Land space planning current situation evaluation method

Abstract

The invention discloses a territorial space planning current situation evaluation method, and belongs to the technical field of territorial space planning. Comprising the following steps: step 1, constructing an initial evaluation unit data set adapted to geographic features; 2, constructing a multi-level index system covering the land use structure, the spatial form and the ecological environment dimension, setting scale sensitivity parameters and threshold intervals for the indexes, and quantitatively analyzing the correlation among the indexes under different scales; step 3, calculating dispersion and spatial heterogeneity of internal indexes of the evaluation units, and realizing self-adaptive segmentation and combination of the evaluation units in combination with association strength between adjacent units; step 4, realizing cooperative calculation of a multi-scale evaluation result through a scale conversion matrix and spatial autocorrelation analysis; 5, optimizing an evaluation result through parameter adjustment; and step 6, displaying the multi-scale evaluation result, the dynamic evolution process of the evaluation unit and the spatial distribution characteristics of each evaluation index.

Description

Technical Field

[0001] The present application relates to the technical field of national land space planning, and more specifically, to a method for assessing the current status of national land space planning. Background Art

[0002] With the acceleration of urbanization and continued population growth, national land and spatial planning has become increasingly important in social development. Reasonable national land and spatial planning can promote economic development, improve the ecological environment, and enhance the quality of life. However, existing assessment methods face numerous limitations when addressing complex and changing geographic characteristics and diverse planning needs. First, traditional assessment methods often rely on single-scale data analysis and lack comprehensive consideration of multi-scale spatial information. This approach fails to fully reflect the dynamic changes and multi-level characteristics of national land and spatial planning, resulting in insufficient accuracy and comprehensiveness in assessment results. Second, existing methods often use a fixed division model for assessment units, failing to adapt to the distribution characteristics of geographic elements and actual needs. This approach can easily lead to significant heterogeneity in characteristics within assessment units, affecting the accuracy of assessment results. Furthermore, when dealing with multiple assessment indicators, traditional assessment methods often overlook the correlations and interactions between indicators, failing to establish a scientific and reasonable indicator system. The setting of indicator weights and the determination of evaluation criteria rely on subjective judgment and lack quantitative analysis, making it difficult to ensure the objectivity and scientific nature of assessment results. Furthermore, the visualization of spatial information is still underutilized in national land and spatial planning assessments. Traditional evaluation results are mostly displayed in static charts, which cannot intuitively reflect the dynamic evolution process and spatial distribution characteristics of the evaluation units, limiting the application value of the evaluation results in planning decisions.

[0003] To sum up, how to establish a multi-scale, multi-dimensional and dynamic method for assessing the current status of national land space planning has become a technical problem that needs to be solved urgently. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the purpose of this application is to provide a method for assessing the current status of land space planning, including the following steps:

[0005] Step 1: construct an initial assessment unit dataset adapted to geographical characteristics;

[0006] Step 2: Construct a multi-level indicator system covering land use structure, spatial form, and ecological environment dimensions, set scale sensitivity parameters and threshold intervals for each indicator, and quantitatively analyze the correlation between indicators at different scales;

[0007] Step 3: Calculate the dispersion and spatial heterogeneity of the internal indicators of the evaluation unit, combine the correlation strength between adjacent units, and realize the adaptive segmentation and merging of the evaluation unit to ensure that the evaluation unit matches the indicator characteristics;

[0008] Step 4: Apply the analytic hierarchy process to determine the weights of indicators at different scales, and achieve collaborative calculation of multi-scale assessment results through scale conversion matrix and spatial autocorrelation analysis;

[0009] Step 5: Carry out cross-validation based on the error tolerance standard, automatically trace and optimize key indicators and evaluation units, and optimize the evaluation results through parameter adjustment;

[0010] Step 6: Use spatial information visualization technology to display multi-scale assessment results, the dynamic evolution process of assessment units, and the spatial distribution characteristics of various assessment indicators, providing an intuitive reference for national land space planning decisions.

[0011] Furthermore, step 1 includes the following steps:

[0012] Collect and pre-process high-resolution remote sensing imagery data, integrate existing basic geographic data, including vector data of terrain, land use, and administrative boundaries, and unify data resolution and projection coordinate systems to ensure data consistency;

[0013] According to the geographical characteristics and assessment requirements of the target area, an initial assessment grid is constructed to characterize the spatial distribution characteristics of geographical elements;

[0014] Quantitatively analyze the changes in terrain slope, land use types, and ecosystem types in different geographical units within the target area, and establish a threshold system for grid optimization;

[0015] Based on the spatial distribution characteristics and attribute similarity of geographic elements, a coarse-grained merging of the initial grid is performed to form a basic evaluation unit;

[0016] Perform spatial topological relationship checks between grid cells, identify and eliminate voids or overlapping areas, correct abnormal topological relationships, and generate an initial evaluation unit dataset.

[0017] Furthermore, step 2 includes the following steps:

[0018] Establish first-level indicators based on land use structure characteristics, including land use type composition ratio, land use mix and construction land density indicators, while also constructing spatial morphology indicators and ecological environment indicators to form a multi-dimensional comprehensive evaluation system;

[0019] Analyze the variation patterns of each indicator value with the scale of the assessment unit, determine the optimal evaluation scale for each indicator, and identify and quantify the corresponding sensitivity parameters;

[0020] Based on regional development characteristics, planning standards and ecological environmental carrying capacity, set grading thresholds for various indicators;

[0021] Conduct indicator correlation analysis at different spatial scales to identify the interaction relationship between evaluation indicators and quantitatively analyze the degree of correlation between indicators;

[0022] Based on the results of correlation analysis, highly correlated indicators are eliminated or merged, and independent evaluation indicators are supplemented to form the final indicator system.

[0023] Furthermore, the correlation analysis of indicators is carried out at different spatial scales to identify the interaction relationship between the evaluation indicators and quantitatively analyze the degree of correlation between the indicators, including the following steps:

[0024] Construct a multi-level spatial unit system, determine the analysis range of each scale, and standardize the evaluation indicators at each scale to eliminate the dimensionality effect;

[0025] Calculate the correlation coefficient matrix between evaluation indicators at each spatial scale level, and preliminarily judge the direction and strength of the relationship between indicators through the positive and negative and numerical values of the correlation coefficients;

[0026] By controlling other variables, we can identify the direct correlation between any two evaluation indicators and eliminate the interference effect of the third variable;

[0027] Construct the grey correlation matrix between indicators and quantify the dynamic correlation characteristics between evaluation indicators;

[0028] Based on the set correlation threshold, screen the pairs of indicators with significant correlation and grade the correlation strength;

[0029] Compare the differences in indicator correlation characteristics at different spatial scales, summarize the scale-dependent laws of indicator relationships, and identify indicator correlations with scale sensitivity;

[0030] Based on the relationship matrix, the interaction network structure between evaluation indicators is visualized, key node indicators and important association paths are identified and labeled, and the overall association characteristics are revealed.

[0031] Furthermore, step 3 includes the following steps:

[0032] Quantitatively calculate the coefficient of variation of indicators in the three dimensions of land use structure, spatial form and ecological environment for each assessment unit;

[0033] Estimate the spatial heterogeneity of elements within a unit, including calculating the landscape diversity index, topographic heterogeneity, land use heterogeneity, and spatial distribution heterogeneity of ecosystem service functions;

[0034] Calculate the index difference between adjacent evaluation units based on the proximity analysis method;

[0035] Set the internal heterogeneity threshold α 内部and the inter-unit difference threshold β, where: when the internal heterogeneity of the evaluation unit is greater than α 内部 When , it is divided into smaller evaluation units; when the difference between adjacent evaluation units is less than β, they are merged into a larger evaluation unit;

[0036] Through iterative optimization, a basic evaluation unit system that meets evaluation needs is formed.

[0037] Furthermore, step 4 includes the following steps:

[0038] A judgment matrix was established through expert scoring, and each indicator was compared pairwise. The eigenvectors were calculated and consistency tests were performed to determine the relative importance weights of each indicator at each scale level.

[0039] Construct a scale conversion matrix to achieve the conversion of evaluation indicators and assessment results between different spatial scales, ensuring the comparability and consistency of evaluation results between different scales;

[0040] Conduct multi-scale spatial agglomeration feature analysis and analyze the spatial distribution patterns of evaluation indicators at different scales;

[0041] Comprehensively consider the indicator weights, scale conversion relationships and spatial autocorrelation characteristics, establish a collaborative calculation model for multi-scale assessment results, and achieve the organic integration of assessment results at different scales;

[0042] The consistency test and uncertainty analysis of the multi-scale assessment results are carried out to verify the reliability and stability of the assessment results at different scales, and the key influencing factors are identified through sensitivity analysis.

[0043] Furthermore, step 5 includes the following steps:

[0044] Taking into account data accuracy, assessment objectives, and actual application needs, formulate error tolerance standards, including setting acceptable error ranges and confidence levels;

[0045] Use the k-fold cross-validation method to perform repeated validation, calculate the error of the evaluation results under different evaluation units and indicator combinations, record the evaluation units and indicators whose errors exceed the tolerance, and establish a complete validation record database;

[0046] Systematically analyze the sources of evaluation errors, including the rationality of the division of evaluation units, the completeness of the indicator system, and the accuracy of scale conversion;

[0047] Based on the cross-validation results, the evaluation parameters are optimized, including adjusting the indicator weight coefficients and correcting the scale conversion parameters. Through multiple rounds of iterative optimization, the evaluation error is continuously reduced.

[0048] The optimized assessment results are fully verified to ensure their accuracy, spatial continuity and matching degree with regional characteristics, so as to form the final optimization plan.

[0049] Furthermore, the k-fold cross-validation method is used to perform repeated validation, calculate the evaluation result errors under different evaluation units and indicator combinations, record the evaluation units and indicators whose errors exceed the tolerance, and establish a complete validation record database, including the following steps:

[0050] The initial evaluation unit data set is grouped and sorted according to the evaluation unit, the k value of the k-fold cross-validation is determined, and the error tolerance threshold of the evaluation indicator is set;

[0051] The data of each evaluation unit is randomly divided into k subsets of similar size, one of which is used as the validation set, and the remaining k-1 subsets are used as the training set;

[0052] For each round of validation, the model is trained using the training set and then predicted on the validation set, calculating the values of all the evaluation indicators of interest.

[0053] Summarize the results of each indicator of each evaluation unit in k verifications, calculate the mean and standard deviation, compare the actual error with the preset tolerance, and mark the situation where the tolerance is exceeded;

[0054] Record all verification process details, including the evaluation unit number, indicator name, original results of k verifications, statistical analysis results, and whether the tolerance is exceeded, to form a structured verification record;

[0055] Based on the verification record database, a summary report of the verification results is generated, focusing on abnormal situations, and analyzing the causes of evaluation units and indicators that exceed the tolerance, providing a basis for subsequent optimization.

[0056] Furthermore, step 6 includes the following steps:

[0057] Build a multi-dimensional data visualization platform that integrates the assessment results database and geographic information system to display assessment results at different scales. It also supports dynamic data updates and real-time rendering to ensure the accuracy and intuitiveness of visualization effects.

[0058] Clearly display the entire process of the evaluation unit from the initial grid to the final optimization result, including the changes in the unit boundary and the dynamic adjustment of the index value, to help decision makers understand the inherent logic of the evaluation unit optimization;

[0059] Use a variety of visual expressions to intuitively present the spatial distribution of evaluation indicators, and enhance the spatial distribution characteristics through color gradients and symbol changes;

[0060] Realize multi-window and data linkage to comprehensively display the spatial characteristics of assessment results in different dimensions, making it easier to analyze the internal relationships of assessment results from multiple perspectives;

[0061] It provides statistical analysis, spatial query and thematic chart generation functions, offering comprehensive support for national land space planning decision-making.

[0062] Compared with the prior art, this application has the following beneficial effects:

[0063] This application uses a multi-level indicator system, spatial heterogeneity analysis, and multi-scale collaborative computing to achieve adaptive segmentation and optimization of assessment units. Combined with spatial information visualization technology, it can intuitively display assessment results, providing a scientific basis and support for national land space planning decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flow chart of a method for assessing the current status of land space planning disclosed in an embodiment of this application. DETAILED DESCRIPTION

[0065] To make the objectives, technical solutions, and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Throughout the drawings, identical or similar reference numerals represent identical or similar elements or elements having identical or similar functions. The described embodiments are only some, not all, of the embodiments of the present invention.

[0066] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0067] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0068] like Figure 1 As shown, a method for assessing the status quo of national land space planning includes the following steps:

[0069] Step 1: construct an initial assessment unit dataset adapted to geographical characteristics;

[0070] Step 2: Construct a multi-level indicator system covering land use structure, spatial form, and ecological environment dimensions, set scale sensitivity parameters and threshold intervals for each indicator, and quantitatively analyze the correlation between indicators at different scales;

[0071] Step 3: Calculate the dispersion and spatial heterogeneity of the internal indicators of the evaluation unit, combine the correlation strength between adjacent units, and realize the adaptive segmentation and merging of the evaluation unit to ensure that the evaluation unit matches the indicator characteristics;

[0072] Step 4: Apply the analytic hierarchy process to determine the weights of indicators at different scales, and achieve collaborative calculation of multi-scale assessment results through scale conversion matrix and spatial autocorrelation analysis;

[0073] Step 5: Carry out cross-validation based on the error tolerance standard, automatically trace and optimize key indicators and evaluation units, and optimize the evaluation results through parameter adjustment;

[0074] Step 6: Use spatial information visualization technology to display multi-scale assessment results, the dynamic evolution process of assessment units, and the spatial distribution characteristics of various assessment indicators, providing an intuitive reference for national land space planning decisions.

[0075] In this embodiment, constructing an initial assessment unit dataset adapted to geographic characteristics is a crucial first step in the current status assessment of national land spatial planning. Geographic characteristics include information such as topography, climate, vegetation, and population distribution, which are all basic data sources for the assessment. The key to constructing the initial dataset is to divide the various regions in the geographic space according to specific criteria to ensure that the assessment units are representative and adaptable. For example, regional divisions can be based on natural geographic units (such as river basins, mountain ranges, etc.) or administrative boundaries (such as townships, counties, etc.), with each unit containing data such as terrain slope, land use type, and ecosystem services. This data not only provides support for subsequent indicator analysis but also reflects the differences in geographical characteristics between different regions, providing a foundation for optimizing spatial planning. When constructing the dataset, in addition to considering geographic characteristics, it is also necessary to focus on data quality control and the integration of multi-source data. For example, remote sensing technology can be used to obtain satellite imagery, combined with ground survey data and socioeconomic data, to achieve multi-dimensional and multi-scale acquisition of geographic information. In addition, data preprocessing techniques (such as data cleaning, data interpolation, and missing value processing) should not be ignored. Through these means, a comprehensive and accurate initial data set can be provided for national land space planning, providing solid data support for subsequent evaluations.

[0076] In this embodiment, land use structure, spatial form, and ecological environment are the three most important dimensions in national land space planning. They determine the potential, quality, and sustainability of spatial development. Land use structure includes land use types (such as residential, commercial, industrial, agricultural, and green space), spatial form focuses on the rationality of spatial layout (such as functional zoning and density distribution), and ecological environment focuses on the ecological carrying capacity and environmental quality of the land. In the process of constructing an indicator system, it is first necessary to set different indicators for each dimension and determine the weights and scale sensitivity parameters of these indicators. The scale sensitivity parameter reflects the variation characteristics of a certain indicator at different spatial scales and is crucial for multi-scale analysis. For example, certain ecological and environmental indicators show significant regional differences at larger scales, while at smaller scales they are strongly influenced by local factors. Therefore, how to flexibly set threshold intervals and analyze the correlation of indicators at different scales is the key to building an effective indicator system. The construction of a multi-level indicator system must not only consider the comprehensiveness and representativeness of the indicators, but also ensure their scientific and operational nature. By combining data analysis and expert experience, the resulting indicator system can quantify and evaluate the current status and potential of land use in different spatial units, providing precise support for spatial planning decisions.

[0077] In this embodiment, the key to step 3 is to quantitatively analyze the spatial characteristics within the evaluation unit. The specific method includes calculating the discreteness and spatial heterogeneity of the indicators. Discreteness refers to the distribution of a certain indicator within the evaluation unit. A higher discreteness means that the indicator has a larger variation in space. Spatial heterogeneity involves the diversity of different geographical features or indicators within the evaluation unit. Higher spatial heterogeneity usually means that there are significant differences in the land use, ecological environment, etc. in the area. By calculating discreteness and spatial heterogeneity, the characteristic distribution pattern within each evaluation unit can be revealed, providing a basis for subsequent segmentation and merging. In order to optimize the adaptability of the evaluation unit, it is an important step to perform adaptive segmentation and merging based on the correlation strength between adjacent units. The correlation strength can be calculated by methods such as spatial autocorrelation analysis. Strongly correlated units can be considered for merging, while units with weaker correlations should be segmented. This process not only ensures the consistency of the spatial characteristics of the evaluation unit, but also provides more accurate regional divisions for subsequent evaluation and decision-making.

[0078] In this embodiment, in the multi-scale evaluation, how to accurately determine the weight of the indicators at each scale is a challenging problem. The analytic hierarchy process (AHP) is an effective decision analysis method that can help determine the weight of each indicator at different scales. Through expert evaluation or data-driven methods, the analytic hierarchy process decomposes complex decision-making problems into multiple levels, and the indicators in each level are comprehensively analyzed through weight relationships to determine the weight distribution at each scale. The core of step 4 is to weight the multi-dimensional and cross-scale indicators through the analytic hierarchy process to ensure that the evaluation results at each scale can reflect the actual spatial characteristics and planning needs. The application of the scale conversion matrix enables the evaluation results of different scales to be unified, and the spatial autocorrelation analysis helps to reveal the relationship between different evaluation units, thereby effectively evaluating and optimizing the overall plan. This process can achieve synergy between the various scales, ensure that the evaluation results at different scales can complement and support each other, and avoid the deviation caused by the evaluation results of a single scale.

[0079] In this embodiment, step 5 is a key link in improving the accuracy and reliability of the evaluation results. By setting the error tolerance standard, errors can be discovered and corrected in a timely manner during the evaluation process, thereby improving the predictive ability and practicality of the model. The cross-validation method can effectively avoid the overfitting problem and test the stability and accuracy of the model by dividing the data set and conducting multiple training and verification. The automatic tracing function can identify the key indicators and evaluation units that affect the results during the evaluation process, and then make optimization adjustments. Through the error tolerance standard, various parameters can be automatically adjusted during the calculation process to minimize the evaluation error. This not only improves the reliability of the evaluation results, but also helps to discover potential influencing factors, providing a more accurate basis for subsequent planning optimization. Optimizing the adjustment process of key indicators and evaluation units can also promote the adaptive learning and self-improvement of the evaluation model.

[0080] In this embodiment, step 6 is to convert the complex multi-scale evaluation results into intuitive visual images through spatial information visualization technology. This process can not only display the evaluation results, but also display the dynamic evolution process of the evaluation unit and the spatial distribution characteristics of each evaluation indicator. Visualization technology uses tools such as GIS, three-dimensional modeling, and heat maps to make the evaluation results more intuitive and easy to understand, providing an intuitive reference for decision makers. Through spatial information visualization, decision makers can clearly see the changes in each evaluation unit at different scales and promptly identify potential planning problems and spatial irrational phenomena. The application of this technology can greatly improve the decision-making efficiency and accuracy of spatial planning, and also provide visual support for public participation and social supervision.

[0081] In summary, this national land spatial planning status assessment method automates the entire process, from data collection to assessment optimization, through six steps. Each step enhances the scientific nature and accuracy of the assessment from a different perspective. Furthermore, through comprehensive multi-scale and multi-dimensional analysis, it effectively improves the reliability and decision-making value of the assessment results. This method not only provides an efficient and scientific assessment tool for national land spatial planning, but also offers solid technical support for future intelligent planning.

[0082] Furthermore, step 1 includes the following steps:

[0083] High-resolution remote sensing imagery data is collected and preprocessed, while existing basic geographic data, including vector data on topography, land use, and administrative boundaries, is integrated. Data resolution and projection coordinate systems are standardized to ensure data consistency. This step is fundamental to spatial planning assessments and provides a wealth of geographic information. High-resolution remote sensing imagery accurately reflects surface changes, including the spatial distribution of various land use types, such as buildings, roads, green spaces, and farmland. Basic geographic data, such as topography (elevation, slope, aspect, etc.) and administrative boundaries, provide an indispensable reference for spatial delineation and regional analysis. These data often have different resolutions and projection coordinate systems, making the key task of the data preprocessing phase to unify their resolution and coordinate systems. Through techniques such as geometric correction, projection transformation, and resampling, these data from various sources are integrated into a consistent format, ensuring spatial consistency and comparability. Data preprocessing also requires steps such as noise filtering, image fusion, and data accuracy assessment to ensure that the raw data provide accurate information for subsequent analysis. Data integration not only improves data quality and utilization efficiency but also provides precise geographic information support for subsequent spatial analysis, helping decision makers better understand the region's spatial characteristics and development potential. Preprocessed data, as the foundation for assessment, is crucial for other steps (such as grid construction and spatial distribution analysis) and determines the accuracy and reliability of subsequent analytical results.

[0084] Based on the geographic characteristics of the target area and assessment requirements, an initial assessment grid is constructed to characterize the spatial distribution of geographic features. Gridding is a common method in spatial data processing, dividing a large geographic area into small cells (grids). Each grid represents a specific spatial region and carries relevant geographic feature information. The construction of the initial assessment grid considers several factors: geographic cell size, spatial resolution, data availability, and computational efficiency. The size of the grid cell determines the accuracy and level of detail of the assessment. In large-scale spatial planning assessments, grid cells are typically larger to reduce computational effort and increase overall assessment speed. In more refined assessments, grid cells are smaller to provide higher spatial resolution and more detailed analysis results. When constructing the initial assessment grid, optimization is required based on the actual geographic characteristics of the target area. For example, in urban areas, grid construction needs to take into account factors such as road networks, building distribution, and population density. In rural areas or nature reserves, topographical undulations and ecological complexity are key considerations for grid division. By combining remote sensing imagery, basic geographic data, and expert experience, a grid system that aligns with regional characteristics can be constructed. This grid will provide basic support for subsequent spatial analysis, indicator calculation and evaluation, ensuring that spatial distribution characteristics can be accurately mapped to each evaluation unit.

[0085] Quantitatively analyze changes in terrain slope, land use types, and ecosystem types across different geographic units within the target area to establish a threshold system for grid optimization. This step transforms geographic information into actionable and evaluable quantitative data, providing a scientific basis for subsequent grid optimization and evaluation. Terrain slope change is a key indicator for assessing regional suitability, particularly in land use planning and ecological and environmental protection, where changes in slope directly impact land developability and ecosystem stability. Remote sensing imagery and digital elevation model (DEM) data allow for precise calculation of terrain characteristics such as slope and elevation for each grid unit, and further analysis of the spatial distribution of slope. Land use type analysis helps understand the distribution of resources within the region, such as residential areas, commercial areas, farmland, and forests. This data not only reflects the spatial distribution of human activities but also reveals the carrying capacity of the ecological environment. Combined with land use data, different land types can be assigned different assessment weights to reflect their importance and specific needs in planning. Ecosystem type analysis focuses on the health of the regional ecological environment, such as vegetation cover and water system distribution. Ecological indices and related models can be used to quantify these characteristics and provide a basis for subsequent planning decisions. Through this analysis, we can establish an optimized threshold system for each grid cell based on spatial characteristics and attribute similarity. This threshold system is used for subsequent grid optimization, dynamically adjusting the grid size and shape to ensure it matches the region's actual geographic characteristics and planning requirements.

[0086] Based on the spatial distribution characteristics and attribute similarity of geographic features, coarse-grained merging of initial grids is performed to form basic assessment units. The goal of coarse-grained merging is to reduce the number of assessment units by rationally merging grid cells with similar geographic features and attributes, thereby reducing computational complexity while maintaining analytical accuracy and regional representativeness. Merging is typically based on spatial distribution characteristics and attribute similarity. For example, adjacent grid cells with the same land use type, similar ecological and environmental characteristics, or similar terrain slopes can be merged into a single basic assessment unit. Coarse-grained merging not only improves computational efficiency but also further enhances the stability of assessment results. During this process, special care must be taken to avoid loss of spatial heterogeneity due to merging, ensuring that the merged assessment units still accurately reflect the spatial distribution and attribute characteristics of the region. These merged assessment units serve as the basis for subsequent refined analysis, simplifying the model structure while ensuring spatial representativeness, thereby improving the operability and accuracy of the assessment.

[0087] Perform spatial topological relationship testing between grid cells, identify and eliminate holes or overlapping areas, correct abnormal topological relationships, and generate an initial evaluation unit dataset. Topological relationship testing involves analyzing the connectivity, adjacency, overlap, etc. between grid cells to ensure that the merged evaluation units do not have problems such as holes, overlaps, or abnormal connections. For example, during the merging process, if adjacent grid cells have holes or overlaps due to data errors or improper merging strategies, these problems will affect the results of subsequent analysis and lead to inaccurate spatial distribution. Through topological relationship testing, these problems can be discovered in a timely manner, and anomalies can be eliminated by adjusting the grid boundaries or re-dividing the grid. In addition, topological relationship testing can also help discover potential connectivity problems between grid cells and ensure that the evaluation units have reasonable spatial connectivity. After correcting the abnormal topological relationships, the generated initial evaluation unit dataset will be more stable and accurate, providing reliable basic data for subsequent spatial analysis and decision support.

[0088] In summary, all sub-steps in Step 1 collectively build a foundation for accurate and efficient national spatial assessment. This process relies not only on the collection and preprocessing of multi-source data but also includes in-depth analysis and optimization of geographic spatial characteristics. Through techniques such as gridding, merging, and topological relationship verification, the accuracy and consistency of assessment units are continuously improved. The resulting assessment unit dataset truly reflects the spatial characteristics of the target area, providing solid support for subsequent spatial planning and decision-making. This entire process demonstrates the core role of geographic information technology in national spatial planning and provides the theoretical and technical foundation for more refined and intelligent spatial management.

[0089] Furthermore, the following formula is used to quantitatively analyze the changes in terrain slope, land use type and ecosystem type of different geographical units in the target area: G i =w1·ΔS i +w2·L i +w3·E i , where G i represents the comprehensive score of the i-th assessment unit, which is the result of weighted calculation of the three indicators: terrain slope, land use type and ecosystem type; w1 is the weight coefficient of terrain slope, which indicates the degree of influence of terrain slope on the comprehensive score; w2 is the weight coefficient of land use type, which indicates the degree of influence of land use type on the comprehensive score; w3 is the weight coefficient of ecosystem type, which indicates the degree of influence of ecosystem type on the comprehensive score; ΔS i It represents the change in terrain slope of the i-th evaluation unit and is expressed as: ΔS i =(1 / n)·∑ r＝1 n ∣S i,r -S i,r-1 |, where n represents the number of slope data points in each evaluation unit; S i,r represents the slope value of the rth position in the i-th evaluation unit; S i,r-1 represents the slope value of the r-1th position in the i-th evaluation unit, which is used to calculate the slope change in the i-th evaluation unit; L i It represents the land use score of the i-th assessment unit and is expressed as follows: L i =∑ g＝1 G W LU,g ·f i,g , where G represents the number of land use types involved in each assessment unit; W LU,g represents the weight coefficient of the g-th land use type; f i,g E represents the area proportion of the g-th land use type in the i-th assessment unit; i It represents the ecosystem score of the i-th assessment unit and is expressed as: E i =∑ s＝1 m W E,s ·f i,s , where m represents the total number of ecosystem types used to calculate the ecosystem score; W E,s represents the weight of the sth ecosystem type in the weighted calculation; f i,s表示 The area proportion of the sth ecosystem type in the ith assessment unit.

[0090] In summary, the above method can quantitatively analyze terrain slope changes, land use types, and ecosystem types in different geographic units within the target area, providing accurate assessment results for national land space planning. This comprehensive scoring method based on weighted calculation not only considers multi-dimensional spatial characteristics, but also can adapt to different planning needs by adjusting weight coefficients, thereby providing flexible and customized decision support in practical applications. With the development of technology, this method is expected to be further combined with remote sensing technology, geographic information systems (GIS), and artificial intelligence to achieve more accurate and intelligent national land space assessment, laying a more solid foundation for sustainable development and scientific decision-making.

[0091] Furthermore, the spatial distribution characteristics and attribute similarity based on geographic elements are expressed as follows: S total (G i ,G j )=αS space (G i ,G j )+(1-α)S attr (G i ,G j ), where S total (G i ,G j ) represents the grid cell G i and G j is the comprehensive similarity measure between them; α is the weight coefficient between spatial similarity and attribute similarity; S space (G i ,G j ) represents the grid cell G i and G j The spatial similarity between them is expressed as: S space (G i ,G j )=1 / [1+d(G i ,G j )], where d(G i ,G j ) is the grid cell G i and G j The spatial distance between attr (G i ,G j ) represents the grid cell G i and G j The attribute similarity between them is expressed as: attr (G i ,G j )=1 / [1+∑ k＝1 N w k (a i,k-a j,k ) 2 ], where N is the number of different attributes considered when calculating attribute similarity; w k is the weight of attribute k, indicating the importance of the attribute in the total similarity calculation; a i,k Represents the grid cell G i The value of attribute k; a j,k Represents the grid cell G j The value of attribute k.

[0092] In summary, combining spatial and attribute similarity to create a comprehensive similarity metric using a weighted fusion approach can provide a more accurate and comprehensive assessment basis for national spatial planning. Spatial similarity helps planners identify spatial connections by considering the relative positions and distances between geographic units; attribute similarity focuses on analyzing the internal attribute characteristics of each unit, reflecting the inherent differences between different regions. The comprehensive similarity metric further integrates this information from both perspectives, providing scientific decision-making support for regional demarcation, resource allocation, and land development.

[0093] Further, step 2 includes the following steps:

[0094] A first-level indicator based on land use structure characteristics is established, including land use type composition ratio, land use mix, and construction land density. Spatial form indicators and ecological environment indicators are also constructed to form a multi-dimensional comprehensive evaluation system. The land use type composition ratio measures the proportion of different land types within the total area, helping to analyze the distribution of different land use types within a region. This indicator can reveal the allocation and optimization of land resources. For example, the rational allocation of residential, commercial, and industrial land can support more efficient urban function operation and resource utilization. Land use mix refers to the degree of spatial interweaving of different land types. A high degree of mix generally indicates efficient and multifunctional land use. This indicator helps analyze the comprehensive development of a city, especially in urban renewal and spatial complex utilization, reflecting the diversity and vitality of land use. Construction land density assesses the compactness of construction land within a unit area and directly affects the city's carrying capacity, environmental quality, and infrastructure construction needs. High construction density can lead to urban congestion and environmental pollution, while a reasonable density can meet economic needs while maintaining a good living environment. Spatial form indicators reflect the spatial morphological characteristics of a city or region through analysis of the spatial layout within the region. These characteristics include the direction of urban expansion, the rationality of functional layout, and spatial accessibility. In the evaluation of spatial form, highly compact cities generally utilize land resources more efficiently, while a reasonable distribution of green space can improve residents' quality of life and ecological well-being. Ecological and environmental indicators focus on assessing a region's ecological carrying capacity and environmental protection level. By analyzing the distribution of ecological functional zones, vegetation cover, and biodiversity, it is possible to identify regional ecological protection needs and assess the sustainability of the ecological environment. For example, high vegetation cover generally indicates better air quality and soil and water conservation capacity, while diverse biomes can enhance ecosystem stability and disaster resilience.

[0095] Analyze how each indicator value varies with the scale of the assessment unit, determine the optimal evaluation scale for each indicator, and identify and quantify the corresponding sensitivity parameters. In actual spatial planning, assessment units at different scales exhibit different characteristics and patterns. As scale changes, the sensitivity of each indicator also changes. This means that some indicators will be more significant at large scales and become insignificant at small scales. Therefore, analyzing how indicator values vary with the scale of the assessment unit allows us to determine the optimal evaluation scale for each indicator and, subsequently, quantify the sensitivity parameters, ensuring that the assessment of each indicator is appropriately reflected in multi-scale analysis. For example, in city-level planning, construction land density is more sensitive because density is directly related to the city's carrying capacity. In nature reserve planning, ecological and environmental indicators are more important evaluation dimensions. By analyzing how these indicators vary with scale, we can adjust the assessment focus in a targeted manner, improving the scientific nature and accuracy of planning.

[0096] Based on regional development characteristics, planning standards, and ecological and environmental carrying capacity, thresholds for each indicator should be set. Setting thresholds for each indicator is a key step in establishing an effective indicator system. By integrating regional development characteristics, planning standards, and ecological and environmental carrying capacity, appropriate thresholds for each indicator can be clearly defined to meet the target at different levels, providing a quantitative basis for regional development planning. For example, thresholds for construction land density can be adjusted based on the city's functional requirements and land carrying capacity. Excessive density can lead to traffic congestion and environmental degradation, so it is important to set appropriate upper limits for density in specific areas. Thresholds for ecological and environmental indicators should be based on the carrying capacity of natural resources. Ecological quality standards for protected areas should be commensurate with the health and sustainable development capacity of the ecosystem. Setting these thresholds not only standardizes the implementation of spatial planning but also helps decision-makers make informed decisions during the planning process, ensuring that regional development meets the requirements of sustainable development.

[0097] Indicator correlation analysis is conducted at different spatial scales to identify interactions between evaluation indicators and quantitatively analyze the degree of correlation between them. Different spatial scales can cause variations in the correlation between evaluation indicators. In particular, at large scales, multiple indicators show stronger connections, while at small scales, these connections are weaker. Therefore, indicator correlation analysis is particularly important in this process. By identifying interactions between different indicators, we can further understand how these indicators influence evaluation results and quantitatively analyze the degree of correlation between them. For example, at large scales, construction land density and land use type composition are typically strongly correlated because land use type and density often influence each other. However, at small scales, regional topography, traffic conditions, and other factors can significantly influence this relationship. Therefore, multi-scale correlation analysis can help identify key indicators for focus in different contexts, thereby optimizing planning and decision-making.

[0098] Based on the results of the correlation analysis, highly correlated indicators are eliminated or merged, and independent evaluation indicators are supplemented to form the final indicator system. After completing the correlation analysis, eliminating or merging highly correlated indicators is an important step in optimizing the indicator system. Too many repeated or highly correlated indicators will complicate the evaluation system, reduce calculation efficiency, and lead to inaccurate results. Eliminating unnecessary related indicators and merging indicators with similar meanings can improve the simplicity and practicality of the model. At the same time, supplementing independent evaluation indicators can enhance the comprehensiveness of the evaluation system, especially in areas such as environmental protection and resource management. Adding independent ecological assessments or socioeconomic indicators can make the evaluation results more scientific and comprehensive.

[0099] In summary, by establishing a comprehensive evaluation system based on multiple dimensions, including land use structure, spatial form, and ecological environment, combined with correlation analysis, threshold setting, and indicator optimization at different scales, Step 2 provides scientific indicator support and data foundation for spatial planning. By flexibly applying these indicators, spatial planning solutions can be quantified and optimized based on specific regional characteristics and development needs, thereby achieving efficient resource utilization and sustainable development within the region. This step not only enhances the scientific nature and accuracy of the planning process but also provides decision makers with a more intuitive and systematic basis for decision-making, promising broad application prospects.

[0100] Furthermore, the optimal evaluation scale is determined for each indicator through the following formula: S * =arg min S [∑ p＝1 M (α p ·SSI p (S)+β p VR p (S)+γp CV p (S))], where S * is the optimal evaluation scale finally selected; S represents the scale of the evaluation unit; M is the total number of indicators, that is, the number of all indicators considered in the evaluation; α p is the weight coefficient of index p at different scales; β p is the weight coefficient of the rate of change of indicator p; γ p is the coefficient of variation weight coefficient of index p; SSI p (S) is the sensitivity of the scale sensitivity index calculation index p under scale S, which is expressed as: SSI p (S)=|I p (S2)-I p (S1)| / [|I p (S1)|+|I p (S2)|], where I p (S2) is the value of index p under scale S2; I p (S1) is the value of index p under scale S1; VR p (S) is the rate of change of index p under scale S, which is expressed as: VR p (S)=[I p (S2)-I p (S1)] / |S2-S1|;CV p (S) is the coefficient of variation of the index p under the scale S, which is expressed as: CV p (S)=σ p (S) / μ p (S), where σ p (S) is the standard deviation of the index p at scale S; μ p (S) is the mean of the indicator p under scale S.

[0101] In summary, determining the optimal evaluation scale for each indicator is a complex and meticulous process, involving comprehensive considerations of multiple aspects. By weighted optimization of parameters such as scale sensitivity, rate of change, and coefficient of variation, the most appropriate scale can be effectively selected for spatial planning assessment. This approach ensures that different indicators provide optimal evaluation results at the appropriate scale, avoiding evaluation errors caused by inappropriate scale selection.

[0102] Furthermore, the correlation analysis of indicators is carried out at different spatial scales to identify the interaction relationship between the evaluation indicators and quantitatively analyze the degree of correlation between the indicators, including the following steps:

[0103] Furthermore, the correlation analysis of indicators is conducted at different spatial scales to identify the interaction relationship between the evaluation indicators and quantitatively analyze the degree of association between the indicators, including the following steps;

[0104] Construct a multi-level spatial unit system, determine the analysis range of each scale, and standardize the evaluation indicators at each scale to eliminate the dimensionality effect;

[0105] Calculate the correlation coefficient matrix between evaluation indicators at each spatial scale level, and preliminarily judge the direction and strength of the relationship between indicators through the positive and negative and numerical values of the correlation coefficients;

[0106] By controlling other variables, we can identify the direct correlation between any two evaluation indicators and eliminate the interference effect of the third variable;

[0107] Construct the grey correlation matrix between indicators and quantify the dynamic correlation characteristics between evaluation indicators;

[0108] Based on the set correlation threshold, screen the pairs of indicators with significant correlation and grade the correlation strength;

[0109] Compare the differences in indicator correlation characteristics at different spatial scales, summarize the scale-dependent laws of indicator relationships, and identify indicator correlations with scale sensitivity;

[0110] Based on the relationship matrix, the interaction network structure between evaluation indicators is visualized, key node indicators and important association paths are identified and labeled, and the overall association characteristics are revealed.

[0111] In summary, analyzing indicator correlations at different spatial scales can provide a deeper understanding of the interactions between various indicators in spatial planning and their variations at different scales. By constructing a multi-level spatial unit system, standardizing it, calculating correlation coefficients, and constructing a gray correlation matrix, we can provide a more accurate basis for decision-making in national land spatial planning. Screening for significantly correlated indicator pairs and grading their strength helps planners identify key indicators and optimize resource allocation. The application of visualization technology can make complex indicator relationship networks more intuitive, further improving the transparency and efficiency of the decision-making process.

[0112] Further, step 3 includes the following steps:

[0113] The coefficient of variation (CV) of indicators for each assessment unit in three dimensions: land use structure, spatial form, and ecological environment is quantitatively calculated. In spatial planning and assessment, the CV of an assessment unit is a core indicator for measuring the degree of variation across multiple dimensions, including land use structure, spatial form, and ecological environment. As a standardized measure of variation, the CV eliminates the influence of units and dimensions on the results, converting data from different dimensions and scales into a relative degree of variation under a unified standard, thereby achieving greater comparability in comprehensive assessments. This process accurately measures the performance of each assessment unit (such as a plot or region) in a specific dimension, quantifying its spatial heterogeneity and providing a quantitative basis for subsequent analysis. By quantitatively calculating the CV, the diversity and differences among assessment units across different dimensions can be effectively measured. For example, in terms of land use structure, the CV can reveal differences in land use types between plots; in terms of spatial form, the CV helps analyze the uneven distribution of elements such as buildings and roads within a region; and in terms of ecological environment, the CV reflects the spatial heterogeneity of natural environmental factors (such as vegetation cover and wetlands). The calculated coefficient of variation provides quantitative support for subsequent classification, clustering, and optimization, ensuring that the evaluation system can accurately capture the potential spatial heterogeneity and environmental differences within the region, thereby providing an important basis for the design of subsequent spatial planning schemes.

[0114] Estimating the spatial heterogeneity of elements within a unit involves calculating the landscape diversity index, topographic heterogeneity, land use heterogeneity, and the spatial distribution heterogeneity of ecosystem services. Spatial heterogeneity refers to the uneven distribution of different elements (such as landscape, topography, and land use types) within a spatial unit. Estimating the spatial heterogeneity of elements within a unit helps reveal the complexity of ecosystems, landscapes, or land use patterns within a region. By calculating the landscape diversity index, topographic heterogeneity, land use heterogeneity, and the spatial distribution heterogeneity of ecosystem services, we can gain a differentiated understanding of the spatial distribution of different elements within a unit. The landscape diversity index is often used to quantify the uniformity of the distribution of different landscape types within a region. Topographic heterogeneity is related to geographical factors such as surface elevation and slope. Land use heterogeneity focuses on the distribution of different land uses within a region. The spatial distribution heterogeneity of ecosystem services reflects the diversity and distribution patterns of natural resources. By estimating spatial heterogeneity, we can identify the distribution patterns and spatial differences of elements within a unit, providing a multi-dimensional reference for regional development planning. For example, the landscape diversity index can help analyze the distribution differences between natural and artificial landscapes within a region, enabling precise design of regional landscape planning. Topographic heterogeneity can help assess the topographic characteristics of different regions and optimize land development plans. Land use heterogeneity analysis can provide a basis for decision-making on land use adjustment and planning during urbanization. And analysis of the spatial distribution heterogeneity of ecosystem service functions provides important guidance for projects such as nature reserves and ecological restoration. By combining these heterogeneity indicators, planners can more scientifically identify and analyze potential spatial problems, such as overdeveloped areas and ecologically degraded areas, thereby better formulating resource allocation and environmental protection strategies.

[0115] Proximity analysis is used to calculate the degree of indicator dissimilarity between adjacent assessment units. In spatial analysis, proximity analysis is used to assess the relationships between different assessment units, particularly the spatial dissimilarity between adjacent units. Calculating the degree of indicator dissimilarity between adjacent assessment units can help identify spatial heterogeneity and its potential impact. For example, by calculating the degree of dissimilarity between adjacent units in land use type, ecological and environmental quality, and land use structure, it is possible to determine whether adjacent units are experiencing issues such as overdevelopment and environmental degradation, thereby providing a basis for regional development strategies. Calculating the degree of dissimilarity between adjacent assessment units can help planners identify and understand the connections and distinctions between different parts of a region. Highly dissimilar units indicate significant differences in ecological environment, land use, or socioeconomic activities, which can impact regional coordination and sustainable development. This step quantifies the differences between adjacent units, further providing a basis for regional integration, consolidation, and optimization. For example, in ecological and environmental protection, if two adjacent regions differ significantly, different protection measures may be required, or more focused support may be given in terms of ecological compensation. This analysis provides quantitative data support for coordinated regional development and facilitates the formulation of more refined management policies.

[0116] Set the internal heterogeneity threshold α 内部 and the inter-unit difference threshold β, where: when the internal heterogeneity of the evaluation unit is greater than α 内部 When the internal heterogeneity threshold and inter-unit heterogeneity threshold are less than β, they are merged into larger units. The goal of setting internal heterogeneity thresholds and inter-unit heterogeneity thresholds is to optimize the division and merging strategies of assessment units by defining reasonable criteria. In spatial planning, when the internal heterogeneity of an assessment unit exceeds the set threshold, it indicates that the spatial structure of the area is too complex to provide effective management and decision support. In this case, it should be considered to be subdivided into smaller units. Conversely, when the heterogeneity of adjacent units is less than the set threshold, the differences between the two units are small, and merging them into a larger unit can be considered. By setting these thresholds, units can be divided and merged more finely, making the assessment system more scientific and reasonable. Setting heterogeneity thresholds and inter-unit heterogeneity thresholds provides a quantitative basis for the division and merging of assessment units, avoids bias in human decision-making, and enhances the objectivity and scientific nature of the assessment. Setting these thresholds allows for flexible treatment of different types of spatial units, improving the accuracy and operability of assessment results. Setting thresholds not only helps optimize the size of spatial units, but also helps adjust the composition of units, making subsequent spatial planning more refined. For example, in an ecological and environmental protection area, if the ecological conditions within the area are complex and highly variable, it is necessary to subdivide it into multiple sub-areas and adopt different protection strategies. For adjacent areas with smaller differences, merging them can be considered to improve the efficiency of resource allocation.

[0117] Through iterative optimization, a basic assessment unit system that meets the assessment requirements is formed. This step requires repeated division and merging of units based on set thresholds until the assessment system fully reflects the spatial characteristics and development needs of the region. Iterative optimization not only helps improve assessment accuracy but also ensures the adaptability and flexibility of the assessment unit system. Through iterative optimization, the final assessment unit system fully accounts for the various heterogeneous factors within the region and forms a highly adaptable and operational spatial division scheme. This process helps gradually eliminate unreasonable unit divisions and ensures that the planning scheme is highly practical and scientific while maintaining spatial rationality. Through repeated adjustments, the final assessment unit system is able to meet the needs of spatial planning assessments at different scales and for different needs, helping to improve the accuracy and effectiveness of the decision-making process. For planning in areas such as resource allocation, land development, and environmental protection, the iteratively optimized unit system provides a more flexible and operational foundation, enhancing the adaptability and foresight of planning results.

[0118] In summary, by quantitatively calculating the coefficient of variation of indicators across multiple dimensions for assessment units, estimating spatial heterogeneity, and analyzing the differences between adjacent units, a refined and flexible system of basic assessment units was established through threshold setting and iterative optimization. This series of analysis and optimization processes not only improves the scientific nature and accuracy of spatial planning and assessment, but also enhances the precision and adaptability of decision-making. These technical approaches enable comprehensive analysis of complex spatial environments, providing strong data support and decision-making basis for regional sustainable development.

[0119] Furthermore, the coefficient of variation of indicators in the three dimensions of land use structure, spatial form and ecological environment of each assessment unit is quantitatively calculated by the following formula: CV total,i =w land CV land,i +w space CV space,i +w eco CV Veco,i , where CV total,i represents the comprehensive variation coefficient of the assessment unit i in the three dimensions of land use structure, spatial form and ecological environment; w land is the weight coefficient of the land use structure dimension, which indicates the relative importance of the land use structure dimension in the comprehensive evaluation; CV land,i is the coefficient of variation of the assessment unit i in the land use structure dimension, which is expressed as: CV land,i =σ land,i / μ land,i , σ land,i is the standard deviation of the land use structure dimension of assessment unit i; μ land,iis the mean value of the land use structure dimension of assessment unit i; w space is the weight coefficient of the spatial morphology dimension, indicating the importance of the spatial morphology dimension in the comprehensive evaluation; CV space,i is the coefficient of variation of the evaluation unit i in the spatial morphological dimension, which is expressed as: CV space,i =σ space,i / μ space,i , σ space,i is the standard deviation of the evaluation unit i in the spatial morphological dimension; μ space,i is the mean value of the evaluation unit i in the spatial morphological dimension; w eco is the weight coefficient of the ecological environment dimension, indicating the importance of the ecological environment dimension in the comprehensive assessment; CV Veco,i It is the coefficient of variation of the evaluation unit i in the ecological environment dimension, which measures the heterogeneity or dispersion of the unit in the ecological environment. It is expressed as: CV Veco,i =σ Veco,i / μ Veco,i , σ Veco,i is the standard deviation of the evaluation unit i in the ecological environment dimension; μ Veco,i is the mean value of the evaluation unit i in the ecological environment dimension;

[0120] The landscape diversity index is expressed as follows: H′=-∑ e＝1 E p e ln(p e ), where H′ is the landscape diversity index, indicating the diversity of the landscape; p e is the proportion of the e-th landscape type;

[0121] The terrain heterogeneity is expressed as: TRI = (1 / F)·∑ f＝1 F ∣Z f -Z f-1 ∣, where TRI is the terrain roughness index, which is used to analyze the heterogeneity of the terrain; F is the number of pixels in the grid cell; Z f is the elevation of the f-th pixel; Z f-1 is the elevation of the f-1th pixel;

[0122] The land use heterogeneity is expressed as follows: I = (U / W)·[(∑ i ∑ j w ij ·(x i -xˉ)·(x j -xˉ)) / (∑ i (x i -xˉ) 2)], where I is the land use heterogeneity; U is the number of grid cells; W is the sum of the weight matrices; w ij is the spatial weight between the i-th and j-th units; x i is the land use type of unit i; x j is the land use type of unit j; xˉ is the average value of all units;

[0123] The spatial distribution heterogeneity of ecosystem service functions can be expressed as follows: ESHI = (1 / U)·∑ i＝1 U (f i / ∑ j＝1 U f j )·(1 / ∑ q＝1 U d iq ), where ESHI is the ecosystem service heterogeneity index; f i is the ecological service function value of unit i; d iq is the spatial distance between the i-th unit and other units q; f j is the ecological service function value of the jth unit;

[0124] The indicator difference formula between adjacent evaluation units is expressed as: D ij =[1 / (1+exp(-δ·Moran(i,j))]·√∑ p＝1 M W p [(X ip -X jp ) / (max(X p )-min(X p )] 2 , where D ij is the difference between the evaluation unit i and the evaluation unit j; W p is the weight of the pth indicator; δ is a control parameter used to adjust the intensity of the influence on the difference between adjacent units; X ip is the value of evaluation unit i under the pth indicator; X jp is the value of evaluation unit j under the pth indicator; max(X p ) is the maximum value of the p-th index; min(X p ) is the minimum value of the pth indicator; Moran(i,j) is the spatial autocorrelation index between evaluation units i and j.

[0125] In summary, quantitatively calculating the coefficient of variation across the three dimensions of land use structure, spatial form, and ecological environment within each assessment unit effectively measures spatial heterogeneity. Indicators such as landscape diversity, topographic heterogeneity, and land use heterogeneity can more accurately capture the distribution patterns of natural and human factors within a region. The spatial heterogeneity of ecosystem service functions and the degree of variability between adjacent units further provide more detailed and targeted decision-making support for regional planning. By applying these comprehensive technical approaches, spatial planning can more scientifically and precisely address regional heterogeneity, promoting the rational allocation of resources and sustainable environmental management.

[0126] Further, step 4 includes the following steps:

[0127] A judgment matrix is established using expert scoring. Each indicator is compared pairwise, and eigenvectors are calculated and then tested for consistency. This determines the relative importance of each indicator at each scale level. This step leverages the experience and knowledge of experts to subjectively score complex issues, thereby quantifying the relative weights of each indicator. By calculating the eigenvectors of the judgment matrix, weights can be assigned to each indicator within each evaluation scale, ensuring that the final evaluation results fully reflect the importance of each indicator. Consistency testing is an essential step in this process, as it verifies the consistency of expert ratings and ensures the reliability and stability of the evaluation system.

[0128] A scale conversion matrix is constructed to convert evaluation indicators and results between different spatial scales, ensuring comparability and consistency of evaluation results across scales. This ensures that evaluation results can be effectively compared and integrated across scales. The evaluation objects and indicators have different meanings at different spatial scales. For example, the macroscale focuses on overall development trends, while the microscale focuses on details. The scale conversion matrix eliminates these differences through standardization and normalization, making evaluation results at different scales consistent and comparable, thereby enhancing the overall integrity and coordination of the results. This step effectively addresses the scale inconsistency commonly encountered in multi-scale assessments and provides a foundation for subsequent comprehensive assessment and multi-scale integration.

[0129] Conduct multi-scale spatial agglomeration analysis to analyze the spatial distribution patterns of evaluation indicators at different scales. This analysis can reveal the clustering and distribution patterns of spatial elements at different scales, providing critical spatial characteristic information for spatial decision-making. By analyzing spatial distribution patterns at different scales, potential hotspots, marginal areas, and spatial gaps within a region can be identified, providing policymakers with detailed spatial distribution characteristic data. This analysis method is highly flexible and operational, allowing for adjustment and optimization based on different evaluation objectives. It has broad application prospects in various fields, including regional development, resource allocation, and environmental protection. This analysis method can achieve optimal allocation of spatial resources and improve the scientific nature and operability of spatial planning.

[0130] By comprehensively considering indicator weights, scale conversion relationships, and spatial autocorrelation characteristics, a collaborative computation model for multi-scale assessment results is established, achieving the organic integration of assessment results at different scales. This collaborative computation model effectively integrates multi-scale assessment results, thereby providing a global spatial assessment conclusion. By considering indicator weights, scale conversion relationships, and spatial autocorrelation characteristics, the model is able to complement and optimize information across different spatial scales. The incorporation of spatial autocorrelation characteristics enables the assessment results to more accurately reflect the interactive relationships between spatial elements, avoiding the localized and one-sided nature that can arise from single-scale assessments. This model not only improves the accuracy and credibility of assessment results but also provides an effective technical means for cross-scale information transfer in multi-scale assessments, making spatial planning and decision-making more responsive to practical needs.

[0131] Consistency testing and uncertainty analysis are performed on the multi-scale assessment results to verify their reliability and stability at different scales, and key influencing factors are identified through sensitivity analysis. The introduction of consistency testing and uncertainty analysis helps to improve the credibility and reliability of multi-scale assessment results. Consistency testing can detect whether there are large deviations between assessment results at different scales, ensuring that the assessment conclusions are highly coordinated and consistent. Uncertainty analysis can reveal possible errors and uncertainties in the assessment process, help decision makers identify potential risks, and provide data support for future optimization. The application of these two technical links can provide effective guarantees for multi-scale assessment results, making the assessment conclusions more stable and reliable, thereby providing a solid basis for policy formulation and spatial planning.

[0132] In summary, the implementation of the above technical steps enables the construction of an efficient multiscale spatial assessment system, ensuring the comparability, consistency, and reliability of assessment results at different scales. The expert scoring method and scale conversion matrix provide quantitative support and a standardized framework for the assessment results. Multiscale spatial clustering feature analysis reveals spatial distribution patterns, and the collaborative computational model integrates the assessment results at each scale. Finally, consistency checks and uncertainty analysis further enhance the credibility and stability of the assessment results. Overall, this assessment system provides a multidimensional and multi-level analytical framework for regional planning and decision-making, helping decision-makers to scientifically understand spatial characteristics and trends and formulate more reasonable and precise planning strategies.

[0133] Further, step 5 includes the following steps:

[0134] Taking into account data accuracy, assessment objectives and actual application needs, error tolerance standards are formulated, including setting acceptable error ranges and confidence levels. By reasonably setting error tolerance standards, the permissible errors of assessment results can be effectively controlled according to actual application needs and data accuracy requirements. This process helps to ensure the scientific nature and practicality of the assessment results, and can also improve the flexibility of the model in practical applications without losing model accuracy. For example, in ecological and environmental management, if the error tolerance is too large, it will lead to wrong decisions and affect the effectiveness of ecological protection measures; in urban planning, a higher error tolerance allows for a wider planning space. Therefore, when setting error tolerance standards, it is necessary to combine the actual application scenarios and consider the accuracy and feasibility of data collection to ensure that the assessment results can provide a basis for decision makers within a reasonable range.

[0135] Using the k-fold cross-validation method, repeated validation is performed, calculating the error in the evaluation results for different combinations of evaluation units and indicators. Any evaluation units or indicators with errors exceeding the tolerance are recorded, and a comprehensive validation record database is established. Using k-fold cross-validation can significantly improve the robustness and generalization of the evaluation model, preventing the model from being limited to specific datasets. Through repeated validation, k-fold cross-validation not only helps identify which evaluation units and indicator combinations contribute most to the error but also helps identify areas of instability during the evaluation process, thus providing a basis for optimization. Recording any evaluation units and indicators with errors exceeding the tolerance helps establish a comprehensive validation database, providing detailed data support for subsequent model adjustments and optimization. This process enhances the transparency and credibility of the evaluation model, providing a more reliable basis for decision-making.

[0136] Systematic analysis of sources of assessment error, including the rationality of assessment unit division, the completeness of the indicator system, and the accuracy of scale conversion, can effectively identify potential weaknesses in the assessment process and provide practical directions for optimizing the model. Analyzing the rationality of assessment unit division can help optimize spatial partitioning and ensure that spatial heterogeneity is fully accounted for. Analyzing the completeness of the indicator system can identify any important omissions and supplement them, thereby enhancing the comprehensiveness of the assessment. Analyzing the accuracy of scale conversion can help adjust the conversion method between scales to ensure consistency and comparability of results across scales. These analyses can significantly improve model accuracy, ensure the reliability and stability of assessment results, and provide a scientific basis for decision-making.

[0137] Based on the cross-validation results, the evaluation parameters are optimized, including adjusting indicator weight coefficients and revising scaling parameters. Through multiple rounds of iterative optimization, the evaluation error is continuously reduced. This iterative optimization not only continuously adjusts parameters based on the validation results, minimizing the model's evaluation error in different situations, but also helps the model automatically identify and correct its deficiencies in certain scenarios. Through this adaptive optimization approach, the model's accuracy and stability are continuously improved, ultimately forming an efficient and accurate evaluation system. The optimized evaluation model can better adapt to various scenarios in real-world applications, improving the scientific nature and practicality of the evaluation. Furthermore, this process significantly reduces errors caused by human intervention and subjective judgment, making the evaluation results more objective and reliable.

[0138] The optimized assessment results are fully verified to ensure their accuracy, spatial continuity, and degree of alignment with regional characteristics, thereby forming a final optimization plan. This verification step is extremely practical, ensuring that the optimized assessment results effectively reflect the region's spatial characteristics and actual needs in practical applications. Accuracy verification ensures the reliability of the assessment results, spatial continuity verification ensures their operability, and verification of the degree of alignment with regional characteristics enhances their practical relevance. Through comprehensive verification, the quality of the assessment model can be maximized, ensuring that the final assessment plan can provide decision makers with a practical basis and scientific support for subsequent spatial planning and resource allocation.

[0139] In summary, by comprehensively considering data accuracy, assessment objectives, and practical application needs, establishing error tolerance standards, employing cross-validation for repeated verification, systematically analyzing and assessing error sources, optimizing assessment parameters, and conducting comprehensive verification, a rigorous validation and optimization framework for multi-scale spatial assessment was established. This process not only improved the accuracy, reliability, and stability of the assessment model but also ensured that the assessment results reflected the spatial characteristics and actual needs of the region. Through multiple rounds of iterative optimization and continuous adjustment, the resulting assessment scheme possessed high accuracy and broad applicability, providing scientific support for regional planning and decision-making.

[0140] Furthermore, the k-fold cross-validation method is used to perform repeated validations, calculate the errors of the evaluation results under different evaluation units and indicator combinations, record the evaluation units and indicators whose errors exceed the tolerance, and establish a complete validation record database, including the following steps.

[0141] The initial evaluation unit datasets were grouped and organized according to the evaluation units. The k value for k-fold cross-validation was determined, and the error tolerance thresholds for the evaluation indicators were set. Data grouping and k-value setting ensured the representativeness and consistency of each dataset during the evaluation process, helping to mitigate the impact of sample imbalance or data bias on the results. Proper tolerance threshold settings not only controlled model error and enhanced the credibility of the evaluation results, but also provided a clear reference standard for subsequent analysis. In this way, the evaluation process maintained high computational efficiency and applicability for practical applications while ensuring model accuracy.

[0142] The data from each evaluation unit is randomly divided into k subsets of similar size. One of these subsets is used as the validation set, and the remaining k-1 subsets are used as the training set. This random partitioning of the dataset and the use of k-fold cross-validation effectively avoids bias caused by fixed partitioning and improves the model's generalization and stability. Each round of independent validation helps comprehensively evaluate the model's performance under different data conditions, thereby avoiding overfitting in practical applications. More importantly, this method can effectively detect and reduce evaluation errors caused by uneven data partitioning, ensuring more objective and reliable model evaluation results.

[0143] For each validation round, the model is trained on the training set and then predicted on the validation set, calculating the values of all relevant evaluation metrics. Calculating evaluation metrics in each validation round allows for in-depth analysis of model performance, revealing differences in performance under different data conditions. By calculating evaluation metrics one by one, it becomes clearer where the model performs well and where optimization is needed. Furthermore, focusing on multiple evaluation metrics allows for a more comprehensive assessment of model performance, reflecting both the strengths and weaknesses of the model and providing a scientific basis for developing optimization plans.

[0144] The results of each evaluation unit for each metric across k validation runs are summarized, and the mean and standard deviation are calculated. The actual error is compared to the preset tolerances, and any cases exceeding the tolerances are marked. By summarizing and analyzing the validation results, it is possible to effectively identify differences in the model's performance across different evaluation units and metrics, and promptly discover potential sources of error. Calculating the mean and standard deviation not only quantifies the accuracy and stability of the evaluation results but also provides a detailed basis for optimization. Marking evaluation units and metrics that exceed tolerances helps focus attention on key issues in model optimization, ensuring that subsequent improvements are more targeted and effective.

[0145] Completely document all verification details, including the evaluation unit number, indicator name, raw results from k verifications, statistical analysis results, and whether tolerances were exceeded, to form a structured verification record. This thorough documentation of the verification process ensures that all data from the evaluation model validation phase is accurately archived, providing a valuable reference for subsequent analysis. This process not only enhances the traceability and transparency of model evaluation but also helps identify and analyze anomalous results during the verification process, ensuring clear optimization directions and appropriate improvement measures.

[0146] Based on the validation record database, a summary report of the validation results is generated, highlighting anomalies and analyzing the causes of out-of-tolerance assessment units and indicators, providing a basis for subsequent optimization. This summary report not only provides a comprehensive review of the model evaluation but also provides a detailed basis for optimization and improvement. By analyzing the causes of anomalies, the report provides in-depth insights for decision makers, effectively guiding subsequent optimization processes and improving the overall accuracy and reliability of the model.

[0147] In summary, the k-fold cross-validation method enables rigorous control and analysis at every stage of data partitioning, model training, evaluation, and optimization. This process not only helps us identify potential model issues but also provides a clear direction for subsequent improvements. Through detailed validation records, metric calculations, and error analysis, we ensure the scientific and rigorous nature of the evaluation process and provide strong support for model optimization. Ultimately, this systematic and structured validation approach can significantly improve the reliability and accuracy of the model in various application scenarios.

[0148] Furthermore, step 6 includes the following steps:

[0149] Build a multi-dimensional data visualization platform that integrates the assessment results database and geographic information system to display assessment results at different scales. It also supports dynamic data updates and real-time rendering to ensure the accuracy and intuitiveness of visualization effects.

[0150] Clearly display the entire process of the evaluation unit from the initial grid to the final optimization result, including the changes in the unit boundary and the dynamic adjustment of the index value, to help decision makers understand the inherent logic of the evaluation unit optimization;

[0151] Use a variety of visual expressions to intuitively present the spatial distribution of evaluation indicators, and enhance the spatial distribution characteristics through color gradients and symbol changes;

[0152] Realize multi-window and data linkage to comprehensively display the spatial characteristics of assessment results in different dimensions, making it easier to analyze the internal relationships of assessment results from multiple perspectives;

[0153] It provides statistical analysis, spatial query and thematic chart generation functions, offering comprehensive support for national land space planning decision-making.

[0154] In summary, by building a multidimensional data visualization platform, combined with a geographic information system (GIS) and dynamic real-time rendering technology, the presentation of assessment results and decision-making support can be effectively enhanced. Multi-window, data linkage, and multi-dimensional visualization methods further enhance the multi-angle analysis capabilities of assessment results, enabling decision-makers to fully understand and grasp the assessment data. The introduction of statistical analysis, spatial query, and thematic chart generation capabilities provides more detailed data analysis and decision-making support, ensuring the scientific and accurate nature of national land space planning decisions. Ultimately, this series of technical means provides decision-makers with comprehensive, multi-dimensional data support, greatly improving the transparency of the assessment process and the scientific nature of decision-making.

[0155] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art will appreciate that modifications may be made to the technical solutions described in the above embodiments, or that some of the technical features may be replaced with equivalents; such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for assessing the status quo of national land space planning, characterized in that: The following steps are involved: Step 1: construct an initial assessment unit dataset adapted to geographical characteristics; Step 2: Construct a multi-level indicator system covering land use structure, spatial form, and ecological environment dimensions, set scale sensitivity parameters and threshold intervals for each indicator, and quantitatively analyze the correlation between indicators at different scales; Step 3: Calculate the dispersion and spatial heterogeneity of the internal indicators of the evaluation unit, combine the correlation strength between adjacent units, and realize the adaptive segmentation and merging of the evaluation unit to ensure that the evaluation unit matches the indicator characteristics; Step 4: Apply the analytic hierarchy process to determine the weights of indicators at different scales, and achieve collaborative calculation of multi-scale assessment results through scale conversion matrix and spatial autocorrelation analysis; Step 5: Carry out cross-validation based on the error tolerance standard, automatically trace and optimize key indicators and evaluation units, and optimize the evaluation results through parameter adjustment; Step 6: Use spatial information visualization technology to display multi-scale assessment results, the dynamic evolution process of assessment units, and the spatial distribution characteristics of various assessment indicators, providing an intuitive reference for national land space planning decisions.

2. A method for assessing the status quo of national land space planning according to claim 1, characterized in that: Step 1 includes the following steps: Collect and pre-process high-resolution remote sensing imagery data, integrate existing basic geographic data, including vector data of terrain, land use, and administrative boundaries, and unify data resolution and projection coordinate systems to ensure data consistency; According to the geographical characteristics and assessment requirements of the target area, an initial assessment grid is constructed to characterize the spatial distribution characteristics of geographical elements; Quantitatively analyze the changes in terrain slope, land use types, and ecosystem types in different geographical units within the target area, and establish a threshold system for grid optimization; Based on the spatial distribution characteristics and attribute similarity of geographic elements, a coarse-grained merging of the initial grid is performed to form a basic evaluation unit; Perform spatial topological relationship checks between grid cells, identify and eliminate voids or overlapping areas, correct abnormal topological relationships, and generate an initial evaluation unit dataset.

3. A method for assessing the status quo of national land space planning according to claim 1, characterized in that: Step 2 includes the following steps: Establish first-level indicators based on land use structure characteristics, including land use type composition ratio, land use mix and construction land density indicators, while also constructing spatial morphology indicators and ecological environment indicators to form a multi-dimensional comprehensive evaluation system; Analyze the variation patterns of each indicator value with the scale of the assessment unit, determine the optimal evaluation scale for each indicator, and identify and quantify the corresponding sensitivity parameters; Based on regional development characteristics, planning standards and ecological environmental carrying capacity, set grading thresholds for various indicators; Conduct indicator correlation analysis at different spatial scales to identify the interaction relationship between evaluation indicators and quantitatively analyze the degree of correlation between indicators; Based on the results of correlation analysis, highly correlated indicators are eliminated or merged, and independent evaluation indicators are supplemented to form the final indicator system.

4. A method for assessing the status quo of national land space planning according to claim 3, characterized in that: Conduct indicator correlation analysis at different spatial scales, identify the interaction relationship between evaluation indicators, and quantitatively analyze the degree of correlation between indicators, including the following steps: Construct a multi-level spatial unit system, determine the analysis range of each scale, and standardize the evaluation indicators at each scale to eliminate the dimensionality effect; Calculate the correlation coefficient matrix between evaluation indicators at each spatial scale level, and preliminarily judge the direction and strength of the relationship between indicators through the positive and negative and numerical values of the correlation coefficients; By controlling other variables, we can identify the direct correlation between any two evaluation indicators and eliminate the interference effect of the third variable; Construct the grey correlation matrix between indicators and quantify the dynamic correlation characteristics between evaluation indicators; Based on the set correlation threshold, screen the pairs of indicators with significant correlation and grade the correlation strength; Compare the differences in indicator correlation characteristics at different spatial scales, summarize the scale-dependent laws of indicator relationships, and identify indicator correlations with scale sensitivity; Based on the relationship matrix, the interaction network structure between evaluation indicators is visualized, key node indicators and important association paths are identified and labeled, and the overall association characteristics are revealed.

5. The method for assessing the current status of national land space planning according to claim 1, characterized in that: Step 3 includes the following steps: Quantitatively calculate the coefficient of variation of indicators in the three dimensions of land use structure, spatial form and ecological environment for each assessment unit; Estimate the spatial heterogeneity of elements within a unit, including calculating the landscape diversity index, topographic heterogeneity, land use heterogeneity, and spatial distribution heterogeneity of ecosystem service functions; Calculate the index difference between adjacent evaluation units based on the proximity analysis method; Set the internal heterogeneity threshold α 内部 and the inter-unit difference threshold β, where: when the internal heterogeneity of the evaluation unit is greater than α 内部 When , it is divided into smaller evaluation units; when the difference between adjacent evaluation units is less than β, they are merged into a larger evaluation unit; Through iterative optimization, a basic evaluation unit system that meets evaluation needs is formed.

6. A method for assessing the status quo of national land space planning according to claim 1, characterized in that: Step 4 includes the following steps: A judgment matrix was established through expert scoring, and each indicator was compared pairwise. The eigenvectors were calculated and consistency tests were performed to determine the relative importance weights of each indicator at each scale level. Construct a scale conversion matrix to achieve the conversion of evaluation indicators and assessment results between different spatial scales, ensuring the comparability and consistency of evaluation results between different scales; Conduct multi-scale spatial agglomeration feature analysis and analyze the spatial distribution patterns of evaluation indicators at different scales; Comprehensively consider the indicator weights, scale conversion relationships and spatial autocorrelation characteristics, establish a collaborative calculation model for multi-scale assessment results, and achieve the organic integration of assessment results at different scales; The consistency test and uncertainty analysis of the multi-scale assessment results are carried out to verify the reliability and stability of the assessment results at different scales, and the key influencing factors are identified through sensitivity analysis.

7. A method for assessing the status quo of national land space planning according to claim 1, characterized in that: Step 5 includes the following steps: Taking into account data accuracy, assessment objectives, and actual application needs, formulate error tolerance standards, including setting acceptable error ranges and confidence levels; Use the k-fold cross-validation method to perform repeated validation, calculate the error of the evaluation results under different evaluation units and indicator combinations, record the evaluation units and indicators whose errors exceed the tolerance, and establish a complete validation record database; Systematically analyze the sources of evaluation errors, including the rationality of the division of evaluation units, the completeness of the indicator system, and the accuracy of scale conversion; Based on the cross-validation results, the evaluation parameters are optimized, including adjusting the indicator weight coefficients and correcting the scale conversion parameters. Through multiple rounds of iterative optimization, the evaluation error is continuously reduced. The optimized assessment results are fully verified to ensure their accuracy, spatial continuity and matching degree with regional characteristics, so as to form the final optimization plan.

8. A method for assessing the status quo of national land space planning according to claim 7, characterized in that: The k-fold cross-validation method is used to perform repeated validation, calculate the evaluation result errors under different evaluation units and indicator combinations, record the evaluation units and indicators whose errors exceed the tolerance, and establish a complete validation record database, including the following steps: The initial evaluation unit data set is grouped and sorted according to the evaluation unit, the k value of the k-fold cross-validation is determined, and the error tolerance threshold of the evaluation indicator is set; The data of each evaluation unit is randomly divided into k subsets of similar size, one of which is used as the validation set, and the remaining k-1 subsets are used as the training set; For each round of validation, the model is trained using the training set and then predicted on the validation set, calculating the values of all the evaluation indicators of interest. Summarize the results of each indicator of each evaluation unit in k verifications, calculate the mean and standard deviation, compare the actual error with the preset tolerance, and mark the situation where the tolerance is exceeded; Record all verification process details, including the evaluation unit number, indicator name, original results of k verifications, statistical analysis results, and whether the tolerance is exceeded, to form a structured verification record; Based on the verification record database, a summary report of the verification results is generated, focusing on abnormal situations, and analyzing the causes of evaluation units and indicators that exceed the tolerance, providing a basis for subsequent optimization.

9. The method for assessing the current status of national land space planning according to claim 1, characterized in that: Step 6 includes the following steps: Build a multi-dimensional data visualization platform that integrates the assessment results database and geographic information system to display assessment results at different scales. It also supports dynamic data updates and real-time rendering to ensure the accuracy and intuitiveness of visualization effects. Clearly display the entire process of the evaluation unit from the initial grid to the final optimization result, including the changes in the unit boundary and the dynamic adjustment of the index value, to help decision makers understand the inherent logic of the evaluation unit optimization; Use a variety of visual expressions to intuitively present the spatial distribution of evaluation indicators, and enhance the spatial distribution characteristics through color gradients and symbol changes; Realize multi-window and data linkage to comprehensively display the spatial characteristics of assessment results in different dimensions, making it easier to analyze the internal relationships of assessment results from multiple perspectives; It provides statistical analysis, spatial query and thematic chart generation functions, offering comprehensive support for national land space planning decision-making.

Images