A power distribution network fault isolation method and system based on power distribution terminal fault information

Abstract

The application relates to the technical field of power distribution faults, and provides a power distribution network fault isolation method and system based on power distribution terminal fault information, which comprises the following steps: analyzing fault recording data to obtain a fault event set containing electrical quantity characteristics at a fault moment and fault indication information; constructing a fault topology graph reflecting the electrical connection relationship between power distribution terminals according to the fault event set, and marking a suspected fault section in the fault topology graph according to the fault indication information; cooperatively analyzing the fault recording data in the suspected fault section, comparing the electrical quantity characteristic differences at the two ends of the suspected fault section, and locating a real fault section; generating an isolation operation instruction according to the position of the real fault section in the fault topology graph; sending the isolation operation instruction to the power distribution terminal, abnormally feeding back the isolation operation instruction, and obtaining an instruction execution closed-loop verification result of the power distribution terminal; and the application can improve the efficiency of power distribution network fault isolation.

Description

Technical Field

[0001] This invention relates to the field of power distribution fault technology, and in particular to a method and system for power distribution fault isolation based on power distribution terminal fault information. Background Technology

[0002] Fault handling in distribution networks heavily relies on fault waveforms and fault indication information uploaded by distribution terminals. Existing fault isolation schemes often rely solely on fault flag bits from a single terminal to roughly determine the fault section, without conducting collaborative verification of synchronous waveform data from multiple terminals. Traditional schemes lack a unified timing calibration and data integrity verification process. The electrical quantity data reported by different distribution terminals exhibits time discrepancies and sampling gaps, directly leading to significant deviations in the initial fault section assessment. Many non-faulty lines are mistakenly marked as suspected fault sections, greatly increasing the computational overhead for subsequent fault identification.

[0003] Existing technologies, after screening suspected fault sections, simply compare the fault current amplitude thresholds on both sides to confirm the fault, without constructing standardized feature vectors to quantify the differences in electrical quantities at both ends. This makes them susceptible to load fluctuations and instantaneous interference currents, leading to missed or false judgments. Furthermore, fault isolation command generation only outputs tripping commands directly based on the location of the fault section, without combining the distribution network topology to generate an ordered operation sequence. After issuance, there is a lack of a complete closed-loop process for command issuance, execution feedback, and anomaly verification. Switch tripping operations are prone to timing conflicts and failures in terminal execution cannot be detected in a timely manner, ultimately prolonging the time for distribution network fault isolation and power restoration, and reducing the reliability of distribution network operation. Therefore, improving the efficiency of distribution network fault isolation has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for distributing network fault isolation based on distribution terminal fault information, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for distributing network fault isolation based on distribution terminal fault information, comprising: A. Analyze the fault recording data of the distribution terminal in the distribution network to obtain a set of fault events containing electrical quantity characteristics and fault indication information at the time of the fault; B. Based on the fault event set, construct a fault topology map reflecting the electrical connection relationship between the power distribution terminals, and mark suspected fault sections in the fault topology map according to the fault indication information; C. Perform collaborative analysis on the fault recording data in the suspected fault section, compare the differences in electrical quantity characteristics at both ends of the suspected fault section, and locate the actual fault section of the distribution network; D. Based on the location of the actual fault section in the fault topology diagram, generate isolation operation instructions for the distribution network; E. Send the isolation operation command to the power distribution terminal, and provide anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal.

[0006] Preferably, the step of parsing the fault recording data of the distribution terminal in the distribution network to obtain a fault event set containing electrical quantity characteristics and fault indication information at the time of the fault includes: The fault recording data reported by the distribution terminal in the distribution network is obtained and timestamped to obtain the synchronous recording data of the distribution terminal. Data integrity verification is performed on the recording channels in the synchronous recording data to obtain the electrical quantity sampling sequence of the power distribution terminal; Traverse the electrical quantity sampling sequence, detect abrupt changes in the sampled values ​​in the electrical quantity sampling sequence, and record the time corresponding to the abrupt change position and the amplitude direction of the sampled values ​​before and after the abrupt change position to obtain the electrical quantity characteristics of the power distribution terminal at the time of the fault. Based on the status value of the fault indication flag bit in the power distribution terminal, the electrical quantity characteristics at the time of the fault are associated and matched with the corresponding fault indication information to obtain the fault event set of the power distribution terminal.

[0007] Preferably, the step of constructing a fault topology map reflecting the electrical connection relationships between the power distribution terminals based on the fault event set, and marking suspected fault sections in the fault topology map based on the fault indication information, includes: Based on the static topology connection relationship of the distribution network, the distribution terminal identifiers in the fault event set are mapped as graph nodes, and the graph nodes are connected to obtain the initial topology graph of the distribution network. Based on the fault indication information of the centralized distribution terminal of the fault event, fault status attributes are assigned to the graph nodes in the initial topology graph to obtain the fault topology graph of the distribution network. In the fault topology map, faulty nodes with the fault state attribute and non-faulty nodes without the fault state attribute are identified, and the connecting edges between the faulty nodes and the non-faulty nodes are marked as suspected fault segments.

[0008] Preferably, in the fault topology graph, identifying faulty nodes with the fault state attribute and non-faulty nodes without the fault state attribute, and marking the connecting edges between the faulty nodes and the non-faulty nodes as suspected fault segments, includes: A connectivity analysis is performed on the fault topology graph to obtain a fault node subgraph of the fault topology graph; Traverse the boundary nodes in the fault node subgraph and find the non-fault node that is directly connected to the boundary node and does not have the fault state attribute. The connecting edge between the boundary node and the non-faulty node is identified as a suspected faulty section.

[0009] Preferably, the step of collaboratively analyzing the fault recording data in the suspected fault section and comparing the differences in electrical quantity characteristics at both ends of the suspected fault section to locate the actual fault section of the distribution network includes: The first end characteristic and the second end characteristic of the suspected fault section are obtained by extracting the fault waveform data reported from both ends of the suspected fault section to characterize the fault current amplitude. The first end feature and the second end feature are normalized to obtain the first end normalized feature and the second end normalized feature of the suspected fault segment. The electrical quantity characteristic difference degree between the two ends of the suspected fault section is calculated based on the first end normalized characteristic quantity and the second end normalized characteristic quantity. The electrical quantity characteristic difference is compared with a preset difference threshold. When the electrical quantity characteristic difference is greater than the difference threshold, the suspected fault section is determined as the actual fault section of the distribution network.

[0010] Preferably, the step of calculating the electrical quantity characteristic difference between the two ends of the suspected fault section based on the first end normalized characteristic quantity and the second end normalized characteristic quantity includes: The first-end normalized feature quantity and the second-end normalized feature quantity are vectorized respectively to obtain the first-end feature vector and the second-end feature vector of the suspected fault segment. Based on the first end feature vector and the second end feature vector, the electrical quantity characteristic difference degree between the two ends of the suspected fault section is calculated, wherein the calculation formula for the electrical quantity characteristic difference degree is: ; in, This indicates the degree of difference in the electrical quantity characteristics. The dimension of the feature vector. Represents the first end feature vector. One portion, Represents the second end feature vector. One portion, Indicates the component index of the eigenvector.

[0011] Preferably, generating the isolation operation command for the distribution network based on the location of the actual faulty section in the fault topology map includes: In the fault topology diagram, the topology of the distribution network is divided into a fault-side sub-diagram and a non-fault-side sub-diagram, with the actual fault section as the boundary. Traverse the fault side subgraph, extract the switch equipment identifiers that are directly connected to the actual fault section in the fault side subgraph, and obtain the first isolation device set of the distribution network; Traverse the non-faulty side subgraph, extract the switch equipment identifiers that are directly connected to the actual faulty section in the non-faulty side subgraph, and obtain the second isolation device set of the distribution network; The first isolation device set and the second isolation device set are merged and deduplicated to obtain the complete isolation device set of the distribution network, and the isolation operation command of the distribution network is generated based on the complete isolation device set.

[0012] Preferably, generating the isolation operation command for the distribution network based on the complete isolation device set includes: Obtain the topology hierarchy information of the complete isolation equipment centralized switching equipment in the fault topology diagram; Arrange the switching devices in ascending order of the topology hierarchy information to obtain the isolation operation sequence of the distribution network; A unique operation sequence number is assigned to the switching equipment in the isolation operation sequence, and the operation sequence number, the switching equipment identifier, and the tripping action identifier are associated and encapsulated to obtain the isolation operation instruction of the distribution network.

[0013] Preferably, the step of sending the isolation operation command to the power distribution terminal and providing anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal includes: The isolation operation command is subjected to integrity verification and command sequence consistency verification, and a verification pass identifier for the isolation operation command is generated; Based on the verification pass identifier, the isolation operation instruction is encapsulated into an instruction data packet with a timing synchronization tag and mapped to the instruction receiving queue of the power distribution terminal. According to the timing synchronization tag, the instruction data packets in the instruction receiving queue are prioritized and the execution window is allocated to obtain the instruction execution scheduling sequence of the power distribution terminal; The instruction execution scheduling sequence is sent to the power distribution terminal, and the instruction reception confirmation identifier and execution status feedback data returned by the power distribution terminal are received. Based on the execution status feedback data, the instruction execution scheduling sequence is marked with abnormal status to obtain the instruction execution closed-loop verification result of the power distribution terminal.

[0014] To address the above problems, the present invention also provides a power distribution network fault isolation system based on power distribution terminal fault information, the system comprising: The fault analysis module is used to analyze the fault recording data of the distribution terminal in the distribution network to obtain a set of fault events containing electrical quantity characteristics and fault indication information at the time of the fault. The suspected fault marking module is used to construct a fault topology map reflecting the electrical connection relationship between the power distribution terminals based on the fault event set, and to mark the suspected fault sections in the fault topology map based on the fault indication information. The real fault location module is used to perform collaborative analysis on the fault recording data in the suspected fault section, compare the differences in electrical quantity characteristics at both ends of the suspected fault section, and locate the real fault section of the distribution network. An isolation command generation module is used to generate isolation operation commands for the distribution network based on the location of the actual fault section in the fault topology diagram. The closed-loop feedback module is used to send the isolation operation command to the power distribution terminal, and to provide anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention relies on complete fault recording data from distribution terminals to complete layered analysis and synchronous calibration. It simultaneously extracts the characteristics of electrical quantity mutations and fault indications at the time of the fault to construct a standardized fault event set. Based on topology graph connectivity analysis, it accurately locates suspected fault sections at the boundary. Then, it calculates the fault difference by the similarity of electrical quantity vectors at both ends to complete the determination of the actual fault section. The entire process relies on multi-terminal data collaborative verification to narrow the fault discrimination range, significantly shortens the fault location time in the distribution network, improves the accuracy of fault section identification, and reduces misjudgment and missed judgment.

[0016] 2. This invention generates standardized isolation operation instructions in an orderly manner based on the topology hierarchy of the actual fault section, and is equipped with a time-sequential instruction scheduling and issuance and a closed-loop verification mechanism for terminal execution status. It automatically sorts out the complete set of isolation devices and outputs an orderly operation sequence, realizing the automatic generation, orderly issuance and execution result tracking of fault isolation instructions. This ensures that the fault isolation operation process is standardized and controllable, accelerates the electrical isolation speed of the fault area, shortens the power outage time of the distribution network, and forms a complete execution verification link through instruction anomaly feedback, ensuring the reliable implementation of isolation operations. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a power distribution network fault isolation method based on power distribution terminal fault information, provided in an embodiment of the present invention. Figure 2This is a functional block diagram of a power distribution network fault isolation system based on power distribution terminal fault information, provided in an embodiment of the present invention. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for distributing network fault isolation based on distribution terminal fault information. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for distributing network fault isolation based on distribution terminal fault information can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a distribution network fault isolation method based on distribution terminal fault information according to an embodiment of the present invention. In this embodiment, the distribution network fault isolation method based on distribution terminal fault information includes: A. Analyze the fault recording data of the distribution terminal in the distribution network to obtain a set of fault events containing electrical quantity characteristics and fault indication information at the time of the fault; In this embodiment of the invention, the step of parsing the fault recording data of the distribution terminal in the distribution network to obtain a fault event set containing electrical quantity characteristics and fault indication information at the time of the fault includes: The fault recording data reported by the distribution terminal in the distribution network is obtained and timestamped to obtain the synchronous recording data of the distribution terminal. Data integrity verification is performed on the recording channels in the synchronous recording data to obtain the electrical quantity sampling sequence of the power distribution terminal; Traverse the electrical quantity sampling sequence, detect abrupt changes in the sampled values ​​in the electrical quantity sampling sequence, and record the time corresponding to the abrupt change position and the amplitude direction of the sampled values ​​before and after the abrupt change position to obtain the electrical quantity characteristics of the power distribution terminal at the time of the fault. Based on the status value of the fault indication flag bit in the power distribution terminal, the electrical quantity characteristics at the time of the fault are associated and matched with the corresponding fault indication information to obtain the fault event set of the power distribution terminal.

[0021] The system retrieves the original fault waveform data uploaded by all power distribution terminals that have triggered fault reports from the data acquisition channel of the power distribution network master station. It also retrieves the standard time value output by the Beidou / GPS unified standard clock source built into the master station system. The system extracts the original timestamps of the first and last sampling points in the waveform data of each power distribution terminal. The time deviation between the two sampling points and the standard time value is calculated. The clock drift rate of the entire waveform recording is calculated using linear interpolation. The timestamp labels of all sampling points in the waveform data are corrected point by point according to the drift rate. This ensures that the sampling time deviation of the waveform data of all power distribution terminals participating in the fault analysis does not exceed 1 millisecond and is aligned to the same time base. Finally, the synchronous waveform data of the power distribution terminals is obtained.

[0022] Extract all waveform recording channel data contained in the synchronous waveform recording data of each power distribution terminal. Check the actual number of sampling points for each channel according to the preset total number of sampling points for a single channel. Remove invalid channel data with missing sampling points and retain all valid waveform recording channel data with complete sampling points. Arrange all sampling values ​​in the valid channels in chronological order to form a continuous electrical quantity sampling data set corresponding to each power distribution terminal, and finally obtain the electrical quantity sampling sequence of the power distribution terminal.

[0023] Each sample value in the electrical quantity sampling sequence is read sequentially according to time. The ratio of the absolute value of the difference between two adjacent sample values ​​to the rated current value of that sample point is calculated. When this ratio is greater than a preset step judgment threshold and three consecutive sample points meet the amplitude jump condition, the location of the first amplitude jump sample point is marked as the abrupt change location. The preset step judgment threshold is 20% of the rated current value. The timestamp value corresponding to the abrupt change location is extracted as the fault time. At the same time, the average amplitude and positive and negative current directions of the sample values ​​of the cycle before and after the abrupt change location are recorded. The above time information and amplitude direction information are integrated and stored to finally obtain the electrical quantity characteristics of the distribution terminal at the fault time.

[0024] The status value of the fault indication flag bit stored inside each power distribution terminal is retrieved. The fault indication flag bit is represented by a 1-bit binary value. A status value of 1 corresponds to a fault trigger state, and a status value of 0 corresponds to a normal operation state. Power distribution terminals with a status value of 1 are selected, and the electrical quantity characteristics of the terminal at the time of the fault are bound one-to-one with the fault indication information such as fault type, fault phase, fault duration, and terminal number. All the terminal fault information that has been bound is summarized into a unified set of data, and finally the fault event set of the power distribution terminal is obtained.

[0025] The beneficial effects include achieving time-series unification and data integrity screening of multi-terminal fault recording data, accurately extracting the electrical quantity change characteristics at the fault moment and associating and binding them with fault indication information to form a standardized fault event set, providing unified and standardized data support for subsequent fault topology construction and fault section determination, and ensuring data consistency and reliability in the subsequent fault analysis process.

[0026] B. Based on the fault event set, construct a fault topology map reflecting the electrical connection relationship between the power distribution terminals, and mark suspected fault sections in the fault topology map according to the fault indication information; In this embodiment of the invention, the step of constructing a fault topology map reflecting the electrical connection relationship between the power distribution terminals based on the fault event set, and marking suspected fault sections in the fault topology map based on the fault indication information, includes: Based on the static topology connection relationship of the distribution network, the distribution terminal identifiers in the fault event set are mapped as graph nodes, and the graph nodes are connected to obtain the initial topology graph of the distribution network. Based on the fault indication information of the centralized distribution terminal of the fault event, fault status attributes are assigned to the graph nodes in the initial topology graph to obtain the fault topology graph of the distribution network. In the fault topology map, faulty nodes with the fault state attribute and non-faulty nodes without the fault state attribute are identified, and the connecting edges between the faulty nodes and the non-faulty nodes are marked as suspected fault segments.

[0027] In the fault topology graph, identifying faulty nodes with the fault state attribute and non-faulty nodes without the fault state attribute, and marking the connecting edges between the faulty nodes and the non-faulty nodes as suspected fault segments, includes: A connectivity analysis is performed on the fault topology graph to obtain a fault node subgraph of the fault topology graph; Traverse the boundary nodes in the fault node subgraph and find the non-fault node that is directly connected to the boundary node and does not have the fault state attribute. The connecting edge between the boundary node and the non-faulty node is identified as a suspected faulty section.

[0028] Retrieve pre-stored CIM-formatted static topology connection relationship data from the distribution network master station, extract all distribution terminal identifiers contained in the fault event set, with each distribution terminal corresponding to a pole-mounted switch or ring main unit switch in the distribution network; convert each distribution terminal identifier into an independent graph node in the topology structure, with graph node attributes including terminal number, corresponding switch equipment number, and associated line number. Based on the distribution line connection relationships between switch equipment recorded in the static topology connection relationship, draw connection edges between two graph nodes with direct electrical connections, each connection edge corresponding to a distribution line segment. After all graph nodes and connection edges are drawn, the initial topology diagram of the distribution network is obtained.

[0029] The fault indication information corresponding to each distribution terminal in the fault event set is extracted one by one. All graph nodes in the initial topology graph are traversed. Graph nodes with corresponding fault indication information are assigned fault status attributes. Graph nodes without associated fault indication information do not have fault status attributes. After all graph nodes have completed the attribute assignment, the fault topology graph of the distribution network is obtained.

[0030] Traverse all fault nodes with fault state attributes in the fault topology graph, select any fault node as the starting node, and use a breadth-first search algorithm to traverse along the connecting edges to both sides to collect all fault nodes that can be directly or indirectly connected to each other and the connecting edges between these nodes. During the traversal, only nodes with fault state attributes are counted, and non-fault nodes are not included. The independent connected topology structure formed by the collection is used as the fault node subgraph of the fault topology graph.

[0031] The association status of all adjacent nodes of each fault node in the fault node subgraph is counted one by one. Fault nodes with adjacent nodes that do not belong to the fault node subgraph are marked as boundary nodes. The search is extended outward along the connection edge of each boundary node to locate non-fault nodes that are directly connected to the boundary nodes and do not have fault status attributes.

[0032] Extract the connection edges between each set of boundary nodes and the corresponding non-faulty nodes, and determine the power distribution line segment corresponding to the connection edge as a suspected fault segment. After all the connection edges that meet the judgment conditions are marked, a complete set of suspected fault segments is formed.

[0033] The beneficial effects are that the topology mapping and status assignment of fault terminals are completed based on the static topology relationship of the distribution network, the boundary nodes of the fault area are accurately located through connectivity analysis, the boundary section between the fault node and the non-fault node is quickly located, the fault investigation scope is effectively narrowed, and accurate candidate objects are provided for subsequent determination of the actual fault section, reducing the computational overhead of the subsequent fault identification process.

[0034] C. Perform collaborative analysis on the fault recording data in the suspected fault section, compare the differences in electrical quantity characteristics at both ends of the suspected fault section, and locate the actual fault section of the distribution network; In this embodiment of the invention, the step of collaboratively analyzing the fault recording data in the suspected fault section and comparing the differences in electrical quantity characteristics at both ends of the suspected fault section to locate the actual fault section of the distribution network includes: The first end characteristic and the second end characteristic of the suspected fault section are obtained by extracting the fault waveform data reported from both ends of the suspected fault section to characterize the fault current amplitude. The first end feature and the second end feature are normalized to obtain the first end normalized feature and the second end normalized feature of the suspected fault segment. The electrical quantity characteristic difference degree between the two ends of the suspected fault section is calculated based on the first end normalized characteristic quantity and the second end normalized characteristic quantity. The electrical quantity characteristic difference is compared with a preset difference threshold. When the electrical quantity characteristic difference is greater than the difference threshold, the suspected fault section is determined as the actual fault section of the distribution network.

[0035] The step of calculating the electrical quantity characteristic difference degree between the two ends of the suspected fault section based on the first end normalized characteristic quantity and the second end normalized characteristic quantity includes: The first-end normalized feature quantity and the second-end normalized feature quantity are vectorized respectively to obtain the first-end feature vector and the second-end feature vector of the suspected fault segment. Based on the first end feature vector and the second end feature vector, the electrical quantity characteristic difference degree between the two ends of the suspected fault section is calculated, wherein the calculation formula for the electrical quantity characteristic difference degree is: ; in, This indicates the degree of difference in the electrical quantity characteristics. The dimension of the feature vector. Represents the first end feature vector. One portion, Represents the second end feature vector. One portion, Indicates the component index of the eigenvector.

[0036] For each suspected fault section, the corresponding power distribution terminal equipment at both ends of the section is located, and the complete fault waveform data reported by each of the two power distribution terminals is retrieved. The sampling interval from 0.5 to 2 power frequency cycles after the fault time is selected as the feature extraction window. Four feature values ​​are extracted from the current sampling data within the window: peak fault current, effective fault current, transient fault current amplitude, and steady-state fault current amplitude. All feature values ​​at one end of the section are grouped into one group, and all feature values ​​at the other end are grouped into another group to obtain the first-end feature value and the second-end feature value of the suspected fault section.

[0037] For each type of feature quantity, the maximum value between the two ends of the feature quantity is selected as the normalization reference value of the feature quantity. The corresponding feature values ​​at the first and second ends are divided by the reference value of the feature quantity to obtain the normalization ratio value of the corresponding feature. The above single feature independent normalization process is performed on the four types of feature quantities to eliminate the comparison bias caused by different dimensions and amplitude levels, and the first end normalized feature quantity and the second end normalized feature quantity of the suspected fault section are obtained.

[0038] Following a fixed arrangement order of fault current peak value, fault current effective value, fault current transient component amplitude, and fault current steady-state component amplitude, the normalized values ​​of each item in the first-end normalized feature quantity are arranged sequentially to form a 4-dimensional ordered numerical sequence. The normalized values ​​of each item in the second-end normalized feature quantity are arranged sequentially in the same order to form a 4-dimensional ordered numerical sequence, thus obtaining the first-end feature vector and the second-end feature vector of the suspected fault section. The feature vector dimension n is 4.

[0039] The values ​​at the same positions in the two feature vectors are compared sequentially, and the matching degree of all corresponding position values ​​is calculated. The overall similarity of the two feature vectors is calculated by combining the matching results of all positions. Then, the similarity value is subtracted from the value to obtain the electrical quantity characteristic difference between the two ends of the suspected fault section.

[0040] The first-end feature vector required for calculating the electrical quantity characteristic difference is obtained by vectorizing the first-end normalized feature quantity; the second-end feature vector required for calculating the electrical quantity characteristic difference is obtained by vectorizing the second-end normalized feature quantity; the first-end normalized feature quantity is obtained by normalizing the first-end feature quantity; the second-end normalized feature quantity is obtained by normalizing the second-end feature quantity; the first-end feature quantity is extracted from the fault waveform data reported from the first end of the suspected fault section and is used to characterize the fault current amplitude; the second-end feature quantity is extracted from the fault waveform data reported from the second end of the suspected fault section and is used to characterize the fault current amplitude; the dimension of the feature vector is determined by the total number of feature quantities included in the calculation during the vectorization process; the component index of the feature vector corresponds to the position of each feature quantity in the feature vector.

[0041] This calculation process is used to obtain the electrical quantity characteristic difference between the two ends of a suspected fault section, serving as a quantitative basis for determining the actual fault section. The calculation involves first multiplying the corresponding feature values ​​in the first-end feature vector and the second-end feature vector one by one; then summing all the multiplication results to obtain the sum of the numerator; squaring the value of each feature value in the first-end feature vector and summing them all to obtain the first-end sum of squares; squaring the value of each feature value in the second-end feature vector and summing them all to obtain the second-end sum of squares; taking the square root of the first-end sum of squares to obtain the first-end modulus value; taking the square root of the second-end sum of squares to obtain the second-end modulus value; multiplying the first-end modulus value and the second-end modulus value to obtain the product of the denominator; dividing the sum of the numerator by the product of the denominator to obtain the cosine similarity value between the two feature vectors; and finally subtracting the cosine similarity value from the first value to output the final result of the electrical quantity characteristic difference calculation.

[0042] The calculation result can be directly compared with the preset difference threshold to determine whether the corresponding suspected fault section is a real fault section.

[0043] When the electrical quantity characteristics at both ends of the suspected fault section are completely identical, the cosine similarity value of the two feature vectors is one, and the calculated electrical quantity characteristic difference value is zero; when the difference in electrical quantity characteristics at both ends of the suspected fault section gradually increases, the cosine similarity value of the two feature vectors gradually decreases, and the calculated electrical quantity characteristic difference value gradually increases; when the electrical quantity characteristics at both ends of the suspected fault section are completely opposite, the cosine similarity value of the two feature vectors is negative one, and the calculated electrical quantity characteristic difference value is two.

[0044] The value of the electrical quantity characteristic difference decreases monotonically as the similarity of the electrical quantity characteristics at both ends increases; the value of the electrical quantity characteristic difference increases monotonically as the difference of the electrical quantity characteristics at both ends increases.

[0045] A pre-set difference threshold is retrieved. This threshold is calibrated based on sample data of typical fault scenarios in the distribution network, with a typical value of 0.3. It can be adjusted within the range of 0.2 to 0.4 according to the voltage level and line length of the distribution network. The calculated electrical quantity characteristic difference is compared with this threshold. When the value of the electrical quantity characteristic difference is greater than the difference threshold, it is determined that the fault current flows in from only one side of the section. This section is the section where the fault point is located, and the corresponding suspected fault section is marked as the fault section. After traversing all suspected fault sections and completing all comparisons, the actual fault section of the distribution network is obtained.

[0046] The beneficial effects are that by extracting the fault current amplitude features and completing normalization and vectorization processing, and using the degree of difference between the features at both ends as the basis for fault judgment, it can effectively avoid misjudgment and missed judgment caused by load fluctuations and instantaneous interference, improve the accuracy of identifying real fault sections, and provide a reliable positioning basis for the subsequent generation of isolation commands.

[0047] D. Based on the location of the actual fault section in the fault topology diagram, generate isolation operation instructions for the distribution network; In this embodiment of the invention, generating the isolation operation command for the distribution network based on the location of the actual fault section in the fault topology map includes: In the fault topology diagram, the topology of the distribution network is divided into a fault-side sub-diagram and a non-fault-side sub-diagram, with the actual fault section as the boundary. Traverse the fault side subgraph, extract the switch equipment identifiers that are directly connected to the actual fault section in the fault side subgraph, and obtain the first isolation device set of the distribution network; Traverse the non-faulty side subgraph, extract the switch equipment identifiers that are directly connected to the actual faulty section in the non-faulty side subgraph, and obtain the second isolation device set of the distribution network; The first isolation device set and the second isolation device set are merged and deduplicated to obtain the complete isolation device set of the distribution network, and the isolation operation command of the distribution network is generated based on the complete isolation device set.

[0048] The step of generating isolation operation instructions for the distribution network based on the complete isolation device set includes: Obtain the topology hierarchy information of the complete isolation equipment centralized switching equipment in the fault topology diagram; Arrange the switching devices in ascending order of the topology hierarchy information to obtain the isolation operation sequence of the distribution network; A unique operation sequence number is assigned to the switching equipment in the isolation operation sequence, and the operation sequence number, the switching equipment identifier, and the tripping action identifier are associated and encapsulated to obtain the isolation operation instruction of the distribution network.

[0049] Locate the connecting edge corresponding to the actual faulty segment in the fault topology graph, disconnect the overall topology from the position of the connecting edge, and perform a full connectivity traversal to both sides of the connecting edge. Define the connected topology structure on one side that contains all nodes with fault state attributes as the faulty side subgraph, and define the connected topology structure on the other side that does not contain nodes with fault state attributes as the non-faulty side subgraph.

[0050] Traverse all graph nodes in the fault-side subgraph, filter out graph nodes that are directly adjacent to the connecting edges corresponding to the actual fault section, extract the switch equipment identifiers corresponding to these graph nodes, and collect all the filtered switch equipment identifiers into independent set data to obtain the first isolation device set of the distribution network.

[0051] Traverse all graph nodes in the non-faulty side subgraph, filter out the graph nodes that are directly adjacent to the connecting edge corresponding to the actual faulty section, extract the switch equipment identifiers corresponding to these graph nodes, and collect all the filtered switch equipment identifiers into independent set data to obtain the second isolation device set of the distribution network.

[0052] All switchgear identifiers contained in the first and second isolation device sets are aggregated into the same set. Duplicate switchgear identifiers are removed to form a set of switchgear identifiers without duplicates, thus obtaining the complete isolation device set of the distribution network.

[0053] Starting from the outgoing switch of the distribution network substation as the first level, the level number increases by 1 for each switch equipment level along the load side of the line. The corresponding topology level values ​​are pre-stored for each graph node in the fault topology diagram. The graph node corresponding to each switch equipment identifier in the complete isolation equipment set is matched one by one, and the topology level value of the corresponding graph node is read to obtain the topology level information of all switch equipment in the complete isolation equipment set.

[0054] Following the rule of operating the switching equipment closest to the power source first and then the switching equipment closest to the load side, all the switching equipment in the complete isolation equipment set are sorted in sequence to form a queue of switching equipment arranged in a fixed execution order, thus obtaining the isolation operation sequence of the distribution network.

[0055] According to the order of the isolation operation sequence, each switchgear in the sequence is assigned a unique and consecutive decimal operation sequence number. The tripping action identifier is represented by the fixed string "OPEN". The operation sequence number, the unique identifier of the switchgear, the tripping action identifier, and the instruction generation timestamp of each switchgear are bound and combined, and encapsulated into a standardized instruction data unit according to the IEC60870-5-104 communication protocol format. All instruction data units are integrated in sequence to obtain the isolation operation instruction of the distribution network.

[0056] The beneficial effects are that by selecting directly connected switching equipment from both sides of the fault section to form a complete isolation range, and by generating an ordered isolation operation sequence based on topology information, timing conflicts in switching operations can be avoided, ensuring that isolation operations are executed in a reasonable order. At the same time, standardized encapsulation forms standardized isolation operation instructions, providing a unified and reliable data basis for subsequent instruction issuance and execution.

[0057] E. Send the isolation operation command to the power distribution terminal, and provide anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal.

[0058] In this embodiment of the invention, sending the isolation operation command to the power distribution terminal and providing anomaly feedback on the isolation operation command to obtain the command execution closed-loop verification result of the power distribution terminal includes: The isolation operation command is subjected to integrity verification and command sequence consistency verification, and a verification pass identifier for the isolation operation command is generated; Based on the verification pass identifier, the isolation operation instruction is encapsulated into an instruction data packet with a timing synchronization tag and mapped to the instruction receiving queue of the power distribution terminal. According to the timing synchronization tag, the instruction data packets in the instruction receiving queue are prioritized and the execution window is allocated to obtain the instruction execution scheduling sequence of the power distribution terminal; The instruction execution scheduling sequence is sent to the power distribution terminal, and the instruction reception confirmation identifier and execution status feedback data returned by the power distribution terminal are received. Based on the execution status feedback data, the instruction execution scheduling sequence is marked with abnormal status to obtain the instruction execution closed-loop verification result of the power distribution terminal.

[0059] Verify the operation sequence number, switch equipment identifier, and tripping action identifier contained in the isolation operation instruction field by field, confirm that all required fields are complete, and complete the instruction integrity verification. Then, verify the arrangement and continuity of the operation sequence numbers in the instruction, confirm that the sequence order conforms to the rule of the topology hierarchy from low to high and that the sequence numbers are not repeated or have gaps, and complete the instruction sequence consistency verification. After both verifications pass, generate the verification pass identifier for the isolation operation instruction.

[0060] After the verification pass flag is identified, a time synchronization tag with a unified reference is attached to each instruction unit in the isolation operation instruction to mark the planned execution time reference of the instruction. The instruction content with the synchronization tag is encapsulated into a standard format instruction data packet, and the instruction data packet is delivered to the instruction receiving queue of the corresponding power distribution terminal according to the power distribution terminal address of each switchgear.

[0061] Extract the timing synchronization tag attached to each instruction data packet in the instruction receiving queue, prioritize all instruction data packets according to the time sequence corresponding to the timing synchronization tag, allocate an independent execution time window to each sorted instruction data packet, limit the start and end time of the corresponding instruction execution, and organize them into an ordered instruction execution queue to obtain the instruction execution scheduling sequence of the power distribution terminal.

[0062] The instruction execution scheduling sequence is sent to the corresponding power distribution terminal through the optical fiber communication link. After the power distribution terminal successfully receives the scheduling sequence, it returns an instruction reception confirmation flag within 10 milliseconds. During the process of the power distribution terminal performing the tripping operation according to the scheduling sequence, it sends back execution status feedback data at two moments: action start and action completion. The feedback data includes four items: switch equipment identifier, action type, action time, and switch position signal.

[0063] The received execution status feedback data is compared and verified one by one with the expected execution status in the instruction execution scheduling sequence. Abnormal statuses are divided into two categories: timeout and execution failure. If the action completion feedback is not returned within the corresponding execution window, it is marked as timeout exception; if the action failure signal is returned, it is marked as execution failure exception. After all instruction items are verified, a complete instruction execution verification record is formed, which includes the total number of instructions, the number of successful executions, the number of exceptions, and the exception details. The closed-loop verification result of instruction execution of the power distribution terminal is obtained.

[0064] The beneficial effects are that the accuracy and compliance of the instruction content are ensured by double verification of the integrity and sequence consistency before the instruction is issued, the orderly scheduling of instructions and the control of the execution window are realized by combining the timing synchronization tag, and the supporting reception confirmation and execution status feedback mechanism completes the whole process tracking. It can promptly identify instruction execution anomalies, form a complete instruction execution closed-loop verification link, and ensure the reliable implementation of fault isolation operations.

[0065] like Figure 2 The diagram shown is a functional block diagram of a power distribution network fault isolation system based on power distribution terminal fault information provided in an embodiment of the present invention.

[0066] The power distribution network fault isolation system based on power distribution terminal fault information described in this invention can be installed in electronic devices. Depending on the functions implemented, the power distribution network fault isolation system based on power distribution terminal fault information may include a fault analysis module, a suspected fault marking module, a real fault location module, an isolation command generation module, and a closed-loop feedback module. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and is stored in the memory of the electronic device.

[0067] In this embodiment, the functions of each module / unit are as follows: The fault analysis module is used to analyze the fault recording data of the distribution terminal in the distribution network to obtain a set of fault events containing electrical quantity characteristics and fault indication information at the time of the fault. The suspected fault marking module is used to construct a fault topology map reflecting the electrical connection relationship between the power distribution terminals based on the fault event set, and to mark the suspected fault section in the fault topology map based on the fault indication information. The real fault location module is used to perform collaborative analysis on the fault recording data in the suspected fault section, compare the differences in electrical quantity characteristics at both ends of the suspected fault section, and locate the real fault section of the distribution network. The isolation command generation module is used to generate isolation operation commands for the distribution network based on the location of the actual fault section in the fault topology diagram. The closed-loop feedback module is used to send the isolation operation command to the power distribution terminal, and to provide anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal.

[0068] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0069] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0070] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0071] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0072] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for fault isolation in a distribution network based on fault information from distribution terminals, characterized in that, The method includes: A. Analyze the fault recording data of the distribution terminal in the distribution network to obtain a set of fault events containing electrical quantity characteristics and fault indication information at the time of the fault; Step A specifically includes: obtaining the status value of the fault indication flag bit stored inside each power distribution terminal, filtering out power distribution terminals with a status value of 1, binding the electrical quantity characteristics of the fault time corresponding to the terminal with the four fault indication information items of the terminal itself: fault type, fault phase, fault duration, and terminal number, summarizing all the terminal fault information that has been bound into a unified set of data, and finally obtaining the fault event set of the power distribution terminal. B. Based on the set of fault events, construct a fault topology map reflecting the electrical connection relationship between the power distribution terminals, and mark suspected fault sections in the fault topology map according to the fault indication information; Step B specifically includes: Based on the static topology connection relationship of the distribution network, the distribution terminal identifiers in the fault event set are mapped as graph nodes, and the graph nodes are connected to obtain the initial topology graph of the distribution network. Based on the fault indication information of the centralized distribution terminal of the fault event, fault status attributes are assigned to the graph nodes in the initial topology graph to obtain the fault topology graph of the distribution network. Traverse all fault nodes with fault state attributes in the fault topology graph, collect all fault nodes that can be directly or indirectly connected to each other and the connecting edges between these nodes, and use the collected independent connected topology structure as the fault node subgraph of the fault topology graph. Mark the fault nodes that have adjacent nodes that do not belong to the fault node subgraph as boundary nodes. Extend the search outward along the connecting edges of each boundary node to locate the non-fault nodes that are directly connected to the boundary nodes and do not have fault state attributes. The connecting edges between the boundary nodes and the non-faulty nodes are identified as suspected faulty sections. C. Perform collaborative analysis on the fault recording data in the suspected fault section, compare the differences in electrical quantity characteristics at both ends of the suspected fault section, and locate the actual fault section of the distribution network; D. Based on the location of the actual fault section in the fault topology diagram, generate isolation operation instructions for the distribution network; E. Send the isolation operation command to the power distribution terminal, and provide anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal; Step E specifically includes: The isolation operation command is subjected to integrity verification and command sequence consistency verification, and a verification pass identifier for the isolation operation command is generated; Based on the verification pass identifier, the isolation operation instruction is encapsulated into an instruction data packet with a timing synchronization tag and mapped to the instruction receiving queue of the power distribution terminal. According to the timing synchronization tag, the instruction data packets in the instruction receiving queue are prioritized and the execution window is allocated to obtain the instruction execution scheduling sequence of the power distribution terminal; The instruction execution scheduling sequence is sent to the power distribution terminal, and the instruction reception confirmation identifier and execution status feedback data returned by the power distribution terminal are received. Based on the execution status feedback data, the instruction execution scheduling sequence is marked with abnormal status to obtain the instruction execution closed-loop verification result of the power distribution terminal.

2. The power distribution network fault isolation method based on power distribution terminal fault information as described in claim 1, characterized in that, The step of parsing the fault recording data of the distribution terminal in the distribution network to obtain a fault event set containing electrical quantity characteristics and fault indication information at the time of the fault also includes: The fault recording data reported by the distribution terminal in the distribution network is obtained and timestamped to obtain the synchronous recording data of the distribution terminal. Data integrity verification is performed on the recording channels in the synchronous recording data to obtain the electrical quantity sampling sequence of the power distribution terminal; Traverse the electrical quantity sampling sequence, detect abrupt changes in the sampled values ​​in the electrical quantity sampling sequence, and record the time corresponding to the abrupt change position and the amplitude direction of the sampled values ​​before and after the abrupt change position to obtain the electrical quantity characteristics of the power distribution terminal at the time of the fault.

3. The power distribution network fault isolation method based on power distribution terminal fault information as described in claim 1, characterized in that, The step of collaboratively analyzing fault recording data in the suspected fault section and comparing the differences in electrical quantity characteristics at both ends of the suspected fault section to locate the actual fault section of the distribution network includes: The first end characteristic and the second end characteristic of the suspected fault section are obtained by extracting the fault waveform data reported from both ends of the suspected fault section to characterize the fault current amplitude. The first end feature and the second end feature are normalized to obtain the first end normalized feature and the second end normalized feature of the suspected fault segment. The electrical quantity characteristic difference degree between the two ends of the suspected fault section is calculated based on the first end normalized characteristic quantity and the second end normalized characteristic quantity. The electrical quantity characteristic difference is compared with a preset difference threshold. When the electrical quantity characteristic difference is greater than the difference threshold, the suspected fault section is determined as the actual fault section of the distribution network.

4. The power distribution network fault isolation method based on power distribution terminal fault information as described in claim 3, characterized in that, The step of calculating the electrical quantity characteristic difference degree between the two ends of the suspected fault section based on the first end normalized characteristic quantity and the second end normalized characteristic quantity includes: The first-end normalized feature quantity and the second-end normalized feature quantity are vectorized respectively to obtain the first-end feature vector and the second-end feature vector of the suspected fault segment. Based on the first end feature vector and the second end feature vector, the electrical quantity characteristic difference degree between the two ends of the suspected fault section is calculated, wherein the calculation formula for the electrical quantity characteristic difference degree is: ； in, This indicates the degree of difference in the electrical quantity characteristics. The dimension of the feature vector. Represents the first end feature vector. One portion, Represents the second end feature vector. One portion, Indicates the component index of the eigenvector.

5. The power distribution network fault isolation method based on power distribution terminal fault information as described in claim 1, characterized in that, The step of generating isolation operation instructions for the distribution network based on the location of the actual fault segment in the fault topology map includes: In the fault topology diagram, the topology of the distribution network is divided into a fault-side sub-diagram and a non-fault-side sub-diagram, with the actual fault section as the boundary. Traverse the fault side subgraph, extract the switch equipment identifiers that are directly connected to the actual fault section in the fault side subgraph, and obtain the first isolation device set of the distribution network; Traverse the non-faulty side subgraph, extract the switch equipment identifiers that are directly connected to the actual faulty section in the non-faulty side subgraph, and obtain the second isolation device set of the distribution network; The first isolation device set and the second isolation device set are merged and deduplicated to obtain the complete isolation device set of the distribution network, and the isolation operation command of the distribution network is generated based on the complete isolation device set.

6. The power distribution network fault isolation method based on power distribution terminal fault information as described in claim 5, characterized in that, The step of generating isolation operation instructions for the distribution network based on the complete isolation device set includes: Obtain the topology hierarchy information of the complete isolation equipment centralized switching equipment in the fault topology diagram; Arrange the switching devices in ascending order of the topology hierarchy information to obtain the isolation operation sequence of the distribution network; A unique operation sequence number is assigned to the switching equipment in the isolation operation sequence, and the operation sequence number, the switching equipment identifier, and the tripping action identifier are associated and encapsulated to obtain the isolation operation instruction of the distribution network.

7. A power distribution network fault isolation system based on power distribution terminal fault information, characterized in that, The system for implementing the power distribution network fault isolation method based on power distribution terminal fault information as described in claim 1 includes: The fault analysis module is used to analyze the fault recording data of the distribution terminal in the distribution network to obtain a set of fault events containing electrical quantity characteristics and fault indication information at the time of the fault. The suspected fault marking module is used to construct a fault topology map reflecting the electrical connection relationship between the power distribution terminals based on the fault event set, and to mark the suspected fault sections in the fault topology map based on the fault indication information. The real fault location module is used to perform collaborative analysis on the fault recording data in the suspected fault section, compare the differences in electrical quantity characteristics at both ends of the suspected fault section, and locate the real fault section of the distribution network. An isolation command generation module is used to generate isolation operation commands for the distribution network based on the location of the actual fault section in the fault topology diagram. The closed-loop feedback module is used to send the isolation operation command to the power distribution terminal, and to provide anomaly feedback on the isolation operation command to obtain the closed-loop verification result of the command execution of the power distribution terminal.

Images