CN117668243A - Nuclear power plant I0 knowledge base intelligent construction method and device - Google Patents
Nuclear power plant I0 knowledge base intelligent construction method and device Download PDFInfo
- Publication number
- CN117668243A CN117668243A CN202311580573.1A CN202311580573A CN117668243A CN 117668243 A CN117668243 A CN 117668243A CN 202311580573 A CN202311580573 A CN 202311580573A CN 117668243 A CN117668243 A CN 117668243A
- Authority
- CN
- China
- Prior art keywords
- nuclear power
- knowledge
- power plant
- structured
- entity
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 17
- 238000000605 extraction Methods 0.000 claims description 68
- 238000000034 method Methods 0.000 claims description 50
- 230000006870 function Effects 0.000 claims description 35
- 238000013136 deep learning model Methods 0.000 claims description 22
- 238000010801 machine learning Methods 0.000 claims description 16
- 230000010354 integration Effects 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 12
- 230000009467 reduction Effects 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 claims description 7
- 230000007547 defect Effects 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims description 3
- 230000000737 periodic effect Effects 0.000 claims description 3
- 239000013589 supplement Substances 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 3
- 238000010977 unit operation Methods 0.000 abstract description 3
- 238000009412 basement excavation Methods 0.000 abstract description 2
- 230000033228 biological regulation Effects 0.000 abstract description 2
- 230000008520 organization Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 12
- 238000012545 processing Methods 0.000 description 12
- 230000005540 biological transmission Effects 0.000 description 4
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009960 carding Methods 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- 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
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure belongs to the technical field of nuclear power, and particularly relates to an intelligent construction method and device for an I0 knowledge base of a nuclear power plant. The intelligent construction method of the nuclear power plant I0 knowledge base adopts an intelligent means to assist in constructing the nuclear power plant I0 knowledge base, can quickly and accurately manifest the implicit knowledge of individuals and organization layers of the nuclear power plant related to I0, realizes deep excavation of the related knowledge of the nuclear power plant related to I0, realizes construction of the nuclear power plant I0 knowledge base by the intelligent means, improves the working efficiency and accuracy of personnel identification I0, is convenient for a master control operator to execute production operation activities according to regulations, a nuclear safety supervision engineer to develop nuclear safety supervision and a planning engineer to reasonably schedule activities causing operation limiting conditions, avoids unacceptable risks caused by unit exceeding operation limit, and ensures unit operation safety in the whole life period.
Description
Technical Field
The invention belongs to the technical field of nuclear power, and particularly relates to an intelligent construction method and device for an I0 knowledge base of a nuclear power plant.
Background
The case where all safety requirements associated with each operation mode are not consistent, for example, the required safety function is not available, or the normal operation limit is exceeded, is defined as "I0" in the nuclear power plant operation specification. I0 is divided into a first group I0 and a second group I0, wherein the first group I0 can be expressed as: beyond important design assumptions concerning nuclear safety that should be followed in operation, or the unavailability of reactor shutdown protection and dedicated safety facility systems; the second group I0 can be expressed as: the unavailability of devices and systems associated with the set of events will directly affect the monitoring, diagnosis, and handling of abnormal situations. In recent years, nuclear power plants have experienced excessive events that generate I0.
Currently, the identification of a nuclear power plant master control operator, a nuclear safety supervision engineer, a planning engineer and each professional work responsible person on I0 mainly depends on consulting operation technical specifications and personal experience, and the personnel need to comprehensively consider from multiple layers such as a unit operation mode, whether working object equipment is available, whether upstream and downstream isolation equipment is available, whether a system is partially unavailable or completely unavailable and the like, judge whether the work can generate I0 and make I0 records, and whether I0 conflicts exist. The I0 work is accurately identified in a short time, the requirements on the working experience of a main control operator, a nuclear safety supervision engineer, a planning engineer and professional work responsible persons of the nuclear power plant are higher, the risk that the unit exceeds the operation limit due to insufficient identification of the I0 in the actual working process is increased, the nuclear safety of the unit is affected, and further the operation event generated by exceeding the operation technical specification when the unit is in a transient state is possibly caused, and the strict requirements of the supervision department on the recording and management of the I0 of the nuclear power plant are not met.
Because the number of systems and equipment involved in the nuclear power plant is large, the interlocking condition is complicated, the unavailable conditions of the systems and the equipment are difficult to enumerate, and the I0 records in different operation modes are different, so that the working efficiency of constructing the I0 knowledge base of the nuclear power plant by manual carding is low, and the accuracy and the integrity are difficult to guarantee.
Therefore, there is a need to improve the accuracy and recognition efficiency of I0 recognition.
Disclosure of Invention
In order to overcome the problems in the related art, the intelligent construction method and device for the I0 knowledge base of the nuclear power plant are provided.
According to an aspect of the disclosed embodiments, there is provided a method for intelligently constructing an I0 knowledge base of a nuclear power plant, the method including:
step 1, collecting related structured, semi-structured and unstructured data of a nuclear power plant I0 through a rule/template and field mapping method;
step 2, performing text preprocessing of Chinese word segmentation, noise reduction and entity disambiguation on related data of the nuclear power plant I0;
step 3, acquiring related structural knowledge of the nuclear power plant I0 by adopting a knowledge extraction model;
and 4, enriching and expanding the related structural knowledge of the nuclear power plant I0 based on a knowledge fusion technology to form a nuclear power plant I0 knowledge base.
In one possible implementation, step 1 includes:
step 11, identifying and collecting I0 related data X in the nuclear power plant operation technical specification by adopting a rule/template method, wherein the X is expressed as unstructured text corpus which is inconsistent with all safety requirements related to each operation mode, and the X comprises unstructured text corpus which is used for describing that the required safety function is not used and unstructured text corpus which is used for describing that the normal operation limit value is exceeded;
step 12, collecting I0 related data Y in a historical work order by adopting a field mapping method, wherein Y comprises a historical periodic trigger class work order, a historical defect class work order and other class work orders related to I0;
step 13, collecting I0 related data Z in experience feedback by adopting a field mapping method, wherein Z comprises a nuclear power plant operation event report related to I0, a nuclear power plant internal event report, a device code, a device name, a system number, a system name and other structured text corpus in a C/D level state report, and a semi-structured and unstructured text corpus related to the I0 such as the nuclear power plant operation event report, the nuclear power plant internal event report, an event name, a abstract, a description, risks and consequences, a direct cause, a root cause, a contributing factor and the like in the C/D level state report.
In one possible implementation, step 2 includes:
step 21, chinese word segmentation is carried out on the semi-structured and unstructured text corpus in the step 1, and word segmentation of the text corpus is realized by adopting a conditional random field model and a machine learning model of a Viterbi algorithm based on a barker word stock and a nuclear power professional word stock;
step 22, performing noise reduction treatment on the semi-structured and unstructured text corpus in step 1, establishing a stop word lexicon, wherein the stop word lexicon comprises punctuation marks, special symbols' except for case letters and numbers, language gas word assisting and the like, and performing noise reduction treatment on the text corpus by adopting the stop word lexicon;
and 23, performing entity disambiguation on the structured, semi-structured and unstructured text corpus in the step 1, wherein the entity disambiguation comprises identical reference words and synonym disambiguation, the identical reference words comprise different reference words of the same system, different reference words of the same device, different reference words of the same component and the like, and the synonyms comprise different nuclear power professional words with the same meaning.
In one possible implementation, step 3 includes:
step 31, inputting I0 related data X in the operation technical specification, constructing an entity extraction algorithm model by adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT deep learning model and the like, extracting to obtain an entity X1 'related to keywords such as different systems, partial function not used, whole function not used and a first group I0/second group I0 in the structured knowledge X1, inputting I0 related data X and an entity X1' in the operation technical specification, constructing a relation extraction algorithm model by adopting a CNN, biLSTM+ Atteneion, bert deep learning model and the like, and extracting to obtain a relation X1 'among all entities X1' in the structured knowledge X1;
step 32, obtaining structural knowledge X2 specified in the operation technical specification of the nuclear power plant by adopting entity extraction, attribute extraction and relation extraction technologies, wherein I0 related data X in the operation technical specification is input, an entity extraction algorithm model is constructed by adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT deep learning model and the like, I0 related data X in the operation technical specification is input by extracting to obtain different systems and a first group I0/second group I0 entity X2 'in the structural knowledge X2, an attribute extraction algorithm model is constructed by adopting a BiLSTM+Atteneion model, I0 related data X, entity X2' and attribute X2 'in the structural knowledge X2 are extracted by adopting a CNN, biLSTM+ Atteneion, bert deep learning model and the like, and a relation X2' between the entity X2 'and the attribute X2' in the structural knowledge X2 is obtained by extracting;
step 33, adopting data integration, entity extraction and relation extraction technology to obtain structured knowledge Y1 contained in historical work order data, wherein the structured knowledge Y1 comprises structured knowledge Y11 for indicating which equipment/components are not needed to cause partial or complete non-use of system safety functions, structured knowledge Y12 for indicating which partial or complete non-use of system/equipment functions correspond to the first group I0/second group I0 conditions, inputting I0 related data Y in the historical work order, adopting a mode of carrying out data integration on structured fields related to the system and the equipment in the historical work order data, directly obtaining I0 related data Y in the structured knowledge Y1, adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT and other deep learning model, constructing an entity extraction algorithm model, extracting to obtain 'component', 'partial function is not needed', 'all function is not needed' and 'first group I0/second group I0 entity Y1', inputting the related data Y0 in the historical work order data, adopting a mode of carrying out data integration on structured fields related to the system and the equipment in the historical work order data, directly obtaining I0 related data Y in the historical work order, inputting the entity Y1', adopting a method of combining a CRF machine learning model with the BERT, biLSTM, latticeLSTM, LEBERT and the like, and adopting a deep learning model to obtain a relation 1' 3, and the relation extraction algorithm model;
step 34, obtaining the structured knowledge Z1 contained in the experience feedback data by adopting data integration, entity extraction and relation extraction technologies, wherein the system and the equipment in the experience feedback data directly realize entity identification by means of structured data integration to improve entity identification accuracy, wherein the structured knowledge Z1 comprises 'which equipment/components do not cause partial or complete elimination of system safety functions' structured knowledge Z11 and 'which partial or complete elimination of system/equipment functions' structured knowledge Z12 corresponding to the first group I0/second group I0 condition ', inputting I0 related data Z in the experience feedback, adopting a data integration mode of structured fields related to the system and the equipment in the experience feedback data, directly obtaining a system and a device entity Z1' in the structured knowledge Z1, inputting I0 related data Z in experience feedback, constructing an entity extraction algorithm model by adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT deep learning model and the like, extracting to obtain a component, a part of functions, a first group of I0/a second group of I0 entities Z1 in the structured knowledge Z1, inputting I0 related data Z, an entity Z1 'and an entity Z1 in the experience feedback, adopting a CNN, biLSTM+ Atteneion, bert deep learning model and the like, and extracting to obtain a relation Z1' between the entity Z1 'and the entity Z1' in the structured knowledge Z1.
In one possible implementation, step 4 includes:
step 41, mutually verifying the extraction result of the structural knowledge X1 of the nuclear power plant operation technical specification in the step 3 and the extraction result of the structural knowledge Y12 and Z12 of the historical work order and the experience feedback data in the step 3, and improving the accuracy of the nuclear power plant I0 knowledge base obtained by extraction;
step 42, adopting the extraction results of the structural knowledge Y11 and Z11 of the historical worksheets and the experience feedback data in the step 3 to expand and supplement the mutually verified structural knowledge X1, Y12 and Z12, and improving the fineness of the extracted nuclear power plant I0 knowledge base;
and 43, carrying out induction arrangement on the I0 structured knowledge X1, X2, Y11, Y12, Z11 and Z12 to finally form a set of relatively accurate and rich nuclear power plant I0 knowledge base.
According to another aspect of the embodiments of the present disclosure, there is provided an intelligent construction device for an I0 knowledge base of a nuclear power plant, the device including:
the acquisition module is used for acquiring related structured, semi-structured and unstructured data of the nuclear power plant I0 through a rule/template and field mapping method;
the preprocessing module is used for performing text preprocessing of Chinese word segmentation, noise reduction and entity disambiguation on the related data of the nuclear power plant I0;
the extraction module is used for acquiring related structural knowledge of the nuclear power plant I0 by adopting a knowledge extraction model;
and the fusion module is used for enriching and expanding the related structural knowledge of the nuclear power plant I0 based on the knowledge fusion technology to form a nuclear power plant I0 knowledge base.
According to another aspect of the embodiments of the present disclosure, there is provided an intelligent construction device for an I0 knowledge base of a nuclear power plant, the device including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to perform the above-described method.
According to another aspect of the disclosed embodiments, there is provided a non-transitory computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the above-described method.
The beneficial effects of the present disclosure are: the intelligent construction method of the nuclear power plant I0 knowledge base adopts an intelligent means to assist in constructing the nuclear power plant I0 knowledge base, can quickly and accurately manifest the implicit knowledge of individuals and organization layers of the nuclear power plant related to I0, realizes deep excavation of the related knowledge of the nuclear power plant related to I0, realizes construction of the nuclear power plant I0 knowledge base by the intelligent means, improves the working efficiency and accuracy of personnel identification I0, is convenient for a master control operator to execute production operation activities according to regulations, a nuclear safety supervision engineer to develop nuclear safety supervision and a planning engineer to reasonably schedule activities causing operation limiting conditions, avoids unacceptable risks caused by unit exceeding operation limit, and ensures unit operation safety in the whole life period.
Drawings
FIG. 1 is a flow chart illustrating a method of intelligently building a knowledge base of a nuclear power plant I0, according to an exemplary embodiment.
FIG. 2 is a flow chart illustrating knowledge extraction in a method for intelligently building an I0 knowledge base in a nuclear power plant, in accordance with an exemplary embodiment.
FIG. 3 is a block diagram illustrating an intelligent construction device for an I0 knowledge base of a nuclear power plant, according to an example embodiment.
Detailed Description
The invention will be described in further detail with reference to the accompanying drawings and specific examples.
FIG. 1 is a flow chart illustrating a method of intelligently building a knowledge base of a nuclear power plant I0, according to an exemplary embodiment. The method may be performed by a terminal device, where the terminal device may be a server, a desktop computer, a notebook computer, or the like, and the embodiment of the present disclosure does not limit the type of the terminal device. As shown in fig. 1, the method comprises the steps of:
step 1, collecting related structured, semi-structured and unstructured data of the nuclear power plant I0 through a rule/template and field mapping method.
For example, step 1 may include steps 11 to 13.
And 11, identifying and collecting I0 related data X in the operation technical specification of the nuclear power plant by adopting a rule/template method. X may be represented as all unstructured text corpora inconsistent with the security requirements associated with the respective modes of operation, and X may include unstructured text corpora for describing that the required security functions are not available, and unstructured text corpora for describing that the normal operation limits are exceeded.
And 12, acquiring I0 related data Y in the historical work order by adopting a field mapping method. Y may include historic periodic trigger class worksheets, historic defect class worksheets, historic other class worksheets related to I0, for example, Y may include structured text corpora such as device codes, device names, system numbers, system names, etc. in the worksheets, and semi-structured and unstructured text corpora such as worksheet task names, worksheet descriptions, isolation safety measures, etc. in the worksheets.
And step 13, acquiring I0 related data Z in experience feedback by adopting a field mapping method. Z may include I0 empirical feedback data including structured text corpora related to I0's nuclear power plant operational event reports, nuclear power plant internal event reports, device codes in C/D level status reports, device names, system numbers, system names, etc., and semi-structured and unstructured text corpora related to I0's nuclear power plant operational event reports, nuclear power plant internal event reports, event names in C/D level status reports, abstracts, descriptions, risks and consequences, direct causes, root causes, contributors, etc.
And 2, performing text preprocessing of Chinese word segmentation, noise reduction and entity disambiguation on the related data of the nuclear power plant I0.
For example, step 2 may include steps 21 to 23.
And 21, performing Chinese word segmentation on the semi-structured and unstructured text corpus in the step 1. Based on the barker word stock and the nuclear power professional word stock, a conditional random field model is combined with a machine learning model of a Viterbi algorithm, so that word segmentation of text corpus is realized.
And 22, performing noise reduction processing on the semi-structured and unstructured text corpus in the step 1. And (3) establishing a stop word lexicon, wherein the stop word lexicon comprises punctuation marks, special symbols '#', case letters and numbers, and the like, and adopting the stop word lexicon to perform noise reduction treatment on the text corpus.
And step 23, performing entity disambiguation on the structured, semi-structured and unstructured text corpus in the step 1. Including the same reference disambiguation and synonym disambiguation. Wherein the same reference numerals comprise different reference numerals of the same system, different reference numerals of the same device, different reference numerals of the same component, etc. Synonyms include the same meaning of different nuclear power professional terms such as "leak", "drip", etc.
And step 3, acquiring the related structural knowledge of the nuclear power plant I0 by adopting a knowledge extraction model. Referring to fig. 2, step 3 may include
And step 31, obtaining structural knowledge X1 in the operation technical specification of the nuclear power plant by adopting entity extraction and relation extraction models. For example, in step 31, I0 related data X may be input, and an entity extraction algorithm model may be constructed by combining a CRF machine learning model and a BERT, biLSTM, latticeLSTM, LEBERT deep learning model, so as to extract and obtain the entities X1' associated with the keywords "different systems", "partially unavailable" and "first group I0/second group I0" in the structured knowledge X1.
And inputting I0 related data X and entities X1' in the operation technical specification, and constructing a relation extraction algorithm model by adopting deep learning models such as CNN, biLSTM+ Atteneion, bert and the like, so that the relation X1' among the entities X1' in the structured knowledge X1 can be extracted.
And step 32, obtaining structural knowledge X2 specified in the operation technical specification of the nuclear power plant by adopting entity extraction, attribute extraction and relation extraction technologies. Wherein:
inputting I0 related data X in the operation technical specification, constructing an entity extraction algorithm model by adopting a method of combining a CRF machine learning model and a BERT, biLSTM, latticeLSTM, LEBERT deep learning model, and extracting to obtain ' different systems ' and ' a first group I0/a second group I0 ' entity X2' in the structured knowledge X2.
And inputting I0 related data X in the operation technical specification, constructing an attribute extraction algorithm model by adopting a BiLSTM+Atteneion model, and extracting to obtain an attribute X2 required by the operation parameters in the structural knowledge X2.
And inputting I0 related data X, an entity X2' and an attribute X2' in the operation technical specification, adopting deep learning models such as CNN, biLSTM+ Atteneion, bert and the like, constructing a relation extraction algorithm model, and extracting to obtain a relation X2' between the entity X2' and the attribute X2' in the structured knowledge X2.
And step 33, obtaining the structured knowledge Y1 contained in the historical work order data by adopting data integration, entity extraction and relation extraction technologies. Wherein the structured knowledge Y1 comprises structured knowledge Y11 for indicating which devices/components are not available which would result in a partial or total unavailability of the system security functions, and structured knowledge Y12 for indicating which system/device partial functions are not available or the total functionality is not available for the corresponding first set I0/second set I0 of situations.
The related data Y of I0 in the historical work order is input, and the data integration mode of the structured fields related to the system and the equipment in the historical work order data is adopted, so that the ' system ' entity and the ' equipment ' entity Y1' in the structured knowledge Y1 can be directly obtained.
And inputting I0 related data Y in the historical worksheet, and constructing an entity extraction algorithm model by adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT deep learning model and the like, wherein "parts", "partial functions are unavailable", "all functions are unavailable" and "a first group I0/a second group I0" entity Y1 "in the structured knowledge Y1 can be extracted.
And inputting the I0 related data Y, the entity Y1' and the entity Y1' in the historical work order, adopting deep learning models such as CNN, biLSTM+ Atteneion, bert and the like, constructing a relation extraction algorithm model, and extracting to obtain a relation Y1' between the entity Y1' and the entity Y1' in the structured knowledge Y1.
And step 34, adopting data integration, entity extraction and relation extraction technology to obtain the structured knowledge Z1 contained in the experience feedback data. The system and the equipment in the experience feedback data can directly realize entity identification in a structured data integration mode so as to improve the accuracy of entity identification. Wherein:
structured knowledge Z1 includes "which devices/components are unavailable will result in system security functions being partially or fully unavailable" structured knowledge Z11 and "which system/device portions are not available or full functions are unavailable" structured knowledge Z12 for the corresponding first set of I0/second set of I0 cases.
And inputting the I0 related data Z in the experience feedback, and directly obtaining a 'system' entity Z1 'and a' equipment 'entity Z1' in the structured knowledge Z1 by adopting a mode of integrating data about structured fields of the system and the equipment in the experience feedback data.
Inputting I0 related data Z in experience feedback, constructing an entity extraction algorithm model by adopting a method of combining a CRF machine learning model and a BERT, biLSTM, latticeLSTM, LEBERT deep learning model, and extracting to obtain a component, a part of unavailable function, an all-function unavailable function and a first group I0/a second group I0 entity Z1 in the structural knowledge Z1.
And inputting I0 related data Z, an entity Z1' and an entity Z1' in experience feedback, adopting deep learning models such as CNN, biLSTM+ Atteneion, bert and the like, constructing a relation extraction algorithm model, and extracting to obtain a relation Z1' between the entity Z1' and the entity Z1' in the structured knowledge Z1.
And 4, enriching and expanding the related structural knowledge of the nuclear power plant I0 based on a knowledge fusion technology, and finally forming a nuclear power plant I0 knowledge base.
For example, step 4 may include steps 41 to 43.
And step 41, mutually verifying the extraction result of the structural knowledge X1 of the nuclear power plant operation technical specification in the step 3 and the extraction result of the structural knowledge Y12 and Z12 of the historical work order and the experience feedback data in the step 3, and improving the accuracy of the nuclear power plant I0 knowledge base obtained by extraction.
And 42, expanding and supplementing the mutually verified structured knowledge X1, Y12 and Z12 by adopting the extraction result of the structured knowledge Y11 and Z11 of the historical worksheet and the experience feedback data in the step 3, and improving the fineness of the extracted nuclear power plant I0 knowledge base.
And 43, carrying out induction arrangement on the I0 structured knowledge X1, X2, Y11, Y12, Z11 and Z12 to finally form a set of relatively accurate and rich nuclear power plant I0 knowledge base.
In one possible implementation manner, an intelligent construction device for an I0 knowledge base of a nuclear power plant is provided, and the device includes:
the acquisition module is used for acquiring related structured, semi-structured and unstructured data of the nuclear power plant I0 through a rule/template and field mapping method;
the preprocessing module is used for performing text preprocessing of Chinese word segmentation, noise reduction and entity disambiguation on the related data of the nuclear power plant I0;
the extraction module is used for acquiring related structural knowledge of the nuclear power plant I0 by adopting a knowledge extraction model;
and the fusion module is used for enriching and expanding the related structural knowledge of the nuclear power plant I0 based on the knowledge fusion technology to form a nuclear power plant I0 knowledge base.
The description of the above apparatus is already described in detail in the description of the above method, and will not be repeated here.
FIG. 3 is a block diagram illustrating an intelligent construction device for an I0 knowledge base of a nuclear power plant, according to an example embodiment. For example, the apparatus 1900 may be provided as a server. Referring to fig. 3, the apparatus 1900 includes a processing component 1922 that further includes one or more processors and memory resources represented by memory 1932 for storing instructions, such as application programs, that are executable by the processing component 1922. The application programs stored in memory 1932 may include one or more modules each corresponding to a set of instructions. Further, processing component 1922 is configured to execute instructions to perform the methods described above.
The apparatus 1900 may further comprise a power component 1926 configured to perform power management of the apparatus 1900, a wired or wireless network interface 1950 configured to connect the apparatus 1900 to a network, and an input/output I/O interface 1958. The device 1900 may operate based on an operating system stored in memory 1932, such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, or the like.
In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 1932, including computer program instructions executable by processing component 1922 of apparatus 1900 to perform the above-described methods.
The present disclosure may be a system, method, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for causing a processor to implement aspects of the present disclosure.
The computer readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of the computer-readable storage medium include the non-exhaustive list of: portable computer disks, hard disks, random access memory RAM, read-only memory ROM, erasable programmable read-only memory EPROM or flash memory, static random access memory SRAM, portable compact disc read-only memory CD-ROM, digital versatile discs DVD, memory sticks, floppy disks, mechanical coding devices, punch cards or in-groove protrusion structures such as those having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media, as used herein, are not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media, for example, optical pulses through fiber optic cables, or electrical signals transmitted through wires.
The computer readable program instructions described herein may be downloaded from a computer readable storage medium to a respective computing/processing device or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. The network interface card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium in the respective computing/processing device.
Computer program instructions for performing the operations of the present disclosure can be assembly instructions, instruction set architecture ISA instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of remote computers, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer through the Internet using, for example, an Internet service provider. In some embodiments, aspects of the present disclosure are implemented by personalizing electronic circuitry, such as programmable logic circuitry, field programmable gate array FPGA, or programmable logic array PLA, with state information of computer readable program instructions, which may execute the computer readable program instructions.
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus systems, and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium having the instructions stored therein includes an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. An intelligent construction method for an I0 knowledge base of a nuclear power plant is characterized by comprising the following steps:
step 1, collecting related structured, semi-structured and unstructured data of a nuclear power plant I0 through a rule/template and field mapping method;
step 2, performing text preprocessing of Chinese word segmentation, noise reduction and entity disambiguation on related data of the nuclear power plant I0;
step 3, acquiring related structural knowledge of the nuclear power plant I0 by adopting a knowledge extraction model;
and 4, enriching and expanding the related structural knowledge of the nuclear power plant I0 based on a knowledge fusion technology to form a nuclear power plant I0 knowledge base.
2. The method of claim 1, wherein step 1 comprises:
step 11, identifying and collecting I0 related data X in the nuclear power plant operation technical specification by adopting a rule/template method, wherein the X is expressed as unstructured text corpus which is inconsistent with all safety requirements related to each operation mode, and the X comprises unstructured text corpus which is used for describing that the required safety function is unavailable and unstructured text corpus which is used for describing that the normal operation limit value is exceeded;
step 12, collecting I0 related data Y in a historical work order by adopting a field mapping method, wherein Y comprises a historical periodic trigger class work order, a historical defect class work order and other class work orders related to I0;
step 13, collecting I0 related data Z in experience feedback by adopting a field mapping method, wherein Z comprises a nuclear power plant operation event report related to I0, a nuclear power plant internal event report, a device code, a device name, a system number, a system name and other structured text corpus in a C/D level state report, and a semi-structured and unstructured text corpus related to the I0 such as the nuclear power plant operation event report, the nuclear power plant internal event report, an event name, a abstract, a description, risks and consequences, a direct cause, a root cause, a contributing factor and the like in the C/D level state report.
3. The method of claim 1, wherein step 2 comprises:
step 21, chinese word segmentation is carried out on the semi-structured and unstructured text corpus in the step 1, and word segmentation of the text corpus is realized by adopting a conditional random field model and a machine learning model of a Viterbi algorithm based on a barker word stock and a nuclear power professional word stock;
step 22, performing noise reduction treatment on the semi-structured and unstructured text corpus in step 1, establishing a stop word lexicon, wherein the stop word lexicon comprises punctuation marks, special symbols' except for case letters and numbers, language gas word assisting and the like, and performing noise reduction treatment on the text corpus by adopting the stop word lexicon;
and 23, performing entity disambiguation on the structured, semi-structured and unstructured text corpus in the step 1, wherein the entity disambiguation comprises identical reference words and synonym disambiguation, the identical reference words comprise different reference words of the same system, different reference words of the same device, different reference words of the same component and the like, and the synonyms comprise different nuclear power professional words with the same meaning.
4. The method of claim 1, wherein step 3 comprises:
step 31, inputting I0 related data X in the operation technical specification, constructing an entity extraction algorithm model by adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT deep learning model and the like, extracting to obtain an entity X1 'related to keywords such as different systems, partial functions unavailable, all functions unavailable and a first group I0/a second group I0 in the structured knowledge X1, inputting I0 related data X and an entity X1' in the operation technical specification, constructing a relation extraction algorithm model by adopting a CNN, biLSTM+ Atteneion, bert deep learning model and the like, and extracting to obtain a relation X1 'among all entities X1' in the structured knowledge X1;
step 32, obtaining structural knowledge X2 specified in the operation technical specification of the nuclear power plant by adopting entity extraction, attribute extraction and relation extraction technologies, wherein I0 related data X in the operation technical specification is input, an entity extraction algorithm model is constructed by adopting a method of combining a CRF machine learning model with a BERT, biLSTM, latticeLSTM, LEBERT deep learning model and the like, I0 related data X in the operation technical specification is input by extracting to obtain different systems and a first group I0/second group I0 entity X2 'in the structural knowledge X2, an attribute extraction algorithm model is constructed by adopting a BiLSTM+Atteneion model, I0 related data X, entity X2' and attribute X2 'in the structural knowledge X2 are extracted by adopting a CNN, biLSTM+ Atteneion, bert deep learning model and the like, and a relation X2' between the entity X2 'and the attribute X2' in the structural knowledge X2 is obtained by extracting;
step 33, adopting data integration, entity extraction and relation extraction technology to obtain structured knowledge Y1 contained in historical work order data, wherein the structured knowledge Y1 comprises structured knowledge Y11 for indicating which equipment/components are unavailable and can cause partial unavailability or total unavailability of system safety functions, structured knowledge Y12 for indicating which partial unavailability or total unavailability of system/equipment functions corresponds to the first group I0/second group I0, I0 related data Y in the historical work order is input, the mode of carrying out data integration on structured fields related to the system and the equipment in the historical work order data is adopted, I0 related data Y in the structured knowledge Y1 is directly obtained, the method of combining deep learning models such as CRF machine learning models and BERT, biLSTM, latticeLSTM, LEBERT is adopted, an entity extraction algorithm model is constructed, I0/second group I0 is obtained by extraction, the mode of 3I 0/second group I0 is adopted, the 3I 0 related data Y1' is input, the 3 ' 3Y 1 is input, the 3Y 1' is input, the 3 ' 3Y 1 is input, and the 3 ' 3Y 1 is input;
step 34, obtaining the structured knowledge Z1 contained in the experience feedback data by adopting data integration, entity extraction and relation extraction technologies, wherein the system and equipment in the experience feedback data directly realize entity identification by means of structured data integration to improve entity identification accuracy, wherein the structured knowledge Z1 comprises a method for combining 'which equipment/components are unavailable and can lead to partial unavailability or total unavailability of safety function of the system' of the structured knowledge Z11 and 'which system/equipment partial functions are unavailable or total unavailability of all functions of the system' of the structured knowledge Z11, a first group of I0/second group of I0 conditions 'structured knowledge Z12, inputting I0 related data Z in the experience feedback data, directly obtaining' system 'and' equipment 'entity Z1' related data Z in the experience feedback data by adopting a mode of integrating data about the system and equipment in the experience feedback data, and adopting a deep learning model of CRF machine learning model and BERT, biLSTM, latticeLSTM, LEBERT and the like to construct an entity extraction algorithm model, extracting to obtain 'component' part 'and' TM 'part' and 'part' of the structured knowledge Z1, and 'I0/second entity' 3 'and' 3Z 1, and a depth relation between 'Z1 and a third group of 3Z 1' of the structured knowledge Z1, and a third group of Z1, and a fourth-step 3.
5. The method of claim 1, wherein step 4 comprises:
step 41, mutually verifying the extraction result of the structural knowledge X1 of the nuclear power plant operation technical specification in the step 3 and the extraction result of the structural knowledge Y12 and Z12 of the historical work order and the experience feedback data in the step 3, and improving the accuracy of the nuclear power plant I0 knowledge base obtained by extraction;
step 42, adopting the extraction results of the structural knowledge Y11 and Z11 of the historical worksheets and the experience feedback data in the step 3 to expand and supplement the mutually verified structural knowledge X1, Y12 and Z12, and improving the fineness of the extracted nuclear power plant I0 knowledge base;
and 43, carrying out induction arrangement on the I0 structured knowledge X1, X2, Y11, Y12, Z11 and Z12 to finally form a set of relatively accurate and rich nuclear power plant I0 knowledge base.
6. An intelligent construction device for an I0 knowledge base of a nuclear power plant, comprising:
the acquisition module is used for acquiring related structured, semi-structured and unstructured data of the nuclear power plant I0 through a rule/template and field mapping method;
the preprocessing module is used for performing text preprocessing of Chinese word segmentation, noise reduction and entity disambiguation on the related data of the nuclear power plant I0;
the extraction module is used for acquiring related structural knowledge of the nuclear power plant I0 by adopting a knowledge extraction model;
and the fusion module is used for enriching and expanding the related structural knowledge of the nuclear power plant I0 based on the knowledge fusion technology to form a nuclear power plant I0 knowledge base.
7. An intelligent construction device for an I0 knowledge base of a nuclear power plant, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to perform the method of any one of claims 1 to 5.
8. A non-transitory computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311580573.1A CN117668243A (en) | 2023-11-22 | 2023-11-22 | Nuclear power plant I0 knowledge base intelligent construction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311580573.1A CN117668243A (en) | 2023-11-22 | 2023-11-22 | Nuclear power plant I0 knowledge base intelligent construction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117668243A true CN117668243A (en) | 2024-03-08 |
Family
ID=90081847
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311580573.1A Pending CN117668243A (en) | 2023-11-22 | 2023-11-22 | Nuclear power plant I0 knowledge base intelligent construction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117668243A (en) |
-
2023
- 2023-11-22 CN CN202311580573.1A patent/CN117668243A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113807098B (en) | Model training method and device, electronic equipment and storage medium | |
| Lami et al. | An automatic tool for the analysis of natural language requirements | |
| US20170270105A1 (en) | Method and apparatus for generating causal explanations using models derived from data | |
| CN113705187A (en) | Generation method and device of pre-training language model, electronic equipment and storage medium | |
| US20230004721A1 (en) | Method for training semantic representation model, device and storage medium | |
| US8739025B2 (en) | Systems and methods for navigating to errors in an XBRL document using metadata | |
| EP3543871A1 (en) | Artificial intelligence-based triple checking method and apparatus, device and storage medium | |
| US20220350690A1 (en) | Training method and apparatus for fault recognition model, fault recognition method and apparatus, and electronic device | |
| CN112612664A (en) | Electronic equipment testing method and device, electronic equipment and storage medium | |
| CN113656590A (en) | Industry map construction method and device, electronic equipment and storage medium | |
| CN116307672A (en) | Fault diagnosis method, device, electronic equipment and medium | |
| Apriyanto et al. | Extraction of Step Performed in Use Case Description as a Reference for Conformity of Sequence Diagrams Using Text Mining (Case Study: SRS APTU) | |
| Zhao et al. | Automated identification of causal relationships in nuclear power plant event reports | |
| US20230185646A1 (en) | Method for early warning of failure, electronic device and non-transitory computer-readable storage medium | |
| CN117668243A (en) | Nuclear power plant I0 knowledge base intelligent construction method and device | |
| CN116881395A (en) | Public opinion information detection method and device | |
| US20230077338A1 (en) | System for natural language processing of safety incident data | |
| CN115687031A (en) | Method, device, equipment and medium for generating alarm description text | |
| CN115344661A (en) | Equipment halt diagnosis method and device, electronic equipment and storage medium | |
| CN111723580B (en) | Power dispatching station information graph validation method based on voice recognition and image recognition | |
| CN114297057A (en) | Design and use method of automatic test case | |
| Wang et al. | Reliability Evaluation of Complex Nuclear Energy Redundancy Systems with Atypical Time-Distribution Events | |
| CN109828775B (en) | WEB management system and method for multilingual translation text content | |
| Ren et al. | Construction Procedural Information Extraction from Textual Sources to Support Scheduling | |
| Liu et al. | Patent Technical Function-effect Representation and Mining Method. |
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 |