Loess plateau human settlement environment assessment algorithm based on big data

Abstract

The invention relates to the technical field of environment assessment and big data analysis, in particular to a loess plateau human settlement environment assessment algorithm based on big data, and the method comprises the steps: calculating four core indexes: soil erosion intensity, facility accessibility, terrain safety and social economic activeness; performing landform automatic partitioning based on topographic features; and then partitioning is carried out, dynamic weights are calculated by adopting a Credit-entropy weight combination method, and finally the human settlement environment quality index is synthesized. The system provides a corresponding hardware implementation architecture. According to the method, the loess landform differentiation rule is converted into the dynamic weighting model, spatial self-adaption of an evaluation standard is realized, dominant restriction factors of human settlement environments in different regions can be accurately identified, and a scientific tool is provided for differential spatial planning.

Description

Technical Field

[0001] This invention relates to the field of environmental assessment and big data analysis technology, specifically to a big data-based algorithm for assessing the human settlement environment in the Loess Plateau. Background Technology

[0002] The Loess Plateau region is characterized by fragmented topography, and the human settlement environment is strongly controlled by different geomorphic units such as plateaus, ridges, mounds, and gullies, exhibiting significant spatial heterogeneity. Existing human settlement environment assessment models, when applied to this region, typically employ a globally uniform indicator system and static weights. This "one-size-fits-all" approach has obvious drawbacks: First, it fails to embed the geographical mechanisms of the Loess Plateau's geomorphic differentiation, and cannot reflect the differences in dominant constraints between different units such as flat plateaus and deeply incised gullies (e.g., plateaus emphasize infrastructure accessibility, while gullies emphasize soil and water security), leading to distorted spatial differentiation in the assessment results. Second, the determination of model weights often lacks spatial adaptability, relying either on subjective experience or employing global statistical methods, failing to accurately quantify the true contribution of each factor in different regional environments. Third, the assessment process has a low degree of automation, making it difficult to efficiently integrate and process multi-source heterogeneous geospatial big data.

[0003] Therefore, developing a proprietary assessment scheme that can achieve spatially adaptive assessment standards and automated processes is crucial for accurately identifying human settlement problems on the Loess Plateau and supporting differentiated spatial governance. Summary of the Invention

[0004] The purpose of this invention is to provide a big data-based algorithm for assessing the human settlement environment on the Loess Plateau, in order to solve the problems mentioned in the background.

[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a big data-based algorithm for assessing the human settlement environment of the Loess Plateau, comprising the following steps:

[0006] S1. Obtain the digital elevation model and multi-source geospatial data of the assessment area, and calculate the following assessment indicators based on the data:

[0007] Soil erosion intensity index The calculation formula is as follows: ,in, As the erosivity factor of rainfall, As a soil erodibility factor, For topographic slope length and gradient factors, For soil and water conservation measures, For vegetation cover and management factors, and the Factors through formula With Normalized Difference Vegetation Index Dynamic association, , For localized experience parameters;

[0008] Comprehensive Accessibility Index It employs an improved Gaussian two-step moving search method for calculation, where the travel time between the supply and demand points is based on the average slope between the two locations. The resistance is corrected, and the corrected model is as follows: ,in, Based on design speed, This is the slope resistance coefficient. This is the corrected traffic speed;

[0009] Terrain-based residential safety index Its calculation formula is ,in, Based on slope The penalty function, Distance from potential geological disaster sites The safety decay function;

[0010] Socioeconomic Activity Index It is synthesized from nighttime light intensity, built-up land density, and population thermal data using principal component analysis.

[0011] S2. Based on the digital elevation model, extract the terrain feature vector of each spatial unit. The terrain feature vector includes at least slope, terrain roughness, and elevation variation coefficient. Use K-means or DBSCAN clustering algorithm to automatically partition all spatial units in the entire domain to form multiple different geomorphic unit type zones. The type zones include at least: plateau, ridge, slope, and gully.

[0012] S3. Based on the partitioning results of step S2, group the global assessment units according to their respective geomorphic unit types; for each geomorphic unit type, based on the assessment index data of all units in its group, independently calculate a set of dedicated assessment index weight vectors for that geomorphic unit type using the Critic-entropy weight combination method.

[0013] S4. Based on the geomorphic unit type to which each assessment unit belongs, select the corresponding dedicated assessment indicator weight vector, and sum the assessment indicator values ​​of that unit using weighted averages to obtain the human settlement environment quality index for that unit. The calculation formula is as follows: ,in, For cell index, Representation unit Landform type For the corresponding weights, This is the normalized value of the indicator;

[0014] S5. Output the spatial distribution results of the human settlement environment quality index.

[0015] Furthermore, the calculation process in step S3 specifically includes:

[0016] S3.1, for the first Geomorphic units are provided One evaluation indicator, Each evaluation unit constitutes a standardized indicator matrix. ;

[0017] S3.2, Calculate the first... Entropy value of each indicator : ,in, This represents the proportion of the standardized value in the indicator column;

[0018] S3.3, Calculate the first Coefficient of difference of each indicator : ;

[0019] S3.4 Calculating the conflict between indicators: Constructing a correlation coefficient matrix Calculate the first The amount of conflict between an indicator and other indicators ,in, As an indicator With indicators The correlation coefficient;

[0020] S3.5, Calculate the first... Information content of each indicator : ;

[0021] S3.6. Normalize the information content to obtain the first... The first landform unit Weight of each indicator : .

[0022] A human settlement environment assessment system for implementing the above method in the Loess Plateau includes:

[0023] Data storage and management server, used to store digital elevation models, multi-source geospatial data and intermediate data;

[0024] A parallel computing cluster, connected to the data storage and management server, is used to execute the evaluation index calculation task described in step S1.

[0025] The automatic landform zoning and dynamic weighting calculation server is connected to the parallel computing cluster and is used to perform the tasks of terrain feature extraction, clustering and zoning, dynamic weight calculation and human settlement environment quality index synthesis described in steps S2 to S4.

[0026] The results are output to the visualization workstation, which is connected to the automatic geomorphological zoning and dynamic weighting calculation server to generate and display thematic maps of the evaluation results.

[0027] Furthermore, the system also includes a task scheduling and workflow management server, which is used to automate the entire process from data preprocessing, index calculation, geomorphological zoning, dynamic weighting evaluation to result output.

[0028] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0029] Deep integration of mechanism and model: By “dynamic weighting driven by geomorphological zoning”, geographical mechanisms are transformed into computable model constraints, making the evaluation criteria adaptable to space and the results more geographically meaningful.

[0030] High evaluation accuracy and interpretability: Partition weighting overcomes the homogenization defect of the global static model, can accurately identify the dominant factors in different regions, and the weight generation process is transparent and traceable.

[0031] High degree of automation and engineering: It provides a complete solution from algorithms to hardware, supports automatic processing of massive data and dynamic monitoring of business operations, and is easy to promote and implement. Detailed Implementation

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] This invention provides a technical solution: a big data-based algorithm for assessing the human settlement environment of the Loess Plateau, the core of which lies in transforming geographical mechanisms (landform differentiation) into mathematical models (dynamic weights), including the following steps:

[0034] S1. Obtain the digital elevation model and multi-source geospatial data of the assessment area, such as digital elevation model, remote sensing imagery, infrastructure locations, transportation network and socio-economic data, and calculate the following assessment indicators based on the data:

[0035] 1. Soil erosion intensity index The revised general soil loss equation applicable to the Loess Plateau was used for calculation.

[0036]

[0037] in, As the erosivity factor of rainfall, As a soil erodibility factor, For topographic slope length and gradient factors, For soil and water conservation measures, For vegetation cover and management factors.

[0038] The The factor is expressed as follows: (The formula is used to correlate the normalized vegetation index with the normalized vegetation index.) Dynamic association:

[0039]

[0040] in, , These are parameters based on localized experience.

[0041] 2. Overall Accessibility Index It uses an improved Gaussian two-step moving search method for calculation, and the key improvement is the introduction of slope resistance to correct the travel time.

[0042] Travel time between supply and demand points depends on the average slope between the two locations. Apply resistance correction to determine the effective travel speed between the two points. The model is as follows:

[0043]

[0044] in, Based on design speed, This is the slope resistance coefficient. The corrected traffic speed. This represents the average slope of the path.

[0045] 3. Terrain-based residential safety index Taking into account both slope stability and geological disaster risks.

[0046]

[0047] in, Based on slope The penalty function, Distance from potential geological disaster sites The safety decay function;

[0048] 4. Socioeconomic Activity Index The data is synthesized by integrating nighttime light intensity, built-up land density, and population thermal data, using principal component analysis.

[0049] S2. Based on the digital elevation model, extract the terrain feature vector for each spatial unit. The terrain feature vector includes at least slope, terrain roughness, and elevation variation coefficient. Use K-means or DBSCAN clustering algorithms to automatically partition all spatial units in the entire domain, forming multiple different geomorphic unit type zones. These type zones include at least: plateau, ridge, slope, and gully. Output a geomorphic type raster map. .

[0050] S3. Based on the partitioning results of step S2, group the overall assessment units according to their respective geomorphic unit types; for each geomorphic unit type, independently calculate a set of dedicated assessment index weight vectors for that geomorphic unit type using the Critic-Entropy Weight Combination Method, based on the assessment index data of all units within its group; the specific steps are as follows:

[0051] S3.1, for the first Geomorphic units are provided One evaluation indicator, Each evaluation unit constitutes a standardized indicator matrix. ;

[0052] S3.2, Calculate the first... Entropy value of each indicator :

[0053]

[0054] in, This represents the proportion of the standardized value in the indicator column;

[0055] S3.3, Calculate the first Coefficient of difference of each indicator

[0056]

[0057] S3.4 Calculating the conflict between indicators: Constructing a correlation coefficient matrix Calculate the first The amount of conflict between an indicator and other indicators

[0058]

[0059] in, As an indicator With indicators The correlation coefficient;

[0060] S3.5, Calculate the first... Information content of each indicator

[0061]

[0062] S3.6. Normalize the information content to obtain the first... The first landform unit Weight of each indicator

[0063]

[0064] S4. Based on the geomorphic unit type to which each assessment unit belongs, select the corresponding dedicated assessment indicator weight vector, and sum the assessment indicator values ​​of that unit using weighted averages to obtain the human settlement environment quality index for that unit. It is calculated by the following formula:

[0065]

[0066] in, For cell index, Representation unit Landform type For the corresponding weights, This is the normalized value of the indicator.

[0067] S5. Output the spatial distribution results of the human settlement environment quality index.

[0068] A human settlement environment assessment system for implementing the above method in the Loess Plateau includes:

[0069] Data storage and management server, used to store digital elevation models, multi-source geospatial data and intermediate data;

[0070] The parallel computing cluster consists of multiple high-performance CPU servers and is connected to the data storage and management server to perform the evaluation index calculation task described in step S1.

[0071] The automatic terrain zoning and dynamic weighting calculation server, equipped with a GPU acceleration card, is connected to the parallel computing cluster and is used to perform the terrain feature extraction, clustering and zoning, dynamic weight calculation and human settlement environment quality index synthesis tasks described in steps S2 to S4.

[0072] The results are output to the visualization workstation, which is connected to the automatic geomorphological zoning and dynamic weighting calculation server to generate and display thematic maps of the evaluation results.

[0073] The system also includes a task scheduling and workflow management server, which is used to automate the entire process from data preprocessing, index calculation, geomorphological zoning, dynamic weighting evaluation to result output.

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the following description uses a loess hilly and gully area in Yulin City, Shaanxi Province as an example.

[0075] 1. Data Preparation and Index Calculation (S1): Acquire 30-meter resolution DEM, satellite remote sensing imagery, township POIs, and road network data for the area. Calculate the soil erosion intensity index on a parallel computing cluster. (set up , Comprehensive accessibility index of facilities (set up , ), terrain and residential safety index Social and economic activity index .

[0076] 2. Automatic Terrain Zoning (S2): Extract slope, roughness, and elevation variation coefficient features. Run K-means clustering (K=4) on the GPU server to obtain a raster map of terrain types. Verification has shown that these correspond to plateaus, ridges, slopes, and gullies, respectively.

[0077] 3. Dynamic weighting and index synthesis (S3, S4): The system automatically calculates weights by partitioning the data. For example, the calculation results for the "gully region" sample show: Weight 0.42, The weight is 0.28; while the "plateau area": Weight 0.38 Weight 0.32. According to the formula... The index value of each raster is synthesized.

[0078] 4. Output Results: Generate The grading map clearly shows that high-quality areas are concentrated on the plateau surface, while low-quality areas overlap with gullies. The report recommends that public service facilities be optimized in the plateau area, and that soil and water conservation projects be prioritized in the gully area.

[0079] 5. System Operation: The task scheduling server can be set to automatically execute the above process every month to achieve dynamic monitoring.

[0080] Finally, it should be noted that the above descriptions are merely preferred embodiments 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 foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A big data-based algorithm for assessing the human settlement environment on the Loess Plateau, characterized in that, Includes the following steps: S1. Obtain the digital elevation model and multi-source geospatial data of the assessment area, and calculate the following assessment indicators based on the data: Soil erosion intensity index The calculation formula is as follows: ,in, As the erosivity factor of rainfall, As a soil erodibility factor, For topographic slope length and gradient factors, For soil and water conservation measures, For vegetation cover and management factors, and the Factors through formula With Normalized Difference Vegetation Index Dynamic association, , For localized experience parameters; Comprehensive Accessibility Index It employs an improved Gaussian two-step moving search method for calculation, where the travel time between the supply and demand points is based on the average slope between the two locations. The resistance is corrected, and the corrected model is as follows: ,in, Based on design speed, This is the slope resistance coefficient. This is the corrected traffic speed; Terrain-based residential safety index Its calculation formula is ,in, Based on slope The penalty function, Distance from potential geological disaster sites The safety decay function; Socioeconomic Activity Index It is synthesized from nighttime light intensity, built-up land density, and population thermal data using principal component analysis. S2. Based on the digital elevation model, extract the terrain feature vector of each spatial unit. The terrain feature vector includes at least slope, terrain roughness, and elevation variation coefficient. Use K-means or DBSCAN clustering algorithm to automatically partition all spatial units in the entire domain to form multiple different geomorphic unit type zones. The type zones include at least: plateau, ridge, slope, and gully. S3. Based on the partitioning results of step S2, group the global assessment units according to their respective geomorphic unit types; for each geomorphic unit type, based on the assessment index data of all units in its group, independently calculate a set of dedicated assessment index weight vectors for that geomorphic unit type using the Critic-entropy weight combination method. S4. Based on the geomorphic unit type to which each assessment unit belongs, select the corresponding dedicated assessment indicator weight vector, and sum the assessment indicator values ​​of that unit using weighted averages to obtain the human settlement environment quality index for that unit. The calculation formula is as follows: ,in, For cell index, Representation unit Landform type For the corresponding weights, This is the normalized value of the indicator; S5. Output the spatial distribution results of the human settlement environment quality index.

2. The big data-based human settlement environment assessment algorithm for the Loess Plateau as described in claim 1, characterized in that, The calculation process in step S3 specifically includes: S3.1, for the first Geomorphic units are provided One evaluation indicator, Each evaluation unit constitutes a standardized indicator matrix. ; S3.2, Calculate the first... Entropy value of each indicator : ,in, This represents the proportion of the standardized value in the indicator column; S3.3, Calculate the first Coefficient of difference of each indicator : ; S3.4 Calculating the conflict between indicators: Constructing a correlation coefficient matrix Calculate the first The amount of conflict between an indicator and other indicators ,in, As an indicator With indicators The correlation coefficient; S3.5, Calculate the first... Information content of each indicator : ; S3.

6. Normalize the information content to obtain the first... The first landform unit Weight of each indicator : .

3. A human settlement environment assessment system for implementing the method of claim 1 on the Loess Plateau, characterized in that, include: Data storage and management server, used to store digital elevation models, multi-source geospatial data and intermediate data; A parallel computing cluster, connected to the data storage and management server, is used to perform the evaluation index calculation task described in step S1 of claim 1. The automatic landform zoning and dynamic weighting calculation server is connected to the parallel computing cluster and is used to perform the tasks of terrain feature extraction, clustering and zoning, dynamic weight calculation and human settlement environment quality index synthesis described in steps S2 to S4 of claim 1. The results are output to the visualization workstation, which is connected to the automatic geomorphological zoning and dynamic weighting calculation server to generate and display thematic maps of the evaluation results.

4. The Loess Plateau human settlement environment assessment system according to claim 3, characterized in that, The system also includes a task scheduling and workflow management server, which is used to automate the entire process from data preprocessing, index calculation, geomorphological zoning, dynamic weighting evaluation to result output.

Images