Power grid equipment fault detailed rule generation method and system based on dynamic knowledge graph

Abstract

The invention discloses a power grid equipment fault detailed rule generation method and system based on a dynamic knowledge graph. The method comprises the following steps: establishing a knowledge graph by utilizing power grid equipment fault text data; extracting features from a power grid equipment fault text through a fault self-diagnosis module by applying an NLP technology, performing fault type diagnosis through text classification, deep learning and an inference algorithm, and continuously optimizing a diagnosis result in combination with operation and maintenance feedback; and according to a fault diagnosis result, automatically generating related supervision rules by utilizing causal reasoning of a knowledge graph. Intelligent diagnosis of power grid equipment faults can be realized, technical supervision suggestions and operation and maintenance measures can be generated, the understanding and operation efficiency of operation and maintenance personnel on the power grid equipment faults is improved, and intelligentization and high efficiency of power grid equipment management are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid equipment fault management, and in particular relates to a method and system for generating power grid equipment fault details based on a dynamic knowledge graph. Background Art

[0002] With the increasing complexity of power systems and the diversification of power grid equipment, traditional methods of fault diagnosis and maintenance management of power grid equipment have been unable to meet the needs of modern power grids. At present, when power grid operation and maintenance personnel conduct equipment fault troubleshooting and formulate maintenance plans, they usually rely on experience and manually consult static data such as equipment manuals and maintenance records. This is inefficient and prone to errors, and it is difficult to quickly respond to complex and changing equipment failures. In addition, the operating status and fault characteristics of power grid equipment will change with equipment aging, environmental changes, and different operating conditions. The traditional static knowledge base cannot be updated in real time and adapt to the emergence of new fault types, resulting in low maintenance efficiency and poor fault diagnosis accuracy.

[0003] In order to improve the efficiency and accuracy of power grid equipment fault diagnosis, intelligent power grid operation and maintenance based on artificial intelligence technology has gradually become a research hotspot in recent years. However, most of the existing intelligent diagnosis methods rely on data-driven machine learning algorithms, and there is still a lack of systems that can effectively process multi-dimensional fault data of power grid equipment and automatically build and evolve knowledge bases. Although some methods attempt to construct equipment fault knowledge graphs, most graph construction methods are relatively static and cannot dynamically respond to the evolution of equipment fault types, nor can they effectively combine fault diagnosis with the generation of intelligent supervision rules. Therefore, how to use advanced knowledge graph technology to realize intelligent diagnosis of power grid equipment faults, dynamically update knowledge bases, and generate personalized intelligent supervision rules is a technical problem that needs to be solved in the current field of power grid equipment management. Summary of the invention

[0004] The present invention provides a method and system for generating power grid equipment fault details based on a dynamic knowledge graph, which is used to solve the technical problem that it is impossible to use advanced knowledge graph technology to realize intelligent diagnosis of power grid equipment faults, dynamically update the knowledge base and generate personalized intelligent supervision details.

[0005] In a first aspect, the present invention provides a method for generating power grid equipment fault details based on a dynamic knowledge graph, comprising:

[0006] Obtaining a power grid equipment fault text, and constructing a dynamic knowledge graph of the power grid equipment according to the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, regulatory details, and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between the nodes;

[0007] Extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result;

[0008] According to a preset causal reasoning mechanism, the dynamic knowledge graph is used to generate regulatory details corresponding to the fault type identification result, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

[0009] In a second aspect, the present invention provides a system for generating power grid equipment fault details based on a dynamic knowledge graph, comprising:

[0010] A construction module is configured to obtain a power grid equipment fault text, and construct a dynamic knowledge graph of the power grid equipment according to the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, supervision details, and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between the nodes;

[0011] An identification module is configured to extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result;

[0012] A generation module is configured to generate regulatory details corresponding to the fault type identification result using the dynamic knowledge graph according to a preset causal reasoning mechanism, wherein the regulatory details include operating procedures, safety standards and operation and maintenance plans.

[0013] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the method for generating fault details of power grid equipment based on a dynamic knowledge graph according to any embodiment of the present invention.

[0014] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of a method for generating fault details of power grid equipment based on a dynamic knowledge graph according to any embodiment of the present invention.

[0015] The method and system for generating fault details of power grid equipment based on dynamic knowledge graphs of the present application can significantly improve the accuracy and efficiency of fault diagnosis of power grid equipment through automated fault diagnosis, intelligent recommendation of maintenance details and real-time updated knowledge graphs, provide technical support for the modern management of the power industry, and have broad application prospects and long-term economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flowchart of a method for generating power grid equipment fault details based on a dynamic knowledge graph provided by an embodiment of the present invention;

[0018] Figure 2 A structural block diagram of a power grid equipment fault details generation system based on a dynamic knowledge graph provided by an embodiment of the present invention;

[0019] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] See also Figure 1 , which shows a flow chart of a method for generating power grid equipment fault details based on a dynamic knowledge graph of the present application.

[0022] like Figure 1 As shown, the method for generating power grid equipment fault details based on dynamic knowledge graph specifically includes the following steps:

[0023] Step S101, obtain the power grid equipment fault text, and construct a dynamic knowledge graph of the power grid equipment based on the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, supervision details and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between nodes.

[0024] In this step, the nodes of the dynamic knowledge graph represent the multi-dimensional information of power grid equipment, including equipment name, key components, fault type, regulatory details, maintenance methods, etc. These nodes are connected by edges. The nodes in the dynamic knowledge graph can be continuously updated through intelligent reasoning to ensure that equipment managers can obtain the latest and most accurate information to diagnose equipment failures and develop maintenance plans.

[0025] The edges of the dynamic knowledge graph represent different types of relationships between nodes, including "equipment-fault" edges, "equipment-component" edges, "component-fault", "fault-repair" edges, "fault-regulatory standard" edges, etc. Each edge has a different weight value, which reflects the closeness of the relationship. For example, some faults are closely related to the type of equipment component, and the weight value is large, indicating that this type of equipment component is more likely to have this type of fault; while some regulatory details are loosely related to the type of fault, and the weight value is small, indicating that the corresponding regulatory details for the fault are diverse and this type of regulatory details is used less frequently. The dynamic adjustment of edge weights can enable the knowledge graph to more accurately reflect the actual fault conditions of the equipment as well as the supervision and maintenance needs.

[0026] It should be noted that the graph neural network is combined with incremental learning technology to dynamically optimize the graph structure. After the initial construction of the knowledge graph, the graph neural network is used to propagate information between devices, automatically learn the potential causal relationship between devices based on the equipment failure history and maintenance data, and adjust the node and edge weights in the graph through back propagation. With the addition of new equipment data, maintenance records and fault reports, an adaptive adjustment algorithm is introduced, and the nodes and edges in the graph are automatically adjusted according to the predetermined algorithm rules, so as to keep the knowledge graph updated and evolved. For example, if a certain fault type is frequently handled by a regulatory detail over a period of time, the weight of the regulatory detail related to the fault type will be increased, and the corresponding operation and maintenance measures will be recommended as a priority. The incremental learning mechanism ensures that new equipment information can be quickly absorbed and integrated into the graph without the need to train from scratch, which significantly improves the efficiency and accuracy of graph evolution, so that the knowledge graph can continuously adapt to new situations as the equipment service life, fault type and operation and maintenance strategy change.

[0027] Step S102: extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result.

[0028] In this step, deep learning-based segmentation models such as Seg-BERT not only focus on word boundaries during the segmentation process, but also dynamically adjust segmentation points based on contextual information to ensure that technical terms or phrase structures are not incorrectly split. Secondly, by using a context-aware irrelevant information filtering mechanism, the semantic contribution of sentences or words is evaluated to automatically filter low semantic relevance information such as redundant descriptions or noise words to improve the purity of the text. Furthermore, by embedding a multi-level feature extraction module, the module combines character-level, word-level, and phrase-level feature extraction strategies to construct a hierarchical feature representation to ensure that micro-semantic information in the text, such as equipment names, equipment components, and macro-structural information such as fault types and operation and maintenance measures, can be fully retained. Finally, through the semantic regularization mechanism, the segmented text is checked for consistency and formatted to ensure the structuring of the text preprocessing results and provide higher-quality semantic input for downstream tasks.

[0029] The multi-head attention mechanism is used to capture the dependencies between words in the text, effectively processing the contextual information in the fault text, especially the semantic relevance of polysemous words, phrase structures, and technical terms. Then, the pre-trained language model SBERT based on the Transformer architecture is used to convert each word in the fault text into a context-related high-dimensional embedding vector, and the fine-grained features related to the fault description are extracted by combining global and local semantic information. At the same time, by dynamically optimizing the weight of the embedding vector, the attention score is used to give higher weights to specific domain terms such as "circuit breaker" and "overload protection", reduce the interference of non-important words such as stop words or high-frequency words, and generate feature representations with higher domain relevance. Finally, the high-dimensional embedding vector is reduced in dimensionality through dynamic pooling and linear projection, and the specific power grid domain rules are introduced in combination with the feature enhancement module to further optimize the accuracy and applicability of the feature representation, providing accurate input support for subsequent classification, reasoning and other tasks.

[0030] Using a deep learning algorithm, the text convolutional neural network model is used to classify the power grid equipment fault text according to the equipment component type, equipment fault type, and fault severity. The text convolutional neural network model can extract local features of the text through sliding convolution kernels, identify key details in the fault text, and convert them into feature vectors that can be used for classification. During the classification process, the text convolutional neural network model extracts multi-level features through a multi-layer network structure, and finally realizes multi-category diagnosis tasks through a normalized exponential function layer.

[0031] Equipment component type: Summarize the components with frequent historical equipment failures and classify them into equipment housing, switches, cooling devices, insulation materials, protective devices, etc.

[0032] Equipment fault type: Classified into different equipment fault types according to the fault characteristics described in the fault text.

[0033] Fault severity: Classify the impact of power grid equipment faults, such as minor faults, moderate faults, and severe faults, in order to determine the subsequent processing priority.

[0034] Step S103, according to a preset causal reasoning mechanism, the dynamic knowledge graph is used to generate regulatory details corresponding to the fault type identification result, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

[0035] In this step, by analyzing multi-dimensional information such as the importance of key components of the equipment, fault type, and fault severity, reasoning is performed based on the nodes and edges in the graph to automatically generate technical details and operation and maintenance measures that are suitable for the current equipment fault status.

[0036] By using the weight information of the relationship edges between nodes in the dynamic knowledge graph, priority is given to recommending regulatory details and operation and maintenance measures related to the fault with higher weights. The severity in the fault classification result determines the priority of the fault handling. For devices with higher fault severity, the recommendation module will prioritize the push of solutions to ensure timely processing. In the reasoning process, the feedback mechanism plays an important role. Whenever the operation and maintenance personnel provide feedback on the implementation effect of the recommended operation and maintenance measures and technical details, the system will collect this feedback information in real time and feed it back to the dynamic knowledge graph. These feedbacks will prompt the weights of relevant nodes and edges in the dynamic knowledge graph to be adjusted, further optimizing the accuracy of the recommendation.

[0037] In a specific embodiment, the fault diagnosis results, recommended supervision details and maintenance plans of power grid equipment are displayed through a graphical interface to help operation and maintenance personnel understand and operate, thereby improving the user experience of the system. In addition, a performance evaluation module is introduced to regularly evaluate the effects of the diagnosis module, recommendation details module and feedback learning module, optimize system performance based on the evaluation results, and ensure the continuous and efficient operation of the system.

[0038] In summary, the method of the present application uses a knowledge graph established using power grid equipment fault text data, which includes nodes such as equipment type and fault type, and displays associations through edges. At the same time, automated reasoning and evolutionary algorithms are used to achieve continuous updating of the graph. The fault self-diagnosis module uses NLP technology to extract features from power grid equipment fault text, and implements fault type diagnosis through text classification, deep learning and reasoning algorithms. The diagnosis accuracy is continuously optimized in combination with operation and maintenance feedback. According to the fault diagnosis results, the knowledge graph causal reasoning is used to automatically generate relevant regulatory details, which can realize intelligent diagnosis of power grid equipment faults and generate technical supervision opinions and operation and maintenance measures, thereby improving the operation and maintenance personnel's understanding of power grid equipment failures and operational efficiency, and realizing intelligent and efficient power grid equipment management.

[0039] See also Figure 2 , which shows a structural block diagram of a power grid equipment fault details generation system based on a dynamic knowledge graph of the present application.

[0040] like Figure 2 As shown, the power grid equipment fault details generating system 200 includes a building module 210 , an identification module 220 and a generating module 230 .

[0041] Among them, the construction module 210 is configured to obtain the power grid equipment fault text, and construct a dynamic knowledge graph of the power grid equipment based on the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, regulatory details and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between nodes; the identification module 220 is configured to extract key features from the power grid equipment fault text, use a fault diagnosis algorithm to identify the key features, and obtain a fault type identification result; the generation module 230 is configured to use the dynamic knowledge graph to generate regulatory details corresponding to the fault type identification result according to a preset causal reasoning mechanism, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

[0042] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0043] In other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the method for generating power grid equipment fault details based on a dynamic knowledge graph in any of the above method embodiments;

[0044] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0045] Obtaining a power grid equipment fault text, and constructing a dynamic knowledge graph of the power grid equipment according to the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, regulatory details, and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between the nodes;

[0046] Extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result;

[0047] According to a preset causal reasoning mechanism, the dynamic knowledge graph is used to generate regulatory details corresponding to the fault type identification result, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

[0048] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of a power grid equipment fault details generation system based on a dynamic knowledge graph, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the power grid equipment fault details generation system based on a dynamic knowledge graph via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0049] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the connection through the bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the method for generating power grid equipment fault details based on the dynamic knowledge graph in the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the power grid equipment fault details generation system based on the dynamic knowledge graph. The output device 340 may include display devices such as display screens.

[0050] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0051] As an implementation mode, the electronic device is applied to a power grid equipment fault detail generation system based on a dynamic knowledge graph, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:

[0052] Obtaining a power grid equipment fault text, and constructing a dynamic knowledge graph of the power grid equipment according to the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, regulatory details, and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between the nodes;

[0053] Extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result;

[0054] According to a preset causal reasoning mechanism, the dynamic knowledge graph is used to generate regulatory details corresponding to the fault type identification result, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

[0055] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating power grid equipment fault details based on dynamic knowledge graph, characterized in that: include: Obtaining a power grid equipment fault text, and constructing a dynamic knowledge graph of the power grid equipment according to the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, regulatory details, and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between the nodes; Extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result; According to a preset causal reasoning mechanism, the dynamic knowledge graph is used to generate regulatory details corresponding to the fault type identification result, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

2. A method for generating power grid equipment fault details based on dynamic knowledge graph according to claim 1, characterized in that: in, The edges in the dynamic knowledge graph include "equipment-failure" edges, "equipment-component" edges, "component-failure" edges, "failure-repair" edges, and "failure-regulatory standards" edges.

3. A method for generating power grid equipment fault details based on dynamic knowledge graph according to claim 1, characterized in that: After constructing a dynamic knowledge graph of power grid equipment according to the power grid equipment fault text, the method further includes: The preset graph neural network propagates information between devices, automatically learns the potential causal relationship between devices based on the equipment failure history and maintenance data, and adjusts the node and edge weights in the dynamic knowledge graph through back propagation; With the addition of new equipment data, maintenance records and fault reports, an adaptive adjustment algorithm is introduced to automatically adjust the nodes and edges in the dynamic knowledge graph according to the algorithm rules to obtain an updated and evolved dynamic knowledge graph.

4. A method for generating power grid equipment fault details based on dynamic knowledge graph according to claim 1, characterized in that: The extracting key features from the power grid equipment fault text, identifying the key features using a fault diagnosis algorithm, and obtaining a fault type identification result includes: Preprocessing the power grid equipment fault text based on a segmentation model of deep learning to obtain at least one segmented text; A pre-trained language model based on the Transformer architecture converts each word in at least one segment text into a context-dependent high-dimensional embedding vector, and extracts fine-grained features related to the fault description by combining global and local semantic information; The fine-grained features in the at least one segmented text are identified according to the text convolutional neural network model to obtain the component type, equipment fault type and fault severity.

5. A method for generating power grid equipment fault details based on dynamic knowledge graph according to claim 4, characterized in that: The method of using the dynamic knowledge graph to generate regulatory details corresponding to the fault type identification result according to the preset causal reasoning mechanism includes: By analyzing the importance of equipment component types, equipment failure types, and failure severity, reasoning is performed based on the nodes and edges in the dynamic knowledge graph to generate technical details and operation and maintenance measures that are suitable for the current equipment failure status.

6. A power grid equipment fault details generation system based on dynamic knowledge graph, characterized in that: include: A construction module is configured to obtain a power grid equipment fault text, and construct a dynamic knowledge graph of the power grid equipment according to the power grid equipment fault text, wherein the nodes in the dynamic knowledge graph include equipment type, fault type, supervision details, and operation and maintenance measures, and the edges in the dynamic knowledge graph represent the association relationship between the nodes; An identification module is configured to extract key features from the power grid equipment fault text, identify the key features using a fault diagnosis algorithm, and obtain a fault type identification result; A generation module is configured to generate regulatory details corresponding to the fault type identification result using the dynamic knowledge graph according to a preset causal reasoning mechanism, and the regulatory details include operating procedures, safety standards and operation and maintenance plans.

7. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images