A method and system for generating secondary physical circuit information model of conventional substation

Abstract

The present invention proposes a conventional substation secondary physical circuit information model generation method and system, including: obtaining a substation CAD drawing; parsing the primitives in the CAD drawing to obtain primitive parameters; calculating the intersection of graphic primitives and splitting the primitives according to the intersection position; calculating the overlapping interval of graphic primitives to split the primitives and remove the overlapping primitives; recalibrating the primitive coordinates using the center point of the primitive as the coordinate origin, and extracting the valid parameter values ​​of the primitives according to the primitive coordinate sequence; using a pre-trained model and a word segmenter to segment the text and convert the words into word vectors, and using a bag of words model to construct the word vectors into sentence vectors; hierarchically clustering the primitives using the primitive parameters after the specification and text vectorization, and outputting a binary clustering tree; starting from the leaf node cluster to the root node layer by layer traversal, assigning actual substation physical components to each node, and forming a conventional substation secondary physical circuit information model. The present invention has a higher degree of automation.

Description

Technical Field

[0001] The present invention belongs to the technical field of substation operation and maintenance, and in particular relates to a method and system for generating a secondary physical circuit information model of a conventional substation. Background Art

[0002] In the "Security Regulations for On-site Work of Relay Protection and Grid Safety Automatic Devices in Power Systems", it is pointed out that "for on-site verification of particularly complex protection devices or those with interlocking circuits, such as busbar protection and circuit breaker failure protection, a test plan approved by the technical person in charge and a relay protection safety measures ticket filled in by the work supervisor and approved by the technical person in charge should be prepared" in terms of preparation before on-site work. The formulation and implementation of relay protection safety measures tickets can encourage personnel at all levels to fulfill their safety responsibilities, understand the working conditions and nature, analyze risk factors, and take protective measures in advance.

[0003] At present, when staff members draft the content of relay protection safety measures, they need to manually read the electrical secondary vector diagram to obtain key information such as the circuit number and the corresponding terminal number according to different work tasks. Since the drafting of the safety measures ticket needs to be completed in a short time, its content is highly professional, with many sub-items and complex order. The manual reading method has a relatively low work efficiency and is prone to omissions and missing items.

[0004] The existing automated compilation method of safety measures tickets requires the establishment of a conventional substation secondary physical circuit information model to provide the ownership and association relationship between various components in the model, so as to further apply rules to generate safety measures text. Taking the patent document with publication number CN114239441A as an example, it first requires the user to "establish a substation secondary physical circuit information model file based on the substation secondary CAD drawing." This model building process is based on manual interaction between the user and the system, and the establishment of the ownership and association relationship of the substation equipment is completed through the point selection, dragging, input and other methods provided by the system, so there is still a lot of repetitive work.

[0005] The AC circuit diagram, control circuit diagram, terminal block diagram and other electrical schematic diagrams and installation wiring diagrams that need to be read when formulating relay protection safety measures tickets are mostly drawn using AUTOCAD software. The graphics are vector graphics, and there are differences in file formats and versions. During the drawing process, the above-mentioned graphics are mostly composed of independent graphics such as points, lines, and text, and the affiliation or association relationship between the graphics is often not defined during the drawing. Therefore, through reasonable technical means, the substation components to which the graphics belong, as well as the affiliation relationship between the substation components, are identified, and the secondary physical circuit information model of the conventional substation is quickly generated, which will provide good technical support for the further automation of the relay protection safety measures ticket generation software.

[0006] There are no reports on related inventions or papers based on electrical quadratic vector element clustering technology.

[0007] No related inventions or papers have been reported on the association of electrical secondary vector diagram element clustering with relay protection safety measures tickets.

[0008] The automatic generation scheme of substation security ticket proposed in the patent document with publication number CN114239441A and the patent document with publication number CN113435388B is similar to the technical background based on the present invention. Among them, the patent document with publication number CN114239441A proposes a method for establishing a so-called "model file" from a conventional substation CAD design file. The entire modeling process involved in this method is a manual interaction between the user and the system. The modeling of the affiliation and association relationship of the substation equipment is completed by the point selection, dragging, input and other methods provided by the system. The electrical secondary vector element hierarchical clustering method proposed in this patent is not involved; the patent document with publication number CN113435388B proposes a so-called "CAD recognition module" to generate a "secondary circuit connection relationship model" from the received CAD file, and perform the subsequent security ticket generation tasks of different security implementation processes based on the model, but does not explain how to obtain the connection relationship model from the CAD drawing.

[0009] The existing process of building the substation secondary physical circuit information model requires too much work, the operation is cumbersome and repetitive, and the applicability is not strong. The problems to be solved include: parsing CAD primitives; regulating CAD primitive parameters; determining the substation components to which CAD primitives containing valid text belong; determining the ownership and association relationship between substation components; and improving the component information in the secondary physical circuit information model based on the text in the CAD primitives. Summary of the invention

[0010] In view of the deficiencies in the prior art, the present invention provides a conventional substation secondary physical circuit information model generation method and system, which reduces the workload required by experts in the process of constructing the substation secondary physical circuit information model, alleviates the tedious and repetitive operations of users in the model construction process, and enhances the applicability of the rule-based automatic generation solution of security measures text.

[0011] To achieve the above object, the present invention adopts the following technical solutions:

[0012] A method for generating a secondary physical circuit information model of a conventional substation comprises the following steps:

[0013] Step 1: Obtain the CAD drawings of the substation related to safety measures;

[0014] Step 2, parsing various primitives in the CAD drawing to obtain all primitive parameters; the primitives include graphic primitives and text primitives;

[0015] Step 3, calculating the intersection points of the graphic primitives and splitting the primitives according to the intersection points; calculating the overlapping intervals of the graphic primitives, splitting the primitives based on the overlapping intervals and removing the overlapping primitives;

[0016] Step 4, calculate the center point of all the primitives after step 3, and use it as the coordinate origin to recalibrate the primitive coordinates, extract the valid parameter values ​​of the primitives according to the primitive coordinate sequence, and obtain the reduced primitive parameters;

[0017] Step 5: Use the pre-trained model and the word segmenter that matches the model to segment the text in the text-class graph primitive and convert each word into a word vector. Use the bag-of-words model to construct the word vector into a sentence vector to obtain the graph primitive parameters after text vectorization.

[0018] Step 6: hierarchically cluster the primitives using the primitive parameters after the reduction and text vectorization, and output a binary clustering tree;

[0019] Step 7: Starting from each leaf node cluster of the binary clustering tree, traverse layer by layer to the root node, and assign each node an actual substation physical component to which the node belongs, and finally form a conventional substation secondary physical circuit information model.

[0020] Furthermore, the graphic primitives include straight lines, broken lines, circles and arcs, and the text primitives include text blocks; the primitive parameters include space attribute parameters, morphological attribute parameters, color attribute parameters and text attribute parameters.

[0021] Furthermore, in step 5, the pre-trained model is specifically ERNIE 3.0Base.

[0022] Furthermore, in step 5, the word segmenter is specifically an AutoTokenizer that matches the pre-trained model ERNIE 3.0Base.

[0023] Furthermore, in step 5, the bag-of-words model is specifically the bag-of-words model BoWEncoder provided by PaddleNLP.

[0024] Furthermore, step 6 specifically includes:

[0025] Step 6.1, select the clustering method;

[0026] Step 6.2, select the proportion of the primitive parameters after reduction and text vectorization in calculating the distance between two primitives in the clustering process;

[0027] Step 6.3, respectively calculating the Euclidean space distance between the parameters of each primitive after the reduction and text vectorization of the two primitives, and using the ratio selected in step 6.2 to perform weighted summation on the Euclidean space distance between the parameters of each primitive after the reduction and text vectorization, and taking the summation result as the distance between the two primitives;

[0028] Step 6.4, inputting parameters of hierarchical clustering, wherein the parameters of hierarchical clustering include the maximum number of clustering iterations, the maximum number of leaf nodes, the clustering tree pruning threshold, and the clustering tree growth period; performing hierarchical clustering on the primitives using the clustering method selected in step 6.1;

[0029] Step 6.5: Use high-dimensional data dimensionality reduction methods to visualize the hierarchical clustering results and obtain a binary clustering tree;

[0030] Step 6.6: Calculate the clustering evaluation index and determine whether to re-cluster based on the clustering evaluation index; if re-clustering is required, adjust the parameters of the hierarchical clustering and restart the hierarchical clustering.

[0031] Furthermore, in step 6.5, the high-dimensional data dimensionality reduction method includes principal component analysis, T-distributed random neighbor embedding method or unified manifold approximation and projection method.

[0032] Furthermore, in step 6.6, the clustering evaluation indicators include silhouette coefficient, CH index, DBI index, cohesion and dispersion.

[0033] A conventional substation secondary physical circuit information model generation system includes a CAD drawing acquisition unit, a CAD drawing parsing unit, a graphic element processing unit, a graphic element parameter reduction unit, a text vectorization unit, a binary clustering tree unit and a model determination unit, wherein:

[0034] The CAD drawing acquisition unit is used to acquire the CAD drawings of the substation related to the safety measures;

[0035] The CAD drawing parsing unit is used to parse various graphic elements in the CAD drawing to obtain all graphic element parameters; the graphic elements include graphic elements and text elements;

[0036] The primitive processing unit is used to calculate the intersection points of the graphic primitives and split the primitives according to the intersection points; calculate the overlapping intervals of the graphic primitives, split the primitives based on the overlapping intervals and remove the overlapping primitives;

[0037] The primitive parameter reduction unit is used to calculate the center point of all primitives and use it as the coordinate origin to recalibrate the primitive coordinates, extract the valid parameter values ​​of the primitives according to the primitive coordinate sequence, and obtain the reduced primitive parameters;

[0038] The text vectorization unit hierarchically clusters the primitives using the primitive parameters after the specification and text vectorization, and outputs a binary clustering tree;

[0039] The method starts from each leaf node cluster of the binary cluster tree and traverses layer by layer toward the root node, and allocates an actual substation physical component to each node, and finally forms a conventional substation secondary physical loop information model.

[0040] Furthermore, it also includes an interactive interface, which includes a graphic element clustering interface, a clustering tree attribution allocation interface and an information model display interface. The graphic element clustering interface includes a parameter adjustment drag bar for hierarchical clustering, clustering options and a parameter input box, a visualization effect diagram of the hierarchical clustering, and a schematic diagram of the CAD file currently being processed. The clustering tree attribution allocation interface includes a hierarchical structure diagram of the clustering tree and the CAD drawing currently being processed. The information model display interface includes a tree list display of the conventional substation secondary physical circuit information model automatically generated through node allocation and the CAD drawing currently being processed.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] (1) Compared with complete manual operation by users, the method of the present invention can significantly reduce the number of user operation steps for building a substation equipment association and attribution relationship model, and has a higher degree of automation;

[0043] (2) The text information of each component of the substation can be directly obtained from the CAD element, which greatly reduces the user's mistakes in filling in text and symbols. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Flow chart of the method of the present invention.

[0045] Figure 2 Schematic diagram of the hierarchical clustering user interface.

[0046] Figure 3 Schematic diagram of the interface for assigning node elements.

[0047] Figure 4 Adjust the page diagram for the model structure. DETAILED DESCRIPTION

[0048] The present invention will now be described in further detail with reference to the accompanying drawings.

[0049] Example 1

[0050] Combination Figure 1 The present invention proposes a method for generating a conventional substation secondary physical circuit information model, comprising the following steps:

[0051] Step 1: Obtain the substation CAD drawings related to safety measures and the historical safety measures ticket files; the historical safety measures ticket files are used as references by the staff in the system and are not involved in the generation process of the automation loop information model;

[0052] Step 2, parsing various primitives in the CAD drawing to obtain all primitive parameters; the primitives include graphic primitives and text primitives, the graphic primitives include straight lines, polylines, circles and arcs, and the text primitives include text blocks; the primitive parameters include space attribute parameters, morphological attribute parameters, color attribute parameters and text attribute parameters;

[0053] Table 1 summarizes the primitive parameters into four attributes: spatial attributes, morphological attributes, color attributes, and text attributes; the details are as follows:

[0054] Table 1 Classification and dimensions of primitive attribute parameters

[0055]

[0056] Step 3, calculating the intersection points of the graphic primitives and splitting the primitives according to the intersection points; calculating the overlapping intervals of the graphic primitives, splitting the primitives based on the overlapping intervals and removing the overlapping primitives;

[0057] Step 4: Calculate the center point of all the primitives and use it as the coordinate origin to recalibrate the primitive coordinates. Extract the valid parameter values ​​of the primitives according to the primitive coordinate sequence to obtain the reduced primitive parameters.

[0058] Step 5. Use the ERNIE 3.0Base (Large-Scale Knowledge Enhanced Pre-Training for Language Understanding And Generation) pre-trained model and the AutoTokenizer tokenizer that matches the model to tokenize the text in the text-like graph and convert each word into a word vector. Use the bag of words model provided by PaddleNLP to construct the word vector (BagofWords, BoW) BoWEncoder into a sentence vector to obtain the graph parameters after text vectorization.

[0059] Step 6: Use the primitive parameters after reduction and text vectorization to hierarchically cluster the primitives and output a binary clustering tree; each node of the tree represents a group of primitives clustered into a cluster, the size of the node represents the number of primitives in the cluster, and the cluster represented by the node is the union of the clusters represented by all its child nodes.

[0060] Step 6 specifically includes:

[0061] Step 6.1, select the clustering method;

[0062] Step 6.2, select the proportion of the primitive parameters after reduction and text vectorization in calculating the distance between two primitives in the clustering process;

[0063] Step 6.3, respectively calculating the Euclidean space distance between the parameters of each primitive after the reduction and text vectorization of the two primitives, and using the ratio selected in step 6.2 to perform weighted summation on the Euclidean space distance between the parameters of each primitive after the reduction and text vectorization, and taking the summation result as the distance between the two primitives;

[0064] Step 6.4: Input the parameters of hierarchical clustering, including the maximum number of clustering iterations, the maximum number of leaf nodes, the clustering tree pruning threshold, and the clustering tree growth period; use the clustering method selected in step 6.1 to perform hierarchical clustering on the primitives; in this embodiment, K-means is used as an example, and the details are as follows:

[0065] Algorithm 1 Hierarchical clustering tree construction algorithm - taking K-means as an example

[0066]

[0067]

[0068] Step 6.5: Use high-dimensional data dimensionality reduction methods to visualize the hierarchical clustering results and obtain a binary clustering tree; high-dimensional data dimensionality reduction methods include: principal component analysis (PCA), T-distributed stochastic neighbor embedding (T-SNE) and unified manifold approximation and projection (UMAP) method; samples belonging to different nodes in the clustering tree will be presented as points of different colors in the visualization graph. If the primitives clustered into a cluster contain several text primitives, the text primitive closest to the cluster center in the feature space is selected and displayed in the center of the cluster.

[0069] Step 6.6, calculate clustering evaluation indicators, including silhouette coefficient, CH index (Calinski-Harabasz Score), DBI index (Davies-Bouldin Score), cohesion and separation, etc. Determine whether to re-cluster based on the clustering evaluation indicators; if re-clustering is required, adjust the parameters of hierarchical clustering and restart hierarchical clustering.

[0070] Step 7: Traverse from each leaf node cluster of the binary clustering tree to the root node layer by layer, and assign each node an actual substation physical component to which the node belongs, and finally form a conventional substation secondary physical circuit information model. Since the number of leaf nodes is limited by the maximum leaf node number threshold, the number of user operations is also relatively limited. The conventional substation secondary physical circuit information model structure we use contains three layers and a total of six types of elements, as shown in Table 2.

[0071] Table 2 Structure of the information model of the secondary physical circuit of a conventional substation

[0072]

[0073] Example 2

[0074] The present invention also develops a conventional substation secondary physical circuit information model generation system that implements the conventional substation secondary physical circuit information model generation method described in Example 1. The method of using the system is as follows:

[0075] Step 1: The user clicks to submit the CAD file to the system through the interactive page;

[0076] Step 2: The user enters the system element clustering page, such as Figure 2 As shown in the figure, the page contains the parameter adjustment slider, clustering options and parameter input box of hierarchical clustering, the visualization effect diagram of hierarchical clustering, and the schematic diagram of the currently processed CAD file. The user completes hierarchical clustering through the following operations:

[0077] Step 2.1, the user selects the basic clustering method to be used through the drop-down box. When the clustering method has additional parameters that need to be filled in, the system will appear with a corresponding text input box;

[0078] Step 2.2, the user adjusts the drag bar to set the distance calculation ratio of the four parameters of the spatial attribute, morphological attribute, color attribute and text attribute of the primitive in the clustering process;

[0079] Step 2.3, the user enters hierarchical clustering parameters such as the maximum number of clustering iterations, the maximum number of leaf nodes, the clustering tree pruning threshold, and the clustering tree growth period through the input box;

[0080] Step 2.4: The user clicks to start hierarchical clustering and waits for the system to respond;

[0081] Step 2.5: The system prompts that the clustering process is finished. The user clicks the drop-down box to adjust the visualization method of the hierarchical clustering and observes the visualization effect diagram of the hierarchical clustering in the system interface. The user interacts with the scatter plot by moving the cursor. The cluster where the cursor hovers in the scatter plot displays the representative text content of the cluster.

[0082] Step 2.6, the user observes the values ​​of clustering effect evaluation indexes such as silhouette coefficient, CH index, DBI index, cohesion and dispersion prompted in the interface;

[0083] Step 2.7: The user judges whether the current clustering effect meets the standard. If not, the user restarts the operation from 0; if so, the user clicks to complete clustering;

[0084] Step 3: The user enters the cluster tree attribution allocation page, such as Figure 3 As shown, the page contains the hierarchical structure diagram of the clustering tree and the currently processed CAD diagram. The user assigns the physical components of the substation to each node of the clustering tree through the following operations:

[0085] Step 3.1, the user observes the node currently in the highlighted state in the cluster tree hierarchy diagram, and observes the highlighted display of the primitives contained in the node in the CAD diagram;

[0086] Step 3.2, the user determines and assigns the type of physical component of the substation to which the current node belongs through the drop-down box;

[0087] Step 3.3, the user clicks to confirm the allocation;

[0088] Step 3.4: The system traverses each node of the clustering tree in order from the leaf node to the root node until all nodes are allocated. When a child node logically contains its parent node during the allocation process, the system will prompt the user that an allocation error has occurred.

[0089] Step 4: The user enters the conventional substation secondary physical circuit information model page, such as Figure 4 As shown, the page contains a tree list display of the conventional substation secondary physical circuit information model automatically generated through node allocation and the currently processed CAD drawing. Users can adjust the model errors caused by hierarchical clustering errors or node allocation errors through the following operations:

[0090] Step 4.1, the user selects a component in the tree list of the conventional substation secondary physical circuit information model and clicks a delete button to delete the component;

[0091] Step 4.2, the user selects a component in the tree list of the conventional substation secondary physical circuit information model and modifies the attribute information of the element in its detail page;

[0092] Step 4.3: The user clicks and drags a component in the tree list of the conventional substation secondary physical circuit information model to adjust the parent component to which the component belongs.

[0093] Step 5: The user clicks the Export Model button on the conventional substation secondary physical circuit information model page to export the conventional substation secondary physical circuit information model file in XML format and stores it in the file system. The model building process ends.

[0094] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A method for generating a secondary physical circuit information model of a conventional substation, characterized in that: The following steps are involved: Step 1: Obtain the CAD drawings of the substation related to safety measures; Step 2, parsing various primitives in the CAD drawing to obtain all primitive parameters; the primitives include graphic primitives and text primitives; Step 3, calculating the intersection points of the graphic primitives and splitting the primitives according to the intersection points; calculating the overlapping intervals of the graphic primitives, splitting the primitives based on the overlapping intervals and removing the overlapping primitives; Step 4, calculate the center point of all the primitives after step 3, and use it as the coordinate origin to recalibrate the primitive coordinates, extract the valid parameter values ​​of the primitives according to the primitive coordinate sequence, and obtain the reduced primitive parameters; Step 5: Use the pre-trained model and the word segmenter that matches the model to segment the text in the text-class graph primitive and convert each word into a word vector. Use the bag-of-words model to construct the word vector into a sentence vector to obtain the graph primitive parameters after text vectorization. Step 6: hierarchically cluster the primitives using the primitive parameters after the reduction and text vectorization, and output a binary clustering tree; specifically, the steps include: Step 6.1, select the clustering method; Step 6.2, select the proportion of the primitive parameters after reduction and text vectorization in calculating the distance between two primitives in the clustering process; Step 6.3, respectively calculating the Euclidean space distance between the parameters of each primitive after the reduction and text vectorization of the two primitives, and using the ratio selected in step 6.2 to perform weighted summation on the Euclidean space distance between the parameters of each primitive after the reduction and text vectorization, and taking the summation result as the distance between the two primitives; Step 6.4, inputting parameters of hierarchical clustering, wherein the parameters of hierarchical clustering include the maximum number of clustering iterations, the maximum number of leaf nodes, the clustering tree pruning threshold, and the clustering tree growth period; performing hierarchical clustering on the primitives using the clustering method selected in step 6.1; Step 6.5: Use high-dimensional data dimensionality reduction methods to visualize the hierarchical clustering results and obtain a binary clustering tree; Step 6.6, calculate the clustering evaluation index, and determine whether to re-cluster according to the clustering evaluation index; if re-clustering is required, adjust the parameters of hierarchical clustering and restart hierarchical clustering; Step 7: Starting from each leaf node cluster of the binary clustering tree, traverse layer by layer to the root node, and assign each node an actual substation physical component to which the node belongs, and finally form a conventional substation secondary physical circuit information model.

2. The conventional substation secondary physical circuit information model generation method according to claim 1, characterized in that: The graphic primitives include straight lines, broken lines, circles and arcs; the text primitives include text blocks; and the primitive parameters include space attribute parameters, morphological attribute parameters, color attribute parameters and text attribute parameters.

3. The conventional substation secondary physical circuit information model generation method according to claim 1, characterized in that: In step 5, the pre-trained model is specifically ERNIE 3.0Base.

4. The conventional substation secondary physical circuit information model generation method according to claim 2, characterized in that: In step 5, the word segmenter is specifically an AutoTokenizer that matches the pre-trained model ERNIE 3.0Base.

5. The method for generating a secondary physical circuit information model of a conventional substation according to claim 1, characterized in that: In step 5, the bag-of-words model is specifically the bag-of-words model BoWEncoder provided by PaddleNLP.

6. The method for generating a conventional substation secondary physical circuit information model according to claim 1, characterized in that: In step 6.5, the high-dimensional data dimensionality reduction method includes principal component analysis, T-distributed random neighbor embedding or unified manifold approximation and projection.

7. The conventional substation secondary physical circuit information model generation method according to claim 1, characterized in that: In step 6.6, the clustering evaluation indicators include silhouette coefficient, CH index, DBI index, cohesion and dispersion.

8. A conventional substation secondary physical circuit information model generation system implementing the model generation method described in any one of claims 1 to 7, characterized in that: It includes a CAD drawing acquisition unit, a CAD drawing parsing unit, a graphic element processing unit, a graphic element parameter reduction unit, a text vectorization unit, a binary clustering tree unit and a model determination unit, wherein: The CAD drawing acquisition unit is used to acquire the CAD drawings of the substation related to the safety measures; The CAD drawing parsing unit is used to parse various graphic elements in the CAD drawing to obtain all graphic element parameters; the graphic elements include graphic elements and text elements; The primitive processing unit is used to calculate the intersection points of the graphic primitives and split the primitives according to the intersection points; calculate the overlapping intervals of the graphic primitives, split the primitives based on the overlapping intervals and remove the overlapping primitives; The primitive parameter reduction unit is used to calculate the center point of all primitives and use it as the coordinate origin to recalibrate the primitive coordinates, extract the valid parameter values ​​of the primitives according to the primitive coordinate sequence, and obtain the reduced primitive parameters; The text vectorization unit hierarchically clusters the primitives using the primitive parameters after the specification and text vectorization, and outputs a binary clustering tree; The model determination unit traverses layer by layer from each leaf node cluster of the binary clustering tree to the root node, and allocates an actual substation physical component to each node, and finally forms a conventional substation secondary physical loop information model.

9. A conventional substation secondary physical circuit information model generation system as claimed in claim 8, characterized in that: It also includes an interactive interface, which includes a graphic element clustering interface, a clustering tree attribution allocation interface and an information model display interface. The graphic element clustering interface includes a parameter adjustment drag bar for hierarchical clustering, clustering options and a parameter input box, a visualization effect diagram of hierarchical clustering, and a schematic diagram of the CAD file currently being processed. The clustering tree attribution allocation interface includes a hierarchical structure diagram of the clustering tree and the CAD drawing currently being processed. The information model display interface includes a tree list display of a conventional substation secondary physical circuit information model automatically generated through node allocation and the CAD drawing currently being processed.

Images