CN113553420B - Power grid fault processing rule recommendation method and system based on knowledge graph - Google Patents
Power grid fault processing rule recommendation method and system based on knowledge graph Download PDFInfo
- Publication number
- CN113553420B CN113553420B CN202010321359.4A CN202010321359A CN113553420B CN 113553420 B CN113553420 B CN 113553420B CN 202010321359 A CN202010321359 A CN 202010321359A CN 113553420 B CN113553420 B CN 113553420B
- Authority
- CN
- China
- Prior art keywords
- fault
- text
- knowledge graph
- event
- attribute
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 147
- 238000012545 processing Methods 0.000 title claims abstract description 138
- 238000004590 computer program Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 239000013598 vector Substances 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 17
- 238000010586 diagram Methods 0.000 description 12
- 230000003993 interaction Effects 0.000 description 7
- 238000007726 management method Methods 0.000 description 7
- 230000000007 visual effect Effects 0.000 description 5
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 238000012958 reprocessing Methods 0.000 description 1
Classifications
-
- 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
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/316—Indexing structures
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- 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
- 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)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses a power grid fault processing procedure recommendation method and system based on a knowledge graph, comprising the steps of carrying out semantic annotation on each fault processing procedure text to obtain a corresponding fault equipment main body, fault attributes comprising fault processing procedure text identifiers and fault event types; obtaining a knowledge graph by respectively utilizing the obtained equipment main body, fault attribute and fault event type of each fault and using the entity, attribute and event to represent topological relation; identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; and retrieving the knowledge graph according to the identification result to obtain each fault processing procedure text. According to the application, the fault related knowledge topology form is stored in the graph database, and the occurrence of human misoperation accidents is avoided by searching in the graph database according to the fault text, so that the accident handling time is effectively shortened, the handling capacity of a dispatcher on the power grid accidents is improved, and the intelligent level of the power grid accident handling is improved.
Description
Technical Field
The application belongs to the technical field of power grid fault processing, and particularly relates to a power grid fault processing rule recommendation method and system based on a knowledge graph.
Background
The power grid fault processing is one of core contents of scheduling daily work, and is an important guarantee for maintaining the safety of a power grid. In recent years, to assist a dispatcher in rapidly processing power grid faults, the capability of the dispatcher in handling power grid operation is improved, and related improvement work is mainly focused on monitoring, analyzing and alarming of power grid fault problems.
However, after a power grid fails or after an operation weak point exists in the power grid in detection, how the dispatcher can accurately obtain the corresponding fault treatment rules of the fault according to the fault is obviously supported insufficiently, and the specific steps are as follows: (1) Acquisition of the fault handling procedure is limited to the dispatcher's personal experience; (2) The existing tool does not organize procedure files according to the business process, and the execution efficiency is required to be further improved; (3) grid fault handling requires direct support tools; (4) The key information of the fault processing rules corresponding to the faults is dispersed, and the effective value of the historical production operation information is fully mined.
The current-stage power grid dispatching is still analysis-experience dispatching essentially, only the perception capability of the tail end is improved, the central processing link still depends on personal experience and business literacy of a dispatcher, timeliness and accuracy of acquiring fault processing rules for faults directly depend on knowledge of the power grid operation characteristics of the dispatcher, familiarity of fault plans, understanding and flexible application of the dispatching rules, and therefore the power grid dispatching process needs to recommend the fault processing rules timely and accurately for specific faults, and intelligent processing capability of the system is improved.
Disclosure of Invention
Aiming at the problems that the acquisition of the fault handling procedure corresponding to the current power grid fault depends on the knowledge of a dispatcher on the running characteristics of the power grid, the familiarity of a fault plan, the understanding and flexible application of the dispatching procedure and the further improvement of the intelligent handling capacity of the system, the application provides a power grid fault handling procedure recommendation method and system based on a knowledge graph.
In order to achieve the technical purpose, the application adopts the following technical design.
On one hand, the application provides a power grid fault processing procedure recommendation method based on a knowledge graph, which comprises the following steps of:
semantic labeling is carried out on each fault processing rule text obtained in advance, so that a fault equipment main body corresponding to each fault processing rule text, fault attributes comprising fault processing rule text identifiers and fault event types are obtained;
obtaining a knowledge graph by respectively utilizing the obtained equipment main body, fault attribute and fault event type of each fault and using the entity, attribute and event to represent topological relation;
identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; and obtaining each fault processing procedure text according to the fault processing procedure text identification.
Further, performing semantic annotation on each pre-acquired fault processing procedure text to obtain a fault equipment main body corresponding to each fault processing procedure text, a fault attribute comprising a fault processing procedure text identifier and a specific fault type, wherein the specific method comprises the following steps of:
taking the file as a unit, and performing word segmentation on each fault processing rule text by using a barker word segmentation algorithm to obtain a word sequence;
each word in the obtained word sequence is subjected to semantic annotation, and the word representing the main body of the fault equipment is marked as a subclass of the father class of the entity; marking words representing fault attributes as subclasses of attribute parent classes; words representing the type of fault event are marked as subclasses of the event parent.
Further, the pre-trained bert semantic model is utilized to identify fault equipment bodies, fault attributes and fault event types contained in the fault text.
Further, retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; the specific method for obtaining each fault handling procedure text according to the procedure text identification comprises the following steps:
retrieving data of each node stored in the knowledge graph to obtain entity nodes and event nodes corresponding to the entity and event names of the faults in the fault text, and inquiring to obtain fault processing rule short text nodes in fault attributes with association relations between the entity and the event nodes; and finally, according to the fault processing procedure text identifier which the short text node belongs to, acquiring a corresponding complete procedure text from a relational database and matching the complete procedure text to acquire each fault processing procedure text.
Further, the method further comprises the steps of obtaining and analyzing an access request of a user, wherein the access request comprises an operation request of a knowledge graph or a fault text of a request fault processing procedure text, and the returned result of the access request is displayed through a graphical display interface to realize the management of the knowledge graph and the display of the fault corresponding to obtain each fault processing procedure text.
In a second aspect, the application provides a power grid fault processing system based on a knowledge graph, wherein the system comprises a logic layer, and the logic layer comprises a logic processing unit and a knowledge graph representing unit;
the logic processing unit is used for carrying out semantic annotation on each fault processing rule text obtained in advance to obtain a fault equipment main body corresponding to each fault processing rule text, a fault attribute comprising a fault processing rule text identifier and a fault event type;
the logic processing unit is also used for identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; and obtaining each fault processing procedure text according to the fault processing procedure text identification.
The knowledge graph representation unit is used for obtaining the knowledge graph by using the obtained equipment main body, fault attribute and fault event type of each fault and using the entity, attribute and event representation topological relation
Optionally, the logic layer further includes a storage unit, where the storage unit includes a graph database and a relational database, where the graph database is used to store a knowledge graph, and the relational database is used to store fault handling procedure text.
The application also provides a computer readable storage medium which stores a computer program, and is characterized in that the computer program realizes the steps of the power grid fault processing procedure recommendation method based on the knowledge graph provided by the technical design when being executed by a processor.
The beneficial technical effects are as follows:
according to the application, the power grid fault processing knowledge graph is constructed, key data in power grid fault processing rules such as power grid operation data, fault processing plans and the like are uniformly analyzed into entities, attributes and events, the entities and the events are represented in a topological relation graph form to obtain the knowledge graph, the fault processing rules corresponding to faults are obtained through searching in a graph database according to fault texts, the problems of inaccuracy, timeliness and low efficiency caused by manual searching of the fault processing rules are avoided, the workload of workers and the occurrence of misoperation accidents caused by inaccurate text of the obtained fault processing rules are reduced, the accident processing time is effectively shortened, the processing capacity of a dispatcher on the power grid accidents is improved, the occurrence and the extension of the power grid accidents are effectively controlled, the transition from empirical dispatching to knowledge-driven dispatching is realized, and the intelligent level of the power grid accident processing is improved.
According to the application, the logic layer, the control layer, the representation layer and the middle layer are used for realizing multidirectional interaction, friendly interaction with a dispatcher is realized, the intelligent processing result of the system is clearly and definitely displayed, and the efficiency of fault processing procedure recommendation and the visualization program are improved. The knowledge graph is managed through the graphical interface, and on the basis of meeting the high expandability of the knowledge graph, the efficient and visual knowledge query and positioning functions are provided for users and application based on the knowledge graph through optimizing an index establishing mode, so that the accuracy and high-efficiency power grid fault processing graph knowledge retrieval is realized.
Drawings
FIG. 1 is a logic architecture diagram of power grid fault processing based on a knowledge graph provided by an embodiment of the application;
FIG. 2 is a diagram illustrating a system prototype design according to an embodiment of the present application;
FIG. 3 is a diagram of a knowledge storage architecture (substation) of the knowledge management section of the present application;
FIG. 4 is a diagram of the knowledge storage architecture (circuit) of the knowledge management section of the present application;
FIG. 5 is a schematic diagram of a fault matching process in an embodiment of the present application;
FIG. 6 is a flow chart of a similarity calculation process in a matching process according to an embodiment of the present application.
Detailed Description
The application is further described below with reference to the drawings and the detailed description.
The power grid fault handling procedure recommendation method based on the knowledge graph comprises the steps of carrying out semantic annotation on each pre-acquired fault handling procedure text to obtain a fault equipment main body corresponding to each fault handling procedure text, fault attributes comprising fault handling procedure text identifiers and fault event types;
obtaining a knowledge graph by respectively utilizing the obtained equipment main body, fault attribute and fault event type of each fault and using the entity, attribute and event to represent topological relation;
rule reasoning, the rule reasoning comprising: identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; and obtaining each fault handling procedure text stored in the relational database according to the fault handling procedure text identification.
The specific embodiment is to identify typical anomalies and fault scenes such as operation anomalies, system oscillations, system disconnection, power flow limit violations, equipment faults and the like, determine fault types, obtain fault attributes and fault equipment topics based on original information and data such as keywords, operation data, fault alarm signals, fault wave recording data and the like when faults occur, support a power grid fault processing knowledge base, and obtain typical processing procedure information under different fault conditions through matching, wherein the typical processing procedure information comprises fault names, post-fault mode descriptions, identification and extraction methods of document text of information such as scheduling processing steps and the like, realize automatic identification and classification of faults to be checked, and return scheduling processing steps in processing procedures of faults which are successful in matching or failed in matching and similar confirmation after similarity calculation.
According to the method, the fault processing knowledge graph based on graph database storage is constructed, the organic combination of the complex connection among knowledge information entities, relations and events is realized, the basic data, the search query and other service functions are provided for the rule reasoning process, meanwhile, on the basis that the high expandability of the knowledge graph is met, the efficient and visual knowledge query and positioning functions are provided for users and the application based on the knowledge graph through the optimized index establishment mode, and the rapid search of the knowledge graph is further realized.
On the basis of the above embodiment, optionally, performing semantic annotation on each fault handling procedure text obtained in advance to obtain a fault equipment main body corresponding to each fault handling procedure text, a fault attribute including a fault handling procedure text identifier, and a specific method of a fault type, including:
taking the file as a unit, and performing word segmentation on each fault processing rule text by using a barker word segmentation algorithm to obtain a word sequence;
each word in the obtained word sequence is subjected to semantic annotation, and the word representing the main body of the fault equipment is marked as a subclass of the father class of the entity; marking words representing fault attributes as subclasses of attribute parent classes; words representing the type of fault event are marked as subclasses of the event parent.
The pre-trained bert semantic model may be utilized to identify the fault device body, fault attributes, and fault event types contained in the fault text.
And constructing a knowledge graph, distinguishing the RDF data characteristics of the entity, the attribute and the relation triples from the traditional RDF data characteristics, and aiming at an industrial application scene, using a graph database based on the attribute graph, wherein the entity node and the relation node both contain attribute information. And dividing the fault processing rule text into words, marking the words, the attributes and the events, excluding repeated parts, and dividing the fault processing rule text into a plurality of short text entities taking short sentences as units by taking punctuation marks as boundaries instead of marking the words. In the form of an attribute graph, different from a triplet form, namely, the entity node and the relation node contain attributes, and the obtained equipment main body, the fault attribute and the fault event type of each fault respectively use the entity, the attribute and the event to represent the topological relation to obtain a knowledge graph. And storing the knowledge graph by using the graph database, and storing massive knowledge data based on the attribute graph data by using the knowledge graph.
The fault treatment procedure recommended by the application comprises the following components: the title comprises: some fault handling plan, fault number (i.e. fault handling procedure text identification id), grid operation mode description (negligible), expected fault (same as header), mode description after fault (this section mainly describes what will happen after fault occurs), scheduling handling steps (what is actually recommended to the user is this section, how to solve the fault, including fault handling steps).
According to the application, the power grid fault processing knowledge graph is constructed, key data in power grid operation data, fault processing plans and other power grid fault processing rules are uniformly analyzed into entities, attributes and events, the entities and the events are represented in a topological relation graph form to obtain the knowledge graph, the fault processing rules corresponding to faults are obtained through searching in a graph database according to fault texts, the problems of inaccuracy and timeliness in manually searching the fault processing rules are avoided, the occurrence of misoperation accidents is reduced, the accident processing time is effectively shortened, the processing capacity of a dispatcher on the power grid accidents is improved, the occurrence and the extension of the power grid accidents are effectively controlled, the transition from empirical dispatching to knowledge driving dispatching is realized, and the intelligent level of the power grid accident processing is improved.
Based on the second embodiment, for the input fault text, using the bert semantic model to semantically identify the entity, attribute and event in the input fault text;
in this embodiment, in order to more quickly and accurately obtain the fault handling procedure text identifier of the fault matching, the present embodiment decomposes the fault handling procedure text into a plurality of short text entities with short sentences as units by using punctuation marks as boundaries, and stores the short text entities as fault attributes as subclasses of attribute parent classes. In the embodiment, retrieving a knowledge graph according to the identification result to obtain a fault processing procedure text identifier of the fault matching; the specific method for obtaining each fault handling procedure text according to the procedure text identification comprises (as shown in fig. 5):
based on the identified entity name, the event name retrieves the entity node, such as a substation, and the event node, such as a total stop fault, associated with the fault from the knowledge graph. Each entity node or event node is associated with a series of fault procedure short text nodes, the nodes comprise a post-fault operation mode in a fault procedure text, the part is decomposed into a series of short sentences without symbols, and each short text node only comprises a sentence phrase and an id of the fault procedure text, namely a long text (namely fault treatment procedure text identification). The complete procedure text is stored in a relational database and can be retrieved and obtained by id;
screening out fault short text nodes shared by the entity nodes and the event nodes, and acquiring fault procedure long text in the relational database according to the id;
inputting a fault text, and carrying out semantic matching on the fault text and a fault post-operation mode part in a long text through a bert semantic model;
and if the matching is successful, returning the scheduling treatment step in the long text as a result of rule reasoning.
Further the method comprises the optional steps (as shown in fig. 6): if the matching fails, searching for similar faults. On the premise that the events are the same, corresponding entities with the same voltage class as the belonging areas in the attributes are screened first. And obtaining the number n1 of low-voltage-class substations connected in the corresponding entity, the number n2 of high-voltage-class substations and the numbers m1 and m2 of the two substations in the entity identified in the original input fault text. And (3) performing cosine similarity calculation on the (n 1, n 2) vector and the (m 1, m 2) vector, obtaining a plurality of entity nodes with higher scores, associating the fault short text nodes with the event nodes and the fault procedure long text, and matching the fault processing procedure text identifiers associated with the entity nodes with higher scores and the event nodes with the fault procedure long text.
The third embodiment provides, on the basis of the above embodiments, the power grid fault handling procedure recommendation method based on a knowledge graph according to the present embodiment further includes obtaining and analyzing an access request of a user, where the access request includes an operation request for the knowledge graph or a fault text requesting a fault handling procedure text, matching a view corresponding to the access request, and forwarding the access request to the corresponding view; and the view function in the view converts the access request into a call request, returns a return result of the call request to the graphical display interface for display, manages the knowledge graph and displays the text of each fault processing procedure corresponding to the fault.
And optionally carrying out structure conversion packaging on the returned result of the logic processing unit and returning to the graphical display interface.
The access request can also realize the management function logic architecture and strategy of knowledge map addition, update, deletion, import and the like, and the graphical interface and operation mode corresponding to the functions.
The entity in the application refers to an equipment main body (shown in figure 3) and comprises a transformer substation, a circuit breaker, a disconnecting link and the like; the attribute refers to information such as voltage level, belonging area and processing procedure text id (namely fault processing procedure text identifier); events refer to various types of faults.
According to the embodiment, by relying on the node data stored in the knowledge graph, the entity and the event name in the input fault text are searched in the knowledge graph, so that the entity node and the event node related to the fault in the graph are obtained, wherein the entity node is a transformer substation, and the event node is a complete stop fault. And inquiring to obtain a fault processing procedure short text node which has an association relation with the entity and the event node. And finally, according to the rule text id of the short text node, acquiring a corresponding complete rule text from a relational database, matching, returning the processing scheduling treatment in the rule text as a rule reasoning result, and displaying the result on a graphical display interface. The embodiment realizes multidirectional interaction, friendly interaction with a dispatcher, clear and definite display of the intelligent processing result of the system and improvement of the failure processing efficiency. The knowledge graph is managed through the graphical interface, and on the basis of meeting the high expandability of the knowledge graph, the efficient and visual knowledge query and positioning functions are provided for users and application based on the knowledge graph through optimizing an index establishing mode, so that the accuracy and high-efficiency power grid fault processing graph knowledge retrieval is realized.
An embodiment four provides a power grid fault processing system based on a knowledge graph, including: the logic layer comprises a logic processing unit and a knowledge graph representation unit;
the logic processing unit is used for carrying out semantic annotation on each fault processing rule text obtained in advance to obtain a fault equipment main body corresponding to each fault processing rule text, a fault attribute comprising a fault processing rule text identifier and a fault event type;
the logic processing unit is also used for identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; and obtaining each fault processing procedure text according to the fault processing procedure text identification.
The knowledge graph representing unit is used for obtaining the knowledge graph by using the obtained equipment main body, fault attribute and fault event type of each fault through using the knowledge graph and using the entity, attribute and event representing topological relation.
Optionally, the logic layer further includes a storage unit, where the storage unit includes a graph database and a relational database, where the graph database is used to store a knowledge graph, and the relational database is used to store fault handling procedure text.
Fifth embodiment as shown in fig. 1 and 2, on the basis of the fourth embodiment, the system further optionally further includes a control layer, a presentation layer, and an intermediate layer:
the control layer is used for acquiring and analyzing an access request of a user, wherein the access request comprises an operation request of a knowledge graph or a fault text of a request fault processing procedure text, and the access request is forwarded to a corresponding view by matching views in a middle layer corresponding to the access request;
the middle layer comprises a view, a view function in the view converts an access request forwarded by a controller into a call request for a logic processing unit in a logic layer, and the logic processing unit accesses a knowledge graph according to the call request, and returns a returned result to a template page in a representation layer for display after structure conversion packaging;
the representation layer is used for displaying the result of the access request through the graphical template page to manage the knowledge graph and display fault corresponding fault processing procedure texts.
The presentation layer enables knowledge visual presentation and friendly interaction. Based on the knowledge data storage structure of the attribute graph, a logic architecture and a strategy of management functions for knowledge addition, updating, deleting, importing and the like, and a graphical interface and an operation mode corresponding to the functions are designed in the representation layer unit.
FIG. 2 is a schematic diagram of a system architecture, in which a controller in a control layer unit matches corresponding view functions according to an input access request. The view in the middle tier element requests data from the logical tier and submits the data returned by the logical tier to the template. The logic processing unit executes specific business logic, namely specific implementation of a power grid fault processing method, and the specific implementation comprises rule reasoning, knowledge graph management and the like. The logic processing unit presents the relevant data on a page representing the layer unit.
Fig. 3 and fig. 4 are knowledge graph structures of the knowledge management part of the present application, where the knowledge graph stores entities (e.g. transformer stations, lines), events (e.g. total stop accidents), included attributes (e.g. voltage levels) and relationships among all fault handling rule texts. Taking the topological structures of the transformer substation entity (shown in fig. 3) and the line entity (shown in fig. 4) in the graph database as examples. The connection relation between the transformer substation topological structure and the adjacent nodes is divided into belonging type, connection type and possession type, wherein the belonging type is connected with the attribute nodes such as voltage class, the belonging type is connected with the control area node, the connection type is connected with the line node, and the possession type is connected with the interval node of the transformer substation or other transformer substations. The connection relation between the line topology structure and the adjacent nodes is divided into a type and a connection type, wherein the type is connected with two attribute nodes of a starting point and an ending point at two ends of the line, and the connection type is connected with transformer substation nodes at two ends of the line.
The logic layer firstly obtains the running characteristics of the power grid and the like according to the fault partition, judges whether the fault names are matched with related fault examples, and if so, retrieves the information of the normal running mode, the running mode after the fault and the like associated with the fault examples from the map, and feeds back the information to the page for selection. If the fault names are not matched, manually searching the matched fault instance according to the equipment names of the power plant of the transformer substation, if the fault instance is successfully searched, operating the fault instance after the fault instance is matched with the fault name, and if the fault instance is not found, reporting the fault instance.
And after the fault occurs, the logic layer identifies the entity and the event corresponding to the fault, and retrieves the corresponding entity and event node in the knowledge graph according to the names of the entity and the event, and the corresponding fault procedure text judges whether the fault is matched with the input fault text to be checked. And if the matching is successful, returning to a scheduling treatment step in the fault procedure text. If not, obtaining similar fault procedures through similarity calculation for reprocessing.
Fig. 5 is a flowchart of the similarity calculation process. The similarity calculation is performed under the condition that the long text of the fault procedure in the last step of rule reasoning is not matched with the input fault text to be checked. The corresponding entities in the attribute are first screened for voltage class that is the same as the region to which they belong. And obtaining the number n1 of low-voltage-class substations connected in the corresponding entity, the number n2 of high-voltage-class substations and the numbers m1 and m2 of the two substations in the entity identified in the original input fault text. And (2) performing cosine similarity calculation on the (n 1, n 2) vector and the (m 1, m 2) vector, and returning a plurality of entity nodes with higher result scores for fault matching in rule pushing.
The application realizes high-speed indexing oriented to massive knowledge by utilizing the knowledge graph. On the basis of fully considering the influence of index feature quality on index performance, a high-speed index technology combining the structural mode and index characteristics of the optimized graph index is researched for minimizing the size of the knowledge query search space. The indexing features are selected according to the following principle:
(1) Preference for features with high pruning forces
(2) Ensuring feature coverage complete query graphs
(3) Feature count is low to ensure shrinkage of index structure
According to the embodiment, the logic layer, the control layer, the representation layer and the middle layer are used for realizing multidirectional interaction, friendly interaction with a dispatcher is realized, the intelligent processing result of the system is clearly and definitely displayed, and the efficiency of fault processing procedure recommendation and the visualization program are improved. The knowledge graph is managed through the graphical interface, and on the basis of meeting the high expandability of the knowledge graph, the efficient and visual knowledge query and positioning functions are provided for users and application based on the knowledge graph through optimizing an index establishing mode, so that the accuracy and high-efficiency power grid fault processing graph knowledge retrieval is realized.
The application can realize the associated search query function by utilizing the knowledge graph, and can realize sub-graph query, hypergraph query and shortest path query based on the graph database architecture. And (3) a map database retrieval technology and means for reachability confirmation, partial matching and the like are used for researching a power grid fault processing map knowledge retrieval technology meeting accuracy and high efficiency. The method is characterized in that the method comprises the steps of compressing organization facing a graph database, effective feature index construction and processing query based on the compressing organization are studied deeply, and the size of a search candidate result set is fully reduced.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are all within the protection of the present application.
Claims (6)
1. The power grid fault processing procedure recommendation method based on the knowledge graph is characterized by comprising the following steps of:
semantic labeling is carried out on each fault processing rule text obtained in advance, so that a fault equipment main body corresponding to each fault processing rule text, fault attributes comprising fault processing rule text identifiers and fault event types are obtained;
obtaining a knowledge graph by respectively utilizing the obtained equipment main body, fault attribute and fault event type of each fault and using the entity, attribute and event to represent topological relation; identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; recommending each fault handling procedure text according to the fault handling procedure text identification;
searching a knowledge graph according to the identification result, and searching similar faults if matching fails in the fault processing procedure text identification process of the fault matching, wherein the specific method is as follows:
on the premise that the events are the same, firstly screening corresponding entities with the same voltage class as the affiliated region in the attribute; obtaining the number of the equipment main bodies with different voltage levels connected in the corresponding entity to form a first vector, and the number of the same equipment main bodies in the entity identified in the original input fault text to form a second vector; performing cosine similarity calculation on the two vectors, and matching fault handling procedure text identifiers associated with a plurality of entity nodes with higher scores and event nodes with fault procedure long texts;
the method further comprises the steps of obtaining and analyzing an access request of a user, wherein the access request comprises an operation request of a knowledge graph or a fault text requesting fault processing procedure text, and a returned result of the access request is displayed through a graphical display interface, so that the knowledge graph is managed, and each fault processing procedure text corresponding to a displayed fault is obtained.
2. The power grid fault handling procedure recommendation method based on the knowledge graph according to claim 1, wherein the specific method for performing semantic annotation on each pre-acquired fault handling procedure text to obtain a fault equipment main body corresponding to each fault handling procedure text, fault attributes including fault handling procedure text identifiers and fault types comprises the following steps:
taking the file as a unit, and performing word segmentation on each fault processing rule text by using a barker word segmentation algorithm to obtain a word sequence;
each word in the obtained word sequence is subjected to semantic annotation, and the word representing the main body of the fault equipment is marked as a subclass of the father class of the entity; marking words representing fault attributes as subclasses of attribute parent classes; words representing the type of fault event are marked as subclasses of the event parent.
3. The knowledge-graph-based power grid fault handling procedure recommendation method according to claim 1, wherein a pre-trained bert semantic model is utilized to identify fault equipment bodies, fault attributes and fault event types contained in fault texts.
4. The power grid fault handling procedure recommendation method based on the knowledge graph according to claim 1, wherein the knowledge graph is retrieved according to the recognition result, and a fault handling procedure text identifier of the fault matching is obtained; the specific method for recommending each fault handling procedure text according to the procedure text identification comprises the following steps:
retrieving data of each node stored in the knowledge graph to obtain entity nodes and event nodes corresponding to the entity and event names of the faults in the fault text, and inquiring to obtain fault processing rule short text nodes in fault attribute nodes with association relations with the entity and the event nodes; and finally, according to the fault processing procedure text identifier which the short text node belongs to, acquiring a corresponding complete procedure text from a relational database and matching the complete procedure text to acquire each fault processing procedure text.
5. The power grid fault processing system based on the knowledge graph is characterized by comprising a logic layer, wherein the logic layer comprises a logic processing unit and a knowledge graph representing unit;
the logic processing unit is used for carrying out semantic annotation on each fault processing rule text obtained in advance to obtain a fault equipment main body corresponding to each fault processing rule text, a fault attribute comprising a fault processing rule text identifier and a fault event type;
the logic processing unit is also used for identifying a fault equipment main body, a fault attribute and a fault event type contained in the fault text; retrieving a knowledge graph according to the identification result, and acquiring a fault processing procedure text identifier of the fault matching; obtaining each fault processing procedure text according to the fault processing procedure text identification;
the knowledge graph representation unit is used for obtaining a knowledge graph by using the obtained equipment main body, fault attribute and fault event type of each fault and using the entity, attribute and event representation topological relation;
the logic layer is also used for searching the knowledge graph to find similar faults if the matching fails in the fault processing procedure text identification process of the fault matching, and the specific method is as follows: on the premise that the events are the same, firstly screening corresponding entities with the same voltage class as the affiliated region in the attribute; obtaining the number of the equipment main bodies with different voltage levels connected in the corresponding entity to form a first vector, and the number of the same equipment main bodies in the entity identified in the original input fault text to form a second vector; performing cosine similarity calculation on the two vectors, and matching fault handling procedure text identifiers associated with a plurality of entity nodes with higher scores and event nodes with fault procedure long texts; the system further includes a control layer, a presentation layer, and an intermediate layer:
the control layer is used for acquiring and analyzing an access request of a user, wherein the access request comprises an operation request of a knowledge graph or a fault text of a request fault processing procedure text, and the access request is forwarded to a corresponding view by matching views in a middle layer corresponding to the access request;
the middle layer comprises a view, a view function in the view converts an access request forwarded by a controller into a call request for a logic processing unit in a logic layer, and returns a result returned by the logic processing unit to a template page in a presentation layer for presentation;
the representation layer is used for displaying the result of the access request through the graphical template page to manage the knowledge graph and display fault corresponding fault processing procedure texts.
6. A computer readable storage medium storing a computer program, which when executed by a processor performs the steps of the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010321359.4A CN113553420B (en) | 2020-04-22 | 2020-04-22 | Power grid fault processing rule recommendation method and system based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010321359.4A CN113553420B (en) | 2020-04-22 | 2020-04-22 | Power grid fault processing rule recommendation method and system based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113553420A CN113553420A (en) | 2021-10-26 |
| CN113553420B true CN113553420B (en) | 2023-11-24 |
Family
ID=78101017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010321359.4A Active CN113553420B (en) | 2020-04-22 | 2020-04-22 | Power grid fault processing rule recommendation method and system based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113553420B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114118879B (en) * | 2022-01-25 | 2022-04-29 | 浙江裕腾百诺环保科技股份有限公司 | Method, system, server and storage medium for recommending environment protection management and control measures |
| CN114757097B (en) * | 2022-04-07 | 2023-09-26 | 国网河北省电力有限公司邯郸供电分公司 | Line fault diagnosis method and device |
| CN115114448B (en) * | 2022-06-02 | 2023-07-14 | 中国电力科学研究院有限公司 | Intelligent multi-mode fusion power consumption inspection method, device, system, equipment and medium |
| CN115240390A (en) * | 2022-07-14 | 2022-10-25 | 华能核能技术研究院有限公司 | Automatic identification system, method, electronic device and medium for alarm response procedure |
| CN115292518A (en) * | 2022-08-18 | 2022-11-04 | 国家电网有限公司 | Power distribution network fault processing method and system based on knowledge type information extraction |
| CN115953146B (en) * | 2022-12-07 | 2024-02-13 | 国家电网有限公司 | Auxiliary decision-making system for fault treatment of power distribution network |
| CN115905577B (en) * | 2023-02-08 | 2023-06-02 | 支付宝(杭州)信息技术有限公司 | Knowledge graph construction method and device and rule retrieval method and device |
| CN116467468B (en) * | 2023-05-05 | 2024-01-05 | 国网浙江省电力有限公司 | Power management system abnormal information handling method based on knowledge graph technology |
| CN117150042B (en) * | 2023-09-01 | 2024-04-16 | 海通证券股份有限公司 | Method, device, equipment and medium for recommending emergency plans based on knowledge graph |
| CN116893924B (en) * | 2023-09-11 | 2023-12-01 | 江西南昌济生制药有限责任公司 | Equipment fault processing method, device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06309172A (en) * | 1993-04-22 | 1994-11-04 | Nec Corp | Knowledge base device and knowledge base constituting method |
| CN103033359A (en) * | 2012-12-19 | 2013-04-10 | 西安交通大学 | Wind turbine generator main gear fault diagnosis method of multi-feature multilevel comprehensive judgment |
| CN110008288A (en) * | 2019-02-19 | 2019-07-12 | 武汉烽火技术服务有限公司 | The construction method in the knowledge mapping library for Analysis of Network Malfunction and its application |
| CN110033101A (en) * | 2019-03-07 | 2019-07-19 | 华中科技大学 | The Fault Diagnosis Method of Hydro-generating Unit and system of knowledge mapping based on fusion feature |
| CN110175239A (en) * | 2019-04-23 | 2019-08-27 | 成都数联铭品科技有限公司 | A kind of construction method and system of knowledge mapping |
| CN110825885A (en) * | 2019-11-13 | 2020-02-21 | 南方电网科学研究院有限责任公司 | Power equipment knowledge graph application system |
| CN110941725A (en) * | 2019-11-29 | 2020-03-31 | 国网湖南省电力有限公司 | Knowledge graph-based hydroelectric generating set fault diagnosis method and system |
-
2020
- 2020-04-22 CN CN202010321359.4A patent/CN113553420B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06309172A (en) * | 1993-04-22 | 1994-11-04 | Nec Corp | Knowledge base device and knowledge base constituting method |
| CN103033359A (en) * | 2012-12-19 | 2013-04-10 | 西安交通大学 | Wind turbine generator main gear fault diagnosis method of multi-feature multilevel comprehensive judgment |
| CN110008288A (en) * | 2019-02-19 | 2019-07-12 | 武汉烽火技术服务有限公司 | The construction method in the knowledge mapping library for Analysis of Network Malfunction and its application |
| CN110033101A (en) * | 2019-03-07 | 2019-07-19 | 华中科技大学 | The Fault Diagnosis Method of Hydro-generating Unit and system of knowledge mapping based on fusion feature |
| CN110175239A (en) * | 2019-04-23 | 2019-08-27 | 成都数联铭品科技有限公司 | A kind of construction method and system of knowledge mapping |
| CN110825885A (en) * | 2019-11-13 | 2020-02-21 | 南方电网科学研究院有限责任公司 | Power equipment knowledge graph application system |
| CN110941725A (en) * | 2019-11-29 | 2020-03-31 | 国网湖南省电力有限公司 | Knowledge graph-based hydroelectric generating set fault diagnosis method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113553420A (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113553420B (en) | Power grid fault processing rule recommendation method and system based on knowledge graph | |
| CN107330125B (en) | Mass unstructured distribution network data integration method based on knowledge graph technology | |
| CN111985653B (en) | Power grid fault knowledge recommendation and knowledge management system and method based on knowledge graph | |
| CN115357726A (en) | Fault disposal plan digital model establishing method based on knowledge graph | |
| CN109977291A (en) | Search method, device, equipment and storage medium based on physical knowledge map | |
| CN105335488A (en) | Knowledge base construction method | |
| CN115757810A (en) | Method for constructing standard ontology of knowledge graph | |
| EP2599017A1 (en) | Electronic document repository system | |
| CN114048308A (en) | Method and device for generating category retrieval report | |
| KR101955376B1 (en) | Processing method for a relational query in distributed stream processing engine based on shared-nothing architecture, recording medium and device for performing the method | |
| CN103150632A (en) | Structuring method for flood control and drought control bulletin generation system based on water conservancy cloud platform | |
| CN115937881A (en) | Method for automatically identifying content of knowledge graph construction standard form | |
| JPH08305724A (en) | Device for managing design supporting information document | |
| CN115827885A (en) | Operation and maintenance knowledge graph construction method and device and electronic equipment | |
| CN115344661A (en) | Equipment halt diagnosis method and device, electronic equipment and storage medium | |
| CN113434658A (en) | Thermal power generating unit operation question-answer generation method, system, equipment and readable storage medium | |
| KR101592670B1 (en) | Apparatus for searching data using index and method for using the apparatus | |
| CN112685608A (en) | CYPHER-based power grid dispatching field knowledge attribute graph model construction method | |
| CN112668836A (en) | Risk graph-oriented associated risk evidence efficient mining and monitoring method and device | |
| CN112784129A (en) | Pump station equipment operation and maintenance data supervision platform | |
| Huang et al. | Knowledge storage and acquisition for industrial cyber-physical systems based on non-relational database | |
| CN115269500B (en) | Ecological environment data storage method, ecological environment data retrieval method and electronic equipment | |
| CN112686633B (en) | Intelligent ticket simulating method | |
| CN113392291B (en) | Service recommendation method and system based on data center | |
| Pan et al. | Research on intelligent search framework technology oriented to monitoring of power grid dispatching equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |