A digital low-voltage distribution network rural transformer area network structure abnormality detection method and system and medium

Abstract

The application discloses a kind of digital low-voltage distribution network rural area network structure abnormality detection method, system and medium, and wherein method includes: obtaining the historical network structure image and current network structure image of digital low-voltage distribution network rural area;Two are converted into gray image;Again color threshold analysis, texture analysis are sequentially carried out, and the final historical network structure gray image and the final current network structure gray image are extracted;The topological feature vector of final historical network structure gray image and final current network structure gray image is respectively extracted;Dynamic time warping distance between topological feature vector is calculated, and compared with its threshold value, whether the historical network structure and current network structure are consistent is judged;If not consistent, superimposed comparison is carried out to historical network structure and current network structure, and intuitive display area network structure change result.The application has the advantages of simple detection method, no manual intervention, time-saving and labor-saving, high accuracy.

Description

Technical Field

[0001] This invention relates to the field of digital power grids, and in particular to a method, system, and medium for detecting network structure anomalies in rural transformer substations of digital low-voltage distribution networks. Background Technology

[0002] The modernization of rural power grids is driving the extension of smart grids to rural areas. Rural distribution areas have wide coverage, sparsely distributed users, long and branching lines, and are influenced by geographical environments (such as mountainous areas and farmland), resulting in complex network structures. Factors such as line additions and subtractions, transformer expansions, user access / exit, and grid connection of distributed photovoltaic and energy storage devices make the rural distribution network structure dynamic and prone to change. Failure to promptly grasp these changes will reduce the efficiency of distribution network operation and maintenance, increase the difficulty of analyzing problems such as three-phase imbalance, overload, and line losses, and make it difficult to ensure the safety and reliability of electricity use for rural users.

[0003] Currently, the main methods for detecting structural anomalies in rural power grids include manual inspection and drone inspection. Manual inspection requires staff to periodically visit rural areas to inspect each line, record anomaly data, and report the anomalies to the distribution network management platform for comparison and correction. This method suffers from problems such as data lag, difficulty in locating anomalies, and high time and labor costs. While drone inspection improves the automation level of rural power grid structural anomaly detection and reduces workload to some extent, it still requires on-site operation. Staff need to judge anomalies by comparing drone images, making automatic anomaly diagnosis difficult. This method is complex, time-consuming, labor-intensive, and has low accuracy. Summary of the Invention

[0004] To address the shortcomings of the existing technology, the present invention aims to provide a method, system, and medium for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network, so as to achieve network structure anomaly detection in rural transformer substations of a digital low-voltage distribution network that is time-saving, labor-saving, and highly accurate without manual intervention.

[0005] Firstly, a method for detecting network structure anomalies in rural distribution areas of a digital low-voltage distribution network is provided, including the following steps:

[0006] S1: Obtain historical and current network structure images of rural distribution areas in a digital low-voltage distribution network;

[0007] S2: Convert the historical network structure image and the current network structure image into grayscale images;

[0008] S3: Perform color threshold analysis and texture analysis in sequence to extract the grayscale image of the final historical network structure and the grayscale image of the final current network structure;

[0009] S4: Extract the topological feature vectors of the final historical network structure grayscale image and the final current network structure grayscale image respectively;

[0010] S5: Calculate the dynamic time warp distance between topological feature vectors, compare the dynamic time warp distance with its threshold, and determine whether the historical network structure is consistent with the current network structure.

[0011] S6: Overlay and compare the historical network structure with the current network structure to intuitively display the results of changes in the network structure of the transformer area.

[0012] According to the first aspect, in some possible implementations, both the historical network structure image and the current network structure image include information on transformer substations, users, and lines. The transformer substations, lines, and users are connected in sequence, and the lines include three-phase four-wire transmission lines and single-phase lines, which are distinguished by lines of varying thickness.

[0013] According to the first aspect, in some possible implementations, the steps of the color threshold analysis are as follows:

[0014] S301: Obtain the brightness value of each pixel in the grayscale image of the historical network structure and the grayscale image of the current network structure;

[0015] S302: Logically compare the brightness value of each pixel in the grayscale image of the historical network structure and the grayscale image of the current network structure with their respective brightness value thresholds. If the pixel brightness value is less than the brightness value threshold, the comparison result is 1, otherwise it is 0.

[0016] S303: Generate the binary matrix of brightness "0-1" for the grayscale image of the historical network structure and the binary matrix of brightness "0-1" for the grayscale image of the current network structure, respectively, according to S302.

[0017] According to the first aspect, in some possible implementations, the steps of the texture analysis are as follows:

[0018] S311: Calculate the local entropy of the "0-1" binary matrix of the brightness of the grayscale image of the historical network structure and the "0-1" binary matrix of the brightness of the grayscale image of the current network structure, and obtain the entropy matrix of the grayscale image of the historical network structure and the entropy matrix of the grayscale image of the current network structure.

[0019] S312: Linearly normalize the values ​​of the entropy matrix of the grayscale image of the historical network structure and the entropy matrix of the grayscale image of the current network structure to the range of [0, 1], and generate the corresponding standard grayscale image of the historical network structure and the standard grayscale image of the current network structure.

[0020] S313: Logically compare each pixel value of the historical network structure standard grayscale image and the current network structure standard grayscale image with the pixel threshold. If the pixel value is greater than the pixel threshold, the comparison result is 1, otherwise it is 0.

[0021] S314: Generate the binary matrix of the standard grayscale image "0-1" of the historical network structure and the binary matrix of the standard grayscale image "0-1" of the current network structure respectively, according to S313.

[0022] According to the first aspect, in some possible implementations, in step S3, the result obtained by color threshold analysis of the historical network structure grayscale image and the result obtained by texture analysis are subjected to a logical AND operation to extract the final historical network structure grayscale image; the result obtained by color threshold analysis of the current network structure grayscale image and the result obtained by texture analysis are subjected to a logical AND operation to extract the final historical network structure grayscale image and the final current network structure grayscale image.

[0023] According to the first aspect, in some possible implementations, the topological feature vector includes line area, line density, average line orientation angle, and standard deviation of line orientation angle.

[0024] According to the first aspect, in some possible implementations, the step of calculating the line area includes:

[0025] S401: Use the 4-neighborhood connectivity rule for binary images;

[0026] S402: Traverse the grayscale image of the network structure line by line, and perform the following operations for each foreground pixel: If the left and upper neighbors of the current pixel are both background pixels, then assign a new label to the current pixel; if the left or upper neighbor is a foreground pixel, then label the current pixel with the same label as the connected neighbors; if the labels of the left and upper neighbors are different, then record the equivalence relationship of the labels; wherein the grayscale image of the network structure is the final historical grayscale image of the network structure or the final current grayscale image of the network structure;

[0027] S403: Use a disjoint-set data structure to merge equivalent labels to ensure that the same connected region has a unique label;

[0028] S404: Calculate the number of foreground pixels in each connected region as the area of ​​each connected region;

[0029] S405: Rank the connected regions from largest to smallest, and take the area of ​​the top K connected regions as the line area, where K is a preset positive integer.

[0030] According to the first aspect, in some possible implementations, the line density is equal to the sum of all pixel values ​​of the network structure grayscale image divided by the number of pixels in the network structure grayscale image; wherein the network structure grayscale image is the final historical network structure grayscale image or the final current network structure grayscale image.

[0031] According to the first aspect, in some possible implementations, the direction angle of the straight line in the grayscale of the line is calculated using Hough transform as the line direction angle.

[0032] According to the first aspect, in some possible implementations, the superposition and comparison of the historical network structure and the current network structure to intuitively display the network structure change results specifically includes: extracting the outlines of the final historical network structure grayscale image and the final current network structure grayscale image respectively, superimposing the two outlines, and then using differentiated colors to mark the abnormal parts to intuitively display the changes in the transformer area network structure.

[0033] Secondly, a digital low-voltage distribution network rural transformer area network structure anomaly detection system is provided, including a memory and a processor connected to each other. The memory stores computer programs or instructions, and the processor executes the computer programs or instructions to implement the steps of the aforementioned digital low-voltage distribution network rural transformer area network structure anomaly detection method.

[0034] Thirdly, a computer-readable storage medium is provided, on which a computer program or instructions are stored, wherein when the computer program or instructions are executed by a computer device, the steps of the method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network as described above are implemented.

[0035] This invention proposes a method, system, and medium for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network. Compared with existing technologies, it has the following advantages: It includes acquiring historical and current network structure images of rural transformer substations in a digital low-voltage distribution network; converting the historical and current network structure images into grayscale images; sequentially performing color threshold analysis and texture analysis to extract the final historical and current network structure grayscale images; extracting topological feature vectors from both the final historical and current network structure grayscale images; calculating the dynamic time warping distance between the topological feature vectors; comparing the dynamic time warping distance with its threshold to determine whether the historical and current network structures are consistent; and superimposing and comparing the historical and current network structures to visually display the network structure change results. This method has the advantages of being simple, easy to program, requiring no manual intervention, saving time and effort, and having high accuracy. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of the method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network provided in this embodiment of the invention;

[0038] Figure 2 The image shows the historical network structure of a rural distribution substation, as provided in an embodiment of the present invention.

[0039] Figure 3 This is an image showing the current network structure of a rural distribution network area, as provided in an embodiment of the present invention.

[0040] Figure 4 This is a grayscale image of the historical network structure diagram of a rural distribution substation provided in an embodiment of the present invention.

[0041] Figure 5 This is a grayscale image of the current network structure diagram of a rural distribution area provided in an embodiment of the present invention.

[0042] Figure 6 The historical network structure grayscale image provided in this embodiment of the invention has a brightness "0-1" binary image.

[0043] Figure 7 This is a binary image of grayscale image brightness "0-1" provided for the current network structure in this embodiment of the invention.

[0044] Figure 8 The grayscale image of the final historical network structure provided in the embodiments of the present invention.

[0045] Figure 9 The grayscale image of the final current network structure provided in the embodiments of the present invention.

[0046] Figure 10 This is a diagram showing the network structure changes in rural transformer substations of a digital low-voltage distribution network, provided as an embodiment of the present invention. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0048] like Figure 1 As shown in the figure, this embodiment of the invention provides a method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network, comprising the following steps:

[0049] S1: Obtain historical and current network structure images of rural distribution areas in the digital low-voltage distribution network.

[0050] In this embodiment, both the historical network structure image and the current network structure image contain information about transformer substations, users, and lines. The transformer substations, lines, and users are connected in sequence. The lines include three-phase four-wire transmission lines and single-phase lines, which are distinguished by lines of different colors and thicknesses. Figure 2 The image shown is a historical network structure image of a rural distribution substation in this embodiment. In this embodiment, due to line structure adjustments, the current network structure image of the rural digital low-voltage distribution substation is as follows: Figure 3 As shown, a three-phase four-wire user and a single-phase user have been added inside the dashed box.

[0051] S2: Convert the historical network structure image and the current network structure image to grayscale images. Converting both images to grayscale is feasible, facilitates subsequent analysis and processing, and reduces the amount of data to be analyzed. Figure 2 and Figure 3 After being converted to grayscale images, they are respectively as follows: Figure 4 and Figure 5 As shown.

[0052] S3: Perform color threshold analysis and texture analysis in sequence to extract the grayscale images of the final historical network structure and the final grayscale image of the current network structure.

[0053] right Figure 4 and Figure 5 The steps for performing color thresholding analysis on the grayscale image shown are as follows:

[0054] S301: Obtain the brightness value of each pixel in the grayscale image of the historical network structure and the grayscale image of the current network structure;

[0055] S302: Logically compare the brightness value of each pixel in the grayscale image of the historical network structure and the grayscale image of the current network structure with their respective brightness value thresholds. If the pixel brightness value is less than the brightness value threshold, the comparison result is 1, otherwise it is 0.

[0056] S303: Generate the binary matrix of brightness "0-1" for the grayscale image of the historical network structure and the binary matrix of brightness "0-1" for the grayscale image of the current network structure, respectively, according to S302.

[0057] In this embodiment, the brightness threshold is equal to the product of the average value of the network structure grayscale image (historical network structure grayscale image or current network structure grayscale image) and a coefficient of 0.6 (in other embodiments, 0.5 or 0.7 can also be selected according to the actual situation). Figure 4 and Figure 5 The brightness thresholds are 151.1026 and 151.0464, respectively. The generated binary images of the historical network structure grayscale image (brightness "0-1") and the current network structure grayscale image (brightness "0-1") are shown below. Figure 6 , Figure 7 As shown.

[0058] right Figure 6 and Figure 7 Continue to conduct texture analysis to enhance the line extraction effect. The specific steps are as follows:

[0059] S311: Calculate the historical network structure grayscale image and the current network structure grayscale image after color threshold analysis (i.e., Figure 6 and Figure 7 The local entropy of the network structure is used to obtain the grayscale image entropy matrix of the historical network structure and the grayscale image entropy matrix of the current network structure.

[0060] S312: Linearly normalize the values ​​of the entropy matrix of the grayscale image of the historical network structure and the entropy matrix of the grayscale image of the current network structure to the range of [0, 1], and generate the corresponding standard grayscale image of the historical network structure and the standard grayscale image of the current network structure.

[0061] S313: Logically compare each pixel value of the historical network structure standard grayscale image and the current network structure standard grayscale image with a pixel threshold (0.4 in this embodiment, and can be selected according to actual needs in other embodiments, such as 0.35 or 0.45). If the pixel value is greater than the pixel threshold, the comparison result is 1, otherwise it is 0.

[0062] S314: Generate the binary matrix of the standard grayscale image "0-1" of the historical network structure and the binary matrix of the standard grayscale image "0-1" of the current network structure respectively, according to S313.

[0063] The final historical network structure grayscale image is extracted by performing a logical AND operation between the binary matrix of brightness "0-1" from the historical network structure grayscale image and the binary matrix of standard grayscale image of the historical network structure. Figure 8 As shown, the final grayscale image of the current network structure is extracted by performing a logical AND operation between the current network structure's grayscale image brightness "0-1" binary matrix and the current network structure's standard grayscale image "0-1" binary matrix. Figure 9 As shown.

[0064] S4: Extract the topological feature vectors of the final historical network structure grayscale image and the final current network structure grayscale image respectively.

[0065] In order to achieve Figure 8 The final historical network structure grayscale image and Figure 9 Finally, the comparison of the grayscale image of the current network structure requires extracting the topological feature vector from the image. In this embodiment, the topological feature vector includes the line area, line density, average line orientation angle, and standard deviation of the line orientation angle.

[0066] The steps for calculating the line area include:

[0067] S401: Use the 4-neighborhood connectivity rule for binary images;

[0068] S402: Traverse the grayscale image of the network structure line by line, and perform the following operations for each foreground pixel: If the left and upper neighbors of the current pixel are both background pixels, then assign a new label to the current pixel; if the left or upper neighbor is a foreground pixel, then label the current pixel with the same label as the connected neighbors; if the labels of the left and upper neighbors are different, then record the equivalence relationship of the labels; wherein the grayscale image of the network structure is the final historical grayscale image of the network structure or the final current grayscale image of the network structure;

[0069] S403: Use a disjoint-set data structure to merge equivalent labels to ensure that the same connected region has a unique label;

[0070] S404: Calculate the number of foreground pixels in each connected region as the area of ​​each connected region;

[0071] S405: Rank the connected regions from largest to smallest, and take the area of ​​the top K connected regions as the line area. K is a preset positive integer. In this embodiment, K is 5. In other embodiments, the value of K can be set according to actual needs, such as 4 or 6.

[0072] In this embodiment, Figure 8 The final calculated line area of ​​the historical network structure grayscale image is [90450, 912, 398, 116, 41]. Figure 9 The final calculated line area of ​​the current network structure grayscale image is [92943, 912, 419, 398, 116]. Because... Figure 8 and Figure 9 The range of changes in the network topology is not large, so the variation in line area is relatively small.

[0073] Wherein, the line density is equal to the sum of all pixel values ​​in the grayscale image of the network structure divided by the number of pixels in the grayscale image of the network structure; where the grayscale image of the network structure is either the final historical grayscale image of the network structure or the final current grayscale image of the network structure. Figure 8 The final line density of the historical network structure grayscale image is 0.0155. Figure 9 The final line density of the grayscale image of the current network structure is 0.0160. Figure 9 Due to the increased line length, the line density is higher. Figure 8 There has been an increase.

[0074] In this embodiment, the direction angle of the straight line in the grayscale of the line is calculated using Hough transform as the line direction angle.

[0075] According to the above eigenvector calculation method Figure 8 The final historical network structure grayscale image and Figure 9 The final topological feature vectors of the grayscale images of the current network structure are [90450, 912, 398, 116, 41, 0.0155, -0.5, 52.11] and [92943, 912, 419, 398, 116, 0.0160, -0.5, 52.11].

[0076] S5: Calculate the dynamic time warping distance between topological feature vectors, compare the dynamic time warping distance with its threshold, and determine whether the historical network structure is consistent with the current network structure.

[0077] The dynamic time warping distance between the above topological feature vectors is calculated, and compared with its threshold to determine whether the historical network structure and the current network structure are consistent. If they are consistent, it indicates that the distribution area network structure has not changed; if they are inconsistent, it indicates that the distribution area network structure has changed. In actual environments, although the network topology of rural distribution areas in low-voltage distribution networks changes frequently, each change is relatively small compared to the overall topology. Therefore, in this embodiment, the threshold for dynamic time warping distance is 0.004 times the length of the topological feature vector data (an empirical value; other values, such as 0.005, can be selected as needed in other embodiments), the length (dimension) of the topological feature vector is 8, and the threshold for dynamic time warping distance is 0.032. Based on this, the calculation... Figure 8 The final historical network structure grayscale image and Figure 9 The final distance between the topological feature vectors of the current network structure grayscale image is 1.4627. Since 1.4627 is greater than the set threshold of 0.032, it is determined that the network structure of the rural distribution area of ​​the digital low-voltage distribution network has changed.

[0078] S6: If the historical network structure is inconsistent with the current network structure, it indicates that the network structure of the transformer area has changed. The historical network structure and the current network structure are superimposed and compared to visually demonstrate the changes in the network structure of the transformer area.

[0079] To visually demonstrate the anomaly results, this embodiment extracts... Figure 8 The final historical network structure grayscale image and Figure 9 Finally, the outline of the current network structure grayscale image is obtained, and the two outlines are superimposed. Then, the abnormal parts are marked with different colors, such as... Figure 10 As shown, the changes in the network structure of rural distribution areas of the digital low-voltage distribution network are intuitively displayed to the staff.

[0080] In summary, the present invention provides a method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network. This method acquires historical and current network structure images of rural transformer substations in a digital low-voltage distribution network; converts both images into grayscale images; performs color threshold analysis and texture analysis sequentially to extract the final grayscale images of the historical and current network structures; extracts topological feature vectors from both images; calculates the dynamic time warping distance between the topological feature vectors; compares the dynamic time warping distance with a threshold to determine if the historical and current network structures are consistent; and overlays and compares the historical and current network structures to visually display the network structure change results. This method has the advantages of being simple to implement, easy to program, requiring no manual intervention, time-saving, labor-saving, and highly accurate.

[0081] Furthermore, this invention also provides a digital low-voltage distribution network rural transformer area network structure anomaly detection system, including a memory and a processor connected to each other. The memory stores computer programs or instructions, and the processor executes the computer programs or instructions to implement the steps of the aforementioned digital low-voltage distribution network rural transformer area network structure anomaly detection method.

[0082] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing a computer program or instructions thereon, wherein when the computer program or instructions are executed by a computer device, the steps of the method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network as described above are implemented.

[0083] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0084] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0085] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0086] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0088] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network, characterized in that, Includes the following steps: S1: Obtain historical and current network structure images of rural distribution areas in a digital low-voltage distribution network; S2: Convert the historical network structure image and the current network structure image into grayscale images; S3: Perform pixel brightness threshold binarization analysis and local entropy texture analysis in sequence to extract the final historical network structure grayscale image and the final current network structure grayscale image; S4: Extract the topological feature vectors of the final historical network structure grayscale image and the final current network structure grayscale image respectively; the topological feature vectors include line area, line density, average line orientation angle, and standard deviation of line orientation angle; S5: Calculate the dynamic time warp distance between topological feature vectors, compare the dynamic time warp distance with its threshold, and determine whether the historical network structure is consistent with the current network structure. S6: Overlay and compare the historical network structure with the current network structure to intuitively display the results of changes in the network structure of the transformer area.

2. The method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network according to claim 1, characterized in that, Both the historical network structure image and the current network structure image contain information about transformer substations, users, and lines. Transformers, lines, and users are connected in sequence. Lines include three-phase four-wire transmission lines and single-phase lines, which are distinguished by lines of different thicknesses.

3. The method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network according to claim 1, characterized in that, The steps of the pixel brightness threshold binarization analysis are as follows: S301: Obtain the brightness value of each pixel in the grayscale image of the historical network structure and the grayscale image of the current network structure; S302: Logically compare the brightness value of each pixel in the grayscale image of the historical network structure and the grayscale image of the current network structure with their respective brightness value thresholds. If the pixel brightness value is less than the brightness value threshold, the comparison result is 1, otherwise it is 0. S303: Generate the binary matrix of the brightness "0-1" of the grayscale image of the historical network structure and the binary matrix of the brightness "0-1" of the grayscale image of the current network structure, respectively, according to S302.

4. The method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network according to claim 3, characterized in that, The steps of the local entropy texture analysis are as follows: S311: Calculate the local entropy of the "0-1" binary matrix of the brightness of the grayscale image of the historical network structure and the "0-1" binary matrix of the brightness of the grayscale image of the current network structure, and obtain the entropy matrix of the grayscale image of the historical network structure and the entropy matrix of the grayscale image of the current network structure. S312: Linearly normalize the values ​​of the entropy matrix of the grayscale image of the historical network structure and the entropy matrix of the grayscale image of the current network structure to the range of [0, 1], and generate the corresponding standard grayscale image of the historical network structure and the standard grayscale image of the current network structure. S313: Logically compare each pixel value of the historical network structure standard grayscale image and the current network structure standard grayscale image with the pixel threshold. If the pixel value is greater than the pixel threshold, the comparison result is 1, otherwise it is 0. S314: Generate the binary matrix of the standard grayscale image "0-1" of the historical network structure and the binary matrix of the standard grayscale image "0-1" of the current network structure respectively, according to S313.

5. The method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network according to claim 1, characterized in that, In step S3, the results obtained from color threshold analysis and texture analysis of the historical network structure grayscale image are combined using a logical AND operation to extract the final historical network structure grayscale image; the results obtained from color threshold analysis and texture analysis of the current network structure grayscale image are combined using a logical AND operation to extract the final current network structure grayscale image.

6. The method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network according to claim 1, characterized in that, The steps for calculating the line area include: S401: Use the 4-neighborhood connectivity rule for binary images; S402: Traverse the grayscale image of the network structure line by line, and perform the following operations for each foreground pixel: If the left and upper neighbors of the current pixel are both background pixels, then assign a new label to the current pixel; if the left or upper neighbor is a foreground pixel, then label the current pixel with the same label as the connected neighbors; if the labels of the left and upper neighbors are different, then record the equivalence relationship of the labels; wherein the grayscale image of the network structure is the final historical grayscale image of the network structure or the final current grayscale image of the network structure; S403: Use a disjoint-set data structure to merge equivalent labels to ensure that the same connected region has a unique label; S404: Calculate the number of foreground pixels in each connected region as the area of ​​each connected region; S405: Rank the connected regions from largest to smallest, and take the area of ​​the top K connected regions as the line area, where K is a preset positive integer.

7. The method for detecting network structure anomalies in rural distribution areas of a digital low-voltage distribution network according to claim 1, characterized in that, The line density is equal to the sum of all pixel values ​​in the grayscale image of the network structure divided by the number of pixels in the grayscale image of the network structure; wherein the grayscale image of the network structure is either the final historical grayscale image of the network structure or the final current grayscale image of the network structure.

8. The method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network according to claim 1, characterized in that, The direction angle of the straight line in the grayscale of the line is calculated using Hough transform and taken as the line direction angle.

9. The method for detecting network structure anomalies in rural distribution areas of a digital low-voltage distribution network according to claim 1, characterized in that, The method of superimposing and comparing the historical network structure and the current network structure to intuitively display the changes in network structure specifically includes: extracting the outlines of the final historical network structure grayscale image and the final current network structure grayscale image respectively, superimposing the two outlines, and then using differentiated colors to mark the abnormal parts to intuitively display the changes in the network structure of the transformer area.

10. A digital low-voltage distribution network rural transformer area network structure anomaly detection system, comprising an interconnected memory and a processor, wherein the memory stores computer programs or instructions, characterized in that, When the processor executes the computer program or instructions, it implements the steps of the method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network as described in any one of claims 1 to 9.

11. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a computer device, they implement the steps of the method for detecting network structure anomalies in rural transformer substations of a digital low-voltage distribution network as described in any one of claims 1 to 9.

Images