators, weights and scores of each evaluation dimension in the multi-indicator evaluation system for farmland agglomeration, a grade classification process is performed to obtain the evaluation model for farmland agglomeration potential. Acquire evaluation data of the target area under a preset multi-indicator evaluation system for farmland agglomeration; the evaluation data includes the evaluation data of each regional unit in the target area corresponding to each evaluation dimension in the multi-indicator evaluation system for farmland agglomeration; Based on the evaluation data, the evaluation results of the farmland agglomeration potential for each regional unit are obtained under the farmland agglomeration potential evaluation model.

4. The method for layout of farmland protection clusters according to claim 1, characterized in that, The preliminary potential areas for farmland agglomeration are analyzed and processed under a coupling coordination degree model to obtain a coupling coordination degree. Based on the coupling coordination degree, screening and spatial filling are performed under a preset coordination degree threshold to obtain the preferred space for key farmland agglomeration areas, including: Based on the preliminary potential area of ​​cultivated land agglomeration, the coupling coordination degree is obtained by analyzing and processing under the coupling coordination degree model to obtain the coupling coordination degree corresponding to each potential area unit in the preliminary potential area of ​​cultivated land agglomeration. Based on the preliminary potential area for farmland agglomeration and the coupling coordination degree, a screening is performed under a preset first coordination degree threshold to obtain the preferred space of the initial key area for farmland agglomeration. Based on the initial selected space of key areas for farmland agglomeration and the coupling coordination degree, spatial filling is performed under a preset second coordination degree threshold to obtain the selected space of key areas for farmland agglomeration.

5. The method for layout of farmland protection clusters according to claim 1, characterized in that, The method involves performing local autocorrelation analysis based on the optimized space of the key cultivated land agglomeration areas to obtain the agglomeration type of each spatial unit in the optimized space of the key cultivated land agglomeration areas, including: Based on the optimized space of the key areas of cultivated land agglomeration, local autocorrelation analysis is performed to obtain the LISA clustering map corresponding to each evaluation dimension in the preset multi-index evaluation system of cultivated land agglomeration. Based on the LISA cluster map, the clustering type of each spatial unit in the preferred space of the key cultivated land clustering area is obtained in each evaluation dimension.

6. The method for layout of farmland protection clusters according to claim 1, characterized in that, The process involves obtaining the minimum arable land requirement area corresponding to the target area, and based on the cluster type and the minimum arable land requirement area, performing spatial optimization processing under a search algorithm to obtain the layout result of the arable land protection cluster area corresponding to the target area, including: Based on the clustering type, a quantitative superposition process is performed to obtain the comprehensive score corresponding to each spatial unit in the preferred space of the key area of ​​farmland clustering. Based on the comprehensive score and the optimized space of key areas for farmland agglomeration, a preliminary set of farmland protection agglomeration areas and a set of map patches are obtained under a preset comprehensive score threshold. Based on the map patch set and the preliminary farmland protection cluster area, the cluster distance corresponding to each map patch in the map patch set is obtained; wherein, the cluster distance is the minimum distance between the map patch and the preliminary farmland protection cluster area. Based on the set of map features, the map features are sorted according to the comprehensive score and the clustering distance to obtain a map feature queue. Obtain the minimum arable land requirement area corresponding to the target area, and based on the map patch queue, the clustering distance and the minimum arable land requirement area, perform a spatial iterative optimization step to obtain the arable land protection cluster area; Based on the farmland protection clusters, the layout results of the farmland protection clusters corresponding to the target area are obtained.

7. The method for layout of farmland protection clusters according to claim 6, characterized in that, The step involves obtaining the minimum arable land requirement area corresponding to the target area, and based on the map patch queue, the clustering distance, and the minimum arable land requirement area, performing a spatial iterative optimization step to obtain the arable land protection cluster area. The spatial iterative optimization step includes: Obtain the status of the map patch queue and the cumulative area of ​​the preliminary farmland protection cluster; If the state is empty or the accumulated area reaches the minimum arable land requirement area, then the preliminary arable land protection cluster is set as the arable land protection cluster, and the spatial iteration optimization step is terminated; otherwise, the current patch is taken from the head of the patch queue. If the clustering distance corresponding to the current patch is not less than the preset first distance threshold, then the current patch is removed from the patch queue and the spatial iteration optimization step is re-executed; otherwise, it is determined whether the distance between the current patch and any preliminary farmland protection cluster in the preliminary farmland protection cluster meets the preset search distance. If the distance between the current patch and any of the preliminary farmland protection clusters in the preliminary farmland protection clusters meets the preset search distance, then the current patch is merged into the preliminary farmland protection clusters, the current patch is removed from the patch queue, and the spatial iterative optimization step is re-executed; otherwise, it is determined whether there are other patches within a range centered on the current patch and with the preset search distance as the radius. If there are other patches within a radius of the current patch and the preset search distance, the current patch is removed from the patch queue and inserted at the end of the patch queue, and the spatial iteration optimization step is re-executed; otherwise, the current patch is removed from the patch queue and the spatial iteration optimization step is re-executed.

8. The method for layout of farmland protection clusters according to claim 1, characterized in that, The process of obtaining the minimum arable land requirement area corresponding to the target area, and performing a spatial iterative optimization step based on the map patch queue, the clustering distance, and the minimum arable land requirement area to generate the minimum arable land requirement area in the arable land protection cluster area includes: Obtain historical population data and, based on this data, perform prediction processing under the logistic model to obtain population change curves; Based on the population change curve, a scale calculation is performed to obtain the minimum arable land requirement area.

9. A layout system for farmland protection clusters, characterized in that, A method for implementing the layout of farmland protection clusters as described in any one of claims 1 to 8 includes: The agglomeration potential evaluation result acquisition module is used to acquire evaluation data of the target area under the preset multi-index evaluation system for farmland agglomeration, and to conduct spatial agglomeration potential evaluation based on the evaluation data to obtain farmland agglomeration potential evaluation results. The preliminary screening and processing module is used to perform screening processing based on the farmland agglomeration potential evaluation results obtained by the agglomeration potential evaluation result acquisition module to obtain preliminary farmland agglomeration potential areas. The preferred space acquisition module is used to analyze and process the preliminary farmland agglomeration potential area obtained by the preliminary screening and processing module under the coupling coordination degree model to obtain the coupling coordination degree, and based on the coupling coordination degree, to perform screening and spatial filling under a preset coordination degree threshold to obtain the preferred space of key farmland agglomeration areas. The local autocorrelation analysis module is used to perform local autocorrelation analysis based on the preferred space of the key areas of farmland agglomeration obtained by the preferred space acquisition module, so as to obtain the agglomeration type of each spatial unit in the preferred space of the key areas of farmland agglomeration; The farmland protection cluster layout result acquisition module is used to obtain the minimum farmland demand area corresponding to the target area, and based on the cluster type obtained by the local autocorrelation analysis module and the minimum farmland demand area, perform spatial optimization processing under the search algorithm to obtain the farmland protection cluster layout result corresponding to the target area.

10. A farmland protection cluster layout system according to claim 9, characterized in that, Before acquiring evaluation data of the target area under a preset multi-indicator evaluation system for farmland agglomeration, and conducting spatial agglomeration potential evaluation based on the evaluation data to obtain the farmland agglomeration potential evaluation result, the construction of the multi-indicator evaluation system for farmland agglomeration includes: Based on preset evaluation dimensions, multiple indicators are selected; the evaluation dimensions include resource endowment matching, service function matching, and regional coordinated development matching. Based on the quantitative indicators in the above indicators, the quantitative indicator scores corresponding to each quantitative indicator are obtained under the segmentation method; based on the qualitative indicators in the above indicators, the qualitative indicator scores corresponding to each qualitative indicator are obtained under the derived Delphi method. The quantitative and qualitative index scores are normalized to obtain the corresponding index scores for each index. Based on the scores of the indicators, the weights corresponding to each indicator are obtained under the CRITIC method; Based on the aforementioned indicators, weights, and indicator scores, a multi-indicator evaluation system for farmland agglomeration is constructed.

Images