Multi-constraint site selection method and system for spoil field

Abstract

The invention discloses a spoil site multi-constraint site selection method and system, and belongs to the technical field of construction management, and the method comprises the steps: carrying out the precise modeling of a terrain for analysis; candidate areas are obtained through screening driven by morphological constraints; performing three-dimensional form deduction on the candidate area based on layered filling dynamics; and performing multi-objective optimization under space conflicts in the three-dimensional form to obtain site selection information. The site selection efficiency is improved in order of magnitude, the labor and time cost is greatly reduced, the limitation of plane site selection is broken through, three-dimensional filling scheme output and multi-constraint condition full coverage are achieved, and the reliability and adaptability of the scheme are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of construction management technology, and in particular to a multi-constraint site selection method and system for spoil disposal sites. Background Technology

[0002] In recent years, the scale of infrastructure construction in mountainous areas has continued to expand. As a core component of earthwork disposal, the scientific selection of spoil disposal sites directly impacts project costs, construction efficiency, and regional ecological security, making it a critical issue urgently needing resolution in linear engineering projects (such as highways and railways) in mountainous areas. While existing technologies have made some progress in spoil disposal site selection, they still suffer from the following significant limitations:

[0003] Insufficient digitalization and automation: Some solutions rely on human experience to propose qualitative site selection principles (such as "trumpet-shaped area" and "body ratio"), without combining digital technologies such as BIM and GIS, and thus cannot achieve automated analysis and refined management of the site selection process;

[0004] Insufficient consideration of multidimensional constraints: Although some solutions combine GIS spatial analysis (such as buffer and overlay analysis) with BIM capacity measurement to achieve automated identification of potential areas, they only focus on depression extraction or single control elements, and do not fully integrate multidimensional constraints such as terrain conditions and engineering suitability, resulting in deviations between the site selection results and actual engineering needs;

[0005] Lack of a multi-factor evaluation system: Some schemes simplify site selection to "filling depression analysis" or remote sensing image screening, ignoring the differences in engineering geology of different depressions (such as foundation bearing capacity and topographic slope), lacking a systematic multi-factor evaluation system, and unable to effectively distinguish between "buildable" and "unbuildable" sites.

[0006] Lack of spatial morphology constraints: Some schemes have constructed a collaborative optimization model of "earthwork allocation-spoil disposal site selection" (such as using particle swarm optimization to optimize economic indicators such as transportation costs and site capacity), but the constraints only focus on economic and technical parameters and do not fully consider spatial constraints such as site morphology (such as fill height and slope gradient), making it difficult to adapt to complex mountainous terrain conditions.

[0007] There is a lack of simulation applications in the site selection stage: existing simulation technologies are mostly focused on stability assessment after the construction of spoil heaps. In the site selection decision stage, there is no integrated simulation model that integrates terrain constraints, layered filling dynamics, and environmental sensitivity analysis. This makes it impossible to dynamically simulate the filling process and potential risks of different site selection schemes, which restricts the practical promotion of intelligent site selection technology.

[0008] In summary, existing technologies have significant shortcomings in areas such as deep application of digital technology, multi-factor constraint modeling, spatial morphology control, and integration of site selection simulation. This results in the scientific rigor and reliability of spoil disposal site selection failing to meet the high requirements of mountainous engineering projects. To address these issues, there is an urgent need for a site selection method that integrates high-precision data, multi-dimensional constraints, and dynamic simulation to achieve an intelligent transformation of spoil disposal sites from "experience-based judgment" to "data-driven, simulation-verified" methods. Summary of the Invention

[0009] This invention proposes a multi-constraint site selection method and system for spoil disposal sites to provide suitable site selection schemes.

[0010] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0011] A multi-constraint site selection method for spoil disposal sites includes: performing accurate terrain modeling for analysis; obtaining candidate regions through morphological constraint-driven screening; performing three-dimensional morphological deduction of the candidate regions based on layered filling dynamics; and performing multi-objective optimization under spatial conflicts in the three-dimensional morphology to obtain site selection information.

[0012] Furthermore, the precise terrain modeling includes: using 3D point clouds as the data source, generating a continuous digital elevation model through rasterization and interpolation to fill holes; and further extracting a contour-preserving sequence of contour lines.

[0013] Furthermore, the three-dimensional point cloud is acquired based on airborne LiDAR and / or photogrammetry; wherein the grid resolution is 0.5~2m; the missing areas of the point cloud are filled based on Kriging interpolation to achieve the interpolation hole filling; the contour interval is 0.5~5m.

[0014] Furthermore, the process of obtaining candidate regions through morphological constraint-driven screening includes: setting quantitative constraints based on engineering site selection experience; using multi-level Boolean filtering of the quantitative constraints; and performing weighted comprehensive scoring on the filtering results for optimal selection.

[0015] Furthermore, the quantification constraints include:

[0016] Enclosure: ,in, This represents the minimum straight-line distance to the neck in the region. Its convex hull perimeter, This indicates that the entrance is narrow and easy to enclose;

[0017] Entrance width: ,in, , The two ends of the neck, The maximum distance threshold between the two endpoints. ;

[0018] Area size: , The measured area is used, and a minimum area threshold is set. To eliminate areas with insufficient capacity;

[0019] Convexity ratio: ,in, The convex hull of the candidate region is used to filter out broken polygons that are severely segmented by ridges.

[0020] Furthermore, the multi-level Boolean filtering of the quantization constraints includes:

[0021] Unify the indices corresponding to the quantitative constraints into a family of Boolean functions:

[0022] , ;

[0023] The corresponding candidate region must satisfy: To advance to the next round of selection.

[0024] Furthermore, the preferred comprehensive score includes:

[0025] For candidate sets that pass the hard constraints Establish a weighted scoring function:

[0026] , , , , where the weight vector , The maximum area of ​​the candidate spoil disposal site; the top 5% of the scored areas are retained, and the rest are automatically eliminated.

[0027] Furthermore, the three-dimensional morphological deduction of the candidate region based on layered stacking dynamics includes:

[0028] Filling parameter settings:

[0029] Let the local DEM after candidate region cropping be a raster field. ,in, For row and column index set, This represents the original ground elevation.

[0030] Introducing construction control parameter vectors Including single-layer compaction thickness Slope ratio and edge platform width ;

[0031] Using a hierarchical iterative algorithm:

[0032] The algorithm in the first layer( First, the benchmark elevation is calculated. To give the elevation of the top layer target, for any internal grid Defines the set of its surrounding 8 grid cells. ;

[0033] Domain Grid If it has been marked as a boundary, then its slope constraint elevation is... ,in, For the first Layer boundary point set;

[0034] Then construct the platform mask. ,like And it expands outwards If there are points with lower elevations within the distance buffer zone, then... Promoted to the 1st New layer boundary; otherwise, mark it as inside the platform and it will not participate in subsequent iterations;

[0035] when or The iteration terminates at time 1, the 2nd iteration. Layer volume ,in, Raster resolution;

[0036] Total capacity Introducing a minimum effective layer threshold If the actual number of iteration layers If the capacity of the area is insufficient, it will be excluded; otherwise, it will be excluded. As the final filling elevation site.

[0037] Furthermore, the multi-objective optimization under spatial conflict in three-dimensional morphology includes:

[0038] Spatial conflict topology modeling:

[0039] Let the set of candidate sites be , of which each It is a two-dimensional polygon;

[0040] Define binary intersection relation ,in, For the tolerance threshold, take ;

[0041] by Given an adjacency matrix, the disjoint-set data structure algorithm is used to... Divided into several non-overlapping conflict groups , The time complexity of this process is ;

[0042] Intragroup multi-objective scoring:

[0043] For each conflict group Construct a three-dimensional evaluation index vector: , , , ;

[0044] in, For normalized area, For convexity ratio, For environmentally sensitive distances, the exponential decay function is used. This refers to the distance between the site and a body of water or farmland. The attenuation coefficient;

[0045] Constructing a comprehensive score using a linear weighting method The weight vector is determined as follows ;

[0046] In each conflict group Inside, sorted in descending order of rating. ,reserve As the only site in this group, all other candidates were automatically marked as invalid.

[0047] Final output Simultaneously, it generates 3D bounding boxes, layered volumes, and boundary vectors for each site.

[0048] A multi-constraint site selection system for spoil disposal sites includes:

[0049] The first module is used for accurate terrain modeling for analysis.

[0050] The second module is used to obtain candidate regions through shape constraint-driven filtering;

[0051] The third module is used to perform three-dimensional morphological deduction of the candidate region based on the layered stacking dynamics.

[0052] The fourth module is used for multi-objective optimization under spatial conflicts in three-dimensional morphology to obtain site selection information.

[0053] By adopting the above technical solution, the present invention has the following beneficial effects:

[0054] 1. This invention performs precise terrain modeling for analysis; obtains candidate regions through morphological constraint-driven screening; performs three-dimensional morphological deduction of the candidate regions based on layered filling dynamics; and performs multi-objective optimization under spatial conflicts in the three-dimensional morphology to obtain site selection information. This achieves an order-of-magnitude improvement in site selection efficiency, significantly reduces labor and time costs, overcomes the limitations of planar site selection, enables the output of three-dimensional filling schemes, fully covers multiple constraints, and significantly enhances the reliability and adaptability of the schemes. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the multi-constraint site selection method for spoil disposal sites proposed in this invention;

[0056] Figure 2 This is a schematic diagram of the multi-constraint site selection process model for spoil disposal sites proposed in this invention. Detailed Implementation

[0057] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0058] like Figure 1 The method for multi-constraint site selection of spoil disposal sites includes: S1, performing accurate terrain modeling for analysis; S2, obtaining candidate areas through morphological constraint-driven screening; S3, performing three-dimensional morphological deduction of the candidate areas based on layered filling dynamics; and S4, performing multi-objective optimization under spatial conflicts in the three-dimensional morphology to obtain site selection information.

[0059] The process of accurate terrain modeling includes: using 3D point clouds as the data source, generating a continuous digital elevation model through rasterization and interpolation to fill holes; and further extracting a contour-preserving sequence of contour lines.

[0060] The three-dimensional point cloud is acquired using airborne LiDAR and / or photogrammetry; wherein the grid resolution is 0.5~2m; missing areas in the point cloud are filled using Kriging interpolation to achieve the interpolation hole filling; the contour interval is 0.5~5m.

[0061] The process of obtaining candidate regions through morphological constraint-driven screening includes: setting quantitative constraints based on engineering site selection experience; using multi-level Boolean filtering of the quantitative constraints; and performing weighted comprehensive scoring on the filtering results for optimal selection.

[0062] The quantization constraints include:

[0063] Enclosure: ,in, This represents the minimum straight-line distance to the neck in the region. Its convex hull perimeter, This indicates that the entrance is narrow and easy to enclose;

[0064] Entrance width: ,in, , The two ends of the neck, The maximum distance threshold between the two endpoints. ;

[0065] Area size: , The measured area is used, and a minimum area threshold is set. To eliminate areas with insufficient capacity;

[0066] Convexity ratio: ,in, The convex hull of the candidate region is used to filter out broken polygons that are severely segmented by ridges.

[0067] The quantization constraint conditions employing multi-level Boolean filtering include:

[0068] Unify the indices corresponding to the quantitative constraints into a family of Boolean functions:

[0069] , ;

[0070] The corresponding candidate region must satisfy: To advance to the next round of selection.

[0071] The preferred comprehensive score includes:

[0072] For candidate sets that pass the hard constraints Establish a weighted scoring function:

[0073] , , , ;

[0074] Among them, the weight vector , is the maximum area of ​​the spoil disposal sites in the candidate set; the top 5% of the scored areas are retained, and the rest are automatically eliminated, and j is the parameter in the summation operation.

[0075] The three-dimensional morphological deduction of the candidate region based on layered stacking dynamics includes:

[0076] Filling parameter settings:

[0077] Let the local DEM after candidate region cropping be a raster field. ,in, For row and column index set, This represents the original ground elevation.

[0078] Introducing construction control parameter vectors Including single-layer compaction thickness Slope ratio and edge platform width Hollow R means the set of real numbers;

[0079] Using a hierarchical iterative algorithm:

[0080] The algorithm in the first layer( First, the benchmark elevation is calculated. To give the elevation of the top layer target, for any internal grid Defines the set of its surrounding 8 grid cells. ;

[0081] Domain Grid If it has been marked as a boundary, then its slope constraint elevation is... ,in, For the first Layer boundary point set;

[0082] Then construct the platform mask. ,like And it expands outwards If there are points with lower elevations within the distance buffer zone, then... Promoted to the 1st New layer boundary; otherwise, mark it as inside the platform and it will not participate in subsequent iterations;

[0083] when or The iteration terminates at time 1, the 2nd iteration. Layer volume ,in, Raster resolution;

[0084] Total capacity Introducing a minimum effective layer threshold If the actual number of iteration layers If the capacity of the area is insufficient, it will be excluded; otherwise, it will be excluded. As the final filling elevation site.

[0085] The multi-objective optimization under spatial conflict in three-dimensional form includes:

[0086] Spatial conflict topology modeling:

[0087] Let the set of candidate sites be , of which each It is a two-dimensional polygon;

[0088] Define binary intersection relation ,in, For the tolerance threshold, take ;

[0089] by Given an adjacency matrix, the disjoint-set data structure algorithm is used to... Divided into several non-overlapping conflict groups , The time complexity of this process is ;

[0090] Intragroup multi-objective scoring:

[0091] For each conflict group Construct a three-dimensional evaluation index vector: , , , ;

[0092] in, For normalized area, For convexity ratio, For environmentally sensitive distances, the exponential decay function is used. This refers to the distance between the site and a body of water or farmland. The attenuation coefficient;

[0093] Constructing a comprehensive score using a linear weighting method The weight vector is determined as follows ;

[0094] In each conflict group Inside, sorted in descending order of rating. ,reserve As the only site in this group, all other candidates were automatically marked as invalid.

[0095] Final output Simultaneously, it generates 3D bounding boxes, layered volumes, and boundary vectors for each site.

[0096] A multi-constraint site selection system for spoil disposal sites includes:

[0097] The first module is used for accurate terrain modeling for analysis.

[0098] The second module is used to obtain candidate regions through shape constraint-driven filtering;

[0099] The third module is used to perform three-dimensional morphological deduction of the candidate region based on the layered stacking dynamics.

[0100] The fourth module is used for multi-objective optimization under spatial conflicts in three-dimensional morphology to obtain site selection information.

[0101] This invention addresses the shortcomings of existing spoil disposal site selection methods, such as reliance on manual experience, lack of quantifiable constraints, and the ability to only output planar locations. It provides a multi-constraint spoil disposal site selection method based on landfill simulation. Through a closed-loop technical system of "quantified engineering experience → intelligent screening → 3D simulation → conflict resolution," it achieves automated generation from the original point cloud to the optimal 3D site. The core technical features of this solution are as follows:

[0102] Overall technical solution concept

[0103] This invention uses high-precision point cloud data as input and employs four main stages: "precise terrain modeling → morphological constraint-driven candidate region selection → three-dimensional morphological deduction of layered filling → multi-objective optimization under spatial conflicts." It transforms engineering experience such as "low in the middle and high around the edges" and "large belly and small opening" into computable constraint functions. Combined with a layered filling dynamics model to predict the stability of the pile, it ultimately outputs a unique optimal site that meets the criteria of "capacity, stability, environmental friendliness, and accessibility." Figure 2 The model shown illustrates the multi-constraint site selection process for spoil disposal sites:

[0104] Accurate terrain modeling: 3D point cloud, DEM, contour lines;

[0105] Form constraint-driven intelligent candidate region selection: candidate regions are selected based on closure degree, area size, entrance width, and convexity ratio;

[0106] Three-dimensional morphological deduction of layered fill: parameters such as single-layer thickness, slope ratio, and edge platform width are used to derive the fill elevation field from the three-dimensional morphological deduction.

[0107] Multi-objective optimization decision-making: Grouping conflict groups (area, convexity ratio, and environmental water conservation) to select the best location and obtain the recommended site.

[0108] Specific technical solution details:

[0109] Accurate terrain modeling:

[0110] Using 3D point clouds (multi-source data such as airborne LiDAR and UAV photogrammetry) as the data source, a continuous digital elevation model (DEM) is generated through rasterization (raster resolution of 0.5~2m, matching point cloud density) and interpolation hole filling (Kriging interpolation to fill missing areas in the point cloud); further, a conformal contour sequence is extracted (contour interval of 0.5~5m, preserving terrain features), laying the foundation for subsequent geometric analysis.

[0111] Shape constraint-driven intelligent candidate region selection:

[0112] The experience of engineering site selection is transformed into quantitative constraints, and the candidate area is accurately locked through "multi-level Boolean filtering + comprehensive scoring".

[0113] Among them, the constraint conditions are quantified:

[0114] The degree of closure is determined by the relative closure coefficient. ,in, This represents the minimum straight-line distance to the neck in the region. Its convex hull perimeter, This indicates that the entrance is narrow and easy to enclose.

[0115] The width of the entrance is given by the two ends of the neck. , The entrance width is: ,Require To ensure that construction machinery can access the site. Among these, This is the maximum distance threshold between the two endpoints.

[0116] Area size, take it directly: And set a minimum area threshold. Remove areas with insufficient capacity.

[0117] Convexity ratio, defined as: ,Require This is to filter out broken polygons that are severely segmented by ridges; among which The convex hull of the candidate region.

[0118] Multi-level Boolean filtering:

[0119] Unify the above indicators into a family of Boolean functions: , ;

[0120] Candidate regions must meet the following requirements Only then can they proceed to the next round of selection.

[0121] This Boolean cascade effectively compresses the search space, reducing the computational complexity from... Down to ,in, This represents the number of contour line nodes.

[0122] Overall score preferred:

[0123] For candidate sets that pass the hard constraints Establish a weighted scoring function:

[0124]

[0125]

[0126]

[0127]

[0128] Where the weight vector Prioritizing capacity while also considering convexity and closure. The top 5% of regions in terms of score are retained, while the rest are automatically eliminated.

[0129] Three-dimensional morphological deduction of layered filling dynamics:

[0130] Using the local DEM of the candidate area as the base map, and coupling construction parameters with geotechnical mechanics experience, the three-dimensional fill morphology is deduced and stability is predicted through a "layer-iteration-convergence" algorithm.

[0131] Filling parameter settings:

[0132] Let the local DEM after candidate region cropping be a raster field. ,in For row and column index set, This represents the original ground elevation.

[0133] Introducing a construction control parameter vector: , respectively represent the compaction thickness of a single layer Slope ratio and edge platform width .

[0134] Hierarchical iterative algorithm:

[0135] The algorithm in the first layer( First, the benchmark elevation is calculated. The target elevation of the top of the layer is given.

[0136] For any internal grid Defines the set of its surrounding 8 grid cells. If the field grid If it has been marked as a boundary, then its slope constraint elevation is... ,in, For the first Layer boundary point set.

[0137] Then construct the platform mask.

[0138] like And its -If there is a lower elevation point within the domain, then Promoted to the 1st New layer boundary; otherwise, mark it as internal to the platform and it will not participate in subsequent iterations. or The iteration terminates at that time.

[0139] No. Layer volume ,in, This refers to the raster resolution.

[0140] Total capacity Introducing a minimum effective layer threshold , This is the maximum effective layer count threshold. If the actual number of iteration layers... If the capacity of the area is insufficient, it will be excluded; otherwise, it will be excluded. As the final filling elevation site.

[0141] Multi-objective optimization under spatial conflict:

[0142] For overlapping parts in the candidate region (such as multiple overlapping depressions), conflict resolution is achieved through "topology modeling + multi-objective scoring".

[0143] Spatial conflict topology modeling:

[0144] Let the set of candidate sites be , of which each Given a two-dimensional polygon, define a binary intersection relation. In the formula For the tolerance threshold, take .by Given an adjacency matrix, the disjoint-set data structure algorithm is used to... Divide into several non-overlapping conflict groups:

[0145]

[0146] The time complexity of this process is... .

[0147] Intragroup multi-objective scoring:

[0148] For each conflict group Construct a three-dimensional evaluation index vector:

[0149]

[0150]

[0151]

[0152]

[0153] in, For normalized area, the larger the area, the higher the spoil volume; The convexity ratio is the ratio of the convexity to the convexity; the closer it is to 1, the more regular the shape and the easier it is to construct. For environmentally sensitive distances, the exponential decay function is used. This refers to the distance between the site and a body of water or farmland. This is the attenuation coefficient.

[0154] Constructing a comprehensive score using a linear weighting method The weight vector is determined as follows: .

[0155] In each conflict group Inside, sorted in descending order of rating: ;reserve As the only site in this group, the other candidates were automatically marked as invalid.

[0156] Final output ;

[0157] Simultaneously, it generates 3D bounding boxes, layered volumes, and boundary vectors for each site.

[0158] This invention addresses the pain points of existing spoil disposal site selection methods, such as heavy reliance on manual labor, insufficient consideration of constraints, and inadequate refinement of schemes. Through an integrated technical system of "morphological constraints—layered filling—spatial conflict resolution," it achieves an intelligent transformation from "experience-based judgment" to "data-driven, simulation-verified" methods. Compared with existing technologies, it has the following significant, quantifiable, and verifiable advantages:

[0159] 1. Site selection efficiency is improved by orders of magnitude, significantly reducing labor and time costs.

[0160] Existing technologies rely on manual screening of candidate areas one by one, which is extremely inefficient when faced with large amounts of data and complex terrain (e.g., long-distance mountain route projects may require dozens of people / days to complete the initial screening). This invention transforms "engineering experience" into "calculable geometric indicators," employing multi-level Boolean filtering and weighted scoring to directly compress the candidate area to less than 5% of its original size, significantly reducing manual intervention. In case applications:

[0161] Case 1 (Route length 104km, 34.14 million point cloud data): The whole process took 8 person-days (2 people working together for 4 working days), which is 5 times more efficient than traditional manual methods;

[0162] Case 2 (Route length over 60km, 23.14 million point cloud data): The entire process can be completed with only 0.5 person-days, improving efficiency by 10 times.

[0163] 2. Overcoming the limitations of planar site selection, enabling the output of three-dimensional filling schemes.

[0164] Existing technologies can only output the "planar location range" of spoil heaps, failing to guide layered filling, slope control, and stability management during construction (requiring manual adjustments later, which can easily lead to safety risks). This invention, through layered iterative elevation equations and a filling dynamics model, achieves for the first time automated deduction "from point cloud data to three-dimensional filling morphology," outputting a scheme including sequence structure, inter-level platforms, slope geometric parameters, and layered volumes (such as key parameters like the "maximum number of stackable layers" and "volume / area of ​​each layer" for the spoil heap). This scheme directly solves the problems of construction workers "not knowing how to fill" (providing intuitive guidance for layered filling) and rework caused by "mismatched morphology" in traditional schemes.

[0165] 3. Comprehensive coverage of multiple constraints significantly enhances the reliability and adaptability of the solution.

[0166] Existing technologies focus only on single constraints such as "capacity and transportation costs," making it difficult to meet the high requirements of mountainous areas with "complex terrain and environmental sensitivity" (e.g., traditional solutions often select areas with "sufficient capacity but steep terrain and prone to landslides"). This invention integrates multiple dimensions of conditions, such as morphological constraints (e.g., slope ratio ≤ 1:1.5), stability constraints (e.g., single-layer fill height ≤ 10m), and environmental constraints (e.g., avoiding ecological red lines), and automatically resolves spatial overlaps through a disjoint-set data structure algorithm to ensure optimal solution.

[0167] In summary, this invention solves the core problems of "low efficiency, crude solutions, and high risk" in existing technologies through a closed-loop design of "data compression → morphological deduction → constraint satisfaction". Its results can be directly applied to various linear engineering projects in mountainous areas, promoting the transformation of spoil disposal site selection from "experience-driven" to "intelligent-driven", and has significant economic value (cost reduction), social value (environmental protection), and technological value (industry upgrading).

[0168] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit of the present invention should fall within the patent scope covered by the present invention.

Claims

1. A multi-constraint site selection method for spoil disposal sites, characterized in that, include: Accurate terrain modeling is performed for analysis. Candidate regions are obtained through morphological constraint-driven filtering; Three-dimensional morphological deduction of the candidate region is performed based on layered stacking dynamics; Multi-objective optimization under spatial conflict in three-dimensional form is performed to obtain site selection information.

2. The multi-constraint site selection method for spoil disposal sites according to claim 1, characterized in that, The process of performing accurate terrain modeling includes: Using 3D point clouds as the data source, a continuous digital elevation model is generated through rasterization and interpolation to fill holes. Further extract the shape-preserving contour line sequence.

3. The multi-constraint site selection method for spoil disposal sites according to claim 2, characterized in that, The 3D point cloud is acquired using airborne LiDAR and / or photogrammetry; wherein... The raster resolution is set to 0.5~2m; The missing areas in the point cloud are filled using the Kriging interpolation method to achieve the interpolation-based hole filling. The contour interval is 0.5~5m.

4. The multi-constraint site selection method for spoil disposal sites according to claim 3, characterized in that, The process of obtaining candidate regions through morphological constraint-driven filtering includes: Set quantitative constraints based on engineering site selection experience; The quantization constraints are obtained by employing multi-level Boolean filtering. The filtering results are optimized through a weighted comprehensive score.

5. The multi-constraint site selection method for spoil disposal sites according to claim 4, characterized in that, The quantization constraints include: Enclosure: ,in, This represents the minimum straight-line distance to the neck in the region. Its convex hull perimeter, This indicates that the entrance is narrow and easy to enclose; Entrance width: ,in, , The two ends of the neck, The maximum distance threshold between the two endpoints. ; Area size: , The measured area is used, and a minimum area threshold is set. To eliminate areas with insufficient capacity; Convexity ratio: ,in, The convex hull of the candidate region is used to filter out broken polygons that are severely segmented by ridges.

6. The multi-constraint site selection method for spoil disposal sites according to claim 5, characterized in that, The quantization constraint conditions employing multi-level Boolean filtering include: Unify the indices corresponding to the quantitative constraints into a family of Boolean functions: ， ； The corresponding candidate region must satisfy: To advance to the next round of selection.

7. The multi-constraint site selection method for spoil disposal sites according to claim 6, characterized in that, The preferred comprehensive score includes: For candidate sets that pass the hard constraints Establish a weighted scoring function: , , , , where the weight vector , The maximum area of ​​the candidate spoil disposal site; the top 5% of the scored areas are retained, and the rest are automatically eliminated.

8. The multi-constraint site selection method for spoil disposal sites according to claim 7, characterized in that, The three-dimensional morphological deduction of the candidate region based on layered stacking dynamics includes: Filling parameter settings: Let the local DEM after candidate region cropping be a raster field. ,in, For row and column index set, This represents the original ground elevation. Introducing construction control parameter vectors Including single-layer compaction thickness Slope ratio and edge platform width ; Using a hierarchical iterative algorithm: The algorithm in the first layer( First, the benchmark elevation is calculated. To give the elevation of the top layer target, for any internal grid Defines the set of its surrounding 8 grid cells. ; Domain Grid If it has been marked as a boundary, then its slope constraint elevation is... ,in, For the first Layer boundary point set; Then construct the platform mask. ,like And it expands outwards If there are points with lower elevations within the distance buffer zone, then... Promoted to the 1st New layer boundary; otherwise, mark it as inside the platform and it will not participate in subsequent iterations; when or The iteration terminates at time 1, the 2nd iteration. Layer volume ,in, Raster resolution; Total capacity Introducing a minimum effective layer threshold If the actual number of iteration layers If the capacity of the area is insufficient, it will be excluded; otherwise, it will be excluded. As the final filling elevation site.

9. The multi-constraint site selection method for spoil disposal sites according to claim 8, characterized in that, The multi-objective optimization under spatial conflict in three-dimensional form includes: Spatial conflict topology modeling: Let the set of candidate sites be , of which each It is a two-dimensional polygon; Define binary intersection relation ,in, For the tolerance threshold, take ; by Given an adjacency matrix, the disjoint-set data structure algorithm is used to... Divided into several non-overlapping conflict groups , The time complexity of this process is ; Intragroup multi-objective scoring: For each conflict group Construct a three-dimensional evaluation index vector: , , , ; in, For normalized area, For convexity ratio, For environmentally sensitive distances, the exponential decay function is used. This refers to the distance between the site and a body of water or farmland. The attenuation coefficient; Constructing a comprehensive score using a linear weighting method The weight vector is determined as follows ; In each conflict group Inside, sorted in descending order of rating. ,reserve As the only site in this group, all other candidates were automatically marked as invalid. Final output Simultaneously, it generates 3D bounding boxes, layered volumes, and boundary vectors for each site.

10. A multi-constraint site selection system for spoil disposal sites, characterized in that, include: The first module is used for accurate terrain modeling for analysis. The second module is used to obtain candidate regions through shape constraint-driven filtering; The third module is used to perform three-dimensional morphological deduction of the candidate region based on the layered stacking dynamics. The fourth module is used for multi-objective optimization under spatial conflicts in three-dimensional morphology to obtain site selection information.

Images