Three-dimensional data contrastive analysis system for pumped storage power station

Abstract

A three-dimensional data comparative analysis system of a pumped storage power station belongs to the technical field of three-dimensional geographic information systems and computer graphics, and comprises a data receiving unit used for receiving a reference elevation model and a target elevation model; the geometric stability analysis unit is used for calculating geometric stability measurement of a to-be-intersected triangle pair from the reference elevation model and the target elevation model; the analysis module is also used for comparing and analyzing the geometric stability measurement and a preset risk threshold value; when the geometric stability measure is smaller than a risk threshold value, a rapid intersection signal is generated; generating a robust calculation signal when the geometric stability measure is greater than or equal to a risk threshold; and the adaptive intersection unit is used for responding to the rapid intersection signal to call a rapid intersection algorithm, or responding to the robust calculation signal to execute coordinate perturbation and then perform intersection operation so as to generate an accurate intersection line, and the engineering design efficiency and the result reliability are remarkably improved.

Description

Technical Field

[0001] This invention relates to the fields of three-dimensional geographic information systems and computer graphics technology, specifically a three-dimensional data comparison and analysis system for pumped storage power stations. Background Technology

[0002] In the construction and management of large-scale engineering projects such as pumped storage power stations, it is necessary to accurately compare the terrain before and after construction to calculate the volume of excavation and filling, and to conduct detailed analyses of geological stability, hydrological impacts, etc. Existing technologies can overlay terrain data from two periods for display and perform simple volume estimation. However, when two complex digital elevation model surfaces exhibit geometric degradation such as tangent contact, small-angle intersection, or local overlap, the intersection algorithm based on standard floating-point operations is prone to numerical instability. This leads to broken intersection lines, self-intersections, or mismatches with the original models. These errors at the geometric calculation level can cause topological errors, resulting in the generated excavation and filling models containing voids, non-manifold edges, and lacking watertightness. Such models can only provide rough volume estimates and cannot be used for subsequent detailed analyses that rely on correct topology. Technicians must invest a significant amount of time in tedious manual repairs, which is inefficient and difficult to guarantee accuracy. Therefore, there is an urgent need for an algorithm that can automatically handle these problems and stably and efficiently generate watertight and topologically correct 3D solid models of excavation and filling.

[0003] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to provide a three-dimensional data comparison and analysis system for pumped storage power stations to solve the problems mentioned in the background art.

[0005] The technical solution of the present invention includes: a data receiving unit, used to receive a reference elevation model and a target elevation model;

[0006] The geometric stability analysis unit is used to calculate the geometric stability measure of the pairs of triangles to be intersected in the benchmark elevation model and the target elevation model; it is also used to compare and analyze the geometric stability measure with a preset risk threshold; when the geometric stability measure is less than the risk threshold, a fast intersection signal is generated; when the geometric stability measure is greater than or equal to the risk threshold, a robust calculation signal is generated.

[0007] The adaptive intersection unit is used to respond to the fast intersection signal to call the fast intersection algorithm, or to respond to the robust calculation signal to perform coordinate perturbation and then perform intersection operation to generate an accurate intersection line.

[0008] Model segmentation building blocks are used to segment and classify the baseline elevation model and the target elevation model based on the precise intersection line in order to build the initial solid model;

[0009] The topology repair unit is used to calculate the topology integrity metric of the initial entity model. When the topology integrity metric is greater than zero, it iteratively repairs the initial entity model until the topology integrity metric is zero, and finally outputs a watertight entity model.

[0010] Preferably, the geometric stability analysis unit is used for:

[0011] The first metric is calculated based on the dot product of the unit normal vectors of the pair of intersecting triangles and the shortest distance between the triangles, combined with the normalized scaling factor; the first metric is then determined as the geometric stability metric.

[0012] Preferably, the normalized scaling factor is defined as the average of the side lengths of all triangular faces of the baseline elevation model and the target elevation model within the current calculation area.

[0013] Preferably, the coordinate perturbation process is as follows:

[0014] Based on the unit normal vector of the triangle to be intersected and the average side length of the model, a perturbation vector is generated; the perturbation vector is applied to the vertices of the triangle to be intersected to remove the geometrically degenerate state.

[0015] Preferably, the model segmentation building unit is used for:

[0016] Using precise intersection lines as boundaries, the baseline elevation model and the target elevation model are divided to obtain multiple sub-grid regions;

[0017] Determine the position of the geometric center point of the sub-mesh region relative to another complete surface to classify the sub-mesh region into a top surface region and a bottom surface region; generate vertical sidewalls along the precise intersection line; stitch the triangular facet sets from the top surface region and the triangular facet sets from the bottom surface region with the vertical sidewalls to construct the initial solid model.

[0018] Preferably, the topology repair unit is used for:

[0019] Based on the total length of all boundary edges and the number of triangles connected to all non-manifold edges in the initial entity model, and combined with the preset penalty weights, a second metric is calculated; the second metric is then determined as the topology integrity metric.

[0020] Preferably, the boundary edge is an edge shared by only one triangular face; the non-manifold edge is an edge shared by two or more triangular faces.

[0021] Preferably, the specific repair process of the topology repair unit includes:

[0022] When the topology integrity metric indicates the existence of a hole formed by boundary edges, the constrained Delaunay triangulation algorithm is invoked to partition and fill the hole.

[0023] When the topological integrity metric indicates the presence of non-manifold edges, the manifold structure is restored by copying the non-manifold edges and their associated vertices and reconnecting the surrounding triangular faces.

[0024] This invention provides an improved three-dimensional data comparison and analysis system for pumped storage power stations, which, compared with existing technologies, has the following improvements and advantages:

[0025] 1. By setting up a geometric stability analysis unit, the potential numerical instability risk can be preemptively quantitatively assessed before the actual intersection operation is performed, eliminating the conditions for the formation of degradation state. No matter how complex the original terrain data is, the system can always generate a geometrically accurate, topologically continuous and non-self-intersecting precise intersection line. The quality of this intersection line is an absolute prerequisite for the success of all subsequent operations and cannot be reliably guaranteed by existing technologies.

[0026] 2. This ensures that the vertical sidewalls generated by subsequent stitching can be perfectly aligned with the top and bottom surfaces, so that the initial solid model constructed is geometrically very close to the real situation, avoiding geometric distortion caused by boundary errors;

[0027] 3. The originality of this invention lies in the active repair mechanism of its topology repair unit, which fully automates the model repair process that originally required a lot of manual intervention, greatly shortens the cycle of topographic comparison analysis for large-scale projects such as pumped storage power stations, and significantly improves the efficiency of engineering design and the reliability of results. Attached Figure Description

[0028] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0029] Figure 1 This is a flowchart of the system of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0031] Example 1:

[0032] Please see Figure 1 The present invention provides a three-dimensional data comparison and analysis system for pumped storage power stations, including: a data receiving unit for receiving a benchmark elevation model and a target elevation model;

[0033] The geometric stability analysis unit is used to calculate the geometric stability measure of the pairs of triangles to be intersected in the benchmark elevation model and the target elevation model; it is also used to compare and analyze the geometric stability measure with a preset risk threshold; when the geometric stability measure is less than the risk threshold, a fast intersection signal is generated; when the geometric stability measure is greater than or equal to the risk threshold, a robust calculation signal is generated.

[0034] The adaptive intersection unit is used to respond to the fast intersection signal to call the fast intersection algorithm, or to respond to the robust calculation signal to perform coordinate perturbation and then perform intersection operation to generate an accurate intersection line.

[0035] Model segmentation building blocks are used to segment and classify the baseline elevation model and the target elevation model based on the precise intersection line in order to build the initial solid model;

[0036] The topology repair unit is used to calculate the topology integrity metric of the initial entity model. When the topology integrity metric is greater than zero, it iteratively repairs the initial entity model until the topology integrity metric is zero, and finally outputs a watertight entity model.

[0037] In this embodiment, the system, through the coordinated operation of a series of functional units, aims to automatically generate a watertight three-dimensional solid model of excavation and filling for analysis of pumped storage power station projects. This process begins with the data receiving unit loading a triangular mesh model representing the terrain at two different stages before and after construction, i.e., a baseline elevation model. With target elevation model The geometric stability analysis unit performs a risk assessment on each pair of potentially intersecting triangles in the two models. Based on this assessment, the adaptive intersection unit is triggered to selectively execute the intersection strategy best suited to the current geometric configuration, ensuring the geometric accuracy of the calculated intersection line. This accurate intersection line forms the geometric benchmark for subsequent steps. The model segmentation and construction unit uses it as a cutting boundary to segment and classify the original terrain surface and construct the initial geometric shapes of the cut and fill entities. For the topological defects that may exist in the initial model due to complex stitching operations, the topology repair unit initiates a self-diagnosis and repair iterative cycle driven by quantitative indicators. This unit continuously quantifies the topological health of the model and specifically repairs all identified defects until the model reaches a completely watertight topologically correct state. This hierarchical processing flow from geometric accuracy to topological integrity systematically solves the inherent defects in existing technologies that lead to model damage due to unstable numerical calculations and require manual intervention, thereby achieving a high degree of automation and high reliability in large-scale engineering terrain comparison analysis.

[0038] Example 2:

[0039] The geometric stability analysis unit is used for:

[0040] The first metric is calculated based on the dot product of the unit normal vectors of the pair of intersecting triangles and the shortest distance between the triangles, combined with the normalized scaling factor; the first metric is then determined as the geometric stability metric.

[0041] The normalized scaling factor is defined as the average of the side lengths of all triangular faces in the current calculation area for the baseline elevation model and the target elevation model.

[0042] In this embodiment, the core function of the geometric stability analysis unit is to predict and quantify potential numerical instability risks before performing geometric intersection operations using a specifically constructed mathematical model; this unit provides each pair of data from the baseline elevation model. and target elevation model Given a triangle to be intersected, calculate a dimensionless measure of its geometric stability. The technical motivation for this metric is that existing technologies are prone to errors in standard floating-point operations when dealing with geometrically degenerate configurations where two triangular facets are nearly parallel or intersect at a small angle, resulting in broken intersection lines. This scheme aims to identify these configurations before calculation so as to enable specialized robust processing strategies.

[0043] This metric The mathematical model is defined as follows: ;

[0044] In this formula, It is a dimensionless measure of geometric stability; and The dimensionless weighting coefficients are determined through calibration on a benchmark dataset containing various known geometric degradation cases. This benchmark dataset preferably contains at least several thousand pairs of triangle samples, covering various known geometric degradation configurations such as coplanar, nearly coplanar, with an absolute value of the dot product of normal vectors greater than 0.999, extremely close vertices or edges, and small angles (less than 5 degrees), and intersections. It also includes a large number of randomly generated ordinary intersection configurations without degradation characteristics. The calibration process can employ an optimization algorithm, such as simulated annealing or gradient descent, whose objective function aims to maximize the geometric stability measure generated by all degradation configurations. The minimum value and the values ​​produced by all non-degenerate configurations The difference between the maximum values ​​is used to obtain the best discrimination, in order to balance the contributions of angle and distance to instability; and They are respectively from the model and The absolute value of the dot product of the unit normal vector of the triangle to be intersected is given. Approaching 1 indicates a degenerative pattern that tends towards parallelism; It is a function that represents the minimum value within the parentheses; Let be the distance from any vertex of the triangle in model DA to the plane containing the triangle in model DB, and The distance from any vertex of a triangle in model DB to the plane containing the triangle in model DA is given in the formula. Used to capture the spatial proximity between triangular faces, representing the minimum of all distances from any vertex of one triangle to the plane containing the other triangle; The normalization scaling factor is defined as the average of the side lengths of all triangular faces of the two models within the current computational region. It aims to eliminate the influence of model scale on the measurement results and ensure... The formula has universality; both sides are dimensionless pure scalars, satisfying the dimensionality consistency requirement.

[0045] At the application level, the geometric stability analysis unit will calculate the... Values ​​and risk thresholds A comparison was made; to ensure the objectivity and reproducibility of the threshold selection, this threshold... The determination is based on statistical principles, specifically defined as the output of all non-degenerate configurations in a benchmark dataset covering various geometric configurations. The upper quartile of the statistical distribution of values;

[0046] To ensure the representativeness of this statistical distribution, the benchmark dataset used here is similar in configuration type and quantity to that used for calibrating the weighting coefficients. and The dataset is kept consistent, thus ensuring the logical consistency and reliability within the risk assessment system;

[0047] like If the geometry is stable, the element generation generates a fast intersection signal; otherwise, if If the value is too high, the unit generates a robust calculation signal. This adaptive decision-making mechanism based on quantitative analysis provides a decisive basis for the subsequent adaptive intersection unit to select the correct processing path, thereby ensuring the stability and accuracy of the intersection line calculation from the root.

[0048] The advancement of this invention lies in its ability to pre-evaluate potential numerical instability risks by incorporating a geometric stability analysis unit before performing the actual intersection operation. Existing technologies typically directly call global intersection algorithms, which, when encountering geometrically degenerate conditions such as tangential contact, small-angle intersection, or local overlap between two digital elevation model surfaces, are prone to producing unpredictable and fragmented intersection lines due to the precision limitations of standard floating-point arithmetic. This invention introduces a geometric stability metric... ;in As a measure of geometric stability, For dimensionless weights, For unit normal vector, For the shortest distance set, The normalized scaling factor, mathematically speaking, integrates the two core factors leading to geometric degradation—angular parallelism and spatial proximity—into a unified, dimensionless risk index. This metric is not an abstract concept but a stably calculable quantity, its composition clearly defined in the specification, and can be directly implemented by those skilled in the art. Based on this metric, the system can drive the adaptive intersection unit to make a deterministic choice between a fast algorithm and a robust computation mode. The coordinate perturbation performed in the robust mode applies a small perturbation vector related to the model resolution to one of the vertices of the triangle face. This eliminates the conditions for the formation of a degraded state; the ultimate benefit is that no matter how complex the original terrain data is, the system can always generate a geometrically accurate, topologically continuous, and non-self-intersecting precise intersection line; the quality of this intersection line is an absolute prerequisite for the success of all subsequent operations, which cannot be reliably guaranteed by existing technologies.

[0049] Example 3:

[0050] The execution process of coordinate perturbation is as follows:

[0051] Based on the unit normal vector of the triangle to be intersected and the average side length of the model, a perturbation vector is generated; the perturbation vector is applied to the vertices of the triangle to be intersected to remove the geometrically degenerate state.

[0052] In response to a robust computation signal, the coordinate perturbation performed by the adaptive intersection unit is a deterministic geometric transformation process aimed at applying a perturbation vector with a defined direction and magnitude. To break the state of geometric degradation;

[0053] The disturbance vector The generation and application process is determined by a mathematical model. Defined; where, This is the perturbation vector applied to each vertex of the triangle to be processed; For example, a preset minimum normal quantity that is related to the precision of machine calculation. To ensure that the magnitude of the disturbance is small enough; The average side length, as mentioned above, represents the local feature scale of the model. This is the unit normal vector of the triangle itself; once generated, this perturbation vector is applied to all vertices of one of the triangles with a high numerical risk.

[0054] To ensure the determinism and reproducibility of the calculation process, when both triangular faces in the pair of intersecting triangles are identified as having high numerical risk, the system stipulates that the values ​​derived from the target elevation model must be used. The perturbation is applied to the triangular facets. This convention avoids randomness in selection and ensures that a unique and accurate intersection line can always be obtained under the same input.

[0055] Due to the direction of the disturbance Perpendicular to the triangular face, this operation is geometrically equivalent to translating the plane along its normal direction by a controllable, small distance. This displacement effectively eliminates the degenerate states such as coplanarity, tangency, or small-angle intersection between the two triangular faces. The technical effect is that it transforms a numerically unstable intersection problem into a numerically stable, conventional problem. Due to the relationship between the magnitude of the perturbation and the model resolution... and machine precision The correlation can be controlled to a negligible range in terms of the final cut and fill volume calculation, thus achieving the robustness of the intersection algorithm without sacrificing engineering accuracy;

[0056] The reason this disturbance is negligible in terms of engineering precision is because the magnitude of the vertex displacement it causes is... In a typical pumped storage power station terrain model, the average side length is... It might be on the order of meters, such as 0.5 meters, while For a very small constant with a precision comparable to that of machine calculation, such as Therefore, the actual displacement of the vertex is only at the sub-millimeter level, which is much smaller than the inherent error of topographic surveying and the model resolution. For a large model containing millions of triangular faces, the total volume calculation error caused by the accumulation of such tiny displacements is usually lower than the accuracy threshold required by the engineering, for example, one-thousandth. Therefore, its impact on the final engineering quantity calculation result can be safely ignored.

[0057] Example 4:

[0058] Model segmentation building blocks are used for:

[0059] Using precise intersection lines as boundaries, the baseline elevation model and the target elevation model are divided to obtain multiple sub-grid regions;

[0060] Determine the position of the geometric center point of the sub-mesh region relative to another complete surface to classify the sub-mesh region into a top surface region and a bottom surface region; generate vertical sidewalls along the precise intersection line; stitch the triangular facet sets from the top surface region and the triangular facet sets from the bottom surface region with the vertical sidewalls to construct the initial solid model.

[0061] After receiving the precise intersection line generated by the adaptive intersection unit, the model segmentation building unit performs a series of ordered geometric operations. Its core function is to construct a three-dimensional solid structure based on one-dimensional intersection line information.

[0062] This unit uses the precise intersection line as the cutting boundary to define the original baseline elevation model. and target elevation model The triangular mesh is used to divide the complete terrain surface into multiple independent sub-mesh regions. To determine the role of these sub-mesh regions in the final entity, the unit then performs a classification step. The implementation path is to calculate the geometric center point of each sub-mesh region and determine the spatial position of the point relative to another complete terrain surface, i.e., above or below.

[0063] To improve the robustness of the classification process, especially when dealing with sub-grid regions composed of narrow or irregularly shaped triangular facets, this system preferably adopts a more robust determination method: calculate the arithmetic mean of the spatial coordinates of all vertices in the sub-grid region and use this average point as the determination point. Alternatively, weighted random sampling can be performed in the sub-grid region based on the area to obtain multiple sampling points, and the classification of the region can be finally determined based on the positional relationship of the majority of sampling points relative to another complete surface, thereby effectively avoiding misjudgment caused by the offset of a single geometric center point.

[0064] Based on this positional relationship, all sub-regions are explicitly categorized as sets of triangular facets that constitute the top or bottom of a cut or fill volume. Taking the construction of a cut volume as an example, after classification, this unit will include all facets categorized as cut top faces originating from... The subgrid region, along with all those classified as cut bottom surfaces originating from... The sub-mesh regions are combined; to ensure solid closure, the cell is then aligned along the precise intersection line that serves as the boundary. and New triangular facets are generated between the corresponding boundary vertices to construct the vertical sidewalls connecting the top and bottom surfaces. This unit geometrically stitches together the set of top triangular facets, the set of bottom triangular facets, and the newly generated vertical sidewalls to form a closed initial solid model. The completion of this process provides a complete geometric object for subsequent topology verification and repair.

[0065] Based on this precise intersection line, the model segmentation building unit can accurately segment and classify the benchmark elevation model and the target elevation model, and construct the initial solid model. Since the intersection line, which serves as the segmentation boundary, is itself accurate, the boundaries of the segmented top and bottom regions are also precise. This directly ensures that the vertical sidewalls generated by subsequent stitching can be perfectly aligned with the top and bottom surfaces. As a result, the constructed initial solid model is geometrically very close to the real situation, avoiding geometric distortion caused by boundary errors.

[0066] Example 5:

[0067] Topology repair units are used for:

[0068] Based on the total length of all boundary edges and the number of triangles connected to all non-manifold edges in the initial entity model, and combined with the preset penalty weights, a second metric is calculated; the second metric is then determined as the topology integrity metric.

[0069] Boundary edges are edges shared by only one triangular face; non-manifold edges are edges shared by two or more triangular faces.

[0070] The specific repair process of the topology repair unit includes:

[0071] When the topology integrity metric indicates the existence of a hole formed by boundary edges, the constrained Delaunay triangulation algorithm is invoked to partition and fill the hole.

[0072] When the topological integrity metric indicates the presence of non-manifold edges, the manifold structure is restored by copying the non-manifold edges and their associated vertices and reconnecting the surrounding triangular faces.

[0073] As a crucial step in ensuring the quality of the final model, the topology repair unit aims to refine the initial entity model built in the previous step into a fully watertight and topologically correct final model. This unit achieves this goal through an iterative repair loop driven by quantitative metrics; the core driver is topology integrity measurement. The technical motivation behind this metric is to provide a computable index that quantifies the severity of topological defects in a model, representing two key topological concepts: watertightness and manifold, thereby enabling automated remediation; a non-zero value... The value directly indicates that the model has defects;

[0074] This metric The mathematical model is defined as follows: ;

[0075] In this formula, It is a dimensionless measure of topological integrity. It is the set of all boundary edges in the model, where a boundary edge is defined as an edge shared by only one triangular face, which forms the boundary of the hole in space; To share nonmanifold edges The number of triangular faces; For boundary edge Length; The average length of all edges in the model is used as a normalization factor to assign a higher penalty weight to larger holes; Let be the set of all non-manifold edges in the model, where a non-manifold edge is defined as an edge shared by two or more triangular faces. ; and The weights are dimensionless penalty weights, determined by calibration on a model library containing typical topological errors. This library includes a series of models with typical topological errors, generated through actual engineering scans or artificial construction, such as holes with multiple boundary loops, T-shaped connection structures composed of non-manifold edges, and non-manifold vertices where multiple triangular facets are pinched together at the same vertex. The calibration process aims to adjust the weights so that the repair program can prioritize defects that have a greater impact on the accuracy of subsequent volume calculations, such as large holes, or defects that are more likely to cause the repair algorithm to fail, such as complex non-manifold structures. This can be achieved, for example, by setting a weighted sum minimization objective function to adjust the priority of the repair strategy for different types of defects. Both sides of this formula are also dimensionless scalars, which conforms to the principle of dimensional consistency.

[0076] In applications, the topology repair unit continuously calculates in a loop. Value; as long as The repair process is then triggered; if If the first term is not zero, the system locates the boundary loop that forms the hole. And call the constrained Delaunay triangulation algorithm to perform region partitioning with the loop as the boundary, thereby filling the hole;

[0077] To ensure the geometric quality of the filled area and its smooth transition with the surrounding surface, the constrained Delaunay triangulation algorithm can be subjected to the following constraints during execution: the angle difference between the normal vector of each newly generated triangular facet and the average normal vector of the adjacent triangular facets on the hole boundary at the location of the triangular facet should be less than a preset threshold, such as 15 degrees; new triangular faces that do not meet this constraint will be rejected or adjusted to avoid producing sharp fill faces that do not conform to the natural terrain features.

[0078] If the second term is not zero, the system locates the specific non-manifold edge. By copying the edge and the vertex and reconnecting the surrounding triangles, the entangled topology is separated, and the manifold properties at that point are restored.

[0079] With a three-sided triangle Shared nonmanifold edges For example, the specific steps for repair may include: creating edges copy and its two endpoints copy Identify the face among the three faces that has the greatest geometric angular difference from the other two faces, for example... ;Will The topological connectivity relationship starts from the shared original edge Modified to share new edges ,Right now The vertex is no longer Instead Update all related The connection information of the relevant edges; through this operation, the original non-manifold structure is successfully separated into two or more independent edge structures that conform to the definition of a manifold, thereby systematically eliminating topological errors;

[0080] This repair process is performed iteratively, with recalculation after each iteration. Value, until Converging to zero; this convergence signifies that all topological defects in the model have been systematically eliminated, thus ensuring that the output solid model is a watertight final product that has undergone rigorous topological verification and repair and can be directly used for high-precision engineering analysis.

[0081] The originality of this invention lies in its active repair mechanism of topology repair units. Existing technologies often produce models with topological defects such as holes and non-manifold edges due to calculation errors. Such models lack watertightness and can only provide rough estimates of cut and fill volumes, making them unsuitable for subsequent refined analyses such as those relying on correct topology for hydrological impacts. Furthermore, technicians must invest significant time in tedious and inefficient manual repairs. This invention introduces a topology integrity metric. ;in For topological integrity measurement, For penalty weighting, For the set of boundary edges, The length of the boundary edge. For the average side length, Let be the set of edges of a nonmanifold. For the shared number of edges and faces; the practical significance of this formula lies in its deconstruction of the topological concept of watertightness into two quantifiable geometric indices: the total size of the holes formed by the boundary edges and the connectivity complexity caused by non-manifold edges; this formula is also explicitly defined in the specification, and its parameters can be directly calculated from the model; if and only if When the value is greater than zero, the repair process is triggered, and based on the terms in the formula that produce non-zero values, specific repair operators are invoked, such as constrained Delaunay triangulation for holes or vertex copying and reconnection for non-manifold edges; this iterative repair process continues until... Converging to zero ensures, in terms of algorithm, that the final output entity model is watertight. This closed-loop diagnosis-repair process transforms model quality from a problem requiring subjective judgment and manual intervention into an objectively measurable and automatically achievable engineering goal.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A three-dimensional data comparison and analysis system for pumped storage power stations, characterized in that, include: The data receiving unit is used to receive the baseline elevation model and the target elevation model. The geometric stability analysis unit is used to calculate the geometric stability measure of the pairs of intersecting triangles derived from the baseline elevation model and the target elevation model. It is also used to compare and analyze geometric stability metrics with preset risk thresholds; when the geometric stability metrics are less than the risk threshold, a fast intersection signal is generated; when the geometric stability metrics are greater than or equal to the risk threshold, a robust calculation signal is generated. The adaptive intersection unit is used to respond to the fast intersection signal to call the fast intersection algorithm, or to respond to the robust calculation signal to perform coordinate perturbation and then perform intersection operation to generate an accurate intersection line. Model segmentation building blocks are used to segment and classify the baseline elevation model and the target elevation model based on the precise intersection line in order to build the initial solid model; The topology repair unit is used to calculate the topology integrity metric of the initial entity model. When the topology integrity metric is greater than zero, it iteratively repairs the initial entity model until the topology integrity metric is zero, and finally outputs a watertight entity model.

2. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 1, characterized in that, The geometric stability analysis unit is used for: The first metric is calculated based on the dot product of the unit normal vectors of the pair of intersecting triangles and the shortest distance between the triangles, combined with the normalized scaling factor; the first metric is then determined as the geometric stability metric.

3. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 2, characterized in that, The normalized scaling factor is defined as the average of the side lengths of all triangular faces in the current calculation area for the baseline elevation model and the target elevation model.

4. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 1, characterized in that, The execution process of the coordinate perturbation is as follows: Based on the unit normal vector of the triangle to be intersected and the average side length of the model, a perturbation vector is generated; the perturbation vector is applied to the vertices of the triangle to be intersected to remove the geometrically degenerate state.

5. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 1, characterized in that, The model segmentation construction unit is used for: Using precise intersection lines as boundaries, the baseline elevation model and the target elevation model are divided to obtain multiple sub-grid regions; Determine the position of the geometric center point of the sub-mesh region relative to another complete surface to classify the sub-mesh region into a top surface region and a bottom surface region; generate vertical sidewalls along the precise intersection line; stitch the triangular facet sets from the top surface region and the triangular facet sets from the bottom surface region with the vertical sidewalls to construct the initial solid model.

6. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 1, characterized in that, The topology repair unit is used for: Based on the total length of all boundary edges and the number of triangles connected to all non-manifold edges in the initial entity model, and combined with the preset penalty weights, a second metric is calculated; the second metric is then determined as the topology integrity metric.

7. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 6, characterized in that, The boundary edge is an edge shared by only one triangular face; the non-manifold edge is an edge shared by two or more triangular faces.

8. The three-dimensional data comparison and analysis system for pumped storage power stations according to claim 1 or 6, characterized in that, The specific repair process of the topology repair unit includes: When the topology integrity metric indicates the existence of a hole formed by boundary edges, the constrained Delaunay triangulation algorithm is invoked to partition and fill the hole. When the topological integrity metric indicates the presence of non-manifold edges, the manifold structure is restored by copying the non-manifold edges and their associated vertices and reconnecting the surrounding triangular faces.

Images