Land space planning data intelligent analysis method and system

Abstract

The invention provides an intelligent analysis method and system for territorial space planning data, and relates to the technical field of data processing, and the method comprises the steps: recognizing an overlapping region through comparative analysis; performing buffer area analysis on the geographic information data, the land utilization data and the urban planning data, and creating buffer areas for rivers, roads and protection areas to identify conflict areas; according to the overlapping region and the conflict region, a corresponding territorial space planning scheme is found through iterative search; performing pattern recognition on the territorial space planning scheme to obtain a planning risk assessment result; according to the planning risk assessment result, the risks are divided into different categories, and each risk is quantified to obtain a specific score of each risk category; and obtaining an overall risk assessment result of the whole territorial space planning scheme according to the scores of the various risks. According to the invention, the accuracy and efficiency of risk assessment can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent analysis method for national land space planning data. Background Art

[0002] In the field of national space planning, data processing and analysis are crucial links. Some traditional national space planning data processing methods face many challenges and defects, which affect the accuracy and efficiency of planning.

[0003] For example, some traditional planning methods lack comprehensive data support. When formulating planning schemes, if we fail to fully utilize and integrate multi-source information such as geographic information data, land use data, and urban planning data, it may lead to the planning scheme being out of touch with the actual situation and difficult to effectively guide practice.

[0004] Secondly, traditional methods have limitations in identifying overlapping and conflicting areas. Due to the lack of efficient spatial overlay analysis and buffer analysis tools, traditional methods sometimes have difficulty accurately identifying overlapping areas between different planning layers and potential conflicting areas, such as conflicts in sensitive areas such as rivers, roads and protected areas, which may lead to obstacles in the actual implementation of planning schemes. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent analysis method for national land space planning data, which can improve the accuracy and efficiency of risk assessment.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] In a first aspect, a method for intelligent analysis of land space planning data is provided, the method comprising:

[0008] Collecting national land space planning data from various data sources, including geographic information data, land use data and urban planning data; pre-processing the national land space planning data to obtain pre-processed standardized data;

[0009] Integrate the pre-processed standardized data to build a national land space planning database;

[0010] Perform spatial overlay analysis on geographic information data, land use data and urban planning data, overlay data from different layers together, and identify overlapping areas through comparative analysis;

[0011] Buffer analysis was performed using geographic information data, land use data, and urban planning data to create buffers for rivers, roads, and protected areas to identify conflict areas;

[0012] According to the overlapping areas and conflicting areas, the corresponding national land space planning scheme is found through iterative search;

[0013] Conduct pattern recognition on national land space planning schemes to obtain planning risk assessment results;

[0014] Based on the planning risk assessment results, risks are divided into different categories and quantified to obtain specific scores for each risk category;

[0015] Based on the scores of various risks, the overall risk assessment results of the entire national land space planning program are obtained.

[0016] Furthermore, according to the overlapping areas and conflicting areas, the corresponding national land space planning scheme is found through iterative search, including:

[0017] Set the parameters of the Gray Wolf Algorithm, including the number of wolf packs, the maximum number of iterations, and the dimension of the search space, which corresponds to the number of variables in the planning scheme; the number of wolf packs represents the number of search agents;

[0018] Initialize the wolf pack, that is, generate an initial set of national land space planning schemes, each of which represents a gray wolf, that is, a solution in the search space;

[0019] Define the objective function used to evaluate the quality of each planning alternative;

[0020] Use the objective function to evaluate the fitness of each individual in the initial wolf pack. According to the fitness, the corresponding individual is determined as wolf a, the second best individual is determined as wolf β, the third best individual is determined as wolf δ, and the remaining individuals are determined as wolf ω. For each iteration:

[0021] Update the positions of wolf a, wolf β and wolf δ, adjust the position of wolf ω, and re-evaluate the fitness of each individual in the adjusted wolf pack; update wolf a, wolf β and wolf δ according to the new fitness value until the preset number of iterations is reached. After the iteration is completed, output wolf a as the corresponding national land space planning plan found.

[0022] Furthermore, the calculation formula of the objective function F(x) is:

[0023]

[0024] Where N represents the total number of regions under consideration; is a Gaussian function used to calculate the distance effect between two regions; D represents the distance between the i-th region and the j-th region; σ is the standard deviation of the Gaussian function; A o represents the overlapping area between the i-th region and the j-th region; k s represents the adjustment coefficient; E represents the soil erosion risk value; T sIndicates the judgment threshold of soil sensitivity; C indicates the vegetation coverage of the current area; V indicates the maximum vegetation coverage; W indicates the distance from the current area to the nearest water source protection area; σ u represents the standard deviation of the decay rate of the control water source protection area proximity score; k s represents the adjustment coefficient of the terrain slope sensitivity score; S represents the terrain slope of the current area; T a It represents the threshold for determining the slope sensitivity; i and j represent the index value of the area.

[0025] Furthermore, pattern recognition is performed on the national land space planning scheme to obtain planning risk assessment results, including:

[0026] Collect historical data related to national land space planning, including planning schemes and implementation results; clean the data to obtain processed data; extract features related to planning risk assessment from the processed data, including geographical location, land use type and environmental sensitivity;

[0027] Divide the processed data into training set, test set and test set; label the processed data, that is, assign a risk level label to each sample based on historical risk assessment results;

[0028] The correlation between the features and the risk level is analyzed through variance analysis; the random forest model is trained using the training set data and the random forest model parameters are adjusted; the performance of the random forest model is evaluated using the test set data to obtain the trained random forest model;

[0029] The new national land space planning scheme data is input into the trained random forest model, and the trained random forest model outputs the risk level prediction results for each planning scheme.

[0030] Furthermore, a risk level label is assigned to each sample based on historical risk assessment results, including:

[0031] Determine the predefined risk level standards; match the historical risk assessment results with the predefined risk level standards to determine the risk level to which each historical sample belongs;

[0032] In the processed data set, a new field is added to store the risk level label;

[0033] According to the risk level of each historical sample, fill in the corresponding risk level label in the newly added label field for each sample.

[0034] Furthermore, the correlation between the characteristics and the risk level was analyzed through variance analysis, including:

[0035] Extract all data belonging to the current risk level group from the data set, calculate the average value of the selected features in the current risk level group; use the square of the difference between each data point in the current risk level group and the average value to find the average value and obtain the variance of the selected features in the current risk level group;

[0036] For each current risk level group, the mean of the selected features was calculated and the variance of the means between groups was calculated;

[0037] According to the variance of the selected features within the current risk level group and the variance of the mean between groups, the intra-group variation is compared with the inter-group variation to obtain the comparison result;

[0038] Based on the comparison results, the correlation between the characteristics and the risk level is determined.

[0039] Furthermore, according to the planning risk assessment results, the risks are divided into different categories and quantified to obtain specific scores for each risk category, including:

[0040] Classify risks into different categories according to their nature, source or scope of impact;

[0041] For each risk category, determine a corresponding quantitative standard indicator;

[0042] Collect historical records related to each risk category and analyze them to extract key information for quantification;

[0043] For each risk category, a quantitative risk assessment is conducted based on its quantitative standards and key information to calculate the specific score for each risk category.

[0044] Second, a national land space planning data intelligent analysis system, including:

[0045] The acquisition module is used to collect national land space planning data from various data sources, including geographic information data, land use data and urban planning data; pre-process the national land space planning data to obtain pre-processed standardized data;

[0046] The construction module is used to integrate the pre-processed standardized data and build a national land space planning database;

[0047] The analysis module is used to perform spatial overlay analysis on geographic information data, land use data and urban planning data, overlay data from different layers together, and identify overlapping areas through comparative analysis; perform buffer zone analysis on geographic information data, land use data and urban planning data, create buffer zones for rivers, roads and protected areas, and identify conflicting areas;

[0048] The search module is used to find the corresponding national land space planning scheme through iterative search according to the overlapping areas and conflicting areas;

[0049] The recognition module is used to perform pattern recognition on national land space planning schemes to obtain planning risk assessment results;

[0050] The classification module is used to classify risks into different categories according to the planning risk assessment results, and quantify each type of risk to obtain the specific score of each risk category; based on the scores of each type of risk, the overall risk assessment results of the entire national land space planning scheme are obtained.

[0051] According to a third aspect, a computing device includes:

[0052] one or more processors;

[0053] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0054] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0055] The above scheme of the present invention includes at least the following beneficial effects.

[0056] Centralized data storage and management can be achieved to improve data accessibility and utilization efficiency. By overlaying data from different layers, overlapping areas can be intuitively identified, revealing the spatial relationship between different planning elements. This helps to discover potential spatial conflicts and optimize planning layouts. Creating buffers for specific geographic elements (such as rivers, roads, and protected areas) helps to identify conflict areas associated with these elements. This can provide important reference information for planning decisions and ensure that planning schemes are coordinated with the needs of environmental protection and social development. Improve the pertinence and effectiveness of planning schemes. Through iterative search, planning schemes that meet the requirements can be quickly found based on the specific conditions of overlapping areas and conflicting areas, reducing blindness and trial-and-error costs in the planning process. Using pattern recognition technology, potential risks in planning schemes can be automatically identified and evaluated, which helps to improve the accuracy and efficiency of risk assessment. By classifying risks and assigning specific scores, the severity and scope of impact of various risks can be understood in more detail. This helps to formulate targeted risk management measures and reduce uncertainty and potential losses during planning implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of a method for intelligent analysis of land space planning data provided by an embodiment of the present invention.

[0058] Figure 2 It is a schematic diagram of a national land space planning data intelligent analysis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0060] like Figure 1 As shown, an embodiment of the present invention provides a method for intelligent analysis of land space planning data, the method comprising the following steps:

[0061] Step 11, collecting national land space planning data from multiple data sources, including geographic information data, land use data and urban planning data; preprocessing the national land space planning data to obtain preprocessed standardized data;

[0062] Step 12, integrating the preprocessed standardized data to build a national land space planning database;

[0063] Step 13, perform spatial overlay analysis on geographic information data, land use data and urban planning data, overlay data of different layers together, and identify overlapping areas through comparative analysis;

[0064] Step 14: Perform buffer analysis on geographic information data, land use data, and urban planning data to create buffers for rivers, roads, and protected areas to identify conflicting areas;

[0065] Step 15, according to the overlapping areas and conflicting areas, find the corresponding national land space planning scheme through iterative search;

[0066] Step 16, performing pattern recognition on the national land space planning scheme to obtain planning risk assessment results;

[0067] Step 17: Based on the planning risk assessment results, the risks are divided into different categories and each type of risk is quantified to obtain a specific score for each risk category;

[0068] Step 18: Based on the scores of various risks, the overall risk assessment results of the entire national land space planning program are obtained.

[0069] In an embodiment of the present invention, step 11, through preprocessing, can clean and format the data to eliminate noise and inconsistency in the data. Step 12, realize centralized storage and efficient management of data, facilitate subsequent data retrieval and analysis, and the establishment of a database improves the maintainability and scalability of the data. Step 13, can intuitively display the spatial relationship between different planning data layers, help planners quickly identify overlapping or intersecting areas, help to discover potential conflicts and optimize spatial layout, and improve the rationality of planning. Step 14, create buffers for specific geographic elements, which helps to identify potential conflict areas related to these elements. Step 15, improve the pertinence and feasibility of planning schemes. Through iterative search, planning schemes that meet specific constraints can be quickly found, reducing the trial and error costs and time consumption in the planning process. Step 16, use pattern recognition technology to automatically identify potential risk points in planning schemes, improve the efficiency and accuracy of risk assessment, help planners discover and solve potential problems in the early stages, and reduce risks in the planning implementation process. Step 17: By classifying risks and assigning specific scores, the severity and scope of each risk can be assessed more carefully, which helps planners develop more accurate risk management strategies and optimize resource allocation and response measures. Step 18: Provide planners with a comprehensive and quantitative risk assessment report to help them understand the risk status of the planning scheme more comprehensively.

[0070] In another preferred embodiment of the present invention, the specific implementation process of the above step 11 is as follows:

[0071] Select data sources and determine which data sources you need to obtain data from, such as satellite remote sensing image providers (such as Landsat, Sentinel, etc.), online map service providers (such as Google Maps, OpenStreetMap, etc.) or government open data platforms. For data sources that require authentication, obtain the necessary API keys or access credentials. Use existing data crawling tools to automatically download the required geographic information data, land use data, and urban planning data from the selected data sources through the API interface. Set the download area range, time range, data resolution and other parameters as needed. Check and delete duplicate records in the dataset to ensure that each data is unique. Identify and correct errors in the data, such as format errors, logical errors, etc. For data records with missing key information, fill or delete them to ensure data integrity. Use GIS software or programming libraries (such as GDAL / OGR) to convert the collected data into a unified format, such as Shapefile (.shp) or GeoJSON. During the conversion process, ensure that the spatial reference system (CRS) and other metadata of the data are consistent for subsequent spatial analysis. If the data source uses a different coordinate system, it needs to be converted to a standard coordinate system, such as WGS84. Ensure that the units of measurement (such as meters, kilometers, etc.) in all data are consistent, and unify the naming and format of attribute fields in the dataset.

[0072] In another preferred embodiment of the present invention, the specific implementation process of the above step 12 is as follows:

[0073] Determine which database management system (DBMS) to use to store and manage national spatial planning data. In this scenario, select PostgreSQL as the relational database management system and use the PostGIS extension to support the storage and query of geospatial data. Install the PostgreSQL database server and ensure that it runs properly. Install the PostGIS extension to support geospatial data types and functions in PostgreSQL. Configure database parameters such as memory allocation and number of connections as needed to optimize performance.

[0074] Identify key entities in the national spatial planning data, such as geographic information, land use type, urban planning area, etc. Create one or more data tables for each entity. For example, you can create a geographic_info table to store geographic information data, a land_use table to store land use data, and a city_planning table to store urban planning data. Define the necessary fields (columns) in each data table, including primary keys, foreign keys, spatial data fields (such as geometry or geography type fields), and other attribute fields. Define the relationship between data tables through foreign key constraints and other relationships to ensure data integrity and consistency. Prepare the preprocessed geographic information data, land use data, and urban planning data in an appropriate format (such as Shapefile, GeoJSON, etc.) and match the field definition in the database schema. Use the tools provided by PostgreSQL (such as pgAdmin, psql command line tool, or other third-party tools) to import the data into the corresponding data table. For the import of large amounts of data, you can consider using the batch import function to improve efficiency. Create spatial indexes for fields that store geospatial data (such as geometry or geography type fields) to speed up spatial query operations. In PostGIS, you can use the CREATE INDEX statement with USING GIST or USING SPGIST to create a spatial index. Based on query requirements, create common indexes for attribute fields that are frequently used in query conditions to improve query performance. Perform some basic query operations to verify that the imported data is complete and correct, and adjust database parameters, query statements, or index strategies as needed to further optimize database performance.

[0075] In another preferred embodiment of the present invention, in step 13, the geographic information data, land use data and urban planning data are spatially overlaid and analyzed, data of different layers are overlaid together, and overlapping areas are identified through comparative analysis. The specific implementation process is as follows:

[0076] Start the GIS analysis module, load the library and tool set, and prepare for data processing; automatically load geographic information data, land use data, and urban planning data layers from the specified path; verify whether the format and structure of the loaded layer data meet the preset standards. If the data does not meet the requirements, record the error information and stop subsequent operations. Check whether the coordinate system of each layer is consistent. If the coordinate system is not unified, automatically convert the coordinates to ensure that all layers are under the same spatial reference.

[0077] According to the analysis requirements, select the appropriate spatial overlay analysis algorithm, such as "Intersect". In the GIS software or programming environment, call the selected overlay analysis algorithm, pass the geographic information data, land use data and urban planning data as input parameters to the algorithm, set the algorithm parameters, such as the name of the output layer, coordinate system, etc.; the algorithm starts to execute, matches the spatial elements in each layer, and calculates the intersection, union or other spatial relationships according to the type of algorithm. In this process, the algorithm compares the geometric shapes of the elements in different layers to determine the spatial relationship between them.

[0078] After the overlay analysis is completed, the algorithm will generate a new layer containing the overlay results. By comparing and analyzing this new layer, the overlapping areas between different data layers can be automatically identified. These overlapping areas are the common spatial parts in multiple input layers. Once the overlapping areas are identified, different methods can be used to mark them, change the color of the overlapping areas to distinguish them from other non-overlapping areas, and add labels or annotations to the overlapping areas.

[0079] Save the identified overlapping areas as new layers or data files, and automatically export the analysis results according to the preset format and path, such as Shapefile, GeoJSON, etc.; during the whole process, record detailed operation logs, including operation steps, timestamps, and possible errors or warning messages; if exceptions or errors are encountered during the execution process, automatically handle them or notify the user according to the preset error handling mechanism.

[0080] In another preferred embodiment of the present invention, in step 14, the geographic information data, land use data and urban planning data are subjected to buffer zone analysis to create buffer zones for rivers, roads and protected areas to identify conflicting areas. The specific implementation process is as follows:

[0081] Select the features that need to be buffered from the geographic information data layer, such as rivers, roads, and protected areas. You can use selection boxes, query conditions, or other selection tools to select specific features. Determine the distance or range of the buffer, which can be a fixed distance value or a dynamically calculated value based on feature attributes, set the unit of the buffer (such as meters, kilometers, etc.), and define the shape and type of the buffer, such as circular, rectangular, or polygonal buffer.

[0082] Use the buffer analysis tool or function in the GIS software, with the selected features and the set buffer parameters as input, to perform a buffer analysis, which will create buffers around the selected features based on the set parameters.

[0083] From the layer list, select the buffer layer created previously, and select the land use data layer and the urban planning data layer as the input of the overlay analysis; set the parameters of the overlay analysis, such as the name of the output layer, the coordinate system, the field retention rules, etc., run the overlay analysis, and wait for the system to complete the processing. After the overlay is completed, the generated new layer will contain the intersection information of the buffer and the land use and urban planning layers. Through attribute query or spatial query tools, filter out those parts that are located inside the buffer and overlap with specific land use types (such as construction land) or urban planning areas. These overlapping parts are potential conflict areas, indicating that construction activities may be too close to rivers, roads or protected areas.

[0084] Visually mark the identified conflict areas, using different color fills or boundary lines to highlight these areas, add text labels or annotations to the conflict areas to explain the type of conflict and potential issues. Select the marked conflict areas and prepare to export them. Use the export function of the GIS software to save the conflict areas to common GIS data formats, such as Shapefile (.shp) or GeoJSON.

[0085] In a preferred embodiment of the present invention, the above step 15, finding the corresponding national land space planning scheme through iterative search according to the overlapping areas and the conflicting areas, may include:

[0086] Step 151, setting parameters of the gray wolf algorithm, including the number of wolf packs, the maximum number of iterations, and the dimension of the search space. The dimension of the search space corresponds to the number of variables in the planning scheme. The number of wolf packs represents the number of search agents, including: selecting an appropriate number of wolf packs (e.g., 20, 30, or more) based on the complexity of the problem and the availability of computing resources. This number will determine the parallelism and diversity of the search process; setting a reasonable maximum number of iterations (e.g., 1000, 2000, etc.) based on the scale of the problem and the accuracy of the solution. This parameter will control the running time and convergence of the algorithm; setting the dimension of the search space based on the number of variables included in the national land space planning scheme. Each dimension corresponds to a planning variable, such as land use type, development intensity, etc.

[0087] Step 152, initialize the wolf pack, i.e. generate an initial set of land space planning schemes, each of which represents a gray wolf, i.e. a solution in the search space, specifically including: randomly generate initial solutions equal to the number of wolves in the search space, each of which represents a land space planning scheme. Ensure that each solution is within a reasonable range and meets the basic constraints of the planning, regard each initial solution as a gray wolf, and assign corresponding attributes to it, such as position, speed, etc.

[0088] Step 153, defining an objective function for evaluating the quality of each planning solution;

[0089] Step 154, using the objective function to evaluate the fitness of each individual in the initial wolf pack, determining the corresponding individual as wolf a according to the fitness, the second best individual as wolf β, the third best individual as wolf δ, and the remaining individuals as ω; for each iteration, updating the positions of wolf a, wolf β, and wolf δ, adjusting the position of wolf ω, and re-evaluating the fitness of each individual in the adjusted wolf pack; updating wolf a, wolf β, and wolf δ according to the new fitness value until the preset number of iterations is reached, and after the iteration is completed, outputting wolf a as the corresponding national land space planning scheme found, specifically including:

[0090] Each individual in the initial wolf pack is evaluated using a defined objective function to obtain a fitness value, which is used to compare the pros and cons of different planning schemes. According to the fitness value, the best individual is selected as wolf a, the second best individual is selected as wolf β, and the third best individual is selected as wolf δ. These wolves will guide other wolves to search. According to the update mechanism of the gray wolf algorithm (such as strategies based on social hierarchy and hunting behavior), the positions of wolf a, wolf β, and wolf δ are updated. The position update of these leader wolves will affect the search direction of the entire wolf pack. According to the position of the leader wolf and the distance and angle relationship between other wolves and the leader wolf, the positions of other wolves (ω wolves) are adjusted. This adjustment process aims to move the wolf pack to a better solution space area. After each position update, the fitness value of each individual in the wolf pack is re-evaluated using the objective function. This step aims to test whether the planning scheme after the position update is better. According to the new fitness value, wolf a, wolf β, and wolf δ are re-determined. If the new individual is better than the current leader wolf, it will be replaced by the new leader wolf, and the above steps will be repeated until the preset maximum number of iterations is reached or other termination conditions are met (such as the fitness value reaches the preset threshold or there is no significant improvement after multiple consecutive iterations, etc.). After the iteration is completed, wolf a is output as the optimal national land space planning solution found.

[0091] In the embodiment of the present invention, the gray wolf algorithm has a strong global search capability and can effectively find the global optimal solution in a complex solution space. In national land space planning, this means that the optimal planning scheme that meets multiple constraints (such as environmental protection, economic development, social benefits, etc.) can be quickly found. The individuals (i.e., planning schemes) in the gray wolf algorithm can adaptively adjust their positions (i.e., planning parameters) according to the evaluation results of the objective function, thereby continuously optimizing the scheme. This adaptability can make the search process more flexible and effectively cope with the uncertainty and dynamic changes in the planning process. Since the individuals in the gray wolf algorithm update their positions by imitating the behavior of the optimal individual a wolf, the second-best individual β wolf, and the third-best individual δ wolf, this method helps to avoid the search process from falling into a local optimal solution, thereby making it more likely to find the global optimal national land space planning scheme. The wolf pack in the gray wolf algorithm represents a group of parallel search agents that explore different areas in the solution space at the same time. This parallelism not only improves the search efficiency, but also helps to maintain the diversity of solutions and increases the possibility of finding planning schemes that are both innovative and practical. The gray wolf algorithm is relatively simple and intuitive, and is easy to program and implement. At the same time, it can also be combined with other optimization techniques or domain knowledge to further improve the efficiency and accuracy of the search. The Gray Wolf Algorithm is insensitive to the choice of the initial solution and can maintain relatively stable performance under different operating environments. This makes it highly reliable and stable when dealing with complex land space planning problems.

[0092] In a preferred embodiment of the present invention, in the above step 153, the calculation formula of the objective function F(x) is:

[0093]

[0094] Where N represents the total number of regions under consideration; is a Gaussian function used to calculate the distance effect between two regions; D represents the distance between the i-th region and the j-th region; σ is the standard deviation of the Gaussian function; A o represents the overlapping area between the i-th region and the j-th region; k s represents the adjustment coefficient; E represents the soil erosion risk value; T s Indicates the judgment threshold of soil sensitivity; C indicates the vegetation coverage of the current area; V indicates the maximum vegetation coverage; W indicates the distance from the current area to the nearest water source protection area; σ u represents the standard deviation of the decay rate of the control water source protection area proximity score; k s represents the adjustment coefficient of the terrain slope sensitivity score; S represents the terrain slope of the current area; T a It represents the threshold for determining the slope sensitivity; i and j represent the index value of the area.

[0095] In the embodiment of the present invention, the function comprehensively considers multiple key factors related to national land space planning, such as the distance between regions, soil erosion risk, vegetation coverage, proximity to water source protection areas, and terrain slope sensitivity. This comprehensive evaluation method can more comprehensively reflect the pros and cons of the planning scheme. By introducing multiple adjustment coefficients k s and the decision threshold T s , this function can be flexibly adjusted according to the actual situation of a specific region. By introducing the Gaussian function To calculate the influence of distance between regions, it can more accurately simulate the impact of spatial distance on planning schemes. This method not only takes into account the direct distance between regions, but also smoothly reflects the attenuation effect of distance on planning effect in the form of Gaussian function. By quantifying multiple factors into specific values ​​and incorporating them into the objective function, the evaluation of planning schemes can be more objective and comparable.

[0096] The determination process of soil erosion risk value E is as follows:

[0097] Obtain the latest land use status data, which are usually provided in vector data (such as Shapefile) or raster data (such as GeoTIFF) formats. If it is not possible to directly obtain ready-made data, remote sensing images can be used for interpretation. Select high-resolution remote sensing images, such as satellite images or drone aerial images, and use professional remote sensing software (such as ENVI, ERDAS Imagine, etc.) for visual interpretation or supervised classification to identify different land use types; access public geographic data platforms, such as EarthExplorer of USGS (United States Geological Survey) or other similar services, and download DEM data covering the study area. These data are usually provided in GeoTIFF format with meter-level or higher accuracy. Ensure that the downloaded DEM data has the same coordinate system and projection as the land use type data.

[0098] Select remote sensing images of appropriate time and resolution according to research needs. For vegetation coverage calculation, use images within the growing season, which can be obtained from commercial satellite image providers (such as DigitalGlobe, Airbus, etc.) or public remote sensing data platforms (such as Sentinel-2, Landsat, etc.). If remote sensing images are used for interpretation, use the classification tools in the remote sensing software (such as maximum likelihood method, support vector machine, etc.) to classify the images; after the classification is completed, perform accuracy assessment to ensure the accuracy of the classification results. The classification results can be verified through field surveys or comparisons with other reliable data sources.

[0099] Load DEM data in GIS software (such as ArcGIS, QGIS, etc.), use the slope analysis tools in GIS software (such as Slope tool) to calculate the slope value of each pixel, which will generate a new slope raster dataset, pre-process the remote sensing image, including radiation calibration, atmospheric correction and other steps to ensure image quality; calculate NDVI (normalized vegetation index), using the vegetation index calculation tool in the remote sensing software; divide the vegetation coverage into different levels (such as low, medium and high coverage) according to the vegetation index value, and generate a vegetation coverage classification dataset; overlay the land use type data, slope data and vegetation coverage data in the GIS software to ensure that they have the same spatial resolution and coordinate system. This step can be achieved by using the overlay analysis tools in the GIS software (such as Overlay, Combine, etc.).

[0100] According to the "Soil Erosion Classification and Grading Standards", combined with land use type, slope and vegetation coverage data, the soil erosion grade is determined for each data point. This can be achieved using the conditional query function in the GIS software to generate a new data set containing soil erosion grade information for each data point; according to the severity and distribution of soil erosion, a specific risk value is assigned to each erosion grade. These risk values ​​can be empirical values ​​obtained through statistical analysis.

[0101] For example, determine the various levels of soil erosion. Soil erosion levels can be divided into slight, mild, moderate, strong and severe, etc., which are divided based on the severity of soil erosion. Next, collect historical soil erosion data in the study area, including the frequency, degree, and scope of erosion, to determine the relative severity of each erosion level. Perform statistical analysis on the historical data to calculate indicators such as the frequency of each erosion level in the historical data, the average erosion depth, and the degree of impact on crops and the ecological environment. These indicators will help quantify the severity of each erosion level. Based on the results of the statistical analysis, assign a specific risk value to each erosion level. These risk values ​​can be numbers between 1 and 10, where 1 represents the lowest risk and 10 represents the highest risk. For example: slight erosion: risk value 1; slight erosion: risk value 3; moderate erosion: risk value 5; strong erosion: risk value 7; severe erosion: risk value 9.

[0102] The threshold T for determining soil sensitivity s The determination process is as follows:

[0103] Collect detailed data on different soil types in the study area, including key indicators such as soil type, soil texture (such as the proportion of sand, loam, and clay), organic matter content, and soil thickness. These data can be obtained through field sampling, laboratory analysis, and existing soil survey data.

[0104] Next, the collected soil basic data, combined with expert knowledge and relevant research results, are used to assess the sensitivity of each soil type. The following factors can be considered in the assessment:

[0105] Soil Type: Different types of soil have different resistance to erosion.

[0106] Soil texture: The size and distribution of soil particles affect the soil's ability to retain water and its resistance to erosion.

[0107] Organic matter content: Organic matter can improve soil structure and increase soil's resistance to erosion.

[0108] Soil thickness: Thicker soil layers generally have a better ability to retain water and fertilizer and are more resistant to erosion. Based on these factors, each soil is rated for sensitivity.

[0109] Based on the results of sensitivity assessment, one or more judgment thresholds are set to distinguish between high and low soil sensitivity. These thresholds can be adjusted and optimized according to actual needs.

[0110] For example, a high sensitivity threshold and a low sensitivity threshold can be set. When the soil sensitivity score is higher than the high sensitivity threshold, the soil is considered to be highly sensitive; when the score is lower than the low sensitivity threshold, it is considered to be low sensitivity; and soils with scores between the two can be classified as medium sensitivity.

[0111] In soil erosion risk analysis, the calculated soil erosion risk value can be compared with the above-set judgment threshold. Based on the comparison results, the soil erosion sensitivity level of a specific area can be determined, thereby providing a basis for subsequent soil protection and management measures. For example, for highly sensitive soil areas, we need to take more stringent protection measures, such as afforestation and increasing ground cover; while for low-sensitivity soil areas, management measures can be relatively loose. Through the above steps, the judgment threshold of soil sensitivity can be specifically determined and applied to actual soil erosion risk analysis.

[0112] The threshold T for determining slope sensitivity a The determination process is as follows:

[0113] Obtain digital elevation model (DEM) data of the study area, which can be obtained from public geographic data platforms or relevant institutions. Next, use GIS software (such as ArcGIS, QGIS, etc.) to process the DEM data to ensure the accuracy and applicability of the data. Use the processed DEM data to perform slope analysis in the GIS software. Through the slope analysis tool, the slope value of each pixel in the study area can be calculated and a slope distribution map can be generated. This slope distribution map will intuitively show the spatial distribution of different slopes in the study area. In order to set the threshold for determining the slope sensitivity, it is necessary to collect historical data related to the slope in the study area. These data can include past soil erosion events, land use changes, vegetation cover, etc. By analyzing these data, the frequency and severity of processes such as soil erosion under different slope conditions can be understood. Based on the analysis results of historical data, combined with expert knowledge and experience, set one or more thresholds for determining the slope sensitivity. These thresholds should be able to reflect the critical point of the influence of slope on processes such as soil erosion. For example, a lower threshold can be set to indicate the starting point where the slope begins to have a significant impact on soil erosion, and a higher threshold can be set to indicate the area where the slope has a very serious impact on soil erosion. In practical applications, the accuracy of the set threshold is verified by comparing the actual soil erosion events in the study area with the slope sensitivity determination results. If it is found that there is a large deviation between the threshold and the actual situation, the threshold can be adjusted appropriately to improve the accuracy of the determination. When evaluating the sensitivity of processes such as soil erosion, the terrain slope of the current area can be compared with the set determination threshold. Based on the comparison results, the sensitivity level of the slope to processes such as soil erosion can be evaluated. For example, if the slope of the current area is higher than the set higher threshold, it can be considered that the area has a very high sensitivity to soil erosion.

[0114] In a preferred embodiment of the present invention, in the above step 16, pattern recognition is performed on the national land space planning scheme to obtain a planning risk assessment result, including:

[0115] Step 161, collect historical data related to national land space planning, including planning schemes and implementation results; clean the data to obtain processed data; extract features related to planning risk assessment from the processed data, including geographical location, land use type and environmental sensitivity, specifically including: obtaining historical data related to national land space planning from databases, archives or relevant research institutions, these data should include past planning scheme documents, implementation results reports, environmental monitoring data, etc.; check the consistency and completeness of the data, delete or fill in missing values, identify and correct outliers or errors in the data, standardize the data format, and ensure that all data use a unified unit of measurement.

[0116] From the cleaned data, locate the field containing geographic location information; use data processing tools or programming languages ​​(such as Python's pandas library) to extract the longitude and latitude information of each data point, and ensure that the longitude and latitude are in the correct format. Usually, the longitude range is -180 to 180, and the latitude range is -90 to 90. Based on the longitude and latitude information, you can use geographic information system (GIS) tools or API services (such as Google Maps Geocoding API) to query the administrative region to which each point belongs. Extract the region name or code in the query result as a new field of the data.

[0117] Identify fields or information related to land use types from the data. Based on the extracted information, classify land use types into residential, industrial, agricultural and other categories. For uncertain or ambiguous situations, manual verification or marking can be performed. Identify fields or information related to ecological protection areas, water sources, and areas prone to geological disasters from the data. Based on the identified sensitive area information, assign an environmental sensitivity level to each data point. The sensitivity level can be set to several levels according to the actual situation, such as high sensitivity, medium sensitivity, low sensitivity, etc. The sensitivity level can be determined based on the spatial analysis method of GIS.

[0118] Step 162, divide the processed data into training set, test set and test set, specifically including: randomly divide the cleaned and feature-extracted data set into training set, validation set and test set. Usually, the training set is used for model training, the validation set is used for model selection, and the test set is used to evaluate model performance; label the processed data, that is, assign a risk level label to each sample according to the historical risk assessment results;

[0119] Step 163, analyzing the correlation between the features and the risk level through variance analysis; training the random forest model using the training set data, adjusting the parameters of the random forest model; and evaluating the performance of the random forest model using the test set data to obtain a trained random forest model, specifically including:

[0120] For categorical features (such as land use type), use One-Hot Encoding or Label Encoding for conversion; use the ANOVA function in statistical software or programming language (such as scipy.stats.f_oneway or statsmodels.formula.api.ols with anova_lm in Python) to perform ANOVA analysis on each feature and risk level, calculate the F value and p value, and judge whether there is a significant correlation between the feature and the risk level based on the p value. According to the ANOVA results, select features that are significantly correlated with the risk level. A p-value threshold (such as 0.05) can be set. Features below this threshold are considered to be significantly correlated with the risk level. Make sure that the training set, validation set, and test set have been divided, and the labels have been encoded, and standardize or normalize the training set to eliminate the impact of the dimension on the model. Initialize the random forest model using the training set data and set the initial model parameters, such as the number of trees, maximum depth, etc.; perform model training and record performance indicators during the training process; use the validation set data to evaluate model performance, adjust model parameters through methods such as grid search (GridSearchCV) or random search (RandomizedSearchCV), and select the parameter combination that performs best on the validation set.

[0121] Based on the performance indicators on the validation set (such as accuracy, recall, F1 score, etc.), select the best performing model. Cross-validation (such as K-fold cross-validation) can be used to more robustly evaluate model performance. Use independent test set data to perform a final evaluation of the selected model, and calculate and record indicators such as accuracy, recall, and F1 score on the test set.

[0122] Step 164, input the new national land space planning scheme data into the trained random forest model, and the trained random forest model outputs the risk level prediction results of each planning scheme, specifically including:

[0123] Obtain the latest national land space planning data and make sure that this data includes key information such as the specific content of the planning, geographic location, land use type, etc. Clean the newly collected data to remove duplicate, invalid or erroneous data. Extract features related to risk assessment, which should be consistent with the features used in model training, including geographic location, land use type, and environmental sensitivity. Ensure that the extracted feature data format and units are consistent with the training data so that the model can process it correctly.

[0124] Preprocess the extracted features, such as standardizing or normalizing, to match the data distribution during model training. If one-hot encoding or label encoding was used to process categorical features during model training, the new data should also be encoded in the same way. Input the preprocessed new planning scheme data into the previously trained random forest model. Run the random forest model to predict the newly input data. The model will output the risk level prediction results for each new planning scheme, which are usually expressed in the form of classification labels (such as low risk, medium risk, high risk) or probability values.

[0125] In an embodiment of the present invention, by using historical data and machine learning algorithms (such as random forest models), the risk assessment of new national land space planning schemes can be automatically performed, greatly improving the accuracy and efficiency of the assessment. Compared with traditional manual assessment methods, this pattern recognition method is more objective, faster, and can process large amounts of data. The method comprehensively considers multiple characteristics such as geographical location, land use type, and environmental sensitivity, so that the risks of planning schemes can be more comprehensively assessed. This multi-dimensional assessment method helps to discover potential risk points. By predicting the risk level of new national land space planning schemes, the method can provide decision makers with timely risk warnings. This helps to discover potential problems before the plan is implemented, so that corresponding preventive measures can be taken to reduce the possibility of risks.

[0126] In a preferred embodiment of the present invention, in the above step 162, a risk level label is assigned to each sample according to the historical risk assessment result, including:

[0127] Step 1621, determine the predefined risk level standard, for example, collect historical data containing risk assessment results from relevant databases or archives; clean the collected historical data to remove samples with duplicate, invalid or missing key information; preprocess the data, including data standardization and normalization, to ensure data consistency and comparability; use statistical methods to perform exploratory analysis on historical data to identify distribution characteristics, outliers, clustering, etc. in the data to initially understand the overall situation of the risk; determine key indicators for dividing risk levels based on the characteristics of historical data and the needs of risk assessment, which may include risk assessment scores, probability of occurrence, impact, etc.; preliminarily set risk level division standards and thresholds based on data exploratory analysis and expert judgment; for example, samples with risk assessment scores within a certain range may be set as "low risk", samples with risk assessment scores within another range may be set as "medium risk", and samples with risk assessment scores within another range may be set as "high risk"; match the historical risk assessment results with the predefined risk level standard to determine the risk level to which each historical sample belongs, specifically including: collecting and organizing historical risk assessment data sets to ensure data integrity and accuracy; and formulating a matching rule or algorithm based on the predefined risk level standard. For example, a series of thresholds can be set. When the risk assessment result falls within a certain threshold range, it is classified as the corresponding risk level. Apply this matching algorithm to match the risk assessment result of each historical sample with the risk level standard, and record the risk level obtained by matching each historical sample.

[0128] Step 1622, in the processed data set, a new field is added to store the risk level label, specifically including: using data processing software (such as Excel, Python pandas library, etc.) to open the cleaned and feature-extracted data set; adding a new column (field) to the data set, named "Risk Level" or a similar name, to store the risk level label of each sample; ensuring that the data type of the new field is suitable for storing text or categorical data, because risk levels are usually expressed in text form (such as "low risk", "medium risk", "high risk").

[0129] Step 1623, according to the risk level of each historical sample, fill in the corresponding risk level label in the newly added label field for each sample, specifically including:

[0130] According to the risk level determined in step 1621, prepare a list or dictionary of risk level labels, for example: {'low risk': 1, 'medium risk': 2, 'high risk': 3}. Traverse the processed data set, for each sample, according to its risk level determined in step 1621, select the corresponding label from the risk level label list or dictionary, and fill the selected risk level label into the "risk level" field newly added in step 1622 for the sample.

[0131] In an embodiment of the present invention, by pre-defining risk level standards, it can be ensured that the risk assessment results of all samples follow the same measurement scale. Simplifying complex risk assessment results into several clear risk level labels can greatly simplify the data analysis process. In machine learning tasks, clear labels can significantly improve the training efficiency of the model. By assigning a risk level label to each sample, classification algorithms such as random forests can more easily learn the relationship between features and risk levels, thereby converging faster and achieving higher prediction accuracy.

[0132] In a preferred embodiment of the present invention, in the above step 163, the correlation between the feature and the risk level is analyzed by variance, including:

[0133] Step 1631, extract all data belonging to the current risk level group from the data set, calculate the average value of the selected features in the current risk level group; use the square of the difference between each data point in the current risk level group and the average value to find the average value, and obtain the variance of the selected features in the current risk level group, specifically including:

[0134] According to the risk level label, all data belonging to the current risk level group are filtered out from the entire data set to ensure that the extracted data contains the selected features that need to be analyzed; for the filtered data in the current risk level group, for the selected features, the average value is calculated, and the average value is the sum of the specific feature values ​​of all data points divided by the number of data points; for each data point in the current risk level group, the difference between its selected feature value and the average value is calculated, and each difference is squared to eliminate the influence of negative values. The average value of all squared differences is calculated to obtain the variance of the selected features in the current risk level group.

[0135] Step 1632, for each current risk level group, calculate the average value of the selected features and calculate the variance of the means between groups, specifically including: for each risk level group in the data set, repeat the average value calculation process in step 1631 to obtain the average value of the selected features in each risk level group, use the feature average value of each risk level group, and calculate the variance between these average values, which reflects the degree of variation of the selected features between different risk level groups.

[0136] Step 1633, based on the variance of the selected features within the current risk level group and the variance of the inter-group mean, compare the intra-group variation with the inter-group variation to obtain a comparison result, specifically including: using the variance of the selected features within each risk level group calculated in step 1631, these variances represent the degree of data variation within the group; using the variance of the inter-group mean calculated in step 1632, this variance represents the degree of variation of the selected features between different risk level groups; comparing the intra-group variance of each risk level group with the variance of the inter-group mean, which is usually accomplished by calculating the F statistic, which is the ratio of the inter-group variance to the intra-group variance.

[0137] Step 1634, based on the comparison results, determine the correlation between the feature and the risk level, specifically including: based on the comparison results in step 1633 (such as the size of the F statistic), the correlation between the selected feature and the risk level can be determined; if the inter-group variation is significantly greater than the intra-group variation, it means that the selected feature has significant differences between different risk level groups, so the feature may have a strong correlation with the risk level. Set a significance level (such as 0.05), and determine whether the comparison result is significant by consulting the critical value table of the F distribution or using statistical software; if the F statistic exceeds the critical value, reject the null hypothesis (that is, the feature value is not related to the risk level), and it is considered that there is a significant correlation between the feature and the risk level.

[0138] In an embodiment of the present invention, features that are significantly correlated with risk levels can be accurately identified through variance analysis. Variance analysis can not only determine whether there is a correlation between features and risk levels, but also quantify the strength of this correlation. By comparing the intra-group variance and the inter-group variance, the degree of variation of features between different risk level groups can be understood, so as to more accurately understand the impact of features on risk levels. In a data set, there may be a large number of features, but not all features are closely related to risk levels. Through variance analysis, features with strong correlation with risk levels can be screened out, data dimensions can be reduced, model complexity can be simplified, and computational efficiency can be improved.

[0139] In a preferred embodiment of the present invention, the above step 17, according to the planning risk assessment results, divides the risks into different categories, quantifies each type of risk, and obtains a specific score for each risk category, including:

[0140] Step 171, classify the risks into different categories according to the nature, source or scope of impact of the risks, including: preliminary identification of all possible risks, which can be accomplished through reviewing project plans, market research, expert consultation, past experience, etc. The basis for determining risk classification can be the nature of the risk (such as technical risk, market risk, financial risk, etc.), source (such as internal risk, external risk) or scope of impact (such as local impact, global impact). Classify the initially identified risks according to the determined classification basis. For example, all risks related to technology implementation are classified into the "technical risk" category, and all risks related to changes in market demand are classified into the "market risk" category, etc.

[0141] Step 172, for each risk category, determine a corresponding quantitative standard indicator, specifically including: for each risk category, select one or more quantitative indicators that can reflect the size of the risk in this category. For example, for "technical risk", you can select indicators such as "technical maturity" and "technical update speed"; for "market risk", you can select indicators such as "market demand volatility" and "number of competitors". Determine the specific calculation method and data source for each quantitative indicator. For example, "technical maturity" can be obtained through expert scoring or technical evaluation reports, and "market demand volatility" can be calculated through market research data. Set a reasonable threshold range for each quantitative indicator so that the size of the risk can be judged based on the indicator value later.

[0142] Step 173, collect historical records related to each risk category, analyze the historical records to extract key information for quantification, specifically including: for each risk category, collect relevant historical record data. These data can come from the company's internal database, project documents, market research reports, etc. Clean and organize the historical record data to ensure the accuracy and completeness of the data. By analyzing the historical record data, extract the key information related to each risk category. For example, for "technical risk", key information such as the time to solve technical problems in past projects and cost overruns can be extracted.

[0143] Step 174, for each risk category, a quantitative risk assessment is performed based on its quantitative standards and key information to calculate a specific score for each risk category, specifically including:

[0144] A scoring range is set for each risk category, for example, 1 to 10 points, where 1 point indicates very low risk and 10 points indicates very high risk. For example, for the indicator "technology maturity", the scoring criteria can be set as follows: 1 point for very mature and widely used technology, 3 points for a certain degree of maturity but limited application, 7 points for technology in the experimental stage and less application, and 10 points for technology completely unverified. The key information extracted in step 173 is matched with the scoring criteria. According to the key information of each risk category, a specific score is assigned to each quantitative standard indicator in accordance with the scoring criteria. If a risk category has multiple quantitative standard indicators, a comprehensive score can be calculated, such as taking the average of all indicator scores or taking a weighted average according to the importance of the indicator. For example, if the "technology risk" category has two indicators, "technology maturity" and "technology update speed", which are 7 points and 5 points respectively, the average of these two scores (6 points) can be calculated as the comprehensive score of "technology risk".

[0145] In an embodiment of the present invention, by classifying risks into different categories, the nature and characteristics of various risks can be more accurately identified and understood. This classification helps assessors to develop more accurate assessment strategies for different categories of risks, thereby improving the accuracy of overall risk assessment. By determining corresponding quantitative standard indicators for each risk category and performing quantitative assessments based on historical records, the originally abstract concept of risk can be converted into specific numerical values. By quantitatively assessing various types of risks and calculating specific scores, risks with higher scores and greater potential impacts can be prioritized based on the scores. This targeted risk response strategy can improve the efficiency of risk management and ensure that limited resources are prioritized in the most critical risk areas, thereby improving the overall effectiveness of risk response.

[0146] In a preferred embodiment of the present invention, the above step 18 may include:

[0147] Collect the specific scores for each risk category from step 174, create a risk score list or risk matrix, fill in the score of each risk category in the corresponding position, and ensure that the list or matrix clearly and accurately reflects the scores of all identified risk categories.

[0148] Perform statistical analysis on the aggregated risk scores. By calculating statistics such as standard deviation and variance, analyze the concentration or dispersion of risks, and identify risk categories with high scores or concentrated distribution, which may be risk hotspots. Review and analyze historical data, especially the frequency of risk occurrence and degree of loss of similar planning projects, and use this as a basis to assign weights to risk categories. Based on expert opinions and historical data analysis results, assign a specific weight value to each risk category. The weight value can be a number between 0 and 1, and the sum of the weights of all risk categories should be 1. Use the weighted average method to calculate the overall risk assessment result; based on the calculated overall risk assessment result, a threshold range can be set to divide the results into different risk levels, such as low risk, medium risk, and high risk.

[0149] like Figure 2 As shown, an embodiment of the present invention further provides a national land space planning data intelligent analysis system, comprising:

[0150] The acquisition module is used to collect national land space planning data from various data sources, including geographic information data, land use data and urban planning data; pre-process the national land space planning data to obtain pre-processed standardized data;

[0151] The construction module is used to integrate the pre-processed standardized data and build a national land space planning database;

[0152] The analysis module is used to perform spatial overlay analysis on geographic information data, land use data and urban planning data, overlay data from different layers together, and identify overlapping areas through comparative analysis; perform buffer zone analysis on geographic information data, land use data and urban planning data, create buffer zones for rivers, roads and protected areas, and identify conflicting areas;

[0153] The search module is used to find the corresponding national land space planning scheme through iterative search according to the overlapping areas and conflicting areas;

[0154] The recognition module is used to perform pattern recognition on national land space planning schemes to obtain planning risk assessment results;

[0155] The classification module is used to classify risks into different categories according to the planning risk assessment results, and quantify each type of risk to obtain the specific score of each risk category; based on the scores of each type of risk, the overall risk assessment results of the entire national land space planning scheme are obtained.

[0156] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0157] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0158] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

Claims

1. A method for intelligent analysis of national land space planning data, characterized in that: The method comprises the following steps: Collecting national land space planning data from various data sources, including geographic information data, land use data and urban planning data; pre-processing the national land space planning data to obtain pre-processed standardized data; Integrate the pre-processed standardized data to build a national land space planning database; Perform spatial overlay analysis on geographic information data, land use data and urban planning data, overlay data from different layers together, and identify overlapping areas through comparative analysis; Buffer analysis was performed using geographic information data, land use data, and urban planning data to create buffers for rivers, roads, and protected areas to identify conflict areas; According to the overlapping areas and conflicting areas, the corresponding national land space planning scheme is found through iterative search; Conduct pattern recognition on national land space planning schemes to obtain planning risk assessment results; Based on the planning risk assessment results, risks are divided into different categories and quantified to obtain specific scores for each risk category; Based on the scores of various risks, the overall risk assessment results of the entire national land space planning program are obtained.

2. The method for intelligent analysis of land space planning data according to claim 1, characterized in that: According to the overlapping areas and conflicting areas, the corresponding national land space planning scheme is found through iterative search, including: Set the parameters of the Gray Wolf Algorithm, including the number of wolf packs, the maximum number of iterations, and the dimension of the search space, which corresponds to the number of variables in the planning scheme; the number of wolf packs represents the number of search agents; Initialize the wolf pack, that is, generate an initial set of national land space planning schemes, each of which represents a gray wolf, that is, a solution in the search space; Define the objective function used to evaluate the quality of each planning alternative; Use the objective function to evaluate the fitness of each individual in the initial wolf pack. According to the fitness, the corresponding individual is determined as wolf a, the second best individual is determined as wolf β, the third best individual is determined as wolf δ, and the remaining individuals are determined as wolf ω. For each iteration: Update the positions of wolf a, wolf β and wolf δ, adjust the position of wolf ω, and re-evaluate the fitness of each individual in the adjusted wolf pack; update wolf a, wolf β and wolf δ according to the new fitness value until the preset number of iterations is reached. After the iteration is completed, output wolf a as the corresponding national land space planning plan found.

3. The method for intelligent analysis of land space planning data according to claim 2 is characterized in that: The calculation formula of the objective function F(x) is: Where N represents the total number of regions under consideration; is a Gaussian function used to calculate the distance effect between two regions; D represents the distance between the i-th region and the j-th region; σ is the standard deviation of the Gaussian function; A o represents the overlapping area between the i-th region and the j-th region; k s represents the adjustment coefficient; E represents the soil erosion risk value; T s Indicates the judgment threshold of soil sensitivity; C indicates the vegetation coverage of the current area; V indicates the maximum vegetation coverage; W indicates the distance from the current area to the nearest water source protection area; σ u represents the standard deviation of the decay rate of the control water source protection area proximity score; k s represents the adjustment coefficient of the terrain slope sensitivity score; S represents the terrain slope of the current area; T a It represents the threshold for determining the slope sensitivity; i and j represent the index value of the area.

4. The method for intelligent analysis of land space planning data according to claim 3 is characterized in that: Conduct pattern recognition on national land space planning schemes to obtain planning risk assessment results, including: Collect historical data related to national land space planning, including planning schemes and implementation results; clean the data to obtain processed data; extract features related to planning risk assessment from the processed data, including geographical location, land use type and environmental sensitivity; Divide the processed data into training set, test set and test set; label the processed data, that is, assign a risk level label to each sample based on historical risk assessment results; The correlation between the features and the risk level is analyzed through variance analysis; the random forest model is trained using the training set data and the random forest model parameters are adjusted; the performance of the random forest model is evaluated using the test set data to obtain the trained random forest model; The new national land space planning scheme data is input into the trained random forest model, and the trained random forest model outputs the risk level prediction results for each planning scheme.

5. The method for intelligent analysis of land space planning data according to claim 4 is characterized in that: Assign a risk level label to each sample based on historical risk assessment results, including: Determine the predefined risk level standards; match the historical risk assessment results with the predefined risk level standards to determine the risk level to which each historical sample belongs; In the processed data set, a new field is added to store the risk level label; According to the risk level of each historical sample, fill in the corresponding risk level label in the newly added label field for each sample.

6. The method for intelligent analysis of land space planning data according to claim 5 is characterized in that: Correlations between characteristics and risk levels were analyzed through variance analysis, including: Extract all data belonging to the current risk level group from the data set, calculate the average value of the selected features in the current risk level group; use the square of the difference between each data point in the current risk level group and the average value to find the average value and obtain the variance of the selected features in the current risk level group; For each current risk level group, the mean of the selected features was calculated and the variance of the means between groups was calculated; According to the variance of the selected features within the current risk level group and the variance of the mean between groups, the intra-group variation is compared with the inter-group variation to obtain the comparison result; Based on the comparison results, the correlation between the characteristics and the risk level is determined.

7. The method for intelligent analysis of land space planning data according to claim 6 is characterized in that: Based on the planning risk assessment results, risks are divided into different categories and quantified to obtain specific scores for each risk category, including: Classify risks into different categories according to their nature, source or scope of impact; For each risk category, determine a corresponding quantitative standard indicator; Collect historical records related to each risk category and analyze them to extract key information for quantification; For each risk category, a quantitative risk assessment is conducted based on its quantitative standards and key information to calculate the specific score for each risk category.

8. A national land space planning data intelligent analysis system, characterized in that: Applied to the method according to any one of claims 1 to 7, comprising: The acquisition module is used to collect national land space planning data from various data sources, including geographic information data, land use data and urban planning data; pre-process the national land space planning data to obtain pre-processed standardized data; The construction module is used to integrate the pre-processed standardized data and build a national land space planning database; The analysis module is used to perform spatial overlay analysis on geographic information data, land use data and urban planning data, overlay data from different layers together, and identify overlapping areas through comparative analysis; perform buffer zone analysis on geographic information data, land use data and urban planning data, create buffer zones for rivers, roads and protected areas, and identify conflicting areas; The search module is used to find the corresponding national land space planning scheme through iterative search according to the overlapping areas and conflicting areas; The recognition module is used to perform pattern recognition on national land space planning schemes to obtain planning risk assessment results; The classification module is used to classify risks into different categories according to the planning risk assessment results, and quantify each type of risk to obtain the specific score of each risk category; based on the scores of each type of risk, the overall risk assessment results of the entire national land space planning scheme are obtained.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images