Digital modeling method and system for secondary screen cabinet terminal strip of transformer substation
By analyzing the topological structure and local neighborhood symmetry of triangular faces, edge confidence and compliance confidence are constructed to screen out unexpected holes and perform precise filling and repair. This solves the problem of distinguishing between unexpected holes and functional openings in the 3D model of secondary screen cabinets, and improves the geometric integrity and accuracy of the model.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-27
- Publication Date
- 2026-04-10
AI Technical Summary
Existing technologies cannot distinguish between accidental holes generated during the conversion of CAD drawings of secondary cabinets into 3D models and the inherent functional openings of the equipment itself, resulting in model distortion and reduced geometric integrity.
By analyzing the topological structure and local neighborhood symmetry of triangular faces, edge confidence and compliance confidence are constructed. Combined with a threshold segmentation algorithm, unexpected holes are screened out and accurately filled and repaired, preserving the original functional openings.
This approach achieves improved accuracy and robustness of the secondary cabinet's 3D model while preserving its geometric integrity and functional openings, providing a reliable geometric feature foundation for subsequent intelligent operation and maintenance.
Smart Images
Figure CN121837543A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of three-dimensional modeling technology, specifically to a digital modeling method and system for terminal blocks of secondary switchgear in substations. Background Technology
[0002] With the expansion of power grid scale and the transformation to intelligent operation and maintenance, the number of secondary equipment in substations has increased significantly. Terminal blocks in secondary cabinets, serving as connection hubs for these secondary equipment, are becoming increasingly complex in their wiring. Traditional operation and maintenance methods relying on static drawings and manual inspection are no longer sufficient to support efficient and accurate circuit management. Current technology typically involves importing CAD drawings of secondary cabinets into 3D modeling software to generate 3D models, and then conducting subsequent operation and maintenance analysis directly based on these models.
[0003] However, during the conversion of CAD drawings to 3D models, unexpected geometric holes are easily generated. Existing technologies repair the model by filling these holes, but there is a key drawback: it cannot distinguish between unexpected holes generated during the conversion and the inherent functional openings of the equipment itself. If original design openings such as ventilation openings and wiring ports are filled in together, it will damage the equipment structure, cause model distortion, and reduce the geometric integrity of the secondary cabinet modeling. Summary of the Invention
[0004] To address the aforementioned technical problems, the purpose of this application is to provide a digital modeling method and system for terminal blocks of secondary switchgear in substations. The specific technical solution adopted is as follows: In a first aspect, embodiments of this application provide a digital modeling method for terminal blocks of secondary switchgear in substations, the method comprising the following steps: In the 3D model, the number of triangular faces connected to each edge on each triangular face is counted to determine the minimum adjacency number and detection edge of each triangular face; by analyzing the symmetry of the distribution of triangular faces in the neighborhood of each triangular face, the asymmetry of each triangular face is determined, and combined with the minimum adjacency number, the detection triangular faces are screened from all triangular faces in the 3D model. For all detected triangular face clusters, the local extension consistency of each cluster is determined by analyzing the consistency of the direction of the detected edge and its two nearest neighbor edges on each detected triangular face in each cluster. The edge regularity index of each cluster is determined by combining the dispersion of the length of all edges on each detected triangular face. In all components of the 3D model, the component to which each cluster belongs is obtained. By judging the consistency between the component to which each cluster belongs and the category to which any other component belongs, feature clusters are selected from all clusters. Based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster, and the difference in edge regularity index between each cluster and any feature cluster, the compliance confidence of each cluster is determined. Based on the compliance confidence level, unexpected clusters are screened out from all clusters, and the corresponding regions of the unexpected clusters in the 3D model are filled with holes.
[0005] Preferably, the method for determining the minimum adjacency number of each triangular face and the detection edge is as follows: The minimum number of triangles connected to all edges on each triangle is recorded as the minimum adjacency number of each triangle, and the edge corresponding to the minimum adjacency number is recorded as the detection edge.
[0006] Preferably, the method for determining the asymmetry of each triangular facet is as follows: Construct a sphere with the centroid of each triangular face as the center and a preset value as the radius. Construct a plane passing through the center of the sphere, parallel to the detection edge and perpendicular to the triangular face, and divide the sphere into two parts. Calculate the absolute difference in the number of triangular faces in the two parts, and record it as the asymmetry of each triangular face.
[0007] Preferably, the step of selecting the detection triangles from all triangles of the 3D model includes: Calculate the sum of the minimum adjacency number of each triangular face and the preset parameter adjustment coefficient, and record the ratio of the asymmetry of each triangular face to the sum as the edge confidence of each triangular side. The edge confidence of all triangular faces in the 3D model of the secondary cabinet is used as the input of the threshold segmentation algorithm, and the segmentation threshold is output. Triangular faces with edge confidence greater than or equal to the segmentation threshold are recorded as detected triangular faces.
[0008] Preferably, the method for determining the local extension consistency of each cluster is as follows: The vector connecting the midpoint of any one of the two nearest neighbor edges on each detection triangle to the midpoint of the detection edge is denoted as the first vector. The vector connecting the midpoint of the detection edge on each detection triangle to the midpoint of the other nearest neighbor edge is denoted as the second vector. The cosine of the angle between the first and second vectors is calculated. The mean of the absolute values of the cosines of the angles between the detection edges on all detection triangles within each cluster is denoted as the local extension consistency of each cluster.
[0009] Preferably, the method for determining the edge regularity index of each cluster is as follows: The dispersion of the length of all sides on each detection triangle is denoted as the side length dispersion of each detection triangle. The mean of the side length dispersion of all detection triangles in each cluster is denoted as the discrete feature value of each cluster. The edge regularity index of each cluster is positively correlated with the local extension consistency and negatively correlated with the discrete eigenvalues.
[0010] Preferably, the step of selecting feature clusters from all clusters includes: In a 3D model, if the component to which the current cluster belongs is of the same type as any component in the 3D model, then the current cluster is denoted as a feature cluster.
[0011] Preferably, the expression for the compliance confidence score of each cluster is: In the formula, This represents the local extension consistency of the u-th cluster; Represents the edge regularity index of the u-th cluster; This represents the normalized value of the distance feature value of the u-th cluster, where the distance feature value is obtained based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster; This represents the minimum value among the differences in edge regularity indices between the u-th cluster and all feature clusters; The matching factor of the u-th cluster is obtained by judging the consistency between the component to which each cluster belongs and the category to which any other component belongs.
[0012] Preferably, the step of filtering out unexpected clusters from all clusters includes: In the 3D model, the compliance confidence scores of all clusters are used as input to the Otsu threshold segmentation algorithm. The output segmentation threshold is denoted as the compliance threshold. Clusters with compliance confidence scores less than or equal to the compliance threshold are denoted as unexpected clusters. Secondly, embodiments of this application also provide a digital modeling system for substation secondary cabinet terminal blocks, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described digital modeling methods for substation secondary cabinet terminal blocks.
[0013] This application has at least the following beneficial effects: This application constructs an edge confidence index by comprehensively analyzing the topological structure and local neighborhood symmetry of triangular faces. This index quantifies the degree of edge sharing and mesh distribution imbalance, and combined with a threshold segmentation algorithm, it can accurately and robustly screen out the detection triangular faces constituting holes and model boundaries from complex models, laying a reliable geometric feature foundation for subsequently distinguishing between original openings and accidental holes. Furthermore, this application divides the detection triangular faces into independent boundary structures through clustering, and comprehensively evaluates them by combining the continuity of boundary orientation and the geometric regularity of the triangular faces themselves. By constructing an edge regularity index, the degree of boundary distortion and deformation is effectively quantified, thereby accurately distinguishing between accidental holes caused by conversion errors and regular openings in the original design, providing a basis for subsequent intelligent repair. This application establishes key geometric morphology criteria. Furthermore, leveraging the standardized characteristics of secondary equipment, it finds similar references for each cluster through component category labels and constructs a compliance confidence level. This index comprehensively quantifies the positional reproducibility and geometric consistency of clusters within similar equipment, thereby accurately determining whether a boundary structure is a functional opening in the original design or an unexpected defect resulting from model conversion. This provides cross-component verification for the final decision. Finally, using the aforementioned compliance confidence level, this application precisely filters out unexpected clusters with low confidence through threshold segmentation. Furthermore, only these hole areas confirmed as unexpected defects are filled and repaired, thereby improving the geometric integrity of the secondary cabinet's 3D model while fully preserving the original functional openings. Attached Figure Description
[0014] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0015] Figure 1 A flowchart illustrating the steps of a digital modeling method for terminal blocks of a substation secondary panel cabinet, provided in one embodiment of this application; Figure 2 A flowchart illustrating the unexpected cluster screening process provided in one embodiment of this application. Detailed Implementation
[0016] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a digital modeling method and system for terminal blocks of a substation secondary switchgear according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0018] The following, in conjunction with the accompanying drawings, details the specific scheme of the digital modeling method and system for the terminal blocks of a substation secondary panel cabinet provided in this application.
[0019] Please see Figure 1 The document illustrates a flowchart of a digital modeling method for terminal blocks of a substation secondary cabinet according to an embodiment of this application. The method includes the following steps: Step S1: Obtain the 3D model of the substation secondary switchgear and perform triangular mesh generation on it.
[0020] In this embodiment, the CAD drawings of the secondary screen cabinet that needs to be 3D modeled are obtained from the database of the secondary screen cabinet modeling system. Then, the CAD drawings are used as input to the 3D modeling software Revit, thereby converting the CAD drawings into a 3D model. The mesh form of the 3D model is a triangular mesh, which is denoted as a triangular face. The centroid coordinates of each triangular face in the 3D model are obtained using the 3D modeling software Revit.
[0021] The process of using Revit software to obtain a 3D model of a secondary cabinet and then meshing it is a well-known technique and will not be elaborated further.
[0022] Step S2: In the 3D model, count the number of triangular faces connected to each edge on each triangular face to determine the minimum adjacency number and detection edge of each triangular face; by analyzing the symmetry of the distribution of triangular faces in the neighborhood of each triangular face, determine the asymmetry of each triangular face, and combine it with the minimum adjacency number to screen out the detection triangular faces from all triangular faces in the 3D model.
[0023] To accurately assess whether the holes in the 3D model belong to the original design holes of the secondary cabinet, it is first necessary to locate the hole area in the 3D model. According to the characteristics of the 3D model mesh division, the hole is a continuous missing area on the mesh surface, that is, there are no triangular facets in the hole area. However, since the hole area is surrounded by multiple triangular facets, the hole area can be initially located by analyzing whether each triangular facet is likely to be located outside the hole.
[0024] Based on the above analysis, this embodiment determines the minimum adjacency number and detection edge of each triangle by counting the number of triangles connected to each edge on each triangle; it also determines the asymmetry of each triangle by analyzing the symmetry of the triangle distribution in the neighborhood of each triangle, and combines this with the minimum adjacency number to filter out the detection triangles from all triangles in the 3D model. The specific process is as follows: In the 3D model of the secondary screen cabinet, for the triangular face located at the edge of the hole, at least one of its three edges is a non-shared edge, that is, the non-shared edge belongs only to the current triangular face and is not shared with other triangular faces.
[0025] Therefore, this embodiment counts the number of triangular faces connected to each edge on each triangular face to determine the minimum number of adjacencies for each triangular face and the detected edges. Specifically: Count the number of triangles connected to each edge of each triangle, and mark the edge with the fewest connected triangles as the detection edge. The number of triangles connected to the detection edge is also marked as the minimum adjacency number of each triangle. The minimum adjacency number reflects the probability that each triangle is surrounded by other triangles. The smaller the minimum adjacency number, the greater the probability that the triangle is located in a hole or on the edge of the model.
[0026] Furthermore, if the current triangular face is on the normal continuous surface of the secondary equipment in the secondary cabinet, the number of triangular faces in the neighborhood of the current triangular face is approximately evenly distributed on both sides of the current triangular face; conversely, if the detection edge of the current triangular face is a hole or model boundary, the distribution of the number of triangular faces in the neighborhood on both sides of the detection edge will be significantly unbalanced.
[0027] Therefore, based on the above analysis, this embodiment determines the asymmetry of each triangular face by analyzing the symmetry of the triangular face distribution within the neighborhood of each triangular face. Specifically: In this embodiment, a sphere with a preset radius R is constructed using the centroid of each triangular facet as its center. A plane parallel to the detection edge, passing through the center of the sphere, and perpendicular to the triangular facets is constructed within the sphere, thus dividing the sphere into two equal parts. Further, the absolute difference in the number of triangular facets within the two parts is calculated and denoted as the asymmetry of each triangular facet. The asymmetry reflects the degree of symmetry in the distribution of the number of triangular facets on both sides within the local neighborhood of each triangular facet. A greater asymmetry indicates a more asymmetrical mesh distribution on both sides of the triangular facet's neighborhood, thus increasing the likelihood that the triangular facet is located in a hole or at the edge of the model. It should be noted that a triangular facet is only counted when it is completely within the sphere.
[0028] It should be noted that in this embodiment, the radius R is set to 5 times the average length of the three sides of each triangular face. Setting the radius R to 5 times the average length of the triangular face is to construct an adaptive neighborhood that is proportional to the local grid scale, thereby achieving robust detection of hole boundaries. This value ensures that the analysis range can not only fully cover the neighboring triangular faces on both sides of the boundary, but also effectively limit it to the local area immediately adjacent to the boundary, avoiding the introduction of interference from distant unrelated structures due to an excessively large range, thus achieving a balance between the completeness of feature capture and the locality of analysis.
[0029] Furthermore, this embodiment determines the edge confidence of each triangular face based on the minimum adjacency number and the asymmetry, specifically: Calculate the normalized value of the minimum adjacency number of each triangular face and the sum of the preset parameter tuning coefficients, and record it as the ratio of the normalized value of the asymmetry of each triangular face to the normalized value of the sum, and record it as the edge confidence of each triangular edge.
[0030] In order to avoid the denominator being 0, the preset parameter adjustment coefficient is taken from the empirical range (0.005, 0.01). The value has little impact on the calculation and can be ignored. In this embodiment, 0.008 is used.
[0031] It should be noted that, to avoid the problem of one parameter being too large and causing other parameters to have a weak effect, the minimum adjacency number and asymmetry degree are normalized separately. In this embodiment, an exponential normalization function is used to normalize the minimum adjacency number and asymmetry degree to _____. In practical applications, as other implementation methods, implementers may also adopt other normalization methods such as the maximum and minimum value normalization method according to specific circumstances. This embodiment does not impose any special restrictions on the selection of normalization methods.
[0032] The process of normalizing data using an exponential normalization function is a well-known technique and will not be elaborated further.
[0033] Based on the edge confidence of each triangular face, we can understand that edge confidence reflects the tightness of each triangular face being surrounded by other triangular faces through the minimum adjacency number; it reflects the symmetry of the mesh distribution on both sides of each triangular face within its local neighborhood through asymmetry; and finally, it comprehensively reflects the probability of each triangular face being located at the boundary position by mapping positive and negative correlations. When the minimum adjacency number of the i-th triangular face is smaller and the asymmetry is larger, the edge confidence is larger, indicating that the i-th triangular face has both a low degree of edge sharing and a high degree of local distribution asymmetry, and therefore the confidence of being located at the edge of a hole is higher. Conversely, when the minimum adjacency number of the i-th triangular face is larger and the asymmetry is smaller, the edge confidence is smaller, indicating that the i-th triangular face has both a high degree of edge sharing and a low degree of local distribution asymmetry, and therefore the probability of being located on a continuous smooth surface inside the model is higher, that is, the confidence of being located at the edge of a hole is lower.
[0034] Furthermore, in this embodiment, the edge confidence of all triangular faces in the three-dimensional model of the secondary screen cabinet is used as the input of the threshold segmentation algorithm, and the segmentation threshold is output. Triangular faces with edge confidence greater than or equal to the segmentation threshold are recorded as detected triangular faces, thereby realizing the localization of triangular faces in the hole and the boundary area of the model.
[0035] It should be noted that there are many commonly used threshold segmentation algorithms. In this embodiment, the Otsu threshold segmentation algorithm is used to divide the triangular face. In practical applications, as other implementation methods, implementers may also use other threshold segmentation algorithms according to specific circumstances. This embodiment does not impose any special restrictions on the selection of threshold segmentation algorithms.
[0036] The process of obtaining the segmentation threshold using the Otsu threshold segmentation algorithm is a well-known technique and will not be described in detail here.
[0037] Thus, this embodiment constructs an edge confidence index by comprehensively analyzing the topological structure and local neighborhood symmetry of the triangular facets. This index, by quantifying the degree of edge sharing and the imbalance of grid distribution, and combined with a threshold segmentation algorithm, can accurately and robustly screen out the detection triangular faces that constitute holes and model boundaries from complex models, laying a reliable geometric feature foundation for subsequently distinguishing between original openings and accidental holes.
[0038] Step S3: For all detected triangular face clusters, the consistency of the direction of each detected edge and its two nearest neighbor edges on each detected triangular face in each cluster is analyzed to determine the local extension consistency of each cluster. Combined with the dispersion of the length of all edges on each detected triangular face, the edge regularity index of each cluster is determined.
[0039] Furthermore, during the conversion of CAD drawings into 3D models, holes accidentally generated due to data recognition errors, layer breaks, or improper conversion tolerance settings typically exhibit irregular changes such as local distortion and violent oscillations in their extension direction at their edges. In contrast, functional openings in the original design of secondary cabinets, such as ventilation openings, wiring ports, heat dissipation vents, and the outline edges of secondary cabinets, are constrained by engineering drawing standards, and their edges usually exhibit smooth boundaries and stable extension direction, characteristic of regular changes. Therefore, the consistency of the edge orientation of triangular faces and their neighboring triangular faces can be analyzed to further determine whether the holes are accidental holes generated during model conversion. Based on the above analysis, this embodiment clusters all detected triangular faces. By analyzing the consistency of the directional orientation between the detected edges and their two nearest neighbor edges on each detected triangular face in each cluster, the local extension consistency of each cluster is determined. Combined with the dispersion of the lengths of all edges on each detected triangular face, the edge regularity index of each cluster is determined. The specific process is as follows: First, cluster all detected triangles. Use all detected triangles as input to the clustering algorithm. Set the neighborhood radius to the mean of all side lengths of all detected triangles, set the minimum number of points to 3, set the metric distance to the distance between the centroids of the detected triangles, and use the elbow rule to determine the number of clusters. Finally, output all clusters.
[0040] It should be noted that in the DBSCAN clustering algorithm, the minimum number of points is set to 3 to ensure that each cluster can form at least one basic and meaningful boundary segment, rather than being formed by a single or two discrete noise points. The boundary of a hole should at least appear as a continuous line segment or loop in space, which requires at least three detection triangles to form a stable topological structure in space. Therefore, it can effectively filter out isolated noise points caused by model errors, and also ensure the effective identification of such continuous structures as the boundary of a real hole.
[0041] It should be noted that there are many commonly used clustering algorithms. In this embodiment, the DBSCAN density clustering algorithm is used to cluster the detected triangular faces. In practical applications, as other implementation methods, implementers may also use other clustering methods such as the DPC density peak clustering algorithm according to specific circumstances. This embodiment does not impose any special restrictions on the selection of clustering algorithms.
[0042] The process of using the DBSCAN density clustering algorithm to detect triangular face clusters, the calculation of Euclidean distance, and the determination of the number of clusters using the elbow rule are all well-known techniques and will not be elaborated further.
[0043] Since the direction of the detection edges of the triangular faces in the original design holes and model edge areas is relatively continuous and known, the direction of the detection edges of the triangular faces in the unexpected hole areas changes abruptly and frequently.
[0044] Therefore, based on the above analysis, this embodiment determines the local extension consistency of each cluster by analyzing the consistency of the directional orientation between the detected edge and its two nearest neighbor edges on each detected triangle face in each cluster. Specifically: Calculate the distance from the center of the detection edge on each detection triangle in each cluster to the midpoint of the detection edge on the other detection surfaces. Record the detection edges corresponding to the smallest and second smallest distances as the nearest neighbors of the detection edges on each detection triangle.
[0045] It should be noted that there are many commonly used methods for calculating the distance between points. In this embodiment, the Euclidean distance from the midpoint of the detection edge on each detection triangle in each cluster to the midpoint of the detection edge on the other detection surfaces is used as the distance between the midpoint of the detection edge on each detection triangle in each cluster to the midpoint of the detection edge on the other detection surfaces. In practical applications, as other implementation methods, implementers may also use other methods for calculating the distance between points, such as Manhattan distance, depending on the specific circumstances. This embodiment does not impose any special restrictions.
[0046] Furthermore, the vector connecting the midpoint of any one of the two nearest neighboring edges on each detection triangle to the midpoint of the detection edge is denoted as the first vector, and the vector connecting the midpoint of the detection edge on each detection triangle to the midpoint of the other nearest neighboring edge is denoted as the second vector. The cosine of the angle between the first and second vectors is calculated, and the mean of the absolute values of the cosine of the angle between the detection edges on all detection triangles within each cluster is denoted as the local extension consistency of each cluster. The greater the local extension consistency, the more consistent the orientation of the detection edge on the triangle with its nearest neighboring edge, and the more it conforms to the regular geometric characteristics of the original design hole or model edge.
[0047] From the component method of the first vector and the second vector, it can be seen that the range of the angle between the first vector and the second vector is within a certain range. .
[0048] Furthermore, since accidental holes are often caused by problems such as CAD conversion distortion, layer breakage, or abnormal mesh division, their edge triangular faces often exhibit twisted, stretched, or distorted characteristics, which in turn makes the difference in the length of the three sides of the triangular face quite significant. Conversely, since the holes and model edges in the original design are modeled in a standardized manner, the length of the three sides of the triangular face will be relatively more uniform.
[0049] Based on the above analysis, this embodiment determines the edge regularity index of each cluster by combining the local extension consistency of each cluster with the dispersion of the length of all edges on each detection triangle. Specifically, the dispersion of the length of the three edges on each detection triangle is calculated and denoted as the edge length dispersion of each detection triangle. The method for calculating the dispersion is not limited to variance, standard deviation, and coefficient of variation. In this embodiment, the variance of the length of the three edges on each detection triangle is used as the dispersion of the length of the triangle edges on each detection triangle.
[0050] Furthermore, the mean of the side length dispersion of all detected triangular faces in each cluster is denoted as the discrete feature value of each cluster. The discrete feature value can reflect the geometric distortion of the side length distribution of triangular faces in each cluster. The larger the discrete feature value, the greater the possibility that the region where the cluster is located is an unexpected hole caused by the corresponding transformation anomaly.
[0051] Since the calculation results of discrete eigenvalues will have physical dimensions, in order to eliminate the influence of dimensions, the discrete eigenvalues of all clusters are used as inputs to the maximum-minimum normalization method to normalize the discrete eigenvalues and obtain the normalized results of the discrete eigenvalues. In practical applications, as other implementation methods, implementers can also adopt other normalization methods according to specific circumstances. This embodiment does not impose any special restrictions.
[0052] The process of normalizing data using the maximum-minimum normalization method is a well-known technique and will not be elaborated further.
[0053] Furthermore, based on the local extension consistency and the discrete eigenvalues, the edge regularity index of each cluster is determined, specifically: The edge regularity index of each cluster is positively correlated with the local extension consistency and negatively correlated with the discrete eigenvalues.
[0054] It should be understood that a positive correlation means that the dependent variable increases as the independent variable increases, and the dependent variable decreases as the independent variable decreases. The specific relationship can be additive or multiplicative, etc., and is determined by the actual application. This application does not impose any special restrictions. A negative correlation means that the dependent variable decreases as the independent variable increases, and the dependent variable increases as the independent variable decreases. The relationship can be subtractive or divisive, etc., and is determined by the actual application.
[0055] Preferably, as one implementation, in this embodiment, the edge regularity index of the u-th cluster is... The expression is: In the formula, This represents the local extension consistency of the u-th cluster; is the normalized value of the discrete feature value of the u-th cluster; This indicates a constant that is pre-defined as being greater than 0. The value is set manually, in this embodiment. The value of is 0.01. Provided that the denominator is not zero and does not excessively affect the calculation result, the implementer may also set it according to the specific situation. This embodiment does not impose any special restrictions.
[0056] Based on the edge regularity index of each cluster, it can be understood that the edge regularity index reflects the degree of geometric distortion of the detected triangular facets within the cluster through discrete eigenvalues, reflects the local continuity of the boundary direction through local extension consistency, and finally comprehensively reflects the probability that the region corresponding to each cluster belongs to an unexpected hole through positive and negative correlation mapping. If the local extension consistency of the u-th cluster is smaller and the discrete eigenvalue is larger, the corresponding edge regularity index is smaller, indicating that the boundary structure corresponding to the u-th cluster is more likely to be an unexpected hole. Conversely, if the local extension consistency of the u-th cluster is larger and the discrete eigenvalue is smaller, the corresponding edge regularity index is larger, indicating that the geometric shape of the boundary structure corresponding to the u-th cluster is more regular and the direction is more continuous, and it is more likely to be an original design opening or model edge.
[0057] Thus, this embodiment divides the detected triangular facets into independent boundary structures through clustering, and comprehensively evaluates them by combining the continuity of the boundary orientation with the geometric regularity of the triangular facets themselves; by constructing an edge regularity index, the degree of distortion and deformation of the boundary is effectively quantified, thereby accurately distinguishing between unexpected holes caused by conversion errors and regular openings in the original design, providing key geometric morphological criteria for subsequent intelligent repair.
[0058] Step S4: Among all components of the 3D model, obtain the component to which each cluster belongs. By judging the consistency between the component to which each cluster belongs and the category to which any other component belongs, feature clusters are selected from all clusters. Based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster, and the difference in edge regularity index between each cluster and any feature cluster, determine the compliance confidence of each cluster.
[0059] Furthermore, since the secondary equipment in the secondary cabinet has standardized features, the structural design of equipment of the same model is strictly consistent. Therefore, the location and number of functional openings of the secondary equipment are highly reproducible. At the same time, since various equipment in the CAD drawings have equipment type labels, when converted into a 3D model, the equipment type labels and related attributes are usually retained with the model conversion. Therefore, by analyzing the location of the holes in the same type of equipment, we can further analyze the possibility that the holes are original design holes.
[0060] Based on the above analysis, this embodiment obtains the component to which each cluster belongs from all components of the 3D model. By judging the consistency between the component to which each cluster belongs and the category to which any other component belongs, feature clusters are selected from all clusters. Based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster, and the difference in edge regularity index between each cluster and any feature cluster, the compliance confidence of each cluster is determined. The specific process is as follows: First, obtain the component to which each cluster belongs, specifically: Obtain the centroid coordinates of all triangular faces contained in the region where each component is located in the 3D model of the secondary cabinet. Calculate the intersection-union ratio (IUR) of the centroid coordinates of all detected triangular faces in each cluster with the centroid coordinates of all triangular faces contained in the region where each component is located. The component with the largest IUR is recorded as the component belonging to each cluster. The method for calculating the IUR is a well-known technique, and its specific calculation process will not be elaborated here.
[0061] Furthermore, the category labels of the components belonging to each cluster in the 3D model are obtained, and the component matching factor is calculated. Any component in the 3D model that is completely consistent with the category label data of its belonging component is matched. If the category label of the current cluster's belonging component is consistent with the category label of any component in the 3D model, then the current cluster's belonging component is considered to belong to the same type as any component in the 3D model. The matching factor of the current cluster's belonging component is then set to 1, and the corresponding cluster is recorded as a feature cluster. If no component that is completely consistent with the category label of each cluster's belonging component is detected in the 3D model, then the matching factor of each cluster's belonging component is set to 0. In this embodiment, the component's category label is the component's Name attribute data, obtained through the 3D modeling software Revit. The component's category label matching method is a brute-force matching algorithm. In practical applications, as other implementation methods, implementers can also use other matching methods such as Jaccard similarity to perform matching, depending on the specific circumstances. This embodiment does not impose any special restrictions.
[0062] The process of obtaining component category labels using the 3D modeling software Revit, and the process of matching category labels using a brute-force matching algorithm, are well-known technologies and will not be elaborated further.
[0063] Furthermore, in this embodiment, the distance feature value of each cluster is determined based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster. Specifically: Calculate the mean of the centroid coordinates of all detected triangular faces within each cluster, and record it as the average centroid coordinate of each cluster. Calculate the distance from the average centroid coordinate to the centroid coordinate of the associated component, and record it as the centroid distance of each cluster. The centroid distance can reflect the distance difference between the boundary structure corresponding to each cluster and the component. Here, averaging the coordinates means averaging the same coordinate values.
[0064] Furthermore, the absolute difference in centroid distance between each cluster and any feature cluster is calculated. The minimum absolute difference is recorded as the distance feature value of each cluster. The distance feature value can reflect the position reproducibility of the boundary structure corresponding to each cluster in similar devices. The smaller the distance feature value, the more consistent the position of the boundary structure corresponding to the cluster in similar devices.
[0065] Furthermore, this embodiment determines the geometric eigenvalues of each cluster based on the difference in edge regularity index between each cluster and any feature cluster. Specifically: The difference between the edge regularity index of each cluster and the edge regularity index of any feature cluster is calculated. The minimum difference is denoted as the geometric feature value of each cluster for ease of description in this embodiment. The geometric feature value can reflect the relative degree of anomaly of the boundary structure of each cluster in the same type of equipment. The smaller the geometric feature value, the more consistent the geometric shape of the boundary structure of the cluster is with the equipment.
[0066] It should be noted that there are many methods to measure the difference between data. In this embodiment, the absolute difference between the edge regularity index of each cluster and the edge regularity index of any feature cluster is taken as the difference between the edge regularity index of each cluster and the edge regularity index of any feature cluster. In practical applications, as other implementation methods, implementers may also use other methods such as the square of the difference to measure the difference between data, depending on the specific circumstances. This embodiment does not impose any special restrictions on the selection of methods for measuring the difference between data.
[0067] Furthermore, based on the distance feature value and the geometric feature value, the compliance confidence level of the boundary structure corresponding to each cluster is determined, specifically: As one implementation method, in this embodiment, the compliance confidence score of the boundary structure corresponding to the u-th cluster is... The expression is: In the formula, Represents the edge regularity index of the u-th cluster; This represents the normalized value of the distance feature value of the u-th cluster; This represents the normalized value of the geometric eigenvalue of the u-th cluster. Represents the matching factor of the u-th cluster; This indicates a preset constant greater than 0, used to prevent the denominator from being 0. The value is set manually, in this embodiment. The value of is 0.01. Provided that the denominator is not zero and does not excessively affect the calculation result, the implementer may also set it according to the specific situation. This embodiment does not impose any special restrictions.
[0068] Based on the compliance confidence score of the boundary structure corresponding to each cluster, it can be understood that the compliance confidence score reflects the positional reproducibility stability of the boundary structure corresponding to each cluster in similar devices through distance feature values, and reflects the geometric consistency of the boundary structure corresponding to each cluster in similar devices through geometric feature values. Finally, by making a ratio, it comprehensively reflects the credibility of the boundary structure corresponding to each cluster as belonging to the original design features. If the edge regularity index of the u-th cluster is larger, while the distance feature value and geometric feature value are smaller, the compliance confidence score is larger, indicating that the boundary structure corresponding to the u-th cluster conforms more to the common features of similar devices in terms of geometric shape and spatial layout, and the possibility that there is a functional opening in the original design is greater. Conversely, if the edge regularity index of the u-th cluster is smaller, while the distance feature value and geometric feature value are larger, the compliance confidence score is smaller, indicating that the boundary structure corresponding to the u-th cluster deviates more from the common features of similar devices in terms of geometric shape and spatial layout, and the possibility that it is an unexpected hole generated during the model conversion process is greater.
[0069] Thus, this embodiment utilizes the standardized characteristics of secondary equipment to find similar references for each cluster through component category labels and constructs a compliance confidence score. This index comprehensively quantifies the position reproducibility and geometric consistency of clusters in similar equipment, thereby accurately determining whether a boundary structure is a functional opening in the original design or an unexpected defect caused by model conversion, providing a strong cross-component verification basis for the final decision.
[0070] Step S5: Based on the compliance confidence level, select the unexpected clusters from all clusters and fill the holes in the corresponding areas of the unexpected clusters in the 3D model.
[0071] Because unexpected holes generated due to conversion anomalies have poor boundary regularity, poor reproducibility of similar devices' positions and geometric shapes, their compliance confidence will be significantly low. Therefore, the compliance confidence of the boundary structure corresponding to each cluster is calculated, and the compliance confidence of all clusters is used as the input of the Otsu threshold segmentation algorithm. The output segmentation threshold is denoted as the compliance threshold. Clusters with compliance confidence less than or equal to the compliance threshold are denoted as unexpected clusters, thereby achieving accurate identification of holes unexpectedly generated during the conversion process.
[0072] For the hole areas corresponding to each unexpected cluster, a hole filling algorithm is used to repair the holes. This accurately repairs the flaws in the model conversion process without damaging the original design openings, improves the geometric integrity of the secondary cabinet's 3D model, and provides reliable support for subsequent intelligent operation and maintenance analysis.
[0073] Preferably, the flowchart of the unexpected cluster screening process provided in this embodiment is as follows: Figure 2 As shown.
[0074] There are many commonly used hole repair algorithms. In this embodiment, the minimum curvature surface repair algorithm is used to repair the hole. In practical applications, as other implementation methods, implementers can also use other hole repair methods such as Poisson reconstruction according to specific circumstances. This embodiment does not impose any special restrictions. The process of using the minimum curvature surface repair algorithm to repair the hole is a well-known technology and will not be described in detail here.
[0075] Thus, this embodiment utilizes the compliance confidence level constructed above to accurately filter out unexpected clusters with low confidence levels through threshold segmentation; furthermore, only these hole areas confirmed as unexpected defects are filled and repaired, thereby effectively improving the geometric integrity of the 3D model while fully preserving the original functional openings, providing a high-fidelity data foundation for subsequent intelligent operation and maintenance.
[0076] Based on the same inventive concept as the above method, this application embodiment also provides a digital modeling system for terminal blocks of substation secondary cabinets, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described digital modeling methods for terminal blocks of substation secondary cabinets.
[0077] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.
[0078] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
[0079] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.
Claims
1. A digital modeling method for terminal blocks of secondary switchgear in substations, characterized in that, The method includes the following steps: Obtain the 3D model of the substation secondary switchgear and perform triangular mesh generation on it; In the 3D model, the number of triangular faces connected to each edge on each triangular face is counted to determine the minimum adjacency number and detection edge of each triangular face; by analyzing the symmetry of the distribution of triangular faces in the neighborhood of each triangular face, the asymmetry of each triangular face is determined, and combined with the minimum adjacency number, the detection triangular faces are screened from all triangular faces in the 3D model. For all detected triangular face clusters, the local extension consistency of each cluster is determined by analyzing the consistency of the direction of the detected edge and its two nearest neighbor edges on each detected triangular face in each cluster. The edge regularity index of each cluster is determined by combining the dispersion of the length of all edges on each detected triangular face. In all components of the 3D model, the component to which each cluster belongs is obtained. By judging the consistency between the component to which each cluster belongs and the category to which any other component belongs, feature clusters are selected from all clusters. Based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster, and the difference in edge regularity index between each cluster and any feature cluster, the compliance confidence of each cluster is determined. Based on the compliance confidence level, unexpected clusters are screened out from all clusters, and the corresponding regions of the unexpected clusters in the 3D model are filled with holes.
2. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The method for determining the minimum adjacency number of each triangular face and the detection edge is as follows: The minimum number of triangles connected to all edges on each triangle is recorded as the minimum adjacency number of each triangle, and the edge corresponding to the minimum adjacency number is recorded as the detection edge.
3. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The method for determining the asymmetry of each triangular face is as follows: Construct a sphere with the centroid of each triangular face as the center and a preset value as the radius. Construct a plane passing through the center of the sphere, parallel to the detection edge and perpendicular to the triangular face, and divide the sphere into two parts. Calculate the absolute difference in the number of triangular faces in the two parts, and record it as the asymmetry of each triangular face.
4. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The process of filtering out the detection triangles from all triangles in the 3D model includes: Calculate the sum of the minimum adjacency number of each triangular face and the preset parameter adjustment coefficient, and record the ratio of the asymmetry of each triangular face to the sum as the edge confidence of each triangular side. The edge confidence of all triangular faces in the 3D model of the secondary cabinet is used as the input of the threshold segmentation algorithm, and the segmentation threshold is output. Triangular faces with edge confidence greater than or equal to the segmentation threshold are recorded as detected triangular faces.
5. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The method for determining the local extension consistency of each cluster is as follows: The vector connecting the midpoint of any one of the two nearest neighbor edges on each detection triangle to the midpoint of the detection edge is denoted as the first vector. The vector connecting the midpoint of the detection edge on each detection triangle to the midpoint of the other nearest neighbor edge is denoted as the second vector. The cosine of the angle between the first and second vectors is calculated. The mean of the absolute values of the cosines of the angles between the detection edges on all detection triangles within each cluster is denoted as the local extension consistency of each cluster.
6. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The method for determining the edge regularity index of each cluster is as follows: The dispersion of the length of all sides on each detection triangle is denoted as the side length dispersion of each detection triangle. The mean of the side length dispersion of all detection triangles in each cluster is denoted as the discrete feature value of each cluster. The edge regularity index of each cluster is positively correlated with the local extension consistency and negatively correlated with the discrete eigenvalues.
7. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The process of selecting feature clusters from all clusters includes: In a 3D model, if the component to which the current cluster belongs is of the same type as any component in the 3D model, then the current cluster is denoted as a feature cluster.
8. The digital modeling method for terminal blocks of secondary switchgear in substations as described in claim 1, characterized in that, The expression for the compliance confidence score of each cluster is: In the formula, This represents the local extension consistency of the u-th cluster; Represents the edge regularity index of the u-th cluster; This represents the normalized value of the distance feature value of the u-th cluster, where the distance feature value is obtained based on the distance from the average centroid coordinates of all detected triangular faces in each cluster to the average centroid coordinates of all detected triangular faces in any feature cluster; This represents the minimum value among the differences in edge regularity indices between the u-th cluster and all feature clusters; The matching factor of the u-th cluster is obtained by judging the consistency between the component to which each cluster belongs and the category to which any other component belongs.
9. The digital modeling method for terminal blocks of secondary switchgear in a substation as described in claim 1, characterized in that, The process of filtering out unexpected clusters from all clusters includes: In the 3D model, the compliance confidence of all clusters is used as the input of the Otsu threshold segmentation algorithm. The output segmentation threshold is denoted as the compliance threshold. Clusters with compliance confidence less than or equal to the compliance threshold are denoted as unexpected clusters.
10. A digital modeling system for terminal blocks of secondary switchgear in a substation, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the digital modeling method for the terminal block of the secondary panel cabinet in a substation as described in any one of claims 1-9.