An Automatic Verification Method and System for Distribution Network Protection Parameters Based on Graph Topology
By automatically parsing SVG graphic files to construct distribution network topology diagrams and performing multi-level protection coordination verification, the efficiency and accuracy problems of the traditional manual verification mode are solved, realizing fast and accurate verification of distribution network protection parameters and improving the stability and reliability of the power grid.
Patent Information
- Application Number
- CN202511181365.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-22
- Publication Date
- 2025-12-02
- Estimated Expiration
- 2045-08-22
AI Technical Summary
Traditional manual verification methods suffer from problems such as long calculation time, frequent data entry errors, and oversights in protection coordination logic in complex multi-branch mesh topology distribution networks with distributed power sources. They cannot automatically parse grid graphic files, resulting in long setting cycles and low reliability.
By automatically parsing SVG graphic files to construct a distribution network topology, and combining protection setting data and zero-sequence current data, a multi-level protection coordination verification algorithm is adopted to achieve automated verification, including data preparation, topology construction, hierarchical analysis and result visualization, supporting line parameter and protection setting verification.
It significantly improves the efficiency and accuracy of distribution network parameter verification, reduces human error, shortens verification time, lowers the technical threshold, ensures that protection coordination logic is consistent with the actual network, and enhances the stability and reliability of power grid operation.
Smart Images

Figure CN120675000B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to an automatic verification method and system for distribution network protection parameters based on graph topology, belonging to the field of power transmission and distribution technology. Background Technology
[0002] In the distribution network operation system, the accuracy of protection settings directly determines the accuracy of fault isolation and the reliability of the power grid. The traditional manual verification mode has significant bottlenecks. It requires manual calculation of the short-circuit current, action time difference, and sensitivity of protection at each level. In complex mesh topologies with multiple branches and distributed power sources, it is easy to miss key coordination nodes. For large-scale networks with thousands of feeders, the calculation time of several hours for a single line leads to a setting cycle of several weeks. Moreover, when manually verifying data, problems such as impedance parameter input errors and confusion of current transformer (CT) ratios frequently occur, creating hidden dangers of over-level tripping or failure to operate. While existing technologies provide basic computing tools, they heavily rely on manually drawn distribution network topology maps and cannot automatically parse network graphic files, such as Scalable Vector Graphics (SVG) or Geographic Information System (GIS) data. Furthermore, they lack intelligent verification capabilities for cross-level protection coordination. For example, key logic such as the timing differences between 10kV outgoing line protection and mainline feeder terminal unit (FTU) protection, and the sensitivity connection between branch line protection and user boundary switches, all require manual derivation, leading to a disconnect between setting strategies and the actual network. Therefore, a deeply automated solution is urgently needed. This solution should directly extract device symbols, connection relationships, and electrical parameters from SVG using an intelligent graphics parsing engine, automatically constructing a topology network with hierarchical attributes. Based on a multi-level coordination verification algorithm, differentiated strategies should be implemented layer by layer (e.g., focusing on time-difference coordination on the high-voltage side and strengthening sensitivity verification on the low-voltage side), completing closed-loop verification of the entire network protection setting logic within seconds, fundamentally solving the dual dilemma of efficiency and accuracy. Summary of the Invention
[0003] The purpose of this invention is to provide an automatic verification method and system for distribution network protection parameters based on graph topology. This method can automatically parse SVG graphics to construct a distribution network topology diagram, and combine protection setting data and zero-sequence current data to perform multi-level protection coordination verification, effectively improving the efficiency and accuracy of distribution network parameter verification.
[0004] This invention is achieved through the following technical solution: An automatic verification method for distribution network protection parameters based on graph topology, comprising the following steps:
[0005] Step 1: Data Preparation: Obtain substation outgoing line protection settings, terminal protection settings, line zero-sequence current data, substation-level topology, line parameters, and operating status data of lines, switches, and distribution transformers;
[0006] Step 2, SVG parsing and distribution network topology construction: Parse the SVG graphic file of the power distribution line, extract equipment information, construct nodes and edges to form the distribution network topology, and complete parameter labeling in conjunction with the line ledger;
[0007] Step 3: Subgraph Extraction and Hierarchical Analysis: Extract nodes based on equipment type codes to generate a connected subgraph. Using the substation outgoing switch as the root node, perform hierarchical analysis and assign levels using a breadth-first search algorithm. Remove edges between nodes at the same level to form a tree structure.
[0008] Step 4: Line Parameter and Protection Setting Verification: This includes line parameter verification and protection setting verification; the line parameter verification includes branch segment verification and switch position verification; the protection setting verification includes overcurrent stage I verification, overcurrent stage II verification, and zero-sequence protection verification.
[0009] Step 5: Result Visualization: Draw a tree topology diagram, color the nodes according to the level, and simultaneously display the anomaly list tables for overcurrent segment I, segment II, and zero-sequence protection, and output an SVG graphic file.
[0010] Specifically, in step one, the substation outgoing line protection settings include the settings and times for overcurrent stage I, II, and III, and the reclosing time; the terminal protection settings include the enabling / disabling status, settings, and time of the switch overcurrent stage I and II, the enabling / disabling status, settings, time, and reclosing parameters of the zero-sequence protection; and the operating status data includes switch action information, power outage and restoration events, and voltage and current measurement data.
[0011] Specifically, in step four, the branch segment verification includes:
[0012] Calculate the total impedance of the branch segment based on the conductor segment parameters;
[0013] Define capacity-length matching degree, divide intervals based on regional characteristics, and identify abnormal routes using the isolated forest algorithm;
[0014] A normalized formula is used to calculate the line anomaly score, and lines that exceed the line anomaly score threshold are marked as pending verification.
[0015] Specifically, in step four, the switch position verification includes:
[0016] When a fault is merged, the switch action information is used to form an action set, and an action pool is established according to the region type and protection strategy.
[0017] Calculate the overlap between the fault area and the power outage area, and mark switches with an overlap lower than the overlap threshold as pending verification;
[0018] Inefficient switches are identified by performing two-dimensional clustering based on the average number of switching operations and the frequency of fault occurrence.
[0019] Specifically, in step four, the overcurrent segment I verification is performed as follows:
[0020] For 10kV lines originating from non-35kV substations, the overcurrent protection setting of the current-level switch is 70-95% of that of the upper-level switch, and the reclosing time of the current-level switch is less than that of the upper-level switch.
[0021] For 10kV lines originating from 35kV substations, the protection setting should be verified to be no less than 70% of the maximum three-phase short-circuit current at the end of the line.
[0022] Specifically, in step four, the zero-sequence protection verification must meet the following requirements: the zero-sequence setting value does not exceed the upper-level zero-sequence setting value, the zero-sequence time does not exceed the upper-level zero-sequence time, the zero-sequence setting value is less than the zero-sequence current on the power supply side and greater than the zero-sequence current on the load side, and the difference between the zero-sequence current on the power supply side and the load side is greater than 3A.
[0023] This invention also provides an automatic verification system for distribution network protection parameters based on graph topology, comprising:
[0024] Data reading module: used to acquire substation outgoing line protection settings, terminal protection settings, line zero-sequence current data, topology relationships and operating status data;
[0025] Data fusion module: used to associate multi-source data based on line name and unique ID, and perform consistency verification;
[0026] Distribution network topology construction module: used to parse SVG graphic files, extract equipment information and connection relationships, and construct a distribution network topology map with parameter annotations;
[0027] Hierarchical analysis engine: used to extract nodes based on device type codes, generate connected subgraphs, divide levels, and form a tree structure;
[0028] Verification module: includes a line parameter verification submodule and a protection setting verification submodule; the line parameter verification submodule is used for branch line segment verification and switch position verification; the protection setting verification submodule is used for overcurrent stage I, overcurrent stage II and zero sequence protection verification;
[0029] Visualization module: Used to draw tree topology diagrams and exception list tables, and output SVG graphic files.
[0030] Furthermore, the data reading module obtains data from the scheduling system, power distribution automation system, and electricity consumption information collection system through the data platform and business platform interfaces, and supports multiple access methods such as files, databases, and interfaces.
[0031] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the instructions are executed by the processor, the processor implements the above-described automatic verification method for distribution network protection parameters based on graph topology.
[0032] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-described automatic verification method for distribution network protection parameters based on graph topology.
[0033] The present invention has the following advantages:
[0034] By automatically parsing SVG graphic files to construct power distribution network topology diagrams, the traditional manual topology drawing time of several hours to several days is shortened to less than 1 minute; the batch processing function supports simultaneous verification of multiple SVG graphic files, compressing the verification cycle of thousands of feeders from several weeks to several hours, significantly reducing labor costs.
[0035] Eliminate the risk of errors in manually entering impedance parameters, CT ratios, and other data. Use intelligent methods such as isolated forest algorithm and two-dimensional clustering to identify abnormal lines and switches, reduce oversights caused by manually deriving protection coordination logic, and reduce the risk of over-level tripping or failure to operate.
[0036] There is no need to rely on manually drawing power distribution network topology diagrams or mastering specialized drawing software; the verification process can be started using only standard SVG graphic files, lowering the technical threshold. Visual integration enables diagram-data linkage, and abnormal nodes are highlighted for easy and quick problem location.
[0037] By using hierarchical pruning technology to simplify the mesh topology into a tree structure, and combining it with differentiated verification strategies based on voltage levels (such as distinguishing between 10kV lines from 35kV and non-35kV substations), it can adapt to complex distribution networks with distributed power sources and multiple branches, ensuring that the protection coordination logic is consistent with the actual network.
[0038] By combining full-link protection with verification (from substation outgoing lines to user-side switches), we ensure step-by-step isolation during faults and transfer of power to non-faulty areas, thereby reducing the scope of power outages and improving the stability and reliability of the distribution network. Attached Figure Description
[0039] Figure 1 This is a flowchart of the method of the present invention.
[0040] Figure 2 This is a schematic diagram of the system modules of the present invention. Detailed Implementation
[0041] The present invention will be further described below with reference to the embodiments. It should be noted that the following embodiments are only used to further illustrate the present invention and should not be construed as limiting the scope of protection of the present invention. Some non-essential improvements and adjustments made by those skilled in the art based on the above-described invention are still within the scope of protection of the present invention.
[0042] Example 1
[0043] like Figure 1 As shown, an automatic verification method for distribution network protection parameters based on graph topology includes:
[0044] Step 1: Data Preparation
[0045] Obtain the protection settings for the outgoing lines of the substation, including the substation, switch, overcurrent stage I, overcurrent stage II and overcurrent stage III settings and times, reclosing time, etc. Among them, overcurrent stage I is instantaneous overcurrent protection, set to avoid the maximum short-circuit current at the end of the line; overcurrent stage II is time-limited overcurrent protection, which cooperates with the overcurrent stage I of the adjacent line; overcurrent stage III is time-limited overcurrent protection, set to avoid the maximum load current.
[0046] Obtain terminal protection settings: including the overcurrent stage I and overcurrent stage II alarms and output activation / deactivation status, settings, and time of the switch; the zero-sequence protection activation / deactivation status, settings, and time; and the reclosing activation / deactivation and time, etc.
[0047] Obtain line zero-sequence current data: including the zero-sequence current on the power supply side and the zero-sequence current on the load side of the line;
[0048] Obtain the substation's topology and line parameters; obtain the operating status of lines, switches, and distribution transformers, including switch operation information, power outage and restoration events at all levels, and voltage and current measurement data.
[0049] The source systems for acquiring data include power dispatching systems, distribution automation systems, power homogenization systems, and electricity consumption information collection systems. The acquisition methods are data middleware and business middleware interfaces.
[0050] Step 2: SVG parsing and distribution network topology construction: Parse the SVG graphic file of the power distribution line, extract the equipment information from the metadata tags, construct nodes and edges, with the node ID being the unique identifier of the equipment and the edges representing the equipment connection relationship, to form the distribution network topology map, and complete the parameter labeling of the distribution network topology map by combining the line ledger.
[0051] Step 3: Subgraph Extraction and Hierarchical Analysis: Extract nodes based on equipment type codes (e.g., 11100000 represents line switches, 30500000 represents substation outgoing switches) to generate a connected subgraph; using the substation outgoing switch (equipment type code 30500000) as the root node, perform hierarchical analysis using a breadth-first search (BFS) algorithm to assign a level to each node; remove edges between nodes at the same level, retain tie switch nodes and edges, mark them as backup connections, add a simulated scenario of tie switch closure during fault verification, evaluate the effectiveness of protection coordination under power transfer conditions, and finally form a tree structure.
[0052] Step 4: Verification of line parameters and protection settings:
[0053] Step 4.1, Line Parameter Verification
[0054] The line parameters that affect protection settings mainly include line length, topology, and switch position. The following section verifies the branch line parameters and switch position.
[0055] (1) Branch segment verification
[0056] Based on the construction and hierarchical analysis results of the distribution network topology, a single-line topology is formed according to the substation-feeder section-branch section-distribution transformer. The ledger and topological relationship of the substation-feeder section-distribution transformer are generally relatively accurate, while the accuracy and rationality of the ledger and topological relationship of the branch section-distribution transformer need to be further verified.
[0057] Calculate the total impedance at any location on the branch line based on the conductor segmentation parameters:
[0058] ,
[0059] ,
[0060] in, For the branch segment to be calculated, Indicates the number of segments. For the first Resistance per unit length The total resistance of the branch segment is... For the first Reactance per unit length of segment For the total reactance of the branch line segment, This represents the segment length.
[0061] Based on the transformer capacity connected to the branch line and the branch line length, the capacity-length matching degree is defined as follows:
[0062] ,
[0063] Where Q represents the total capacity of the distribution transformers connected to the branch line segment. The capacity-length matching degree of the branch segment to be calculated is given. Considering regional characteristics such as urban, township, and rural areas, load density, and load growth, the capacity-length matching degree ρ is aggregated according to region. The capacity-length matching degree results are standardized, and intervals are divided according to the total transformer capacity and the branch segment length. Within a divided interval, the feature space is randomly divided according to the capacity-length matching degree. Sample features are extracted multiple times using a sampling mode with or without replacement, and t binary trees are constructed. Abnormal lines that do not meet the accuracy and rationality of the parameters are isolated earlier in the recursive process. The isolated forest identifies abnormal lines as observations with shorter average paths on the trees.
[0064] Line anomaly scores are calculated using a normalized method:
[0065] ,
[0066] in, For the first The number of line samples in this range Anomaly scores in each line sample For the first The path length of each route sample in the binary tree. For the first The mean path length of each route sample across t binary trees. for The average path length on a sample binary tree. The closer the score is to 1, the more obvious the abnormal characteristics are. This indicates that the capacity-length matching degree between the transformer capacity connected to the corresponding branch segment and the line length is low. There may be errors in line length, branch-transformer topology errors, unreasonable branch segmentation, etc. Set a line abnormality score threshold, and further verify lines that exceed the line abnormality score threshold.
[0067] (2) Switch position verification
[0068] When the switch position is configured properly, it should be able to correctly isolate the faulty area and transfer power to the non-faulty area in the event of a fault; otherwise, isolation may fail or the faulty area may be expanded.
[0069] Based on the established topological hierarchy, the action information of each switch during a fault is merged to form a set of actions for each level of switches during the fault, including the opening and closing status and frequency of each level of switches on the line during the fault. Depending on the implemented strategy, action pools are established for similar fault sets within the region. For example, if a line has a protection strategy but no automation strategy, all switches on that line are ordered from substation to end point as K=[K1,K2,…,K…]. s The number of times the switch tripped during a certain fault is F = [F1, F2, ..., F]. s, the switch closing times are \(H = [H_1, H_2, \ldots, H]\) s , the fault range is \(G = [G_1, G_2, \ldots, G]\) s , the affected range is \(W = [W_1, W_2, \ldots, W]\) s , where \(s\) is the number of automation and protection switches on this line, \(K\) is the switch sorting on the line, \(K\) s is the \(s\)th switch, \(F\) is the number of breaker trips of each level of switches during the fault, \(F\) s is the number of breaker trips of the \(s\)th switch, \(H\) is the number of switch closings of each level of switches during the fault, \(H\) s is the number of switch closings of the \(s\)th switch, \(G\) is the number of substations in the fault range under each level of switches during the fault, \(G\) s is the number of substations in the fault range under the \(s\)th switch during the fault, \(W\) is the number of substations with power outages in the protection range of each level of switches during the fault, \(W\) s is the number of substations with power outages in the protection range of the \(s\)th switch during the fault. According to the regional type classification methods such as urban areas, towns, and rural areas, collect the switch opening and closing states and times during faults of similar strategy lines within a certain range, and establish an action pool \(R\). Based on the action pool \(R\), further verify the rationality of the switch position configuration.
[0070] Calculate the overlap degree \(\gamma\) between the fault area and the power outage area for each fault:
[0071] ,
[0072] Set the overlap degree threshold \(y\). If \(\gamma < y\) occurs more than 2 times, or the average value of the overlap degree within the calculation period is less than the overlap degree threshold, it indicates a low overlap degree, an expanded power outage range during the fault, and a poor effect of the corresponding switch configuration position. It is necessary to further verify whether the switch position and protection setting values are reasonable.
[0073] Further, conduct an evaluation of the switch action efficiency, cluster the switch opening and closing times in similar faults, and identify high-efficiency and low-efficiency switch groups.
[0074] Construct a feature engineering: average switch opening and closing times = (switch opening and closing times \((F + H)\)) / number of faults, fault occurrence frequency = number of faults \(c\) occurring within the protection range of the switch at this level / number of faults. It can be determined that the fault occurs within the protection range of the \(c\)th switch through the number of substations \(G\) in the fault range under each level of switches during the fault.
[0075] According to the two features of the average switch opening and closing times and the fault occurrence frequency, set the weights of the two features, and use a clustering method, such as kMeans or random forest, to perform two-dimensional weighted clustering on the switches. Pay attention to two types of switches: low action times / high fault frequencies and high action times / low fault frequencies. It is necessary to further verify whether the switch position and protection setting values are reasonable.
[0076] Step 4.2, Protection Setting Verification
[0077] (1) Overcurrent Section I Verification
[0078] For 10kV lines originating from non-35kV substations: trace upwards from the terminal node to find the nearest upstream switch (including substation outgoing switches) that has been put into overcurrent protection stage I. Verification should be performed according to the principle that the current setting should not exceed 95% of the upstream setting and should not be lower than 70% of the upstream setting (or 70% of the maximum three-phase short-circuit current at the end of the line), and the reclosing time should be less than the upstream reclosing time. For 10kV lines originating from 35kV substations, only the coordination between the terminal node and the upstream non-root node (non-substation outgoing switches) should be verified, using the lower limit of the setting as 70% of the maximum three-phase short-circuit current at the end of the line.
[0079] (2) Overcurrent stage II verification
[0080] Tracing upwards from the terminal node, find the nearest upstream switch that has been put into overcurrent stage II protection (overcurrent stage III setting is used for substation outgoing switches), and verify it according to the principle that the current setting does not exceed the upstream setting and is not lower than 90% of the upstream setting.
[0081] (3) Zero-sequence protection verification
[0082] Tracing upwards from the terminal node, find the nearest upstream switch that has been put into zero-sequence protection, and verify it according to the following rules: the zero-sequence setting value does not exceed the upstream zero-sequence setting value, the zero-sequence time does not exceed the upstream zero-sequence time, the zero-sequence setting value is less than the zero-sequence current on the power supply side, the zero-sequence setting value is greater than the zero-sequence current on the load side, and the difference between the zero-sequence current on the power supply side and the zero-sequence current on the load side is greater than 3A.
[0083] Step 5: Result Visualization: Draw a tree topology diagram, color the nodes according to the level, display three tables of verification results on the right (overcurrent segment I, overcurrent segment II, and zero-sequence protection anomaly list), output an SVG graphic file and save it.
[0084] This method can automatically extract topology relationships from SVG graphic files and automatically complete the overcurrent I, overcurrent II, and zero-sequence automatic verification of line terminal switches, greatly improving the accuracy of switch action in distribution network protection.
[0085] (1) Automated topology construction: The distribution network topology is automatically constructed by parsing the SVG graphic file of the distribution line. No manual drawing is required, which completely eliminates the tedious and time-consuming process of drawing or importing complex distribution network topology diagrams in special software. Only a standard SVG graphic file is required, and the system can automatically identify and extract electrical connection relationships, reducing the topology construction time from several hours or even days to less than 1 minute. This reduces human error. The automated parsing directly extracts information from the source graphic, minimizing the topology structure errors introduced by manual intervention and significantly improving the accuracy of the input data for subsequent verification. There is no need to master the complex operation of specific drawing software. Topology construction can be started simply by preparing or exporting a standard SVG graphic file, which reduces the technical threshold and operational complexity.
[0086] (2) Hierarchical pruning technology: Based on breadth-first search (BFS), hierarchical analysis is performed to prune connections at the same level, forming a tree structure and simplifying the verification logic. By intelligently identifying electrical levels (such as substation outgoing lines -> main lines -> branch lines -> users) and accurately pruning redundant connections within the same voltage level (such as tie lines at the same level), the original network is automatically simplified into a clear, loop-free tree-like radial structure.
[0087] (3) Multi-level protection coordination verification: Considering the coordination between substation outgoing line protection and terminal protection, different strategies are adopted according to voltage level. Through the contents described in steps 4.1 and 4.2, the full-link coordination verification is realized. The complete coordination chain from the protection device of the 10kV outgoing line switch of the substation (master station level) to the feeder terminal (FTU / DTU) protection (such as overcurrent and zero-sequence protection) and user-side protection (such as user boundary switch) in the middle and later sections of the line is systematically considered to ensure that selective and step-by-step tripping can be achieved in case of fault. The refined verification strategy applies different verification rules and criteria to the protection equipment characteristics and coordination requirements of different voltage levels (such as 110kV / 35kV substation outgoing line, 10kV main line, 10kV branch line, 0.4kV user side).
[0088] For example: for lines with frequent power outages or low reliability, focus on switch position verification and protection setting verification; for lines with many ground faults, focus on zero-sequence protection verification; for substation outgoing lines, focus on the differential coordination with the protection at the head end of the lower-level main line; for main line protection, focus on the differential coordination with branch line protection and user boundary protection, as well as sensitivity verification; for user-side protection, focus on the rationality of its own settings and coordination with the upper-level protection; improve the reliability of the protection system by minimizing the risks of cascading tripping (lower-level faults causing upper-level tripping) and failure to operate (protection failing to operate when it should) through the coordination verification in steps 4.1-4.2, and optimize line parameters and protection setting settings.
[0089] (4) Visual Integration: The distribution network topology diagram and the verification result table are integrated into one diagram to intuitively display abnormal information. The abstract verification results (table data) and the intuitive distribution network topology diagram are closely combined in the same view to achieve diagram-data linkage. Anomalies are accurately located by highlighting (e.g., color change, flashing, special marking) the specific equipment (switches, line segments) or protection devices where the problem is found on the distribution network topology diagram, thereby locating the fault point or the point of poor coordination; the results are presented intuitively by clearly marking key verification results (e.g., unreasonable time difference, setting over limit) next to the equipment or line or linking to the detailed result table, making complex technical conclusions clear at a glance.
[0090] (5) Batch processing: Supports batch processing of multiple SVG graphic files, automatically generates summary reports, and can import and automatically process all SVG graphic files of the entire area, multiple feeders or the entire project at one time. The processing capability far exceeds that of single file operation, greatly improving work efficiency. For scenarios that require analysis of a large number of lines or the entire network protection coordination (such as setting value quarterly, planning evaluation, and post-renovation acceptance), the batch processing function compresses the manual work that originally required several weeks to several hours or even less, freeing up a lot of human resources. The summary report provides a macro view of the entire network protection coordination status, which makes it easy for managers to quickly grasp the overall weak links and common problems, and to prioritize and allocate resources.
[0091] Example 2
[0092] like Figure 2 As shown, an automatic verification system for distribution network protection parameters based on graph topology includes the following modules:
[0093] The data reading module is responsible for collecting and accessing all data, providing fundamental data support for subsequent verification. The data scope includes:
[0094] Substation outgoing line protection settings: including substation name, switch number, overcurrent stage I / II / III settings and operating time, reclosing time, etc. (where overcurrent stage I is instantaneous overcurrent protection, set to avoid the maximum short-circuit current at the end of the line; overcurrent stage II is time-limited overcurrent protection, coordinated with overcurrent stage I of adjacent lines; overcurrent stage III is time-limited overcurrent protection, set to avoid the maximum load current).
[0095] Terminal protection settings include alarms and output activation / deactivation status, settings, and operating time for switch overcurrent stage I / II, activation / deactivation status, settings, and operating time for zero-sequence protection, and reclosing activation / deactivation status and time, etc.
[0096] Line zero-sequence current data: zero-sequence current on the power supply side and zero-sequence current on the load side of the line.
[0097] Topology and operational data: Equipment topology relationships below the substation, line parameters (length, impedance, etc.), and operational status of lines, switches, and transformers (switch action information, power outage and restoration events, voltage / current measurement data, etc.).
[0098] Data sources and access methods: Data is obtained from source systems such as scheduling systems, power distribution automation systems, and electricity consumption information collection systems through the data platform and business platform interfaces. Multiple access methods are supported, including files (such as SVG graphic files), databases (structured fixed value data), and interfaces (real-time running data).
[0099] The data fusion module is used to realize the association and consistency verification of multi-source data, ensure data continuity, and provide a unified data foundation for topology construction and verification.
[0100] Association rules: Based on line name, unique ID (such as switch number, line ID) and the mapping relationship between the dispatching system and the distribution automation system, the substation outgoing line protection settings, terminal protection settings, zero-sequence current data, topology parameters, etc. are associated into a unified dataset;
[0101] Consistency verification: Automatically verifies data format (such as time format, current unit) and logical correlation (such as the matching of switch number with its corresponding circuit), eliminates duplicate or conflicting data, and ensures data accuracy.
[0102] The power distribution network topology construction module is used to automatically parse SVG graphics and construct power distribution network topology maps.
[0103] The hierarchical analysis engine is used to extract subgraphs and divide hierarchically into distribution network topology diagrams, simplifying the verification logic.
[0104] The verification module is used to perform automated verification of line parameters and protection settings;
[0105] The verification module includes:
[0106] The line parameter verification submodule is used to perform branch segment verification and switch position verification.
[0107] The protection setting verification submodule is used to perform overcurrent stage I verification, overcurrent stage II verification, and zero-sequence protection verification.
[0108] The visualization module is used to present the verification results in a visual format.
[0109] Example 3
[0110] This embodiment provides an electronic device, including a memory and a processor. The memory stores computer-readable instructions. When the instructions are executed by the processor, the processor implements the automatic verification method for distribution network protection parameters based on graph topology as described in Embodiment 1.
[0111] Example 4
[0112] This embodiment provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the automatic verification method for distribution network protection parameters based on graph topology as described in Embodiment 1.
[0113] The above description merely illustrates preferred embodiments of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make modifications or alterations to the above-disclosed content to create equivalent embodiments. However, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention, without departing from the scope of the present invention, shall still fall within the protection scope of the present invention.
Claims
1. An automatic verification method for distribution network protection parameters based on graph topology, characterized in that, Includes the following steps: Step 1: Data Preparation: Obtain substation outgoing line protection settings, terminal protection settings, line zero-sequence current data, substation-level topology, line parameters, and operating status data of lines, switches, and distribution transformers; Step 2, SVG parsing and distribution network topology construction: Parse the SVG graphic file of the power distribution line, extract equipment information, construct nodes and edges to form the distribution network topology, and complete parameter labeling in conjunction with the line ledger; Step 3: Subgraph Extraction and Hierarchical Analysis: Extract nodes based on equipment type codes to generate a connected subgraph. Using the substation outgoing switch as the root node, perform hierarchical analysis and assign levels using a breadth-first search algorithm. Remove edges between nodes at the same level to form a tree structure. Step 4: Verification of line parameters and protection settings: This includes verification of line parameters and verification of protection settings; The line parameter verification includes branch segment verification and switch position verification; The branch segment verification includes: calculating the total impedance of the branch segment based on the conductor segmentation parameters; defining the capacity-length matching degree; dividing the interval based on regional characteristics; and identifying abnormal lines through the isolated forest algorithm. The line anomaly score is calculated using a normalized formula, and lines that exceed the line anomaly score threshold are marked as pending verification. The switch position verification includes: When a fault occurs, the switch action information is used to form an action set, and an action pool is established according to the region type and protection strategy. The overlap between the fault area and the power outage area is calculated, and switches with an overlap lower than the overlap threshold are marked as pending verification. Two-dimensional clustering is performed based on the average number of switch opening and closing times and the frequency of fault occurrence to identify inefficient switches. The protection setting verification includes overcurrent stage I verification, overcurrent stage II verification, and zero-sequence protection verification. Step 5: Result Visualization: Draw a tree topology diagram, color the nodes according to the level, and simultaneously display the anomaly list tables for overcurrent segment I, segment II, and zero-sequence protection, and output an SVG graphic file.
2. The automatic verification method for distribution network protection parameters according to claim 1, characterized in that, In step one, the substation outgoing line protection settings include the settings and times for overcurrent stage I, II, and III, and the reclosing time; the terminal protection settings include the on / off status, settings, and time of the switch overcurrent stage I and II, the on / off status, settings, time, and reclosing parameters of the zero-sequence protection; and the operating status data includes switch action information, power outage and restoration events, and voltage and current measurement data.
3. The automatic verification method for distribution network protection parameters according to claim 1, characterized in that, In step four, the overcurrent segment I verification is performed: For lines not in 35kV substations, the current setting value should be 70%-95% of the previous setting value, and the reclosing time should be less than that of the previous setting value. For outgoing lines of 35kV substations, the verification setting value shall not be less than 70% of the maximum three-phase short-circuit current at the end of the line.
4. The automatic verification method for distribution network protection parameters according to claim 1, characterized in that, In step four, the zero-sequence protection verification must meet the following requirements: the zero-sequence setting value does not exceed the upper-level zero-sequence setting value, the zero-sequence time does not exceed the upper-level zero-sequence time, the zero-sequence setting value is less than the zero-sequence current on the power supply side and greater than the zero-sequence current on the load side, and the difference between the zero-sequence current on the power supply side and the load side is greater than 3A.
5. A system for implementing the automatic verification method for distribution network protection parameters according to any one of claims 1-4, characterized in that, include: Data reading module: used to acquire substation outgoing line protection settings, terminal protection settings, line zero-sequence current data, topology relationships and operating status data; Data fusion module: used to associate multi-source data based on line name and unique ID, and perform consistency verification; Distribution network topology construction module: used to parse SVG graphic files, extract equipment information and connection relationships, and construct a distribution network topology map with parameter annotations; Hierarchical analysis engine: used to extract nodes based on device type codes, generate connected subgraphs, divide levels, and form a tree structure; Verification module: includes a line parameter verification submodule and a protection setting verification submodule; The line parameter verification submodule is used for branch line segment verification and switch position verification; The protection setting verification submodule is used for overcurrent stage I, overcurrent stage II and zero-sequence protection verification; Visualization module: Used to draw tree topology diagrams and exception list tables, and output SVG graphic files.
6. The system according to claim 5, characterized in that, The data reading module obtains data from the scheduling system, power distribution automation system, and electricity consumption information collection system through the data platform and business platform interfaces, and supports multiple access methods such as files, databases, and interfaces.
7. An electronic device comprising a memory and a processor, wherein the memory stores computer-readable instructions, characterized in that, When the instruction is executed by the processor, the processor implements the automatic verification method for distribution network protection parameters as described in any one of claims 1-4.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the automatic verification method for distribution network protection parameters as described in any one of claims 1-4.
Citation Information
Patent Citations
Constant value checking and evaluating system for power distribution network
CN115021199A
Power distribution network topology automatic identification method and system based on big data
CN116089849A