Power distribution automation operation and maintenance method based on electric power big data

By combining a fault feature rule set and topology reachability model based on power big data with an operation and maintenance scheduling model, rapid fault location and intelligent operation and maintenance decision-making for complex power distribution networks are achieved. This solves the problems of inaccurate diagnosis and low response efficiency under traditional operation and maintenance methods, and improves the fault handling capability of the power distribution system.

CN121172983APending Publication Date: 2025-12-19TAIAN POWER SUPPLY CO OF STATE GRID SHANDONG ELECTRIC POWER CO
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202511323598.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-16
Publication Date
2025-12-19

AI Technical Summary

Technical Problem

Traditional power distribution automation operation and maintenance methods are inaccurate in fault diagnosis and have low dispatch response efficiency in complex network topologies. In particular, operation and maintenance are lagging in areas with distributed power sources, which poses a risk of power outages for users.

Method used

By collecting data in real time through the feeder terminal, and combining it with the fault feature rule set and topology reachability model, a fault diagnosis result dataset is constructed. Then, the operation and maintenance scheduling model is used to generate fault isolation strategies and maintenance task priority lists to achieve intelligent operation and maintenance decision-making.

Benefits of technology

It improves the accuracy and timeliness of fault diagnosis, reduces dispatch response time, and enhances operation and maintenance collaboration efficiency and power supply reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121172983A_ABST
    Figure CN121172983A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of power distribution automation operation and maintenance, in particular to a power distribution automation operation and maintenance method based on electric power big data. The method comprises the following steps: collecting and processing power distribution operation data, calling a fault feature rule set to analyze the power distribution operation data, and identifying and detecting a power distribution fault event; based on the electrical topological structure, the protection action information and the power distribution node on-off state of the current power distribution network, a fault feeder line and a fault section are identified, a switching value logic relation and a topological accessibility model are used for power distribution positioning of a fault event area, and a fault diagnosis result data set is output; and generating a fault isolation operation and maintenance strategy and an overhaul task priority list by using an operation and maintenance scheduling model based on the fault diagnosis result data set and the load transfer capacity, the field operation capacity and the historical overhaul resource distribution information of the power distribution station. According to the method, fault diagnosis and an intelligent scheduling mechanism are fused, and the fault response speed and the operation and maintenance cooperation efficiency of the power distribution system under complex working conditions are improved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution automation operation and maintenance, and particularly relates to a power distribution automation operation and maintenance method based on power big data. BACKGROUND

[0002] Under the background of continuous development of power distribution automation technology, the power distribution system is gradually evolving towards digitization, visualization and intelligentization; the power distribution automation operation and maintenance technology usually relies on feeder terminals, fault indicators, power distribution master station systems and operation and maintenance dispatching platforms to realize real-time monitoring, fault identification and remote control of the operation state of the power distribution network; with the increasing complexity of the power grid load structure, the proportion of distributed power supply access is increasing year by year, the power distribution network topology changes frequently, and the traditional operation and maintenance mode relying on manual judgment and fixed rules faces the bottleneck problems of low response efficiency, insufficient diagnostic accuracy and poor dispatching coordination, so it is urgent to introduce intelligent analysis and decision mechanism based on power big data to improve the fault handling capability and operation and maintenance level of the power distribution system;

[0003] In the actual operation process of the power distribution system, the following two key problems still need to be solved: firstly, when a fault occurs in the power distribution network, due to the complex network topology structure, numerous branches and limited protection information, the propagation path of the fault signal in the network is difficult to accurately model, resulting in problems such as inaccurate section identification and incorrect type judgment during fault diagnosis; secondly, even if the fault location is completed, the traditional dispatching system relies on manual experience to develop isolation and recovery strategies, lacks systematic dispatching optimization model support, and is prone to problems such as fault isolation response delay, unreasonable load transfer path selection and low efficiency of maintenance resource allocation, especially in high-proportion distributed power supply access or load-intensive areas, which may further amplify the operation and maintenance response lag and user power outage risk. SUMMARY

[0004] The purpose of the present application is to provide a power distribution automation operation and maintenance method based on power big data to solve the two key problems raised in the background.

[0005] To achieve the above purpose, the technical solution of the present application is: a power distribution automation operation and maintenance method based on power big data, comprising:

[0006] S1, real-time acquisition and processing of power distribution operation data through feeder terminals, intelligent switches and monitoring equipment, and calling of a fault feature rule set to analyze the power distribution operation data and identify and detect power distribution fault events;

[0007] S2, based on the electrical topology structure, protection action information and power distribution node opening state of the current power distribution network, identifying the fault feeder and fault section, and using the switch quantity logical relationship and topology reachability model to locate the fault event area and output the fault diagnosis result data set;

[0008] S3, generating a fault isolation operation and maintenance strategy and a maintenance task priority list using an operation and maintenance scheduling model based on the fault diagnosis result data set and the load transfer capability, on-site operation capability and historical maintenance resource distribution information of the power distribution station;

[0009] S4, executing the fault isolation operation and maintenance strategy according to the maintenance task priority list, collecting power distribution state quantity changes and strategy response feedback information in real time, and archiving them to an operation and maintenance knowledge base.

[0010] Preferably, the power distribution operation data includes line voltage, line current, switch state, power quality parameter, power factor, frequency, device operating temperature and historical alarm state information.

[0011] Preferably, the fault feature rule set is called in parallel by the power distribution automation master station system based on the operation data, including condition judgment rules of current mutation, switch jump, protection action, voltage drop, zero sequence current offset and abnormal state of power distribution node, which are as follows:

[0012] A current mutation judgment rule based on the fact that the amplitude of line current mutation in a continuous sampling period exceeds a set threshold; a switch jump judgment rule based on the fact that the switch state changes from closed to open or from open to closed; a protection action judgment rule based on the protection device start signal; a voltage drop judgment rule based on the fact that the voltage drop amplitude exceeds a set proportion; a zero sequence current offset judgment rule based on the fact that the zero sequence current offset from the reference threshold value; and a power distribution node state abnormality judgment rule based on the fact that the connection state of the power distribution node in the topology model is abnormal.

[0013] Preferably, the topology reachability model is constructed by the power distribution automation master station system, based on the electrical connection relationship between each power distribution node in the power distribution network, the on-off state of the feeder switch and the protection device configuration information, to generate a comprehensive topology graph structure including node connection graph, protection action graph and fault flow path graph, and based on the graph search algorithm to judge the reachability of the fault propagation path and output a candidate path set of the fault influence area.

[0014] Preferably, the switch quantity logical relationship is established by a logical judgment module configured in the power distribution automation master station system, which matches the on-off state of the upper switch, branch switch and terminal switch of each power distribution node on the same feeder through combination relationship, and verifies the path consistency of the opening and closing change sequence before and after the fault through the construction of state link graph and time sequence reasoning chain, which is used to assist in judging the fault section.

[0015] Preferably, the fault diagnosis result dataset comprises a structured diagnosis field; the structured diagnosis field comprises at least a fault feeder number, a fault section start node number, a fault section end node number, a fault judgment timestamp, a fault type label and a diagnosis confidence level, and is used to provide a fault isolation strategy generation and power supply path reconstruction calculation for an operation and maintenance scheduling system.

[0016] Preferably, a scheduling optimization model used in the operation and maintenance scheduling system is constructed based on the fault diagnosis result dataset.

[0017] The optimization objective function takes minimizing fault isolation time, minimizing user power supply interruption capacity and minimizing maintenance resource path length as optimization objectives, and the constraint conditions comprise a topological connectivity constraint, a distribution station switching capacity constraint and a field operation ability boundary condition.

[0018] The scheduling optimization model adopts a mixed integer linear programming method to solve the strategy.

[0019] Preferably, the load transfer capacity is dynamically evaluated by a distribution automation master station system in combination with a current operation mode, a standby power output capacity, a transformer load rate and a feeder switch operable state, and is used as a path feasibility screening factor in the scheduling optimization model to determine whether a feasible path set for implementing power supply recovery operation is available.

[0020] Preferably, the maintenance task priority list is comprehensively scored and sorted by the operation and maintenance scheduling system according to a fault type label, a fault section covered load amount, a distribution node power supply level and a real-time position of a maintenance resource in the fault diagnosis result dataset.

[0021] The comprehensive scoring and sorting process uses a priority score function to construct a multi-factor weighted model, and outputs a sorted maintenance operation plan list and a scheduling work order data, and finally obtains the maintenance task priority list.

[0022] Preferably, the response feedback information comprises an execution state of a remote control device, a maintenance progress of a field operation terminal, distribution node state quantity change information and an operation timestamp record; and the response feedback information is archived to an operation and maintenance knowledge base by the operation and maintenance scheduling system after the fault isolation and power supply recovery operation is completed.

[0023] The operation and maintenance knowledge base is used to update a parameter weight of the optimization scheduling optimization model.

[0024] Compared with the prior art, the above technical solutions of the present application have the following beneficial technical effects:

[0025] 1、In the application, based on the parallel calling of the fault feature rule set by the power distribution automation master station system and the fusion calculation of the topological accessibility model, the rapid and accurate positioning of the fault feeder and the fault section under the complex power distribution topological structure can be realized, and in the power distribution network scene with high proportion of distributed power access and complex electrical connection relationship, the timeliness and accuracy of fault diagnosis are improved.

[0026] 2、In the application, by constructing a scheduling optimization model, the intelligent generation of the fault isolation path, the power supply recovery path and the maintenance priority by the operation and maintenance scheduling system is realized, the scheduling response time and the operation resource conflict are reduced, and the overall operation and maintenance coordination efficiency and the power supply reliability in the complex fault handling process are improved. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 A flowchart of an embodiment of the application is provided. DETAILED DESCRIPTION

[0028] As shown in Figure 1 The power distribution automation operation and maintenance method based on power big data provided by the application has the following specific implementation steps:

[0029] S1, real-time acquisition and processing of power distribution operation data by feeder terminal, intelligent switch and monitoring equipment, calling of fault feature rule set for analyzing power distribution operation data, and identification of detected power distribution fault events;

[0030] S2, based on the electrical topological structure, protection action information and power distribution node breaking state of the current power distribution network, identification of the fault feeder and the fault section, and use of the switch quantity logical relationship and the topological accessibility model for power distribution positioning of the fault event area, and output of the fault diagnosis result data set;

[0031] S3, based on the fault diagnosis result data set and the load transfer capacity, on-site operation capacity and historical maintenance resource distribution information of the power distribution station, use of the operation and maintenance scheduling model to generate the fault isolation operation and maintenance strategy and the maintenance task priority list;

[0032] S4, execution of the fault isolation operation and maintenance strategy according to the maintenance task priority list, real-time acquisition of power distribution state quantity changes and strategy response feedback information, and archiving to the operation and maintenance knowledge base.

[0033] In the S1 of the embodiment, the power distribution operation data includes line voltage, line current, switch state, power quality parameter, power factor, frequency, device operating temperature and historical alarm state information.

[0034] The fault feature rule set is called in parallel by the power distribution automation master station system based on the operation data, including condition judgment rules of current mutation, switch jump, protection action, voltage drop, zero sequence current offset and power distribution node state anomaly, and specifically as follows:

[0035] A current jump determination rule based on a magnitude of a line current exceeding a set threshold in a continuous sampling period; a switch jump determination rule based on a switch state changing from a closed position to an open position and from an open position to a closed position; a protection action determination rule based on a protection device starting signal; a voltage drop determination rule based on a line voltage drop exceeding a set proportion; a zero sequence current deviation determination rule based on a zero sequence current deviating from a reference threshold; and a distribution node state abnormality determination rule based on a connection state of a distribution node being abnormal in a topology model.

[0036] In the embodiment S2, the topology reachability model is constructed by a distribution automation master station system, based on electrical connection relationships between distribution nodes in a distribution network, on-off states of feeder switches, and on protection device configuration information, to generate a comprehensive topology graph structure including a node connection graph, a protection action graph, and a fault flow path graph, and to determine reachability of a fault propagation path based on a graph search algorithm, and to output a candidate path set of a fault influence area.

[0037] In the embodiment, the topology reachability model is constructed based on graph theory principles, and an adjacency matrix is generated using a real-time model topology library of a distribution network in actual deployment, and node connectivity is dynamically adjusted in combination with real-time switch states; in a multi-source power supply, ring network structure, or distributed power supply reverse power supply situation, there can be multiple solutions for a fault propagation path, therefore, the system introduces a protection action graph and a fault flow direction restriction to filter out candidate paths that are not electrically logical, and finally outputs a high-confidence reachable path set as an input for subsequent section positioning.

[0038] The switch quantity logical relationship is established by a logical determination module configured in the distribution automation master station system, the logical determination module matches combination relationships of on-off states of upper switches, branch switches, and end switches of distribution nodes on a same feeder, and performs path consistency verification on on-off change sequences before and after a fault occurs through construction of a state link graph and a time sequence reasoning chain, to assist in determining a fault section.

[0039] In the embodiment, the modeling process of the switch quantity logical relationship includes not only upstream and downstream relationships of device physical positions, but also operation sequences and response time sequence differences, and state transition rules before and after a fault occurs are determined through construction of a state link graph and introduction of a time sequence reasoning chain; if a downstream branch trips before an upper switch, there is a time logic conflict in the path, and the path is excluded, to ensure logical consistency and verifiability of fault section positioning.

[0040] The fault diagnosis result data set comprises a structured diagnosis field; the structured diagnosis field at least comprises a fault feeder number, a fault section start node number, a fault section end node number, a fault judgment timestamp, a fault type label, and a diagnosis confidence level, and is used to provide a fault isolation strategy generation and power supply path reconstruction calculation for an operation and maintenance scheduling system.

[0041] In the embodiment, the diagnosis confidence level in the fault diagnosis result data set is a hierarchical label automatically calculated based on the consistency of system triggering under multiple rule conditions and the uniqueness of the positioning path, and is divided into three levels of high, medium, and low; the hierarchical label is used to guide whether the operation and maintenance scheduling system needs manual review of the result or directly inputs the fault isolation operation; for example, when the path is uniquely identified and the rule consistency is high, the system can automatically issue a scheduling suggestion, and when the confidence level is low, the priority is adjusted to a low level and enters a manual verification process.

[0042] In the embodiment S3, a scheduling optimization model used in the operation and maintenance scheduling system constructs an optimization objective function based on the fault diagnosis result data set.

[0043] The optimization objective function takes minimizing fault isolation time, minimizing user power supply interruption capacity, and minimizing maintenance resource path length as optimization objectives, and the constraint conditions include a topological connectivity constraint, a distribution station switching capacity constraint, and a field operation ability boundary condition.

[0044] The scheduling optimization model uses a mixed integer linear programming method to solve the strategy.

[0045] In the embodiment, the scheduling optimization model uses a modular objective function definition mechanism to allow flexible adjustment of the objective function weight according to different operation and maintenance scenarios. The mixed integer linear programming method combines graph search structure, defines node operation state and path combination mode through variables, realizes linkage optimization of device-level operation decision and network-level power supply reconstruction, and under a large-scale network, the model can improve scheduling calculation efficiency by heuristic pre-solution constraint boundary, and ensure that the isolation strategy is generated within a response time of seconds.

[0046] The load transfer capacity is dynamically evaluated by the distribution automation master station system in combination with the current operation mode, standby power output capacity, transformer load rate, and feeder switch operable state, and is used as a path feasibility screening factor in the scheduling optimization model to determine whether there is a feasible path set for implementing power supply recovery operation.

[0047] In the embodiment, the dynamic assessment of load transfer capability is based on the linkage operation results of the power distribution automation master station and the real-time power flow analysis system; the output state of the standby power source, the current load rate of the main transformer and the allowed transient overload capacity are collected in real time, and the standby interconnection operability of the adjacent feeder is judged, and a power supply capability matrix is constructed to indicate the power supply capability level of each possible recovery node, which is used as a path feasibility constraint condition for the dispatching model to screen out electrically unreachable paths.

[0048] The maintenance task priority list is scored and sorted by the operation and maintenance scheduling system according to the fault type label, fault section covered load, power distribution node power supply level and real-time location of maintenance resources contained in the fault diagnosis result data set;

[0049] The process of the comprehensive scoring and sorting uses a priority scoring function to construct a multi-factor weighted model, and outputs the sorted maintenance work plan list and dispatch work order data, and finally obtains the maintenance task priority list.

[0050] In the embodiment, the maintenance task priority scoring function comprehensively considers the task urgency and resource accessibility factors, and can embed an artificial weight template in the scheduling model for fine tuning; the maintenance task priority scoring function is expressed in the form of a multi-factor weighted model, including: a load importance coefficient corresponding to the power supply level of the power distribution node, a time cost function converted from the physical distance between the task and the resource, and a matching degree between the task quantity and the maintenance team capacity.

[0051] In the embodiment S4, the response feedback information includes the execution state of the remote control device, the maintenance progress of the field operation terminal, the power distribution node state quantity change information and the operation timestamp record; the response feedback information is archived to the operation and maintenance knowledge base by the operation and maintenance scheduling system after the completion of the fault isolation and power supply recovery operation;

[0052] The operation and maintenance knowledge base is used to update the parameter weight of the optimized scheduling optimization model.

[0053] In the embodiment, the operation timestamp record is generated by the system through remote command issuance and on-site confirmation of two nodes to ensure the completeness of the operation closed loop; the maintenance progress is automatically updated by the field operation terminal based on the task completion flag and the geographic location check-in mechanism to form a structured timeline type operation log; after archiving the information, the operation and maintenance knowledge base automatically marks the strategy execution effect level based on the historical execution effect analysis algorithm, and re-trains the key parameters in the diagnosis model and the scheduling model.

[0054] The embodiments of the application are described in detail above in combination with the drawings, but the application is not limited thereto, and various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the purpose of the application.

Claims

1. A power distribution automation operation and maintenance method based on power big data, characterized in that, The method comprises the following steps: S1, collecting and processing power distribution operation data in real time through a feeder terminal, an intelligent switch and a monitoring device, calling a fault feature rule set to analyze the power distribution operation data, and identifying and detecting a power distribution fault event; S2, identifying a fault feeder and a fault section based on an electrical topology structure of a current power distribution network, protection action information and a power distribution node breaking state, locating a fault event area by using a switch quantity logical relationship and a topology reachability model, and outputting a fault diagnosis result data set; S3, generating a fault isolation operation strategy and a maintenance task priority list by using an operation and maintenance scheduling model based on the fault diagnosis result data set and a load transfer capability, a field operation capability and historical maintenance resource distribution information of a power distribution station; S4, executing the fault isolation operation strategy according to the maintenance task priority list, collecting power distribution state quantity changes and strategy response feedback information in real time, and archiving them to an operation and maintenance knowledge base. 2.The power grid big data based power distribution automation operation and maintenance method of claim 1, wherein: The power distribution operation data comprises line voltage, line current, switch state, power quality parameter, power factor, frequency, device operation temperature and historical alarm state information. 3.The power grid big data based power distribution automation operation and maintenance method of claim 2, wherein: The fault feature rule set is called in parallel by a power distribution automation master station system based on the operation data, and comprises condition judgment rules of current mutation, switch jump, protection action, voltage drop, zero sequence current offset and power distribution node state anomaly, and specifically as follows: A current mutation judgment rule based on a current mutation amplitude of line current exceeding a set threshold in a continuous sampling period; a switch jump judgment rule based on a switch state changing from a closed position to an open position and from the open position to the closed position; a protection action judgment rule based on a protection device start signal; a voltage drop judgment rule based on a line voltage drop amplitude exceeding a set proportion; a zero sequence current offset judgment rule based on a zero sequence current offset amount from a reference threshold; and a power distribution node state anomaly judgment rule based on a connection state anomaly of a power distribution node in a topology model.

4. The power distribution automation operation and maintenance method based on power big data according to claim 3, characterized in that: The topology reachability model is constructed by the power distribution automation master station system, and is generated based on electrical connection relationships between power distribution nodes in a power distribution network, on-off states of feeder switches and protection device configuration information, and comprises a comprehensive topology graph structure including a node connection graph, a protection action graph and a fault flow path graph, and a reachability judgment of a fault propagation path is performed based on a graph search algorithm, and a fault influence area candidate path set is output.

5. The power distribution automation operation and maintenance method based on power big data according to claim 4, characterized in that: The switch quantity logical relationship is established by a logical judgment module configured in the power distribution automation master station system, the logical judgment module matches combination relationships of on-off states of upper switches, branch switches and terminal switches of power distribution nodes on a same feeder, performs path consistency verification on breaking change sequences before and after a fault occurs by constructing a state link graph and a time sequence reasoning chain, and is used for auxiliary judgment of a fault section.

6. The power distribution automation operation and maintenance method based on power big data according to claim 5, characterized in that: The fault diagnosis result data set comprises a structured diagnosis field; The structured diagnosis field at least comprises a fault feeder number, a fault section start node number, a fault section end node number, a fault judgment timestamp, a fault type label and a diagnosis confidence level, and is used for providing a fault isolation strategy generation and a power supply path reconstruction calculation to an operation and maintenance scheduling system.

7. The power grid data-based power distribution automation operation method of claim 6, wherein: The dispatch optimization model used in the operation and maintenance scheduling system constructs an optimization objective function based on the fault diagnosis result dataset; The optimization objective function takes minimizing fault isolation time, minimizing user power interruption capacity, and minimizing repair resource path length as optimization objectives, and the constraint conditions include topological connectivity constraints, distribution station switching capacity constraints, and field operation ability boundary conditions; The dispatch optimization model uses a mixed integer linear programming method to solve the strategy. 8.The power grid big data based power distribution automation operation and maintenance method of claim 7, wherein: The load transfer capacity is dynamically evaluated by the distribution automation master station system in combination with the current operation mode, standby power output capacity, transformer load rate, and feeder switch operable state, and is used as a path feasibility screening factor in the dispatch optimization model to determine whether a feasible path set for implementing power supply restoration operations exists. 9.The power grid big data based power distribution automation operation and maintenance method of claim 8, wherein: The repair task priority list is comprehensively scored and sorted by the operation and maintenance scheduling system according to the fault type label, fault section covered load, distribution node power supply level, and real-time location of repair resources included in the fault diagnosis result dataset. The process of comprehensive scoring and sorting uses a priority scoring function to construct a multi-factor weighted model, outputs the sorted repair work plan list and dispatch work order data, and finally obtains the repair task priority list.

10. The power distribution automation operation and maintenance method based on power big data according to claim 9, characterized in that: The response feedback information includes the execution state of the remote control device, the repair progress of the field operation terminal, the distribution node state quantity change information, and the operation timestamp record; the response feedback information is archived to the operation and maintenance knowledge base by the operation and maintenance scheduling system after the fault isolation and power supply restoration operations are completed. The operation and maintenance knowledge base is used to update the parameter weights of the optimization dispatch optimization model.

Citation Information

Cited By

  • AI drive power distribution fault positioning system and method

    CN121770157A

  • An AI-driven power distribution fault location system and method

    CN121770157B

  • Method for judging power loss event type of station power

    CN121856692A