CN114792140A - Transformer substation defect analysis system based on knowledge graph - Google Patents
Transformer substation defect analysis system based on knowledge graph Download PDFInfo
- Publication number
- CN114792140A CN114792140A CN202210326061.1A CN202210326061A CN114792140A CN 114792140 A CN114792140 A CN 114792140A CN 202210326061 A CN202210326061 A CN 202210326061A CN 114792140 A CN114792140 A CN 114792140A
- Authority
- CN
- China
- Prior art keywords
- defect
- knowledge
- module
- information
- graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 342
- 238000004458 analytical method Methods 0.000 title claims abstract description 65
- 238000010276 construction Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims description 40
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 11
- 238000012423 maintenance Methods 0.000 claims description 11
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 238000003745 diagnosis Methods 0.000 claims description 7
- 230000008676 import Effects 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 description 6
- 238000012217 deletion Methods 0.000 description 5
- 230000037430 deletion Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000011218 segmentation Effects 0.000 description 5
- 238000009825 accumulation Methods 0.000 description 4
- 238000012850 discrimination method Methods 0.000 description 4
- 230000002708 enhancing effect Effects 0.000 description 4
- 230000002950 deficient Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000001376 precipitating effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013479 data entry Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a transformer substation equipment defect analysis system based on a knowledge graph, which comprises a defect registration module, a case management module, a knowledge acquisition module, a knowledge graph construction module and an intelligent reasoning module, wherein the defect registration module, the case management module, the knowledge acquisition module, the knowledge graph construction module and the intelligent reasoning module are sequentially transmitted by signals, and the system comprises: the defect registration module is used for registering defect information; the case management module is used for managing historical defect case information; the knowledge acquisition module is used for deconstructing the defect cases into structured data and acquiring effective knowledge; the knowledge graph building module is used for building an equipment defect characteristic knowledge graph; the intelligent reasoning module is used for matching and reasoning the current defect characteristic information with the existing defect knowledge and expert experience in the knowledge base to find out the defect reason; by means of the knowledge map building module, algorithm processing efficiency is improved, precision of the inference algorithm based on the intelligent inference module is greatly improved, and defects of equipment can be accurately positioned in practice, so that equipment operation quality is improved.
Description
Technical Field
The invention relates to the field of transformer substation equipment defect analysis, transformer substation equipment defect management and natural language processing, in particular to a transformer substation equipment defect analysis system based on a knowledge graph.
Background
With the development of science and technology, the defect analysis of substation equipment becomes more and more important, the current common method is a universal question-answering system, a large amount of corpus is generally used as a basis to realize short dialogue question-answering, the system is often used in the entertainment field and less in the equipment defect maintenance field, one reason is that the system is often difficult to extract the accurate meaning in the user question, and the intention of the user cannot be accurately analyzed; the traditional industrial equipment defect analysis system has certain defects in the aspects of knowledge representation, knowledge storage and knowledge reasoning; secondly, the industrial fault maintenance association field is wide, and the question-answering system can only inquire key words of question sentences in a traditional knowledge base according to questions provided by users, so that the positions of answers are positioned.
Therefore, in the process of implementing the invention, the inventor finds that a transformer substation equipment defect analysis system based on the knowledge graph needs to be provided, adopts an expert knowledge representation and knowledge storage method based on the knowledge graph, combines an intelligent reasoning technology to accurately diagnose and analyze the equipment defect condition, and ensures the stability of equipment operation.
Disclosure of Invention
In order to overcome the defects and shortcomings in the prior art, the invention provides a transformer substation equipment defect analysis system based on knowledge graphs, which can continuously enrich the knowledge graphs through diagnosis accumulation, defect registration, case management and knowledge extraction, thereby being capable of adapting to continuously improved user requirements.
The technical scheme of the invention is as follows:
the utility model provides a transformer substation equipment defect analysis system based on knowledge graph, transformer substation equipment defect analysis system includes defect registration module, case management module, knowledge acquisition module, knowledge graph construction module and the intelligent reasoning module that the signal transmitted in proper order, wherein:
the defect registration module is used for registering defect information;
the case management module is used for managing historical defect case information;
the knowledge acquisition module is used for deconstructing the defect cases into structured data and acquiring effective knowledge;
the knowledge graph building module is used for building an equipment defect characteristic knowledge graph;
the intelligent reasoning module is used for matching and reasoning the current defect characteristic information with the defect knowledge and the expert experience in the knowledge base to find out the defect reason.
Therefore, the substation equipment defect analysis system based on the knowledge graph of the embodiment of the invention generates effective knowledge by automatically, quickly and accurately deconstructing the defect cases, thereby precipitating expert knowledge in the accumulation process of the defect cases, enhancing the expression capability of the expert knowledge by constructing the substation equipment defect analysis knowledge graph, improving the processing speed, reducing the redundancy of the knowledge graph through the knowledge graph construction module, enhancing the knowledge correlation in the graph, improving the analysis processing efficiency, greatly improving the precision by means of the intelligent reasoning module, accurately positioning the equipment defects in practical application, and improving the equipment operation quality. Furthermore, a defect analysis system with better user experience can be provided, a professional and mature substation equipment fault maintenance scheme is provided for a user, and stable operation of equipment is guaranteed.
Preferably, the registered defect information includes device information for generating a defect, data for causing a defect, and phenomenon characteristic information, a defect cause, and defect type information.
Preferably, the specific implementation manner of obtaining the defect case according to the registered defect information is as follows: and (3) inputting the defect information by an operation and maintenance worker in a manual inputting mode, displaying the defect information in a defect registration list after inputting, clicking a generated case, generating document case information from the defect registration information by a program and using a template engine Freemarker technology, starting a case auditing link, formally generating a defect case from the defect registration information after auditing is passed, and releasing the defect case to the line.
Preferably, the knowledge acquisition module acquires the knowledge by the following three ways:
analyzing and cleaning expert knowledge maintained in an Excel file by an Excel import mode to generate effective knowledge; manually maintaining expert knowledge through a system page; and generating effective knowledge by deconstructing the defect case.
Preferably, the method for deconstructing the defect case comprises the following steps:
s31, splitting the content of the defect case and cleaning the text;
s32, deconstructing defect content, and acquiring all object attributes of the defect content;
and S33, storing the content of the defect case and the corresponding object attribute into the corresponding data table entry according to the obtained object attribute.
Therefore, according to the transformer substation equipment defect analysis system based on the knowledge graph, disclosed by the embodiment of the invention, the defect cases can be analyzed without manual participation, all object attribute information in the defect cases can be automatically and accurately extracted, the information after deconstruction is conveniently inquired and understood by people, the whole process is automatically completed without manual intervention, and the system is high in accuracy of automatic deconstruction data.
Preferably, the object attribute of the defect content obtained in step S32 includes a transformer substation name, an equipment voltage level, an equipment type, an equipment structure type, defect component information, a defect type, defect cause information, a determination method, defect description information, a determination technology standard, a defect elimination measure, defect finder information, and defect handler information corresponding to the defect case.
Preferably, the transformer substation name is obtained by identifying the content of the title in a regular expression mode, and determining the transformer substation name according to the result of regular identification; the deletion measures are obtained by traversing the whole identified content by utilizing the prestored deletion trigger words, inquiring the content related to the deletion measures in the text content and extracting the part of content.
Preferably, the subsystem for deconstructing defect cases comprises the following three modules:
a, a defect content splitting module, wherein the defect content splitting module splits the defect content and splits the defect content into the following parts according to the document format of the defect case: title, discovery processing and equipment conditions are obtained, and the split content is cleaned and then transmitted to an object attribute identification module;
b, an object attribute identification module, wherein the object attribute identification module identifies the content of the defect document, acquires all object attribute information corresponding to the content of the defect, and transmits the object information to a corresponding storage module;
and c, a deconstruction result storage module, wherein the deconstruction result storage module forms a corresponding mapping relation for all the object attribute information acquired by the object attribute identification model and stores the mapping relation in the database through corresponding data table items.
Preferably, the knowledge graph construction module analyzes and constructs the reconstructed structured data by using a graph calculation engine technology to realize the analysis and construction processing of the relevant knowledge graph in the field of substation equipment defect analysis, abstracts the relation between each type of equipment, defect characteristic data and information of defect reasons, defect types and solutions into a graph, combines the semi-structured characteristics of the graph data, adopts a graph vertex as a center and is based on a graph calculation engine to perform operation optimization, and establishes the equipment defect characteristic knowledge graph.
Preferably, the intelligent reasoning module performs matching reasoning on the current defect feature information, the existing defect knowledge in the knowledge base and the expert experience according to the knowledge map constructed by the knowledge map construction module, so as to find out the defect reason.
Preferably, the process of matching inference mainly comprises the following steps:
s51, selecting the structure type of the substation equipment;
s52, inputting substation equipment information and carrying out familial defect analysis;
s53, selecting a characteristic part, inquiring a knowledge graph according to the selected characteristic part, finding out a judgment method and a technical standard which are indirectly related to the characteristic part, and pushing the judgment method;
s54, selecting a judgment method, inquiring data and phenomenon characteristics which have indirect relation with the judgment method in the knowledge graph, and displaying;
s55, the user correspondingly selects the fault equipment in the transformer substation field according to the characteristics shown by the fault equipment in the transformer substation field in the S54 result;
s56, through the data and the phenomenon characteristics selected in the step S55, the data and the phenomenon characteristics corresponding to the defect reasons in the knowledge map are hit, if the hit rate is satisfied, the defect reason with the highest hit rate is output as a matching result, and if the hit rate is not satisfied, the step S57 is continued;
and S57, pushing the related features from top to bottom according to the hit rate sequence of the defect reasons in the step S56 until the matching degree meets the requirement, outputting a diagnosis result, and ending the reasoning process, or outputting a prompt of an operation suggestion of the next step until the related features are pushed completely, and ending the reasoning process.
Preferably, the logic for performing the familial defect analysis calculation in step S52 includes the following steps:
s521, the case management module maintains the inspected defect cases;
s522, each approved defect case has a system label;
s523, matching the system labels of all cases according to the input substation equipment information, and counting the times of the same equipment manufacturer, the same equipment model, the same defect part and the same defect type;
s524, determining whether the familial defect rule is satisfied, outputting a result of the familial defect analysis if the familial defect rule is satisfied, and continuing to execute the step S53 if the familial defect rule is not satisfied.
The invention has the beneficial effects that:
the invention discloses a transformer substation equipment defect analysis system based on a knowledge graph, which generates effective knowledge by automatically, quickly and accurately deconstructing defect cases, thereby precipitating expert knowledge in the accumulation process of the defect cases, constructing a transformer substation equipment defect analysis knowledge graph by a graph calculation engine, enhancing the expression capability of the expert knowledge, improving the processing speed, reducing the redundancy of the knowledge graph by a knowledge graph construction module, enhancing the knowledge relevance in the graph, improving the algorithm processing efficiency, greatly improving the precision of an inference algorithm based on an intelligent inference module, and accurately positioning equipment defects in practice, thereby improving the equipment operation quality; particularly, by combining the extraction template, the semantic rule and the regular expression and adopting a mode of assisting various resource word banks, the object attribute information in the defect case can be extracted under the condition of no labeled data or a small amount of labeled data, all the object attribute information in the defect case can be automatically and accurately extracted without manual participation in analyzing the defect case, the information which is convenient for people to inquire and know after deconstruction is formed, the whole process is automatically completed without manual intervention, the accuracy of the system for automatically deconstructing the data is high, and the accuracy and the precision of the system are further improved.
Drawings
The invention is described in further detail below with reference to the drawings and the detailed description.
Fig. 1 is a diagram of a knowledge graph structure and feature selection step established after knowledge extraction in step S4 and step S5 in an embodiment of a transformer substation equipment defect analysis system based on a knowledge graph of the present invention;
fig. 2 is a flowchart illustrating the familial defect analysis in step S5 in an embodiment of the system for analyzing substation equipment defects based on knowledge-graph according to the present invention;
fig. 3 is a diagram of a data feature matching logic structure in step S5 in an embodiment of a transformer substation equipment defect analysis system based on a knowledge graph according to the present invention;
FIG. 4 is a flow chart of the operation of a transformer substation equipment defect analysis system based on a knowledge graph according to the present invention;
FIG. 5 is a flowchart illustrating a substation equipment defect case deconstruction scheme of a substation equipment defect analysis system according to the present invention;
fig. 6 is a schematic diagram of the overall architecture of a substation equipment defect analysis system based on a knowledge graph according to an embodiment of the present invention;
fig. 7 is a schematic overall flow chart of a method for analyzing defects of substation equipment based on a knowledge graph according to an embodiment of the present invention;
fig. 8 is a schematic overall flow chart of the defect case deconstruction in step S3 of the method for analyzing the defects of the substation equipment based on the knowledge graph according to the embodiment of the present invention;
fig. 9 is a schematic overall flow chart of matching inference in step S5 in the method for analyzing substation equipment defects based on knowledge graph according to an embodiment of the present invention;
fig. 10 is a schematic overall flowchart of the familial defect analysis calculation performed in step S52 of the method for analyzing the defects of substation equipment based on a knowledge graph according to an embodiment of the present invention;
fig. 11 is a schematic main flowchart of the defect case deconstruction method in step S3 in the embodiment of the system for analyzing substation equipment defects based on knowledge graph according to the present invention.
Detailed Description
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
In the following description, a detailed structure will be presented for a thorough understanding of embodiments of the invention. It is apparent that the implementation of the embodiments of the present invention is not limited to the specific details familiar to those skilled in the art. The following detailed description of the preferred embodiments of the invention, however, the invention can be practiced otherwise than as specifically described.
Referring to fig. 1 to 5, fig. 1 is a diagram illustrating a structure of a knowledge graph and a feature selection step established after knowledge extraction in step (4) and step (5) in an embodiment of a transformer substation equipment defect analysis system based on a knowledge graph according to the present invention; FIG. 2 is a flowchart of the familial defect analysis in step (5) of an embodiment of the present invention of a knowledge-graph-based substation equipment defect analysis system; fig. 3 is a diagram of a data feature matching logic in step (5) of an embodiment of a system for analyzing substation equipment defects based on a knowledge graph according to the present invention; FIG. 4 is a flow chart of the operation of a knowledge-graph-based substation equipment defect analysis system of the present invention; FIG. 5 is a flowchart of a substation equipment defect case deconstruction scheme of a substation equipment defect analysis system based on a knowledge graph of the present invention; the transformer substation equipment defect analysis system based on the knowledge graph provided by the embodiment comprises a defect registration module, a case management module, a knowledge acquisition module, a knowledge graph construction module and an intelligent reasoning module, wherein the defect registration module, the case management module, the knowledge acquisition module, the knowledge graph construction module and the intelligent reasoning module are sequentially transmitted through signals, and the transformer substation equipment defect analysis system comprises: the defect registration module is used for registering defect information; the case management module is used for managing historical defect case information; the knowledge acquisition module is used for deconstructing the defect cases into structured data and acquiring effective knowledge; the knowledge graph building module is used for building an equipment defect characteristic knowledge graph; the intelligent reasoning module is used for matching and reasoning the current defect characteristic information with the existing defect knowledge and expert experience in the knowledge base to find out the defect reason.
With reference to fig. 6 and 7, the specific operation of this embodiment includes the following steps:
step S1: and the defect registration module is used for recording expert experience knowledge related to defect diagnosis and maintenance in a manual recording mode. Effective defect maintenance related information is formed through cleaning, screening and feature extraction, registered defect generation defect cases are stored in a defect case management module through a system, specifically, the registered defect information comprises equipment information for generating defects, data caused by the defects, phenomenon feature information, defect reason and defect type information, all the information is input by operation and maintenance personnel in a manual input mode, after the input, the information is displayed in a defect registration list, then a generated case is clicked, a program is used, a template engine Freemarker technology is used for generating document case information from the defect registration information, a case auditing link is started, and after the auditing is passed, the defect registration information formally generates defect cases and releases the defect cases to the line;
step S2: in the defect case management module, the user can conveniently use the processing mode of historical defects for reference by introducing defect cases into the user and collecting and searching the cases generated by defect registration to process the existing similar defects; the method has the main significance that through managing historical defect information, mature expert knowledge can be conveniently used for reference and extracted, main source users of defect cases upload and register the defects, the users can collect the common defects or the defect cases corresponding to the defects with larger reference significance, the users can conveniently check the defects later, and when the users find new defects, the users can search the module through the found defect information so as to use historical solutions of the defects;
step S3: and in the knowledge acquisition module, expert experience knowledge related to defect diagnosis and analysis, such as staff system page maintenance, Excel import and defect case deconstruction, is received, and the comprehensive information is subjected to cleaning, screening and feature extraction to form effective defect feature related information. Specifically, the acquisition is performed in three ways: analyzing and cleaning expert knowledge maintained in an Excel file by an Excel import mode to generate effective knowledge; manually maintaining expert knowledge through a system page; effective knowledge is generated by automatically, quickly and accurately deconstructing the defect cases;
referring to fig. 8, the method for deconstructing the defect case includes the following steps: s31, splitting the content of the defect case, and cleaning the text; s32, deconstructing the defect content, and acquiring all object attributes of the defect content; s33, storing the defect case content and the corresponding object attribute into the corresponding data table entry according to the obtained object attribute; further, the object attributes obtained in step S32 include a substation name, an equipment voltage level, an equipment type, an equipment structure type, defect component information, a defect type, defect cause information, a discrimination method, defect description information, a discrimination technical standard, a defect elimination measure, defect finder information, defect handler information, and the like corresponding to the defect case, and when the substation name information is obtained in step S32, the content of the header is identified by using a regular expression, and the substation name is determined according to a result of the regular identification; when the defect eliminating measures are obtained, traversing the whole identified content by utilizing the prestored defect eliminating trigger words, inquiring the content of the defect eliminating measures in the text content and extracting the part of the content; when the defect case content is stored in step S33, the obtained discrimination method is first stored in the corresponding data entry, and then the defect description information corresponding to the discrimination method is stored in the entry.
The deconstruction scheme of the defect case comprises three modules: a. the defect content splitting module splits the defect content and splits the defect content into the following parts according to the document format of the defect case: title, discovery processing and equipment conditions are obtained, and the split content is cleaned and then transmitted to an object attribute identification module; b. the object attribute identification module identifies the content of the defective document, acquires all object attribute information corresponding to the defective content and transmits the object information to the corresponding storage module; c. the deconstruction result storage module forms a corresponding mapping relation for all the object attribute information acquired by the object attribute identification model and stores the mapping relation in a database through corresponding data table items; the defect case deconstruction scheme formed according to the scheme can automatically acquire object information in the defect cases to form information convenient for people to inquire and understand the deconstructed information. The whole process is automatically finished without manual intervention, and the system automatically deconstructs data with high accuracy.
In specific implementation, referring to fig. 11, the defect case deconstruction may include the following steps:
s311, training a neural network model aiming at the field of substation equipment, wherein the model is used for word segmentation, part of speech tagging and entity identification of texts.
S312, constructing a plurality of different word banks:
1) the defect content segmentation identification library is used for segmenting the text;
2) the triggering word bank of the defect characteristics is used for positioning the defect description;
3) a trigger word bank of negative words for screening non-defective descriptions;
4) the trigger word bank of the defect reasons is used for positioning the defect reasons;
5) the deletion triggering word stock is used for positioning deletion measures;
6) the part of speech extension library is used for forward extension of defect characteristics to complete defect description;
7) and the special name word library is used for correcting information analysis errors caused by the model word segmentation errors.
S313, segmenting the content by using the defect content segmentation identification library, segmenting the paragraphs, utilizing the sentences, segmenting the clauses by using the marks such as' and the like, and segmenting the clauses by using the trained model, performing word segmentation, part of speech tagging and entity recognition to obtain an analysis result of the clauses.
S321, according to the analysis result: if the result has the defect characteristic words, performing forward traversal expansion of defect description by using a set part-of-speech expansion library through part-of-speech judgment, and splicing words of the part-of-speech in the part-of-speech expansion library to the front of the defect characteristic words; if the part of speech of the previous word is not in the part of speech expansion library, the splicing is terminated, complete defect description is formed, and the formed defect description comprises the position where the defect occurs and the defect which occurs; if the result has a negative word, the defect description is filtered out, which indicates that the defect does not occur.
S322, according to the analysis result: and if the result has the defect reason characteristic words and no negative words, determining that the sentence description is the defect reason information.
S323, matching the sentence content by extracting the template, for example: the information of the defect discrimination method can be extracted through the' pass.. the detection.
And S331, forming a corresponding mapping relation of all the object attribute information acquired by the defect content deconstruction module, storing the mapping relation into a database through corresponding data table entries, and establishing the structured knowledge base.
Therefore, the object attribute information in the defect case can be extracted under the condition of no marking data or a small amount of marking data, all the object attribute information in the defect case can be automatically and accurately extracted without manual participation in analyzing the defect case, information which is convenient for people to inquire and know after deconstruction is formed, the whole process is automatically finished without manual intervention, and the system automatically deconstructs data with high accuracy.
Step S4, in a knowledge graph construction module, analyzing and constructing a related knowledge graph in the field of transformer substation equipment defect analysis on the structured data reconstructed in the step S3 by using a graph calculation engine technology, abstracting the relation of information such as various types of equipment, defect characteristic data, defect reasons, defect types, solutions and the like into a graph, combining the semi-structured characteristics of graph data, performing operation optimization by using a graph vertex as a center and based on a graph calculation engine, and constructing a related transformer substation equipment defect characteristic knowledge graph;
and step S5, matching and reasoning the current defect characteristic information with the existing defect knowledge and expert experience in the knowledge base in the intelligent reasoning module according to the knowledge graph constructed in the step S4 to find out the defect reasons. Firstly, selecting defect characteristics according to step prompts. And traversing and matching the stored defect knowledge in the knowledge graph and the input defect characteristics by adopting a matching algorithm for the selected defect characteristic information, and judging whether the matchable knowledge exists in the knowledge graph. If yes, judging whether conflict resolution is needed, eliminating conflict according to the predefined map knowledge matching priority level, outputting a defect analysis reasoning result, and finishing the reasoning process. And outputting information such as matched defects and defect solutions. And the defect fact is manually recorded into a defect registration module as a new defect.
Referring to fig. 9, in an embodiment of the present invention, the matching process is mainly divided into the following steps: s51, selecting the structure type of the substation equipment; s52, inputting substation equipment information, and performing familial defect analysis, wherein the familial defect analysis calculation logic comprises the following steps: s521, the case management module maintains the checked defect cases, S522, each checked defect case has a system label, S523, the system labels of all cases are matched according to the input substation equipment information, the times of the same equipment manufacturer, the same equipment model, the same defect part and the same defect type are counted, S524, whether a familial defect rule is met is judged, if yes, a familial defect analysis result is output, and if not, the following step S53 is continuously executed; selecting a characteristic part, inquiring a knowledge graph according to the selected characteristic part, finding out a judging method and a technical standard which have indirect relation with the characteristic part, and pushing the judging method; selecting a judgment method, inquiring data and phenomenon characteristics which have indirect relation with the judgment method in the knowledge graph, and displaying the data and the phenomenon characteristics; the user correspondingly selects the fault equipment in the transformer substation according to the characteristics shown by the field fault equipment in the transformer substation in the step S54; through the data and the phenomenon characteristics selected in the step S55, the data and the phenomenon characteristics corresponding to the defect reasons in the knowledge graph are hit, if the hit rate is satisfied, the defect reason with the highest hit rate is output as a matching result, and if the hit rate is not satisfied, the step S57 is continued; according to the step S56, the hit rate of the defect reasons is sorted, the related features are pushed from top to bottom until the matching degree meets the requirement, a diagnosis result is output, the reasoning process is ended, or the prompt of the operation suggestion of the next step is output until the related features are pushed completely (for example, if the matching degree of the defect reasons is low, the diagnosis cannot be made, the following work can be suggested to be further carried out, namely, a1, carrying out grounding current detection according to grounding current detection, a2, carrying out insulation test according to insulation test, and/or a3, carrying out appearance inspection according to transformer appearance inspection standard), and the reasoning process is ended. Outputting information such as matched defects, defect solutions and the like, and simultaneously taking the defect fact as a new defect to be manually input into a defect registration module for later use.
The transformer substation equipment defect analysis based on the knowledge graph is specifically to build the knowledge graph based on expert experience knowledge in the transformer substation equipment maintenance professional field, analyze equipment defect characteristics consulted by a user by using a graph calculation engine technology, a natural language processing technology and an intelligent reasoning technology, and give corresponding defect information and maintenance suggestions and the like by inquiring the knowledge graph.
The invention discloses a transformer substation equipment defect analysis system based on a knowledge graph, which is characterized in that effective knowledge is generated by automatically, quickly and accurately deconstructing defect cases, so that expert knowledge is precipitated in the accumulation process of the defect cases, a transformer substation equipment defect analysis knowledge graph is constructed by a graph calculation engine, the expression capability of the expert knowledge is enhanced, the processing speed is improved, the redundancy of the knowledge graph is reduced by a knowledge graph construction module, the knowledge correlation in the graph is enhanced, the algorithm processing efficiency is improved, the precision of an inference algorithm based on an intelligent inference module is greatly improved, the equipment defect can be accurately positioned in practice, and the equipment operation quality is improved.
Finally, it should be noted that: although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can make modifications and equivalents to the embodiments of the present invention without departing from the spirit and scope of the present invention, and such modifications and equivalents are within the scope of the appended claims.
Claims (10)
1. The utility model provides a transformer substation equipment defect analysis system based on knowledge graph, its characterized in that, transformer substation equipment defect analysis system includes defect registration module, case management module, knowledge acquisition module, knowledge graph construction module and the intelligent reasoning module that the signal transmitted in proper order, wherein:
the defect registration module is used for registering defect information;
the case management module is used for managing historical defect case information;
the knowledge acquisition module is used for deconstructing the defect cases into structured data and acquiring effective knowledge;
the knowledge graph building module is used for building an equipment defect characteristic knowledge graph;
the intelligent reasoning module is used for matching and reasoning the current defect characteristic information with the defect knowledge and the expert experience in the knowledge base to find out the defect reason.
2. The knowledge-graph-based substation equipment defect analysis system of claim 1, wherein the registered defect information comprises equipment information that generated a defect, data that was caused by the defect, and phenomenon signature information, defect cause, and defect type information.
3. The knowledge-graph-based substation equipment defect analysis system of claim 1, wherein the specific implementation manner of obtaining the defect case according to the registered defect information is as follows: and the operation and maintenance personnel input the defect information in a manual input mode, after the input, the defect information is displayed in a defect registration list, then a generated case is clicked, the defect registration information is generated into document case information, a case auditing link is started, and after the auditing is passed, the defect registration information formally generates a defect case and releases the defect case to the line.
4. The knowledge-graph-based substation equipment defect analysis system of claim 1, wherein the knowledge acquisition module acquires by three means:
analyzing and cleaning expert knowledge maintained in an Excel file by an Excel import mode to generate effective knowledge; manually maintaining expert knowledge through a system page; and/or generating effective knowledge by deconstructing the defect cases.
5. The knowledge-graph-based substation equipment defect analysis system of claim 4, wherein the method of deconstructing defect cases comprises the steps of:
s31, splitting the content of the defect case and cleaning the text;
s32, deconstructing defect content, and acquiring all object attributes of the defect content;
and S33, storing the content of the defect case and the corresponding object attribute into the corresponding data table entry according to the obtained object attribute.
6. The knowledge-graph-based substation equipment defect analysis system of claim 4, wherein the subsystem that deconstructs defect cases comprises the following three modules:
a, a defect content splitting module, wherein the defect content splitting module splits the defect content and splits the defect content into the following parts according to the document format of the defect case: title, discovery processing and equipment conditions are obtained, and the split content is cleaned and then transmitted to an object attribute identification module;
b, an object attribute identification module, wherein the object attribute identification module identifies the content of the defect document, acquires all object attribute information corresponding to the content of the defect, and transmits the object information to a corresponding storage module;
and c, a deconstruction result storage module, wherein the deconstruction result storage module forms a corresponding mapping relation for all the object attribute information acquired by the object attribute identification model and stores the mapping relation in the database through corresponding data table items.
7. The substation equipment defect analysis system based on the knowledge graph of claim 1, wherein the knowledge graph construction module analyzes and constructs the reconstructed structural data by using a graph calculation engine technology to realize the relevant knowledge graph in the field of substation equipment defect analysis, abstracts the relation between various types of equipment and defect characteristic data and information of defect reasons, defect types and solutions into a graph, combines the semi-structured characteristics of graph data, and adopts a graph vertex as a center and a graph-based calculation engine to perform operation optimization and establish the equipment defect characteristic knowledge graph.
8. The transformer substation equipment defect analysis system based on the knowledge graph of claim 1, wherein the intelligent inference module performs matching inference on current defect feature information, existing defect knowledge in a knowledge base and expert experience according to the knowledge graph constructed by the knowledge graph construction module to find out the defect reasons.
9. The knowledge-graph-based substation equipment defect analysis system according to claim 8, wherein the process of matching inference mainly comprises the steps of:
s51, selecting the structure type of the substation equipment;
s52, inputting substation equipment information and performing familial defect analysis;
s53, selecting characteristic parts, inquiring a knowledge map according to the selected characteristic parts, finding out a judging method and a technical standard which have indirect relation with the characteristic parts, and pushing the judging method;
s54, selecting a judgment method, inquiring data and phenomenon characteristics which have indirect relation with the judgment method in the knowledge graph, and displaying;
s55, the user correspondingly selects the results in the step S54 according to the characteristics shown by the substation field fault equipment;
s56, through the data and the phenomenon characteristics selected in the step S55, the data and the phenomenon characteristics corresponding to the defect reasons in the knowledge map are hit, if the hit rate is satisfied, the defect reason with the highest hit rate is output as a matching result, and if the hit rate is not satisfied, the step S57 is continued;
and S57, pushing the related features from top to bottom according to the hit rate sequence of the defect reasons in the step S56 until the matching degree meets the requirement, outputting a diagnosis result, and ending the reasoning process, or outputting a prompt of an operation suggestion of the next step until the related features are pushed completely, and ending the reasoning process.
10. The knowledge-graph-based substation equipment defect analysis system of claim 9, wherein the performing familial defect analysis calculation logic in step S52 comprises the steps of:
s521, the case management module maintains the inspected defect cases;
s522, each defect case which is approved has a system label;
s523, matching the system labels of all cases according to the input substation equipment information, and counting the times of the same equipment manufacturer, the same equipment model, the same defect part and the same defect type;
and S524, judging whether the familial defect rule is satisfied, if so, outputting a familial defect analysis result, and if not, continuing to execute the step S53.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210326061.1A CN114792140A (en) | 2022-03-30 | 2022-03-30 | Transformer substation defect analysis system based on knowledge graph |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210326061.1A CN114792140A (en) | 2022-03-30 | 2022-03-30 | Transformer substation defect analysis system based on knowledge graph |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114792140A true CN114792140A (en) | 2022-07-26 |
Family
ID=82461159
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210326061.1A Pending CN114792140A (en) | 2022-03-30 | 2022-03-30 | Transformer substation defect analysis system based on knowledge graph |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114792140A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116993327A (en) * | 2023-09-26 | 2023-11-03 | 国网安徽省电力有限公司经济技术研究院 | Defect positioning system and method for transformer substation |
CN117522371A (en) * | 2023-09-27 | 2024-02-06 | 华能核能技术研究院有限公司 | Nuclear power plant defect automatic judging and executing professional method and system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108537344A (en) * | 2018-02-01 | 2018-09-14 | 贵州电网有限责任公司 | Secondary device intelligence O&M method based on closed loop information management |
CN110705710A (en) * | 2019-04-17 | 2020-01-17 | 中国石油大学(华东) | Knowledge graph-based industrial fault analysis expert system |
WO2021008180A1 (en) * | 2019-07-16 | 2021-01-21 | 扬州大学 | Software defect knowledge-oriented knowledge search method |
CN113112216A (en) * | 2021-03-03 | 2021-07-13 | 广东电网有限责任公司 | Prejudgment analysis method for equipment defects |
CN113112164A (en) * | 2021-04-19 | 2021-07-13 | 特变电工股份有限公司新疆变压器厂 | Transformer fault diagnosis method and device based on knowledge graph and electronic equipment |
CN113987139A (en) * | 2021-09-24 | 2022-01-28 | 南京航空航天大学 | Knowledge graph-based visual query management system for software defect cases of aircraft engine FADEC system |
CN114117061A (en) * | 2021-10-27 | 2022-03-01 | 南京信息工程大学 | River facies knowledge graph reverse-deducing method based on data mining and tree structure |
-
2022
- 2022-03-30 CN CN202210326061.1A patent/CN114792140A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108537344A (en) * | 2018-02-01 | 2018-09-14 | 贵州电网有限责任公司 | Secondary device intelligence O&M method based on closed loop information management |
CN110705710A (en) * | 2019-04-17 | 2020-01-17 | 中国石油大学(华东) | Knowledge graph-based industrial fault analysis expert system |
WO2021008180A1 (en) * | 2019-07-16 | 2021-01-21 | 扬州大学 | Software defect knowledge-oriented knowledge search method |
CN113112216A (en) * | 2021-03-03 | 2021-07-13 | 广东电网有限责任公司 | Prejudgment analysis method for equipment defects |
CN113112164A (en) * | 2021-04-19 | 2021-07-13 | 特变电工股份有限公司新疆变压器厂 | Transformer fault diagnosis method and device based on knowledge graph and electronic equipment |
CN113987139A (en) * | 2021-09-24 | 2022-01-28 | 南京航空航天大学 | Knowledge graph-based visual query management system for software defect cases of aircraft engine FADEC system |
CN114117061A (en) * | 2021-10-27 | 2022-03-01 | 南京信息工程大学 | River facies knowledge graph reverse-deducing method based on data mining and tree structure |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116993327A (en) * | 2023-09-26 | 2023-11-03 | 国网安徽省电力有限公司经济技术研究院 | Defect positioning system and method for transformer substation |
CN116993327B (en) * | 2023-09-26 | 2023-12-15 | 国网安徽省电力有限公司经济技术研究院 | Defect positioning system and method for transformer substation |
CN117522371A (en) * | 2023-09-27 | 2024-02-06 | 华能核能技术研究院有限公司 | Nuclear power plant defect automatic judging and executing professional method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108804521B (en) | Knowledge graph-based question-answering method and agricultural encyclopedia question-answering system | |
CN106295807B (en) | A kind of method and device of information processing | |
CN108932945B (en) | Voice instruction processing method and device | |
CN114792140A (en) | Transformer substation defect analysis system based on knowledge graph | |
CN107330125A (en) | The unstructured distribution data integrated approach of magnanimity of knowledge based graphical spectrum technology | |
CN110209828B (en) | Case query method, case query device, computer device and storage medium | |
US20220164531A1 (en) | Quality assessment method for automatic annotation of speech data | |
CN112597038B (en) | Software defect prediction method and system | |
CN112163553B (en) | Material price accounting method, device, storage medium and computer equipment | |
CN104347071B (en) | Method and system for generating reference answers of spoken language test | |
CN113626573B (en) | Sales session objection and response extraction method and system | |
CN111026884A (en) | Dialog corpus generation method for improving quality and diversity of human-computer interaction dialog corpus | |
CN112765974A (en) | Service assisting method, electronic device and readable storage medium | |
CN116108190A (en) | Intelligent operation and maintenance-oriented power transformer knowledge graph construction method | |
CN111104503A (en) | Construction engineering quality acceptance standard question-answering system and construction method thereof | |
Mu et al. | NERO: A text-based tool for content annotation and detection of smells in feature requests | |
CN113626557A (en) | Intelligent law enforcement recommendation auxiliary system based on element labeling and BERT and RCNN algorithms | |
CN115392255A (en) | Few-sample machine reading understanding method for bridge detection text | |
CN115114417A (en) | Automatic scoring method and device for Chinese blank filling questions | |
CN113987141A (en) | Question-answering system answer reliability instant checking method based on recursive query | |
CN112035680B (en) | Knowledge graph construction method of intelligent auxiliary learning machine | |
CN111209375B (en) | Universal clause and document matching method | |
CN111274404B (en) | Small sample entity multi-field classification method based on man-machine cooperation | |
CN113778875A (en) | System test defect classification method, device, equipment and storage medium | |
CN113722421A (en) | Contract auditing method and system and computer readable storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |