CN115098625A - Accurate matching system and method for NCR and completion report data of nuclear power plant - Google Patents
Accurate matching system and method for NCR and completion report data of nuclear power plant Download PDFInfo
- Publication number
- CN115098625A CN115098625A CN202210657113.3A CN202210657113A CN115098625A CN 115098625 A CN115098625 A CN 115098625A CN 202210657113 A CN202210657113 A CN 202210657113A CN 115098625 A CN115098625 A CN 115098625A
- Authority
- CN
- China
- Prior art keywords
- data
- completion report
- matching
- historical
- power plant
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000004519 manufacturing process Methods 0.000 claims abstract description 127
- 238000012163 sequencing technique Methods 0.000 claims abstract description 3
- 230000011218 segmentation Effects 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 8
- 238000010169 landfilling Methods 0.000 abstract 1
- 230000002159 abnormal effect Effects 0.000 description 7
- 239000000945 filler Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention particularly relates to an accurate matching method for nuclear power plant NCR and completion report data, which comprises the following steps: acquiring historical completion report data to form a nuclear power plant completion report database; matching the production non-conformity data with historical completion report data in a completion report database of the nuclear power plant through an NCR (nuclear power plant network record) and completion report data accurate matching model by taking the production non-conformity data as input data, and sequencing the matched historical completion report data according to the obtained historical completion report data matching value W; pushing the matching score W to be greater than the set W Limiting the score Historical completion report data. The method for accurately matching the NCR and completion report data of the nuclear power plant realizes intelligent and accurate pushing of the completion report data of the defect work order of the nuclear power plant, and reduces the searching time of a production non-conforming landfilling person.
Description
Technical Field
The invention relates to the technical field of management of production non-conformity items of a nuclear power plant, in particular to a system and a method for accurately matching NCR (nuclear reactor) and completion report data of the nuclear power plant.
Background
Quality defects of systems, equipment and structures found or generated in the production operation process of the nuclear power plant still cannot meet the original design requirements or relevant acceptance criteria after being processed in the form of a work ticket, and are managed as production non-conforming items (called NCR for short). At present, when filling out production non-conformity items, production non-conformity item fillers mainly make reference by manually searching relevant information such as defect reasons and work completion details of similar defect type worksheet completion reports (completion reports for short) to make better processing schemes and measures. However, the retrieval mode adopted at present is a traditional keyword search mode, and the most desirable reference data information cannot be obtained by using the traditional keyword search mode. In the process of filling out the production non-conforming items, if the work order completion report of the historical defect types similar to the production non-conforming items needing to be developed at present can be intelligently and accurately pushed, the searching time of a filler can be reduced, meanwhile, the quality defect problems of a system, equipment and a structure are more quickly processed, and the safety of a nuclear power plant is ensured.
Disclosure of Invention
Based on the above, it is necessary to provide a system and a method for accurately matching completion report data of a nuclear power plant, aiming at the problems of long time consumption and low accuracy rate when a production nonconformity filler fills out a production nonconformity item and manually searches for similar completion report data for reference.
In order to achieve the above purpose, the invention provides the following technical scheme:
an accurate matching method for nuclear power plant NCR and completion report data comprises the following steps:
step 1, obtaining historical completion report data to form a nuclear power plant completion report database;
step 2, taking the production non-conformity data as input data, matching the production non-conformity data with historical completion report data in a nuclear power plant completion report database through a nuclear power plant NCR and completion report data accurate matching model, and sequencing the matched historical completion report data according to the obtained historical completion report data matching value W;
step 3, pushing the matching score W to be larger than the set W Limiting the score Historical completion report data.
Further, the input field data of the production non-compliant data includes "power plant", "equipment code", "equipment name", and "non-compliant name" field data; the query field data of the completion report data comprises field data of 'power plant', 'work order type', 'equipment code', 'equipment name' and 'work order task title'; the pushed historical completion report data display field data comprise field data of 'power plant', 'unit state', 'recommendation degree', 'equipment code', 'equipment name', 'state', 'work order task number' and 'work order task title'.
Furthermore, the nuclear power plant NCR and completion report data accurate matching model matches input production non-conforming item data with historical completion report data based on semantic similarity matching and equipment coding field data matching.
Further, the semantic similarity matching comprises the following steps:
(1) performing word segmentation processing on input word description field data of production non-conforming item data and historical completion report data through a Chinese word segmentation engine optimized by a nuclear power professional word stock to respectively obtain respective word segmentation phrases;
(2) respectively calculating the word frequency vector and the sentence vector of the word segmentation phrase obtained in the step (1) based on a TF-IDF algorithm;
(3) cosine similarity calculation is carried out on the word frequency vectors and sentence vectors obtained in the step (2), and semantic similarity matching scores are obtained and output; the larger the semantic similarity matching score is, the higher the similarity is, and conversely, the lower the similarity is.
Further, the accurate matching model of the nuclear power plant NCR and the completion report data matches production non-conformity data with historical completion report data in a nuclear power plant completion report database through the following steps, and matches a score W according to the obtained historical completion report data:
step 1, based on the ' equipment code ' field data matching principle, carrying out ' equipment code ' field data matching on extracted equipment code ' field data for matching production non-conforming item data and ' equipment code ' field data for matching historical completion report data to obtain a matching score W of the ' equipment code ' field data of the historical completion report 1 ；
Step 2, based on semantic similarity matching principle, producing non-conforming data and historySemantic similarity matching is carried out on the finished report data to obtain a semantic similarity matching score W of the historical finished report 2 ；
Step 3, according to the historical completion report 'equipment code' field data matching score W 1 Semantic similarity match score W with historical completion report 2 Calculating historical completion report data match score W ＝ Historical completion report 'equipment code' field data match score W 1 + historical completion report semantic similarity match score W 2 ；
Step 4, sorting the matched historical completion report data according to the matching score W of the historical completion report data, and pushing that the matching score W is higher than the set value W Limiting the score Historical completion report data.
Further, when the matched production non-conformity data and the "power plant" field data of the historical completion report data are the same, or the "power plant" field data are different and are not "CNP-300", "CANDU 6", "fast reactor", "AP-1000", "VVER V-428M" or "CNP-1000", the accurate matching model of the nuclear power plant NCR and the completion report data extracts the "equipment code" field data for matching the production non-conformity data from the "equipment code" field data or the "non-conformity name" field data of the production non-conformity data; and extracting the 'equipment code' field data for matching the historical completion report data from the 'equipment code' field data or the 'work order task title' field data of the historical completion report data.
Further, the accurate matching model of the nuclear power plant NCR and the completion report data is used for matching the equipment code field data of the equipment code field data for the production of the non-conforming data matching and the historical completion report data based on the following equipment code field data matching principle:
when the production non-conforming item data is the same as the 'equipment code' field data for matching the historical completion report data, the matching score W of the 'equipment code' field data of the historical completion report 1 ＝α 1 ；
The production non-conformity item data and the historical completion report data are matched by the matching value W of the 'equipment code' field data of the historical completion report, except that the 'unit number' codes in the 'equipment code' field data are different, and other codes are all the same 1 ＝α 1 -0.05；
In the field data of 'equipment code' for matching production non-conforming item data with historical completion report data, except that the codes of 'unit number' and 'equipment position number' are different, the other codes are all the same, and the matching score W of the field data of 'equipment code' of the historical completion report 1 ＝α 1 -0.1；
When the situations are not met, the historical completion report 'equipment code' field data matching score W 1 ＝0。
Further, when the NCR and the accurate matching model of the completion report data of the nuclear power plant cannot extract the equipment code field data for matching the production non-conforming data or the equipment code field data for matching the historical completion report data, the equipment code field data matching score W of the historical completion report 1 ＝0。
Further, when the matched data of the field of the production non-conforming item is different from the data of the field of the 'power plant' of the completion report data and is 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000', only the semantic similarity of the name of the non-conforming item and the name of the equipment 'of the production non-conforming item are considered to be matched with the semantic similarity of the task title of the' worksheet 'and the name of the equipment' of the completion report, and at the moment, the matching score W of the data of the field of the 'equipment code' of the completion report of the historical defect type worksheet is considered to be matched with 1 ＝0。
Further, the model for accurately matching the nuclear power plant NCR with the completion report data performs semantic similarity matching on the production non-conforming data and the historical completion report data under the following conditions:
1. the matched production non-conformity data and the 'power plant' field data of the historical completion report data are different and are 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000';
2. the matched production non-compliance data and the historical completion report data have the same field data of "power plant" and are not "CNP-300", "CANDU 6", "fast reactor", "AP-1000", "VVER V-428M" and "CNP-1000".
Further, the accurate matching model of the nuclear power plant NCR and the completion report data carries out semantic similarity matching on production non-conforming data and historical completion report data based on the following semantic similarity matching principle:
step 1, when the field data of the power plant of the unmatched production unmatched item data and the historical completion report data are different and are CNP-300, CANDU6, fast reactor, AP-1000, VVER V-428M and CNP-1000, semantic similarity matching is carried out on the field data of the unmatched production unmatched item name and equipment name and the field data of the work order task title and equipment name of the historical completion report data, the scores of the similarity are normalized, and the semantic similarity matching score W of the historical completion report is obtained 2 ；
Step 2, when the field data of the matched production non-conforming item data and the historical completion report data are the same and are not CNP-300, CANDU6, fast reactor, AP-1000, VVER V-428M and CNP-1000 and the field data of the production non-conforming item data and the equipment code field data for matching the historical completion report data cannot be extracted, the field data of the production non-conforming item data and the equipment name field data of the production non-conforming item data and the historical completion report data are subjected to semantic similarity matching, the scores of the similarity are normalized to obtain the semantic similarity matching score W of the historical completion report 2 ；
Step 3, the matched production non-conformity item data is the same as the 'power plant' field data of the historical completion report data andnot 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000', when extracting the 'equipment code' field data for matching the production non-conforming item data and the 'equipment code' field data for matching the historical completion report data, semantic similarity matching is carried out on the 'non-conforming item name' and the 'equipment name' field data of the production non-conforming item data after removing the 'equipment code' field data and the 'work order task title' and the 'equipment name' field data of the historical completion report data after removing the 'equipment code' field data, the similarity is divided and normalized to obtain a semantic similarity matching score W of the historical completion report 2 In this case, the historical completion report semantic similarity match score W 2 Maximum value of alpha 2 ；α 2 +α 2 ＝1。
The invention also provides a nuclear power plant NCR and completion report data accurate matching system used for the method, which comprises an inquiry module, an output module and a nuclear power plant completion report database, wherein the inquiry module is internally provided with a nuclear power plant NCR and completion report data accurate matching model;
the query module is used for inputting production non-conforming item data by a user; the system is used for matching the production non-conformity data with historical completion report data in a nuclear power plant completion report database through a nuclear power plant NCR and completion report data accurate matching model according to the production non-conformity data to obtain a historical completion report data matching value W; the matching module is used for sorting the matched historical completion report data according to the historical completion report data matching score W and then sending the sorted historical completion report data to the output module;
the output module is provided with a historical completion report data limit score W Limiting the score The output module is used for receiving historical completion report data sent by the query module and pushing W higher than set W Limiting the score Historical completion report data.
The invention has the beneficial technical effects that:
the system and the method for accurately matching the NCR and the completion report data of the nuclear power plant improve the level and the depth of data mining based on semantic similarity matching and equipment coding field data matching, automatically and accurately match and push the completion report data of the nuclear power plant matched with production non-conforming data, reduce the search time of production non-conforming data reporters, improve the accuracy and the reliability of retrieval and pushing of relevant data of the production non-conforming data, effectively solve the problem of low utilization rate of data resources in a nuclear power safety production management platform, and have very important significance for effectively controlling and managing production non-conforming data reporters and constructing good nuclear safety culture.
Drawings
FIG. 1 is a flow chart of an accurate matching method for the data of the nuclear power plant NCR and the completion report of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the detailed description.
Referring to fig. 1, the invention provides an accurate matching method for NCR and completion report data of a nuclear power plant, comprising the following steps:
1. acquiring historical completion report data to form a nuclear power plant completion report database;
2. formulating a matching rule and a data assignment principle of the NCR and the completion report data of the nuclear power plant;
3. the method comprises the steps that production non-conforming item data are used as input data, and a nuclear power plant NCR and completion report data matching model is established through matching rules and data assignment principles of the nuclear power plant NCR and completion report data;
4. sorting the importance of the matched historical completion report data according to the obtained historical completion report data matching score W;
5. pushing the matching score W to be greater than the set W Limiting the score Historical completion report data.
Further, the input field data of the production non-conforming data includes field data of "power plant", "equipment code", "equipment name" and "non-conforming item name"; the query field data of the completion report data comprises field data of 'power plant', 'work order type', 'equipment code', 'equipment name' and 'work order task title'; the pushed historical completion report data display field data comprise field data of 'power plant', 'unit state', 'recommendation degree', 'equipment code', 'equipment name', 'state', 'work order task number' and 'work order task title'.
Furthermore, the nuclear power plant NCR and completion report data accurate matching model matches input production non-conforming item data with historical completion report data based on semantic similarity matching and equipment coding field data matching.
Further, the semantic similarity matching comprises the following steps:
(1) performing word segmentation processing on input word description field data of production non-conforming item data and historical completion report data through a Chinese word segmentation engine optimized by a nuclear power professional word stock to respectively obtain respective word segmentation phrases;
(2) respectively calculating the word frequency vector and the sentence vector of the word segmentation phrase obtained in the step (1) based on a TF-IDF algorithm;
(3) cosine similarity calculation is carried out on the word frequency vectors and sentence vectors obtained in the step (2), and semantic similarity matching scores are obtained and output; the larger the semantic similarity matching score is, the higher the similarity is, and conversely, the lower the similarity is.
Further, the accurate matching model of the nuclear power plant NCR and the completion report data matches production non-conformity data with historical completion report data in a nuclear power plant completion report database through the following steps, and matches a score W according to the obtained historical completion report data:
step 1, based on the 'equipment code' field data matching principle, carrying out 'equipment code' field data matching on the extracted 'equipment code' field data for matching the production non-conforming item data and the 'equipment code' field data for matching the historical completion report data to obtain a matching score W of the 'equipment code' field data of the historical completion report 1 ；
Step 2, matching based on semantic similarityMatching principle, semantic similarity matching is carried out on the data of the production non-conforming items and the historical completion report data to obtain a semantic similarity matching score W of the historical completion report 2 ；
Step 3, according to the historical completion report 'equipment code' field data matching score W 1 Semantic similarity match score W with historical completion report 2 Calculating historical completion report data match score W ＝ Historical completion report "device code" field data match score W 1 + historical completion report semantic similarity match score W 2 ；
Step 4, sorting the matched historical completion report data according to the matching score W of the historical completion report data, and pushing that the matching score W is higher than the set value W Limiting the score Historical completion report data.
Further, under the condition that the matched production non-conforming item data and the power plant field data of the historical completion report data are the same and the equipment code field data are normal, the accurate matching model of the nuclear power plant NCR and the completion report data extracts the equipment code field data for matching the production non-conforming item data from the equipment code field data or the non-conforming item name field data of the production non-conforming item data; extracting 'equipment code' field data for matching historical completion report data from 'equipment code' field data or 'work order task title' field data of the historical completion report data; and based on the following 'equipment code' field data matching principle, carrying out 'equipment code' field data matching on the production non-conforming data and the historical completion report data:
when the production non-conforming item data is the same as the 'equipment code' field data for matching the historical completion report data, the matching score W of the 'equipment code' field data of the historical completion report 1 ＝α 1 ；
The production non-conformity item data and the historical completion report data are matched by the 'equipment code' field data, except that the 'unit number' codes are different, other codes are all the same, and the 'equipment code' field data of the historical completion report is matchedScore W 1 ＝α 1 -0.05；
In the field data of 'equipment code' for matching production non-conforming item data with historical completion report data, except that the codes of 'unit number' and 'equipment position number' are different, the other codes are all the same, and the matching score W of the field data of 'equipment code' of the historical completion report 1 ＝α 1 -0.1；
When the situations are not met, the historical completion report 'equipment code' field data matching score W 1 ＝0。
Further, when the matched data of the field of the production non-conforming item is different from the data of the field of the 'power plant' of the completion report data and is 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000', only the semantic similarity of the name of the non-conforming item and the name of the equipment 'of the production non-conforming item is considered to be matched with the semantic similarity of the title of the' work order task 'and the name of the equipment' of the completion report, and at the moment, the matching score W of the data of the field of the 'equipment code' of the completion report of the historical defect type work order is considered 1 ＝0。
Further, when the matched production non-conformity data is the same as the 'plant' field data of the historical completion report data, and the 'equipment code' field data is abnormal or no data, the accurate matching model of the nuclear power plant NCR and the completion report data extracts the 'equipment code' field data for the matching of the production non-conformity data and the 'equipment code' field data for the matching of the historical completion report data in the following way:
when the 'equipment code' field data of the production non-conforming data is normal, the 'equipment code' field data of the completion report data is abnormal, and the 'equipment code' field data meeting the equipment code rule exists in the 'worksheet task title' field data of the completion report data, extracting the 'equipment code' field data of the production non-conforming data as the 'equipment code' field data for matching the production non-conforming data, and extracting the 'equipment code' field data meeting the equipment code rule in the 'non-conforming name' field data of the historical completion report data as the 'equipment code' field data for matching the historical completion report data;
when the 'equipment code' field data of the production non-conforming item data is abnormal, the 'equipment code' field data meeting the equipment code rule exists in the 'non-conforming item name' field data of the production non-conforming item data, and the 'equipment code' field data of the historical completion report data is normal, the 'equipment code' field data meeting the equipment code rule in the 'non-conforming item name' field data of the production non-conforming item data is extracted as the 'equipment code' data for matching the production non-conforming item data, and the 'equipment code' field data of the historical completion report data is extracted as the 'equipment code' field data for matching the historical completion report data;
when the 'equipment code' field data of the production non-conforming item data is abnormal, the 'equipment code' field data meeting the equipment code rule exists in the 'non-conforming item name' field data of the production non-conforming item data, the 'equipment code' field data of the historical completion report data is abnormal, and when the 'work order task title' field data of the historical completion report data contains 'equipment coding' field data meeting the equipment coding rule, extracting 'equipment code' field data meeting the equipment code rule in 'non-conforming item name' field data of the production non-conforming item data as 'equipment code' data for matching the production non-conforming item data, and extracting 'equipment code' field data meeting the equipment code rule in 'non-conforming item name' field data of the historical finished report data as 'equipment code' field data for matching the historical finished report data.
Further, when the NCR and the accurate matching model of the completion report data of the nuclear power plant cannot extract the equipment code field data for matching the production non-conforming data or the equipment code field data for matching the historical completion report data, the equipment code field data matching score W of the historical completion report 1 ＝0；
When the 'equipment code' field data which do not accord with the item data are not produced normally, and the 'equipment code' field data which do not meet the equipment code rule are not produced in the 'non-conformity item name' field data which do not accord with the item data; or the 'equipment code' field data of the historical completion report data are abnormal, and the 'equipment code' field data of the equipment code rule are not met in the 'worksheet task title' field data of the historical completion report data, the accurate matching model of the completion report data of the nuclear power plant cannot extract the 'equipment code' field data for matching the production non-conforming data or the 'equipment code' field data for matching the historical completion report data.
Further, the NCR and completion report data accurate matching model carries out semantic similarity matching on production non-conforming item data and historical completion report data under the following conditions:
1. the matched production non-conformity data and the 'power plant' field data of the historical completion report data are different and are 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000';
2. the matched production non-compliance data and the "plant" field data of the historical completion report data are the same and are not "CNP-300", "CANDU 6", "fast reactor", "AP-1000", "VVER V-428M" and "CNP-1000".
Further, the model for accurately matching the NCR with the finished report data performs semantic similarity matching on the production non-conforming item and the finished report data based on the following semantic similarity matching principle:
1. when the field data of the matched production non-conforming items and the historical completion report data are different and are ' CNP-300 ', ' CANDU6 ', ' fast reactor ', ' AP-1000 ', ' VVER V-428M ' and ' CNP-1000 ', semantic similarity matching is carried out on the field data of the production non-conforming items ' non-conforming item name ' + ' equipment name ' and the field data of the work order task title ' + ' equipment name ' of the completion report data, the similarity is divided and normalized to obtain the semantic similarity of the historical completion reportSimilarity match score W 2 In this case, the historical completion report semantic similarity match score W 2 The highest value is 1;
2. when the field data of the NCR data and the work order completion report data which are matched are the same and are not 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000', the field data of the 'equipment code' for matching the NCR data or the field data of the 'equipment code' for matching the historical completion report data can not be extracted,
performing semantic similarity matching on the field data of the 'non-conforming item name' + 'equipment name' of the production non-conforming item data and the field data of the 'work order task title' + 'equipment name' of the historical completion report data, and normalizing the score of the similarity to obtain the semantic similarity matching score W of the historical completion report 2 In this case, the historical completion report semantic similarity match score W 2 The highest value is 1;
3. the matched field data of the production non-conforming item data and the field data of the completion report data are the same and are not 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000', when the field data of the equipment code for matching the production non-conforming item data and the field data of the equipment code for matching the completion report data are extracted,
carrying out semantic similarity matching on the 'non-conforming item name' + 'equipment name' field data of the production non-conforming item data after the 'equipment coding' field data is removed and the 'work order task title' + 'equipment name' field data of the completion report data after the 'equipment coding' field data is removed, and normalizing the scores of the similarity to obtain the semantic similarity matching score W of the historical completion report 2 In this case, the historical completion report semantic similarity match score W 2 Maximum value of alpha 2 ；α 2 +α 2 ＝1。
Further, when the "device code" field data which do not conform to the item data are not normally produced, and the "device code" field data which do not satisfy the device code rule in the "non-conforming item name" field data which do not conform to the item data are produced; or the 'equipment code' field data of the historical completion report data are abnormal, and when the 'equipment code' field data meeting the equipment code rule are not in the 'work order task title' field data of the historical completion report data, the 'equipment code' field data for matching the production non-conformity data or the 'equipment code' field data for matching the historical completion report data cannot be extracted by the NCR and the accurate matching model of the completion report data.
Further, when the matching of 'equipment' and 'semanteme' of the production non-conforming item data and the historical completion report data is completed, the similarity matching score W is calculated 1 +W 2 . And pushing historical completion report data with the matching score W larger than the set W limit score.
The invention also provides a nuclear power plant NCR and completion report data accurate matching system used for the method, which comprises an inquiry module, an output module and a nuclear power plant completion report database, wherein the inquiry module is embedded with a nuclear power plant completion report data accurate matching model;
the query module is used for inputting production non-conforming item data by a user; the system is used for matching production non-conforming data with historical completion report data in a completion report database of the nuclear power plant through a nuclear power plant NCR and completion report data accurate matching model according to the production non-conforming data to obtain a historical completion report data matching value W; the matching module is used for sorting the matched historical completion report data according to the historical completion report data matching score W and then sending the sorted historical completion report data to the output module;
the output module is provided with a historical completion report data limit score W Limiting the score The output module is used for receiving historical completion report data sent by the query module and pushing W higher than set W Limiting the score Historical completion report data.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, many variations and modifications can be made without departing from the spirit of the invention, which falls within the scope of the invention. Therefore, the protection scope of the present patent should be subject to the appended claims.
Claims (11)
1. A method for accurately matching NCR (nuclear power plant reference signal) with completion report data is characterized by comprising the following steps:
step 1, obtaining historical completion report data to form a completion report database of a nuclear power plant;
step 2, taking the production non-conformity data as input data, matching the production non-conformity data with historical completion report data in a nuclear power plant completion report database through a nuclear power plant NCR and completion report data accurate matching model, and sequencing the matched historical completion report data according to the obtained historical completion report data matching value W;
step 3, pushing the matching score W to be larger than the set W Limiting the score Historical completion report data.
2. The method for accurately matching nuclear power plant NCR with completion report data according to claim 1, wherein the input field data of production non-compliant data includes "power plant", "equipment code", "equipment name" and "non-compliant name" field data; the query field data of the completion report data comprises field data of 'power plant', 'work order type', 'equipment code', 'equipment name' and 'work order task title'; the pushed historical completion report data display field data comprise field data of 'power plant', 'unit state', 'recommendation degree', 'equipment code', 'equipment name', 'state', 'work order task number' and 'work order task title'.
3. The method for accurately matching nuclear power plant NCR with completion report data as recited in claim 2, wherein the accurate matching model of nuclear power plant NCR with completion report data matches input production non-compliance data with historical completion report data based on semantic similarity matching and "device code" field data matching.
4. The method for accurately matching nuclear power plant NCR with completion report data according to claim 3, wherein the semantic similarity matching comprises the following steps:
step 1, performing word segmentation processing on input word description field data of production non-conforming item data and historical completion report data through a Chinese word segmentation engine optimized by a nuclear power professional lexicon to respectively obtain respective word segmentation phrases;
step 2, respectively calculating word frequency vectors and sentence vectors of the word segmentation phrases obtained in the step (1) based on a TF-IDF algorithm;
step 3, performing cosine similarity calculation on the word frequency vector and the sentence vector obtained in the step (2) to obtain and output a semantic similarity matching score; the larger the semantic similarity matching score is, the higher the similarity is, and conversely, the lower the similarity is.
5. The method for accurately matching nuclear power plant NCR with completion report data according to claim 4, characterized in that the accurate matching model of nuclear power plant NCR with completion report data matches production non-compliance data with historical completion report data in a nuclear power plant completion report database by the following steps, and matches score value W according to the obtained historical completion report data:
step 1, based on the 'equipment code' field data matching principle, carrying out 'equipment code' field data matching on the extracted 'equipment code' field data for matching the production non-conforming item data and the 'equipment code' field data for matching the historical completion report data to obtain a matching score W of the 'equipment code' field data of the historical completion report 1 ；
Step 2, based on the semantic similarity matching principle, performing semantic similarity matching on the production non-conforming item data and the historical completion report data to obtain the semantic similarity of the historical completion reportDegree match score W 2 ；
Step 3, according to the historical completion report 'equipment code' field data matching score W 1 Semantic similarity match score W with historical completed report 2 Calculating historical completion report data match score W ＝ Historical completion report 'equipment code' field data match score W 1 + historical completion report semantic similarity match score W 2 ；
Step 4, sorting the matched historical completion report data according to the matching score W of the historical completion report data, and pushing that the matching score W is higher than the set value W Limiting the score Historical completion report data.
6. The accurate matching method of the nuclear power plant NCR and the completion report data as claimed in claim 5, characterized in that when the matched production non-conformity data is the same as the "plant" field data of the historical completion report data, or the "plant" field data is different and is not "CNP-300", "CANDU 6", "fast reactor", "AP-1000", "VVER V-428M" or "CNP-1000", the accurate matching model of the nuclear power plant NCR and the completion report data extracts the "equipment code" field data for matching the production non-conformity data from the "equipment code" field data or the "non-conformity name" field data of the production non-conformity data; and extracting the 'equipment code' field data for matching the historical completion report data from the 'equipment code' field data or the 'work order task title' field data of the historical completion report data.
7. The method for accurately matching nuclear power plant NCR with completion report data according to claim 6, characterized in that the accurate matching model of nuclear power plant NCR and completion report data is used for matching the device code field data for the production of the device code field data for the matching of non-conforming data and the device code field data for the matching of historical completion report data based on the following device code field data matching principle:
can not be producedMatching score W of 'equipment code' field data of historical completion report when coincidence data is identical to 'equipment code' field data for matching historical completion report data 1 ＝α 1 ；
The production non-conformity item data and the historical completion report data are matched by the matching value W of the 'equipment code' field data of the historical completion report, except that the 'unit number' codes in the 'equipment code' field data are different, and other codes are all the same 1 ＝α 1 -0.05；
In the field data of 'equipment code' for matching production non-conforming item data with historical completion report data, except that the codes of 'unit number' and 'equipment position number' are different, the other codes are all the same, and the matching score W of the field data of 'equipment code' of the historical completion report 1 ＝α 1 -0.1；
When the situations are not met, historical completion report 'equipment code' field data matching score W 1 ＝0。
8. The method for accurately matching nuclear power plant NCR with completion report data as claimed in claim 7, wherein when the model for accurately matching nuclear power plant NCR with completion report data cannot extract the field data of 'equipment code' for matching production non-conforming data or the field data of 'equipment code' for matching historical completion report data, the matching score W of the field data of 'equipment code' of historical completion report 1 ＝0。
9. The method of claim 7, wherein the model for accurately matching nuclear power plant NCR with completion report data is used for matching semantic similarity between production non-compliance data and historical completion report data when:
(1) the field data of the power plant for the matched production non-conformity data and the historical completion report data are different and are 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000';
(2) the field data of the production non-conformity data and the historical completion report data which are matched are the same and are not 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000'.
10. The method for accurately matching nuclear power plant NCR with completion report data according to claim 9, wherein the accurate matching model of nuclear power plant NCR with completion report data performs semantic similarity matching on production non-conforming data and historical completion report data based on the following semantic similarity matching principle:
step 1, when the field data of the power plant of the unmatched production unmatched item data and the historical completion report data are different and are CNP-300, CANDU6, fast reactor, AP-1000, VVER V-428M and CNP-1000, semantic similarity matching is carried out on the field data of the unmatched production unmatched item name and equipment name and the field data of the work order task title and equipment name of the historical completion report data, the scores of the similarity are normalized, and the semantic similarity matching score W of the historical completion report is obtained 2 ；
Step 2, when the field data of the matched production non-conforming item data and the historical completion report data are the same and are not CNP-300, CANDU6, fast reactor, AP-1000, VVER V-428M and CNP-1000 and the field data of the production non-conforming item data and the equipment code field data for matching the historical completion report data cannot be extracted, the field data of the production non-conforming item data and the equipment name field data of the production non-conforming item data and the historical completion report data are subjected to semantic similarity matching, the scores of the similarity are normalized to obtain the semantic similarity matching score W of the historical completion report 2 ；
Step 3, matching production non-conformity item data and historical completion report dataThe field data of the 'power plant' are identical and are not 'CNP-300', 'CANDU 6', 'fast reactor', 'AP-1000', 'VVER V-428M' and 'CNP-1000', when the field data of the 'equipment code' for matching the production non-conforming item data and the field data of the 'equipment code' for matching the historical completion report data are extracted, the field data of the 'non-conforming item name' and the 'equipment name' of the production non-conforming item data after the field data of the 'equipment code' are removed and the field data of the 'work order task title' and the 'equipment name' of the historical completion report data after the field data of the 'equipment code' are removed are subjected to semantic similarity matching, and the score of the similarity is normalized to obtain the semantic similarity matching score W of the historical completion report 2 In this case, the historical completion report semantic similarity match score W 2 Maximum value of alpha 2 ；α 2 +α 2 ＝1。
11. The system for accurately matching nuclear power plant NCR with completion report data according to any one of claims 1-10, comprising a query module, an output module and a nuclear power plant completion report database, wherein the query module is embedded with a nuclear power plant NCR and completion report data accurate matching model;
the query module is used for inputting production non-conforming item data by a user; the system is used for matching the production non-conformity data with historical completion report data in a nuclear power plant completion report database through a nuclear power plant NCR and completion report data accurate matching model according to the production non-conformity data to obtain a historical completion report data matching value W; the matching module is used for sorting the matched historical completion report data according to the historical completion report data matching score W and then sending the sorted historical completion report data to the output module;
the output module is provided with a historical completion report data limit score W Limiting the score The output module is used for receiving historical completion report data sent by the query module and pushing W higher than set W Limiting the score Historical completion report data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210657113.3A CN115098625A (en) | 2022-06-10 | 2022-06-10 | Accurate matching system and method for NCR and completion report data of nuclear power plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210657113.3A CN115098625A (en) | 2022-06-10 | 2022-06-10 | Accurate matching system and method for NCR and completion report data of nuclear power plant |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115098625A true CN115098625A (en) | 2022-09-23 |
Family
ID=83291930
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210657113.3A Pending CN115098625A (en) | 2022-06-10 | 2022-06-10 | Accurate matching system and method for NCR and completion report data of nuclear power plant |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115098625A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107809322A (en) * | 2016-09-06 | 2018-03-16 | 中兴通讯股份有限公司 | The distribution method and device of work order |
| US20210287177A1 (en) * | 2020-03-10 | 2021-09-16 | Moseley Ltd. | Automatic monitoring and reporting system |
| CN114298460A (en) * | 2021-11-15 | 2022-04-08 | 深圳市东信时代信息技术有限公司 | Material work order assignment processing method, device, equipment and storage medium |
| CN114462399A (en) * | 2020-11-09 | 2022-05-10 | 中核核电运行管理有限公司 | Accurate matching method for quality defect report and state report of nuclear power plant |
| CN114462737A (en) * | 2020-11-09 | 2022-05-10 | 中核核电运行管理有限公司 | Accurate matching method applied to nuclear power plant work order task and operation event report |
| CN114600136A (en) * | 2019-09-25 | 2022-06-07 | 奥恩全球运营欧洲股份公司新加坡分公司 | System and method for automated operation of due diligence analysis to objectively quantify risk factors |
-
2022
- 2022-06-10 CN CN202210657113.3A patent/CN115098625A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107809322A (en) * | 2016-09-06 | 2018-03-16 | 中兴通讯股份有限公司 | The distribution method and device of work order |
| CN114600136A (en) * | 2019-09-25 | 2022-06-07 | 奥恩全球运营欧洲股份公司新加坡分公司 | System and method for automated operation of due diligence analysis to objectively quantify risk factors |
| US20210287177A1 (en) * | 2020-03-10 | 2021-09-16 | Moseley Ltd. | Automatic monitoring and reporting system |
| CN114462399A (en) * | 2020-11-09 | 2022-05-10 | 中核核电运行管理有限公司 | Accurate matching method for quality defect report and state report of nuclear power plant |
| CN114462737A (en) * | 2020-11-09 | 2022-05-10 | 中核核电运行管理有限公司 | Accurate matching method applied to nuclear power plant work order task and operation event report |
| CN114298460A (en) * | 2021-11-15 | 2022-04-08 | 深圳市东信时代信息技术有限公司 | Material work order assignment processing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108304372B (en) | Entity extraction method and device, computer equipment and storage medium | |
| CN111858842A (en) | Judicial case screening method based on LDA topic model | |
| CN113157903A (en) | Multi-field-oriented electric power word stock construction method | |
| CN114266256A (en) | Method and system for extracting new words in field | |
| CN112541077A (en) | Processing method and system for power grid user service evaluation | |
| CN116049359A (en) | Duplicate checking algorithm based on document content analysis | |
| CN110704638A (en) | Clustering algorithm-based electric power text dictionary construction method | |
| CN107239455B (en) | Core word recognition method and device | |
| CN115098673A (en) | Business document information extraction method based on variant attention and hierarchical structure | |
| CN114418327A (en) | Automatic order recording and intelligent order dispatching method for customer service system | |
| CN116628173B (en) | Intelligent customer service information generation system and method based on keyword extraction | |
| CN114462736A (en) | Experience feedback intelligent recommendation method for nuclear power plant radiation work license application | |
| CN112036150A (en) | Electricity price policy term analysis method, storage medium and computer | |
| CN108460119B (en) | System for improving technical support efficiency by using machine learning | |
| CN115098625A (en) | Accurate matching system and method for NCR and completion report data of nuclear power plant | |
| CN115982316A (en) | Multi-mode-based text retrieval method, system and medium | |
| CN114610882A (en) | Abnormal equipment code detection method and system based on electric power short text classification | |
| CN113342949A (en) | Matching method and system of intellectual library experts and topic to be researched | |
| CN112488593A (en) | Auxiliary bid evaluation system and method for bidding | |
| CN112860815A (en) | Finance and tax informatization data processing system based on big data | |
| CN112270185A (en) | Text representation method based on topic model | |
| CN111814457A (en) | Power grid engineering contract text generation method | |
| CN117235137B (en) | Professional information query method and device based on vector database | |
| CN116842021B (en) | Data dictionary standardization method, equipment and medium based on AI generation technology | |
| CN113139106B (en) | Event auditing method and device for security check |
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 |