Power system topology similarity data deviation correction method

Abstract

The invention discloses a power system topology similarity data deviation correction method, and relates to the field of power systems. According to the power system topology similarity data deviation correction method, a power system architecture model is abstracted into a simple topology model, corresponding data and graphs are abstracted from the topology model, the processed data are mapped to a corresponding topology space, a characteristic relation is obtained, key characteristics are extracted and compared, and difference data are identified. According to the method, differential data are separated by utilizing topological characteristics, traceablility is carried out on the differential data, the differential data are reversely mapped to a corresponding power system model from a topological model, so that data deviation in each topological level in the power system model is captured, data modification is carried out aiming at data value deviation, and the power system model is obtained. The power system is corrected by utilizing the topological similarity data deviation, the data reading amount and the adjustment difficulty are reduced, multi-system unit batch operation is realized for a large comprehensive system, and the adjustment period is shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method for correcting topology similarity data deviation of a power system. Background Art

[0002] As the main control method for centralized control in the power transmission and distribution network, the power system faces problems such as frequent topology changes and line switching. The workload of data analysis under different topological states is large. It is necessary to analyze the impact of topology on system operation data. In response to different power application environments, the power system also needs to make certain adjustments.

[0003] In the prior art, the correction process of the power system is usually adjusted according to the direct application performance or data feedback of the power system. For some power plants or independent power users, the entire power system model is small, the adjustment difficulty is low, and it can be directly operated. However, for some comprehensive large-scale power systems, such as the civil power system of an entire city, it covers a wide range and involves a lot of content, and the topological relationship within the range is intricate and not a simple linear relationship. At this time, if it is adjusted directly according to the data feedback of the power system, on the one hand, the data reading volume is large, and on the other hand, the adjustment cycle is long and the difficulty is great. After the power system is divided into units, data analysis is performed according to the topological similarity data deviation of each power system unit, and then the topological model is used to verify and correct the data, which can reduce the data reading volume and can perform batch operations on multiple power unit systems, so that the correction cycle of the entire power system is short and the difficulty is small. For this purpose, a power system topology similarity data deviation correction method is provided to solve the above problems. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method for correcting the deviation of power system topology similarity data, which solves the problem that a comprehensive large-scale power system covers a wide range, involves a lot of content, and the topological relationship within the range is intricate and not a simple linear relationship. If it is adjusted directly based on the data feedback of the power system, on the one hand, the data reading volume is large, and on the other hand, the adjustment cycle is long and difficult.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for correcting deviation of power system topology similarity data, comprising the following steps;

[0006] Step 1: Obtain a power system architecture model;

[0007] Step 2: Topologically abstract the power system;

[0008] Step 3: Abstract data and graphics from the topology model;

[0009] Step 4: compare topological data and separate difference data;

[0010] Step 5: Feedback the deviations of each level of the topology model based on the difference data;

[0011] Step 6: Summarize the topological deviations to obtain the system data deviation items and deviation values, and correct the deviation contents;

[0012] The deviation correction process includes:

[0013] Data cleaning, removing and correcting duplicate, incomplete and erroneous data. The whole process includes deduplication, filling missing data, correcting erroneous data and standardization;

[0014] Data verification, using statistical analysis, cross-validation and historical data comparison, using the power system basic model data and the acquired power system data model to compare data, so as to verify the consistency and accuracy of the acquired power data after correction;

[0015] Data error correction, by detecting and correcting errors in the data, including data quality, data error detection and correction, verifying the original data through mathematical statistics methods, deleting the error data, tuning the remaining data, and finally making the power system data close to the standard model;

[0016] Quantile mapping method, when there are obvious differences between the power system data from different models and the acquired power system model, the deviation of the current power model data is corrected through the cumulative distribution function relationship between the historical model data and multiple groups of actual analysis data.

[0017] Preferably, in step one, when acquiring the power system architecture model, the type of the target power system is identified, and then the levels of the main architecture framework of the power system are obtained, the model data is listed in the form of a tree diagram, and finally a complete power system architecture model is drawn.

[0018] Preferably, in step 2, the process of topological abstraction of the power system includes the following steps:

[0019] S1. Establish the topology of the power grid. Connect the various components in the power system to form the topology of the power grid. Use different types of annotation lines to mark the relationship and orientation between each node, and overall describe the physical connection relationship between the various components in the power grid.

[0020] S2. Topology analysis: simplifying the power system network described based on the switch and node model into a bus and branch model, converting the complex power system network data into a simpler network topology structure, and realizing simplified processing of the topology structure;

[0021] S3, drawing a power grid topology diagram according to the simplified topology relationship diagram, forming a connection mode description of the entire power grid based on the loop reliability through the power grid topology diagram drawn in S1, and calculating the reliability index of the power grid as a whole and the hierarchical partitions;

[0022] S4. Use data processing tools to abstract the topology diagram, use professional drawing tools to quickly build a topology relationship network, then fill in the corresponding components, and then mark the annotation lines, so as to abstract the complex power system model into a simple topology model;

[0023] In the abstract process based on node merging, when calculating the electrical distance, let the impedance between node i and node j be Z ij , the electrical distance d between nodes can be calculated ij To measure their proximity, we use where |Z IJ | represents the modulus of impedance, when the electrical distance d ij Merge nodes when the number is less than the threshold;

[0024] When calculating the merged node, we need to merge nodes i and j into a new node k. For the injected power P and Q of the node, we have P k =P i +P j , Q k =Q i +Q j ; where and are the combined equivalent resistance and reactance.

[0025] Preferably, in step 3, the process of abstracting data and graphics from the topology model includes:

[0026] S1. Data preparation, collecting and processing raw data. For image data, the input nodes will be arranged in a square grid corresponding to the pixel matrix for the data that constitutes the image;

[0027] S2, topological representation, mapping the processed data to the corresponding topological space and extracting the topological corresponding points of the data;

[0028] S3, feature extraction, based on the topological representation, classification is performed according to the corresponding point identification, feature relationships are obtained, and then key features are extracted;

[0029] S4, image data generation, visualize the extracted features, generate graphical representations, and use graphical libraries and visualization tools to convert topological features into intuitive graphical representations, including network graphs, graphical representations of nodes and edges;

[0030] S5, data definition, classify and identify the generated graphics and data, and correspond them according to data types and topological hierarchical relationships;

[0031] In the whole process, the original data set is set to D = {d1, d2, ..., dn}, where each data point d has a feature vector The target topological space consists of a node set V = {v1, U2, …, υp} and an edge set E = {(va, ub) | va, u€V}.

[0032] In the process of abstracting data and graphs from the topological model, node degree calculation needs to be performed;

[0033] For an undirected graph, the node degree refers to the number of edges connected to the node. Let the degree of node v be d(v). If there is an adjacency matrix A representing the connectivity of the graph, for a graph with n nodes,

[0034] For directed graphs, there are in-degree and out-degree. In-degree is the number of edges pointing to the node, and out-degree is the number of edges emanating from the node. Let the in-degree of node v be d in (υ), out-degree is d out (υ), then,

[0035] Preferably, in step four, the topological data comparison is a specific topological data comparison of the power system, which includes line-transformer relationships, household-transformer relationships and phase relationships. Through topological data analysis, the overall structure and pattern of the data are captured, and the similarities and differences between the data are compared and identified, thereby discovering the potential structure and relationship in the data.

[0036] Preferably, in the process of topological data comparison, for the similarity function defining the node attributes, for nodes with numerical attributes, the Euclidean distance measurement method is used, and the attribute vectors of two nodes A and B are respectively and

[0037] For nodes with classification attributes, the Jaccard coefficient method can be used. For the classification attribute set S of two nodes, A and S B ,

[0038] Preferably, the line-transformer relationship refers to the connection relationship between the power line and the transformer, and the line-transformer relationship is determined based on a K nearest neighbor algorithm and rules;

[0039] The household-to-transformer relationship is the process of converting high-voltage power into low-voltage power through a transformer and supplying it to users. By matching and comparing the load data of users and transformers, accurate identification of the household-to-transformer relationship is achieved, providing accurate data support for load forecasting and electric energy metering of the power system.

[0040] The phase relationship refers to the relative phase difference between the voltage and current of each node in the power system. By measuring and calculating the node voltage phase difference, the phase relationship between the nodes can be determined, providing accurate data support for the stability analysis and short-circuit calculation of the power system.

[0041] Preferably, in step four, after completing the topological data comparison, the topological characteristics of the data are identified by analyzing the topological properties of the data, including connectivity and holes, so as to understand the overall structure and local differences of the data. Then, the difference data is identified by comparing the topological characteristics of different data sets, so as to tolerate small-scale deformation and distortion of the data, and is applied to the difference data processing and separation of topological relationships of complex types of power systems.

[0042] Preferably, in step five, the corresponding difference data is separated by analyzing the topological characteristics of the same data set, and then the difference data is traced and reversely mapped from the topological model to the corresponding power system model, thereby capturing the data deviations in each topological level in the power system model.

[0043] Preferably, the data deviation includes the deviation of the data item and the deviation of the data value. With respect to the data item deviation, the corresponding data item is supplemented according to the topological mapping relationship, and the corresponding data item is associated with the power system architecture. With respect to the data value deviation, the threshold is adjusted for the corresponding data, and the model data is automatically modified within the effective threshold range.

[0044] The present invention discloses a method for correcting deviation of power system topology similarity data, which has the following beneficial effects:

[0045] 1. The method for correcting the topological similarity data deviation of the power system obtains the architecture of the power system, abstracts the power system architecture model into a simple topological model, reduces the data volume and process complexity, then abstracts the corresponding data and graphics from the topological model, maps the processed data into the corresponding topological space, extracts the topological corresponding points of the data, and then classifies them according to the corresponding point identifiers on the basis of the topological representation to obtain the feature relationship, and then extracts the key features, compares the extracted key features, identifies the difference data, and then separates the difference data using the topological features, traces the difference data, and reversely maps it from the topological model to the corresponding power system model, thereby capturing the data deviation in each topological level in the power system model, adjusting the threshold of the corresponding data for the data value deviation, and automatically modifying the model data within the effective threshold range, thereby realizing the use of topological similarity data deviation to correct the power system, reducing the data reading volume and adjustment difficulty, and for large-scale comprehensive systems, it can realize synchronous batch operation of multiple system units and shorten the adjustment cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 This is a flow chart of a method for correcting the deviation of power system topology similarity data according to the present invention;

[0048] Figure 2 The data deviation correction flow chart of the present invention is as follows;

[0049] Figure 3 It is a flow chart of the power system topology abstraction method of the present invention;

[0050] Figure 4 Abstracting and generating a flow chart for data and images in the topological model of the present invention;

[0051] Figure 5 A schematic diagram of a power system topology data comparison project according to the present invention;

[0052] Figure 6 This is a simplified topological relationship diagram of the power system of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] The embodiment of the present application provides a method for correcting the deviation of power system topology similarity data, thereby solving the problem that a large-scale comprehensive power system covers a wide range, involves many contents, and the topological relationship within the range is intricate and not a simple linear relationship. If it is adjusted directly according to the data feedback of the power system, on the one hand, the data reading volume is large, and on the other hand, the adjustment cycle is long and difficult.

[0055] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0056] An embodiment of the present invention discloses a method for correcting deviation of power system topology similarity data.

[0057] According to the attached Figure 1-6 As shown, the following steps are included;

[0058] Step 1: Obtain a power system architecture model;

[0059] Step 2: Topologically abstract the power system;

[0060] Step 3: Abstract data and graphics from the topology model;

[0061] Step 4: compare topological data and separate difference data;

[0062] Step 5: Feedback the deviations of each level of the topology model based on the difference data;

[0063] Step 6: Summarize the topological deviations to obtain the system data deviation items and deviation values, and correct the deviation contents;

[0064] The deviation correction process includes:

[0065] Data cleaning, removing and correcting duplicate, incomplete and erroneous data. The whole process includes deduplication, filling missing data, correcting erroneous data and standardization;

[0066] Data verification, using statistical analysis, cross-validation and historical data comparison, using the power system basic model data and the acquired power system data model to compare data, so as to verify the consistency and accuracy of the acquired power data after correction;

[0067] Data error correction, by detecting and correcting errors in the data, including data quality, data error detection and correction, verifying the original data through mathematical statistics methods, deleting the error data, tuning the remaining data, and finally making the power system data close to the standard model;

[0068] Quantile mapping method, when there are obvious differences between the power system data from different models and the acquired power system model, the deviation of the current power model data is corrected through the cumulative distribution function relationship between the historical model data and multiple groups of actual analysis data.

[0069] In step one, when acquiring the power system architecture model, the type of the target power system is identified, and then the main architecture framework of the power system is obtained to determine which levels it contains, and the model data is listed in the form of a tree diagram, and finally a complete power system architecture model is drawn.

[0070] In step 2, the process of topological abstraction of the power system includes the following steps;

[0071] S1. Establish the topology of the power grid. Connect the various components in the power system to form the topology of the power grid. Use different types of annotation lines to mark the relationship and orientation between each node, and overall describe the physical connection relationship between the various components in the power grid.

[0072] S2. Topology analysis: simplifying the power system network described based on the switch and node model into a bus and branch model, converting the complex power system network data into a simpler network topology structure, and realizing simplified processing of the topology structure;

[0073] S3, drawing a power grid topology diagram according to the simplified topology relationship diagram, forming a connection mode description of the entire power grid based on the loop reliability through the power grid topology diagram drawn in S1, and calculating the reliability index of the power grid as a whole and the hierarchical partitions;

[0074] S4. Use data processing tools to abstract the topology diagram, use professional drawing tools to quickly build a topology relationship network, then fill in the corresponding components, and then mark the annotation lines, so as to abstract the complex power system model into a simple topology model;

[0075] In the abstract process based on node merging, when calculating the electrical distance, let the impedance between node i and node j be Z ij , the electrical distance d between nodes can be calculated ij To measure their proximity, we use where |Z IJ | represents the modulus of impedance, when the electrical distance d ij Merge nodes when the number is less than the threshold;

[0076] When calculating the merged node, we need to merge nodes i and j into a new node k. For the injected power P and Q of the node, we have P k =P i +P j , Q k =Q i +Q j ; where and are the combined equivalent resistance and reactance.

[0077] In step 3, the process of abstracting data and graphics from the topology model includes:

[0078] S1. Data preparation, collecting and processing raw data. For image data, the input nodes will be arranged in a square grid corresponding to the pixel matrix for the data that constitutes the image;

[0079] S2, topological representation, mapping the processed data to the corresponding topological space and extracting the topological corresponding points of the data;

[0080] S3, feature extraction, based on the topological representation, classification is performed according to the corresponding point identification, feature relationships are obtained, and then key features are extracted;

[0081] S4, image data generation, visualize the extracted features, generate graphical representations, and use graphical libraries and visualization tools to convert topological features into intuitive graphical representations, including network graphs, graphical representations of nodes and edges;

[0082] S5, data definition, classify and identify the generated graphics and data, and correspond them according to data types and topological hierarchical relationships;

[0083] In the whole process, the original data set is set to D = {d1, d2, ..., dn}, where each data point d has a feature vector The target topological space consists of a node set V = {v1, U2, …, υp} and an edge set E = {(va, ub) | va, u€V}.

[0084] In the process of abstracting data and graphs from the topological model, node degree calculation is required;

[0085] For an undirected graph, the node degree refers to the number of edges connected to the node. Let the degree of node υ d(υ) be, if there is an adjacency matrix A representing the connectivity of the graph, for a graph with n nodes,

[0086] For a directed graph, it is divided into in-degree and out-degree. In-degree is the number of edges pointing to the node, and out-degree is the number of edges emanating from the node. Let the in-degree of node υ be din (υ), out-degree is d out (υ), then,

[0087] In step 4, topological data comparison is a specific topological data comparison of the power system, which includes line-transformer relationships, household-transformer relationships, and phase relationships. Through topological data analysis, the overall structure and pattern of the data are captured, and the similarities and differences between the data are compared and identified, thereby discovering the potential structure and relationship in the data.

[0088] In the process of topological data comparison, for the similarity function that defines the node attributes, the nodes with numerical attributes use the Euclidean distance measurement method. Suppose the attribute vectors of two nodes A and B are and

[0089] For nodes with classification attributes, the Jaccard coefficient method can be used. For the classification attribute set S of two nodes, A and S B ,

[0090] The line-to-transformer relationship refers to the connection relationship between the power line and the transformer, which is determined based on the K nearest neighbor algorithm and rules;

[0091] The user-to-transformer relationship is the process of converting high-voltage power into low-voltage power through transformers and supplying it to users. By matching and comparing the load data of users and transformers, the user-to-transformer relationship can be accurately identified, providing accurate data support for load forecasting and power metering of the power system.

[0092] Phase relationship refers to the relative phase difference between voltage and current in the power system. By measuring and calculating the node voltage phase difference, the phase relationship between nodes can be determined, providing accurate data support for the stability analysis and short-circuit calculation of the power system.

[0093] In step 4, after completing the topological data comparison, the topological characteristics of the data are identified by analyzing the topological properties of the data, including connectivity and holes, to understand the overall structure and local differences of the data. Then, by comparing the topological characteristics of different data sets, the difference data is identified, thereby tolerating small-scale deformation and distortion of the data, and applying it to the difference data processing and separation of the topological relationships of complex types of power systems.

[0094] In step five, the corresponding difference data is separated by analyzing the topological characteristics of the same data set, and then the difference data is traced and reversely mapped from the topological model to the corresponding power system model, thereby capturing the data deviations in each topological level in the power system model.

[0095] Data deviation includes the deviation of data items and the deviation of data values. For data item deviation, the corresponding data items are supplemented according to the topological mapping relationship, and the corresponding data items are associated with the power system architecture. For data value deviation, the threshold is adjusted for the corresponding data, and the model data is automatically modified within the valid threshold range.

[0096] The method for correcting the deviation of the data of the topological similarity of the power system obtains the architecture of the power system, and then processes the power system architecture model, so that the complex system model is abstracted into a simple topological model, reducing the amount of data and the complexity of the process, and then abstracts the corresponding data and graphics from the topological model, maps the processed data into the corresponding topological space, extracts the topological corresponding points of the data, and then classifies them according to the corresponding point identifiers on the basis of the topological representation, obtains the feature relationship, and then extracts the key features, compares the extracted key features, and identifies the topological features of the data by analyzing the topological properties of the data, including connectivity and holes, so as to understand the overall structure and local differences of the data, and then identifies the difference data by comparing the topological features of different data sets;

[0097] Then, the topological features are used to separate the difference data, trace the source of the difference data, and reversely map it from the topological model to the corresponding power system model, so as to capture the data deviations in each topological level in the power system model. According to the data value deviation, the threshold is adjusted to the corresponding data, and the model data is automatically modified within the effective threshold range, so as to realize the use of topological similarity data deviation to correct the power system, reduce the amount of data reading and adjustment difficulty, and for large-scale comprehensive systems, it can realize synchronous batch operation of multiple system units and shorten the adjustment cycle.

[0098] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for correcting deviation of power system topology similarity data, characterized in that: The steps include: Step 1: Obtain a power system architecture model; Step 2: Perform topological abstraction on the power system; Step 3: Abstract data and graphics from the topology model; Step 4: compare topological data and separate difference data; Step 5: Feedback the deviations of each level of the topology model based on the difference data; Step 6: Summarize the topological deviations to obtain the system data deviation items and deviation values, and correct the deviation contents; The deviation correction process includes: Data cleaning, removing and correcting duplicate, incomplete and erroneous data. The whole process includes deduplication, filling missing data, correcting erroneous data and standardization; Data verification, using statistical analysis, cross-validation and historical data comparison, using the power system basic model data and the acquired power system data model to compare data, so as to verify the consistency and accuracy of the acquired power data after correction; Data error correction, by detecting and correcting errors in the data, including data quality, data error detection and correction, verifying the original data through mathematical statistics methods, deleting the error data, tuning the remaining data, and finally making the power system data close to the standard model; Quantile mapping method, when there are obvious differences between the power system data from different models and the acquired power system model, the deviation of the current power model data is corrected through the cumulative distribution function relationship between the historical model data and multiple groups of actual analysis data.

2. A method for correcting deviation of power system topology similarity data according to claim 1, characterized in that: In the step one, when acquiring the power system architecture model, the type of the target power system is identified, and then the main architecture framework of the power system is obtained to determine which levels it contains, and the model data is listed in the form of a tree diagram, and finally a complete power system architecture model is drawn.

3. A method for correcting the deviation of power system topology similarity data according to claim 1, characterized in that: In the step 2, the process of topological abstraction of the power system includes the following steps: S1. Establish the topology of the power grid. Connect the various components in the power system to form the topology of the power grid. Use different types of annotation lines to mark the relationship and orientation between each node, and overall describe the physical connection relationship between the various components in the power grid. S2. Topology analysis: simplifying the power system network described based on the switch and node model into a bus and branch model, converting the complex power system network data into a simpler network topology structure, and realizing simplified processing of the topology structure; S3, drawing a power grid topology diagram according to the simplified topology relationship diagram, forming a connection mode description of the entire power grid based on the loop reliability through the power grid topology diagram drawn in S1, and calculating the reliability index of the power grid as a whole and the hierarchical partitions; S4. Use data processing tools to abstract the topology diagram, use professional drawing tools to quickly build a topological relationship network, then fill in the corresponding components, and then mark the annotation lines, so as to abstract the complex power system model into a simple topology model. In the abstract process based on node merging, when calculating the electrical distance, let the impedance between node i and node j be Z ij , the electrical distance d between nodes can be calculated ij To measure their proximity, we use where |Z IJ | represents the modulus of impedance, when the electrical distance d ij Merge nodes when the number is less than the threshold; When calculating the merged node, we need to merge nodes i and j into a new node k. For the injected power P and Q of the node, we have P k =P i +P j , Q k =Q i +Q j ; where and are the combined equivalent resistance and reactance.

4. The method for correcting the deviation of power system topology similarity data according to claim 1, characterized in that: In step 3, the process of abstracting data and graphics from the topology model includes: S1. Data preparation, collecting and processing raw data. For image data, the input nodes will be arranged in a square grid corresponding to the pixel matrix for the data that constitutes the image; S2, topological representation, mapping the processed data to the corresponding topological space and extracting the topological corresponding points of the data; S3, feature extraction, based on the topological representation, classification is performed according to the corresponding point identification, feature relationships are obtained, and then key features are extracted; S4, image data generation, visualize the extracted features, generate graphical representations, and use graphical libraries and visualization tools to convert topological features into intuitive graphical representations, including network graphs, graphical representations of nodes and edges; S5, data definition, classify and identify the generated graphics and data, and correspond them according to data types and topological hierarchical relationships; In the whole process, the original data set is set as D = {d1, d2, …, dn}, where each data point d; With characteristic vector The target topological space consists of a node set V = {v1, U2, …, υp} and an edge set E = {(va, ub) | va, u€V}.

5. A method for correcting deviation of power system topology similarity data according to claim 4, characterized in that: In the process of abstracting data and graphs from the topological model, node degree calculation needs to be performed; For an undirected graph, the node degree refers to the number of edges connected to the node. Let the degree of node v be d(v). If there is an adjacency matrix A representing the connectivity of the graph, for a graph with n nodes, For directed graphs, there are in-degree and out-degree. In-degree is the number of edges pointing to the node, and out-degree is the number of edges emanating from the node. Let the in-degree of node v be d in (v), out-degree is d out (v) then, 6. A method for correcting deviation of power system topology similarity data according to claim 1, characterized in that: In step 4, the topological data comparison is a specific topological data comparison of the power system, which includes line-transformer relationships, household-transformer relationships, and phase relationships. Through topological data analysis, the overall structure and pattern of the data are captured, and the similarities and differences between the data are compared and identified, thereby discovering the potential structure and relationship in the data.

7. A method for correcting deviation of power system topology similarity data according to claim 5, characterized in that: In the process of topological data comparison, for the similarity function that defines the node attributes, the nodes with numerical attributes use the Euclidean distance measurement method, assuming that the attribute vectors of two nodes A and B are and For nodes with classification attributes, the Jaccard coefficient method can be used. For the classification attribute set S of two nodes, A and S B , 8. A method for correcting deviation of power system topology similarity data according to claim 5, characterized in that: The line-to-transformer relationship refers to the connection relationship between the power line and the transformer, and the line-to-transformer relationship is determined based on the K nearest neighbor algorithm and rules; The household-to-transformer relationship is the process of converting high-voltage power into low-voltage power through a transformer and supplying it to users. By matching and comparing the load data of users and transformers, accurate identification of the household-to-transformer relationship is achieved, providing accurate data support for load forecasting and electric energy metering of the power system. The phase relationship refers to the relative phase difference between the voltage and current of each node in the power system. By measuring and calculating the node voltage phase difference, the phase relationship between the nodes can be determined, providing accurate data support for the stability analysis and short-circuit calculation of the power system.

9. A method for correcting deviation of power system topology similarity data according to claim 1, characterized in that: In step 4, after completing the topological data comparison, the topological characteristics of the data are identified by analyzing the topological properties of the data, including connectivity and holes, to understand the overall structure and local differences of the data. Then, by comparing the topological characteristics of different data sets, the difference data is identified, thereby tolerating small-scale deformation and distortion of the data, and applying it to the difference data processing and separation of the topological relationships of complex types of power systems.

10. A method for correcting deviation of power system topology similarity data according to claim 1, characterized in that: In the step five, the corresponding difference data is separated by analyzing the topological characteristics of the same data set, and then the difference data is traced and reversely mapped from the topological model to the corresponding power system model, so as to capture the data deviations in each topological level in the power system model, and for the data item deviations, the corresponding data items are supplemented according to the topological mapping relationship, and the corresponding data items are associated with the power system architecture. For the data value deviations, the thresholds are adjusted for the corresponding data, and the model data is automatically modified within the effective threshold range.

Images