Measurement system for natural resource engineering land assessment

Abstract

The invention provides a measurement system for natural resource engineering land evaluation, and relates to the technical field of data processing, and the system comprises a region modeling module which is used for receiving topographic data, landform data and soil data, and carrying out the optimization selection of model parameters through a genetic algorithm, so as to obtain final model parameters; constructing a three-dimensional model of the measurement area by utilizing the final model parameters; the finite element analysis module is used for acquiring a three-dimensional model and dividing the three-dimensional model into a plurality of feature regions by applying a finite element analysis method; and determining a corresponding region division strategy by using a particle optimization algorithm, performing geographical and soil characteristic analysis on each feature region, and extracting key indexes. The accuracy and efficiency of land assessment can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a measurement system for natural resource engineering land evaluation. Background Art

[0002] With the continuous development of natural resource engineering technology, land evaluation and measurement have become a key link in the rational development and utilization of land resources. Some traditional land evaluation methods still rely on manual on-site surveys and basic data analysis. Although they can meet the evaluation requirements to a certain extent, there are some deficiencies.

[0003] For example, some traditional land evaluations lack efficient and intelligent optimization means in regional modeling. The selection of model parameters mostly depends on empirical judgment, lacking scientific basis, and it is difficult to ensure the accuracy and applicability of model construction. This has limited the in-depth and detailed evaluation work to a certain extent.

[0004] Moreover, some traditional regional division strategies are relatively rough. Therefore, it is difficult to accurately capture the subtle differences in land characteristics, which makes the analysis of the geographical and soil characteristics of the characteristic regions not deep enough and the extraction of key indicators not accurate enough. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a measurement system for natural resource engineering land evaluation, which can improve the accuracy and efficiency of land evaluation.

[0006] To solve the above technical problem, the technical solution of the present invention is as follows:

[0007] A measurement system for natural resource engineering land evaluation, the system includes:

[0008] A data collection module, used to obtain topographic data, geomorphic data, and soil data of the land to be evaluated;

[0009] A regional modeling module, used to receive topographic data, geomorphic data, and soil data, optimize and select model parameters through a genetic algorithm to obtain final model parameters; use the final model parameters to construct a three-dimensional model of the measurement area;

[0010] A finite element analysis module, used to obtain the three-dimensional model, and divide it into multiple characteristic regions by applying the finite element analysis method; use a particle optimization algorithm to determine the corresponding regional division strategy, analyze the geographical and soil characteristics of each characteristic region, and extract key indicators;

[0011] A data processing module, used to receive key indicators, and evaluate the data in each characteristic region through cluster analysis to identify potential problems in land use.

[0012] Further, terrain data, landform data, and soil data are received, and genetic algorithms are used to optimize and select model parameters to obtain the final model parameters, including:

[0013] Determine the basic framework and model parameter range of the three-dimensional model;

[0014] Encode the parameters and convert them into the gene string form of genetic algorithms;

[0015] Randomly generate an initial population, where each individual in the population represents a set of model parameters;

[0016] Define a fitness function for evaluating the quality of each individual, i.e., each set of model parameters; according to the fitness function, select the corresponding individuals in the population as parents;

[0017] Randomly select two parent individuals, generate new offspring individuals through crossover operations, randomly change the genes of the offspring individuals, add the newly generated offspring individuals to the population to form a new population, and repeat the fitness evaluation, selection, and genetic operations until the preset number of iterations is reached to obtain the final model parameters.

[0018] Further, the calculation formula for the fitness function F(x) is:

[0019]

[0020] where w s represents the weight coefficient of soil quality; S(x i ) represents the soil quality score of the i-th region; n represents the total number of sample regions; w f represents the weight coefficient of soil fertility; F(x i ) represents the soil fertility score of the i-th region; w a represents the weight coefficient of the accuracy of regional division; represents the category of the i-th region in the model division; C i represents the true category of the i-th region; represents the indicator function, which takes 1 when the model prediction is consistent with the true category, and 0 otherwise; w e represents the weight coefficient of the silhouette coefficient; S(i) represents the silhouette coefficient of the i-th region; i represents the index value.

[0021] Further, obtain the three-dimensional model and apply the finite element analysis method to divide it into multiple characteristic regions, including:

[0022] Extract all vertex coordinates from the three-dimensional model to form a point set, perform triangulation on the point set to generate a convex hull composed of tetrahedrons in three-dimensional space, and ensure that no points in the point set are inside the circumcircle of each tetrahedron; screen the tetrahedrons to obtain a triangular mesh approximate to the surface of the original three-dimensional model;

[0023] Identify different material regions based on soil and geomorphic data, map the different material regions onto the three-dimensional model to form different material partitions;

[0024] For each material partition, retrieve the corresponding elastic modulus, Poisson's ratio, and density properties from the database and assign the properties to the corresponding grid cells to obtain a three-dimensional model in which each grid cell has the corresponding material properties;

[0025] Determine the type of dynamic analysis and the corresponding analysis parameters;

[0026] Perform calculations through the finite element analysis algorithm. After the calculation is completed, extract the key indicators in the results;

[0027] Analyze the key indicators to identify characteristic regions with stress concentration, severe deformation, or significant changes in soil properties;

[0028] According to the characteristic regions, divide the three-dimensional model into multiple sub-regions. Each sub-region has similar mechanical behaviors or soil properties, and assign a unique identifier to each sub-region to obtain the three-dimensional model of the finally divided multiple characteristic regions.

[0029] Furthermore, extract all vertex coordinates from the three-dimensional model to form a point set, perform triangulation on the point set to generate a convex hull composed of tetrahedrons in three-dimensional space, and ensure that no points in the point set are inside the circumcircle of each tetrahedron; screen the tetrahedrons to obtain a triangular mesh approximate to the surface of the original three-dimensional model, including:

[0030] Extract all vertex coordinates in the three-dimensional model to form a point set, and set the initial temperature, temperature drop rate, termination temperature, and the number of iterations L at each temperature of the simulated annealing algorithm;

[0031] At the current temperature, randomly select four points in the point set to form a tetrahedron;

[0032] Calculate the first evaluation function value of the tetrahedron and perform L iterations at the current temperature;

[0033] In each iteration, a new tetrahedron configuration is randomly generated, and the value of the second evaluation function for the new configuration is calculated. If the value of the second evaluation function satisfies the Metropolis criterion, the new configuration is accepted, and the current final configuration is updated. The temperature is decreased until the preset number of iterations is reached. After simulated annealing optimization, a set of final tetrahedron configurations is obtained;

[0034] The tetrahedrons approximated to the surface of the original 3D model are selected from the final tetrahedron configurations to form a triangular mesh.

[0035] Further, the soil data includes soil type, texture, and organic matter content, and the geomorphic data includes terrain elevation, slope, and aspect.

[0036] Further, based on the soil and geomorphic data, different material regions are identified and mapped onto the 3D model to form different material partitions, including:

[0037] Key features are extracted from the soil data, including soil type, particle size distribution, soil texture, organic matter content, and pH value; geomorphic features are extracted from the geomorphic data, including terrain elevation, slope, aspect, and geomorphic form;

[0038] The key features and geomorphic features are integrated to form a dataset;

[0039] The features in the dataset are standardized, and the sample points in the dataset are clustered by K-means clustering to identify groups of sample points with similar soil and geomorphic characteristics. Each group represents a similar region;

[0040] The material type of each region is determined according to the clustering center to divide each similar region into different material types, including sandy soil areas, clay areas, and rock areas;

[0041] The similar regions with similar soil and geomorphic characteristics are aligned and matched with the 3D model;

[0042] On the 3D model, different material partitions are created according to the mapped material region boundaries. Each partition should represent a region with unique soil and geomorphic characteristics and be assigned corresponding material properties.

[0043] Further, a particle optimization algorithm is used to determine the corresponding region division strategy, and the geographical and soil characteristics of each feature region are analyzed to extract key indicators, including:

[0044] The size of the particle swarm is set, that is, the number of particles; each particle represents a region division strategy and has two attributes: position and velocity; the position of the particle represents the parameters of the region division, and the velocity represents the moving direction and rate of the particle in the search space;

[0045] Construct a fitness function to evaluate the advantages and disadvantages of each regional division strategy;

[0046] In each iteration, calculate the fitness value of each particle according to the fitness function, update the individual position and global position of each particle; according to the individual position and global position, adjust the speed and position of each particle. When the preset number of iterations is reached, the corresponding particles are obtained; the regional division strategy represented by the corresponding particles is used as the final strategy;

[0047] According to the final strategy, divide the entire region into several characteristic regions, and for each characteristic region, use soil analysis methods for analysis to extract key indicators.

[0048] Furthermore, the key indicators include terrain elevation range, slope distribution, soil type and distribution, and soil fertility.

[0049] Furthermore, receive the key indicators, and evaluate the data within each characteristic region through cluster analysis to identify potential problems in land use, including:

[0050] Receive key indicator data, including terrain elevation range, slope distribution, soil type and distribution, and soil fertility;

[0051] Divide the key indicator data into different clusters through the K-means clustering algorithm, and each cluster represents a region with similar characteristics;

[0052] Analyze the characteristics of each clustering cluster, count the key indicators, and identify land use problems existing in each cluster, including soil erosion risk, insufficient soil fertility, and terrain limitations.

[0053] The above solution of the present invention has at least the following beneficial effects.

[0054] Through the data collection module, it is possible to comprehensively obtain the terrain, landform, and soil data of the land to be evaluated. These data are the basis for land evaluation, ensuring the accuracy and reliability of the evaluation.

[0055] The regional modeling module uses the genetic algorithm to optimize the selection of model parameters, thereby constructing a more accurate three-dimensional model. This optimization method can improve the adaptability and accuracy of the model and better reflect the actual situation of the land.

[0056] The finite element analysis module can divide the three-dimensional model into multiple characteristic regions and use the particle optimization algorithm to determine the division strategy. This refined analysis method helps to more deeply understand the geographical and soil characteristics of each region and extract key indicators.

[0057] The data processing module evaluates the data within each feature area through cluster analysis and can automatically identify potential problems in land use. This intelligent data processing method not only improves the evaluation efficiency. Description of the Drawings

[0058] Figure 1 It is a schematic diagram of a measurement system for natural resource engineering land evaluation provided by an embodiment of the present invention. Detailed Implementation Modes

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

[0060] As Figure 1 shown, an embodiment of the present invention provides a measurement system for natural resource engineering land evaluation, including:

[0061] A data collection module for obtaining topographic data, geomorphic data, and soil data of the land to be evaluated;

[0062] A regional modeling module for receiving topographic data, geomorphic data, and soil data, optimizing and selecting model parameters through a genetic algorithm to obtain final model parameters; constructing a three-dimensional model of the measurement area using the final model parameters;

[0063] A finite element analysis module for obtaining the three-dimensional model and dividing it into multiple feature areas by applying the finite element analysis method; determining the corresponding area division strategy using a particle optimization algorithm, analyzing the geographical and soil characteristics of each feature area, and extracting key indicators;

[0064] A data processing module for receiving the key indicators and evaluating the data within each feature area through cluster analysis to identify potential problems in land use.

[0065] In the embodiments of the present invention, the data collection module can collect multi-dimensional data such as the terrain, landform, and soil of the land to be evaluated, ensuring that the basic data for evaluation is comprehensive and accurate. By means of automated data collection, manual operations are reduced, and the efficiency and accuracy of data collection are improved. Using the genetic algorithm to optimize the selection of model parameters can ensure that the constructed three-dimensional model more accurately reflects the actual land situation. The constructed three-dimensional model provides an intuitive visualization tool, which helps to better understand and analyze land characteristics. Through the precise model; applying the finite element analysis method to divide the land into multiple characteristic regions helps to more carefully understand the local characteristics of the land. Using the particle optimization algorithm to determine the regional division strategy can more efficiently extract the geographical and soil key indicators of each characteristic region; targeted analysis of each characteristic region can reveal specific problems and potential in land use. By clustering analysis to evaluate the data within each characteristic region, similar and different data groups can be identified, and then the data processing process can be optimized; clustering analysis helps to quickly identify potential problems in land use and provide early warnings for timely countermeasures.

[0066] To obtain accurate terrain data, an airborne laser measurement system (such as ARS-200) was used. This system can quickly obtain high-precision point cloud data by integrating sensors such as a high-precision laser scanner, GNSS positioning system, and IMU inertial navigation system. The system successfully obtained the terrain data within 1 square kilometer in 30 minutes. Through remote sensing technology, high-resolution remote sensing images were obtained. These images clearly show the landform features such as hills, farmland, and reservoirs in the surveyed area. Combining with ground surveys, the landform data was classified and labeled in detail; to obtain soil data, the methods of soil sampling and laboratory analysis were used. Sampling points were evenly distributed in the surveyed area, and soil samples were collected at each point. Subsequently, these samples were sent to the laboratory for detailed physical and chemical property analysis, including key indicators such as soil type, organic matter content, and pH value.

[0067] In a preferred embodiment of the present invention, the terrain data, landform data, and soil data are received, and the genetic algorithm is used to optimize the selection of model parameters to obtain the final model parameters, including:

[0068] Determine the basic framework and model parameter range of the 3D model, specifically including: Select a grid-based model because it can flexibly represent the complex details of terrain and landforms and is compatible with many existing Geographic Information Systems (GIS) and visualization tools. The grid-based model divides the geospatial into a series of regular grid cells. Each grid cell can store various attribute information related to that location, such as height, slope, soil type, etc. After determining the grid-based 3D model structure, determine the model parameters and optimize them through genetic algorithms. Terrain height is a basic parameter describing the undulation of the ground surface. By optimizing the height values of each grid cell, the details of the actual terrain can be simulated more precisely. Slope represents the degree of surface inclination and has an important impact on natural processes such as water flow direction, soil erosion, and vegetation distribution. The search range of the slope can vary from flat (0 degrees) to steep (close to 90 degrees), depending on the actual terrain characteristics of the study area. Geomorphic types (such as mountains, plains, hills, etc.) are comprehensive indicators describing the surface form. In the model, this parameter can be optimized by assigning different geomorphic type labels to each grid cell. The search range covers all possible geomorphic types within the study area. Soil thickness affects multiple aspects such as vegetation growth, hydrological cycle, and land use patterns. When optimizing the soil thickness parameter, the corresponding thickness range can be determined according to different soil types and geological conditions; Soil moisture is an important parameter reflecting the soil water condition and is crucial for agricultural production, ecological restoration, etc. The search range of the moisture can be determined according to the rainfall pattern, soil type, and vegetation cover of the study area.

[0069] Encode the parameters and convert them into the gene string form of the genetic algorithm, specifically including: Convert each model parameter into an encoding form that the genetic algorithm can handle. This usually means converting continuous numerical parameters into discrete binary or other forms of encoding. For example, a floating-point number representing height may be converted into a fixed-length binary string; Concatenate all the encoded parameters to form a "gene string" or "chromosome" representing the entire set of model parameters. This gene string will serve as the basic representation of an individual in the genetic algorithm.

[0070] Randomly generate an initial population, where each individual in the population represents a set of model parameters, specifically including: Determine the size of the initial population. The population size affects the search ability and convergence speed of the genetic algorithm. Within the set parameter range, randomly generate a set of model parameters and encode them into the gene string form. In this way, an individual is formed. Repeat this process until a sufficient number of individuals are generated to form the initial population.

[0071] Define a fitness function for evaluating the quality of each individual, i.e., each set of model parameters; according to the fitness function, select the corresponding individuals in the population as parents. Specifically, for each individual in the population, use the model parameters decoded from its gene string to calculate the value of the fitness function. This value will serve as the basis for subsequent selection and genetic operations.

[0072] Randomly select two parent individuals, generate new offspring individuals through crossover operations, randomly change the genes of the offspring individuals, and add the newly generated offspring individuals to the population to form a new population. Repeat the fitness evaluation, selection, and genetic operations until the preset number of iterations is reached to obtain the final model parameters. Specifically, formulate a selection strategy, such as roulette wheel selection, tournament selection, etc., to select parents according to the fitness values of individuals. The goal of the strategy is to select individuals with better performance while maintaining the diversity of the population; according to the selection strategy, select a certain number of individuals from the current population as parents for subsequent crossover and mutation operations; randomly select two parent individuals and determine one or more crossover points. These points will be used to cut the gene strings of the parents and exchange the corresponding parts to generate new offspring individuals; cut the gene strings of the parents at the crossover points, exchange the cut parts, and then recombine these parts into new gene strings to generate offspring individuals; randomly change (mutate) the genes of the offspring individuals with a certain probability to increase the diversity of the population. Mutation can be bit flipping or more complex operations; add the newly generated offspring individuals to the original population (possibly replacing some old individuals) to form a new population. This new population will be used for the next round of fitness evaluation, selection, and genetic operations; repeat the above fitness evaluation, selection, and genetic operations, that is, continuously evaluate the fitness of the new population, select parents, perform crossover and mutation operations, and generate a new population; when the preset number of iterations is reached, stop the iteration process, select the individual with the highest fitness from the population in the last round of iteration, and decode its gene string into model parameters. These parameters are the final model parameters optimized by the genetic algorithm.

[0073] In the embodiments of the present invention, through the optimization process of the genetic algorithm, the most suitable model parameters can be found, so that the constructed 3D model can more accurately reflect the actual land conditions. This improvement in accuracy contributes to the accuracy of subsequent land evaluations. The genetic algorithm is an automated optimization method that can automatically search for the optimal solution within the preset parameter range, reducing the process of manual intervention and trial-and-error, and improving work efficiency. The genetic algorithm has strong global optimization capabilities and is not easily trapped in local optimal solutions. Therefore, it can find a model parameter combination closer to the global optimum, making the model have better generalization ability and prediction accuracy. The genetic algorithm has good applicability to different types of land data (such as terrain, landform, soil, etc.) and different 3D model frameworks, and can adapt to different optimization requirements by adjusting the fitness function and genetic operations. Since the genetic algorithm is based on the principle of population evolution, it is not sensitive to the selection of initial parameters. Even if the quality of individuals in the initial population is not high, it can gradually approach the optimal solution through multiple iterations.

[0074] In a preferred embodiment of the present invention, the calculation formula of the fitness function F(x) is:

[0075]

[0076] where w s represents the weight coefficient of soil quality; S(x i ) represents the soil quality score of the i-th region; n represents the total number of sample regions; w f represents the weight coefficient of soil fertility; F(x i ) represents the soil fertility score of the i-th region; w a represents the weight coefficient of the accuracy part of the regional division; represents the category of the i-th region in the model division; C i represents the true category of the i-th region; represents the indicator function, which takes 1 when the model prediction is consistent with the true category, and 0 otherwise; w e represents the weight coefficient of the silhouette coefficient; S(i) represents the silhouette coefficient of the i-th region; i represents the index value.

[0077] In the embodiments of the present invention, this function can comprehensively evaluate the land from multiple dimensions such as soil quality, soil fertility, accuracy of regional division, and silhouette coefficient, so as to obtain a comprehensive and objective evaluation result. By adjusting the weight coefficients of each part, certain evaluation indicators can be emphasized or weakened according to actual needs, making the evaluation more flexible and customizable. By introducing the evaluation part of the accuracy of regional division This function can quantify the matching degree between the predicted category of the model and the true category, thus reflecting the accuracy of the evaluation. The introduction of the silhouette coefficient S(i) enables the evaluation to take into account the clustering effect of the data. The silhouette coefficient is a way to evaluate the quality of the clustering effect. The larger its value, the better the clustering effect, that is, the closer the distance between samples of the same class and the farther the distance between samples of different classes. Among them, the soil quality score S(x i ) is calculated as follows:

[0078]

[0079] Among them, A i represents the soil pH; A min and A max are the minimum and maximum values of the soil pH respectively; O i represents the soil organic matter content; O min and O max are the minimum and maximum values of the soil organic matter content respectively; T i represents the soil texture score; T min and T max are the minimum and maximum values of the soil texture score respectively; w A , w O and w T are weight coefficients; among them,

[0080]

[0081] Among them, w S , w c and w M are weight coefficients; S i , C i and M i are the actual percentage contents of sand, clay and silt in sample i respectively; S min and S max are the minimum and maximum values of sand respectively; C min and C max are the minimum and maximum values of clay respectively; M min and M max are the minimum and maximum values of silt respectively. Among them, the soil fertility score F(x i ) is calculated as follows:

[0082]

[0083] Among them, N i represents the nitrogen content; P i represents the phosphorus content; K i represents the potassium content; w N , w P and wK is the weight coefficient; N min and N max are the minimum and maximum values of nitrogen content respectively; P min and P max are the minimum and maximum values of phosphorus content respectively; K min and K max are the minimum and maximum values of potassium content respectively.

[0084] In a preferred embodiment of the present invention, a three-dimensional model is obtained and divided into multiple feature regions by applying the finite element analysis method, including:

[0085] All vertex coordinates are extracted from the three-dimensional model to form a point set, and the point set is triangulated to generate a convex hull composed of tetrahedrons in three-dimensional space, and no points in the point set are contained inside the circumcircle of each tetrahedron; the tetrahedrons are screened to obtain a triangular mesh approximate to the surface of the original three-dimensional model;

[0086] According to the soil and geomorphic data, different material regions are identified and mapped onto the three-dimensional model to form different material partitions;

[0087] For each material partition, the corresponding elastic modulus, Poisson's ratio, and density properties are retrieved from the database and assigned to the corresponding grid cells to make the three-dimensional model of each grid cell have the corresponding material properties, specifically including: establishing a database that stores the physical properties of various materials, including elastic modulus, Poisson's ratio, and density, etc. The database can be classified according to material types (such as soil, rock, concrete, etc.) and provide detailed physical property values for each material. Enter the physical property data of various materials into the database and ensure the accuracy and reliability of these data. The data sources can be literature materials, experimental data, or standard values provided by professional institutions. According to the previously identified material regions, determine the material region to which each grid cell belongs, which can be achieved by comparing the spatial position of the grid cell and the spatial range of the material region. Set the database query conditions according to the material region to which the grid cell belongs. This usually involves specifying the material type or other relevant attributes to retrieve the correct physical property values from the database, execute the database query, obtain the physical property data corresponding to the specific material region, including elastic modulus, Poisson's ratio, and density, etc., match the retrieved physical property data with the corresponding grid cells, and assign these properties to the corresponding grid cells.

[0088] Determine the type of dynamic analysis and the corresponding analysis parameters, specifically including: collecting seismic wave data of the area where the land to be evaluated is located, including historical earthquake records, synthetic seismic waves, etc., and these data should reflect the seismic activity characteristics and potential seismic hazards of the area. According to the seismic wave data and the requirements of engineering seismic design, set the amplitude (peak acceleration, velocity or displacement), main frequency range and seismic duration of the seismic wave. Determine the input direction of ground motion, such as one-way, two-way or three-way ground motion input, to simulate the complex situation under actual earthquake action. Analyze the support conditions of the structure in the actual project, such as the foundation type, connection method, etc., and set the boundary conditions of the model according to the structural support conditions and analysis requirements. For example, for a structure with a fixed foundation, fixed support boundary conditions can be set at the bottom of the model. After setting the boundary conditions, conduct verification to ensure its accuracy and rationality, which can be achieved by comparing the boundary condition settings of similar projects, conducting preliminary calculation analysis, etc. Apply the set seismic response analysis parameters and boundary conditions to the three-dimensional finite element model.

[0089] Perform calculations using the finite element analysis algorithm. After the calculations are completed, extract the key indicators from the results, specifically including: confirm the current file format of the assigned 3D model. If it is not a format supported by the finite element analysis software (such as.stl,.step,.iges, etc.), use CAD software or conversion tools to perform format conversion; during the conversion process, ensure that the geometric accuracy and material property information of the model are retained; open SolidWorks or other finite element analysis software and use its provided import function to load the 3D model in the converted format. During the import process, the software provides some options to adjust the import settings of the model, such as units, scale, etc., and configure them according to the actual situation to ensure the accuracy of the model. After import, check whether the geometric shape and material properties of the model are consistent with the original model to ensure that there is no loss or deformation; after importing the model, the software automatically generates a mesh for the model. The automatically generated mesh may not necessarily meet the requirements of the analysis accuracy; therefore, check the mesh quality. The software provides some tools to evaluate the mesh quality, such as mesh density, element quality factor, etc.; if it is found that the mesh quality is poor, such as having overly large elements or deformed elements, perform mesh refinement or optimization, which can be achieved by adjusting mesh parameters, manually splitting or merging elements, etc. Set an appropriate time step according to the purpose of the analysis and the dynamic characteristics of the model. The size of the time step directly affects the accuracy and computational efficiency of the analysis; if the model contains high-frequency vibrations or rapidly changing dynamic responses, a smaller time step needs to be set to capture these details. Select an iterative solver according to the scale and complexity of the problem; ensure that the previously defined boundary conditions (such as fixed supports, free boundaries, etc.) have been correctly applied to the model, which usually involves selecting specific parts of the model and assigning the corresponding boundary conditions to them. Apply the seismic load to the model, which requires defining parameters such as the direction, amplitude, and time history of the seismic wave; specify the data that needs to be recorded and output during the analysis process. These data include key indicators such as stress distribution, displacement distribution, vibration frequency, etc. By setting appropriate output options, the required result data can be directly obtained after the analysis is completed; after completing all the settings, start the analysis process, which usually involves clicking the "Run" or similar button in the software. During the analysis run, monitor the calculation progress and resource usage, which can be achieved by viewing the progress bar, calculation time estimate, or resource occupancy provided by the software; if any problems are encountered, such as calculation errors or insufficient resources, adjust the analysis settings or increase the computing resources in a timely manner to ensure that the analysis can proceed smoothly. After the analysis is completed, view the analysis results generated by the software. These results are usually presented in the form of charts, curves, and data tables. Extract the key indicators from the analysis results, such as the maximum stress value, maximum displacement value, etc. These indicators can be directly read or calculated through the post-processing tools provided by the software.

[0090] Analyze key indicators to identify characteristic areas with stress concentration, severe deformation, or significant changes in soil properties, specifically including:

[0091] Organize the key indicator data extracted from the above steps to ensure data integrity and accuracy, which includes stress distribution data, displacement distribution data, soil property change data, etc.; select appropriate data analysis tools according to the type and characteristics of the data, which includes professional finite element analysis post-processing software, data visualization tools, or statistical analysis software; view the stress distribution in the model, usually presented through color contour maps or isograms, which can visually show the distribution and magnitude of stress in different regions. In the stress distribution map, look for points or regions where the stress value is significantly higher than the surrounding areas, and these are the characteristic areas of stress concentration, which are caused by factors such as sudden changes in structural geometry, differences in material properties, or external load effects. Analyze the reasons for the identified stress concentration areas to understand which factors lead to stress concentration, so as to carry out optimization design or take corresponding measures to reduce the degree of stress concentration later. View the displacement distribution in the model, which can also be presented through color contour maps or isograms, and these charts can show the magnitude and distribution of displacements of the structure in different directions. In the displacement distribution map, look for points or regions where the displacement value is significantly larger than the surrounding areas, and these are the characteristic areas of severe deformation. These areas may be caused by factors such as insufficient structural stiffness, excessive external load, or unreasonable support conditions. Evaluate the impact of the identified severely deformed areas on the overall performance and safety of the structure to determine the degree of influence, so as to take corresponding measures to strengthen the structural stiffness or adjust the support conditions later. If the model includes soil or foundation parts, then it is necessary to view the changes in soil properties, which includes changes in indicators such as soil stress, strain, pore water pressure, etc. In the soil data map, look for areas where the characteristic values change significantly, and these are the characteristic areas of significant changes in soil properties, and these changes are caused by factors such as geological condition differences, groundwater activities, or external load effects. Conduct geological analysis on the identified areas with significant changes in soil properties to understand which geological factors lead to the changes in properties, so as to carry out foundation treatment or take corresponding measures to ensure the safety of the structure later.

[0092] According to the characteristic regions, the three-dimensional model is divided into multiple sub-regions, each sub-region having similar mechanical behaviors or soil properties. A unique identifier is assigned to each sub-region to obtain the three-dimensional model of the multiple characteristic regions after final division, which specifically includes: obtaining the characteristic regions identified in the previous steps where stress concentration, severe deformation, or significant changes in soil properties occur, and these regions are important bases for dividing the sub-regions. Based on the nature of the characteristic regions, determine the criteria for dividing the sub-regions. For example, division can be made according to the similarity of stress levels, displacement magnitudes, or soil types. According to the analysis results of the characteristic regions, determine the geometric boundaries of each sub-region, and these boundaries should be able to clearly distinguish regions with different mechanical behaviors or soil properties. Use the division tools in professional three-dimensional modeling software or finite element analysis software to divide the model into sub-regions according to the determined geometric boundaries, ensuring the accuracy of geometric shapes and the consistency of topological relationships during the division process. After the division is completed, verify each sub-region to ensure that it has similar mechanical behaviors or soil properties, and this can be achieved by comparing the key index data within the sub-region. To facilitate subsequent management and analysis work, design a clear and easy-to-understand sub-region identifier system, which can be in the form of numbers, letters, or combinations, ensuring that each sub-region has a unique identifier. According to the designed identifier system, assign a unique identifier to each divided sub-region, and this identifier will be used in subsequent data recording, result querying, visualization display, and other links; establish a sub-region information database or table to record key data such as the identifier, geometric information, mechanical behavior, or soil properties of each sub-region.

[0093] In the embodiment of the present invention, by extracting vertex coordinates from the three-dimensional model and performing triangulation, high-precision tetrahedral meshes can be generated, and these meshes can more accurately represent the shape and details of the original three-dimensional model. Combining soil and geomorphic data, map different material regions onto the three-dimensional model and retrieve the corresponding material properties from the database for assignment, which enables each mesh element to have physical properties matching the actual materials, thereby improving the accuracy and authenticity of finite element analysis. The type of dynamic analysis and corresponding analysis parameters are determined in the process, which makes the analysis process more flexible and can adjust the focus and accuracy of the analysis according to actual needs to meet diverse engineering analysis requirements. Through finite element analysis, characteristic regions with stress concentration, severe deformation, or significant changes in soil properties can be accurately identified. At the same time, according to these characteristic regions, the three-dimensional model is divided into multiple sub-regions, each sub-region having similar mechanical behaviors or soil properties, which helps to manage and analyze the model more precisely. The entire process has a high degree of automation and can greatly improve the efficiency and accuracy of engineering analysis.

[0094] In a preferred embodiment of the present invention, all vertex coordinates are extracted from the three-dimensional model to form a point set, and the point set is triangulated to generate a convex hull composed of tetrahedrons in three-dimensional space, and the interior of the circumcircle of each tetrahedron does not contain points in the point set; the tetrahedrons are screened to obtain a triangular mesh approximate to the surface of the original three-dimensional model, including:

[0095] Extract all vertex coordinates in the three-dimensional model to form a point set, and set the initial temperature, temperature reduction rate, termination temperature of the simulated annealing algorithm, and the number of iterations L at each temperature. Specifically, it includes: opening the target three-dimensional model using three-dimensional modeling software or related tools, extracting the coordinate information of all vertices in the model through the functions or APIs of the software, saving the extracted vertex coordinates into a set to form a point set, and this point set will be used for subsequent tetrahedron generation and optimization processes; determining the initial temperature T0 of the simulated annealing algorithm, which is usually a relatively high value to ensure that the algorithm can explore a wider solution space at the beginning; setting the temperature reduction rate α, which determines the speed of temperature reduction after each iteration, setting the termination temperature Tf, when the temperature drops to this value, the algorithm will stop iterating, and setting the number of iterations L at each temperature, that is, how many different tetrahedron configurations are tried at each temperature.

[0096] At the current temperature, randomly select four points from the point set to form a tetrahedron; calculate the first evaluation function value of the tetrahedron. At the current temperature, perform L iterations; in each iteration, randomly generate a new tetrahedron configuration, calculate the second evaluation function value of the new configuration, if the second evaluation function value satisfies the Metropolis criterion, accept the new configuration, and update the current final configuration, reduce the temperature until the preset number of iterations is reached. After simulated annealing optimization, a set of final tetrahedron configurations is obtained; screen out the tetrahedrons approximate to the surface of the original three-dimensional model from the final tetrahedron configurations to form a triangular mesh; The soil data includes soil type, texture and organic matter content, and the geomorphic data includes terrain elevation, slope and aspect, specifically including:

[0097] At the current temperature, randomly select four non-repeating points from the point set, use these four points to form a tetrahedron, and calculate the first evaluation function value of the tetrahedron to evaluate the quality or adaptability of the tetrahedron, and enter the iterative process: at the current temperature, repeat the following steps L times:

[0098] Randomly generate a new tetrahedron configuration (i.e., select four new points to form a tetrahedron), calculate the value of the second evaluation function for the new configuration, and determine whether to accept the new configuration according to the Metropolis criterion. If the evaluation function value of the new configuration is better (i.e., lower or higher, depending on the definition of the evaluation function), or satisfies a certain probability condition, then accept the new configuration; if the new configuration is accepted, update the current final configuration to the new configuration; reduce the temperature, usually using the temperature reduction rate α to calculate the new temperature, and repeat the calculation of the first evaluation function value and the iterative operation for this tetrahedron until the preset number of iterations is reached or the temperature drops to the termination temperature Tf; after simulated annealing optimization, obtain a set of final tetrahedron configurations, and screen out the tetrahedrons that are approximately the same as the surface of the original 3D model from this set of configurations according to a certain criterion (such as the proximity of the tetrahedron to the surface of the original 3D model), and use the surface triangles of the selected tetrahedrons to form a triangular mesh. The finally obtained triangular mesh can be used to represent the approximate surface of the original 3D model. Among them, the calculation formula for the first evaluation function value is:

[0099] F 1 = w g ·(∑ i ||P i - P′ i || 2 ) + w s ·(∑ i w s ·(T i - T′ i ) 2 ) + w m ·(∑ i w m ·(G i - G′ i ) 2 );

[0100] Among them, ||P i - P′ i || 2 represents the Euclidean distance, P i and P′ i are the coordinates of the corresponding points. The smaller the error, the better the matching between the model and the original data; T i is the soil property value of the i-th region (such as soil texture or organic matter content), and T′ i is the corresponding property value in the original soil data; G i is the geomorphic property value of the i-th region (such as elevation or slope, etc.), and G′ i is the corresponding property value in the original geomorphic data; w g , w s and w m are weight coefficients.

[0101] In an embodiment of the present invention, by extracting vertex coordinates in a three-dimensional model and forming a point set, and then performing triangulation, a convex hull composed of tetrahedrons is generated. The simulated annealing algorithm is used to optimize the tetrahedron configuration. By setting reasonable initial temperature, temperature reduction rate, termination temperature, and the number of iterations at each temperature, a better tetrahedron configuration can be searched globally. This optimization method not only improves the search efficiency but also helps to avoid falling into local optimal solutions. After being optimized by simulated annealing, tetrahedrons approximated to the surface of the original three-dimensional model are selected from the final tetrahedron configuration, and then a triangular mesh is formed. The triangular mesh generated by this method can approximate the surface of the original three-dimensional model more accurately.

[0102] In a preferred embodiment of the present invention, according to soil and geomorphic data, different material regions are identified and mapped onto the three-dimensional model to form different material partitions, including:

[0103] Key features are extracted from the soil data, including soil type, particle size distribution, soil texture, organic matter content, and pH value; geomorphic features are extracted from the geomorphic data, including terrain elevation, slope, aspect, and geomorphic form, specifically including: collecting soil data of the target area, which includes soil type maps, soil investigation reports, or laboratory test data, etc.; extracting key features from the collected data, such as soil type (e.g., sandy soil, clay soil, loam soil, etc.), particle size distribution (data obtained through particle size analysis), soil texture (classification based on particle size distribution, such as sandy, loamy, clayey, etc.), organic matter content (obtained through chemical analysis), and pH value (soil acidity); obtaining geomorphic data of the target area, which includes digital elevation model (DEM), topographic maps, or remote sensing images, etc.; calculating geomorphic features based on the obtained data. For example, terrain elevation, slope, and aspect can be extracted from the DEM; geomorphic forms, such as ridges, valleys, plains, etc., can be identified through terrain analysis; the calculated geomorphic feature data are integrated to form a unified data set.

[0104] The key features and geomorphic features are integrated to form a data set, specifically including: merging the soil and geomorphic feature data extracted in the above steps to form a comprehensive data set; checking and cleaning outliers, missing values, or duplicate values in the data set to ensure the accuracy and integrity of the data; if the soil and geomorphic data are based on different sampling points or grids, they are associated to the same spatial resolution or position through spatial interpolation method.

[0105] Standardize the features in the dataset and cluster the sample points in the dataset through K-means clustering to identify groups of sample points with similar soil and geomorphic characteristics. Each group represents a similar area, specifically including: Standardize the features in the dataset, for example, using the z-score standardization method to eliminate the dimensional difference and numerical range difference between different features; Apply the K-means clustering algorithm to cluster the sample points in the dataset. First, determine the number of clusters K (representing the number of similar areas expected to be identified), then initialize the cluster centers, and continuously optimize the positions of the cluster centers through an iterative process until the convergence condition is reached; After clustering, each sample point will be assigned to the group where the nearest cluster center is located, and these groups represent areas with similar soil and geomorphic characteristics.

[0106] Judge the material type of each area according to the cluster centers to divide each similar area into different material types, including sandy soil areas, clay areas, and rock areas, specifically including: Analyze the characteristic values of the center points of each cluster, and judge the material type represented by this cluster according to these characteristic values (such as soil type, texture, organic matter content, and terrain elevation, slope, etc.), such as sandy soil areas, clay areas, or rock areas, etc.; According to the clustering results and the judgment of the material type, divide the similar areas into different material type areas to ensure that the soil and geomorphic characteristics within each area are relatively consistent.

[0107] Align and match the similar areas with similar soil and geomorphic characteristics with the 3D model, specifically including: Determine that the extracted soil and geomorphic data are aligned with the target 3D model in spatial coordinates. If necessary, perform spatial transformation of the data (such as translation, rotation, or scaling) to achieve precise alignment; Match the divided material areas with the corresponding areas of the 3D model, which can be achieved by mapping the boundaries of the material areas to the surface of the 3D model or through volume matching.

[0108] On the 3D model, create different material partitions according to the mapped boundaries of the material areas. Each partition should represent an area with unique soil and geomorphic characteristics and assign corresponding material attributes to it, specifically including: According to the matching results, map the boundaries of the material areas to the surface of the 3D model to form different partition boundaries. On the 3D model, create different material partitions according to the mapped boundaries. Each partition should represent an area with unique soil and geomorphic characteristics, and assign corresponding material attributes to each created material partition, such as color, texture, or other physical attributes, so as to clearly distinguish different material areas in visualization or subsequent analysis.

[0109] In the embodiments of the present invention, by integrating real soil and landform data into a three-dimensional model, the authenticity and accuracy of the model can be significantly improved. This enhanced authenticity helps to better understand and analyze the natural environment of a specific area. Through material zoning, decision-makers can more easily identify which areas are suitable for specific types of development or protection, thereby optimizing resource allocation and utilization. In environmental impact assessments or ecological studies, knowing the soil and landform conditions of different areas can greatly accelerate the assessment process. For example, when assessing the environmental impact of a construction project, it is possible to quickly determine which areas may be more affected, thus concentrating efforts on more detailed analysis.

[0110] In a preferred embodiment of the present invention, a particle optimization algorithm is used to determine the corresponding area division strategy, and the geographical and soil characteristics of each feature area are analyzed to extract key indicators, including:

[0111] Set the size of the particle swarm, that is, the number of particles; each particle represents an area division strategy and has two attributes: position and velocity; the position of the particle represents the parameters of the area division, and the velocity represents the moving direction and rate of the particle in the search space. Specifically, according to the complexity of the problem and the availability of computing resources, set a suitable size of the particle swarm, that is, the number of particles N, which will affect the exploration ability and computing efficiency of the algorithm; randomly generate N particles, each particle represents a possible area division strategy, and each particle has two attributes: position and velocity. The position represents the parameters of the area division (such as boundary coordinates, number of areas, etc.), and the velocity represents the moving direction and rate of the particle in the search space.

[0112] Construct a fitness function to evaluate the advantages and disadvantages of each area division strategy. Among them, the specific calculation formula of the fitness function is:

[0113]

[0114] Where, Max i and Min i are the maximum and minimum elevations of the i-th area respectively; n represents the total number of divided areas; σ s，i is the standard deviation of the slope within the i-th area; H i is the Shannon entropy of the i-th area; σ f，i is the standard deviation of the soil fertility of the i-th area; E i is the error rate of the i-th area division, which is calculated by comparing the differences between the predicted class and the actual class; w 1 、w 2 、w 3 、w 4 and w 5 are weight coefficients.

[0115] In each iteration, the fitness value of each particle is calculated according to the fitness function, and the individual position and global position of each particle are updated; according to the individual position and global position, the velocity and position of each particle are adjusted. When the preset number of iterations is reached, the corresponding particle is obtained; the region division strategy represented by the corresponding particle is used as the final strategy, specifically including: in each iteration, the fitness function is used to calculate the fitness value of each particle, and this value reflects the quality of the region division strategy represented by the particle; according to the fitness value of each particle, the individual best position and global best position of each particle are updated. The individual best position is the best position found by the particle during the search process, and the global best position is the best position found among all particles; according to the individual best position and global best position, the velocity and position of each particle are adjusted, which usually involves some parameters such as the inertia weight, individual learning factor, and social learning factor. These parameters are used to balance the global search ability and local search ability of the algorithm. When the preset number of iterations is reached or other termination conditions are met, the iteration stops. At this time, the particle corresponding to the global best position is the required optimal region division strategy. In the particle swarm optimization algorithm, the velocity and position of the particle are updated through the following formulas. Update velocity:

[0116]

[0117] where is the velocity of particle i in the (t + 1)-th round, w is the inertia weight, c 1 and c 2 are the learning factors; r 1 and r 2 are random numbers, pbest i is the historical best position of particle i, and gbest is the global best position. Update position:

[0118]

[0119] where is the position of particle i in the (t + 1)-th round, is the position of particle i in the t-th round.

[0120] According to the final strategy, the entire area is divided into several characteristic regions. For each characteristic region, soil analysis methods are used for analysis to extract key indicators; the key indicators include terrain elevation range, slope distribution, soil type and distribution, and soil fertility. Specifically, according to the position of the particle corresponding to the global optimal position (i.e., the optimal regional division strategy), the entire area is divided into several characteristic regions, and these characteristic regions should have similar geographical and soil characteristics; for each characteristic region, detailed geographical and soil characteristic analysis is carried out using soil analysis methods (such as laboratory tests, remote sensing image analysis, etc.), which includes the extraction of key indicators such as terrain elevation range, slope distribution, soil type and distribution, and soil fertility; through analysis, the key indicators of each characteristic region are extracted, and these indicators will be used for subsequent decision-making support, resource management, environmental protection, and other work. For example, a reasonable agricultural planting plan or land use plan can be formulated according to soil type and fertility conditions.

[0121] In the embodiment of the present invention, the particle swarm optimization algorithm has excellent global search ability and can quickly find an approximate optimal solution in a complex solution space, which helps to efficiently determine the optimal strategy in the regional division problem, reducing calculation time and resource consumption. Through the evaluation of the fitness function, the algorithm can adaptively adjust the speed and position of the particles, thereby continuously optimizing the regional division strategy. This self-adaptability enables the algorithm to flexibly divide according to the geographical and soil characteristics of different regions, improving the accuracy and pertinence of the division. Based on the finally determined regional division strategy, detailed geographical and soil characteristic analysis can be carried out for each characteristic region, which helps to accurately extract key indicators such as terrain elevation range, slope distribution, soil type and distribution, and soil fertility.

[0122] In a preferred embodiment of the present invention, key indicators are received, and the data within each characteristic region is evaluated through cluster analysis to identify potential problems in land use, including:

[0123] Receive key indicator data, including terrain elevation range, slope distribution, soil type and distribution, and soil fertility;

[0124] The key indicator data is divided into different clusters through the K-means clustering algorithm, and each cluster represents a region with similar characteristics. Specifically, it is necessary to determine the number of clusters (i.e., the K value), which can be determined by methods such as the elbow method or the silhouette score to determine the optimal number of clusters; using the determined K value, run the K-means clustering algorithm on the selected features, and the algorithm will divide the data into K clusters. The data points in each cluster have similarity on the selected features, and the clustering effect is evaluated by calculating clustering quality indicators (such as the silhouette coefficient).

[0125] Analyze the characteristics of each cluster, count the key indicators, and identify the land use problems existing in each cluster, including soil erosion risk, insufficient soil fertility, and topographic limitations. Specifically, for each cluster, analyze its key indicators (such as average elevation, slope, soil type distribution, and average soil fertility, etc.), which can be completed by calculating the statistics of each cluster (such as mean, median, standard deviation, etc.); based on the characteristic analysis of each cluster, identify potential land use problems. For example:

[0126] If a certain cluster has a low elevation and a steep slope, then there may be a high risk of soil erosion in this area; by comparing the soil fertility indicators of different clusters, it can be identified which areas may have insufficient soil fertility and require additional nutrient supplementation; for areas with extremely high elevation or extremely large slope, there may be topographic limitations, making these areas unsuitable for conventional agricultural production; display the clustering results and the identified problems through maps or other visualization tools to facilitate more intuitive understanding and interpretation. According to the identified problems, formulate specific land use improvement strategies or management suggestions for each cluster. For example, implement soil and water conservation measures in areas with high soil erosion risk, or apply more fertilizers in areas with insufficient soil fertility, etc.

[0127] In the embodiment of the present invention, through the cluster analysis of key indicators (such as topographic elevation, slope, soil type, and fertility), it is possible to more accurately identify areas with different characteristics. This helps to formulate more reasonable land use strategies according to the specific conditions of each area, thereby improving the overall land use efficiency. Cluster analysis can group data points with similar characteristics into one category, thus highlighting the potential land use problems in each category. For example, some clusters may show a high risk of soil erosion or insufficient soil fertility, which provides a clear problem position for decision-makers and facilitates the adoption of targeted measures. By analyzing the characteristics of different clusters, resources (such as fertilizers, water resources, etc.) can be more reasonably allocated to meet the specific land use needs of each area, which not only avoids waste of resources but also ensures that the land is properly maintained and managed.

[0128] The above is the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle described in the present invention, several improvements and refinements can still be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A surveying system for land assessment in natural resource engineering, characterized in that: The system comprises: A data collection module for obtaining topographic data, geomorphic data, and soil data of the land to be assessed; The regional modeling module is used to receive terrain data, landform data and soil data, optimize and select model parameters through genetic algorithms to obtain final model parameters; and use the final model parameters to construct a three-dimensional model of the measurement area; Finite element analysis module, used to obtain the three-dimensional model and divide it into multiple characteristic areas using the finite element analysis method; use the particle optimization algorithm to determine the corresponding regional division strategy, analyze the geographical and soil characteristics of each characteristic area, and extract key indicators; The data processing module is used to receive key indicators and evaluate the data in each characteristic area through cluster analysis to identify potential problems in land use.

2. A natural resource engineering land assessment measurement system according to claim 1, characterized in that: Receive terrain data, landform data and soil data, and optimize the model parameters through genetic algorithms to obtain the final model parameters, including: Determine the basic framework and model parameter range of the 3D model; Encode the parameters and convert them into the gene string form of the genetic algorithm; Randomly generate an initial population, each individual in the population represents a set of model parameters; Define a fitness function for evaluating the quality of each individual, i.e. each set of model parameters; select the corresponding individual in the population as the parent according to the fitness function; Randomly select two parent individuals, generate new offspring individuals through crossover operation, randomly change the genes of the offspring individuals, add the newly generated offspring individuals to the population to form a new population, repeat fitness evaluation, selection and genetic operations until the preset number of iterations is reached to obtain the final model parameters.

3. A natural resource engineering land assessment measurement system according to claim 2, characterized in that: The calculation formula of the fitness function F(x) is: Among them, w s Represents the weight coefficient of soil quality; S(x i ) represents the soil quality score of the i-th area; n represents the total number of sample areas; w f Represents the weight coefficient of soil fertility; F(x i ) represents the soil fertility score of the i-th region; w a The weight coefficient representing the accuracy of the area division; Indicates the category of the i-th region in the model division; C i represents the true category of the i-th region; represents the indicator function, which takes 1 when the model prediction is consistent with the true category, otherwise it takes 0; w e Represents the weight coefficient of the silhouette coefficient; S(i) represents the silhouette coefficient of the i-th region; i represents the index value.

4. A natural resource engineering land assessment measurement system according to claim 3, characterized in that: Obtain a 3D model and apply finite element analysis to divide it into multiple characteristic areas, including: Extract all vertex coordinates from the three-dimensional model to form a point set, triangulate the point set to generate a convex hull composed of tetrahedrons in three-dimensional space, and the circumscribed circle of each tetrahedron does not contain the points in the point set; screen the tetrahedrons to obtain a triangular mesh that approximates the surface of the original three-dimensional model; According to soil and landform data, different material areas are identified and mapped onto the 3D model to form different material partitions; For each material partition, the corresponding elastic modulus, Poisson's ratio and density properties are retrieved from the database, and the properties are assigned to the corresponding grid cells so that each grid cell has a three-dimensional model of the corresponding material properties; Determine the type of dynamic analysis and the corresponding analysis parameters; Calculate using finite element analysis algorithms, and extract key indicators from the results after the calculations are completed; Analyze key indicators and identify characteristic areas of stress concentration, severe deformation or obvious changes in soil properties; According to the characteristic area, the three-dimensional model is divided into multiple sub-areas, each of which has similar mechanical behavior or soil characteristics, and a unique identifier is assigned to each sub-area to obtain the three-dimensional model of multiple characteristic areas after the final division.

5. A natural resource engineering land assessment measurement system according to claim 4, characterized in that: Extract all vertex coordinates from the three-dimensional model to form a point set, and triangulate the point set to generate a convex hull composed of tetrahedrons in the three-dimensional space, and the circumscribed circle of each tetrahedron does not contain the points in the point set; The tetrahedrons are screened to obtain a triangular mesh that approximates the surface of the original 3D model, including: Extract all vertex coordinates in the three-dimensional model to form a point set, set the initial temperature, temperature drop rate, termination temperature and the number of iterations L at each temperature of the simulated annealing algorithm; At the current temperature, four points in the point set are randomly selected to form a tetrahedron; Calculate the first evaluation function value of the tetrahedron, and perform L iterations at the current temperature; In each iteration, a new tetrahedral configuration is randomly generated, and the second evaluation function value of the new configuration is calculated. If the second evaluation function value satisfies the Metropolis criterion, the new configuration is accepted, and the current final configuration is updated, and the temperature is lowered until the preset number of iterations is reached. After simulated annealing optimization, a set of final tetrahedral configurations is obtained; Tetrahedrons that approximate the surface of the original three-dimensional model are screened out from the final tetrahedral configuration to form a triangular mesh.

6. A natural resource engineering land assessment measurement system according to claim 5, characterized in that: Soil data include soil type, texture and organic matter content, and geomorphic data include terrain elevation, slope and aspect.

7. A natural resource engineering land assessment measurement system according to claim 6, characterized in that: Based on soil and landform data, different material areas are identified and mapped onto the 3D model to form different material partitions, including: Extract key features from soil data, including soil type, particle size distribution, soil texture, organic matter content, and pH; extract geomorphic features from geomorphic data, including terrain elevation, slope, aspect, and geomorphic morphology; Integrate key features and geomorphic characteristics to form a data set; The features in the data set were standardized, and the sample points in the data set were clustered by K-means clustering to identify groups of sample points with similar soil and geomorphic characteristics, each group representing a similar area; The material type of each area is determined according to the cluster center, so as to divide each similar area into different material types, including sand area, clay area and rock area; Align and match similar areas with similar soil and geomorphic properties to the 3D model; On the 3D model, different material partitions are created according to the mapped material area boundaries. Each partition should represent an area with unique soil and geomorphic properties and be assigned corresponding material properties.

8. A natural resource engineering land assessment measurement system according to claim 7, characterized in that: The particle optimization algorithm is used to determine the corresponding regional division strategy, analyze the geographical and soil characteristics of each characteristic area, and extract key indicators, including: Set the size of the particle swarm, that is, the number of particles; each particle represents a region division strategy and has two attributes: position and speed; the position of the particle represents the parameters of the region division, and the speed represents the moving direction and speed of the particle in the search space; Construct a fitness function to evaluate the pros and cons of each area division strategy; In each iteration, the fitness value of each particle is calculated according to the fitness function, and the individual position and global position of each particle are updated; the speed and position of each particle are adjusted according to the individual position and the global position, and when the preset number of iterations is reached, the corresponding particle is obtained; the area division strategy represented by the corresponding particle is used as the final strategy; According to the final strategy, the whole area is divided into several characteristic areas, and for each characteristic area, soil analysis methods are used to analyze to extract key indicators.

9. A natural resource engineering land assessment measurement system according to claim 8, characterized in that: Key indicators include terrain elevation range, slope distribution, soil type and distribution, and soil fertility.

10. A natural resource engineering land assessment measurement system according to claim 9, characterized in that: Receive key indicators and evaluate the data within each characteristic area through cluster analysis to identify potential problems in land use, including: Receive key indicator data, including terrain elevation range, slope distribution, soil type and distribution, and soil fertility; The key indicator data are divided into different clusters through the K-means clustering algorithm, and each cluster represents an area with similar characteristics; The characteristics of each cluster were analyzed, key indicators were counted, and land use problems in each cluster were identified, including soil erosion risks, insufficient soil fertility, and terrain restrictions.

Images