CN115510181A - Nuclear power plant instant knowledge management method and system - Google Patents

Abstract

The application discloses a method and a system for managing real-time knowledge of a nuclear power plant, which comprises the steps of firstly, acquiring real-time knowledge data; then loading the instant knowledge data into an instant knowledge database according to the time sequence; and finally, searching in the knowledge database according to a preset index word, and displaying the information of the search result. The method comprises the steps of firstly identifying, screening and extracting various instant information of the nuclear power unit acquired by channels such as an early meeting, a production week meeting, an experience feedback example meeting and a major repair daily report of the nuclear power plant to generate a plurality of instant learning type knowledge documents, constructing an instant knowledge database, and then realizing full-database full-text fast retrieval and display of the instant knowledge documents through indexing of the constructed instant knowledge database, so that professionals in the nuclear power plant can quickly learn and know nuclear power knowledge and skills related to own duties, and the pertinence, timeliness, practicability and effectiveness of knowledge management of nuclear power knowledge training are improved.

Description

Nuclear power plant instant knowledge management method and system

Technical Field

The application relates to the technical field of nuclear power personnel training, in particular to a method and a system for managing instantaneous knowledge of a nuclear power plant.

Background

The training of nuclear power plant personnel is special, mainly because nuclear power project construction operation cycle is long, the technological complicacy is decided with supervision strict grade factor, and nuclear power plant is also high to the knowledge skill requirement of each professional in addition, especially in the debugging or operation initial stage of nuclear power plant, especially first heap project, the complicated problem that its technological advance, uniqueness brought is increasing day by day, a large amount of interior external experience feedback, management requirement, operation essential and hot incident etc. form the unit instant knowledge be difficult to internalize in personnel's ability and work document in short-term, its concrete reason is as follows:

1) The traditional learning and training mode has an inherent period, needs fixed time and space for training course development, training organization, student participation and the like, is time-consuming and involves a large amount of coordination work, and is difficult to meet the requirement of training timeliness;

2) Most of traditional training is passive, the related knowledge is integrated and solidified in advance, and the focus, depth and breadth of the training can meet the individual knowledge requirements of students;

3) The traditional training mode is large in knowledge amount, difficult to internalize in short-time learning and slow in knowledge updating iteration.

In summary, how to make the nuclear power instant knowledge points explicit, instrumental and fragmented better helps each professional of the nuclear power plant to immediately master and apply the nuclear power knowledge, and continuously improves the personnel ability becomes the key for solving the problem.

Disclosure of Invention

The technical problem solved in advance by the application is how to realize scientific management of the real-time knowledge of the nuclear power plant.

In a first aspect, an embodiment provides a method for managing instantaneous knowledge of a nuclear power plant, including:

acquiring instant knowledge data; the types of the instant knowledge data comprise conference summary data, operation parameter data of each nuclear power generating unit, parameter change data of each nuclear power generating unit and/or overhaul data of a nuclear power plant;

loading at least one instant knowledge data into an instant knowledge database according to the sequence of time corresponding to the instant knowledge data; the instant knowledge database is established according to historically acquired instant knowledge data;

searching in the knowledge database according to a preset index word, and displaying search result information; the retrieval result information comprises document id, query index entries, result summaries and/or scores of each document; the document score is used to identify the trustworthiness of the displayed document.

In a second aspect, an embodiment provides a computer-readable storage medium having a program stored thereon; the program is executable by a processor to implement the plant immediate knowledge management method of the first aspect.

In a third aspect, an embodiment provides a nuclear power plant immediate knowledge management system, including:

the instant knowledge acquisition device is used for acquiring instant knowledge data; the types of the instant knowledge data comprise conference summary data, operation parameter data of each nuclear power generating unit, parameter change data of each nuclear power generating unit and/or overhaul data of a nuclear power plant;

the data loading device is used for loading the instant knowledge data into the instant knowledge database according to the time sequence; the instant knowledge database is established according to the historically acquired instant knowledge data;

the data retrieval device is used for retrieving in the knowledge database according to a preset index word and displaying retrieval result information; the retrieval result information comprises a document id, a query index entry, a result abstract and/or each document score; the document score is used to identify the trustworthiness of the document.

According to the method for managing the real-time knowledge of the nuclear power plant in the embodiment, various real-time information of the nuclear power unit acquired by channels such as the early meeting, the production week meeting, the experience feedback example meeting, the major repair daily report and the like of the nuclear power plant is firstly identified, screened and extracted to generate a plurality of real-time learning type knowledge documents, a real-time knowledge database is constructed, and then full-library full-text quick retrieval and display of the real-time knowledge documents are realized through indexes of the constructed real-time knowledge database, so that professionals of the nuclear power plant can quickly learn and know core knowledge and skills such as principles, equipment failure modes, possible risks, intervention methods, historical feedback, operation key points and the like behind the ongoing core work, the occurring actual problems, the current unit or equipment states and the like of the nuclear power plant related to own duties, and the pertinence, timeliness, practicability and the effectiveness of knowledge management are improved.

Drawings

FIG. 1 is a schematic flow chart of a method for real-time knowledge management in a nuclear power plant, according to an embodiment;

FIG. 2 is a flow diagram illustrating the construction of an instant knowledge database according to one embodiment;

FIG. 3 is a diagram illustrating a mapping of a list of instant knowledge points in one embodiment;

FIG. 4 is a block diagram of an embodiment of a point of knowledge list with attributes;

FIG. 5 is a diagram illustrating a listing of instant knowledge documents in one embodiment;

FIG. 6 is a diagram illustrating a forward index of the instant knowledge database, in accordance with an embodiment;

FIG. 7 is a diagram illustrating a display of an inverted index search result list of the instant knowledge database in one embodiment;

FIG. 8 is a schematic diagram of a configuration connection of an immediate knowledge management system of a nuclear power plant in one embodiment;

FIG. 9 is a schematic diagram of a structural connection of an immediate knowledge management system of a nuclear power plant in another embodiment.

Detailed Description

The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. Wherein like elements in different embodiments are numbered with like associated elements. In the following description, numerous details are set forth in order to provide a better understanding of the present application. However, those skilled in the art will readily recognize that some of the features may be omitted or replaced with other elements, materials, methods in different instances. In some instances, certain operations related to the present application have not been shown or described in detail in order to avoid obscuring the core of the present application from excessive description, and it is not necessary for those skilled in the art to describe these operations in detail, so that they may be fully understood from the description in the specification and the general knowledge in the art.

Furthermore, the features, operations, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. Also, the various steps or actions in the method descriptions may be transposed or transposed in order, as will be apparent to one of ordinary skill in the art. Thus, the various sequences in the specification and drawings are for the purpose of describing certain embodiments only and are not intended to imply a required sequence unless otherwise indicated where such sequence must be followed.

The ordinal numbers used herein for the components, such as "first," "second," etc., are used merely to distinguish between the objects described, and do not have any sequential or technical meaning. The term "connected" and "coupled" when used in this application, unless otherwise indicated, includes both direct and indirect connections (couplings).

The method for managing the instant knowledge of the nuclear power plant, disclosed by the embodiment of the application, effectively solves the problem of knowledge timeliness existing in the traditional learning and training mode through rapid manufacturing and generation of the instant knowledge content, is strong in pertinence and rapid in iteration, has the depth and breadth more fit with the operation needs of a nuclear power unit and the personalized knowledge needs of students, and can work around the center of the unit. In addition, through the retrieval and display of the instant knowledge, the 'dominance' and 'instrumentization' of the nuclear power knowledge are realized, the complex knowledge is broken into zero, the cognitive load of each professional is reduced, and the convenience of knowledge use is improved.

Example one

Referring to fig. 1, a schematic flow chart of an embodiment of an instant knowledge management method for a nuclear power plant, which is used for generating, retrieving and displaying instant knowledge of a nuclear power plant, includes:

step 101, acquiring instant knowledge data.

And identifying, screening and/or extracting the instant knowledge document according to a preset format and type to generate instant knowledge data. The instant knowledge documents comprise conference summary documents, operation parameter documents of each nuclear power unit, parameter change documents of each nuclear power unit and/or overhaul documents of a nuclear power plant. Each instant knowledge document is generated according to preset compiling rules, upgrading rules and releasing rules and has fixed document types and templates. The types of the instant knowledge data comprise conference summary data, operation parameter data of each nuclear power unit, parameter change data of each nuclear power unit and/or overhaul data of a nuclear power plant. The format of the instant knowledge data comprises structured data, semi-structured data and/or unstructured data, the structured data has fixed fields and formats, and text at a specified position in the structured data has a determined meaning; the semi-structured data has no fixed format and length, but comprises key identification with characteristic keywords, wherein the key identification is used for identifying that the text at the preset position has a determined meaning; unstructured data has no fixed format and length, and no feature keys. For unstructured data, all texts of the unstructured data are read in a direct streaming mode, and the read all texts serve as a unit to be participled; adopting flow reading for the semi-structured data, segmenting a text at a preset position according to key identification in the semi-structured data to obtain text segments, and taking each text segment as a new unit to be participled; and for the structured data, directly extracting texts from all positions according to the format and meaning of the structured data, and taking the extracted texts as word segmentation units.

In one embodiment, the instant knowledge data is generated after the word units to be segmented are screened, segmented and/or given definitions according to a preset special dictionary. In one embodiment, the instant knowledge data is generated directly for the word segmentation unit. In one embodiment, the instant knowledge data is generated after the word units to be segmented are screened, segmented and/or given definitions according to the preset nuclear power equipment function position coding rule.

The acquisition method of the conference summary data comprises the following steps:

firstly, acquiring an early meeting summary, a production week summary, an experience feedback regular meeting summary and/or a temporary emergency meeting summary of a nuclear power plant; and then converting the preschool era, the production week era, the experience feedback example era and/or the temporary emergency meeting era into a meeting era text according to a preset text format, and taking the meeting era text as meeting era data, wherein the meeting era data comprises meeting time, meeting place, meeting participants, the types and titles of the meeting participants, the number of nuclear power generating units related to the meeting and/or the revision content of the operating parameters of the nuclear power generating units. In addition, the operation parameter data of each nuclear power generating unit is the operation parameter data of each nuclear power generating unit, and the parameter change data of each nuclear power generating unit is the operation parameter change data of each nuclear power generating unit, wherein the operation parameter change data comprises data before change and data after change. The overhaul data of the nuclear power plant comprises overhaul daily report data, overhaul summary data, overhaul plan refueling data, overhaul refueling completion data, overhaul plan test data and overhaul test completion data.

The following describes the acquisition of the instant knowledge data by using an actual embodiment, which specifically includes:

generally, with the improvement of the standardization level of the working process of the nuclear power industry, information of daily early meetings, weekly meetings, experience feedback examples, various daily newspapers and the like of a nuclear power plant basically has a relatively fixed format, and most of the information is structured forms or semi-structured texts and also has a few unstructured texts.

Please refer to table 1, which is a table showing structured data in an empirical feedback example, specifically:

TABLE 1

Type (B)	Event name	Human factor	Focus on the point	SPV information	Remarks for note
						NG	[KA]Master control trigger 2K RT1228KA	NA	The equipment is subjected to accidental fault alarm, and the contact failure of the channel can be alarmed.	No SPV	NA
…	…	…	…	…	…

The data in the structured data has fixed fields and formats (such as tables), and the data with determined meanings at specified positions (rows and columns) can be extracted according to a preset format.

Although the data in the semi-structured data has no fixed format and length, the data has obvious characteristic keyword identification, and the text in the area near the characteristic keyword has specific meaning. The following takes the information of a certain power plant in a certain morning meeting as an example:

as "[ unit number ] [ foreign matter information notification ] [ STA attention ] x [ year ], when 2PTR x EX regular maintenance (5Y) was performed, 2PTR x EX was found to have sponge bars of 8mm × 1mm and white cloth of 7mm × 3mm after opening the apparatus, NG No. 20000 x.

And as another example, "[ problem description ] the field inspection finds that the 2CFI × DG scraper 2CFI × DG stops at the bottom, the rope is sent to the switch to act, and the primary analysis failure cause is that the scraper does not stop in time and further triggers the rope loosening protection action because the lower limit of the 2CFI × DG scraper is not normally triggered. Filling with NG:20000*******. [ MEANS FOR ASSEMBLING THE INDUCTION OF THE SET ] the equipment related to the cold source needs to be processed as soon as possible. … act as a requirement while handling fault by OPI, analyzing whether common mode fault causes exist, and making a scheme for improving equipment reliability for a long time. ".

Although the above example has no fixed format and length, fixed characteristic keyword fields such as [ unit number ], [ problem description ], [ unit influence analysis ], [ historical information ], [ action requirement ], and the like can be used for specific text field positioning and extraction;

the data in the unstructured data has no fixed format and length and no obvious characteristic identification. For example:

and filling water and exhausting after the 2PTR first row maintenance, please use OPO to estimate the water filling volume finely, stare the liquid level of the spent pool in the water filling process, avoid being lower than the requirement of the technical specification, and control the rhythm and the risk.

Aiming at the instant information texts such as the unstructured data, the structured data, the semi-structured data and the like, the instant knowledge points containing multiple dimensions can be extracted from the instant information texts by matching with a predefined special dictionary, a nuclear power equipment coding rule and the like.

In one embodiment, the special dictionary includes special words in addition to the general Chinese and English words. Please refer to table 2, which is a demonstration table of the special dictionary segments in an embodiment, specifically including:

TABLE 2

Entry word	Categories	Weight of	Associated specialties/departments
				PTR	Department/specialty/field	10	Running; maintenance
OPO	Special words/terms	9	Operation of
				Water filling and air exhausting	Special words/terms	10	Instrument control
Spent pool	Special words/terms	10	Running; maintenance
				Liquid level of the spent pool	Special words/terms	10	Running; maintenance
Specification of the technology	Special words/terms	10	Running; maintenance
				Risks	General words	8	NA
KA	Alarm device		9					Running; maintenance
				…	…	…	…

The special words can be obtained from materials such as a unit design file (such as a system code PTR for a 'fuel water pool cooling and purifying system'), a traditional training teaching material, industry standard words (such as foreign matters, foreign matter prevention, human factors, SPV (single weakness), wrong interval and the like), a self-defined code of a nuclear power plant (such as department abbreviations (OPI for an instrument control department, OPO for an operation department), KA alarm, characteristic keywords of a semi-structured text and the like, and the special dictionary can be supplemented and upgraded as required; each entry of the special word comprises attributes such as weight, category, associated specialty/department and the like, and can be mapped to specific attributes of the knowledge point according to the category of the entry. In addition, the device coding rule such as "2CFI × DG" indicates a unit device function position code, 2 indicates a unit number, CFI indicates a system name, × (4-digit number) indicates a device code, and DG indicates a device type.

And step 102, loading the data into a database.

And loading at least one piece of instant knowledge data into an instant knowledge database according to the sequence of time corresponding to the instant knowledge data, wherein the instant knowledge database is established according to the historically acquired instant knowledge data.

In an embodiment, the chronological order of the time corresponding to the immediate knowledge data is based on the occurrence time of the immediate knowledge data, for example, the immediate knowledge data includes first immediate knowledge data and second immediate knowledge data, where the first immediate knowledge data and the second immediate knowledge data are adjustment data for different values of the same parameter of the same equipment of the nuclear power plant, and are loaded into the immediate knowledge database according to the chronological order of the first immediate knowledge data and the second immediate knowledge data.

In one embodiment, the immediate knowledge database includes an immediate knowledge point list, wherein the immediate knowledge database maps the immediate knowledge data to the immediate knowledge point list according to the type of the entry and the specific attribute of the entry, the knowledge point list is used for recording the immediate knowledge points, and each immediate knowledge point is distinguished according to the type of the entry and the specific attribute of the entry. In one embodiment, each point of immediate knowledge may be associated with an immediate knowledge document. In one embodiment, the point of knowledge includes a term name, a system to which the point of knowledge belongs, a directly related department, a potentially influencing department, a person responsible for knowledge point identification, a term source, and/or raw data text.

In one embodiment, the instant knowledge database further comprises at least one instant knowledge document list used for recording instant knowledge document contents, inverted indexes, forward indexes and/or special dictionaries, wherein each instant knowledge document list comprises globally unique document ID, document codes, names, subjects, versions, affiliated systems, associated knowledge points and/or document types. In one embodiment, the recorded instant knowledge document comprises at least one instant knowledge point list.

In one embodiment, the immediate knowledge database further comprises an inverted index library and/or a forward index library. The inverted index library is used for forming a word list with preset text attributes by using a special dictionary for all pure text characters contained in the instant knowledge document after word segmentation, and the preset text attributes comprise a document ID list, a text word frequency and/or a text position. The forward index library comprises mapping information of the text content segment and the specific position of the text content segment in the corresponding recording instant knowledge document, so that the text content segment is positioned to the specific position of the text content segment in the corresponding recording instant knowledge document through content. In one embodiment, the search uses a two-engine mode, an inverted index search mode for gross searches across knowledge documents across the full corpus, and a forward index search mode for fine searches within knowledge documents.

In one embodiment, the immediate knowledge database further includes basic knowledge documents, where the basic knowledge documents include nuclear power knowledge-related teaching materials, nuclear power plant design files, nuclear power plant technical specifications, nuclear power plant digitization programs, and/or nuclear power plant experience feedback documents.

Referring to fig. 2, a schematic diagram of a process for constructing an instant knowledge database in an embodiment is shown, where in an embodiment, a process for constructing an instant knowledge database includes several steps of acquiring instant information, extracting instant knowledge points, generating an instant knowledge point list with several strips and several attributes, generating several instant knowledge documents according to needs, and constructing a knowledge base.

(1) The acquisition of the instant message comprises the following steps: various types of instant information data of the nuclear power unit are obtained from channels such as an early meeting, a production week meeting, an experience feedback example meeting, various daily reports and the like of the nuclear power plant, and the unit instant information comprises three types of unstructured texts, structured texts and semi-structured texts.

(2) The instant knowledge point extraction comprises the following steps: the method comprises the steps of identifying and processing the acquired instant messages, identifying and extracting key messages from the instant messages, and mainly comprising three steps of text preprocessing, text word segmentation and core word extraction. The text preprocessing method comprises the following specific steps:

for unstructured data, directly reading all texts in a streaming manner to serve as a unit to be participled;

for semi-structured data, reading in a streaming mode, segmenting texts according to characteristic keyword fields of the data, wherein each text segment is used as a unit to be segmented;

for the structured data, text information is directly extracted from each position according to the format and meaning of the structured data, part of the text information is directly stored as a knowledge point attribute without word segmentation (text word segmentation can be skipped), and part of the text information, such as important or longer text fields of 'event name', 'concern point' and the like in the table 1, is respectively used as a unit to be word segmented for word segmentation processing.

For each unit text to be segmented, firstly, extracting a device function position code (if the unit text to be segmented exists) according to a device code rule, analyzing attribute information such as a system name, the device code, a device type and the like of the unit text to be segmented according to the device function position code rule, and storing the attribute information as a core word. If from text "[ KA ] master control trigger 2KRT1228KA", device function position code "2KRT1228KA" can be extracted, and attribute information such as its system name (KRT), device code (1228), device type (KA) and the like can be obtained. In one embodiment, a special dictionary is used for word segmentation of the word unit to be segmented, weight calculation and sequencing are carried out on word segmentation results, words with weights larger than a specific threshold value are screened out to serve as core words, and the core words and entry categories of the core words in the dictionary are mapped to specific attributes of the instant knowledge point list. For example, for water filling and air exhausting after the text 2PTR first row maintenance, please use OPO to estimate the water filling volume finely, stare the liquid level of the spent pool in the water filling process, avoid being lower than the requirement of the technical specification, and control the rhythm and the risk. "after performing word segmentation by a special dictionary, a word segmentation list containing entries and their weights can be obtained, such as" 2, 1.3939Y PTR, 9.1607|, 6.1665| overhaul, 6.2943|, 1.6213|, 0.1764| water charging, 6.3417| water charging and air discharging, 9.6109|, 1.676 OPO, 8.9065| fine, 5.837| estimation, 5.7629| volume, 5.8862| water charging, 7.1573 | process, 5.7258 | in, 2.3324 | staring, 5.8315 | liquid level of the pool, 9.8744 avoids, 5.0245 is lower, 6.3462 technical specification, 9.2545 requires, 5.9502 controls, 6.3292 is good, 1.1519 rhythm, 5.3149 and 1.0415 risk, 8.9625 ' screens the sorting according to a threshold (such as 8.000 threshold), and then core words such as ' spent pool liquid level, water and gas filling and discharging, technical specification, PTR, risk, OPO ' and information such as categories, weights, associated professions/departments and the like attached to the special dictionary entries can be obtained. It should be noted that the text word segmentation result depends on a specific word segmentation algorithm and dictionary quality, the prior art is mature, and only examples for easy understanding are provided here, and detailed principles are not described again.

(3) Generating a plurality of instant knowledge point lists with a plurality of attributes comprises the following steps:

referring to fig. 3, a mapping diagram of an embodiment of an instant knowledge point list is shown, where the instant knowledge point list includes attributes such as an instant knowledge point name, a system, a department/potential influence department, a person responsible for knowledge point identification, an instant information source, and an original text. The attributes of the core words, the categories thereof, the associated professions/departments and the like are obtained through the identification processing, and then the attributes can be mapped into the specific attributes of the instant knowledge point list by combining with a division table (mapped to specific responsible personnel) of the nuclear power plant system.

Referring to fig. 4, a schematic structural diagram of an instant knowledge point list with several attributes in an embodiment is shown, where the instant knowledge point list with several attributes can obtain names of instant knowledge points, systems to which the instant knowledge points belong, associated departments/specialties, and responsible persons, and also associate sources of instant messages and original texts, so as to facilitate inspection and processing in subsequent steps.

(4) Generating a plurality of instant knowledge documents according to the requirement comprises:

and according to the instant knowledge point list with the plurality of attributes, assigning a specific responsible person for identification and analysis and associating the knowledge points, if the knowledge points exist in the knowledge base, ignoring or upgrading the instant knowledge document as a case, and if the knowledge points are new knowledge points (not in the knowledge base), organizing and writing the instant knowledge document aiming at the knowledge points.

And establishing a uniform edition compiling and publishing guide rule of the instant knowledge document, and specifying the type, the template, the document coding rule, the instant knowledge point association requirement, the theme naming rule, the document directory and index, the background information, the case and example, the reference file, the page watermark, the identified/found field problem processing, the edition compiling/edition upgrading/examining requirement, the publishing object and the like of the instant knowledge document.

The type of the instant knowledge (JIT) document comprises a knowledge card, a tool card, a case card, a security analysis card, a human factor prevention knowledge card, an SPV case card and the like. Knowledge card: a JIT card for knowledge classes such as basic knowledge, basic principle, characteristic knowledge and the like of a nuclear power station; tool card: a JIT card which is used for summarizing and refining a certain type or a series of contents and used for being inquired by an operator as a tool; case card: deep analysis or extended analysis is carried out on the unit case, and a JIT card of the matters concerned by the operating personnel is provided; a security analysis card: a specialized knowledge card for security analysis; human factor prevention knowledge card and SPV case card: a knowledge card for human error and SPV (single point weak device) cases.

In order to standardize the format and facilitate retrieval, each type of instant knowledge document has a fixed document template, and a proper document type and template can be selected according to the knowledge point attributes of the instant knowledge document.

Each instant knowledge document may be associated with a plurality of knowledge point attributes based on its content. Each instant knowledge document is provided with fixed compiling, checking and examiner, the completion time limit is preset in each link, the document space is limited to improve the timeliness of the instant knowledge document, and after the compiling, checking and checking of the instant knowledge document are finished, the instant knowledge document in a specific format (such as a PDF format) is generated and enters a knowledge base construction link.

(5) Constructing the immediate knowledge database comprises:

the instant knowledge database consists of a plurality of instant knowledge documents, and mainly comprises an instant knowledge document list. Please refer to fig. 5, which is a diagram illustrating an embodiment of an instant knowledge document list. The instant knowledge database comprises instant knowledge document contents, an inverted index library, a forward index library, a special dictionary and the like. The instant knowledge document list comprises attributes of globally unique document ID, document code, name, subject, version, belonging system, associated knowledge point, type and the like. The instant knowledge document content mainly comprises a document ID and content data which is stored in a binary form in an encrypted and compressed mode and keeps the original format of the document; the inverted index library mainly comprises a word list with a plurality of attributes generated after all pure text characters contained in the instant knowledge document are segmented by using a special dictionary, wherein the attributes mainly comprise a document ID list (all documents in which the word appears), word frequency (the number of times the word appears in all documents), positions (character offset values of the word in specific documents) and the like. The input of the inverted index library is the instant document content of plain text characters without any format information. When in query search, firstly, the content to be searched is segmented, the related documents containing the search words and the specific character positions of the related documents in the documents can be quickly matched through the inverted index list, the relevancy of the search words is calculated, and the front and rear character contents of the search words are intercepted through the character positions of the search words in the documents to generate the content abstract. The forward index library comprises mapping information of a plurality of content fragments (the granularity of the content fragments can be paragraphs, lines or specific characters/words) and specific positions (pages, x, y, width and height) of the content fragments in the PDF format document, and the content fragments can be positioned to the specific positions in the document.

Referring to fig. 6, a schematic diagram of forward indexing of the instant knowledge database in an embodiment is shown, where when paragraphs of an instant knowledge document are used as the index granularity, besides the paragraphs are used as the forward index granularity, the text in each line may be used as the index granularity, or each word after word segmentation is used as the index granularity. The different granularities mainly affect the creation speed of the forward index, the size of the index file and the analysis speed of the retrieval result.

In one embodiment, when a new or upgraded instant knowledge document is published, the relevant information of the instant knowledge document is written into a knowledge document list, the content of the instant knowledge document is preserved in an original format and is encrypted and compressed in a binary mode, the pure character content of the document is read to generate and update an inverted index, the content of the document with the read format is analyzed, the forward index is generated and updated in a specific granularity (paragraph, line and word), an index base is built, and relevant personnel or potential users are informed that the instant knowledge document is published. As an optional scheme, in an embodiment, the knowledge base may include other nuclear power knowledge documents, such as teaching materials, design files, technical specifications, digital programs, experience feedback, and the like, in addition to the instant knowledge document, so as to implement retrieval and display of a full-knowledge type and expand an application range of nuclear power knowledge.

And 103, searching and displaying a database.

And searching in the knowledge database according to a preset index word, and displaying search result information, wherein the search result information comprises a document id, a query index entry, a result abstract and/or a score of each document. Note that the document score is used to identify the trustworthiness of the displayed document. In one embodiment, the instant knowledge document that retains the original format is opened and displayed according to the document id.

In one embodiment, on the basis of the instant knowledge database, retrieval and display of nuclear power knowledge points can be realized by constructing a query interface, an inverted index query result display interface and an instant knowledge document display interface of the instant knowledge database. In one embodiment, a dual-engine mode is used for knowledge base query, a user inputs query contents through a query interface, the query interface is called to obtain a plurality of query words after the query contents are segmented through a special dictionary, firstly, an inverted index search engine is started, the query words are retrieved and subjected to relevance scoring in an inverted index base, after results with scores lower than a specified threshold value are filtered, search results are ranked according to the relevance scoring, and an inverted index search result list is obtained. The sequencing rule is as follows: sorting a plurality of matched parts in the same document in a reverse order according to the scores of the parts, and calculating the comprehensive score of the document; and sorting different documents in a reverse order according to the comprehensive scores of the documents.

Referring to fig. 7, a schematic diagram illustrating an embodiment of an inverted index search result list of an instant knowledge database is shown, where in an embodiment, the inverted index search result list mainly includes attributes such as document id, query term, abstract, score, and the like. The specific instant knowledge document can be associated through the document id. And the inverted index query result display interface is used for displaying the ordered search result list generated by the inverted index search, and the selected instant knowledge document can be opened through the document id of the search result list. When the instant knowledge document is opened through the reverse index query result display interface, a forward index search mode is synchronously started, a plurality of query words which are subjected to word segmentation of the special dictionary are input into a forward index library for retrieval, correlation calculation and result filtering with the score lower than a specified threshold value, and then the results are sorted, so that a forward index search result list with preset index granularity (such as each row in the document) can be obtained, and the forward index search result sorting can be sorted according to the score and can also be sorted according to the position of the search result appearing in the document.

In one embodiment, the instant knowledge document display interface is used for displaying instant knowledge documents with original formats (such as PDF formats), when the instant knowledge document display interface is opened from the reverse index query result display interface, in addition to displaying the instant knowledge documents with the original formats (such as PDF formats), a forward index search result list is also displayed, and the search words can be positioned by the forward index search result list to be displayed at corresponding positions of the instant knowledge documents. And inputting a new query word in the instant knowledge document display interface, and starting a new in-document search by using a forward index search mode.

In the method for managing the instantaneous knowledge of the nuclear power plant, disclosed in one embodiment of the application, firstly, instantaneous knowledge data is obtained; then loading the instant knowledge data into an instant knowledge database according to the time sequence; and finally, searching in the knowledge database according to a preset index word, and displaying the information of the search result. The method comprises the steps of firstly identifying, screening and extracting various kinds of instant information of the nuclear power unit acquired by channels such as an early meeting, a production week meeting, an experience feedback example meeting, a major repair daily report and the like of the nuclear power plant to generate a plurality of instant learning type knowledge documents, constructing an instant knowledge database, and then quickly retrieving and displaying the whole database full text of the instant knowledge documents through indexing the constructed instant knowledge database, so that professionals in the nuclear power plant can quickly learn and know core work, actual problems, current unit or equipment states and other core knowledge and skills such as principles, equipment failure modes, possible risks, intervention methods, historical feedback, operation key points and the like which are related to own duties and are carried out by the nuclear power plant, and the pertinence, timeliness, practicability and effectiveness of nuclear power knowledge training are improved.

Those skilled in the art will appreciate that all or part of the functions of the various methods in the above embodiments may be implemented by hardware, or may be implemented by computer programs. When all or part of the functions of the above embodiments are implemented by a computer program, the program may be stored in a computer-readable storage medium, and the storage medium may include: a read only memory, a random access memory, a magnetic disk, an optical disk, a hard disk, etc., and the program is executed by a computer to realize the above functions. For example, the program may be stored in a memory of the device, and when the program in the memory is executed by the processor, all or part of the functions described above can be implemented. In addition, when all or part of the functions in the above embodiments are implemented by a computer program, the program may be stored in a storage medium such as a server, another computer, a magnetic disk, an optical disk, a flash disk, or a removable hard disk, and may be downloaded or copied to a memory of a local device, or may be version-updated in a system of the local device, and when the program in the memory is executed by a processor, all or part of the functions in the above embodiments may be implemented.

Referring to fig. 8, a schematic structural connection diagram of an instant knowledge management system of a nuclear power plant in an embodiment is shown, and in an embodiment of the present application, a system for managing instant knowledge of a nuclear power plant is further disclosed, which is used for implementing management of the instant knowledge of the nuclear power plant by applying the above-mentioned method for managing instant knowledge of a nuclear power plant, and includes an instant knowledge acquisition device 100, a data loading device 200, and a data retrieval device 300. The instant knowledge acquisition device 100 is used for acquiring instant knowledge data, wherein the types of the instant knowledge data comprise conference summary data, operation parameter data of each nuclear power unit, parameter change data of each nuclear power unit and/or overhaul data of a nuclear power plant. The data loading device 200 is configured to load the instant knowledge data into an instant knowledge database according to a chronological order, where the instant knowledge database is established according to the historically acquired instant knowledge data. The data retrieval device 300 is configured to perform retrieval in the knowledge database according to a preset index word, and display information of a retrieval result. The retrieval result information comprises a document id, a query index entry, a result abstract and/or each document score, and the document scores are used for identifying the credibility of the documents.

Example two

Referring to fig. 9, which is a schematic diagram illustrating a structural connection of an instant knowledge management system of a nuclear power plant in another embodiment, the instant knowledge management system of a nuclear power plant includes an instant information collection module 30, an instant knowledge point extraction module 40, an instant knowledge point evaluation and assignment module 50, an instant knowledge document generation and publishing module 60, a knowledge base management module 10, and a retrieval and display module 20. The instant information acquisition module 30 is responsible for regularly acquiring three types of instant information including unstructured texts, structured texts and semi-structured texts of the nuclear power generating unit from channels such as an early meeting of a nuclear power plant, a production week meeting, an experience feedback example meeting, various daily reports and the like. The input of the instant information acquisition module is various format (such as PDF, PPT, excel, word, outlook mail and other formats) materials of a nuclear power plant early meeting, a production week meeting, an experience feedback example meeting, a daily newspaper and the like, a plurality of instant information texts are extracted from the materials and output to the instant knowledge point extraction module.

The instant knowledge point extraction module 40 is responsible for preprocessing the instant message text, segmenting the text into words, extracting core words and generating an instant knowledge point list. Firstly, preprocessing unstructured, semi-structured and structured text data to obtain a plurality of units to be participled; secondly, for each unit to be participled, firstly extracting an equipment function position code (if the equipment function position code exists) in the unit to be participled according to an equipment code rule, analyzing attribute information such as a system name, the equipment code, an equipment type and the like of the unit to be participled according to the equipment function position code, storing the attribute information as a core word, and mapping the attribute output to specific attributes of a knowledge point list; and thirdly, reserving the functional position codes of the equipment (as a whole), performing word segmentation on all or the rest of the unit to be segmented by using a special dictionary, performing weight calculation and sequencing on segmentation results, screening words with weights larger than a specific threshold value as core words, and mapping a plurality of matched core words into specific attributes of the knowledge points through entry categories of the core words. In one embodiment, the input of the instant message collection module 30 may be a user query word that meets a specific condition and is input by the user through the search and display module. For example, after the user inputs the query word, if the search result (or the knowledge point matching result) of the search and display module is smaller than the specified threshold (if the search result is not obtained), the text of the query word and the information (professional and department) of the user can be used as a unit to be segmented, and the instant knowledge point list is generated through the second step and the third step. In one embodiment, the output of the instant knowledge point extraction module is a plurality of instant knowledge point lists with a plurality of attributes.

The instant knowledge point evaluation and assignment module 50 is responsible for evaluating the instant knowledge point lists with several attributes generated by the instant knowledge point extraction module and deciding whether to assign to a specific responsible person to create or update a knowledge document. And inquiring the knowledge base aiming at each instant knowledge point and the key attributes thereof, and evaluating the instant knowledge points according to the coincidence degree of the inquiry result and the knowledge points to be evaluated. The evaluation result is divided into covered, to-be-updated and to-be-created. And (3) covering: the content of the existing instant knowledge document can fully cover the knowledge points to be evaluated without further processing; and (3) waiting for updating: the method comprises the steps that although a part of instant knowledge documents contain knowledge points to be evaluated, the pertinence, the depth and the breadth of the instant knowledge documents cannot meet the learning and using requirements of the instant knowledge points, and the existing relevant instant knowledge documents need to be upgraded and perfected; to be created: and indicating that the related content of the instant knowledge point to be evaluated is not found in the existing knowledge document, and establishing the instant knowledge document aiming at the instant knowledge point to be evaluated. And when the knowledge point evaluation result is to be updated or created, assigning a designated person to write or update the instant knowledge document according to the attribute of the person in charge in the instant knowledge point list, and entering an instant knowledge document generating and publishing module.

The instant knowledge document generation and publishing module 60 is responsible for compiling, checking, examining and publishing the instant knowledge document. The instant knowledge document types comprise a knowledge card, a tool card, a case card, a security analysis card, a human factor prevention knowledge card, an SPV case card and the like, each type of instant knowledge document has a fixed template and format (such as PDF, VISIO and Excel), the instant knowledge document types are generated according to compiling upgrade and release guide rules of the unified instant knowledge document, a proper document type and template are selected according to knowledge point attributes of the instant knowledge document, each instant knowledge document can be associated with a plurality of knowledge point attributes according to the content of the instant knowledge document, when the instant knowledge document is released, instant knowledge document related information is extracted and generated according to the format of a knowledge base, a modification interface of a knowledge base management module is called to write the instant knowledge document into the knowledge base or replace the corresponding instant knowledge document in the knowledge base, and finally related personnel or potential users are informed that the instant knowledge document is released.

The knowledge base management module 10 is responsible for managing the contents of a knowledge base, and the knowledge base comprises an instant knowledge document list 12, instant knowledge document contents 13, a reverse index base 14, a forward index base 15 and a special dictionary 16. The knowledge base management module 10 provides a modification (add \ delete \ modify) interface 17 and a query interface 11 for external module calls.

The modification (add/delete/modify) interface 17 is used to write the related information of the instant knowledge document into the instant knowledge document list, keep the content of the instant knowledge document in the original format, encrypt and compress the storage in a binary mode, read the pure character content of the instant knowledge document to generate and update the inverted index, analyze and read the content of the instant knowledge document with formats (such as PDF, VISIO, excel) and generate and update the forward index library with specific granularity (paragraph, line, word). In addition, the modification interface is also used for updating the dedicated dictionary.

The query interface 11 is used to retrieve knowledge base content, score relevance, filter results, and return a search result list. The query interface uses a double-search engine mode, firstly, the words of the query content are segmented by a special dictionary to obtain a plurality of query words, and then a reverse index search engine or a forward index search engine is started according to requirements, wherein the reverse index search engine is mainly used for general search of a whole library (across documents), and the forward index search engine is mainly used for fine search in the documents.

In one embodiment, the knowledge base may include other nuclear power knowledge documents, such as teaching materials, design files, technical specifications, digital programs, experience feedback, and the like, in addition to the instant knowledge document. And adding a document category attribute in the knowledge base document list for screening/distinguishing the types of the knowledge documents, realizing the retrieval and display of the full-knowledge type and expanding the application range of nuclear power knowledge.

The retrieval and display module 20 is used for receiving user query input, calling a knowledge base query interface to retrieve and displaying search results or instant knowledge documents. In one embodiment, the retrieve and display module 20 includes two search modes, an inverted index search mode for a full-corpus summary search and a forward index search mode for a fine search within a document. In the reverse index search mode, searching and relevance scoring are carried out on query words in a reverse index library, after results with scores lower than a specified threshold value are filtered, the search results are ranked according to the scores, a reverse index search result list is generated, the reverse index search result list mainly comprises attributes such as document id, query words, abstracts and scores, and specific instant knowledge documents can be associated through the document id. In the forward index search mode, after inputting a plurality of query words which have been subjected to word segmentation of the special dictionary into a forward index library for retrieval, calculating correlation, filtering results with scores lower than a specified threshold value, and then sorting the results in a reverse order, a forward index search result list with preset index granularity (such as each row in a document) can be obtained. In addition, in addition to generating the search result list in reverse order according to relevance, the search result list may also be generated in the order of where the search results appear within the document.

The display module of the retrieval and display module 20 comprises a query interface (for receiving and displaying query text input by a user), an inverted index search result display interface and an instant knowledge document display interface. The instant knowledge document display interface is used for displaying the instant knowledge document with an original format (such as a PDF format), and the interface also comprises a forward index search result display interface, so that the in-document search is realized, and the instant knowledge document can be positioned to the corresponding position of the instant knowledge document to be displayed through a forward index search result list. In one embodiment, the retrieval and display module 20 further includes a conventional combined search function of search term filtering, search in result, conditional search, fuzzy search, etc., and may filter the search results according to document classification, display the top of the knowledge document containing specific search terms/labels/classifications, highlight label display of the matched search terms on the display interface of the inverted index search result and the display interface of the instant knowledge document, and perform functions of collecting/canceling collection/managing collection of the instant knowledge document.

In one embodiment, the instant knowledge document display interface includes a content feedback function, and when a user finds that the content of the instant knowledge document is wrong, deviated or has good suggestions, the content-related information (including an author, a document position, document content, tags and the like) can be extracted to generate a data packet and sent to the author (such as through mail) in a specific format so as to ensure that the knowledge document is continuously improved.

In one embodiment, the retrieval interface can comprise modules such as a weekly attention module and the like, and can package and push related knowledge points of the recent hot spot work of the unit to related personnel.

In one embodiment, the knowledge base management module further has a function enhancement of a nuclear power equipment function position code, and specifically includes:

the nuclear power plant equipment function position codes have uniqueness in various nuclear power knowledge documents and nuclear power DCS control logic, and can represent specific equipment and affiliated systems, so that the equipment function position codes are subjected to targeted function expansion in the instant knowledge management system of the nuclear power plant. For example:

establishing a nuclear power plant equipment function position coding database, wherein the data source of the function position coding database is a unit DCS control logic database or a system design file;

and associating, complementing and correcting the device function position codes in the search terms: when a user inputs a search word, the system name, the function position code, the equipment type and other automatic completion functions can be performed, for example, when the user inputs 'KC', the system name 'KCO/KCM/KCQ' is automatically associated; when the user inputs 'RCV 131', searching the database function position code, and automatically completing the function position code 'RCV 1314/RCV1315/…'; when the device function position code input by the user does not exist, the function position code closest to the search word is automatically associated from the device function position code database, such as the RCV1314VR does not exist, and is automatically corrected to the RCV1314 VP.

Expansion of device function location codes in search terms: the pressurized water reactor nuclear power plant is generally arranged in a plurality of loops, the types of equipment of all the loops are consistent, the equipment is distinguished by columns in codes, and related knowledge is universal. For example, "RCP1110 PO/RCP 2110 PO/RCP 3110 PO/RCP 4110PO" represents 4 loop (4 rows) of equipment, which have the same function, the same type, and the same maintenance method, and the knowledge points (skills, notices, experience feedback, etc.) in one row can be applied to the other row. Based on the method, when a user inputs a search word of 'RCP 2110 PO', the function position code of the equipment is automatically expanded and associated to be 'RCP 1110PO | RCP2110PO | RCP3110PO | RCP4110 PO' 4 equipment to be searched in a knowledge base, and a search result containing one of any 4 rows of equipment is returned;

the device function position code weight and mark self-adaptation in the search word: and in the dual-engine mode, the weight of each attribute is self-adaptive to the function position code of the equipment in the search word. When the search word contains the equipment function position code, the equipment function position code is divided into a system name, an equipment code and an equipment type for searching, and the weight of the system name is increased in an inverted index search mode (full-base search), so that the instant knowledge document of the system can be recalled. When a forward index search mode (accurate search) is used, the weight of the system name is reduced, the weight of the equipment coding is improved, the search efficiency is improved, accurate matching of the equipment function position code is ensured, and the interference of the system name to the search result is reduced. For example, when searching the device function position code "KRT1228KA", in the inverted index mode, the weight of "KRT" is increased, so that the knowledge document containing the "KRT" system is easier to be searched, when opening the knowledge document from the inverted index search result interface, because the knowledge document is expected to definitely contain "KRT", when searching the inside of the knowledge document, the weight of "KRT" is decreased and the weights of "1228" and "KA" are increased, so as to search and locate the search content more quickly, and reduce the interference of the system name "KRT" to the search result.

In one embodiment, the device functional position coding enhancement comprises:

when the knowledge document is displayed, the device function position code enhancement is carried out on the text content in the view or in the selected range, if the device function position code is detected in the text content, other knowledge documents containing the device function position code are listed, and related pages of the device function position code are opened by associating with other related application software. For example, in an instant knowledge document display interface, when a user selects a certain segment of text of an instant knowledge document, the device function position code recognition is performed, and if the device function position code is found to exist in the selected text, the following function enhancements are performed:

1) Identifying an instant knowledge document list containing the equipment function position code (including equipment function position code extension) in a knowledge base;

2) And (4) associating other application programs, and opening and positioning a page related to the device function position codes, for example, opening and positioning a page where the device function position codes of DCS control logic offline query analysis software or digital program query software are located.

Through the retrieval and display module, the retrieval and display of the nuclear power knowledge points can be realized.

In an embodiment, the real-time knowledge management system of the nuclear power plant can be deployed in the same computer, and all functions are realized by a single computer. Further, as an alternative, a client server mode can also be used, the knowledge base is deployed on a special server, and the retrieval and display module is deployed on the client, so that the real-time search and display functions of multiple users are realized.

The real-time knowledge management system for the nuclear power plant disclosed in the embodiment can achieve the following beneficial effects:

1. through the rapid production, generation and use of the content of the instant knowledge document, the problem of knowledge timeliness existing in the traditional learning and training mode is effectively solved, the pertinence is strong, the iteration is rapid, the depth and the breadth of the instant knowledge document are more suitable for the operation needs of a nuclear power unit and the personalized knowledge needs of students, and the instant knowledge document can work around the center of the unit;

2. through the innovation of the retrieval and display modes of the instant knowledge, the 'dominance' and 'instrumentization' of the nuclear power knowledge are realized, the complex knowledge is disassembled into zero, the cognitive load of each professional is reduced, and the convenience of knowledge use is improved.

The invention has been successfully implemented in the third-generation EPR (European pressurized water reactor) nuclear power plant, and realizes the organic integration of the inquiry and use of the nuclear power instant knowledge and the actual work of the unit. Before daily work, all professionals can quickly know core knowledge such as operation principles, fault modes, possible risks, intervention methods, historical feedback, operation key points and the like by inquiring relevant nuclear power instant knowledge documents, and the pertinence, timeliness and knowledge management efficiency of nuclear power knowledge are improved.

The present invention has been described in terms of specific examples, which are provided to aid understanding of the invention and are not intended to be limiting. For a person skilled in the art to which the invention pertains, several simple deductions, modifications or substitutions may be made according to the idea of the invention.

Claims (10)

1. A nuclear power plant instant knowledge management method is characterized by comprising the following steps:

acquiring instant knowledge data; the types of the instant knowledge data comprise conference summary data, operation parameter data of each nuclear power generating unit, parameter change data of each nuclear power generating unit and/or overhaul data of a nuclear power plant;

loading at least one instant knowledge data into an instant knowledge database according to the sequence of time corresponding to the instant knowledge data; the instant knowledge database is established according to historically acquired instant knowledge data;

searching in the knowledge database according to a preset index word, and displaying search result information; the retrieval result information comprises a document id, a query index entry, a result abstract and/or each document score; the document score is used to identify the trustworthiness of the displayed document.

2. The method of immediate knowledge management of a nuclear power plant of claim 1, wherein the obtaining of immediate knowledge data comprises:

identifying, screening and/or extracting the instant knowledge document according to a preset format and type to generate instant knowledge data; the instant knowledge documents comprise conference summary documents, nuclear power unit operation parameter documents, nuclear power unit parameter change documents and/or nuclear power plant overhaul documents;

the instant knowledge document is generated according to preset compiling rules, upgrading rules and releasing rules and has fixed document types and templates.

3. The immediate knowledge management method for a nuclear power plant as claimed in claim 2, wherein the acquisition method for the meeting summary data comprises:

acquiring an early meeting summary, a production week summary, an experience feedback regular meeting summary and/or a temporary emergency meeting summary of a nuclear power plant;

converting the preschool era, the production week era, the experience feedback regular meeting era and/or the temporary emergency meeting era into a meeting era text according to a preset text format, and taking the meeting era text as the meeting era data; the conference summary data comprises conference time, a conference place, participants, the types and the titles of the participants, and the revision content of the number of the nuclear power generating unit and/or the operation parameters of the nuclear power generating unit related to the conference;

the operation parameter data of each nuclear power unit is the operation parameter data of each nuclear power unit;

the parameter change data of each nuclear power generating unit are operation parameter change data of each nuclear power generating unit, and the operation parameter change data comprise data before change and data after change;

the overhaul data of the nuclear power plant comprise overhaul daily report data, overhaul summary data, overhaul plan refueling data, overhaul refueling completion data, overhaul plan test data and overhaul test completion data.

4. The method of claim 2, wherein the identifying, screening and/or extracting the immediate knowledge documents in a predetermined format and type to generate the immediate knowledge data comprises:

the format of the instant knowledge data comprises structured data, semi-structured data and/or unstructured data; the structured data has fixed fields and formats, and the text at the specified position in the structured data has a determined meaning; the semi-structured data has no fixed format and length, but comprises key identification with characteristic key words, and the key identification is used for identifying that the text at a preset position has a determined meaning; the unstructured data has no fixed format and length and no characteristic keywords;

for the unstructured data, reading all texts of the unstructured data by adopting a direct streaming mode, wherein the read all texts serve as a unit to be participled;

performing stream reading on the semi-structured data, segmenting a text at a preset position according to a key identifier in the semi-structured data to obtain text segments, and taking each text segment as a new unit to be segmented;

for the structured data, directly extracting texts from all positions according to the format and meaning of the structured data, and taking the extracted texts as word segmentation units;

screening, segmenting and/or defining the unit to be segmented according to a preset special dictionary, and then generating the instant knowledge data;

and directly generating the instant knowledge data for the word segmentation unit.

5. The method of managing instantaneous knowledge of a nuclear power plant of claim 4, wherein the loading at least one of the instantaneous knowledge data into the instantaneous knowledge database in chronological order of the time corresponding to the instantaneous knowledge data comprises:

the instant knowledge database comprises an instant knowledge point list;

the instant knowledge database maps the instant knowledge data to the instant knowledge point list according to the category of the entries and the specific attribute of the entries;

the knowledge point list is used for recording instant knowledge points, and each instant knowledge point is distinguished according to the category of the vocabulary entry and the specific attribute of the vocabulary entry; the instant knowledge points comprise entry names, affiliated systems, directly related departments, potential influence departments, personnel responsible for knowledge point identification, entry sources and/or original data texts.

6. The immediate knowledge management method of a nuclear power plant as claimed in claim 5, wherein the immediate knowledge database further comprises at least one instant knowledge document list for recording the content of an instant knowledge document, an inverted index, a forward index and/or a special dictionary;

each instant knowledge document list comprises a globally unique document ID, a document code, a name, a subject, a version, a system to which the instant knowledge document list belongs, an associated knowledge point and/or a document type;

the recording instant knowledge document comprises at least one instant knowledge point list.

7. The method of immediate knowledge management of a nuclear power plant of claim 6, wherein the immediate knowledge database further comprises an inverted index repository and/or a forward index repository;

the inverted index library is used for forming a word list with preset text attributes by using a special dictionary to perform word segmentation on all plain text characters contained in the instant knowledge document, and the preset text attributes comprise a document ID list, text word frequency and/or text positions;

the forward index library comprises mapping information of text content segments and specific positions of the text content segments in the corresponding recording instant knowledge document, and the mapping information is used for positioning the text content segments to the specific positions of the text content segments in the corresponding recording instant knowledge document through contents.

8. The method of immediate knowledge management of a nuclear power plant of claim 7, wherein the immediate knowledge database further includes basic knowledge documents including nuclear power knowledge related textbooks, nuclear power plant design files, nuclear power plant technical specifications, nuclear power plant digitization procedures, and/or nuclear power plant experience feedback documents.

9. A computer-readable storage medium, characterized in that the medium has a program stored thereon; the program is executable by a processor to implement the nuclear plant immediate knowledge management method of any of claims 1-8.

10. An instantaneous knowledge management system for a nuclear power plant, comprising:

the instant knowledge acquisition device is used for acquiring instant knowledge data; the types of the instant knowledge data comprise conference summary data, operation parameter data of each nuclear power generating unit, parameter change data of each nuclear power generating unit and/or overhaul data of a nuclear power plant;

the data loading device is used for loading the instant knowledge data into the instant knowledge database according to the time sequence; the instant knowledge database is established according to the historically acquired instant knowledge data;

the data retrieval device is used for retrieving in the knowledge database according to a preset index word and displaying retrieval result information; the retrieval result information comprises document id, query index entries, result summaries and/or scores of each document; the document score is used to identify the trustworthiness of the document.

Images