CN115357726A - Fault disposal plan digital model establishing method based on knowledge graph - Google Patents
Fault disposal plan digital model establishing method based on knowledge graph Download PDFInfo
- Publication number
- CN115357726A CN115357726A CN202210973549.3A CN202210973549A CN115357726A CN 115357726 A CN115357726 A CN 115357726A CN 202210973549 A CN202210973549 A CN 202210973549A CN 115357726 A CN115357726 A CN 115357726A
- Authority
- CN
- China
- Prior art keywords
- plan
- knowledge
- model
- power grid
- information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000006243 chemical reaction Methods 0.000 claims abstract description 25
- 230000004927 fusion Effects 0.000 claims abstract description 24
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000000605 extraction Methods 0.000 claims description 25
- 238000010276 construction Methods 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 17
- 230000008859 change Effects 0.000 claims description 10
- 230000014509 gene expression Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000005065 mining Methods 0.000 claims description 5
- 238000003058 natural language processing Methods 0.000 claims description 4
- 238000013441 quality evaluation Methods 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 claims description 3
- 230000002457 bidirectional effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000004590 computer program Methods 0.000 description 8
- 230000006872 improvement Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000008520 organization Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 239000012634 fragment Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
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
- G06F16/36—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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of power grid fault disposal, in particular to a fault disposal plan digital model establishing method based on a knowledge graph. The method comprises the following steps: structure conversion: firstly, extracting key information in a plan, and converting unstructured text information of the plan into structured information; fusion modeling: and performing correlation fusion on the structured plan information and the power grid model to form a digital fault handling plan. The invention designs and utilizes knowledge map technology to extract, express and manage the knowledge of the fault disposal information, and is used for assisting the dispatching personnel to dispose the fault, thereby effectively improving the emergency processing capability and the intelligent dispatching level of the power grid; by adopting a semantic recognition technology, an unstructured plan is converted into computer-recognizable structured information, a fault handling plan digital model is constructed based on a knowledge graph, and then the plan model is effectively associated with a power grid model and operation information, so that basic model support is provided for digital plan management.
Description
Technical Field
The invention relates to the technical field of power grid fault disposal, in particular to a fault disposal plan digital model establishing method based on a knowledge graph.
Background
The power grid fault handling plan has important guiding significance for efficient and orderly implementation of power grid accident emergency work, and is an important means for regulating and controlling operation personnel to deal with power grid sudden faults, orderly implement power grid emergency state correction control and maintain safe and stable operation of a power grid.
At present, the traditional document form management is still adopted for the fault handling plans by all levels of scheduling, the management is extensive, and the application efficiency is low. The handling and scheduling work after the power grid fault mainly depends on the subjective decision of scheduling personnel, and the scheduling personnel analyzes the state and parameter change condition of the power grid after the fault in real time, finds out the reason of the fault and makes corresponding fault handling measures. This kind of handling mode needs the scheduling personnel to look up and remember a large amount of fault handling information that exists in the form of non- (semi-) structured text repeatedly, such as system stability requirement, operation mode after the fault, fault handling key point etc.. Although the traditional text retrieval method using keywords for matching can provide a paragraph positioning function, retrieval results lack fragmentation and organization, situations of incomplete retrieval and questions of answers often occur, negligence and omission easily occur, and the efficiency of fault emergency treatment work is reduced. With the rapid development of power systems, the power grid structure and the operation mode become more complex, the disposal difficulty after a fault is continuously improved, and the conventional scheduling decision mechanism relying on manual experience is more and more difficult to deal with the rapid fault analysis and fault disposal of a complex large power grid. The power system needs to extract and refine unstructured fault handling text data into knowledge by means of intelligent technology, and organize the knowledge into a structured and visualized representation form. When a power grid fault occurs, a dispatcher is helped to quickly analyze the accident reason, comprehensively master key information of fault processing, and make an auxiliary decision so as to improve the emergency handling capacity of the power grid.
In view of this, we propose a method for establishing a digital model of a fault handling plan based on a knowledge graph.
Disclosure of Invention
The invention aims to provide a fault handling plan digital model building method based on a knowledge graph to solve the problems in the background technology.
In order to solve the above technical problem, an object of the present invention is to provide a method for establishing a digital model of a fault handling plan based on a knowledge graph, including the following steps:
s1, structure conversion: firstly, extracting key information in a plan, classifying and combining the key information according to business requirements to obtain a key information structural relationship, then performing semantic analysis and learning training by adopting artificial intelligence technologies such as a knowledge map, natural language processing, deep learning and the like, converting unstructured text information of the plan into structured information, and storing the structured information according to the key information structural relationship;
s2, fusion modeling: and performing association fusion on the structured plan information and the power grid model, and modeling based on the association relation to form a digital fault handling plan.
As a further improvement of the technical solution, in S1, the specific method for structure conversion includes the following steps:
s1.1, analyzing a plan text: analyzing faults which may occur in the operation of a power grid, forming a power grid fault disposal plan in a text format according to a preset plan template style, and performing text analysis according to the characteristics of a plan text;
s1.2, converting a plan structure: and converting the unstructured plan into structured information which can be identified by a computer, and constructing a plan knowledge graph according to the structured information.
As a further improvement of the technical scheme, in S1.1, a plan text is composed of five contents, namely a plan number, a fault name, a stability requirement, a post-fault operation mode, and a fault handling key; wherein: the stability requirement is used to describe stability regulations to which the fault handling work needs to comply; the operation mode after the fault is used for describing the change condition and reason of the operation mode of the power grid after the expected fault occurs; the main points of fault handling are used for describing handling measures and emergency state adjustment required by the power grid to recover normal operation.
In addition, the plan text has the following features:
the construction of the plan text is complex and the types are various;
the plan text contains a large number of proper nouns and special terms in the power field, such as plant station equipment, a power dispatching mechanism, dispatching instructions, equipment indexes and the like;
the entity in the plan text is usually formed by nesting a plurality of nouns such as regions, stations, voltage levels and the like, and the boundary of the entity is fuzzy.
As a further improvement of the technical solution, in S1.2, in the plan structure conversion, the specific method for constructing the plan knowledge graph includes the following steps
S1.2.1, constructing a plan knowledge graph-framework: analyzing and selecting a construction mode of a knowledge graph; firstly, designing a mode layer of a knowledge graph in a top-down mode by analyzing the content of a plan text; then, under the guidance of the mode layer, a data layer is constructed in a bottom-up mode; designing a proper extraction method aiming at the characteristics of the pre-arranged plan text, extracting three knowledge elements of entities, relations and attributes to form a series of high-quality fact expressions, and mapping the high-quality fact expressions into related concept nodes through a bottom storage mode of a knowledge graph;
s1.2.2, constructing a plan knowledge map-mode layer: detailed analysis is carried out on the text content of the pre-arranged plan, and the concept type of the key indexes of the power grid regulation and control service, the related attributes and the relationship among the concepts are extracted, so that a domain knowledge system is formed;
s1.2.3, and the construction of a plan knowledge graph-data: the graphic symbols are used for describing the real objects and the relations of the power grid regulation and control information in a visual form, the basic composition units of the graphic symbols are triples of 'entity-relation-entity' and 'entity-attribute-property value', and the entities are mutually connected through the relations to form a reticular knowledge structure.
As a further improvement of the technical solution, in S1.2.3, the basic method for constructing the plan knowledge graph-data includes the following steps:
information extraction: the method is characterized in that bidirectional, parallel and unsupervised training is carried out on text corpora to realize word vector conversion and semantic content extraction, namely, entities, relations and entity attributes are extracted from semi-structured and non-structured data such as alarm information clauses, fault handling plans, expert experiences, overhaul ticket work contents, fixed value items in fixed value lists and the like and converted into structured information, and ontology knowledge expression is formed on the basis, and related key technologies comprise: entity extraction, relation extraction and attribute extraction;
and (3) knowledge fusion: the relation among the information units after information extraction is flat, hierarchy and logic are lacked, a large number of redundant and even wrong information fragments exist, and multi-source description information about the same entity or concept is fused through knowledge fusion, wherein the multi-source description information comprises entity link, entity disambiguation, reference resolution and knowledge merging;
knowledge processing: the integrated and fused fact data does not form a structured and networked power grid regulation and control information knowledge system, and knowledge processing is required, and the method mainly comprises body construction, quality evaluation and feature mining.
As a further improvement of the technical solution, in S2, the specific method for fusion modeling includes the following steps:
s2.1, generating a plan knowledge map model: a knowledge graph model is formed by performing structural conversion on a fault handling plan, and a relation model is constructed according to power grid regulation and control services;
s2.2, constructing a knowledge graph model based on the CIM: the knowledge graph of the power grid basic model is used as a supporting model and can be used for converting data into the knowledge graph;
s2.3, forming a fused knowledge graph model: forming a structured plan knowledge graph through the structured conversion of the fault handling plan; carrying out association fusion on the plan structured model and the power grid model to form a plan digital model based on the knowledge graph;
and S2.4, considering the digital logic relation of the operation mode.
As a further improvement of the technical solution, in S2.2, after the knowledge graph model based on the CIM is analyzed by the power CIM model, the data and the relationships of the devices such as the bus, the station, the transformer, the line, the disconnecting switch, and the grounding switch can be acquired.
As a further improvement of the technical solution, in S2.3, in the fused knowledge graph model, the power grid base model mainly includes: the method comprises the following steps of (1) information such as a CIM (common information model), a power grid topological graph, alarm event relations, users, operators, a scheduling mechanism, operation risks, overhaul, fixed values and the like; the power grid model related to the plan association further comprises: grid equipment, historical event libraries, expert experience, and the like.
As a further improvement of the present technical solution, in S2.4, the digitalized logical relationship of the operation modes is considered: after the digital conversion of the fault handling plan is realized, modeling needs to be carried out on the digital logic relation considering the operation mode so as to conveniently carry out the safety analysis of the fault handling strategy by combining the current operation mode; after the file is analyzed, an ontology model is created, and the analyzed content can be independently objectified.
Another object of the present invention is to provide a platform device for operating a method for building a digital knowledge-graph-based fault handling plan model, which includes a processor, a memory, and a computer program stored in the memory and executed on the processor, where the processor is configured to implement the steps of the method for building a digital knowledge-graph-based fault handling plan model when executing the computer program.
It is a further object of the present invention to provide a computer-readable storage medium, which stores a computer program, which when executed by a processor, implements the steps of the method for building a digital knowledge-graph-based fault handling plan model.
Compared with the prior art, the invention has the beneficial effects that:
1. in the method for establishing the fault disposal plan digitalized model based on the knowledge map, the knowledge map technology is used for extracting, representing and managing the knowledge of the fault disposal information, and is used for assisting the dispatching personnel in disposing the fault, so that the emergency processing capability and the dispatching intelligence level of the power grid can be effectively improved;
2. according to the method for establishing the digital model of the fault disposal plan based on the knowledge graph, a semantic recognition technology is adopted, a text type disposal plan, namely an unstructured plan is converted into computer-recognizable structural information, then a digital model of the fault disposal plan is established based on the knowledge graph, the digital model comprises a document structure, various content key points and quantized information of the fault plan, the digital logic relation of an operation mode is considered, then the plan model is effectively associated with a power grid model and the operation information, and basic model support is provided for digital plan management.
Drawings
FIG. 1 is a schematic diagram of an exemplary overall process of the present invention;
FIG. 2 is an architecture diagram of an exemplary protocol knowledge-graph-framework construct in accordance with the present invention;
FIG. 3 is an architecture diagram of an exemplary plan knowledge graph-schema layer construction in accordance with the present invention;
FIG. 4 is an architecture diagram of an exemplary protocol knowledge graph-data layer construction in accordance with the present invention;
FIG. 5 is a diagram of an exemplary knowledge-graph structure in accordance with the present invention;
FIG. 6 is a schematic diagram of an exemplary grid topology to knowledge graph conversion in accordance with the present invention;
FIG. 7 is a block diagram of an exemplary electronic computer platform assembly in accordance with the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
As shown in fig. 1 to 7, the present embodiment provides a method for establishing a digital model of a failure handling plan based on a knowledge-graph, which includes the following steps:
s1, structure conversion: firstly, extracting key information in a plan, classifying and combining the key information according to business requirements to obtain a key information structural relationship, then performing semantic analysis and learning training by adopting artificial intelligence technologies such as a knowledge map, natural language processing, deep learning and the like, converting unstructured text information of the plan into structured information, and storing the structured information according to the key information structural relationship;
s2, fusion modeling: and performing association fusion on the structured plan information and the power grid model, and modeling based on the association relation to form a digital fault handling plan.
In the structure conversion step, the whole document is divided primarily, so that the accuracy of semantic analysis can be improved; the clear data structure is convenient for management and application of information, and is also convenient for associated modeling with a power grid model and quick query of relational data.
In this embodiment, in S1, the specific method for structure conversion includes the following steps:
s1.1, analyzing a plan text: analyzing faults which may occur in the operation of a power grid, forming a power grid fault disposal plan in a text format according to a preset plan template style, and performing text analysis according to the characteristics of a plan text; the method comprises the steps that a change situation of a power grid operation mode after a fault is simulated, an approximate influence range and a power grid weak link of the expected fault are determined, and fault handling measures are made in a targeted mode through scheduling experience;
s1.2, converting a plan structure: and converting the unstructured plan into structured information which can be identified by a computer, and constructing a plan knowledge map according to the structured information.
Further, in S1.1, a plan text is composed of five contents, namely a plan number, a fault name, a stability requirement, a fault-later operation mode and a fault handling key point; wherein: the stability requirement is used to describe stability regulations to which the fault handling work needs to comply; the operation mode after the fault is used for describing the change condition and reason of the operation mode of the power grid after the expected fault occurs; the main points of fault handling are used for describing handling measures and emergency state adjustment required by the power grid to recover normal operation.
Specifically, the effect of composing the plan text is to: in order to improve the fault emergency handling capacity of a power grid, ensure that the fault emergency handling work is carried out efficiently and orderly, reduce the influence of accidents on the society to the maximum extent, a controller analyzes the faults which may occur in the operation of the power grid, simulates the change situation of the operation mode of the power grid after the faults, determines the approximate influence range and weak links of the power grid of the expected faults, and establishes fault handling measures in a targeted manner through scheduling experience to form a power grid fault handling plan in a text format.
The preset pattern template has various patterns, for example, as shown in the following table:
compared with texts in the daily field, the plan text has the following characteristics:
the construction of the plan text is complex and the types are various; the method comprises the following steps of not only containing structured information such as a plan number, a fault name and stability requirements, but also containing unstructured information such as a fault-post-operation mode and fault handling key points; and the operation mode after the fault and the key points of fault handling can be further subdivided according to the description objects and the handling measures.
The plan text contains a large number of proper nouns and special terms in the power field, such as plant station equipment, a power dispatching mechanism, dispatching instructions, equipment indexes and the like; some natural language processing technologies suitable for the daily field, such as text word segmentation, part of speech tagging and the like, are difficult to be directly applied to a power grid fault handling plan text.
Entities in the plan text are usually formed by nesting a plurality of nouns such as regions, stations, voltage levels and the like, and the boundary of the entities is fuzzy; when the power grid fault handling knowledge graph is constructed, the characteristics are fully considered, a targeted text preprocessing method is adopted, a set of knowledge extraction model suitable for the power field is constructed, and the conversion of the pre-arranged plan text from the traditional knowledge extraction method based on manual and rule templates to the intelligent knowledge extraction method based on deep learning is realized.
In this embodiment, in S1.2, in the plan structure conversion, the specific method for constructing the plan knowledge graph includes the following steps S1.2.1: analyzing and selecting a construction mode of a knowledge graph; firstly, designing a mode layer of a knowledge graph in a top-down mode by analyzing the content of a plan text; then, under the guidance of the mode layer, a data layer is constructed in a bottom-up mode; designing a proper extraction method aiming at the characteristics of the pre-arranged plan text, extracting three knowledge elements of entities, relations and attributes to form a series of high-quality fact expressions, and mapping the fact expressions to related concept nodes in a bottom storage mode of a knowledge graph (the construction process is shown in figure 2);
s1.2.2, construction of a plan knowledge graph-mode layer: detailed analysis is carried out on the text content of the plan, and the concept type of the key indexes of the power grid regulation and control business, the related attributes and the relationship among the concepts are extracted, so that a domain knowledge system is formed;
the mode layer is a knowledge organization structure of a knowledge graph and is a data model for describing entities in the field, relationships among the entities and attributes; as shown in fig. 3, the mode layer of the grid fault handling knowledge graph is composed of a plurality of core elements such as fault names, system stability requirements, post-fault operation modes, fault handling key points, and the like, and the interrelation among the core elements;
s1.2.3, and the construction of a plan knowledge graph-data: describing real objects and relations of the power grid regulation and control information in a visual form by using graphic symbols, wherein basic composition units of the graphic symbols are triples of 'entity-relation-entity' and 'entity-attribute-property value', and the entities are mutually connected through relations to form a reticular knowledge structure;
the knowledge graph is a structured semantic knowledge base, fact data are extracted from the fused multi-source regulation and control information based on a knowledge graph technology, and a digital plan based on the knowledge graph is formed by combining an association algorithm; and the feature library construction needs to be continuously iterated, and each iteration comprises feature information extraction, knowledge fusion and knowledge processing.
Among them, in s1.2.1, the construction modes of the knowledge graph can be generally divided into 3 modes, i.e., top-down mode, bottom-up mode and mixed mode. The top-down construction mode is that firstly, a mode layer of the knowledge graph is constructed, a knowledge organization framework of the knowledge graph is predefined, and then, under the guidance of the mode layer, a series of knowledge extraction methods are used for extracting knowledge instances from a data source and adding the knowledge instances into a knowledge base. In contrast, the bottom-up construction mode does not define the mode layer of the knowledge graph in advance, but generalizes the mode layer from the data source, and gradually forms and continuously updates the organizational structure of the concept and the relationship between the concepts in the knowledge extraction process. The mixed construction mode combining the two modes is that a predefined mode layer is arranged in the initial stage, and meanwhile, along with the extraction of knowledge, the mode layer is improved and updated according to the knowledge organization structure of the data source, so that a knowledge organization structure with higher confidence coefficient is formed, and the knowledge map is more complete and reliable.
However, since the core constituent elements of the protocol text are relatively fixed, the core elements may be further subdivided into various types of unstructured information. Therefore, in the present embodiment, a top-down and bottom-up combined method is adopted to construct the grid fault handling knowledge graph.
Further, as shown in fig. 4, S1.2.3, the basic method of project knowledge map-data construction includes the following steps:
(1) Information extraction: the method is characterized in that bidirectional, parallel and unsupervised training is carried out on text corpora to realize word vector conversion and semantic content extraction, namely, entities, relations and entity attributes are extracted from semi-structured and non-structured data such as alarm information clauses, fault handling plans, expert experiences, overhaul ticket work contents, fixed value items in fixed value lists and the like and converted into structured information, and ontology knowledge expression is formed on the basis, and related key technologies comprise: entity extraction, relationship extraction and attribute extraction. The names of equipment such as a transformer substation, a main transformer, a bus, a circuit breaker and the like are entities, and the relationship is one, operation, occurrence and the like;
(2) And (3) knowledge fusion: the relation between information units after information extraction is flat, hierarchy and logic are lacked, a large number of redundant and even wrong information fragments exist, and multi-source description information about the same entity or concept is fused through knowledge fusion, wherein the multi-source description information comprises entity link, entity disambiguation, reference resolution and knowledge merging.
The entity link is mainly used for correlating data extracted by information extraction; the entity combination mainly comprises the combination of structured data and data of an external knowledge base, a knowledge system can be uniformly constructed based on a CIM model structure, and if other specialties or special topic construction knowledge bases can be subjected to knowledge fusion in the future, such as other special topics of a regulation and control group, protection specialties, mode specialties, automation specialties and the like;
semantic disambiguation is to disambiguate concepts and eliminate redundant and erroneous concepts, thereby ensuring the quality of the feature library. For example, possible semantics of 'change' include 'transformer substation', 'transformer' and 'change', which belong to the word polysemy and need entity disambiguation, and 'xx change' in regulation and control information refers to '500 kVxx transformer substation'; if the number 1 main transformer and the number #1 main transformer of the same transformer substation refer to the same transformer equipment, wherein the number 1 and the number #1 belong to one object and multiple words and need to be subjected to common reference disambiguation; the variable in the term is closely related to the main in the term to form a characteristic term main transformer, and the bifurcations need to be referred back; the sequence number is represented by a #1 in the main transformer, the number 1 is removed, and the following steps are noted: the use of the other words above "number 1" is not affected.
(3) Knowledge processing: the integrated and fused fact data do not form a structured and networked power grid regulation and control information knowledge system, and need knowledge processing, which mainly comprises ontology construction, quality evaluation and feature mining.
Specifically, the items of knowledge processing include:
constructing an ontology: the ontology is an explicit specification about a certain entity concept system, in the embodiment, an alarm event system based on main network regulation and control information can be regarded as the ontology, the ontology construction is a data set for describing the regulation and control information and is suitable for identification, fault handling and the like of a power grid alarm event, and the ontology is a framework of a feature library; and (3) training a correlation model of the body by adopting deep learning, such as a ternary combination of 'transformer substation-circuit breaker' and 'breaker-occurrence-tripping' and a correlation system for discovering and analyzing abnormal events, faults and other events and intelligently monitoring nodes, operations, contents and the like in the whole process.
Characteristic excavation: feature mining is needed to be carried out when the body is built, integrated global regulation and control information is needed when the knowledge map library is initially built, feature mining of an alarm event is needed, redundant information irrelevant to the event is removed, and a feature library for supporting event identification is formed; the main technical route is to utilize historical events, fault handling plans and expert experiences, deeply excavate the regulation and control information after association and fusion based on the topological relation of a power grid, and sort out types, basic attributes and the like; the types of the influence degree of the power grid are divided into five types: abnormity, accident, out-of-limit, displacement and notification, wherein the notification signal is prompt information which does not affect the normal operation of the system, scheduling operators put emphasis on other 4 types, and the accident and abnormal signals are important signals which need to be monitored in real time and immediately processed; the out-of-limit signal mainly reflects that the telemetering amount exceeds an upper limit interval and a lower limit interval of the alarm; displacement is important information needing real-time monitoring; and realizing a key characteristic model of the alarm information by adopting a multi-time-precision data association rule algorithm, wherein the key characteristic model comprises characteristic behaviors, characteristic factors and characteristic relations.
And (3) quality evaluation: and then, quantifying the credibility of the alarm event feature library information by adopting a confidence coefficient propagation algorithm, discarding lower confidence coefficients such as entity attribute deviation and logic between relations which do not accord with objective facts, and the like, and ensuring the quality of the feature library.
In this embodiment, in S2, the specific method for fusion modeling includes the following steps:
s2.1, generating a plan knowledge map model: the method comprises the steps of forming a knowledge graph model by performing structural conversion on a fault handling plan, and constructing a relation model according to power grid regulation and control services;
s2.2, constructing a knowledge graph model based on the CIM: the knowledge graph of the power grid basic model is used as a supporting model and can be used for converting data into the knowledge graph;
s2.3, forming a fused knowledge graph model: forming a structured plan knowledge graph through the structured conversion of the fault handling plan; carrying out association fusion on the plan structured model and the power grid model to form a plan digital model based on the knowledge graph;
and S2.4, considering the digital logic relation of the operation mode.
The knowledge graph is a knowledge representation method, graph data are used for storage, the knowledge graph is essentially a structured semantic knowledge base, modeling is carried out in the form of triples, and the triples are connected with each other through common entities or attributes to form a network knowledge structure. Compared with the traditional knowledge organization and management mode, the data organization structure of the knowledge graph based on the graph supports more efficient data retrieval, can process complex and various associated expressions, and can simulate the human thinking process to carry out semantic analysis. Therefore, by constructing a domain knowledge graph of power grid fault handling, fault handling knowledge is organized and stored in a graph form and is used as a carrier for understanding human knowledge by a machine, semantic search and auxiliary decision making are carried out by using a computer, and intelligent information service and application are provided for fault handling work.
Further, in S2.2, after the knowledge graph model based on CIM is analyzed by the power CIM model, the data and relationships of the devices such as the bus, the plant, the transformer, the line, the disconnecting switch, and the grounding switch can be obtained.
The knowledge graph of the digitalized plan relates to the fusion with the analysis of power grid equipment and power grid topology, so that the knowledge graph of a power grid basic model is required to be supported, namely a knowledge graph model based on CIM is required to be constructed.
As shown in fig. 6, the power grid topology is constructed based on the CIM model, and the power grid topology can be converted into the knowledge graph.
Further, in S2.3, in the fused knowledge graph model, the power grid base model mainly includes: the method comprises the following steps of (1) information such as a CIM (common information model), a power grid topological graph, alarm event relations, users, operators, a scheduling mechanism, operation risks, overhaul, fixed values and the like; the power grid model related to the plan association further comprises: grid equipment, historical event libraries, expert experience, and the like.
Further, in S2.4, the digital logical relationship of the operation modes is considered, that is: after the digital conversion of the fault handling plan is realized, modeling needs to be carried out on the digital logic relation considering the operation mode so as to conveniently carry out the safety analysis of the fault handling strategy by combining the current operation mode; after the file is analyzed, an ontology model is created, and the analyzed content can be independently objectified. The content is generally the combination of the change description of the multi-section equipment operation mode, and the normal mode mainly describes equipment outage, accompanying outage and the like. The main forms include equipment name outage, power transformation name, line name with full load, substation power failure, equipment name to running state, etc.
As shown in fig. 7, the present embodiment also provides an operation platform device of the method for building a digital fault handling plan model based on a knowledge graph, which includes a processor, a memory, and a computer program stored in the memory and running on the processor.
The processor comprises one or more processing cores, the processor is connected with the memory through the bus, the memory is used for storing program instructions, and the steps of the fault handling plan digital modeling method based on the knowledge map are realized when the processor executes the program instructions in the memory.
Alternatively, the memory may be implemented by any type or combination of volatile and non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
In addition, the present invention also provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps of the above-mentioned method for establishing the digital model of the fault handling plan based on the knowledge-graph.
Optionally, the present invention also provides a computer program product containing instructions which, when run on a computer, cause the computer to perform the steps of the above-described aspects of the method for knowledge-graph based fault handling plan digital modeling.
It will be understood by those skilled in the art that the processes for implementing all or part of the steps of the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and the preferred embodiments of the present invention are described in the above embodiments and the description, and are not intended to limit the present invention. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (9)
1. The fault handling plan digital model building method based on the knowledge graph is characterized by comprising the following steps: the method comprises the following steps:
s1, structure conversion: firstly, extracting key information in a plan, classifying and combining the key information according to business requirements to obtain a key information structural relationship, then performing semantic analysis and learning training by adopting artificial intelligence technologies such as a knowledge map, natural language processing, deep learning and the like, converting unstructured text information of the plan into structured information, and storing the structured information according to the key information structural relationship;
s2, fusion modeling: and performing association fusion on the structured plan information and the power grid model, and modeling based on the association relation to form a digital fault handling plan.
2. The method of knowledge-graph-based fault handling plan digital modeling according to claim 1, wherein: in S1, the specific method of structure conversion includes the following steps:
s1.1, analyzing a plan text: analyzing faults which may occur in the operation of a power grid, forming a power grid fault disposal plan in a text format according to a preset plan template style, and performing text analysis according to the characteristics of a plan text;
s1.2, converting a plan structure: and converting the unstructured plan into structured information which can be identified by a computer, and constructing a plan knowledge graph according to the structured information.
3. The method of knowledge-graph-based fault handling plan digital modeling according to claim 2, wherein: in the S1.1, a plan text consists of five contents, namely a plan number, a fault name, a stability requirement, a fault operation mode and a fault handling key point; wherein: the stability requirement is used to describe stability regulations to which the fault handling work needs to comply; the operation mode after the fault is used for describing the change condition and reason of the operation mode of the power grid after the expected fault occurs; the main points of fault handling are used for describing handling measures and emergency state adjustment required by the power grid to recover normal operation.
4. The method of knowledge-graph-based fault handling plan digital modeling according to claim 2, wherein: in the step S1.2, in the plan structure conversion, the specific method for constructing the plan knowledge graph includes the following steps
S1.2.1, constructing a plan knowledge graph-framework: analyzing and selecting a construction mode of a knowledge graph; firstly, designing a mode layer of a knowledge graph in a top-down mode by analyzing the content of a plan text; then, under the guidance of the mode layer, a data layer is constructed in a bottom-up mode; designing a proper extraction method aiming at the characteristics of the pre-arranged plan text, extracting three knowledge elements of entities, relations and attributes to form a series of high-quality fact expressions, and mapping the fact expressions to related concept nodes in a bottom storage mode of a knowledge graph;
s1.2.2, construction of a plan knowledge graph-mode layer: detailed analysis is carried out on the text content of the pre-arranged plan, and the concept type of the key indexes of the power grid regulation and control service, the related attributes and the relationship among the concepts are extracted, so that a domain knowledge system is formed;
s1.2.3, and construction of a predetermined case knowledge graph-data: the graphic symbols are used for describing the real objects and the relations of the power grid regulation and control information in a visual form, the basic composition units of the graphic symbols are triples of 'entity-relation-entity' and 'entity-attribute-property value', and the entities are mutually connected through the relations to form a reticular knowledge structure.
5. The method of knowledge-graph-based fault handling plan digital modeling according to claim 4, wherein: in S1.2.3, the basic method for constructing the plan knowledge map-data comprises the following steps:
information extraction: bidirectional, parallel and unsupervised training is carried out on the text corpus to realize word vector conversion and semantic content extraction, and ontology knowledge expression is formed on the basis;
and (3) knowledge fusion: fusing multi-source description information about the same entity or concept through knowledge fusion, wherein the multi-source description information comprises entity link, entity disambiguation, reference resolution and knowledge merging;
knowledge processing: the integrated and fused fact data does not form a structured and networked power grid regulation and control information knowledge system, and knowledge processing is required, and the method mainly comprises body construction, quality evaluation and feature mining.
6. The method of knowledge-graph-based fault handling plan digital modeling according to claim 1, wherein: in S2, the specific method for fusion modeling comprises the following steps:
s2.1, generating a plan knowledge map model: the method comprises the steps of forming a knowledge graph model by performing structural conversion on a fault handling plan, and constructing a relation model according to power grid regulation and control services;
s2.2, constructing a knowledge graph model based on the CIM: the knowledge graph of the power grid basic model is used as a supporting model and can be used for converting data into the knowledge graph;
s2.3, forming a fused knowledge graph model: forming a structured plan knowledge graph through the structured conversion of the fault handling plan; performing association fusion on the plan structured model and the power grid model to form a plan digital model based on the knowledge graph;
and S2.4, considering the digital logic relation of the operation mode.
7. The method of knowledge-graph-based fault handling plan digital modeling according to claim 6, wherein: in S2.2, in the knowledge graph model based on CIM, after the analysis by the power CIM model, the data and relationships of the devices such as the bus, the plant, the transformer, the line, the disconnecting switch, and the grounding switch can be obtained.
8. The method of knowledge-graph-based fault handling plan digital modeling according to claim 6, wherein: in S2.3, in the fused knowledge graph model, the power grid base model mainly includes: the method comprises the following steps of (1) information such as a CIM (common information model), a power grid topological graph, alarm event relations, users, operators, a scheduling mechanism, operation risks, overhaul, fixed values and the like; the power grid model related to the plan association further comprises: grid equipment, historical event libraries, expert experience, and the like.
9. The method of knowledge-graph-based fault handling plan digital modeling according to claim 6, wherein: in S2.4, considering the digital logic relationship of the operation mode: after the digital conversion of the fault handling plan is realized, modeling needs to be carried out on the digital logic relation considering the operation mode so as to conveniently carry out the safety analysis of the fault handling strategy by combining the current operation mode; after the file is analyzed, an ontology model is created, and the analyzed content can be independently objectified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210973549.3A CN115357726A (en) | 2022-08-15 | 2022-08-15 | Fault disposal plan digital model establishing method based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210973549.3A CN115357726A (en) | 2022-08-15 | 2022-08-15 | Fault disposal plan digital model establishing method based on knowledge graph |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115357726A true CN115357726A (en) | 2022-11-18 |
Family
ID=84033010
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210973549.3A Pending CN115357726A (en) | 2022-08-15 | 2022-08-15 | Fault disposal plan digital model establishing method based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115357726A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115953146A (en) * | 2022-12-07 | 2023-04-11 | 国家电网有限公司 | Power distribution network fault processing aid decision system |
| CN116521700A (en) * | 2023-06-28 | 2023-08-01 | 广州泓盈信息科技有限公司 | Updating method, system and storage medium for maintenance information base of electric power system |
| CN116611523A (en) * | 2023-07-18 | 2023-08-18 | 山东师范大学 | Method and system for predicting interpretable faults of turbofan engine |
| CN116644810A (en) * | 2023-05-06 | 2023-08-25 | 国网冀北电力有限公司信息通信分公司 | Power grid fault risk treatment method and device based on knowledge graph |
| CN116662559A (en) * | 2023-02-15 | 2023-08-29 | 北京北明数科信息技术有限公司 | Case knowledge graph construction platform and method based on big data technology |
| CN116774569A (en) * | 2023-07-25 | 2023-09-19 | 博纯材料股份有限公司 | Artificial intelligence-based method and system for updating operation system of oxygen-argon separation equipment |
| CN117077780A (en) * | 2023-10-19 | 2023-11-17 | 国网四川省电力公司信息通信公司 | Evaluation method for optimizing communication private network faults based on particle swarm optimization and knowledge graph |
-
2022
- 2022-08-15 CN CN202210973549.3A patent/CN115357726A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115953146A (en) * | 2022-12-07 | 2023-04-11 | 国家电网有限公司 | Power distribution network fault processing aid decision system |
| 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 |
| CN116662559A (en) * | 2023-02-15 | 2023-08-29 | 北京北明数科信息技术有限公司 | Case knowledge graph construction platform and method based on big data technology |
| CN116644810A (en) * | 2023-05-06 | 2023-08-25 | 国网冀北电力有限公司信息通信分公司 | Power grid fault risk treatment method and device based on knowledge graph |
| CN116644810B (en) * | 2023-05-06 | 2024-04-05 | 国网冀北电力有限公司信息通信分公司 | Power grid fault risk treatment method and device based on knowledge graph |
| CN116521700A (en) * | 2023-06-28 | 2023-08-01 | 广州泓盈信息科技有限公司 | Updating method, system and storage medium for maintenance information base of electric power system |
| CN116521700B (en) * | 2023-06-28 | 2023-09-08 | 广州泓盈信息科技有限公司 | Updating method, system and storage medium for maintenance information base of electric power system |
| CN116611523A (en) * | 2023-07-18 | 2023-08-18 | 山东师范大学 | Method and system for predicting interpretable faults of turbofan engine |
| CN116611523B (en) * | 2023-07-18 | 2023-09-26 | 山东师范大学 | Method and system for predicting interpretable faults of turbofan engine |
| CN116774569A (en) * | 2023-07-25 | 2023-09-19 | 博纯材料股份有限公司 | Artificial intelligence-based method and system for updating operation system of oxygen-argon separation equipment |
| CN116774569B (en) * | 2023-07-25 | 2024-04-05 | 博纯材料股份有限公司 | Artificial intelligence-based method and system for updating operation system of oxygen-argon separation equipment |
| CN117077780A (en) * | 2023-10-19 | 2023-11-17 | 国网四川省电力公司信息通信公司 | Evaluation method for optimizing communication private network faults based on particle swarm optimization and knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115357726A (en) | Fault disposal plan digital model establishing method based on knowledge graph | |
| CN112612902B (en) | Knowledge graph construction method and device for power grid main equipment | |
| CN111860882B (en) | Method and device for constructing power grid dispatching fault processing knowledge graph | |
| CN111985653B (en) | Power grid fault knowledge recommendation and knowledge management system and method based on knowledge graph | |
| CN113553420A (en) | Power grid fault handling regulation recommendation method and system based on knowledge graph | |
| CN112434532B (en) | Power grid environment model supporting man-machine bidirectional understanding and modeling method | |
| CN115293507A (en) | Situation knowledge graph construction method and system for power grid dispatching operation management | |
| CN110489562A (en) | A kind of dispatching of power netwoks regulation regulation knowledge modeling method and system based on ontology | |
| CN114077674A (en) | Power grid dispatching knowledge graph data optimization method and system | |
| CN113095050A (en) | Intelligent ticketing method, system, equipment and storage medium | |
| CN111832977A (en) | Maintenance application automatic ticketing method based on natural language parsing | |
| CN113761927B (en) | Power grid fault handling real-time auxiliary decision-making method, system, equipment and storage medium | |
| Tang et al. | Automatic schema construction of electrical graph data platform based on multi-source relational data models | |
| Jiang et al. | Construction of substation engineering design knowledge graph based on “ontology seven-step method” | |
| CN110086166B (en) | Method and system for representing stable operation quota of power grid | |
| Xiao et al. | Research on the construction and implementation of power grid fault handling knowledge graphs | |
| CN115033705A (en) | Power grid regulation and control risk early warning information knowledge graph design method and system | |
| CN115358369A (en) | Monitoring alarm event identification method based on convolutional neural network model | |
| Shan et al. | Research on deep learning based dispatching fault disposal robot technology | |
| CN115239153A (en) | Emergency command decision-making method, system and medium for chemical industry park accidents | |
| CN112860872B (en) | Power distribution network operation ticket semantic compliance verification method and system based on self-learning | |
| CN115481964A (en) | Intelligent ticket forming method and device | |
| Wang et al. | Ontology-based knowledge modelling of power grid dispatching regulations | |
| Yang et al. | Knowledge Modeling of power grid regulation based on reasoning map | |
| Wen et al. | An intelligent analysis method of security and stability control strategy based on the knowledge graph |
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 |