High-standard farmland construction time sequence optimization method

Abstract

The invention discloses a high-standard farmland construction time sequence optimization method, and relates to the technical field of agricultural information, and the method comprises the steps: superposing a permanent basic farmland distribution diagram layer and a high-standard farmland distribution diagram layer, and recognizing a permanent basic farmland region which is not constructed into a high standard; obtaining multi-source basic data in the region, and representing the grain yield by adopting an LAI average value after S-G filtering; constructing a high-standard farmland construction ecological toughness evaluation index system; the combined weight of each evaluation index is calculated according to the combination of an entropy weight method and a CRITIC method, and a set pair analysis model is established to calculate the farmland ecological toughness; a spatial interaction relationship between grain yield and farmland ecological toughness is obtained based on a bivariate local spatial autocorrelation analysis method, a preferential plot of high-standard farmland construction is identified in combination with a self-organizing mapping neural network and a K-means method, and a construction time sequence arrangement and spatial layout optimization scheme is formed. According to the invention, the problem of high-standard farmland construction priority can be solved.

Description

Technical Field

[0001] The present invention relates to the field of agricultural information technology, and in particular to a time sequence optimization method for high-standard farmland construction. Background Art

[0002] Existing technologies face two bottlenecks: Spatially, traditional site selection methods rely on single indicators such as land flatness and soil fertility, failing to systematically integrate multi-source heterogeneous data (such as remote sensing, topography, and meteorology) to construct a comprehensive evaluation system encompassing both grain yield and farmland ecological resilience. Furthermore, they neglect the correlation between measured values ​​of existing farmland indicators and optimal conditions for high-standard farmland construction, resulting in inaccurate identification of priority construction areas. Temporally, an optimization model integrating productivity improvement and ecological resilience has not been established, making it difficult to coordinate construction batches using clustering algorithms, hindering the construction and management of high-standard farmland.

[0003] This is further reflected in the fact that, on the one hand, spatial evaluation lacks a systematic framework, making it difficult to integrate the synergistic relationship between food production capacity representation and ecological resilience, and ignoring the interactive effect of bivariate spatial autocorrelation, resulting in insufficient basis for decision-making on land priority; on the other hand, the time sequence arrangement relies on a fixed decomposition model, and fails to introduce an objective empowerment model and clustering optimization mechanism, making it difficult to promote the transformation and upgrading of permanent basic farmland in batches and stages and build high-standard farmland. Summary of the Invention

[0004] In view of this, the present invention provides a method for optimizing the timing of high-standard farmland construction to solve related problems in the prior art.

[0005] The present invention provides a method for optimizing the timing of high-standard farmland construction, comprising: S1. Perform spatial overlay analysis on the permanent basic farmland distribution layer and the high-standard farmland distribution layer to identify areas where high-standard permanent basic farmland has not been built; S2. Acquire multi-source basic data for areas where high-standard permanent basic farmland has not been built; wherein the multi-source basic data includes: LAI data, soil fertility improvement data, drought and flood disaster prevention data, and land leveling data; and reconstruct the LAI data using SG filtering to calculate the average LAI value during the crop growth period in the areas where high-standard permanent basic farmland has not been built; the average LAI value is used to represent grain yield; S3. Based on multi-source basic data, construct an evaluation index system for the ecological resilience of high-standard farmland construction; calculate the weights of each evaluation index using a combination of the entropy weight method and the CRITIC method, and establish a set pair analysis model to calculate the ecological resilience of farmland; S4. Obtaining the spatial interaction relationship between the grain yield and the farmland ecological resilience based on a bivariate local spatial autocorrelation analysis method; S5. Based on the spatial interaction relationship, combined with the self-organizing map neural network and K-means method, identify the priority plots for high-standard farmland construction in areas where high-standard permanent basic farmland has not yet been built, and form a time-optimization plan for high-standard farmland construction.

[0006] In an optional embodiment, the S1 specifically includes: The permanent basic farmland distribution layer and the high-standard farmland distribution layer are spatially superimposed, and the high-standard farmland distribution layer is removed from the permanent basic farmland distribution layer through GIS difference operation to output the permanent basic farmland distribution layer that has not been built to high standards.

[0007] In an optional embodiment, the soil fertility improvement data includes: soil organic matter, soil total nitrogen content, soil total phosphorus content, soil texture, soil pH value, soil bulk density; The drought and flood disaster resistance data include: key climate factors, SPEI index, and distance to rivers and reservoirs; The land leveling data includes: sub-dimensional index, elevation, slope, terrain undulation, effective soil layer thickness, and road accessibility.

[0008] In an optional embodiment, the SPEI index is calculated using temperature data and precipitation data obtained from the China Meteorological Service Data Sharing Service Network; the key climate factors are obtained by performing correlation analysis on the LAI data with 19 bioclimatic variables, sorting them from high to low according to the correlation, and selecting the top 5 bioclimatic variables.

[0009] In an optional implementation, the S3 includes: S31. Construct an ecological resilience evaluation index system for high-standard farmland construction based on soil fertility improvement data, drought and flood disaster prevention data, and land leveling data; wherein the ecological resilience evaluation index system for high-standard farmland construction includes the name of each evaluation indicator, the corresponding indicator value, the source type, and the indicator type to which it belongs; the indicator type is used to characterize the correlation between the indicator value corresponding to the evaluation indicator and the ecological resilience of the farmland; S32. Based on the correlation between the index value corresponding to each evaluation index and the farmland ecological resilience, the index value corresponding to each evaluation index is normalized to obtain a normalized value corresponding to each evaluation index; S33. Based on S32, the combined weights of the evaluation indicators are calculated by combining the entropy weight method and the CRITIC method; S34. Based on S33, a set pair analysis model is established; and according to the set pair analysis model, the farmland ecological resilience of each unit to be evaluated is calculated.

[0010] In an optional embodiment, the S34 includes: The evaluation problem of high-standard farmland ecological resilience is denoted as ;in, is the evaluation scheme set, is the total number of evaluation options; is the evaluation index set, is the total number of evaluation indicators; is the unit set to be evaluated, For the Units to be evaluated, is the total number of units to be evaluated; the optimal evaluation indicators of each evaluation scheme are compared and determined within the same unit to be evaluated to form the optimal scheme set , the worst evaluation indicators in each evaluation scheme constitute the worst scheme set ; ;

[0011] in, 、 are the identity and opposition of the optimal solution set and the worst solution set respectively; 、 Respectively p Evaluation indicators With the collection [ ]’s identity and opposition, For the The weight of each evaluation indicator; Evaluation plan and the optimal solution set Relative closeness for: ;

[0012] in, To evaluate the program evaluation schemes, .

[0013] In an optional embodiment, the standardization process of the evaluation index is as follows: When the correlation between the index value corresponding to the evaluation index and the farmland ecological resilience is positive, the calculation formula of the standardized value is: ;

[0014] When the correlation between the index value corresponding to the evaluation index and the farmland ecological resilience is negative, the calculation formula of the standardized value is: ;

[0015] When the index value corresponding to the evaluation index is close to the set threshold and meets the set limit, the calculation formula of the standardized value is: ;

[0016] in, For the The indicator value corresponding to each evaluation indicator; For multiple units to be evaluated The maximum value of For multiple units to be evaluated The minimum value of To set the threshold; Comparison operation for maximum and minimum values.

[0017] In an optional embodiment, the S4 includes: GeoDA software was used to conduct a bivariate local spatial autocorrelation analysis on the target unit to be evaluated. The spatial interaction relationship between the farmland ecological resilience of the target unit to be evaluated and the grain yield of other surrounding units to be evaluated was obtained. The spatial interaction relationship included four types: high-high area, high-low area, low-high area, and low-low area. The calculation formula of the spatial interaction relationship is: ;

[0018] in, I is the bivariate global autocorrelation coefficient, t is the total number of grids, is the spatial weight matrix; and Variables: farmland ecological resilience In the grid Value and food production In the grid The value of is the variance of farmland ecological resilience and grain yield.

[0019] In an optional implementation, the S5 includes: S51. Set high-high areas as high-priority construction areas, high-low and low-high areas as secondary-priority construction areas, and low-low areas as non-priority construction areas; S52. Using a self-organizing map neural network and K-means, the standardized values ​​corresponding to each evaluation indicator are clustered into areas dominated by soil fertility improvement, areas dominated by drought and flood disaster prevention, and areas dominated by land leveling; S53. Based on S52 and S53, a timing optimization plan for high-standard farmland construction is obtained.

[0020] The present invention has the following beneficial effects: This invention addresses the issue of prioritizing high-standard farmland construction. Spatially, it establishes a comprehensive evaluation model driven by the fusion of multi-source heterogeneous data. By coupling remote sensing imagery, topography, meteorological factors, and soil moisture data, it establishes a dual-objective collaborative framework for food production capacity and ecological resilience. Bivariate spatial autocorrelation analysis is used to analyze the spatial interaction effects of evaluation indicators, enabling precise identification of priority construction areas. Temporally, an entropy weighting method and a multi-correlation coefficient integrated weighting model are employed to objectively quantify the weights of indicators such as production capacity improvement potential and ecological resilience, generating a phased order for high-standard farmland construction and effectively advancing the transformation of permanent basic farmland in stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 1 is a flow chart of a method for optimizing the timing of high-standard farmland construction according to an embodiment of the present invention; Figure 2 is a correlation diagram between bioclimatic variables and LAI according to an embodiment of the present invention; Figure 3 is a bivariate local spatial autocorrelation graph according to an embodiment of the present invention; Figure 4 This is a priority sequence diagram for high-standard farmland construction according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0024] like Figure 1 As shown, the present invention provides a method for optimizing the timing of high-standard farmland construction, comprising: Step S1: Perform spatial overlay analysis on the permanent basic farmland distribution layer and the high-standard farmland distribution layer to identify areas where high-standard permanent basic farmland has not been built.

[0025] In an optional embodiment, step S1 specifically includes: The permanent basic farmland distribution layer and the high-standard farmland distribution layer are spatially superimposed, and the high-standard farmland distribution layer is removed from the permanent basic farmland distribution layer through GIS difference operation to output the permanent basic farmland distribution layer that has not been built to high standards.

[0026] Preferably, step S1 can perform spatial overlay analysis on a geographic information system (GIS) platform or a remote sensing data processing platform.

[0027] Taking the ArcGIS platform as an example, first import the permanent basic farmland distribution layer as the first data layer (A) and the high-standard farmland distribution layer as the second data layer (B) into the ArcGIS platform; wherein the first data layer (A) contains the boundary range and attribute information of the permanent basic farmland, and the second data layer (B) contains the boundary range and attribute information of the high-standard farmland; then, call the ArcGIS "Erase Tool" with the first data layer (A) as the input feature and the second data layer (B) as the erase feature, perform a spatial difference operation, and generate a difference result layer (C); finally, define the difference result layer (C) as the "permanent basic farmland distribution layer that has not been built to high standards", which contains the boundary range and attribute information of the permanent basic farmland that has not met the high-standard farmland construction standards, that is, the permanent basic farmland area that has not been built to high standards is obtained.

[0028] Step S2: Acquire multi-source basic data for areas where high-standard permanent basic farmland has not been built; wherein the multi-source basic data include: LAI data (leaf area index), soil fertility improvement data, drought and flood disaster prevention data, and land leveling data; and reconstruct the LAI data using SG filtering to calculate the average LAI value during the crop growth period in the areas where high-standard permanent basic farmland has not been built; the average LAI value is used to represent grain yield.

[0029] Preferably, when removing the high-standard farmland distribution layer from the permanent basic farmland distribution layer, it is also necessary to remove sloping farmland with a slope of more than 25 degrees, strictly controlled farmland, ecological protection red line, and returning farmland to forest, grassland, lake, and pasture layers at the same time. These areas strictly restrict the construction of high-standard farmland.

[0030] Among them, sloping farmland with a slope of more than 25 degrees can obtain ASTER GDEM 30M resolution digital elevation data from the geospatial data cloud. Select 3D analyst tools→Raster surface→Slope in the GIS platform to generate slope data, set the data above 25° to 0, and use the raster calculator to multiply it with the permanent basic farmland distribution layer that has not been built to a high standard to eliminate sloping farmland with a slope of more than 25 degrees.

[0031] In an optional embodiment, the soil fertility improvement data include: soil organic matter, soil total nitrogen content, soil total phosphorus content, soil texture, soil pH value, soil bulk density; the drought and flood disaster resistance data include: key climate factors, SPEI index (i.e., standardized precipitation evapotranspiration index), distance to rivers and reservoirs; the land leveling data include: sub-dimensional index, elevation, slope, terrain undulation, effective soil layer thickness, and road accessibility.

[0032] Preferably, LAI data can be collected in Google Earth Engine using the MCD15A3H dataset of the main crop growing season in areas where high-standard permanent basic farmland has not been built, i.e., LAI data, with a spatial resolution of 500 m and a temporal resolution of 8 days.

[0033] In an optional embodiment, the SPEI index is calculated using temperature and precipitation data obtained from the China Meteorological Service Data Sharing Service Network; the key climate factors are obtained by performing correlation analysis between the LAI data and 19 bioclimatic variables, sorting them from high to low according to the correlation, and selecting the top 5 bioclimatic variables.

[0034] Specifically, the calculation process of the SPEI index is as follows: First, monthly temperature and precipitation data from meteorological stations were downloaded and interpolated using ANUSPLIN software to obtain monthly potential evapotranspiration (PET). Second, GDAL was installed in Anaconda, and the corresponding function in Climate_Indices was called to calculate the SPEI to obtain the grid-scale SPEI index.

[0035] It should be noted that when optimizing the timing of standard farmland construction in areas where high-standard permanent basic farmland has not been built, it is necessary to divide the areas where high-standard permanent basic farmland has not been built into multiple grids (plots), with each grid as a unit to be evaluated (for subsequent analysis and processing), and multi-source basic data acquisition is carried out in grid units.

[0036] There are 19 bioclimatic variables (including extreme or limiting environmental factors) in total (Bio_1 to Bio_19), such as Figure 2As shown, they are annual mean temperature (Bio_1), mean diurnal temperature range (Bio_2), isothermality (Bio_3), temperature seasonality (Bio_4), maximum temperature of the hottest month (Bio_5), minimum temperature of the coldest month (Bio_6), annual temperature range (Bio_7), mean temperature of the wettest quarter (Bio_8), mean temperature of the driest quarter (Bio_9), mean temperature of the hottest quarter (Bio_10), mean temperature of the coldest quarter (Bio_11), annual precipitation (Bio_12), precipitation of the wettest month (Bio_13), precipitation of the driest month (Bio_14), precipitation seasonality (Bio_15), precipitation of the wettest quarter (Bio_16), precipitation of the driest quarter (Bio_17), precipitation of the hottest quarter (Bio_18), and precipitation of the coldest quarter (Bio_19). Then, these 19 bioclimatic variables were subjected to Spearman correlation analysis with the LAI data and ranked from high to low according to the correlation. Finally, the top five bioclimatic variables were selected as key climate factors (with the highest correlation).

[0037] Preferably, Bio_5, Bio_8, Bio_10, Bio_16 and Bio_18 are selected as key climate factors.

[0038] In addition, the distance to the river reservoir is based on the “Euclidean distance” tool in ArcGIS to obtain the buffer raster data of the river reservoir, and the “zonal statistics” tool is used to perform distance statistics to obtain the minimum distance from each unit to be evaluated to the river reservoir.

[0039] The fractal dimension index is calculated using Fragstats software. The larger the fractal dimension, the more complex the shape of the unit (plot) to be evaluated. The closer the value is to 1, the stronger the self-similarity of the plot, the more regular the shape, and the closer it is to simplicity, indicating that the degree of interference is greater.

[0040] (1);

[0041] in, For the The fractal dimension of a plot, For the The perimeter of the plot, For the The area of ​​a plot of land.

[0042] Road accessibility was determined using the Near tool in ArcGIS → Analysis Tools → Proximity to obtain the minimum distance from each unit to be evaluated to the road.

[0043] Step S3: Based on multi-source basic data, construct an ecological resilience evaluation index system for high-standard farmland construction; and calculate the weight of each evaluation index based on the combination of the entropy weight method and the CRITIC method, and establish a set pair analysis model to calculate the ecological resilience of farmland.

[0044] In an optional embodiment, step S3 includes: S31. Build an ecological resilience evaluation index system for high-standard farmland construction based on soil fertility improvement data, drought and flood disaster prevention data, and land leveling data; As shown in Table 1, the ecological resilience evaluation index system for high-standard farmland construction includes the name of each evaluation indicator, the corresponding indicator value, the source type, and the indicator type to which it belongs; the indicator type is used to characterize the correlation between the indicator value corresponding to the evaluation indicator and the ecological resilience of farmland.

[0045] Table 1 ;

[0046] Specifically, indicator types can include positive, negative, and neutral indicators. Positive indicators indicate a positive correlation between the target evaluation indicator's value and farmland ecological resilience; negative indicators indicate a negative correlation between the target evaluation indicator's value and farmland ecological resilience; and neutral indicators indicate how close the target evaluation indicator's value is to a set threshold. The closer the indicator value is to the threshold, the closer the evaluation target is to the ideal.

[0047] S32. Based on the correlation between the indicator value corresponding to each evaluation indicator and the farmland ecological resilience, the indicator value corresponding to each evaluation indicator is standardized to obtain a standardized value corresponding to each evaluation indicator.

[0048] In an optional implementation, the standardization process of the evaluation index is as follows: When the correlation between the index value corresponding to the evaluation index and the farmland ecological resilience is positive, the calculation formula of the standardized value is: (2); When the correlation between the index value corresponding to the evaluation index and the farmland ecological resilience is negative, the calculation formula of the standardized value is: (3); When the index value corresponding to the evaluation index is close to the set threshold and meets the set limit, the calculation formula of the standardized value is: (4); in, For the The indicator value corresponding to each evaluation indicator; For multiple units to be evaluated The maximum value of For multiple units to be evaluated The minimum value of To set the threshold; Comparison operation for maximum and minimum values.

[0049] S33: Based on step S32, the combined weights of the evaluation indicators are calculated by combining the entropy weight method and the CRITIC method. ;in, W is the combined weight set of each evaluation index, For the The combined weight of the evaluation indicators, .

[0050] Preferably, the entropy weight method and the CRITIC method are combined to determine the weight of each evaluation indicator, and the combined weight is calculated using the multiplication synthesis form, as follows: (5); in, For the The combined weight of the evaluation indicators; is the first The weight of the evaluation index; The first The weight value of the evaluation index.

[0051] S34. Based on S33, a set pair analysis model is established; and according to the set pair analysis model, the farmland ecological resilience of each unit to be evaluated is calculated, that is, the relative progress of each unit to be evaluated with the optimal solution set.

[0052] In an optional implementation, S34 includes: The evaluation problem of high-standard farmland ecological resilience is denoted as ;in, is the evaluation scheme set, is the total number of evaluation options; is the evaluation index set, is the total number of evaluation indicators; is the unit set to be evaluated, For the Units to be evaluated, is the total number of units to be evaluated; the optimal evaluation indicators of each evaluation scheme are compared and determined within the same unit to be evaluated to form the optimal scheme set , the worst evaluation indicators in each evaluation scheme constitute the worst scheme set ; Set in The connection degree on is: ;

[0053] in, For the evaluation schemes; 、 、 are the degree of identity, difference and opposition of the optimal solution set and the worst solution set respectively; 、 Respectively p Evaluation indicators With the collection [ ]’s identity and opposition, For the The weight of each evaluation indicator; Evaluation plan and the optimal solution set Relative closeness , that is, farmland ecological resilience is: ; reflect and the optimal solution set The degree of connection, The larger the value, the closer the unit to be evaluated is to the optimal solution, that is, The higher the value, the higher the ecological resilience of high-standard farmland.

[0054] Step S4: Based on the bivariate local spatial autocorrelation analysis method, the spatial interaction relationship between grain yield and farmland ecological resilience is obtained.

[0055] In an optional embodiment, step S4 includes: GeoDA software was used to conduct a bivariate local spatial autocorrelation analysis on the target unit to be evaluated. The spatial interaction relationship between the farmland ecological resilience of the target unit to be evaluated and the grain yield of other surrounding units to be evaluated was obtained. The spatial interaction relationship included four types: high-high area, high-low area, low-high area, and low-low area. The calculation formula of the spatial interaction relationship is: ;

[0056] in, I is the bivariate global autocorrelation coefficient, t is the total number of grids, is the spatial weight matrix; and Variables: farmland ecological resilience In the grid Value and food production In the grid The value of is the variance of farmland ecological resilience and grain yield. Among them, the obtained bivariate local spatial autocorrelation diagram is as follows Figure 3 shown.

[0057] Step S5: Based on spatial interaction relationships, combined with the self-organizing map neural network and K-means method, identify the priority plots for high-standard farmland construction in areas where high-standard permanent basic farmland has not yet been built, and form a time sequence optimization plan for high-standard farmland construction.

[0058] In an optional implementation, S5 includes: S51. Set the high-high area as a high-priority construction area, the high-low and low-high areas as secondary-priority construction areas, and the low-low area as a non-priority construction area. Figure 4 As shown; S52. Use self-organizing map neural network and K-means to cluster the standardized values ​​corresponding to each evaluation indicator and divide them into the dominant area for soil fertility improvement, the dominant area for resisting drought and flood disasters, and the dominant area for land leveling.

[0059] Specifically, self-organizing map neural network and K-means were used, self-organizing map analysis was performed through the Kohonen package of R, and K-means analysis was performed through SPSS. The standardized values ​​corresponding to each evaluation index of the three categories of soil fertility improvement, drought and flood disaster resistance, and land leveling were clustered, and the areas were divided into soil fertility improvement-dominant areas, drought and flood disaster resistance-dominant areas, and land leveling-dominant areas.

[0060] S53. Based on S52 and S53, a timing optimization plan for high-standard farmland construction is obtained.

[0061] For example, based on the priority classification in step S51 and the cluster analysis results in step S52, a timeline plan with three phases, three leading categories, and five implementation steps is constructed. In the near term (1-3 years), the focus will be on high-priority construction areas, with categorized advancements based on soil fertility improvement (60%), drought and flood mitigation (30%), and land leveling (10%). Core projects will include deep tillage, canal anti-seepage measures, and field consolidation, along with supporting technical pre-research on organic fertilizer substitution and smart irrigation system pilots. In the medium term (4-6 years), trials of returning straw to farmland and pre-buried ecological ditches will be conducted in secondary priority construction areas to build basic capacity. In the long term (7-10 years), ecological restoration and the deployment of smart farmland base stations will be promoted in non-priority construction areas. The plan innovatively constructs an "engineering package" system and a digital twin platform, and realizes dynamic adjustments through annual NDVI remote sensing monitoring and the ESG evaluation system. It is expected that within 10 years, the quality level of cultivated land will be improved by 3.8 levels, the irrigation guarantee rate will reach 100%, and the carbon emission intensity will be reduced by 45%, forming a three-dimensional space-time-type management and control plan.

[0062] In summary, this invention can address the issue of prioritizing high-standard farmland construction. On a spatial scale, it establishes a comprehensive evaluation model driven by the fusion of multi-source heterogeneous data. By coupling remote sensing imagery, topography, meteorological factors, and soil moisture data, it establishes a dual-objective collaborative framework for food production capacity and ecological resilience. Bivariate spatial autocorrelation analysis is used to analyze the spatial interaction effects of evaluation indicators, enabling accurate identification of priority construction areas. On a temporal scale, an entropy weighting method and a multi-correlation coefficient integrated weighting model are employed to objectively quantify the weights of indicators such as production capacity improvement potential and ecological resilience, generating a phased order for the construction of high-standard farmland and effectively advancing the transformation of permanent basic farmland in stages.

[0063] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for optimizing the timing of high-standard farmland construction, characterized in that: include: S1. Perform spatial overlay analysis on the permanent basic farmland distribution layer and the high-standard farmland distribution layer to identify areas where high-standard permanent basic farmland has not been built; S2. Acquire multi-source basic data for areas where high-standard permanent basic farmland has not been built; wherein the multi-source basic data includes: LAI data, soil fertility improvement data, drought and flood disaster prevention data, and land leveling data; and reconstruct the LAI data using SG filtering to calculate the average LAI value during the crop growth period in the areas where high-standard permanent basic farmland has not been built; the average LAI value is used to represent grain yield; S3. Based on multi-source basic data, construct an evaluation index system for the ecological resilience of high-standard farmland construction; calculate the weights of each evaluation index using a combination of the entropy weight method and the CRITIC method, and establish a set pair analysis model to calculate the ecological resilience of farmland; S4. Obtaining the spatial interaction relationship between the grain yield and the farmland ecological resilience based on a bivariate local spatial autocorrelation analysis method; S5. Based on the spatial interaction relationship, combined with the self-organizing map neural network and K-means method, identify the priority plots for high-standard farmland construction in areas where high-standard permanent basic farmland has not yet been built, and form a time-optimization plan for high-standard farmland construction.

2. The method according to claim 1, characterized in that Said S1 specifically includes: The permanent basic farmland distribution layer and the high-standard farmland distribution layer are spatially superimposed, and the high-standard farmland distribution layer is removed from the permanent basic farmland distribution layer through GIS difference operation to output the permanent basic farmland distribution layer that has not been built to high standards.

3. The method according to claim 1, characterized in that The soil fertility improvement data include: soil organic matter, soil total nitrogen content, soil total phosphorus content, soil texture, soil pH value, and soil bulk density; The drought and flood disaster resistance data include: key climate factors, SPEI index, and distance to rivers and reservoirs; The land leveling data includes: sub-dimensional index, elevation, slope, terrain undulation, effective soil layer thickness, and road accessibility.

4. The method according to claim 3, characterized in that The SPEI index is calculated using temperature and precipitation data obtained from the China Meteorological Service Data Sharing Service Network. The key climate factors are obtained by performing correlation analysis between LAI data and 19 bioclimatic variables, sorting them from high to low according to the correlation, and selecting the top five bioclimatic variables.

5. The method according to claim 4, characterized in that The S3 includes: S31. Construct an ecological resilience evaluation index system for high-standard farmland construction based on soil fertility improvement data, drought and flood disaster prevention data, and land leveling data; wherein the ecological resilience evaluation index system for high-standard farmland construction includes the name of each evaluation indicator, the corresponding indicator value, the source type, and the indicator type to which it belongs; the indicator type is used to represent the correlation between the indicator value corresponding to the evaluation indicator and the ecological resilience of the farmland; S32. Based on the correlation between the index value corresponding to each evaluation index and the farmland ecological resilience, the index value corresponding to each evaluation index is normalized to obtain a normalized value corresponding to each evaluation index; S33. Based on S32, the combined weights of the evaluation indicators are calculated by combining the entropy weight method and the CRITIC method; S34. Based on S33, a set pair analysis model is established; and according to the set pair analysis model, the farmland ecological resilience of each unit to be evaluated is calculated.

6. The method according to claim 5, characterized in that The S34 includes: The evaluation problem of high-standard farmland ecological resilience is denoted as ;in, is the evaluation scheme set, is the total number of evaluation options; is the evaluation index set, is the total number of evaluation indicators; is the unit set to be evaluated, For the Units to be evaluated, is the total number of units to be evaluated; the optimal evaluation indicators of each evaluation scheme are compared and determined within the same unit to be evaluated to form the optimal scheme set , the worst evaluation indicators in each evaluation scheme constitute the worst scheme set ; ； in, 、 are the identity and opposition of the optimal solution set and the worst solution set respectively; 、 Respectively p Evaluation indicators With the collection [ ]’s identity and opposition, For the The weight of each evaluation indicator; Evaluation plan and the optimal solution set Relative closeness for: ； in, To evaluate the program evaluation schemes, .

7. The method according to claim 5, characterized in that The standardization process of the evaluation indicators is as follows: When the correlation between the index value corresponding to the evaluation index and the farmland ecological resilience is positive, the calculation formula of the standardized value is: ； When the correlation between the index value corresponding to the evaluation index and the farmland ecological resilience is negative, the calculation formula of the standardized value is: ； When the index value corresponding to the evaluation index is close to the set threshold and meets the set limit, the calculation formula of the standardized value is: ； in, For the The indicator value corresponding to each evaluation indicator; For multiple units to be evaluated The maximum value of For multiple units to be evaluated The minimum value of To set the threshold; Comparison operation for maximum and minimum values.

8. The method according to claim 1, characterized in that The S4 includes: GeoDA software was used to conduct a bivariate local spatial autocorrelation analysis on the target unit to be evaluated. The spatial interaction relationship between the farmland ecological resilience of the target unit to be evaluated and the grain yield of other surrounding units to be evaluated was obtained. The spatial interaction relationship included four types: high-high area, high-low area, low-high area, and low-low area. The calculation formula of the spatial interaction relationship is: ； in, I is the bivariate global autocorrelation coefficient, t is the total number of grids, is the spatial weight matrix; and Variables: farmland ecological resilience In the grid Value and food production In the grid The value of is the variance of farmland ecological resilience and grain yield.

9. The method according to claim 5, characterized in that The S5 includes: S51. Set high-high areas as high-priority construction areas, high-low and low-high areas as secondary-priority construction areas, and low-low areas as non-priority construction areas; S52. Using a self-organizing map neural network and K-means, the standardized values ​​corresponding to each evaluation indicator are clustered into areas dominated by soil fertility improvement, areas dominated by drought and flood disaster prevention, and areas dominated by land leveling; S53. Based on S52 and S53, a timing optimization plan for high-standard farmland construction is obtained.

Images