CN113268604A - Self-adaptive expansion method and system for knowledge base - Google Patents
Self-adaptive expansion method and system for knowledge base Download PDFInfo
- Publication number
- CN113268604A CN113268604A CN202110545292.7A CN202110545292A CN113268604A CN 113268604 A CN113268604 A CN 113268604A CN 202110545292 A CN202110545292 A CN 202110545292A CN 113268604 A CN113268604 A CN 113268604A
- Authority
- CN
- China
- Prior art keywords
- fault
- knowledge
- power grid
- regulation
- knowledge base
- 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 37
- 238000000605 extraction Methods 0.000 claims abstract description 31
- 238000005516 engineering process Methods 0.000 claims abstract description 30
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 27
- 230000001105 regulatory effect Effects 0.000 claims abstract description 25
- 238000003745 diagnosis Methods 0.000 claims abstract description 24
- 238000003058 natural language processing Methods 0.000 claims abstract description 22
- 239000013598 vector Substances 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims description 28
- 238000012423 maintenance Methods 0.000 claims description 17
- 230000001276 controlling effect Effects 0.000 claims description 16
- 230000003044 adaptive effect Effects 0.000 claims description 15
- 238000012384 transportation and delivery Methods 0.000 claims description 9
- 230000001364 causal effect Effects 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 6
- 238000007670 refining Methods 0.000 claims description 5
- 230000007547 defect Effects 0.000 claims description 4
- 230000002457 bidirectional effect Effects 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 230000007787 long-term memory Effects 0.000 claims description 3
- 230000000737 periodic effect Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000006403 short-term memory Effects 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 abstract description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 7
- 230000009286 beneficial effect Effects 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 239000002184 metal Substances 0.000 description 4
- 239000002904 solvent Substances 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 210000004556 brain Anatomy 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000015654 memory Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000001052 transient effect 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
Abstract
The self-adaptive expansion method and system of the knowledge base comprise the following steps: collecting a regulation and control plan, a power grid operation record, an overhaul plan and a regulation and control operation instruction book, and processing a regulation and control corpus based on a natural language processing algorithm to generate a regulation and control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors; performing off-line expansion on the knowledge base by using a regulated corpus; according to the online diagnosis result of the system-level fault of the power grid, fault disposal knowledge identification is carried out on the online fault disposal knowledge map by a fault disposal knowledge extraction model; and performing online expansion on the knowledge base according to the recognition result. The method has the advantages that the fault handling knowledge of the offline knowledge base is refined and enriched, the fault handling knowledge base is dynamically updated on line, the problem that the offline established regulation knowledge base cannot cover real-time faults is solved, the regulation knowledge base is adaptively expanded and updated through an online knowledge rolling technology, and the regulation work efficiency and the standardization degree are improved.
Description
Technical Field
The invention relates to the technical field of power grid management control, in particular to a method and a system for adaptively expanding a knowledge base.
Background
China highly attaches importance to the development of artificial intelligence industry, and the level of integration and innovation of artificial intelligence and various fields of various industries in the economic society is continuously improved. With the construction and the improvement of new technologies such as a regulation cloud platform and the like, the data volume and the computing capacity of the power system are greatly improved recently, and favorable conditions are created for developing the technical research of the artificial intelligent robot in the field of power regulation.
With the increasingly obvious integration characteristics of large power grids, the coupling relationship between different elements in the system and the external environment is continuously enhanced, and a series of deep development and changes of the power system and the regulation and control service enable the complexity of the scheduling control strategy and the regulation and control rule to be continuously improved, so that higher requirements are provided for the automation and the intellectualization of the regulation and control service. The power dispatching control center is a 'command brain' integrating high-value data, analysis rules, expert experience and calculation decisions, the existing regulation and control mode mainly takes manual experience analysis as a main mode, a dispatcher needs to perform experience knowledge correlation on massive and diverse data and scheme models, more repetitive 'human brain labor' is needed, and the efficiency is low, so that intelligent regulation and control are realized, and the working intensity of the regulator is reduced.
In the prior art, the construction of a multistage scheduling fault co-processing knowledge base requires extracting knowledge such as fault processing key points, processing logic, processing steps and the like from information such as power grid operation information, scheduling rules and priori knowledge. Considering that the power grid operation information is historical data, the plan is compiled based on a typical operation mode, the experience of a dispatcher needs to be specifically analyzed according to problems, and a fault handling knowledge base established by applying the data is difficult to cover all fault handling situations, so how to expand the knowledge base on line to cover the real-time fault handling situations is a difficult point and a key for regulating and controlling the technical development of the knowledge base in the field. In addition, the technical aspect of the knowledge base in the control field still has the following problems: the regulation knowledge base relies on an offline recognition and expansion technology on the basis of historical data, and the expansion efficiency, the recognition accuracy and the intelligent degree are all to be improved; in addition, fault handling based on the existing regulation knowledge base is mainly completed by depending on the experience of a regulator, and the fault handling method is not only lack of standardized, flow-based and intelligent methods, systems, media and equipment, low in handling efficiency and incapable of covering the problem of real-time faults.
Disclosure of Invention
In order to solve the defects in the prior art, the invention aims to provide a knowledge base self-adaptive expansion method and a knowledge base self-adaptive expansion system, which are used for realizing the automatic generation of a regulation knowledge base by using an artificial intelligence method and eliminating the influence caused by manual operation; the real-time analysis of scheduling services and the online identification of fault handling decisions are realized, the problem that the regulation and control knowledge base established offline cannot cover real-time faults is solved, and the regulation and control knowledge base is adaptively expanded and updated through an online knowledge rolling technology.
The invention adopts the following technical scheme.
The self-adaptive expansion method of the knowledge base comprises the following steps:
step 1, collecting a regulation and control plan, a power grid operation record, a maintenance plan and a regulation and control operation instruction book, and processing a regulation and control corpus based on a natural language processing algorithm to generate a regulation and control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors;
step 2, extracting characteristic data from a regulated corpus by a natural language processing technology to form a knowledge graph, and performing off-line expansion on a knowledge base by using the knowledge graph;
step 3, collecting power grid fault off-line simulation data and power grid fault historical operation data in real time, and constructing a power grid system level fault simple event set; based on a complex event processing algorithm, a power grid system-level fault simple event set is utilized to carry out online diagnosis on the power grid system-level fault; constructing an online fault handling knowledge graph according to an online diagnosis result of the system-level fault of the power grid;
step 4, identifying fault disposal knowledge of the online fault disposal knowledge map by using a fault disposal knowledge extraction model; the knowledge base is expanded online with the identified fault handling knowledge.
Preferably, the first and second electrodes are formed of a metal,
in the step 1, the regulation and control plan, the power grid operation record, the maintenance plan and the regulation and control operation instruction book are text data and are obtained by searching a regulation and control intranet according to a search request in a triggering or periodic mode.
The step 1 comprises the following steps:
step 1.1, taking the collected text data as input, performing word segmentation and part-of-speech primary labeling processing, and outputting regulating and controlling corpus words;
step 1.2, using the regulating and controlling linguistic data words as input, and expressing the correlation relation of the regulating and controlling linguistic data words and the function of the regulating and controlling linguistic data words in the sentence by using a nested tree structure;
and 1.3, converting the regulating and controlling linguistic data words from symbolic information in a text form into digital information in a vector form based on the distributed representation model.
Preferably, the first and second electrodes are formed of a metal,
the step 2 comprises the following steps:
step 2.1, extracting historical fault data and power grid fault characteristic information from a power grid operation record through a natural language processing technology, and constructing a knowledge graph of a circuit trial delivery auxiliary decision by using a knowledge representation method;
step 2.2, extracting a line trial delivery strategy from a regulation plan and a regulation operation instruction book through a natural language processing technology, and constructing a fault disposal knowledge graph on the basis of a knowledge graph of the line trial delivery auxiliary decision;
step 2.3, extracting whether live working, maintenance operation and body defects exist in the fault equipment from the maintenance plan through a natural language processing technology, and expanding the fault disposal knowledge graph to form a fault decision knowledge graph;
and 2.4, performing off-line expansion on the knowledge base by using the fault decision knowledge graph.
Preferably, the first and second electrodes are formed of a metal,
in step 3, a time sequence constraint relation exists among all events in the simple event set; designing a power grid system fault trigger based on a complex event processing algorithm, so that a set of all simple events which can be matched by the trigger forms a power grid system level fault causal relationship;
and determining a power grid system-level fault online diagnosis result according to the causal relationship, and constructing an online fault disposal knowledge graph.
Preferably, the first and second electrodes are formed of a metal,
in step 4, the fault handling knowledge comprises: fault handling key points and fault handling relations; the failure handling key points are failure handling entities, and the failure handling relationship is a logical relationship between the failure handling key points.
Step 4 comprises the following steps:
step 4.1, for a single fault object, acquiring an equipment list and an event list which are electrically connected with a power grid where the fault object is located; taking an equipment list as a fault handling key point training set and taking an event list as a fault handling relation training set;
step 4.2, performing primary training on the fault handling knowledge extraction model by using the fault handling key point training set based on the bidirectional long and short term memory network-conditional random field algorithm, and outputting a recognition result of the fault handling key point by the fault handling knowledge extraction model obtained by the primary training;
4.3, performing secondary training on the fault handling knowledge extraction model after the primary training by using the fault handling relationship training set based on the convolutional neural network algorithm, and outputting a recognition result of the fault handling relationship by using the fault handling knowledge extraction model obtained by the secondary training;
step 4.4, identifying the online fault handling knowledge map by using the trained fault handling knowledge extraction model, and fusing the fault handling key points and the fault handling relation into fault handling knowledge by taking the identified fault handling key points as features based on a feature level information fusion algorithm;
and 4.5, carrying out online expansion on the knowledge base subjected to offline expansion by using the fault handling knowledge.
The knowledge base self-adaptive expansion system comprises:
the off-line knowledge base generation module is used for processing the control corpus based on a natural language processing algorithm by using a control plan, a power grid operation record, an overhaul plan and a control operation instruction book collected by external equipment to generate a control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors; performing off-line expansion on the knowledge base by using a regulated corpus;
the online knowledge base generation module is used for acquiring an online fault handling knowledge map in real time from a power grid system-level fault online diagnosis system by using external equipment; identifying fault disposal knowledge for the online fault disposal knowledge map by using a fault disposal knowledge extraction model; performing online expansion on the knowledge base by using the identified fault handling knowledge;
in the online knowledge base generation module, a power grid system-level fault online diagnosis system is used for acquiring power grid fault offline simulation data and power grid fault historical operation data in real time and constructing a power grid system-level fault simple event set; based on a complex event processing algorithm, a power grid system-level fault simple event set is utilized to carry out online diagnosis on the power grid system-level fault; and constructing an online fault handling knowledge graph according to the online diagnosis result of the system-level fault of the power grid.
And the online knowledge base generation module is used for identifying fault handling key points and logic from the power grid operation record log, the maintenance plan and the external environment information by using the fault handling knowledge extraction model, and refining and dynamically updating the fault handling knowledge of the offline knowledge base on line.
Compared with the prior art, the method has the beneficial effects that a knowledge base self-adaptive expansion technology based on online knowledge rolling extraction is provided for solving the problem that the knowledge base established offline is difficult to completely cover the real-time fault handling condition. Identifying fault handling key points and logic from information such as a power grid operation record log, a maintenance plan and an external environment by establishing a fault handling knowledge rolling extraction model, refining and enriching fault handling knowledge of an offline knowledge base, and realizing online dynamic update of the fault handling knowledge base; the regulation and control knowledge base is adaptively expanded and updated through an online knowledge rolling technology, and the regulation and control working efficiency and the standardization degree are also improved.
The method has the beneficial effects that the method for adaptively expanding the knowledge base is applied to the construction of the knowledge base for the multi-stage scheduling fault cooperative disposal, the power grid operation service data, the scheduling fault disposal principle and experience are combined, the knowledge base in the regulation and control field is constructed on the basis of the knowledge map technology, and efficient data support is provided for information query, decision reasoning and the like of power grid fault disposal; the method comprises the following steps of researching a knowledge model construction technology for single equipment fault disposal, taking fault equipment as a center, triggering and fusing knowledge base association knowledge, a power grid fault analysis result and the like on line, constructing an equipment fault disposal knowledge model, and providing knowledge guidance for equipment fault disposal; the single equipment fault-oriented treatment guiding technology is researched, fault treatment processes based on a regulation knowledge base are considered, and the severity of fault influence is considered, so that automatic guidance of a dispatcher to perform real-time rapid treatment of faults according to priorities is achieved.
Drawings
FIG. 1 is a flow chart diagram of the knowledge base adaptive expansion method of the present invention.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
In the preferred embodiment of the invention, an extra-high voltage alternating current-direct current hybrid power grid is taken as a research object, a multi-stage scheduling fault co-disposal knowledge base suitable for a power grid system needs to be constructed, and the construction of the knowledge base extracts knowledge such as fault disposal key points, disposal logic, disposal steps and the like from information such as power grid operation information, scheduling regulations and priori knowledge.
As shown in fig. 1, the method for adaptive expansion of knowledge base includes:
step 1, collecting a regulation and control plan, a power grid operation record, a maintenance plan and a regulation and control operation instruction book, and processing a regulation and control corpus based on a natural language processing algorithm to generate a regulation and control corpus word vector; and constructing a regulation corpus by utilizing the regulation corpus word vectors.
In particular, the amount of the solvent to be used,
in the step 1, the regulation and control plan, the power grid operation record, the maintenance plan and the regulation and control operation instruction book are text data and are obtained by searching a regulation and control intranet according to a search request in a triggering or periodic mode.
The step 1 comprises the following steps:
step 1.1, taking the collected text data as input, performing word segmentation and part-of-speech primary labeling processing, and outputting regulating and controlling linguistic data words.
And step 1.2, using the regulating and controlling linguistic words as input, and expressing the correlation relation of the regulating and controlling linguistic words and the function of the regulating and controlling linguistic words in the sentence by using the nested tree structure.
And 1.3, converting the regulating and controlling linguistic data words from symbolic information in a text form into digital information in a vector form based on the distributed representation model.
In the aspect of knowledge guidance, a natural language processing technology is applied to extract knowledge from text materials such as scheduling rules, fault plans and operation instruction books, and the efficiency and the standard degree of regulation and control work are improved through the application of an artificial intelligence technology.
And 2, extracting characteristic data from the regulated corpus by a natural language processing technology to form a knowledge graph, and performing off-line expansion on the knowledge base by using the knowledge graph.
In particular, the amount of the solvent to be used,
the step 2 comprises the following steps:
and 2.1, extracting historical fault data and power grid fault characteristic information from the power grid operation records through a natural language processing technology, and constructing a knowledge graph of the auxiliary decision of line trial delivery by using a knowledge representation method.
And 2.2, extracting a line trial delivery strategy from the regulation plan and the regulation operation instruction book through a natural language processing technology, and constructing a fault disposal knowledge graph on the basis of the knowledge graph of the line trial delivery auxiliary decision.
And 2.3, extracting whether live working, maintenance operation and body defects exist in the fault equipment from the maintenance plan through a natural language processing technology, and expanding the fault disposal knowledge graph to form a fault decision knowledge graph.
And 2.4, performing off-line expansion on the knowledge base by using the fault decision knowledge graph.
Step 3, collecting power grid fault off-line simulation data and power grid fault historical operation data in real time, and constructing a power grid system level fault simple event set; based on a complex event processing algorithm, a power grid system-level fault simple event set is utilized to carry out online diagnosis on the power grid system-level fault; and constructing an online fault handling knowledge graph according to the online diagnosis result of the system-level fault of the power grid.
In particular, the amount of the solvent to be used,
in step 3, a time sequence constraint relation exists among all events in the simple event set; designing a power grid system fault trigger based on a complex event processing algorithm, so that a set of all simple events which can be matched by the trigger forms a power grid system level fault causal relationship;
in the preferred embodiment, after the extra-high voltage direct current latch occurs, a large amount of event information of different types is generated by different control systems and different applications in a certain time window, incidence relations such as sequence, aggregation, dependence and cause and effect exist among events, the correlation of the events needs to be judged, the events are rapidly deduced to reflect the operation essence of a power grid, the scene characteristics have good matching degree with the characteristics of the CEP technology, and the technology is applied to diagnosis of faults of the extra-high voltage direct current latch and is beneficial to improving the analysis capability and processing speed of system-level faults.
And determining a power grid system-level fault online diagnosis result according to the causal relationship, and constructing an online fault disposal knowledge graph.
The operation characteristics of the extra-high voltage alternating current-direct current hybrid power grid present global characteristics, a single equipment fault easily causes a whole-grid chain reaction, the relation among all events relates to an electromagnetic and electromechanical transient process, the internal mechanism is very complex, the simple judgment and analysis are carried out only by the event time sequence, the wrong judgment is possible, meanwhile, because the events belong to small sample events, the training and learning cannot be carried out, how to intelligently identify the internal cause-effect relation among all the events is achieved, and then a system-level fault analysis result is given. And the coverage of various real-time faults is realized through the online diagnosis of the system-level faults of the power grid.
The knowledge graph has efficient reasoning capability, and can integrate prior knowledge of regulation and control rules, plans, experiences, operation instruction books and the like into available knowledge to form the power grid fault handling knowledge graph for guiding fault handling.
Step 4, identifying fault disposal knowledge of the online fault disposal knowledge map by using a fault disposal knowledge extraction model; the knowledge base is expanded online with the identified fault handling knowledge.
In particular, the amount of the solvent to be used,
in step 4, the fault handling knowledge comprises: fault handling key points and fault handling relations; the failure handling key points are failure handling entities, and the failure handling relationship is a logical relationship between the failure handling key points.
Step 4 comprises the following steps:
step 4.1, for a single fault object, acquiring an equipment list and an event list which are electrically connected with a power grid where the fault object is located; taking an equipment list as a fault handling key point training set and taking an event list as a fault handling relation training set;
step 4.2, performing primary training on the fault handling knowledge extraction model by using the fault handling key point training set based on the bidirectional long and short term memory network-conditional random field algorithm, and outputting a recognition result of the fault handling key point by the fault handling knowledge extraction model obtained by the primary training;
4.3, performing secondary training on the fault handling knowledge extraction model after the primary training by using the fault handling relationship training set based on the convolutional neural network algorithm, and outputting a recognition result of the fault handling relationship by using the fault handling knowledge extraction model obtained by the secondary training;
step 4.4, identifying the online fault handling knowledge map by using the trained fault handling knowledge extraction model, and fusing the fault handling key points and the fault handling relation into fault handling knowledge by taking the identified fault handling key points as features based on a feature level information fusion algorithm;
and 4.5, carrying out online expansion on the knowledge base subjected to offline expansion by using the fault handling knowledge.
A knowledge base adaptive expansion system implementing a knowledge base adaptive expansion method, comprising:
the off-line knowledge base generation module is used for processing the control corpus based on a natural language processing algorithm by using a control plan, a power grid operation record, an overhaul plan and a control operation instruction book collected by external equipment to generate a control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors; performing off-line expansion on the knowledge base by using a regulated corpus;
the online knowledge base generation module is used for acquiring an online fault handling knowledge map in real time from a power grid system-level fault online diagnosis system by using external equipment; identifying fault disposal knowledge for the online fault disposal knowledge map by using a fault disposal knowledge extraction model; performing online expansion on the knowledge base by using the identified fault handling knowledge;
in the online knowledge base generation module, a power grid system-level fault online diagnosis system is used for acquiring power grid fault offline simulation data and power grid fault historical operation data in real time and constructing a power grid system-level fault simple event set; based on a complex event processing algorithm, a power grid system-level fault simple event set is utilized to carry out online diagnosis on the power grid system-level fault; and constructing an online fault handling knowledge graph according to the online diagnosis result of the system-level fault of the power grid.
And the online knowledge base generation module is used for identifying fault handling key points and logic from the power grid operation record log, the maintenance plan and the external environment information by using the fault handling knowledge extraction model, and refining and dynamically updating the fault handling knowledge of the offline knowledge base on line.
A computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
processing a control corpus based on a natural language processing algorithm by using a control plan, a power grid operation record, an overhaul plan and a control operation instruction book collected by external equipment to generate a control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors; performing off-line expansion on the knowledge base by using a regulated corpus;
acquiring an online fault handling knowledge graph in real time by using external equipment; identifying fault disposal knowledge for the online fault disposal knowledge map by using a fault disposal knowledge extraction model; the knowledge base is expanded online with the identified fault handling knowledge.
A computer device comprising a memory and a processor; the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of:
processing a control corpus based on a natural language processing algorithm by using a control plan, a power grid operation record, an overhaul plan and a control operation instruction book collected by external equipment to generate a control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors; performing off-line expansion on the knowledge base by using a regulated corpus;
acquiring an online fault handling knowledge graph in real time by using external equipment; identifying fault disposal knowledge for the online fault disposal knowledge map by using a fault disposal knowledge extraction model; the knowledge base is expanded online with the identified fault handling knowledge.
An information data processing terminal is used for realizing a knowledge base self-adaptive expansion method.
Compared with the prior art, the method has the beneficial effects that a knowledge base self-adaptive expansion technology based on online knowledge rolling extraction is provided for solving the problem that the knowledge base established offline is difficult to completely cover the real-time fault handling condition. Identifying fault handling key points and logic from information such as a power grid operation record log, a maintenance plan and an external environment by establishing a fault handling knowledge rolling extraction model, refining and enriching fault handling knowledge of an offline knowledge base, and realizing online dynamic update of the fault handling knowledge base; the regulation and control knowledge base is adaptively expanded and updated through an online knowledge rolling technology, and the regulation and control working efficiency and the standardization degree are also improved.
The method has the beneficial effects that the method for adaptively expanding the knowledge base is applied to the construction of the knowledge base for the multi-stage scheduling fault cooperative disposal, the power grid operation service data, the scheduling fault disposal principle and experience are combined, the knowledge base in the regulation and control field is constructed on the basis of the knowledge map technology, and efficient data support is provided for information query, decision reasoning and the like of power grid fault disposal; the method comprises the following steps of researching a knowledge model construction technology for single equipment fault disposal, taking fault equipment as a center, triggering and fusing knowledge base association knowledge, a power grid fault analysis result and the like on line, constructing an equipment fault disposal knowledge model, and providing knowledge guidance for equipment fault disposal; the single equipment fault-oriented treatment guiding technology is researched, fault treatment processes based on a regulation knowledge base are considered, and the severity of fault influence is considered, so that automatic guidance of a dispatcher to perform real-time rapid treatment of faults according to priorities is achieved.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are merely preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for limiting the scope of the present invention, and on the contrary, any improvement or modification made based on the spirit of the present invention should fall within the scope of the present invention.
Claims (10)
1. The self-adaptive expansion method of the knowledge base is characterized in that,
the method comprises the following steps:
step 1, collecting a regulation and control plan, a power grid operation record, a maintenance plan and a regulation and control operation instruction book, and processing a regulation and control corpus based on a natural language processing algorithm to generate a regulation and control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors;
step 2, extracting characteristic data from a regulated corpus by a natural language processing technology to form a knowledge graph, and performing off-line expansion on a knowledge base by using the knowledge graph;
step 3, collecting power grid fault off-line simulation data and power grid fault historical operation data in real time, and constructing a power grid system level fault simple event set; based on a complex event processing algorithm, a power grid system-level fault simple event set is utilized to carry out online diagnosis on the power grid system-level fault; constructing an online fault handling knowledge graph according to an online diagnosis result of the system-level fault of the power grid;
step 4, identifying fault disposal knowledge of the online fault disposal knowledge map by using a fault disposal knowledge extraction model; the knowledge base is expanded online with the identified fault handling knowledge.
2. The method of adaptive expansion of a knowledge base according to claim 1,
in the step 1, the regulation and control plan, the power grid operation record, the maintenance plan and the regulation and control operation instruction book are text data and are obtained by searching a regulation and control intranet according to a search request in a triggering or periodic mode.
3. The method of adaptive expansion of a knowledge base according to claim 2,
the step 1 comprises the following steps:
step 1.1, taking the collected text data as input, performing word segmentation and part-of-speech primary labeling processing, and outputting regulating and controlling corpus words;
step 1.2, using the regulating and controlling linguistic data words as input, and expressing the correlation relation of the regulating and controlling linguistic data words and the function of the regulating and controlling linguistic data words in the sentence by using a nested tree structure;
and 1.3, converting the regulating and controlling linguistic data words from symbolic information in a text form into digital information in a vector form based on the distributed representation model.
4. The method of adaptive expansion of a knowledge base according to claim 1,
the step 2 comprises the following steps:
step 2.1, extracting historical fault data and power grid fault characteristic information from a power grid operation record through a natural language processing technology, and constructing a knowledge graph of a circuit trial delivery auxiliary decision by using a knowledge representation method;
step 2.2, extracting a line trial delivery strategy from a regulation plan and a regulation operation instruction book through a natural language processing technology, and constructing a fault disposal knowledge graph on the basis of a knowledge graph of the line trial delivery auxiliary decision;
step 2.3, extracting whether live working, maintenance operation and body defects exist in the fault equipment from the maintenance plan through a natural language processing technology, and expanding the fault disposal knowledge graph to form a fault decision knowledge graph;
and 2.4, performing off-line expansion on the knowledge base by using the fault decision knowledge graph.
5. The method of adaptive expansion of a knowledge base according to claim 1,
in step 3, a time sequence constraint relation exists among all events in the simple event set; designing a power grid system fault trigger based on a complex event processing algorithm, so that a set of all simple events which can be matched by the trigger forms a power grid system level fault causal relationship;
and determining a power grid system-level fault online diagnosis result according to the causal relationship, and constructing an online fault disposal knowledge graph.
6. The method of adaptive expansion of a knowledge base according to claim 1,
in step 4, the failure handling knowledge includes: fault handling key points and fault handling relations; wherein the failure handling main points are failure handling entities, and the failure handling relationship is a logical relationship between the failure handling main points.
7. The method of adaptive expansion of a knowledge base according to claim 6,
step 4 comprises the following steps:
step 4.1, for a single fault object, acquiring an equipment list and an event list which are electrically connected with a power grid where the fault object is located; taking an equipment list as a fault handling key point training set and taking an event list as a fault handling relation training set;
step 4.2, performing primary training on the fault handling knowledge extraction model by using the fault handling key point training set based on the bidirectional long and short term memory network-conditional random field algorithm, and outputting a recognition result of the fault handling key point by the fault handling knowledge extraction model obtained by the primary training;
4.3, performing secondary training on the fault handling knowledge extraction model after the primary training by using the fault handling relationship training set based on the convolutional neural network algorithm, and outputting a recognition result of the fault handling relationship by using the fault handling knowledge extraction model obtained by the secondary training;
step 4.4, identifying the online fault handling knowledge map by using the trained fault handling knowledge extraction model, and fusing the fault handling key points and the fault handling relation into fault handling knowledge by taking the identified fault handling key points as features based on a feature level information fusion algorithm;
and 4.5, carrying out online expansion on the knowledge base subjected to offline expansion by using the fault handling knowledge.
8. A knowledge base adaptive expansion system implemented by using the knowledge base adaptive expansion method of any one of claims 1 to 7, the knowledge base adaptive expansion system comprising:
the off-line knowledge base generation module is used for processing the control corpus based on a natural language processing algorithm by using a control plan, a power grid operation record, an overhaul plan and a control operation instruction book collected by external equipment to generate a control corpus word vector; constructing a regulation corpus by utilizing the regulation corpus word vectors; performing off-line expansion on the knowledge base by using a regulated corpus;
the online knowledge base generation module is used for acquiring an online fault handling knowledge map in real time from a power grid system-level fault online diagnosis system by using external equipment; identifying fault disposal knowledge for the online fault disposal knowledge map by using a fault disposal knowledge extraction model; the knowledge base is expanded online with the identified fault handling knowledge.
9. The system for adaptive expansion of knowledge bases according to claim 8,
in the online knowledge base generation module, the power grid system-level fault online diagnosis system is used for acquiring power grid fault offline simulation data and power grid fault historical operation data in real time and constructing a power grid system-level fault simple event set; based on a complex event processing algorithm, a power grid system-level fault simple event set is utilized to carry out online diagnosis on the power grid system-level fault; and constructing an online fault handling knowledge graph according to the online diagnosis result of the system-level fault of the power grid.
10. The system for adaptive expansion of knowledge bases according to claim 8,
and the online knowledge base generation module is used for identifying fault handling key points and logic from the power grid operation record log, the maintenance plan and the external environment information by using the fault handling knowledge extraction model, and refining and dynamically updating the fault handling knowledge of the offline knowledge base on line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110545292.7A CN113268604A (en) | 2021-05-19 | 2021-05-19 | Self-adaptive expansion method and system for knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110545292.7A CN113268604A (en) | 2021-05-19 | 2021-05-19 | Self-adaptive expansion method and system for knowledge base |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113268604A true CN113268604A (en) | 2021-08-17 |
Family
ID=77231801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110545292.7A Pending CN113268604A (en) | 2021-05-19 | 2021-05-19 | Self-adaptive expansion method and system for knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113268604A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113850017A (en) * | 2021-08-27 | 2021-12-28 | 国家电网有限公司 | System-level fault analysis system and method based on power flow change map |
| CN114430365A (en) * | 2022-04-06 | 2022-05-03 | 北京宝兰德软件股份有限公司 | Fault root cause analysis method and device, electronic equipment and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017076263A1 (en) * | 2015-11-03 | 2017-05-11 | 中兴通讯股份有限公司 | Method and device for integrating knowledge bases, knowledge base management system and storage medium |
| US20170286914A1 (en) * | 2016-04-05 | 2017-10-05 | Facebook, Inc. | Systems and methods to develop training set of data based on resume corpus |
| WO2018072563A1 (en) * | 2016-10-18 | 2018-04-26 | 中兴通讯股份有限公司 | Knowledge graph creation method, device, and system |
| CN108446367A (en) * | 2018-03-15 | 2018-08-24 | 湖南工业大学 | A kind of the packaging industry data search method and equipment of knowledge based collection of illustrative plates |
| CN109189946A (en) * | 2018-11-06 | 2019-01-11 | 湖南云智迅联科技发展有限公司 | A method of the description of equipment fault sentence is converted into knowledge mapping expression |
| CN110717018A (en) * | 2019-04-15 | 2020-01-21 | 中国石油大学(华东) | Industrial equipment fault maintenance question-answering system based on knowledge graph |
| CN110929149A (en) * | 2019-11-12 | 2020-03-27 | 广东工业大学 | Industrial equipment fault maintenance recommendation method and system |
| CN110968700A (en) * | 2019-11-01 | 2020-04-07 | 数地科技(北京)有限公司 | Domain event map construction method and device fusing multi-class affairs and entity knowledge |
| US20200118014A1 (en) * | 2018-10-13 | 2020-04-16 | Predii Inc. | Adaptable Systems and Methods for Discovering Intent from Enterprise Data |
| WO2020143186A1 (en) * | 2019-01-10 | 2020-07-16 | 平安科技（深圳）有限公司 | Recommendation system training method and apparatus, and computer device and storage medium |
| CN112269901A (en) * | 2020-09-14 | 2021-01-26 | 合肥中科类脑智能技术有限公司 | Fault distinguishing and reasoning method based on knowledge graph |
| CN112612902A (en) * | 2020-12-23 | 2021-04-06 | 国网浙江省电力有限公司电力科学研究院 | Knowledge graph construction method and device for power grid main device |
-
2021
- 2021-05-19 CN CN202110545292.7A patent/CN113268604A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017076263A1 (en) * | 2015-11-03 | 2017-05-11 | 中兴通讯股份有限公司 | Method and device for integrating knowledge bases, knowledge base management system and storage medium |
| US20170286914A1 (en) * | 2016-04-05 | 2017-10-05 | Facebook, Inc. | Systems and methods to develop training set of data based on resume corpus |
| WO2018072563A1 (en) * | 2016-10-18 | 2018-04-26 | 中兴通讯股份有限公司 | Knowledge graph creation method, device, and system |
| CN108446367A (en) * | 2018-03-15 | 2018-08-24 | 湖南工业大学 | A kind of the packaging industry data search method and equipment of knowledge based collection of illustrative plates |
| US20200118014A1 (en) * | 2018-10-13 | 2020-04-16 | Predii Inc. | Adaptable Systems and Methods for Discovering Intent from Enterprise Data |
| CN109189946A (en) * | 2018-11-06 | 2019-01-11 | 湖南云智迅联科技发展有限公司 | A method of the description of equipment fault sentence is converted into knowledge mapping expression |
| WO2020143186A1 (en) * | 2019-01-10 | 2020-07-16 | 平安科技（深圳）有限公司 | Recommendation system training method and apparatus, and computer device and storage medium |
| CN110717018A (en) * | 2019-04-15 | 2020-01-21 | 中国石油大学(华东) | Industrial equipment fault maintenance question-answering system based on knowledge graph |
| CN110968700A (en) * | 2019-11-01 | 2020-04-07 | 数地科技(北京)有限公司 | Domain event map construction method and device fusing multi-class affairs and entity knowledge |
| CN110929149A (en) * | 2019-11-12 | 2020-03-27 | 广东工业大学 | Industrial equipment fault maintenance recommendation method and system |
| CN112269901A (en) * | 2020-09-14 | 2021-01-26 | 合肥中科类脑智能技术有限公司 | Fault distinguishing and reasoning method based on knowledge graph |
| CN112612902A (en) * | 2020-12-23 | 2021-04-06 | 国网浙江省电力有限公司电力科学研究院 | Knowledge graph construction method and device for power grid main device |
Non-Patent Citations (2)
| Title |
|---|
| 仇建飞;: "打造智能互联网化的支撑客服系统的设计与实现", 电子技术与软件工程, no. 11, 1 June 2020 (2020-06-01) * |
| 姬源;谢冬;周思明;王向东;: "电力领域语义搜索系统的构建方法", 计算机系统应用, no. 04, 15 April 2016 (2016-04-15) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113850017A (en) * | 2021-08-27 | 2021-12-28 | 国家电网有限公司 | System-level fault analysis system and method based on power flow change map |
| CN113850017B (en) * | 2021-08-27 | 2024-04-02 | 国家电网有限公司 | System-level fault analysis system and method based on tide change map |
| CN114430365A (en) * | 2022-04-06 | 2022-05-03 | 北京宝兰德软件股份有限公司 | Fault root cause analysis method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110277086B (en) | Voice synthesis method and system based on power grid dispatching knowledge graph and electronic equipment | |
| Wehenkel | Machine learning approaches to power-system security assessment | |
| Du et al. | Power load forecasting using BiLSTM-attention | |
| CN111435366A (en) | Equipment fault diagnosis method and device and electronic equipment | |
| CN113268604A (en) | Self-adaptive expansion method and system for knowledge base | |
| CN112599124A (en) | Voice scheduling method and system for power grid scheduling | |
| CN112860872A (en) | Self-learning-based method and system for verifying semantic compliance of power distribution network operation tickets | |
| CN115357726A (en) | Fault disposal plan digital model establishing method based on knowledge graph | |
| CN113283602A (en) | Power grid system fault diagnosis method based on offline knowledge learning and online dynamic correction | |
| CN107103363A (en) | A kind of construction method of the software fault expert system based on LDA | |
| Ouyang et al. | A multi-stage intelligent system for unit commitment | |
| Huang et al. | An intelligent system for advanced dynamic security assessment | |
| Teng et al. | On-line fault diagnosis technology and application based on deep learning of fault characteristic of power grid | |
| CN113379214A (en) | Method for automatically filling and assisting decision of power grid accident information based on affair map | |
| CN115358369A (en) | Monitoring alarm event identification method based on convolutional neural network model | |
| Zhao et al. | Construction and application of digital twin model of hydropower plant based on data-driven | |
| CN116245033A (en) | Artificial intelligent driven power system analysis method and intelligent software platform | |
| JP7314330B2 (en) | Training method, apparatus, equipment and storage medium for power grid system dispatching model | |
| CN110646710B (en) | Intelligent power grid fault diagnosis method and device, computer equipment and storage medium | |
| CN114461735A (en) | Industrial and mining data classification method and device and computer equipment | |
| CN113991843A (en) | Anti-misoperation method, system, equipment and medium for power grid dispatching operation | |
| CN113076217A (en) | Disk fault prediction method based on domestic platform | |
| Liu et al. | Design of equipment condition maintenance knowledge base in power IoT based on edge computing | |
| CN113434667B (en) | Text classification method based on distribution network automation terminal text classification model | |
| Huang et al. | The application of artificial intelligence in energy Internet |
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 |