Fault risk perception-oriented knowledge graph construction method and system, storage medium and processor

By constructing a knowledge graph oriented towards fault risk perception, the limitations of relying on expert experience in traditional fault handling methods are overcome, enabling rapid fault diagnosis and repair, and improving equipment availability and maintenance efficiency.

CN118917396APending Publication Date: 2024-11-08GUANGXI POWER GRID CO LTD NANNING POWER SUPPLY BUREAU
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411036348.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Traditional fault handling methods rely on expert experience, which makes it difficult to fully cover complex and dynamic fault situations. Furthermore, the vast and diverse operational data makes it difficult to extract useful information, thus limiting the ability to perceive fault risks and make decisions.

Method used

A knowledge graph for fault risk perception is constructed. Through data collection, preprocessing, entity and relation extraction, a triple model is established. Real-time fault information and historical data are combined to provide decision support, and the knowledge graph is continuously updated.

Benefits of technology

It enables rapid fault diagnosis and repair, improves the reliability of decision-making and equipment availability, and optimizes maintenance efficiency and cost control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917396A_ABST
    Figure CN118917396A_ABST
Patent Text Reader

Abstract

The invention discloses a knowledge graph construction method and system oriented to fault risk perception, a storage medium and a processor, and relates to the technical field of graph construction, and the method comprises the steps: data collection and integration: collecting power infrastructure data and operation and maintenance logs, and extracting useful information in the risk perception direction for preprocessing operation; knowledge extraction and relationship modeling: extracting entities and relationships from the preprocessed data, and constructing relationships between the entities; constructing a knowledge graph of fault risk perception based on the triple, and storing the knowledge graph in a graph database; and in combination with real-time fault information and historical data, decision support and risk perception are provided for a current fault, and a new fault is input into the knowledge graph as training data so as to update a more comprehensive knowledge query graph. According to the method, the equipment fault knowledge graph is constructed and maintained, so that the perception, prediction and decision support of the equipment fault risk are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of graph construction technology, and in particular to a knowledge graph construction method and system for fault risk perception, a storage medium, and a processor. Background Art

[0002] In modern power systems, the normal operation of various power infrastructure and equipment plays a vital role in ensuring the continuity of people's lives and production. However, equipment failures often lead to production stagnation, service interruptions, and safety risks, causing huge economic losses and electricity difficulties for enterprises and people's livelihood. Therefore, effectively perceiving and managing the failure risks of power equipment and systems has become a research topic that needs to be solved urgently.

[0003] Traditional fault handling methods are mainly based on expert experience and system rules. Power operation and maintenance personnel generally rely on their own experience and professional knowledge to judge and handle faults. This limitation based on data and historical experience makes fault judgment and handling susceptible to subjective factors. Faults are often complex and dynamic processes. Single experience and rules cannot fully cover various fault situations. Secondly, operation and maintenance data is huge and diverse, making it difficult to directly extract useful information, which limits the ability to perceive fault risks and make decisions.

[0004] In view of this, a knowledge graph construction method and system for fault risk perception, a storage medium and a processor are needed. Summary of the invention

[0005] The purpose of this invention is to address the above problems and propose a knowledge graph construction method and system for fault risk perception. By constructing and maintaining an equipment fault knowledge graph, the perception, prediction and decision support of equipment fault risks can be achieved. The specific technical solution is as follows:

[0006] A knowledge graph construction method for fault risk perception includes:

[0007] Data collection and integration: Collect power infrastructure data and operation and maintenance logs, and extract useful information in the direction of risk perception for preprocessing operations;

[0008] Knowledge extraction and relationship modeling: extract entities and relationships from preprocessed data and build connections between entities;

[0009] Build a knowledge graph of fault risk perception based on triples and store it in a graph database;

[0010] Combining real-time fault information and historical data, it provides decision support and risk perception for current faults, and inputs new faults into the knowledge graph as training data to update a more comprehensive knowledge query graph.

[0011] Preferably, the power infrastructure data includes installation, configuration and connection information related to power system facilities; the operation and maintenance log includes normal data and fault event data.

[0012] Preferably, the preprocessing operation includes deduplication, error correction, gap filling, word segmentation, tagging and format standardization of the data.

[0013] Preferably, extracting entities and relationships from preprocessed data and building relationships between entities includes:

[0014] The processed data is distributed in the ratio of training set: test set: validation set = 7:2:1, and expert experience is introduced into the training set for joint extraction of entity-relationships.

[0015] Preferably, different attention mechanisms are set during relation extraction.

[0016] Preferably, the knowledge graph for building fault risk awareness based on triples further includes:

[0017] Identify specific entities in infrastructure and operations based on the characteristics of knowledge;

[0018] At the same time, for each identified entity, the association relationship between the entity and other entities in the text is extracted through keywords;

[0019] Extract in the form of head entity-relationship-tail entity triples according to pre-defined extraction rules;

[0020] Entities and relationships are used as nodes and edges of the knowledge graph, and the connection relationship between different entities in the power failure risk perception system is established based on the TransE model;

[0021] The cosine similarity function and scoring function are used to evaluate and optimize the trained knowledge graph model.

[0022] A knowledge graph construction system for fault risk perception, applied to the method, comprising a data acquisition module, a data preprocessing module, a model construction module and a graph generation module;

[0023] The data acquisition module is used to collect power infrastructure data and operation and maintenance log data;

[0024] The preprocessing module is used to clean, integrate, annotate and convert power infrastructure data and operation and maintenance log data;

[0025] The model building module is used to build a knowledge graph relationship model for fault risk perception;

[0026] The map generation module is used to generate a risk perception map for the power system valley.

[0027] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for constructing a knowledge graph for fault risk perception.

[0028] A processor is used to run a program, wherein the program executes the knowledge graph construction method for fault risk perception when running.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. The present invention realizes decision support and risk perception of faults through comparative analysis of real-time fault data and historical data, accelerates the diagnosis and repair process of faults, reduces the impact of faults on production and operations, and improves the reliability of decision-making and the availability of equipment;

[0031] 2. The present invention calculates the correlation scores between entities based on expert experience and attention mechanism, screens out more matching relationship-entities, and provides more reliable fault perception data input for subsequent model training;

[0032] 3. The present invention combines actual application and feedback data, and the system continuously updates and maintains the knowledge graph to improve the accuracy and adaptability of the system; at the same time, according to the analysis of risk perception and historical data, the maintenance plan is reasonably arranged to provide data support for equipment maintenance and resource allocation plans, thereby improving maintenance efficiency and cost control. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0034] Figure 1 It is a logical architecture diagram of the present invention;

[0035] Figure 2 It is a schematic diagram of the structure of the power system fault knowledge graph of the present invention.

[0036] Figure 3 This is a system framework diagram of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0038] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0039] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0040] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0041] Example 1

[0042] See also Figure 1-2 This embodiment discloses a method for constructing a knowledge graph for fault risk perception, which specifically includes the following steps:

[0043] Step 1: Data collection and integration: Based on power infrastructure data and operation and maintenance logs, extract useful information in the direction of risk perception for preprocessing operations to improve the accuracy and consistency of knowledge;

[0044] Specifically, in the data collection module in step 1, the data sources are mainly installation, configuration and connection information related to the power system infrastructure, including design documents, engineering drawings, equipment specifications and delivery reports, as well as normal data and fault event data in operation and maintenance logs, including equipment status changes, alarm events, maintenance reports and repair records;

[0045] Specifically, in step 1, the data integration module cleans and preprocesses the collected power infrastructure data and operation and maintenance logs, including deduplication, error correction, gap filling, word segmentation, labeling, and format standardization of the data to ensure the accuracy and effectiveness of data use.

[0046] Step 2: Knowledge extraction and relationship modeling: extract entities and relationships from preprocessed data and build connections between entities;

[0047] Specifically, the system collects and preprocesses data, distributes the processed data in the ratio of training set: test set: validation set = 7:2:1, and introduces expert experience into the training set for joint extraction of entity-relationships;

[0048] Step 3: Construct a knowledge graph for fault risk perception based on triples and store it in a graph database to enhance knowledge storage and query capabilities.

[0049] Specifically, building a fault-aware knowledge graph includes the following two parts:

[0050] S1: Knowledge extraction is the process of extracting useful information from preprocessed data. Named entities, as the primary task of natural language processing, aim to identify specific entities in infrastructure and operation and maintenance based on the characteristics of knowledge, usually fault manifestations, properties, levels, etc. At the same time, for each identified entity, the relationship between the entity and other entities (1 or more) is extracted in the text through keywords. Among them, the relationship includes both the physical connection relationship between infrastructure facilities, such as transformers and transmission lines; and the implicit relationship between operating status, such as the relationship between transformer alarm events and maintenance records;

[0051] Assume that the entity set is S = {s1,s2,...,s i ,...,s n},s i represents the i-th entity, where each entity has multiple relationships, namely r i ={r1,r2,...,r j ,...,r m}, r j Represents entity r i The jth relationship;

[0052] S2: The relationship between entities is extracted in the form of triples (head entity-relationship-tail entity) according to predefined extraction rules. The entities in the data are represented by vectors. Each entity obtains a set of relationship scores by calculating the relationship vector between it and other entities. Based on these scores, the relationship with high confidence is selected as the extracted triple. At the same time, considering the inconsistency of the relationship weights between different entities, different attention mechanisms are set when extracting relationships;

[0053] Assume that the entity vector is E Si ={E1,E2,...,E m}, the attention mechanism is ω i ={ω1,ω2,...,ω m}.

[0054] In addition, in step 3, entities and relationships are used as nodes and edges of the knowledge graph. Based on the TransE model, the connection relationship between different entities in the power failure risk perception system is established. The cosine similarity function and scoring function are used to evaluate and optimize the trained knowledge graph model, thereby improving the effect and generalization ability of the model:

[0055]

[0056]

[0057] The value range of the cosine similarity function is [-1, 1]. When the value is closer to 1 (the better), it means that the two entities are very similar in semantics. When the value is closer to 0 or -1, it means that the two entities are completely different in semantics.

[0058] During the training process, the goal of the model is to make the score function value of the true triple as low as possible. In theory, the value is [0,∞). When the value of f is smaller (better), it means that h+r is closer to t, then the triple is more likely to be correct. Conversely, when the value of f is larger, it means that h+r is farther away from t, then the triple is more likely to be wrong.

[0059] As new data accumulates in practical applications, new knowledge is added to the existing knowledge graph, and the model is continuously optimized through entity recognition and relationship extraction to keep the knowledge graph updated and complete.

[0060] Step 4: Combine real-time fault information and historical data to provide decision support and risk perception for current faults, and input new faults into the knowledge graph as training data to update a more comprehensive knowledge query graph.

[0061] Specifically, by combining real-time fault information and historical data for risk perception and predictive analysis, and providing historical solutions and potential risk predictions, it helps operation and maintenance personnel to identify and handle faults in a timely manner, reducing downtime and production losses. At the same time, the fault perception knowledge graph provides more comprehensive fault analysis and decision support, optimizes equipment maintenance plans and resource allocation, and improves equipment reliability and operating efficiency.

[0062] S1: Input real-time information into the trained model to perform real-time fault risk perception and prediction analysis. The model gives the risk level and potential threat of the fault based on the characteristics of the real-time data, and performs similar fault matching based on the fault code, equipment status, etc. according to the fault information and corresponding solutions in the historical data, and provides relevant historical solutions for the current fault. At the same time, a feedback mechanism is established to feed back the fault handling results and solution usage to the system in order to further improve the performance of the model and provide more accurate risk analysis and solution support.

[0063] S2: Based on the fault prediction model, by comparing the similarity between the current fault characteristics and the historical fault characteristics, the probability of potential risks of the current fault is analyzed and identified, and warnings and suggestions on possible additional faults or abnormal conditions are provided. As fault data accumulates, the fault perception model is continuously iterated and updated to improve the accuracy and effectiveness of model perception.

[0064] Example 2

[0065] like Figure 3 As shown, a knowledge graph construction system for fault risk perception is applied to the method described above, including a data acquisition module, a data preprocessing module, a model construction module and a graph generation module;

[0066] The data acquisition module is used to collect power infrastructure data and operation and maintenance log data;

[0067] The preprocessing module is used to clean, integrate, annotate and convert power infrastructure data and operation and maintenance log data;

[0068] The model building module is used to build a knowledge graph relationship model for fault risk perception;

[0069] The map generation module is used to generate a risk perception map for the power system valley.

[0070] Example 3

[0071] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for constructing a knowledge graph for fault risk perception.

[0072] Example 4

[0073] A processor is used to run a program, wherein the program executes the knowledge graph construction method for fault risk perception when running.

[0074] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0075] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0076] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0077] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0078] 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 replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A knowledge graph construction method for fault risk perception, characterized in that: include: Data collection and integration: Collect power infrastructure data and operation and maintenance logs, and extract useful information in the direction of risk perception for preprocessing operations; Knowledge extraction and relationship modeling: extract entities and relationships from preprocessed data and build connections between entities; Build a knowledge graph of fault risk perception based on triples and store it in a graph database; Combining real-time fault information and historical data, it provides decision support and risk perception for current faults, and inputs new faults into the knowledge graph as training data to update a more comprehensive knowledge query graph.

2. The method for constructing a knowledge graph for fault risk perception according to claim 1, characterized in that: The power infrastructure data includes installation, configuration and connection information related to power system facilities; the operation and maintenance log includes normal data and fault event data.

3. The method for constructing a knowledge graph for fault risk perception according to claim 2, characterized in that: The preprocessing operations include data deduplication, error correction, missing fill, word segmentation, tagging and format standardization.

4. The method for constructing a knowledge graph for fault risk perception according to claim 1, characterized in that: The extraction of entities and relations from the preprocessed data and the construction of connections between entities include: The processed data is distributed in the ratio of training set: test set: validation set = 7:2:1, and expert experience is introduced into the training set for joint extraction of entity-relationships.

5. The method for constructing a knowledge graph for fault risk perception according to claim 4 is characterized in that: It also includes setting different attention mechanisms when extracting relations.

6. The method for constructing a knowledge graph for fault risk perception according to claim 4, characterized in that: The knowledge graph for fault risk perception based on triples also includes: Identify specific entities in infrastructure and operations based on the characteristics of knowledge; At the same time, for each identified entity, the association relationship between the entity and other entities in the text is extracted through keywords; Extract in the form of head entity-relationship-tail entity triples according to pre-defined extraction rules; Entities and relationships are used as nodes and edges of the knowledge graph, and the connection relationship between different entities in the power failure risk perception system is established based on the TransE model; The cosine similarity function and scoring function are used to evaluate and optimize the trained knowledge graph model.

7. A knowledge graph construction system for fault risk perception, characterized in that: The method applied to any one of claims 1 to 6, comprising a data acquisition module, a data preprocessing module, a model building module and a graph generation module; The data acquisition module is used to collect power infrastructure data and operation and maintenance log data; The preprocessing module is used to clean, integrate, annotate and convert power infrastructure data and operation and maintenance log data; the model building module is used to build a knowledge graph graph relationship model for fault risk perception; The map generation module is used to generate a risk perception map for the power system valley.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the knowledge graph construction method for fault risk perception according to any one of claims 1 to 6.

9. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the knowledge graph construction method for fault risk perception described in any one of claims 1 to 6.

Citation Information

Cited By

  • Security intelligent decision optimization method based on knowledge graph model

    CN120653784A

  • Personal risk database construction system and method

    CN121456039A

  • Road electromechanical equipment fault knowledge graph construction and dynamic updating method based on heterogeneous multi-source data

    CN121920490A