CN114580694A - Collective dose prediction model based on radiation work license management - Google Patents

Collective dose prediction model based on radiation work license management Download PDF

Info

Publication number
CN114580694A
CN114580694A CN202011390103.5A CN202011390103A CN114580694A CN 114580694 A CN114580694 A CN 114580694A CN 202011390103 A CN202011390103 A CN 202011390103A CN 114580694 A CN114580694 A CN 114580694A
Authority
CN
China
Prior art keywords
rwp
overhaul
historical
total
target
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.)
Granted
Application number
CN202011390103.5A
Other languages
Chinese (zh)
Other versions
CN114580694B (en
Inventor
肖薇
张冀兰
杨加东
刘华
葛旭阳
张廉
杨萌
杨沥铭
刘晗
张立侠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CNNC Nuclear Power Operation Management Co Ltd
Original Assignee
CNNC Nuclear Power Operation Management Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CNNC Nuclear Power Operation Management Co Ltd filed Critical CNNC Nuclear Power Operation Management Co Ltd
Priority to CN202011390103.5A priority Critical patent/CN114580694B/en
Publication of CN114580694A publication Critical patent/CN114580694A/en
Application granted granted Critical
Publication of CN114580694B publication Critical patent/CN114580694B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention belongs to the technical field of nuclear power operation and maintenance, and particularly relates to a collective radiation dose prediction model based on radiation work license management. The technical scheme is as follows: acquiring target overhaul, historical overhaul RWP information, historical overhaul total aggregate dosage and total working hours; matching the information in the RWP item by item; setting main fields and weights of the target overhaul RWP, and calculating the matching degree and total matching score of each field in the historical RWP and the target RWP item by item; and selecting the historical overhaul RWP with the highest score, and extracting the total collective dose and the total work of the historical overhaul RWP as the prediction result of the target overhaul RWP. The method is mainly suitable for collective dose management in the field of radiation protection of nuclear power plants during overhaul, and can meet the requirement of radiation protection indexes related to radiation dose formulated by radiation protection departments of the nuclear power plants; manpower and time for index formulation can be saved, and burden of management engineers is reduced.

Description

Collective dose prediction model based on radiation work license management
Technical Field
The invention belongs to the technical field of nuclear power operation and maintenance, and particularly relates to a collective dose prediction model based on radiation work license management.
Background
According to the national standards, registrars, licensees and human users should be responsible for arranging professional exposure monitoring and evaluation of workers according to the specifics of the practices and sources for which they are responsible. The personal radiation dose monitoring is an important component of radiation protection of the nuclear power station, provides important reference basis for improving radiation protection measures, improving radiation protection effect and implementing radiation protection optimization, and simultaneously provides help for nuclear power station managers to make production plans, evaluate radioactive hazards and perform medical treatment.
The radiation protection management level of a company is fully evaluated through the indexes, and the nuclear power plant staff are ensured to be healthy and safe through comprehensive and effective radiation protection management measures adapted to the indexes. The radiation protection indexes related to the dosage of the overhaul individual mainly comprise collective dosage, total working hours and the like. At present, radiation protection indexes such as overhaul collective dosage and the like of a nuclear power plant are formulated by mainly referring to historical index values and historical experience data and formulating radiation protection indexes related to overhaul personal dosage through manpower and experience, and the radiation protection indexes related to overhaul personal dosage cannot be automatically predicted without a data model. In actual work, the using methods of various nuclear power plants are different, and the condition that the overall dose of overhaul breaks through the prediction index often occurs. Therefore, there is a need for a collective radiation dose prediction model to enable informatization and automation of individual dose-related radiation protection index predictions, and to improve the accuracy of the predictions.
Disclosure of Invention
The invention aims to provide a collective radiation dose prediction model based on radiation work license management, which aims at solving the problems of difficulty in making dose-related radiation protection indexes of a nuclear power plant, no data model and no informatization means, relies on the standardized management flow of Chinese nuclear power, and is based on the radiation work license (RWP for short) management of a safety production platform of the nuclear power plant.
The technical scheme of the invention is as follows:
a collective radiation dose prediction method based on radiation work license management sequentially comprises the following steps:
s1, acquiring target overhaul RWP information and historical overhaul RWP information, and summarizing and calculating the total collective dose and total working hours of each RWP in historical overhaul;
s2, matching the information in the target overhaul RWP and the historical overhaul RWP acquired in the S1 item by item; when the historical overhaul RWP completely consistent with the target overhaul RWP exists, taking the total aggregate dose and the total working hours of the historical overhaul RWP as the total aggregate dose and the total working hours prediction result of the project standard overhaul RWP; when the historical overhaul RWP completely consistent with the target overhaul RWP does not exist, the next step is carried out to carry out similarity matching;
s3, matching the similarity between the target overhaul RWP and the historical overhaul RWP,
s3.1, setting fields to be matched in the target overhaul RWP and weight assignment w of each field;
s3.2, calculating the matching degree v of each field in the historical overhaul RWP item by item and the target overhaul RWP, calculating the matching score u-wv of each field, and calculating the total matching score u of the single historical overhaul RWPGeneral assemblyIs the sum of field match scores;
s3.3 Total matching score u of each item of historical overhaul RWP calculated from S3.2General assemblyThe highest u is selectedmaxAnd extracting the total aggregate dose and the total working hours of the corresponding historical overhaul RWP as a prediction result of the total aggregate dose and the total working hours of the target overhaul RWP.
Further, before similarity matching between the target overhaul RWP and the historical overhaul RWP, the S3 identifies whether a preventive maintenance number is included in a work order associated with the target overhaul RWP; if not, the step S3 is carried out to match the similarity of the target overhaul RWP and the historical overhaul RWP; if yes, preventive maintenance number matching is carried out, a work order with the same preventive maintenance number is searched in work orders related to the historical overhaul RWP, and the searched total collective dose and total working hours of the historical overhaul RWP are used as the prediction result of the target overhaul RWP; and if not, the step S3 is carried out to match the similarity of the target overhaul RWP and the historical overhaul RWP.
Further, the fields in S3 for setting matching in the target overhaul RWP include RWP description, device code, job category, and room.
Further, when the historical overhaul RWP information in S1 is obtained, multiple matching historical overhaul records are selected from the nuclear power plant information management system according to the target overhaul object, the unit type, and the overhaul type, and the historical overhaul RWP information is obtained from the selected multiple matching historical overhaul records.
Further, the step S1 includes a step of correcting the acquired historical overhaul RWP information, in which a signed correction value is entered for the collective dose and the total man-hour corresponding to the work category in the historical overhaul RWP information according to factors in the work category, i.e., the special work or the abnormal event in the historical overhaul, so as to correct the total collective dose and the total man-hour of the historical overhaul RWP referred to in S2.
Further, the historical overhaul RWP information in S1 further includes information of neutron dose and internal irradiation dose.
Further, the total matching score u of a single historical overhaul RWP in S3.2General assemblyA total score u of RWP description match for the historical overhaulDescription of the inventionAnd the total score u of the matching degree of the equipment numbersDeviceThe score u of the degree of matching of job typeType of operationScore u of degree of matching with roomType of operationAnd (4) summing.
Further, the historical overhaul RWP describes a total score u of the degree of matchDescription of the inventionThe calculation steps of (2) are as follows:
s3.2.1 learning and training the machine learning model based on the nuclear power professional lexicon, and then matching the target overhaul RWP with the historical overhaul RWP, thereby obtaining the initial score u of the matching degree described by the non-universal RWPDescription of the invention’；
S3.2.2 selecting one or more keywords from RWP of target overhaul RWP, comparing one by one whether these keywords exist in RWP description of historical overhaul RWP, if yes, setting similarity v of the keywordKeywordIf not, the similarity v of the keyword is set as 1KeywordThen, a matching degree score w of the keyword is calculated as 0Keyword＝wKeyword 1×vKeyword 1+wKeyword 2×vKeyword 3+……+wKeyword n×vKeyword n(n is a natural number);
s3.2.3 keyword matching score w calculated in step 3.2.2KeywordAnd (3) preliminarily scoring the matching degree obtained by the calculation in the step (3.2.1)Description of the invention' make a correction to obtain a corrected match score u for the RWP description of the historical overhauled RWPDescription of the invention", the calculation formula is:
udescription of the invention”＝wDescription of the preferred embodiment×vDescription of the preferred embodiment+wKeyword；
S3.2.4 comparison of u calculated in step 3.2.3Description of the invention"and when vDescription of the inventionCalculated when 1 is satisfiedDescription of the invention', take the total score u of matching degree of RWP description of the smaller of the two as the historical overhaul RWPDescription of the invention。
Further, the step 3.2.1 adopts a semantic similarity technology to match the similarity of the target overhaul RWP and the historical overhaul RWP.
Further, the semantic similarity technique includes the steps of: firstly, performing word segmentation processing on word descriptions in RWP descriptions of a target overhaul RWP and a historical overhaul RWP by a Chinese word segmentation engine after nuclear power professional lexicon optimization; then, the word frequency vectors of the two vectors are respectively calculated based on the TF-IDF algorithm, and the similarity calculation v is carried out on the two vectorsDescription of the inventionThereby resulting in matching of the RWP descriptionsDegree preliminary score uDescription of the invention', the calculation formula is: u'Description of the invention＝wDescription of the invention×vDescription of the preferred embodiment。
Further, S3.2.1 includes a step of removing redundant information in the RWP descriptions of the target overhaul RWP and the historical overhaul RWP, that is, removing literal information in the RWP descriptions that overlaps with other fields before matching the target overhaul RWP and the historical overhaul RWP.
Further, the total score u of the device number matching degreeDeviceThe calculation steps are as follows:
s3.2.5 extracting the equipment number from the work order of the target overhaul RWP and the historical overhaul RWP;
s3.2.6 matching the S3.2.5 extracted target overhaul RWP with the equipment number in the historical overhaul RWP, setting the number of the extracted equipment numbers of the target overhaul RWP as n, and the equipment number matching degree score u of the historical RWPDevice iThe calculation formula is as follows:
when all the device numbers except the first unit number in the historical RWP are the same, the device number matching degree score of the historical RWP
Figure RE-GDA0002968676090000051
When the first digit group number and the middle three digit code of a single equipment number in the historical RWP are different and the rest are the same, the equipment number matching degree score of the historical RWP
Figure RE-GDA0002968676090000052
When the device number except the first unit number, the three-bit system code and the two-bit device type code in the single device number in the historical RWP are different, the device number matching degree score of the historical RWP
Figure RE-GDA0002968676090000053
If the value of the "RWP description" match score equals w in step 3.2.4 when the device number is "XXX 000XXXDescription of the inventionThen the device number match score u for the historical RWPDevice iIs set to wDevice；
If the conditions are not met, the matching is not successful, and the equipment number matching degree score u of the historical RWPDevice iIs 0;
s3.2.7 calculating the total score u of the matching degree of the device number of the historical RWPDeviceThe calculation formula is as follows:
Figure RE-GDA0002968676090000054
further, the S3.2.5 rule for extracting the device number from the work order associated with the target overhaul RWP and the historical overhaul RWP is: if the 'equipment number' is not null and is a single equipment number, taking the value as the equipment number; if the equipment number is not null and is a work item code, identifying and extracting all equipment numbers related to the work item; if the equipment number is empty, automatically identifying whether the equipment number meeting the power plant number rule exists in the worker RWP description; otherwise, the device code match score of the historical RWP is 0.
Further, the matching degree score u of the job typeType of operationThe calculation steps are as follows: first, when the "job type" of the target overhaul RWP has n different job types separated by commas and matches m job types in the historical overhaul RWP, the "job type" matching degree score u of the historical overhaul RWPType of operationThe calculation formula is as follows:
Figure RE-GDA0002968676090000055
if the "RWP description" match score in step 3.2.4 equals w when the "job type" of the target overhaul RWP and the historical overhaul RWP do not match successfullyDescription of the inventionAnd the matching degree score value u of the job type of the historical overhaul RWPType of operationIs set as wType of operationOtherwise, the matching degree score u of the job type of the history overhaul RWPType of operationIs set to 0.
Further, the room matching degree score calculation rule is as follows:
when the value of the "RWP description" match score in step 3.2.4 equals wDescription of the inventionThe value u of the matching degree score of the "room" of the historical RWPRoomIs set as wRoom；
The value u of the match score for the "room" of the historical overhaul RWP when the "room" of the target overhaul RWP is emptyRoomSet to 0;
when the two conditions are not met, the 'room' fields of the target overhaul RWP and the historical overhaul RWP need to be matched, and when the fields are completely matched, the numerical value u of the matching degree score of the 'room' of the historical overhaul RWPRoomIs wRoomOtherwise, the score is 0.
Further, the computer system in S3.3 synthesizes and synthesizes the total matching scores u of each historical overhaul RWPGeneral assemblyAnd intelligently recommending historical data, and taking the total aggregate dose and the total working hours in the historical overhaul RWP with the highest matching degree score as the predicted total aggregate dose and the predicted total working hours of the target overhaul RWP.
Further, if there are n (n >1) historical overhaul RWPs with the highest similarity score in the RWP matching data finally recommended by the computer system, and the total aggregate dose and total man-hours of the historical overhaul RWPs are respectively E1, E2, … Ei …, En, H1, H2, … Hi … and Hn, the predicted total aggregate dose and total man-hour of the target RWP are arithmetically averaged by the n historical RWP total aggregate doses and total man-hours, and the calculation formula is as follows:
Figure RE-GDA0002968676090000061
Figure RE-GDA0002968676090000062
further, the total aggregate dose and total man-hour prediction result of the target major repair RWP obtained in S3.3 is the total aggregate dose and total man-hour prediction result of the single RWP in the target major repair, and the target major repair RWPs are added to obtain the total major repair aggregate dose and total man-hour prediction result.
Further, the target overhaul total overhaul collective dose PEGeneral assemblyPH of total man-hoursGeneral assemblyThe calculation formula is as follows:
Figure RE-GDA0002968676090000071
Figure RE-GDA0002968676090000072
wherein PEXn is the individual RWP collective dose, PE predicted by similarity matchNeutron of neutronThe neutron dose predicted value, PE, is overhauled for the targetInner partPredicted value of irradiation dose, EC, for the target major repairGeneral assemblyTotal working class collective dose correction, ECGeneral assemblyThe calculation formula of (2) is as follows:
Figure RE-GDA0002968676090000073
Figure RE-GDA0002968676090000074
ECnthe sum of the correction values for the working class collective doses entered for S2.1 of each RWP; HCnIs the sum of the correction values of the total working hours of each RWP.
The invention has the beneficial effects that:
according to the invention, radiation protection index prediction related to personal dose in overhaul of the nuclear power plant is realized and historical similar RWP, collective dose and total working hour push are carried out on a single RWP through intelligent experience feedback and big data analysis technologies. The method is mainly suitable for personal dose management in the field of radiation protection of the nuclear power plant during overhaul, can also be used for predicting personal radiation dose in daily maintenance, and can meet the requirement of radiation protection indexes related to the personal radiation dose formulated by radiation protection departments of the nuclear power plant. By adopting the prediction method to carry out personal radiation dose management, the manpower and time for index formulation can be saved, the burden of management engineers is reduced, and the RWP data is analyzed to realize the purpose of predicting the radiation protection index related to the personal dose in the overhaul of the nuclear power plant as accurately as possible.
Detailed Description
The present invention will be described in further detail with reference to specific examples.
A radiation work license (RWP) is a license system, i.e., RWP is handled whenever radioactivity related work is performed. For each work order in the control area, one RWP, i.e., a single certificate, needs to be handled. The embodiment provides a collective radiation dose prediction method based on radiation work license management, which sequentially comprises the following steps:
s1, acquiring RWP information of target overhaul and RWP information of historical overhaul, and summarizing and calculating total collective dose and total working hours of each RWP in the historical overhaul;
the RWPs include a non-common RWP that cannot be used commonly in each overhaul, that is, information included in the RWP in each overhaul differs, and a common RWP that can be used commonly in each overhaul, that is, information included in the RWP in each overhaul is completely identical.
The information in the non-generic RWP mainly includes: a RWP description, a device number, a room, a job type, and a preventive maintenance number, the information in the generic RWP including the RWP description.
And aiming at the target overhaul, recommending the previous overhaul meeting the conditions or selecting partial overhaul from the previous overhaul as a data source of the historical overhaul according to the information such as the overhaul object, the unit type, the overhaul type and the like. The historical overhaul and related data thereof can be directly and automatically selected by setting the preference combination condition as a default item by a user, and can also be manually selected in each prediction process.
1) Selecting a machine set: because the completeness of data preparation of each nuclear power plant is different and the condition of less overhaul sample data possibly exists, a user is supported to select single-unit or multi-unit data as input; the user can also set the preference combination condition as a default option;
2) and (3) selecting major repair types: due to the fact that the arrangement difference between the overhaul project and the overhaul project is large, the nuclear power plant classifies the overhaul, mainly comprises conventional overhaul, unconventional overhaul, five-year overhaul and ten-year overhaul, and supports a user to select a single overhaul type or all overhaul types as input; the combination of preferences may also be set by the user as a default option.
S2, correcting RWP information of historical overhaul;
s2.1 Total collective dose and Total manhours correction
The radiation protection department of the nuclear power plant can input a correction value with a symbol for the collective dose and the total working hours of each working category under the working category label of the historical overhaul according to the conditions of special work, abnormal matters and the like in the historical overhaul, so that the total collective dose and the total working hours of RWP (remote warning report) of the historical overhaul are corrected, and the influence of the special work and the abnormal conditions on the overhaul dose and working hour prediction is reduced;
s2.2 neutron dose and internal radiation dose
When predicting the collective dose index for the target overhaul, the neutron dose and the internal irradiation dose of the past overhaul should be considered in addition to the EPD dose. Especially in heavy water reactor nuclear power plants, tritium internal irradiation accounts for a large proportion during overhaul. Therefore, the radiation protection department can also input the neutron dose and the internal irradiation dose of the historical overhaul for predicting the total aggregate dose of the target overhaul;
s3, similarity matching is carried out on the information in the target overhaul non-general RWP (hereinafter referred to as 'target RWP') acquired in the S1 and the information in the historical overhaul non-general RWP (hereinafter referred to as 'historical RWP') acquired after being corrected in the S2 item by item, and the method comprises the following steps:
s3.1, fields needing specific matching in the non-general RWP and weight assignment of each field are determined;
first, fields for matching are selected from non-general RWP, and four words of RWP description, device code, job type and room are selected mainly in the embodimentA segment; then, according to the importance of the field information, carrying out weight assignment on each field, wherein the weight assignment is wDescription of the invention(wKeyword)、wDevice、wType of operation、 wRoomThe sum of all field weight assignments is 1.
The information amount of different fields is different and the possibility of information cross-overlapping exists, for example, the RWP description may have device encoding information, and in order to improve the efficiency and accuracy of matching, redundant information in different fields of each RWP may be removed when determining the fields, and may also be removed in the subsequent field matching process.
The RWP description is typically a textual description of the work order task and in some cases may be the same as the work order task header. In order to further improve matching accuracy, one or more keywords can be specified in the "RWP description" of the target RWP, and weight assignment is performed on each keyword.
S3.2 preventive maintenance work order matching
When the work order associated with the target RWP has a preventive maintenance number (PM number for short), the work order with the same PM number is searched in the work orders associated with the historical RWP, and the matching score u of the work order of the historical RWP preventive maintenance is usedGeneral assemblySetting the RWP to be 1, namely directly selecting the historical RWP as a target RWP matching object, and directly entering the step 5 by taking the total aggregate dose and the total labor hour of the historical RWP as a prediction result of the target RWP total aggregate dose and the total labor hour;
when the work order associated with the target RWP does not have a preventive maintenance number, matching other fields;
s3.3 RWP description matching
And automatically carrying out similarity matching on the character part of the historical RWP after redundant information is removed from the RWP description to obtain a matching score of the historical overhaul RWP description. The method comprises the following specific steps:
s3.3.1 redundant information in RWP descriptions of the target overhaul and the historical RWP, namely character information coincident with other fields, is removed, and the most common redundant information is the equipment number, so as to improve the accuracy of similarity matching.
S3.3.2 related to similarity matchingThe method mainly comprises the steps of carrying out learning training on the basis of nuclear power professional lexicon through a machine learning model, and then matching S3.3.1 non-universal RWP with redundant information removed, so that a primary matching degree score u described by the non-universal RWP is obtainedDescription of the invention’；
In the embodiment, a semantic similarity technology is adopted for similarity matching, firstly, a Chinese word segmentation engine optimized by a nuclear power professional lexicon completes word segmentation processing on word descriptions in RWP descriptions of a target RWP and a historical RWP, then word frequency vectors of the target RWP and the historical RWP are respectively calculated based on a TF-IDF algorithm, and similarity v is carried out on the two vectorsDescription of the inventionCalculating (the larger the calculated numerical value is, the higher the similarity is, and conversely, the lower the similarity is), so as to obtain the preliminary matching degree score u described by the non-general RWPDescription of the invention', the calculation formula is:
u′description of the preferred embodiment＝wDescription of the preferred embodiment×vDescription of the invention
S3.3.3 comparing RWP description of history RWP with whether there is the keyword determined in step 3.1, if so, the similarity v of the keywordKeywordIf the irrelevant key word is 1, the similarity v isKeywordThen, a matching score w of the keyword is calculated as 0Keyword n×vKeyword n(n is a natural number);
s3.3.4 the preliminary score of matching degree calculated in step 3.3.2 is corrected by the sum of the matching scores of the keywords calculated in step 3.3.3, so as to obtain the corrected matching score u described by the RWP of the historical RWPDescription of the preferred embodiment", the calculation formula is:
udescription of the invention”＝wDescription of the invention×vDescription of the invention+wKeyword 1×vKeyword 1+wKeyword 2×vKeyword 3+……+wKeyword n×vKeyword n
S3.3.5 comparison of u calculated in step 3.3.4Description of the invention"and when vDescription of the inventionCalculated when 1 is satisfiedDescription of the invention' the smaller of the two is taken as the total matching degree score u of the RWP description of the history RWPDescription of the inventionI.e., a smaller total score u of degree of match for RWP descriptions as historical RWPDescription of the inventionThat is to say uDescription of the preferred embodimentIf the value of (A) is greater than wDescription of the inventionThe value of (1) is then wDescription of the invention。
S3.4, matching device codes, specifically comprising the following steps:
s3.4.1 extracting the device number from the work order associated with the target RWP and the historical RWP, the rule is:
if the 'equipment number' is not null and is a single equipment number, taking the value as the equipment number;
if the "equipment number" is not null and is a work item code (which means that the RWP takes multiple equipment as work objects), then all equipment numbers associated with the work item are identified and extracted;
if the equipment number is empty, automatically identifying whether the equipment number meeting the power plant number rule exists in the worker RWP description;
otherwise, the device code matching score of the historical RWP is 0;
s3.4.2 device number matching score calculation: and matching the target RWP with the equipment numbers in the historical RWP, wherein when the target RWP identifies n equipment numbers, the matching principle is as follows:
when all the device numbers except the first unit number in the historical RWP are the same, the device number matching score u of the historical RWPDevice iThe calculation formula is as follows:
Figure RE-GDA0002968676090000121
when the device number in the history RWP is different except the first digit group number and the middle three digit code, and the rest is the same, the device number matching score u of the history RWPDevice iThe calculation formula is as follows:
Figure RE-GDA0002968676090000122
when the device number of the historical RWP is different from the first unit number, the three-bit system code and the two-bit device type code in the single device numberMatching score uDevice iThe calculation formula is as follows:
Figure RE-GDA0002968676090000123
when the device number is "XXX 000 XXX", if the value of the "RWP description" matching score in step 3.3.5 equals wDescription of the inventionThen the device number of the history RWP matches the score uDevice iIs set to wDevice；
If the conditions are not met, the matching is not successful, and the device number of the historical RWP matches the score uDevice iIs 0;
and calculating the total score of the equipment number matching degree of the historical RWP, wherein the calculation formula is as follows:
Figure RE-GDA0002968676090000124
s3.5, matching other fields, and specifically comprising the following steps:
s3.5.1 "Job type" field match
Firstly, removing information including 'other' and irregular information in 'operation type' so as to improve matching efficiency and accuracy;
then, when the "job type" of the target RWP has n different job types separated by commas and matches m job types in the history RWP, the "job type" of the history RWP matches the score uType of operationThe calculation formula is as follows:
Figure RE-GDA0002968676090000131
if the "RWP description" match score in step 3.3.5 equals w when the target RWP does not successfully match the "job type" of the historical RWPDescription of the inventionThen the matching score value u of the job type of the history RWPType of operationIs set as wType of operationOtherwise, the matching score u of the job type of the history RWPType of operationIs set to 0.
S3.5.2 the matching requirement for the "room" field is:
when the value of the "RWP description" match score in step 3.3.5 equals wDescription of the preferred embodimentThe value u of the matching score for the "room" of the historical RWPRoomIs set as wRoom；
When a "room" of the target RWP is empty, the value u of the matching score of the "room" of the history RWPRoomSet to 0;
when neither of the two conditions is satisfied, the "room" fields of the target RWP and the historical RWP need to be matched, and when the two fields are completely matched, the value u of the matching score of the "room" of the historical RWPRoomIs wRoomOtherwise, the score is 0;
s3.6 adding the scores of the fields to obtain the total matching score u of the historical RWPGeneral assemblyThe calculation formula is as follows:
ugeneral assembly＝uDescription of the preferred embodiment+uDevice+uType of operation+uRoom；
S4. non-generic RWP collective dose prediction
The computer system synthesizes the matching results of the plurality of historical RWP obtained in S3 to intelligently recommend the historical data, and takes the total collective dose and the total working hours in the historical RWP with the highest matching score as the prediction result of the total collective dose and the total working hours of the target RWP. In the RWP matching data finally recommended by the system, if n (n >1) historical RWPs with the highest similarity score exist, and the total aggregate dose and the total man-hour are respectively E1, E2, … Ei …, En and H1, H2, … Hi … and Hn, the predicted total aggregate dose and the predicted total man-hour of the target RWP are the arithmetic mean of the n historical RWP total aggregate doses and the total man-hour, and the calculation formula is as follows:
Figure RE-GDA0002968676090000141
Figure RE-GDA0002968676090000142
to further improve the prediction accuracy, EC may be further modified based on the schedule of the special work for the target major repair by setting the revised values of the collective dose and total man-hour information under the "work category" of the historical RWPGeneral assemblyThe influence of the special work of the target major repair on the target major repair dosage and working hour prediction is reduced;
s5. general RWP prediction without associated work order
Reading the total aggregate dose and total man-hour information of the historical overhaul generic RWP matched with the target overhaul generic RWP as the predicted total aggregate dose PE of the target overhaul generic RWPGeneral purposeAnd total man-hours PHGeneral purpose；
S6, predicting neutron dose and internal irradiation dose
The optimal predicted values of the neutron dose and the internal irradiation dose of the target overhaul are given by adopting algorithms such as a trend extrapolation method, a regression prediction method, a Kalman filtering prediction model or a combined prediction model;
s7, correcting the total work category
When the collective dose and the total working hours are predicted according to the working type or working class label, the value obtained by summing the collective dose and the total working hours under the working type label of the previous overhaul and the collective dose and the total working hours correction value recorded by a user is used as model input to predict the dose; the user is supported to freely select whether to select the data for correction; total work class collective dose, total man-hour correction parameter ECGeneral assembly、HCGeneral assemblyThe calculation formula is as follows:
Figure RE-GDA0002968676090000151
Figure RE-GDA0002968676090000152
ECns2.1 entry for RWPCorrection of working class collective dose of (a); HCnA corrected value for the total working hours of each RWP working class;
s8, the whole overhaul period comprises a plurality of target RWP, the total aggregate dose and the total labor hour prediction result of each target RWP are added according to the total aggregate dose and the total labor hour prediction result of the target RWP obtained in the step 4, and the total prediction result of the aggregate dose and the total labor hour in the whole overhaul period, namely the final overhaul aggregate dose PE is obtainedGeneral assemblyPH of total man-hoursGeneral assemblyThe calculation formula is as follows:
Figure RE-GDA0002968676090000153
Figure RE-GDA0002968676090000154
wherein PEXn is the collective dose of RWP predicted by similarity match, PENeutron of neutronThe neutron dose predicted value, PE, is overhauled for the targetInner partPredicted value of the irradiation dose, EC, for the target in-overhaulGeneral assemblyThe total correction introduced for collective dose for some special work classes through radiation protection experience.
The embodiment is an automatic radiation protection index prediction model related to personal dose in overhaul of a nuclear power plant, and recommends collective dose and total working hours for a single RWP. By identifying labels related to radiation protection indexes of nuclear power plant overhaul collective dosage and total man hour, historical target overhaul RWP, work order and personal dosage data are used as input, target overhaul RWP and work order data are analyzed through intelligent experience feedback and big data analysis technologies, predicted collective dosage and total man hour of a single RWP are output, manual correction data introduction is supported, and finally radiation protection index prediction related to nuclear power plant overhaul personal dosage is achieved. By using the method, under the condition that the overhaul work order of the nuclear power plant is basically frozen, the single RWP collective dose and the total working hours of the overhaul work order can be predicted, and the RWP collective dose and the total working hours can be accumulated and corrected to obtain the predicted values of the collective dose, the total working hours, the neutron collective dose and the internal irradiation collective dose of the whole overhaul work.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. It is intended that the present invention cover the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalents.

Claims (19)

1. A collective radiation dose prediction method based on radiation work license management is characterized by sequentially comprising the following steps:
s1, acquiring target overhaul RWP information and historical overhaul RWP information, and summarizing and calculating the total collective dose and total working hours of each RWP in historical overhaul;
s2, matching the information in the target overhaul RWP and the historical overhaul RWP acquired in the S1 item by item; when the historical overhaul RWP completely consistent with the target overhaul RWP exists, taking the total aggregate dose and the total working hour of the historical overhaul RWP as the total aggregate dose and the total working hour prediction result of the project standard overhaul RWP; when the historical overhaul RWP completely consistent with the target overhaul RWP does not exist, the next step is carried out to carry out similarity matching;
s3, matching the similarity between the target overhaul RWP and the historical overhaul RWP,
s3.1, setting fields to be matched in the target overhaul RWP and weight assignment w of each field;
s3.2 calculating the matching degree v of each field in the historical overhaul RWP item by item and the target overhaul RWP, calculating the matching score u of each field to wv, and calculating the total matching score u of the single historical overhaul RWPGeneral (1)Is the sum of field matching scores;
s3.3 Total matching score u of each item of historical overhaul RWP calculated from S3.2General assemblyThe highest u is selectedmaxAnd extracting the total aggregate dose and the total working hours of the corresponding historical overhaul RWP as a prediction result of the total aggregate dose and the total working hours of the target overhaul RWP.
2. A collective radiation dose prediction method based on radiation work license management as set forth in claim 1, characterized in that: before similarity matching between the target overhaul RWP and the historical overhaul RWP, the S3 identifies whether a preventive maintenance number exists in a work order related to the target overhaul RWP or not; if not, the step S3 is carried out to match the similarity of the target overhaul RWP and the historical overhaul RWP; if yes, preventive maintenance number matching is carried out, work orders with the same preventive maintenance number are searched in work orders related to the historical overhaul RWP, and the searched total aggregate dose and total working hours of the historical overhaul RWP are used as the total aggregate dose and total working hour prediction results of the target overhaul RWP; if not, the step S3 is carried out to match the similarity of the target overhaul RWP and the historical overhaul RWP.
3. A collective radiation dose prediction method based on radiation work license management as set forth in claim 2, characterized in that: the fields set in S3 for matching in the target overhaul RWP include RWP description, device code, job category, and room.
4. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 1, 2 or 3, characterized in that: and when the historical overhaul RWP information in the S1 is obtained, selecting multiple matched historical overhaul records from the nuclear power plant information management system according to the information of the target overhaul object, the unit type and the overhaul type, and obtaining the historical overhaul RWP information from the multiple matched historical overhaul records.
5. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 1, 2 or 3, characterized in that: the S1 further includes a step of correcting the acquired historical overhaul RWP information, that is, a signed correction value is entered for the collective dose and the total man-hour corresponding to the work category in the historical overhaul RWP information according to factors in the work category, that is, the special work or the abnormal event in the historical overhaul, so as to correct the total collective dose and the total man-hour of the historical overhaul RWP referred to in S2.
6. A collective radiation dose prediction method based on radiation work license management as set forth in claim 5, characterized in that: the historical overhaul RWP information in S1 further includes information of neutron dose and internal irradiation dose.
7. A collective radiation dose prediction method based on radiation work license management as set forth in claim 6, characterized in that: the total matching score u of a single historical overhaul RWP in S3.2General (1)A total score u of RWP description match for the historical overhaulDescription of the inventionAnd the total score u of the matching degree of the equipment numbersDeviceJob type matching degree score uType of operationScore u of degree of matching with roomType of operationAnd (4) summing.
8. A collective radiation dose prediction method based on radiation work permit management as set forth in claim 7, characterized in that: the historical overhaul RWP description matching degree total score uDescription of the inventionThe calculation steps are as follows:
s3.2.1 learning and training the machine learning model based on the nuclear power professional lexicon, and then matching the target overhaul RWP with the historical overhaul RWP, thereby obtaining the initial score u of the matching degree described by the non-universal RWPDescription of the invention’；
S3.2.2 selecting one or more keywords from RWP of target overhaul RWP, comparing one by one whether these keywords exist in RWP description of historical overhaul RWP, if yes, setting similarity v of the keywordKeywordIf not, the similarity v of the keyword is set as 1KeywordThen, a matching degree score w of the keyword is calculated as 0Keyword＝wKeyword 1×vKeyword 1+wKeyword 2×vKeyword 3+……+wKeyword n×vKeyword n(n is a natural number);
s3.2.3 keyword matching score w calculated in step 3.2.2KeywordAnd (3) preliminarily scoring the matching degree obtained by the calculation in the step (3.2.1)Description of the invention' make a correction to obtain a corrected match score u for the RWP description of the historical overhauled RWPDescription of the invention", the calculation formula is:
udescription of the invention”＝wDescription of the invention×vDescription of the invention+wKeyword；
S3.2.4 comparison of u calculated in step 3.2.3Description of the invention"and when vDescription of the inventionCalculated when 1 is satisfiedDescription of the invention' the smaller of the two is taken as the total matching degree score u of the RWP description of the historical overhaul RWPDescription of the invention。
9. A collective radiation dose prediction method based on radiation work permit management as set forth in claim 8, characterized in that: and 3.2.1, matching the similarity of the target overhaul RWP and the historical overhaul RWP by adopting a semantic similarity technology.
10. A collective radiation dose prediction method based on radiation work permit management as set forth in claim 9, characterized in that: the semantic similarity technique comprises the following steps: firstly, performing word segmentation processing on word descriptions in RWP descriptions of a target overhaul RWP and a historical overhaul RWP by a Chinese word segmentation engine after nuclear power professional lexicon optimization; then, the word frequency vectors of the two vectors are respectively calculated based on the TF-IDF algorithm, and the similarity calculation v is carried out on the two vectorsDescription of the inventionTo obtain a preliminary matching score u of the RWP descriptionDescription of the invention', the calculation formula is: u'Description of the invention＝wDescription of the invention×vDescription of the invention。
11. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 10, characterized in that: s3.2.1, the method also includes the step of removing redundant information in RWP descriptions of the target overhaul and the historical overhaul RWP, namely removing literal information coincident with other fields in the RWP descriptions before matching the target overhaul RWP and the historical overhaul RWP.
12. A collective radiation dose prediction method based on radiation work permit management as set forth in claim 8, characterized in that: the total score u of the matching degree of the equipment numbersDeviceThe calculation steps are as follows:
s3.2.5 extracting the equipment number from the work order of the target overhaul RWP and the historical overhaul RWP;
s3.2.6 matching the S3.2.5 extracted target overhaul RWP with the equipment number in the historical overhaul RWP, setting the number of the extracted equipment numbers of the target overhaul RWP as n, and the equipment number matching degree score u of the historical RWPDevice iThe calculation formula is as follows:
when all the device numbers except the first unit number in the historical RWP are the same, the device number matching degree score of the historical RWP
Figure FDA0002810934710000041
When the first digit group number and the middle three digit code of a single equipment number in the historical RWP are different and the rest are the same, the equipment number matching degree score of the historical RWP
Figure FDA0002810934710000042
When the device number in the historical RWP, except the first unit number, the three-bit system code and the two-bit device type code, is different, the device number matching degree score of the historical RWP
Figure FDA0002810934710000043
If the value of the "RWP description" match score equals w in step 3.2.4 when the device number is "XXX 000XXXDescription of the inventionThen the device number match score u for the historical RWPDevice iIs set to wDevice；
If the conditions are not met, the matching is not successful, and the equipment number matching degree score u of the historical RWPDevice iIs 0;
s3.2.7 calculating the total score u of the matching degree of the device number of the historical RWPDeviceThe calculation formula is as follows:
Figure FDA0002810934710000044
13. a collective radiation dose prediction method based on radiation work permit management as claimed in claim 12, characterized in that: the S3.2.5 rule for extracting the equipment number from the work order associated with the target overhaul RWP and the historical overhaul RWP is as follows: if the 'equipment number' is not null and is a single equipment number, taking the value as the equipment number; if the equipment number is not null and is a work item code, identifying and extracting all equipment numbers related to the work item; if the equipment number is empty, automatically identifying whether the equipment number meeting the power plant number rule exists in the worker RWP description; otherwise, the device code match score of the historical RWP is 0.
14. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 8, 12 or 13, characterized in that: a matching degree score u of the job typeType of operationThe calculation steps are as follows: first, when the "job type" of the target overhaul RWP has n different job types separated by commas and matches m job types in the historical overhaul RWP, the "job type" matching degree score u of the historical overhaul RWPType of operationThe calculation formula is as follows:
Figure FDA0002810934710000051
if the "RWP description" match score in step 3.2.4 equals w when the "job type" of the target overhaul RWP and the historical overhaul RWP do not match successfullyDescription of the inventionAnd the matching degree score value u of the job type of the historical overhaul RWPType of operationIs set as wType of operationOtherwise, the matching degree score u of the job type of the history overhaul RWPType of operationIs set to 0.
15. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 14, characterized in that: the calculation rule of the room matching degree score is as follows:
when the value of the "RWP description" match score in step 3.2.4 equals wDescription of the inventionThe value u of the matching degree score of the "room" of the historical RWPRoomIs set as wRoom；
The value u of the match score for the "room" of the historical overhaul RWP when the "room" of the target overhaul RWP is emptyRoomSet to 0;
when the two conditions are not met, the 'room' fields of the target overhaul RWP and the historical overhaul RWP need to be matched, and when the fields are completely matched, the numerical value u of the matching degree score of the 'room' of the historical overhaul RWPRoomIs wRoomOtherwise, the score is 0.
16. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 15, characterized in that: the computer system in S3.3 synthesizes the total matching score u of each historical overhaul RWPGeneral assemblyAnd intelligently recommending historical data, and taking the total aggregate dose and the total working hours in the historical overhaul RWP with the highest matching degree score as the predicted total aggregate dose and the predicted total working hours of the target overhaul RWP.
17. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 16, characterized in that: in the RWP matching data finally recommended by the computer system, if n (n is greater than 1) historical overhaul RWPs with the highest similarity score exist, and the total aggregate dose and the total man-hour are respectively E1, E2, … Ei …, En and H1, H2, … Hi … and Hn, the predicted total aggregate dose and the predicted total man-hour of the target RWP are the arithmetic mean of the total aggregate dose and the total man-hour of the n historical RWPs, and the calculation formula is as follows:
Figure FDA0002810934710000061
Figure FDA0002810934710000062
18. a collective radiation dose prediction method based on radiation work permit management as claimed in claim 17, characterized in that: and 3.3, the total collective dose and total working hour prediction result of the target overhaul RWP obtained in the step S is the total collective dose and total working hour prediction result of a single RWP in the target overhaul, and the total collective dose and total working hour prediction results of the target overhaul RWP are added, so that the total overhaul collective dose and total working hour prediction result are obtained.
19. A collective radiation dose prediction method based on radiation work permit management as claimed in claim 18, characterized in that: the target overall overhaul total overhaul collective dose PEGeneral assemblyTotal man-hour PHGeneral assemblyThe calculation formula is as follows:
Figure FDA0002810934710000063
Figure FDA0002810934710000064
wherein PEXn is the individual RWP collective dose, PE predicted by similarity matchNeutron of neutronThe neutron dose predicted value, PE, is overhauled for the targetInner partPredicted value of irradiation dose, EC, for the target major repairGeneral assemblyTotal working class collective dose correction, ECGeneral assemblyThe calculation formula of (2) is as follows:
Figure FDA0002810934710000071
Figure FDA0002810934710000072
ECnthe sum of the correction values for the working class collective doses entered for S2.1 of each RWP; HCnIs the sum of the correction values of the total working hours of each RWP.
CN202011390103.5A 2020-12-02 2020-12-02 Collective dose prediction model based on radiation work license management Active CN114580694B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011390103.5A CN114580694B (en) 2020-12-02 2020-12-02 Collective dose prediction model based on radiation work license management

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011390103.5A CN114580694B (en) 2020-12-02 2020-12-02 Collective dose prediction model based on radiation work license management

Publications (2)

Publication Number Publication Date
CN114580694A true CN114580694A (en) 2022-06-03
CN114580694B CN114580694B (en) 2024-07-19

Family

ID=81766956

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011390103.5A Active CN114580694B (en) 2020-12-02 2020-12-02 Collective dose prediction model based on radiation work license management

Country Status (1)

Country Link
CN (1) CN114580694B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013065213A (en) * 2011-09-16 2013-04-11 Toshiba Corp Simulation result correcting device and correcting method
CN106600105A (en) * 2016-11-07 2017-04-26 中广核（深圳）辐射监测技术有限公司 System and method for lean management of dosage data of nuclear power plant
CN108535759A (en) * 2018-03-16 2018-09-14 广东核电合营有限公司 The remote radiation protecting, monitoring method, apparatus and system at million kilowatt nuclear power station
CN109472395A (en) * 2018-10-15 2019-03-15 中广核（深圳）辐射监测技术有限公司 Dose prediction method and terminal are radiated between a kind of nuclear power station overhaul time
CN111833988A (en) * 2020-07-14 2020-10-27 北京安德医智科技有限公司 Radiation parameter determination method and device, electronic equipment and storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013065213A (en) * 2011-09-16 2013-04-11 Toshiba Corp Simulation result correcting device and correcting method
CN106600105A (en) * 2016-11-07 2017-04-26 中广核（深圳）辐射监测技术有限公司 System and method for lean management of dosage data of nuclear power plant
CN108535759A (en) * 2018-03-16 2018-09-14 广东核电合营有限公司 The remote radiation protecting, monitoring method, apparatus and system at million kilowatt nuclear power station
CN109472395A (en) * 2018-10-15 2019-03-15 中广核（深圳）辐射监测技术有限公司 Dose prediction method and terminal are radiated between a kind of nuclear power station overhaul time
CN111833988A (en) * 2020-07-14 2020-10-27 北京安德医智科技有限公司 Radiation parameter determination method and device, electronic equipment and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
沈恩伟 等: "核电厂电子剂量计系统优化改造", 机械与电子, 31 December 2017 (2017-12-31), pages 38 *

Also Published As

Publication number Publication date
CN114580694B (en) 2024-07-19

Similar Documents

Publication Publication Date Title
CN111339126B (en) Medical data screening method and device, computer equipment and storage medium
Liu et al. Occupational health and safety risk assessment using an integrated TODIM‐PROMETHEE model under linguistic spherical fuzzy environment
CN111695836B (en) Clinical trial online operation management and control integrated system
CN106372225A (en) Information processing device and method based on high-value comparison base
CN113657548A (en) Medical insurance abnormity detection method and device, computer equipment and storage medium
CN117236688A (en) Building safety accident risk prevention and early warning system based on big data technology
CN111984788A (en) Electric power system violation management method and device and electric power equipment
CN114492926A (en) Method and system for analyzing and predicting potential safety hazard text of coal mine
CN112950344A (en) Data evaluation method and device, electronic equipment and storage medium
CN115170052A (en) Engineering management method and system based on artificial intelligence
Macedo et al. Identifying low-quality patterns in accident reports from textual data
US20190148007A1 (en) System and Method Associated with Determining Physician Attribution Related to In-Patient Care Using Prediction-Based Analysis
CN114462736B (en) Intelligent recommendation method for experience feedback of nuclear power plant radiation work license application
CN114580694B (en) Collective dose prediction model based on radiation work license management
Ma et al. Mining construction accident reports via unsupervised NLP and Accimap for systemic risk analysis
CN114580693B (en) Radiation dose prediction method based on historical data
CN109460868A (en) A kind of construction method of the enterprise safety operation early warning system using gray scale prediction
KR20140080594A (en) Method for evaluating patents using engine and evaluation server
Jabour et al. Monitoring public health reporting: data tracking in cancer registries
CN110782997A (en) Drug risk assessment model based on adverse reaction monitoring report
CN118246775B (en) Coal mine index management configuration method and system
Stanimirović et al. Digital transformation as one of the instruments for overcoming the public health crisis: the role and use of ehealth solutions during the Covid-19 pandemic in Slovenia
CN115080687B (en) Nuclear power plant work order task and quality defect report information accurate matching method
RU2700397C1 (en) Computer-implemented method for automated processing and analysis of data to assess the efficiency of execution of instructions
CN118365459B (en) Intelligent matching system, method, equipment and medium for business insurance claim rules

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant