Power distribution network operation checking method and system based on knowledge graph

Abstract

The invention discloses a power distribution network operation checking method and system based on a knowledge graph, and belongs to the field of power systems, and the method comprises the steps: obtaining a real-time anti-misoperation knowledge graph corresponding to a power distribution network in response to an operation instruction of the power distribution network; obtaining an operation event corresponding to the operation instruction, and extracting a checking rule corresponding to the operation event from the real-time anti-misoperation knowledge graph; performing anti-error checking on the operation event according to the checking rule to obtain an anti-error checking result corresponding to the operation event; and acquiring real-time operation data of the power distribution network according to the anti-error checking result, and performing abnormality checking on the operation event according to the real-time operation data to obtain an abnormality checking result corresponding to the operation event. The problem that the accuracy and the efficiency are not high due to the fact that checking is difficult to deal with a complex and real-time changing power distribution network environment and potential risks are prone to being neglected in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular to a distribution network operation verification method and system based on a knowledge graph. Background Art

[0002] In modern power systems, with the development of renewable energy, the operating characteristics of distribution networks are gradually evolving toward a high proportion of renewable energy and a high number of power electronic devices. Consequently, the operation and regulation of distribution networks are becoming increasingly complex and challenging to manage, placing higher demands on real-time operation and regulation. This is crucial to ensuring the safe operation of power equipment and preventing electrical accidents caused by human error.

[0003] In the existing technology, the operation of the distribution network involves multiple links such as equipment switching, line scheduling, and fault handling. In order to ensure the compliance and safety of the operation, most traditional operation verification methods rely on manual inspection or static verification mechanisms based on rule engines. These methods usually use preset rules and experience judgments to perform operation verification. For example, a common practice is to ensure that the operation complies with regulations through steps such as manual monitoring of equipment status, checking operation sequence, and confirming equipment permissions, or to use a rule engine to perform simple verification based on input operating parameters (such as equipment status, operating conditions, etc.). However, this method is often not flexible enough and difficult to cope with the complex and real-time changing distribution network environment. It is also easy to ignore potential risks, resulting in low verification accuracy and efficiency. Summary of the Invention

[0004] The present invention provides a distribution network operation verification method and system based on knowledge graph, which can solve the problems in the existing technology of low accuracy and efficiency caused by the difficulty of verification in coping with the complex and real-time changing distribution network environment and the easy neglect of potential risks.

[0005] The present invention discloses a distribution network operation verification method based on knowledge graph, comprising:

[0006] Responding to an operation instruction of the distribution network, obtaining a real-time anti-misoperation knowledge graph corresponding to the distribution network;

[0007] Obtaining an operation event corresponding to the operation instruction, and extracting a verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph;

[0008] Performing error-proofing verification on the operation event according to the verification rules to obtain an error-proofing verification result corresponding to the operation event;

[0009] The real-time operation data of the distribution network is acquired according to the error prevention check result, so as to perform an abnormality check on the operation event according to the real-time operation data and obtain an abnormality check result corresponding to the operation event.

[0010] The present invention discloses a distribution network operation verification method based on a knowledge graph, which responds to the operation instruction of the distribution network to obtain the real-time anti-misoperation knowledge graph of the distribution network, so as to use the real-time anti-misoperation knowledge graph to perform anti-misoperation verification on the operation event in the operation instruction. Among them, based on the real-time anti-misoperation knowledge graph, the verification rules corresponding to the operation event are determined to perform anti-misoperation verification, which can realize automatic verification and improve the efficiency of anti-misoperation verification. At the same time, the anti-misoperation verification knowledge graph can realize a multi-dimensional and refined description of the distribution network, thereby improving the accuracy of anti-misoperation verification. Then, after obtaining the anti-misoperation verification result corresponding to the operation event by using the knowledge graph, the operation event is abnormally verified according to the real-time operation data of the distribution network to identify the potential risks of the operation event and improve the accuracy of the operation verification.

[0011] As a preferred example, the step of responding to an operation instruction of the distribution network and obtaining a real-time anti-misoperation knowledge graph corresponding to the distribution network includes:

[0012] Responding to an operation instruction of the distribution network, obtaining device data of the distribution network; wherein the device data includes the operating status, operation record, fault record and personnel authority corresponding to each device in the distribution network;

[0013] According to the device data, the devices in the distribution network, the operations corresponding to the devices, the personnel corresponding to the operations, and the faults corresponding to the devices are taken as entities of the knowledge graph;

[0014] According to the device data, the attribute data of each entity and the entity relationship between each two entities are obtained, so as to construct a real-time anti-misoperation knowledge graph corresponding to the distribution network according to the entity relationship, the entities and the attribute data.

[0015] In the above scheme, in response to the operation instructions of the distribution network, various types of data are obtained from the distribution network, including the operating status, operation records, fault records and personnel permissions corresponding to each device in the distribution network, so as to ensure the integrity of the information and realize the multi-dimensional and refined description of the distribution network by the knowledge graph, thereby improving the accuracy of the operation event verification using the knowledge graph.

[0016] As a preferred example, the obtaining of the operation event corresponding to the operation instruction and the extraction of the verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph include:

[0017] Parsing the operation instruction to obtain an operation event in the operation instruction; wherein the operation event includes an operation type, a target operation device, an operator, and an operation time;

[0018] Extracting, from the nodes of the real-time anti-misoperation knowledge graph according to the operation event, several operation entities corresponding to the operation event and attribute information of each of the operation entities;

[0019] The verification rule corresponding to each of the operation entities is obtained from a preset rule library.

[0020] In this approach, when an operation request is triggered, entity recognition and relationship tracking are performed within the knowledge graph based on query conditions such as the operation type, operating device, performer information, and time information contained in the operation request. This allows for dynamic verification information retrieval and improves verification efficiency. Next, verification rules corresponding to the extracted operation entities are matched from a pre-set rule library, achieving a one-to-one verification between verification rules and entities, improving verification accuracy.

[0021] As a preferred example, performing error-proof verification on the operation event according to the verification rule to obtain an error-proof verification result corresponding to the operation event includes:

[0022] Obtaining a preliminary error prevention verification result of the operation event according to the verification rule and the attribute information;

[0023] When the operation event is determined to be correct based on the preliminary error-proofing verification result, the attribute information of several operation entities is input into the pre-built reasoning model to obtain the legal reasoning result of the operation event, and the legal reasoning result is used as the error-proofing verification result corresponding to the operation event.

[0024] As a preferred example, the calculation expression of the inference model is:

[0025]

[0026] Wherein, P represents the legal reasoning result, indicating whether the operation event is legal; X i Represents the attribute data corresponding to each operation entity; the W i represents the weight corresponding to each operation entity; the σ represents the activation function.

[0027] In this solution, pre-stored rules in the rule library are used to initially screen out non-compliant operations, improving verification efficiency. After a preliminary assessment of compliance, a reasoning model is then used to calculate the legality score of the operation, ensuring the accuracy of the operation verification.

[0028] As a preferred example, the step of inputting the attribute information of the plurality of operation entities into a pre-built reasoning model to obtain a legal reasoning result of the operation event includes:

[0029] Performing a weighted summation on each of the plurality of attribute data and its corresponding weight to obtain an intermediate score value;

[0030] The intermediate score value is mapped to a probability value of legitimacy through a preset activation function; wherein the calculation expression of the activation function is:

[0031]

[0032] Wherein, P represents the probability value of legitimacy; X represents the intermediate score value.

[0033] In this solution, the inference model uses an activation function to perform a binary classification judgment on operation compliance, improving the accuracy of operation verification. The model first performs a weighted summation of all operation-related features (i.e., the attribute data of the operation) and their corresponding weights to obtain an intermediate score. This score is then mapped to a probability of legality using an immediate function to complete the operation legality verification, ensuring accuracy.

[0034] As a preferred example, performing error-proof verification on the operation event according to the verification rule to obtain an error-proof verification result corresponding to the operation event further includes:

[0035] Obtaining the risk level corresponding to each attribute data according to a preset classification function;

[0036] The risk levels corresponding to the attribute data are normalized to obtain a probability distribution value of the risk level corresponding to the operation event.

[0037] In the above scheme, based on the legality judgment, the legal operation is further evaluated by risk classification, so as to enable reverse reasoning of the subsequent distribution network operation according to the risk level prediction, thereby improving the efficiency of distribution network maintenance.

[0038] As a preferred example, the acquiring of real-time operating data of the distribution network according to the error prevention verification result, performing abnormality verification on the operation event according to the real-time operating data, and obtaining an abnormality verification result corresponding to the operation event, includes:

[0039] When the operation event is determined to be legal according to the error prevention check result, collecting real-time operation data of the distribution network after executing the operation event;

[0040] Acquire historical operating data of the distribution network, and acquire a deviation between the real-time operating data and the historical operating data;

[0041] When the degree of deviation is greater than a preset threshold, the operation event is determined to be an abnormal operation event.

[0042] In the above scheme, after judging the legality of the operation itself, in order to avoid the impact of abnormal operations on the safety of the distribution network, it is also necessary to judge the abnormality of the operation based on the real-time operating data of the distribution network after executing the operation, and then identify the potential risks of the operation and improve the accuracy of operation identification.

[0043] As a preferred example, obtaining the deviation between the real-time operation data and the historical operation data includes:

[0044] The historical operation data and the real-time operation data are respectively standardized according to a preset standardization formula to obtain standardized historical operation data and standardized real-time operation data; wherein the standardization formula is as follows:

[0045]

[0046] Wherein, X′ is the data after standardization; X represents the original data; μ represents the mean difference of the historical operation data; represents the mean deviation of the historical operating data;

[0047] The standardized historical operation data and the standardized real-time operation data are respectively subjected to dimensionality reduction processing, so as to calculate the deviation degree based on the reduced dimensionality historical operation data and the reduced dimensionality real-time operation data obtained after the dimensionality reduction; wherein the calculation formula of the deviation degree is:

[0048]

[0049] Among them, D represents the deviation, X i is the data point of the current operation, μ is the mean of the historical operation data, and N is the number of data points.

[0050] In the above scheme, data standardization and dimensionality reduction techniques are used to ensure that data at different time points can be compared on the same scale, thereby improving the accuracy of anomaly detection.

[0051] The present invention also discloses a distribution network operation verification system based on knowledge graph, which includes a graph acquisition module, a rule retrieval module, an anti-error verification module and an abnormality verification module;

[0052] The graph acquisition module is used to respond to the operation instructions of the distribution network and obtain the real-time anti-misoperation knowledge graph corresponding to the distribution network;

[0053] The rule retrieval module is used to obtain the operation event corresponding to the operation instruction, and extract the verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph;

[0054] The error prevention check module is used to perform an error prevention check on the operation event according to the check rule to obtain an error prevention check result corresponding to the operation event;

[0055] The abnormality verification module is used to obtain the real-time operation data of the distribution network according to the anti-error verification result, so as to perform abnormality verification on the operation event according to the real-time operation data and obtain the abnormality verification result corresponding to the operation event.

[0056] The present invention discloses a distribution network operation verification system based on a knowledge graph, which responds to the operation instructions of the distribution network to obtain the real-time anti-misoperation knowledge graph of the distribution network, so as to use the real-time anti-misoperation knowledge graph to perform anti-misoperation verification on the operation events in the operation instructions. Among them, based on the real-time anti-misoperation knowledge graph, the verification rules corresponding to the operation events are determined to perform anti-misoperation verification, which can realize automatic verification and improve the efficiency of anti-misoperation verification. At the same time, the anti-misoperation verification knowledge graph can realize a multi-dimensional and refined description of the distribution network, thereby improving the accuracy of anti-misoperation verification. Then, after obtaining the anti-misoperation verification result corresponding to the operation event by using the knowledge graph, the operation event is abnormally verified according to the real-time operation data of the distribution network to identify the potential risks of the operation event and improve the accuracy of the operation verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the implementation. Obviously, the drawings described below are only some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0058] Figure 1 This is a flow chart of a method for verifying distribution network operations based on a knowledge graph provided by an embodiment of the present invention;

[0059] Figure 2 This is a schematic diagram of the structure of a distribution network operation verification system based on a knowledge graph provided by an embodiment of the present invention;

[0060] Figure 3 It is a flow chart of a distribution network operation verification method based on a knowledge graph provided in another embodiment of the present invention. DETAILED DESCRIPTION

[0061] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0063] In the description of the embodiments of this application, the technical terms "first" and "second" are used only to distinguish different objects and should not be understood to indicate or imply relative importance or implicitly specify the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined.

[0064] References to "embodiments" herein mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0065] In the description of the embodiments of this application, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0066] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0067] In the description of the embodiments of the present application, unless otherwise expressly specified or limited, technical terms such as "installed," "connected," "connected," and "fixed" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integration; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; internal connections between two components or interactions between two components. Those skilled in the art can understand the specific meanings of the above terms in the embodiments of the present application based on specific circumstances.

[0068] See also Figure 1 To address the problems of low accuracy and efficiency in existing technologies, which are difficult to cope with the complex and real-time changing distribution network environment and prone to ignoring potential risks, an embodiment of the present invention provides a distribution network operation verification method based on a knowledge graph, including:

[0069] Step 101: Respond to an operation instruction of the distribution network and obtain a real-time anti-misoperation knowledge graph corresponding to the distribution network.

[0070] In this embodiment, this step mainly includes: responding to the operation instructions of the distribution network, obtaining the equipment data of the distribution network; wherein, the equipment data respectively corresponds to the operating status, operation records, fault records and personnel permissions of each device in the distribution network; based on the equipment data, the equipment in the distribution network, the operations corresponding to the equipment, the personnel corresponding to the operations and the faults corresponding to the equipment are used as entities of the knowledge graph; based on the equipment data, the attribute data of each entity and the entity relationship between each of the entities are obtained, so as to construct a real-time anti-misoperation knowledge graph corresponding to the distribution network based on the entity relationship, the entity and the attribute data.

[0071] In this embodiment, the above steps respond to the operation instructions of the distribution network and obtain various types of data from the distribution network, including the operating status, operation records, fault records and personnel permissions corresponding to each device in the distribution network, so as to ensure the integrity of the information and realize the multi-dimensional and refined description of the distribution network by the knowledge graph, thereby improving the accuracy of the operation event verification using the knowledge graph.

[0072] Step 102: Obtain an operation event corresponding to the operation instruction, and extract a verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph.

[0073] In this embodiment, this step mainly includes: parsing the operation instruction to obtain the operation event in the operation instruction; wherein, the operation event includes the operation type, target operation device, operator and operation time; according to the operation event, extracting several operation entities corresponding to the operation event and the attribute information of each operation entity from the node of the real-time anti-misoperation knowledge graph; obtaining the verification rules corresponding to each operation entity from the preset rule library.

[0074] In this embodiment, when an operation request is triggered, this step uses the operation type, operation device, execution personnel information, and time information contained in the operation request as query conditions to perform entity recognition and relationship tracking in the knowledge graph, thereby achieving dynamic verification information retrieval and improving verification efficiency. Next, verification rules corresponding to the extracted operation entities are matched from a preset rule library, achieving a one-to-one verification between verification rules and entities, thereby improving verification accuracy.

[0075] Step 103: Perform error-proofing verification on the operation event according to the verification rule to obtain an error-proofing verification result corresponding to the operation event.

[0076] In this embodiment, this step mainly includes: obtaining a preliminary error-proofing verification result of the operation event based on the verification rules and the attribute information; when the operation event is determined to be correct based on the preliminary error-proofing verification result, inputting the attribute information of several operation entities into a pre-built reasoning model to obtain a legal reasoning result of the operation event, and using the legal reasoning result as the error-proofing verification result corresponding to the operation event; wherein the calculation expression of the reasoning model is:

[0077]

[0078] Wherein, P represents the legal reasoning result, indicating whether the operation event is legal; X i Represents the attribute data corresponding to each operation entity; the W i represents the weight corresponding to each operation entity; the σ represents the activation function.

[0079] In one implementation of this embodiment, each of the plurality of attribute data and its corresponding weight is weighted and summed to obtain an intermediate score value; the intermediate score value is mapped to a probability value of legitimacy through a preset activation function; wherein the calculation expression of the activation function is:

[0080]

[0081] Wherein, P represents the probability value of legitimacy; X represents the intermediate score value.

[0082] In one implementation of this embodiment, it also includes: obtaining the risk level corresponding to each of the attribute data according to a preset classification function; normalizing the multiple risk levels corresponding to the multiple attribute data to obtain a risk level probability distribution value corresponding to the operation event.

[0083] In this embodiment, the above steps use the pre-saved rules in the rule base to preliminarily screen out non-standard operations, thereby improving the efficiency of verification. Then, under the condition of preliminary evaluation of the specifications, the inference model is used to calculate the legality score of the operation to ensure the accuracy of the operation verification. Among them, the inference model uses an activation function to complete the binary classification judgment of the operation compliance, thereby improving the accuracy of the operation verification. Among them, the model first performs a weighted summation of all operation-related features, i.e., the attribute data of the operation and its corresponding weight to obtain an intermediate score, and then maps it to a probability value of legality through an immediate function to complete the verification of the legality of the operation and ensure the accuracy of the verification. On the basis of the legality judgment, the legal operation is further evaluated by risk grading, so that the subsequent reverse reasoning of the operation can be performed according to the risk level prediction, thereby improving the efficiency of distribution network maintenance.

[0084] Step 104: Acquire real-time operating data of the distribution network according to the error prevention check result, perform an abnormality check on the operation event according to the real-time operating data, and obtain an abnormality check result corresponding to the operation event.

[0085] In this embodiment, this step mainly includes: when the operation event is determined to be legal based on the anti-error verification result, collecting the real-time operation data of the distribution network after executing the operation event; obtaining the historical operation data of the distribution network, and obtaining the deviation between the real-time operation data and the historical operation data; when the deviation is greater than a preset threshold, determining that the operation event is an abnormal operation event.

[0086] The historical operation data and the real-time operation data are respectively standardized according to a preset standardization formula to obtain standardized historical operation data and standardized real-time operation data; wherein the standardization formula is as follows:

[0087]

[0088] Wherein, X′ is the data after standardization; X represents the original data; μ represents the mean difference of the historical operation data; represents the mean deviation of the historical operating data;

[0089] The standardized historical operation data and the standardized real-time operation data are respectively subjected to dimensionality reduction processing, so as to calculate the deviation degree based on the reduced dimensionality historical operation data and the reduced dimensionality real-time operation data obtained after the dimensionality reduction; wherein the calculation formula of the deviation degree is:

[0090]

[0091] Among them, D represents the deviation, X i is the data point of the current operation, μ is the mean of the historical operation data, and N is the number of data points.

[0092] In this embodiment, after determining the legitimacy of the operation itself, the aforementioned steps further determine the anomaly of the operation based on the real-time operational data of the distribution network after the operation is executed. This identifies the potential risks of the operation and improves the accuracy of operation identification. Data normalization and dimensionality reduction techniques are used to ensure that data at different time points can be compared on the same scale, thereby improving the accuracy of anomaly detection.

[0093] like Figure 2 As shown, based on the above-mentioned method item embodiment, a corresponding device item embodiment is provided; this embodiment provides a distribution network operation verification system based on a knowledge graph, including a graph acquisition module 201, a rule retrieval module 202, an anti-error verification module 203 and an abnormality verification module 204.

[0094] The graph acquisition module 201 is used to respond to the operation instructions of the distribution network and obtain the real-time anti-misoperation knowledge graph corresponding to the distribution network.

[0095] The rule retrieval module 202 is used to obtain the operation event corresponding to the operation instruction and extract the verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph.

[0096] The error prevention check module 203 is used to perform an error prevention check on the operation event according to the check rule to obtain an error prevention check result corresponding to the operation event.

[0097] The abnormality verification module 204 is used to obtain the real-time operation data of the distribution network according to the error prevention verification result, so as to perform abnormality verification on the operation event according to the real-time operation data and obtain the abnormality verification result corresponding to the operation event.

[0098] It can be understood that the above-mentioned device item embodiments correspond to the method item embodiments, which can implement a distribution network operation verification method based on knowledge graph provided by any method item embodiment.

[0099] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Furthermore, in the drawings of the device embodiments provided in this embodiment, the connection relationship between modules indicates that they have a communication connection, which may be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement the present invention without inventive effort.

[0100] Based on the above-mentioned embodiment of a distribution network operation verification method based on a knowledge graph, this embodiment also provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, a distribution network operation verification method based on a knowledge graph of this embodiment is implemented.

[0101] For example, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement this embodiment. The one or more module elements may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0102] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0103] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0104] Based on the above-mentioned method embodiment, this embodiment also provides a computer-readable storage medium, including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a distribution network operation verification method based on a knowledge graph as described in the method embodiment.

[0105] Wherein, the module / unit integrated in the device / terminal equipment, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the process in the method of this embodiment can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0106] This embodiment provides a distribution network operation verification method, system, device and medium based on a knowledge graph, which responds to the operation instruction of the distribution network to obtain the real-time anti-misoperation knowledge graph of the distribution network, so as to use the real-time anti-misoperation knowledge graph to perform anti-misoperation verification on the operation event in the operation instruction. Among them, based on the real-time anti-misoperation knowledge graph, the verification rules corresponding to the operation event are determined to perform anti-misoperation verification, which can realize automatic verification and improve the efficiency of anti-misoperation verification. At the same time, the anti-misoperation verification knowledge graph can realize a multi-dimensional and refined description of the distribution network, thereby improving the accuracy of anti-misoperation verification. Then, after obtaining the anti-misoperation verification result corresponding to the operation event by using the knowledge graph, the operation event is abnormally verified according to the real-time operation data of the distribution network to identify the potential risks of the operation event and improve the accuracy of the operation verification.

[0107] Example 2

[0108] Existing distribution network operation verification methods primarily rely on manual intervention or static rule engines, and are unable to dynamically adapt to real-time changes and complex operational scenarios within the distribution network. As distribution networks evolve toward intelligent systems, traditional operation verification methods are unable to promptly respond to changes in various real-time data within the distribution network, nor can they flexibly adjust operation verification rules to address complex operational situations. Consequently, they lack both real-time and intelligent capabilities.

[0109] In order to solve the problems existing in the above-mentioned existing technologies, this embodiment provides a distribution network operation verification method based on knowledge graph, which utilizes advanced knowledge graph technology, combines real-time data and dynamic algorithms, and improves the real-time and intelligence level of distribution network operation verification, thereby realizing the automation and intelligent verification of the operation process, ensuring the compliance and safety of the operation, and significantly improving efficiency.

[0110] Specifically, please refer to the implementation process of the distribution network operation verification method. Figure 3 , mainly including steps 301 to 305:

[0111] Step 301: Collect equipment data, operation data, fault data and personnel data of the distribution network to construct a knowledge graph for error prevention and verification corresponding to the distribution network.

[0112] In this embodiment, the operating status, operation records, fault records and personnel permissions corresponding to each device in the distribution network are obtained, so that the devices in the distribution network, the operations corresponding to the devices, the personnel corresponding to the operations and the faults corresponding to the devices are used as entities of the knowledge graph; the attribute data of each entity and the entity relationship between each of the entities are obtained based on the device data, so as to construct a real-time anti-misoperation knowledge graph corresponding to the distribution network based on the entity relationship, the entities and the attribute data.

[0113] Specifically, in this embodiment, first, various types of data of the distribution network are collected, including the operating status of the equipment, operation records, fault history, personnel permissions, etc. Among them, the construction of the knowledge graph depends on the extraction of key entities and their associations from the structured data of the distribution network. Entity extraction uses named entity recognition (NER) technology, combined with data sources such as equipment databases, operation logs, and personnel permission tables, to automatically identify core entities such as equipment, operations, personnel, and faults in the distribution network. For equipment entities, they are classified and archived based on information such as equipment type, operating status, and topological relationships; for operation entities, they are labeled based on operation instructions, operation sequence, and applicable equipment; for personnel entities, mapping relationships are established based on information such as identity permissions and operation permissions.

[0114] Secondly, after the entities are constructed, the edge relationships in the knowledge graph are screened using predefined rules and data-driven learning models. The rule-based approach involves setting constraints between entities. For example, if a device in the operation log is affected by a certain operation at a certain moment, an "operation-device" relationship is constructed in the knowledge graph; if a person has performed a certain type of operation, a "personnel-operation" relationship is constructed. The data-driven approach uses graph neural networks (GNNs) to train historical data on the distribution network, and automatically mines potential associations between entities through pattern recognition technology, such as "a certain device usually affects the load in a certain area under a certain fault state," thereby improving the dynamic adaptability of the knowledge graph.

[0115] In this embodiment, the above data is collected in real time across all aspects of the distribution network to ensure information integrity. Using a graph database, this data is converted into nodes and edges to construct a knowledge graph. In this knowledge graph, nodes represent various entities in the distribution network, such as equipment, operations, faults, and personnel, while edges represent the relationships between nodes. For example, the relationship between equipment and operations, the relationship between equipment and faults, and the relationship between operations and personnel permissions.

[0116] In this embodiment, in order to ensure the timeliness and accuracy of the graph, the knowledge graph needs to be dynamically updated. As the distribution network operates, the equipment status, operating data, and fault information will continue to change. Therefore, the nodes and edges in the graph also need to be updated as new data is input. The update formula can be expressed as:

[0117] ΔG={ΔV,ΔE}

[0118] Here, ΔG represents the incremental update of the graph, while ΔV and ΔE represent the changes in nodes and edges, respectively. This way, the system can update the graph in a timely manner, ensuring that each operation verification is based on the latest data.

[0119] Step 302: Obtain an operation event in the operation instruction of the distribution network, obtain attribute data corresponding to the operation event from the anti-mischeck knowledge graph, and obtain the verification rule corresponding to the operation event in combination with the pre-built rule knowledge base.

[0120] In this embodiment, the operation instruction is parsed to obtain the operation event in the operation instruction; wherein, the operation event includes the operation type, target operation device, operator and operation time; the operation event extracts several operation entities corresponding to the operation event and the attribute information of each operation entity from the node of the real-time anti-misoperation knowledge graph; and the verification rule corresponding to each operation entity is obtained from the preset rule library.

[0121] Specifically, in this embodiment, when the distribution network triggers and receives an operation request, it uses the operation type, operation object (such as a specific device), execution personnel information, and timestamp contained in the operation request as query conditions to perform entity recognition and relationship tracking in the knowledge graph. Using the structured parameters in the operation request, the graph database's graph traversal and indexing mechanism automatically retrieves entity nodes related to the operation and their associated relationships.

[0122] Next, the relationships between entities stored in the knowledge graph are equipped with semantic labels, and the associated nodes (such as historical operation nodes, personnel authority nodes, and equipment operation status nodes) are located through the operation objects, and the weight of the edge is used as a screening mechanism to extract the entity set most closely related to the current operation. The extracted historical records, equipment status, and permission information all call the latest state of the graph in real time without the need for separate caching, thereby realizing dynamic verification information calling. In this embodiment, after extracting the entity set, the historical records, equipment status, and permission constraint information required for each entity are dynamically extracted through the semantic edge relationship between entities, ensuring the real-time and accuracy of the verification judgment.

[0123] The characteristic information of the operation request includes but is not limited to the operation object (equipment), operation type (switch, dispatch, maintenance, etc.), execution personnel, timestamp, etc. Based on this characteristic information, entities in the knowledge graph are matched and historical records, equipment status, and permission constraints related to the operation are dynamically extracted to ensure data integrity and real-time performance during the inference process.

[0124] Next, after the entity set is extracted, a verification rule corresponding to each entity is obtained from a predefined rule library, and the verification rule corresponds to the attribute data of the entity one-to-one.

[0125] Step 303: Preliminary screening of the operation events is performed according to the verification rules to obtain preliminary screening results of the operation events.

[0126] In this embodiment, a preliminary error prevention verification result of the operation event is obtained according to the verification rule and the attribute information.

[0127] Specifically, by comparing attribute fields such as device status and personnel permissions extracted from the knowledge graph with the conditions set in the rule base, a rule-level verification of the legality of operations is achieved. There is a one-to-one correspondence between predefined rules and entity attributes, and the rule verification process can be automatically executed and responded to in real time.

[0128] Among them, in the reasoning and verification process, predefined rules are used to preliminarily screen out situations that do not comply with operating specifications. For example: if the equipment is in maintenance mode, switch operations are prohibited; if the operator does not have execution authority, the operation is rejected. When performing operation verification, the entity attribute information extracted from the knowledge graph (such as equipment status field, personnel authority level) is mapped and compared with the predefined rules. For example, if the rule stipulates that a certain operation must be "only allowed to be executed when the equipment operating status is 'standby'", the system will match the status attribute value of the device node in the knowledge graph to determine whether the condition is met; if the rule stipulates that "the operator's authority level must be level two or above", the authority level of the personnel node is extracted from the graph and compared with the rule threshold. If any rule condition is not met, the system will directly interrupt the reasoning process and output a verification failure message.

[0129] Step 304: Based on the preliminary screening result, obtain the legality and compliance reasoning result of the operation event through a preset reasoning model.

[0130] In this embodiment, when the operation event is determined to be correct based on the preliminary error prevention verification results, the attribute information of the multiple operation entities is input into a pre-built inference model to obtain a legitimate inference result for the operation event, which is then used as the error prevention verification result for the operation event. Simultaneously, the risk level corresponding to each piece of attribute data is obtained based on a preset classification function. The risk levels corresponding to the multiple pieces of attribute data are normalized to obtain a probability distribution value for the risk level corresponding to the operation event.

[0131] Specifically, in this embodiment, when the operation event is determined to be correct according to predefined rules during the reasoning verification process, the reasoning model is used to calculate the legality score of the operation for the complex decision-making scenarios that may result from the operation event. The reasoning formula is as follows:

[0132]

[0133] Among them, P is the inference result, indicating whether the operation is legal; X i For operation-related data (such as device status, operation permissions, etc.); W i is the weight of each type of data, and σ is the activation function.

[0134] In this embodiment, the above reasoning model can quickly determine whether the operation is legal and give the result. The weight parameter W in the initial stage of the model is iThe weights are set based on expert experience or statistical analysis, for example, 0.4 for device status, 0.3 for operation history, and 0.3 for personnel permissions. During operation, feedback is provided based on the results of each operation verification (success, failure, or warning), and weights are optimized and updated using a gradient descent method. For different operation types (such as switching and maintenance operations), the system classifies the operation type and calls the corresponding sub-model and weight allocation strategy to ensure that the model's decision logic matches the operational scenario.

[0135] The inference model uses the Sigmoid activation function to perform a binary classification judgment of operation compliance. The model first performs a weighted summation of all operation-related features (such as device status, personnel permissions, etc.) and their corresponding weights to obtain an intermediate score x, which is then mapped to a probability value P of legality through the Sigmoid function. The calculation expression of the probability value of the intermediate score set is as follows:

[0136]

[0137] Among them, x represents the middle score; P represents the probability value of legality, which is used to determine whether the current operation is legal. If P is higher than the set threshold, the operation is considered compliant; otherwise, it is considered illegal.

[0138] Based on the compliance judgment, the legal operation can be further evaluated by risk classification. In this scenario, the Softmax activation function is introduced to classify different risk levels (such as high risk, medium risk, and low risk). The input of Softmax is a set of independent score values {x1, x2, ..., x k Each score is calculated by combining different features and is finally normalized into a risk level probability distribution:

[0139]

[0140] Among them, the represents the independent scoring values corresponding to different attribute data; and K represents the number of the independent scoring values.

[0141] In this embodiment, the legality and compliance reasoning mechanism can be used in scenarios such as scheduling assistance and early warning response, without affecting the basic legality judgment function.

[0142] Step 305: Based on the legal compliance reasoning result, real-time operating data of the distribution network after executing the operation event is collected to perform abnormal reasoning on the operation event based on the real-time operating data to obtain an abnormal verification result of the operation event.

[0143] In this embodiment, when the operation event is determined to be legal based on the anti-error verification result, the real-time operation data of the distribution network after executing the operation event is collected; the historical operation data of the distribution network is obtained, and the deviation between the real-time operation data and the historical operation data is obtained; when the deviation is greater than a preset threshold, the operation event is determined to be an abnormal operation event.

[0144] Specifically, during anomaly detection, historical and real-time operation data are used for comparative analysis. Historical operation data refers to static information stored in the knowledge graph, such as the device's historical operating status, past operation logs, and fault records. It has strong temporal and statistical characteristics. Real-time operation data includes dynamic data such as the device's current status, sensor readings, and operating instructions. It is updated frequently, and the distribution of values may differ significantly from historical data.

[0145] In order to ensure the alignment of historical data and real-time data when calculating the deviation, the data is normalized so that the data in different time periods have the same scale. The normalization formula is as follows:

[0146]

[0147] Where X′ is the standardized data, X is the original data, μ is the historical data mean, and σ is the standard deviation. This method allows for horizontal comparison of operational data from different time periods, improving the accuracy of anomaly detection.

[0148] In addition, to address the issue of inconsistent dimensions between historical data and real-time data, dimensionality reduction methods such as principal component analysis (PCA) are used to project high-dimensional feature data into a low-dimensional space, ensuring that both types of data are calculated in the same dimension. The reduced-dimensional data set is used to calculate the deviation:

[0149]

[0150] Among them, D represents the deviation, X i is the data point of the current operation, μ is the mean of the historical operation data, and N is the number of data points. If the deviation is greater than the predetermined threshold, the operation is considered abnormal.

[0151] Once an abnormal operation is detected, an early warning will be triggered and relevant personnel will be notified to handle it. The early warning rules are as follows:

[0152]

[0153] The early warning information will include the cause of the abnormality, the scope of impact and corrective measures to ensure that operators take timely action.

[0154] In one embodiment of this invention, a feedback mechanism and self-optimization function are introduced to improve the accuracy and response speed of distribution network operation verification. The results of the operation verification (such as success, failure, and anomaly) and the feedback data generated during the verification process are obtained and analyzed. By learning new operation data, the inference algorithm and verification rules are continuously optimized. During the self-optimization process, the algorithm weights are adjusted using methods such as reinforcement learning to gradually improve the effectiveness of verification and anomaly detection. Through this mechanism, it can continuously adapt to the dynamic changes of the distribution network and improve accuracy and timeliness.

[0155] This embodiment provides a knowledge graph-based distribution network operation verification method. First, this intelligent knowledge graph-based approach eliminates manual intervention and relies on real-time data and reasoning mechanisms for dynamic verification, significantly improving operation accuracy and efficiency. Second, by constructing a knowledge graph for the distribution network, it structures and stores information such as device status, operation records, personnel permissions, and historical faults. This graph is dynamically updated with real-time data to promptly respond to changes in operation requests within the distribution network, providing real-time, accurate verification judgments and avoiding the limitations of static rules. Next, a reasoning algorithm automatically analyzes operation data, combining historical distribution network data with device status to automatically identify potential risks and anomalies. Based on factors such as the device's real-time status, operation history, and personnel permissions, it accurately determines whether operational requirements are met. This reasoning process not only improves verification efficiency but also identifies potentially missed abnormal operations, reducing bias in human judgment. Finally, the system possesses strong adaptive learning capabilities. By learning from feedback from historical operation data, it continuously optimizes verification rules and anomaly detection algorithms, improving the accuracy and response speed of distribution network operation verification. Compared with traditional static rules, the self-optimization mechanism enables the verification process to continuously improve with the accumulation of data and changes in the operating environment, always maintaining high accuracy and timeliness.

[0156] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A distribution network operation verification method based on knowledge graph, characterized in that: include: Responding to an operation instruction of the distribution network, obtaining a real-time anti-misoperation knowledge graph corresponding to the distribution network; Obtaining an operation event corresponding to the operation instruction, and extracting a verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph; Performing error-proofing verification on the operation event according to the verification rules to obtain an error-proofing verification result corresponding to the operation event; The real-time operation data of the distribution network is acquired according to the error prevention check result, so as to perform an abnormality check on the operation event according to the real-time operation data and obtain an abnormality check result corresponding to the operation event.

2. A distribution network operation verification method based on knowledge graph according to claim 1, characterized in that: The step of responding to the operation instruction of the distribution network and obtaining the real-time anti-misoperation knowledge graph corresponding to the distribution network includes: Responding to an operation instruction of the distribution network, obtaining device data of the distribution network; wherein the device data includes the operating status, operation record, fault record and personnel authority corresponding to each device in the distribution network; According to the device data, the devices in the distribution network, the operations corresponding to the devices, the personnel corresponding to the operations, and the faults corresponding to the devices are taken as entities of the knowledge graph; According to the device data, the attribute data of each entity and the entity relationship between each two entities are obtained, so as to construct a real-time anti-misoperation knowledge graph corresponding to the distribution network according to the entity relationship, the entities and the attribute data.

3. A distribution network operation verification method based on knowledge graph according to claim 2, characterized in that: The obtaining of the operation event corresponding to the operation instruction and extracting the verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph includes: Parsing the operation instruction to obtain an operation event in the operation instruction; wherein the operation event includes an operation type, a target operation device, an operator, and an operation time; Extracting, from the nodes of the real-time misoperation prevention knowledge graph according to the operation event, several operation entities corresponding to the operation event and attribute information of each of the operation entities; The verification rule corresponding to each of the operation entities is obtained from a preset rule library.

4. A distribution network operation verification method based on knowledge graph according to claim 3, characterized in that: The performing error-proofing verification on the operation event according to the verification rule to obtain an error-proofing verification result corresponding to the operation event includes: Obtaining a preliminary error prevention verification result of the operation event according to the verification rule and the attribute information; When the operation event is determined to be correct based on the preliminary error-proofing verification result, the attribute information of several operation entities is input into the pre-built reasoning model to obtain the legal reasoning result of the operation event, and the legal reasoning result is used as the error-proofing verification result corresponding to the operation event.

5. A distribution network operation verification method based on knowledge graph according to claim 4, characterized in that: The calculation expression of the inference model is: Wherein, P represents the legal reasoning result, indicating whether the operation event is legal; X i Represents the attribute data corresponding to each operation entity; the W i represents the weight corresponding to each operation entity; the σ represents the activation function.

6. A distribution network operation verification method based on knowledge graph according to any one of claims 4-5, characterized in that: Inputting the attribute information of the plurality of operation entities into a pre-built reasoning model to obtain a legal reasoning result of the operation event includes: Performing a weighted summation on each of the plurality of attribute data and its corresponding weight to obtain an intermediate score value; The intermediate score value is mapped to a probability value of legitimacy through a preset activation function; wherein the calculation expression of the activation function is: Wherein, P represents the probability value of legitimacy; X represents the intermediate score value.

7. A distribution network operation verification method based on knowledge graph according to claim 6, characterized in that: The performing error prevention verification on the operation event according to the verification rule to obtain an error prevention verification result corresponding to the operation event further includes: Obtaining the risk level corresponding to each attribute data according to a preset classification function; The risk levels corresponding to the attribute data are normalized to obtain a probability distribution value of the risk level corresponding to the operation event.

8. The method for verifying distribution network operation based on knowledge graph according to claim 1, characterized in that: The acquiring of real-time operating data of the distribution network according to the error prevention check result, performing an abnormality check on the operation event according to the real-time operating data, and obtaining an abnormality check result corresponding to the operation event, includes: When the operation event is determined to be legal according to the error prevention check result, collecting real-time operation data of the distribution network after executing the operation event; Acquire historical operating data of the distribution network, and acquire a deviation between the real-time operating data and the historical operating data; When the degree of deviation is greater than a preset threshold, the operation event is determined to be an abnormal operation event.

9. A distribution network operation verification method based on knowledge graph according to claim 8, characterized in that: The obtaining of the deviation between the real-time operation data and the historical operation data includes: The historical operation data and the real-time operation data are respectively standardized according to a preset standardization formula to obtain standardized historical operation data and standardized real-time operation data; wherein the standardization formula is as follows: Wherein, X′ is the data after standardization; X represents the original data; μ represents the mean difference of the historical operation data; represents the mean deviation of the historical operating data; The standardized historical operation data and the standardized real-time operation data are respectively subjected to dimensionality reduction processing, so as to calculate the deviation degree based on the reduced dimensionality historical operation data and the reduced dimensionality real-time operation data obtained after the dimensionality reduction; wherein the calculation formula of the deviation degree is: Among them, D represents the deviation, X i is the data point of the current operation, μ is the mean of the historical operation data, and N is the number of data points.

10. A distribution network operation verification system based on knowledge graph, characterized in that: It includes the map acquisition module, rule retrieval module, error prevention verification module and anomaly verification module; The graph acquisition module is used to respond to the operation instructions of the distribution network and obtain the real-time anti-misoperation knowledge graph corresponding to the distribution network; The rule retrieval module is used to obtain the operation event corresponding to the operation instruction, and extract the verification rule corresponding to the operation event from the real-time anti-misoperation knowledge graph; The error prevention check module is used to perform an error prevention check on the operation event according to the check rule to obtain an error prevention check result corresponding to the operation event; The abnormality verification module is used to obtain the real-time operation data of the distribution network according to the anti-error verification result, so as to perform abnormality verification on the operation event according to the real-time operation data and obtain the abnormality verification result corresponding to the operation event.

Images