CN112819053B - Model library establishing method and device, diagnosis method and device and prediction method - Google Patents
Model library establishing method and device, diagnosis method and device and prediction method Download PDFInfo
- Publication number
- CN112819053B CN112819053B CN202110095211.8A CN202110095211A CN112819053B CN 112819053 B CN112819053 B CN 112819053B CN 202110095211 A CN202110095211 A CN 202110095211A CN 112819053 B CN112819053 B CN 112819053B
- Authority
- CN
- China
- Prior art keywords
- model
- physical parameters
- diagnosis
- regression
- classification
- 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.)
- Active
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000013145 classification model Methods 0.000 claims abstract description 121
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 33
- 238000011156 evaluation Methods 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 13
- 238000013136 deep learning model Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 229910052739 hydrogen Inorganic materials 0.000 claims description 8
- 239000001257 hydrogen Substances 0.000 claims description 8
- 230000010354 integration Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004088 simulation Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 125000004435 hydrogen atom Chemical class [H]* 0.000 claims 2
- 150000002431 hydrogen Chemical class 0.000 description 5
- 239000002826 coolant Substances 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000977 initiatory effect Effects 0.000 description 3
- 241000321453 Paranthias colonus Species 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- 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)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Human Resources & Organizations (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- Evolutionary Biology (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a method for establishing a diagnosis model library, which comprises the following steps: determining a diagnosis target of the serious accident; acquiring time sequences of M physical parameters under N originating events to form N sequence sets; sequentially selecting T time sequences with the same physical parameters from the N sequence sets respectively to form a data set, wherein T is more than or equal to 2 and less than M; and obtaining a classification model and a regression model corresponding to the T physical parameters, and storing the T physical parameters, the classification model corresponding to the T physical parameters, the regression model and the corresponding relations among the T physical parameters, the regression model and the three into a model library to form a diagnosis model library. Further, a device for establishing the diagnosis model library, a method and a device for diagnosing serious accidents of the nuclear power plant and a method for predicting the progress of the serious accidents of the nuclear power plant are also provided. The diagnosis model obtained based on the establishment method of the diagnosis model library can accurately diagnose the initial event of the serious accident and the parameters of the corresponding initial event according to a limited number of meters when the serious accident happens.
Description
Technical Field
The invention belongs to the technical field of nuclear power serious accident analysis, and particularly relates to a method and a device for establishing a diagnosis model library, a method and a device for diagnosing serious accidents of a nuclear power plant and a method for predicting the progress of the serious accidents of the nuclear power plant.
Background
The severe accident management guidelines (Severe Accident Management Guidelines, SAMG) are guidelines for the nuclear power plant to take steps to alleviate the consequences of core damage accidents by the personnel in the master control room and technical support center when severe accidents occur.
One of the key factors for successful implementation of SAMG is the need to obtain the physical parameters of the power plant status by the meter timely, comprehensively and accurately. However, in the serious accident, the instrument for monitoring the physical parameters may be damaged or the data may be inaccurate, so that the accuracy of the existing serious accident diagnosis result is not high.
Disclosure of Invention
The invention aims to solve the technical problems of the prior art, and provides a method and a device for establishing a diagnosis model library, a method and a device for diagnosing serious accidents of a nuclear power plant and a method for predicting the progress of the serious accidents of the nuclear power plant, so that when the serious accidents happen, the originating events and corresponding parameters of the serious accidents can be accurately diagnosed according to the physical parameters monitored by a limited number of meters, and accurate data are provided for effective implementation of SAMG.
In a first aspect, an embodiment of the present invention provides a method for establishing a diagnostic model library, including: determining a diagnosis target of the serious accident, wherein the diagnosis target of the serious accident comprises N initial events and mapping relations among corresponding parameters of the N initial events; obtaining time sequences of M physical parameters under N starting events to form N sequence sets, wherein each sequence set comprises M columns of data, each column of data is a single time sequence of the physical parameters, and the M physical parameters comprise steam generator water level, steam generator pressure, primary loop pressure, reactor core temperature, water level in a pressure vessel, containment water level, containment hydrogen, containment pressure and containment temperature measured by an instrument, and N and M are positive integers; sequentially selecting T time sequences with the same physical parameter from N sequence sets respectively to form a data set, wherein the data set comprises N first sequence sets, each first sequence set comprises T column data, T is more than or equal to 2 and less than M, and T is a positive integer; and obtaining classification models and regression models corresponding to the T physical parameters, and storing the T physical parameters, the classification models corresponding to the T physical parameters, the regression models and the corresponding relations among the T physical parameters, the regression models and the three into a model library to form a diagnosis model library, wherein the number of the classification models is one.
Preferably, after storing the T physical parameters, the classification model corresponding to the T physical parameters, the regression model, and the correspondence between the three to the model library to form the diagnosis model library, the method for establishing the diagnosis model library further includes: and sequentially selecting T+t time sequences with the same physical parameters from the N sequence sets, wherein t=1, 2,3, & gt, M-T, traversing the value of T, acquiring a classification model and a regression model corresponding to each T+t physical parameter, and storing the corresponding relation among each T+t physical parameter, the classification model corresponding to the T+t physical parameter, the regression model and the three of the classification model, the regression model into a diagnosis model library.
Preferably, the acquiring the time sequence of the M physical parameters under the N originating events forms N sequence sets, which specifically includes: and calling a serious accident integrated analysis program to respectively perform simulation calculation on parameters corresponding to the N initial events, and acquiring time sequences of the M physical parameters under the N initial events to form N sequence sets.
Preferably, the sequentially selecting T time sequences with the same physical parameter from the N sequence sets to form a data set specifically includes: and sequentially and respectively selecting T time sequences with the same physical parameters, wherein the correlation with the diagnostic target is larger than a first correlation threshold value, and the instrument availability is larger than the first instrument threshold value, from the N sequence sets to form a data set.
Preferably, the obtaining a classification model and a regression model corresponding to the T physical parameters specifically includes: performing deep learning model training and performance evaluation on the data set by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to the T physical parameters; and performing performance evaluation on each model in the classification model set and the regression model set corresponding to the T physical parameters, obtaining a classification model and a regression model corresponding to the performance evaluation value exceeding the performance threshold, and determining the classification model and the regression model corresponding to the T physical parameters. Selecting t+t time sequences with the same physical parameter from the N sequence sets in turn, wherein t=1, 2,3,..m-T, traversing the value of T, and obtaining a classification model and a regression model corresponding to each t+t physical parameters, specifically including: sequentially selecting T+t time sequences with the same physical parameters from the N sequence sets, wherein t=1, 2,3, & gt, M-T, traversing the value of T, and respectively performing deep learning model training and performance evaluation on each selected time sequence by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to each T+t physical parameters; and performing performance evaluation on each model in the classification model set and the regression model set corresponding to each T+t physical parameters, obtaining classification models and regression models corresponding to performance evaluation values exceeding a performance threshold, and determining the classification models and the regression models corresponding to each T+t physical parameters, wherein the number of the classification models is the same as the selection frequency of T.
In a second aspect, an embodiment of the present invention further provides a method for diagnosing a serious accident in a nuclear power plant, including: acquiring a diagnosis model library established by the establishment method of the diagnosis model library in the first aspect; and after the serious accident occurs, acquiring diagnosis results of the serious accident starting event and corresponding parameters according to real-time data of physical parameters monitored by the on-site available instrument and a preset classification model and regression model in a diagnosis model library.
Preferably, the acquiring the diagnosis result of the serious accident originating event and the corresponding parameters thereof according to the real-time data of the physical parameters monitored by the available meters on site and the predetermined classification model and regression model in the diagnosis model library specifically comprises: selecting a classification model and a regression model corresponding to physical parameters monitored by an available instrument on site from a diagnosis model library as the preset classification model and regression model; inputting real-time data of physical parameters monitored by an available instrument on site into the preset classification model and regression model to obtain a diagnosis result of a serious accident starting event and corresponding parameters, wherein the classification model is used for diagnosing the serious accident starting event, and the regression model is used for diagnosing the corresponding parameters of the serious accident starting event.
In a third aspect, an embodiment of the present invention further provides a process prediction method for a severe accident in a nuclear power plant, including: the diagnosis results of the severe accident originating event and the corresponding parameters obtained by the diagnosis method of the severe accident of the nuclear power plant according to the second aspect are input into the severe accident integration analysis program to obtain the predicted value of the severe accident process, wherein the predicted value comprises the positive and negative effects of the intervention means.
In a fourth aspect, an embodiment of the present invention further provides an apparatus for establishing a diagnostic model library, including a determining module, an obtaining module, a first selecting module, and a constructing module. The determining module is used for determining a diagnosis target of the serious accident, wherein the diagnosis target of the serious accident comprises N initial events and mapping relations among corresponding parameters of the N initial events. The acquisition module is connected with the determination module and is used for acquiring time sequences of M physical parameters under N initial events to form N sequence sets, wherein each sequence set comprises M columns of data, each column of data is a single time sequence of the physical parameters, and the M physical parameters comprise steam generator water level, steam generator pressure, loop pressure, core temperature, water level in a pressure vessel, containment water level, containment hydrogen, containment pressure and containment temperature which are measured by the instrument, and N and M are positive integers. The first selecting module is connected with the acquiring module and used for sequentially selecting T time sequences with the same physical parameters from N sequence sets respectively to form a data set, wherein the data set comprises N first sequence sets, each first sequence set comprises T column data, T is more than or equal to 2 and less than M, and T is a positive integer. The construction module is connected with the first selection module and used for acquiring a classification model and a regression model corresponding to the T physical parameters and storing the T physical parameters, the classification model corresponding to the T physical parameters, the regression model and the corresponding relation among the T physical parameters and the regression model into a model library to form a diagnosis model library, wherein the number of the classification models is one.
In a fifth aspect, an embodiment of the present invention further provides a device for diagnosing a serious accident in a nuclear power plant, including the device for establishing a diagnostic model library, the second selection module, and the diagnostic module described in the fourth aspect. And the diagnostic model library establishing device is used for establishing the diagnostic model library. The second selecting module is connected with the establishing device of the diagnosis model library and is used for selecting a classification model and a regression model corresponding to physical parameters monitored by the available instrument on site from the diagnosis model library as the preset classification model and regression model. The diagnosis module is connected with the second selection module and is used for inputting real-time data of physical parameters monitored by the available instruments on site into the preset classification model and the regression model to obtain diagnosis results of the serious accident starting event and corresponding parameters, wherein the classification model is used for diagnosing the serious accident starting event, and the regression model is used for diagnosing the parameters corresponding to the serious accident starting event.
The invention relates to a method and a device for establishing a diagnosis model library, a method and a device for diagnosing serious accidents of a nuclear power plant, and a method for predicting the progress of the serious accidents of the nuclear power plant, wherein the diagnosis targets of the serious accidents are determined; obtaining time sequences of M physical parameters under N initial events according to a diagnosis target to form N sequence sets; sequentially selecting T time sequences with the same physical parameters from N sequence sets respectively to form a data set, wherein N and M are positive integers, T is more than or equal to 2 and less than M, and T is a positive integer; and obtaining a classification model and a regression model corresponding to the T physical parameters according to the data set, and storing the classification model and the regression model in a model library to form a diagnosis model library. After serious accidents happen, the initial events and corresponding parameters of the serious accidents are accurately diagnosed according to the limited number of physical parameters monitored by the available instruments and the classification model and the regression model in the diagnosis model library, so that accurate data are provided for effective implementation of the SAMG.
Drawings
Fig. 1: a flowchart of a method for establishing a diagnostic model library according to embodiment 1 of the present invention;
fig. 2: a structural diagram of a diagnostic model library creation apparatus according to embodiment 4 of the present invention;
fig. 3: the structural diagram of a diagnosis device for a severe accident in a nuclear power plant according to embodiment 5 of the present invention is shown.
Detailed Description
The present invention will be described in further detail below with reference to the drawings and examples for better understanding of the technical scheme of the present invention to those skilled in the art.
Example 1:
as shown in fig. 1, the present embodiment provides a method for establishing a diagnostic model library, including:
step 101, determining a diagnosis target of the serious accident, wherein the diagnosis target of the serious accident comprises mapping relations among N initial events and corresponding parameters thereof.
In this embodiment, a diagnosis target of the serious accident is determined, that is, a mapping relationship between N originating events and their corresponding parameters is determined. As shown in the originating event list in table 1 (not limited to the originating events and parameters thereof listed in the table), for the "primary event of a loop coolant system break" in the table, there are corresponding parameter values such as break size, break position, and blowout coefficient, and specific parameter values corresponding to N originating events of a serious accident can be determined by referring to the literature.
TABLE 1
| Sequence number | Initiating an event | Parameters of originating event |
| 1 | A circuit coolant system breach | Size of break, location of break, coefficient of blowout, etc |
| 2 | Water supply pipeline breach | Size of the break, location of the break, etc |
| 3 | Main steam pipeline break | Size of the break, location of the break, etc |
| 4 | Steam generator heat transfer tube rupture | Fracture position, number of fractures, etc |
| ... | ...... | ...... |
| N-1 | ...... | ...... |
| N | ...... | ...... |
Step 102, obtaining time sequences of M physical parameters under N initial events to form N sequence sets, wherein each sequence set comprises M columns of data, each column of data is a time sequence of single physical parameters, and the M physical parameters comprise steam generator water level, steam generator pressure, primary loop pressure, reactor core temperature, water level in a pressure vessel, containment water level, containment hydrogen, containment pressure and containment temperature measured by an instrument, and N and M are positive integers.
In this embodiment, the physical parameters refer to parameters that can still be monitored by the meter after a serious accident, such as steam generator water level, steam generator pressure, primary loop pressure, core temperature, pressure vessel water level, containment hydrogen, containment pressure, containment temperature, and other physical parameters. In this embodiment, by calling a severe accident integration analysis program, simulation calculation is performed on parameters corresponding to N originating events, so as to obtain time sequences of M physical parameters under N originating events, and form N sequence sets. For example, a time series of M physical parameters developed at the onset of a loop coolant system breach is obtained, resulting in a series set comprising M columns of data, wherein each column of data is a time series of physical parameters. Common severe accident integration analysis programs include MAAP developed by Fauske & Associates, usa, meloc developed by Sandia national laboratory, usa, ASTEC developed by IRSN in france and GRS in germany, etc., and physical parameters may be set in the input card of the severe accident integration analysis program.
Step 103, sequentially selecting T time sequences with the same physical parameter from N sequence sets respectively to form a data set, wherein the data set comprises N first sequence sets, each first sequence set comprises T column data, T is more than or equal to 2 and less than M, and T is a positive integer.
In this embodiment, a time series of four physical parameters, i.e., steam generator water level, primary circuit pressure, core temperature, and water level in the pressure vessel, is selected from the 1 st set of sequences, where t=4; then selecting the time sequences of the four same physical parameters from the 2 nd sequence set; sequentially selecting until the time sequences of the four same physical parameters are selected from the Nth sequence set, and obtaining a data set, wherein the data set comprises N first sequence sets, and each first sequence set comprises a time sequence of the steam generator water level, the primary circuit pressure, the core temperature and the water level in the pressure vessel.
Optionally, T time sequences with the same physical parameter, which have a correlation with the diagnostic target greater than the first correlation threshold and an instrument availability greater than the first instrument threshold, are sequentially selected from the N sequence sets, respectively, to form a data set.
In this embodiment, physical parameters with correlation with the diagnostic target greater than the first correlation threshold and availability of the meters greater than the first meter threshold are selected, so that the obtained classification model and regression model corresponding to the T physical parameters are matched with the important meter combination with the best availability. It should be noted that, the correlations between the M physical parameters and the diagnostic target are obtained in advance according to experiments or experience, and a list is ordered according to the correlations, where the list includes the physical parameters, the correlations, and the mapping relationship between the physical parameters and the correlations. In addition, an ordered list of meter availability is obtained in advance according to tests or experience, and the meters have a corresponding relationship with the physical parameters.
Step 104, obtaining classification models and regression models corresponding to the T physical parameters, and storing the T physical parameters, the classification models corresponding to the T physical parameters, the regression models and the corresponding relations among the T physical parameters, the regression models and the three models into a model library to form a diagnosis model library, wherein the number of the classification models is one.
In this embodiment, the data set obtained in step 103 is subjected to training and performance evaluation of a deep learning model, and for diagnosis of a severe accident originating event, classification models corresponding to 4 physical parameters are established, wherein the number of the classification models is one; for the diagnosis of the serious accident initiation event parameters, regression models corresponding to 4 physical parameters are established, the number of the regression models exceeds N, for example, the number of the regression models for diagnosing the corresponding parameters of the 'primary loop coolant system break' initiation event is three, and the regression models are respectively used for diagnosing the break position, the break size and the blowout coefficient. And storing the 4 physical parameters (namely the steam generator water level, the loop pressure, the reactor core temperature and the water level in the pressure vessel), the classification model corresponding to the 4 physical parameters, the regression model and the corresponding relation among the three to a model library so as to form a diagnosis model library. The diagnosis of the original event of the serious accident and the parameters thereof can be realized by using the classification model and the regression model in the diagnosis model library.
Optionally, step 104 specifically includes: performing deep learning model training and performance evaluation on the data set obtained in the step 103 by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to the T physical parameters; and performing performance evaluation on each model in the classification model set and the regression model set corresponding to the T physical parameters, obtaining a classification model and a regression model corresponding to the performance evaluation value exceeding the performance threshold, and determining the classification model and the regression model corresponding to the T physical parameters.
Optionally, after storing the T physical parameters, the classification model corresponding to the T physical parameters, the regression model, and the correspondence between the three to the model library to form the diagnosis model library, the method for establishing the diagnosis model library further includes:
and 105, sequentially selecting T+t time sequences with the same physical parameters from the N sequence sets, wherein t=1, 2,3, & gt, traversing the value of T by M-T, acquiring a classification model and a regression model corresponding to each T+t physical parameter, and storing the T+t physical parameters, the classification model and the regression model corresponding to the T+t physical parameters and the correspondence relation among the T+t physical parameters into a diagnosis model library.
In this embodiment, when a serious accident occurs, since the situation that the instrument on the site for monitoring the physical parameter is damaged or the data is inaccurate is uncertain, and the more the types of the physical parameters in the training data set are, the higher the accuracy of the corresponding classification model and regression model is, so that the classification model and regression model corresponding to various situations are established to perfect the model in the diagnosis model library. On the basis of the embodiment, 4+1 time sequences with the same physical parameters (such as steam generator water level, loop pressure, reactor core temperature, water level in a pressure vessel and containment water level) are selected from N sequence sets respectively to form a data set, deep learning model training and performance evaluation are carried out on the data set, and a classification model and a regression model corresponding to the 4+1 physical parameters are obtained and stored in a diagnosis model library; then, respectively selecting 4+2 time sequences with the same physical parameters (such as steam generator water level, loop pressure, reactor core temperature, water level in a pressure vessel, containment water level and containment hydrogen) from the N sequence sets to form a data set, performing deep learning model training and performance evaluation on the data set, acquiring a classification model and a regression model corresponding to the 4+2 physical parameters, and storing the classification model and the regression model into a diagnosis model library; and traversing the value of t until a classification model and a regression model corresponding to the M physical parameters are obtained. It should be noted that, although in the present embodiment, the classification model and the regression model corresponding to each physical parameter combination are obtained in a manner of increasing one physical parameter, the classification model and the regression model corresponding to each physical parameter combination may also be obtained in a manner of increasing a plurality of physical parameters.
Optionally, step 105 specifically includes: sequentially selecting T+t time sequences with the same physical parameters from the N sequence sets, wherein t=1, 2,3, & gt, M-T, traversing the value of T, and respectively performing deep learning model training and performance evaluation on each selected time sequence by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to each T+t physical parameters; and performing performance evaluation on each model in the classification model set and the regression model set corresponding to each T+t physical parameters, obtaining a classification model and a regression model corresponding to the performance evaluation value exceeding the performance threshold, and determining the classification model and the regression model corresponding to each T+t physical parameters, wherein the number of the classification models is the same as the selection frequency (namely the traversal frequency) of T. In addition, the deep learning model training and performance evaluation uses a common deep learning framework, such as deep learning4J, tensorFlow, pytorch, and uses the LSTM algorithm integrated within these frameworks.
In the method for establishing the diagnosis model library of the embodiment, the data set corresponding to the T physical parameters is obtained by simulating the initial event and the corresponding parameters thereof, and the data set is subjected to deep learning model training and performance evaluation to obtain the classification model and the regression model corresponding to the T physical parameters and store the classification model and the regression model in the model library so as to form the diagnosis model library. After serious accidents happen, the initial events and corresponding parameters of the serious accidents are accurately diagnosed according to the physical parameters monitored by T limited number of available meters and the classification model and the regression model in the diagnosis model library, so that accurate data are provided for effective implementation of the SAMG. In addition, t+t time sequences with the same physical parameter are sequentially selected from the N sequence sets, wherein t=1, 2,3, & gt, M-T, traversing the values of T, obtaining classification models and regression models corresponding to the t+t physical parameters, and storing the classification models and the regression models in a diagnosis model library, so that the number of models and the quality of the models in the diagnosis model library are perfected, the models matched with the physical parameters monitored by the instruments can be selected for diagnosis according to the available instruments on site, the available instruments on site are utilized to the maximum extent, and the accuracy of diagnosis results based on the initial events and the corresponding parameters of the initial events can be further improved.
Example 2:
the embodiment provides a method for diagnosing serious accidents of a nuclear power plant, which comprises the following steps:
step 201, obtaining a diagnostic model library established by the diagnostic model library establishment method described in embodiment 1.
Step 202, after a serious accident occurs, acquiring diagnosis results of a serious accident originating event and corresponding parameters according to real-time data of physical parameters monitored by available meters on site and a predetermined classification model and regression model in a diagnosis model library.
Optionally, step 202 specifically includes: selecting a classification model and a regression model corresponding to physical parameters monitored by an available instrument on site from a diagnosis model library as a preset classification model and a preset regression model; inputting real-time data of physical parameters monitored by an available instrument on site into a preset classification model and a regression model to obtain diagnosis results of the serious accident starting event and corresponding parameters, wherein the classification model is used for diagnosing the serious accident starting event, and the regression model is used for diagnosing the corresponding parameters of the serious accident starting event.
In this embodiment, after a serious accident occurs, when the available meters on the site monitor t+5 physical parameters, a classification model and a regression model corresponding to the t+5 physical parameters are selected from a diagnosis model library as a predetermined classification model and a regression model, and real-time data of the t+5 physical parameters monitored by the available meters on the site are input to the predetermined classification model and regression model, so as to obtain a diagnosis result of the serious accident originating event and the parameters corresponding to the event.
In the method for diagnosing serious accidents of the nuclear power plant provided by the embodiment, the diagnosis results of the initial event and the parameters thereof are obtained according to the perfect diagnosis model in the diagnosis model library and the real-time data of the physical parameters monitored by the available instrument on site. Because the diagnosis model library is provided with the diagnosis models matched with different physical parameter combinations, the data monitored by the available meters on site are utilized to perform diagnosis maximally, and the accuracy of the diagnosis result is higher.
Example 3:
the embodiment provides a process prediction method for severe accidents of a nuclear power plant, which comprises the following steps: the diagnosis results of the severe accident origin event and the corresponding parameters thereof obtained by the diagnosis method of the severe accident in the nuclear power plant according to embodiment 2 are inputted into the severe accident integration analysis program to obtain the predicted value of the severe accident process, and the predicted value includes the positive and negative effects of the intervention means. Serious accident integration analysis programs include MAAP developed by Fauske & Associates, MELCOR developed by Sandia national laboratory, USA, ASTEC developed by IRSN, france, and GRS, germany, etc.
Example 4:
as shown in fig. 2, the present embodiment provides a diagnostic model library creating apparatus 4, which includes a determining module 41, an obtaining module 42, a first selecting module 43, and a creating module 44.
The determining module 41 is configured to determine a diagnosis target of the serious accident, where the diagnosis target of the serious accident includes mapping relationships between N originating events and corresponding parameters thereof. Diagnostic targets are for example table 1 described in example 1.
The obtaining module 42 is connected to the determining module 41, and is configured to obtain time sequences of M physical parameters under N originating events to form N sequence sets, where each sequence set includes M columns of data, each column of data is a time sequence of a single physical parameter, and the M physical parameters include a steam generator water level, a steam generator pressure, a loop pressure, a core temperature, a pressure vessel water level, a containment water level, containment hydrogen, containment pressure, and containment temperature measured by the meter, where N and M are positive integers.
Optionally, the obtaining module 42 is configured to invoke the integrated analysis program for serious accident to perform simulation calculation on parameters corresponding to the N originating events, respectively, so as to obtain a time sequence of M physical parameters under the N originating events.
The first selecting module 43 is connected to the obtaining module 42, and is configured to sequentially select T time sequences having the same physical parameter from the N sequence sets, respectively, to form a data set, where the data set includes N first sequence sets, each first sequence set includes T column data, T is greater than or equal to 2 and less than M, and T is a positive integer.
Optionally, the selecting module 43 is configured to sequentially select, from the N sequence sets, T time sequences having the same physical parameter and having a correlation with the diagnostic target greater than the first correlation threshold and a meter reliability greater than the first meter threshold, respectively, to form a data set.
The construction module 44 is connected to the first selection module 43, and is configured to obtain classification models and regression models corresponding to the T physical parameters, and store the T physical parameters, the classification models corresponding to the T physical parameters, the regression models, and the correspondence between the three to the model library, so as to form a diagnosis model library, where the number of classification models is one, and the number of regression models exceeds N.
Optionally, the construction module 44 is further configured to sequentially select t+t time sequences with the same physical parameters from the N sequence sets, where t=1, 2,3,..m-T, traverse the value of T, obtain a classification model and a regression model corresponding to each t+t physical parameter, and store the corresponding relationship among each t+t physical parameter, the classification model corresponding to each t+t physical parameter, the regression model, and the three to the diagnosis model library.
Example 5:
as shown in fig. 3, the present embodiment provides a diagnostic apparatus for a severe accident in a nuclear power plant, which includes the diagnostic model library creating apparatus 4, the second selecting module 51 and the diagnostic module 52 described in embodiment 4,
and the diagnostic model library establishing device 4 is used for establishing the diagnostic model library.
The second selecting module 51 is connected to the diagnostic model library creating device 4, and is configured to select, from the diagnostic model library, a classification model and a regression model corresponding to physical parameters monitored by the available meters in the field, as the predetermined classification model and regression model.
The diagnosis module 52 is connected to the second selection module 51, and is configured to input real-time data of physical parameters monitored by the available meters on site into a predetermined classification model and a regression model, and obtain a diagnosis result of a serious accident event and corresponding parameters thereof, where the classification model is used for diagnosing the serious accident event, and the regression model is used for diagnosing the parameters corresponding to the serious accident event.
It is to be understood that the above embodiments are merely illustrative of the application of the principles of the present invention, but not in limitation thereof. Various modifications and improvements may be made by those skilled in the art without departing from the spirit and substance of the invention, and are also considered to be within the scope of the invention.
Claims (9)
1. A method of establishing a diagnostic model library, comprising:
determining a diagnosis target of the serious accident, wherein the diagnosis target of the serious accident comprises N initial events and mapping relations among corresponding parameters of the N initial events;
obtaining time sequences of M physical parameters under N starting events to form N sequence sets, wherein each sequence set comprises M columns of data, each column of data is a single time sequence of the physical parameters, and the M physical parameters comprise steam generator water level, steam generator pressure, primary loop pressure, reactor core temperature, water level in a pressure vessel, containment water level, containment hydrogen, containment pressure and containment temperature measured by an instrument, and N and M are positive integers;
sequentially selecting T time sequences with the same physical parameter from N sequence sets respectively to form a data set, wherein the data set comprises N first sequence sets, each first sequence set comprises T column data, T is more than or equal to 2 and less than M, and T is a positive integer;
obtaining classification models and regression models corresponding to the T physical parameters, and storing the T physical parameters, the classification models corresponding to the T physical parameters, the regression models and the corresponding relations among the T physical parameters, the regression models and the three into a model library to form a diagnosis model library, wherein the number of the classification models is one,
the method specifically includes the steps of obtaining time sequences of M physical parameters under N originating events to form N sequence sets, wherein the time sequences include:
calling a serious accident integrated analysis program to respectively perform simulation calculation on parameters corresponding to N initial events, obtaining time sequences of M physical parameters under the N initial events to form N sequence sets,
the obtaining a classification model and a regression model corresponding to the T physical parameters specifically comprises the following steps:
performing deep learning model training and performance evaluation on the data set by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to the T physical parameters;
and performing performance evaluation on each model in the classification model set and the regression model set corresponding to the T physical parameters, obtaining a classification model and a regression model corresponding to the performance evaluation value exceeding the performance threshold, and determining the classification model and the regression model corresponding to the T physical parameters.
2. The method according to claim 1, wherein after storing the T physical parameters, the classification model corresponding to the T physical parameters, the regression model, and the correspondence relationship among the three to the model library to form the diagnosis model library, further comprising:
and sequentially selecting T+t time sequences with the same physical parameters from the N sequence sets, wherein t=1, 2,3, & gt, M-T, traversing the value of T, acquiring a classification model and a regression model corresponding to each T+t physical parameter, and storing the corresponding relation among each T+t physical parameter, the classification model corresponding to the T+t physical parameter, the regression model and the three of the classification model, the regression model into a diagnosis model library.
3. The method for building a diagnostic model library according to claim 2, wherein T time sequences having the same physical parameter are selected from the N sequence sets in sequence, respectively, to form a data set, specifically comprising:
and sequentially and respectively selecting T time sequences with the same physical parameters, wherein the correlation with the diagnostic target is larger than a first correlation threshold value, and the instrument availability is larger than the first instrument threshold value, from the N sequence sets to form a data set.
4. The method for creating a diagnostic model library according to claim 2, wherein,
selecting t+t time sequences with the same physical parameter from the N sequence sets in turn, wherein t=1, 2,3,..m-T, traversing the value of T, and obtaining a classification model and a regression model corresponding to each t+t physical parameters, specifically including:
sequentially selecting T+t time sequences with the same physical parameters from the N sequence sets, wherein t=1, 2,3, & gt, M-T, traversing the value of T, and respectively performing deep learning model training and performance evaluation on each selected time sequence by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to each T+t physical parameters;
and performing performance evaluation on each model in the classification model set and the regression model set corresponding to each T+t physical parameters, obtaining classification models and regression models corresponding to performance evaluation values exceeding a performance threshold, and determining the classification models and the regression models corresponding to each T+t physical parameters, wherein the number of the classification models is the same as the selection frequency of T.
5. A method for diagnosing a serious accident in a nuclear power plant, comprising:
obtaining a diagnostic model library established by the diagnostic model library establishment method of any one of claims 1 to 4;
and after the serious accident occurs, acquiring diagnosis results of the serious accident starting event and corresponding parameters according to real-time data of physical parameters monitored by the on-site available instrument and a preset classification model and regression model in a diagnosis model library.
6. The method for diagnosing a serious accident in a nuclear power plant according to claim 5, wherein the obtaining the diagnosis results of the serious accident event and the corresponding parameters thereof according to the real-time data of the physical parameters monitored by the available meters on site and the predetermined classification model and regression model in the diagnosis model library specifically comprises:
selecting a classification model and a regression model corresponding to physical parameters monitored by an available instrument on site from a diagnosis model library as the preset classification model and regression model;
inputting real-time data of physical parameters monitored by an available instrument on site into the preset classification model and regression model to obtain a diagnosis result of a serious accident starting event and corresponding parameters, wherein the classification model is used for diagnosing the serious accident starting event, and the regression model is used for diagnosing the corresponding parameters of the serious accident starting event.
7. A process prediction method for a severe accident in a nuclear power plant, comprising:
the diagnosis results of the severe accident origin event and the corresponding parameters thereof obtained by the diagnosis method of the severe accident of the nuclear power plant according to claim 5 or 6 are inputted into the severe accident integration analysis program to obtain the predicted value of the severe accident process, and the predicted value includes the positive and negative effects of the intervention means.
8. The device for establishing the diagnosis model library is characterized by comprising a determining module, an acquiring module, a first selecting module and a constructing module,
the determining module is used for determining a diagnosis target of the serious accident, wherein the diagnosis target of the serious accident comprises N initial events and mapping relations among corresponding parameters of the N initial events;
the acquisition module is connected with the determination module and is used for acquiring time sequences of M physical parameters under N initial events to form N sequence sets, wherein each sequence set comprises M columns of data, each column of data is a single time sequence of the physical parameters, and the M physical parameters comprise steam generator water level, steam generator pressure, loop pressure, reactor core temperature, water level in a pressure container, containment water level, containment hydrogen, containment pressure and containment temperature which are measured by an instrument, and N and M are positive integers;
the first selecting module is connected with the acquiring module and is used for sequentially selecting T time sequences with the same physical parameters from N sequence sets respectively to form a data set, wherein the data set comprises N first sequence sets, each first sequence set comprises T column data, T is more than or equal to 2 and less than M, and T is a positive integer;
the construction module is connected with the first selection module and is used for acquiring a classification model and a regression model corresponding to the T physical parameters, and storing the T physical parameters, the classification model corresponding to the T physical parameters, the regression model and the corresponding relation among the T physical parameters and the regression model into a model library to form a diagnosis model library, wherein the number of the classification models is one,
wherein the acquisition module is also used for calling a serious accident integration analysis program to respectively simulate and calculate parameters corresponding to the N initial events to acquire time sequences of M physical parameters under the N initial events,
the construction module is also used for carrying out deep learning model training and performance evaluation on the data set by adopting a long-short-term memory algorithm to obtain a classification model set and a regression model set corresponding to the T physical parameters, carrying out performance evaluation on each model in the classification model set and the regression model set corresponding to the T physical parameters, obtaining a classification model and a regression model corresponding to the performance evaluation value exceeding a performance threshold, and determining the classification model and the regression model corresponding to the T physical parameters.
9. A diagnosis device for severe accidents of a nuclear power plant is characterized by comprising the establishment device of the diagnosis model library, a second selection module and a diagnosis module of claim 8,
diagnostic model library establishing means for establishing a diagnostic model library,
a second selecting module connected with the establishing device of the diagnosis model library and used for selecting a classification model and a regression model corresponding to the physical parameters monitored by the available instrument in the field from the diagnosis model library as a preset classification model and a regression model,
the diagnosis module is connected with the second selection module and is used for inputting real-time data of physical parameters monitored by the available meters on site into the preset classification model and regression model to obtain diagnosis results of the serious accident originating event and the corresponding parameters thereof,
the classification model is used for diagnosing the serious accident starting event, and the regression model is used for diagnosing the corresponding parameters of the serious accident starting event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110095211.8A CN112819053B (en) | 2021-01-25 | 2021-01-25 | Model library establishing method and device, diagnosis method and device and prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110095211.8A CN112819053B (en) | 2021-01-25 | 2021-01-25 | Model library establishing method and device, diagnosis method and device and prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112819053A CN112819053A (en) | 2021-05-18 |
| CN112819053B true CN112819053B (en) | 2024-04-16 |
Family
ID=75859269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110095211.8A Active CN112819053B (en) | 2021-01-25 | 2021-01-25 | Model library establishing method and device, diagnosis method and device and prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112819053B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114996821A (en) * | 2022-06-28 | 2022-09-02 | 中建八局装饰工程有限公司 | Curtain wall cavity air tightness judgment method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103366047A (en) * | 2013-06-24 | 2013-10-23 | 中国核电工程有限公司 | Severe accident countermeasure computing and analyzing method for nuclear power plant |
| CN104915768A (en) * | 2015-06-02 | 2015-09-16 | 中科华核电技术研究院有限公司 | Method and system for diagnosing and responding and supporting nuclear power plant severe accident |
| CN109597396A (en) * | 2018-11-26 | 2019-04-09 | 国网湖北省电力有限公司电力科学研究院 | A kind of distribution transforming on-line fault diagnosis method based on high amount of traffic and transfer learning |
| CN112001440A (en) * | 2020-08-20 | 2020-11-27 | 苏州鸿哲智能科技有限公司 | Fault diagnosis logic algorithm and system |
-
2021
- 2021-01-25 CN CN202110095211.8A patent/CN112819053B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103366047A (en) * | 2013-06-24 | 2013-10-23 | 中国核电工程有限公司 | Severe accident countermeasure computing and analyzing method for nuclear power plant |
| CN104915768A (en) * | 2015-06-02 | 2015-09-16 | 中科华核电技术研究院有限公司 | Method and system for diagnosing and responding and supporting nuclear power plant severe accident |
| CN109597396A (en) * | 2018-11-26 | 2019-04-09 | 国网湖北省电力有限公司电力科学研究院 | A kind of distribution transforming on-line fault diagnosis method based on high amount of traffic and transfer learning |
| CN112001440A (en) * | 2020-08-20 | 2020-11-27 | 苏州鸿哲智能科技有限公司 | Fault diagnosis logic algorithm and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112819053A (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hofer et al. | An approximate epistemic uncertainty analysis approach in the presence of epistemic and aleatory uncertainties | |
| Prošek et al. | Review of quantitative accuracy assessments with fast Fourier transform based method (FFTBM) | |
| EP0358994A1 (en) | Corrosion-erosion trend monitoring and diagnostic system | |
| US6195624B1 (en) | Corrosion analysis system and method | |
| CN104915768A (en) | Method and system for diagnosing and responding and supporting nuclear power plant severe accident | |
| Guan et al. | Life time extension of turbine rotating components under risk constraints: A state-of-the-art review and case study | |
| CN103279657B (en) | A kind of product accelerated degradation test Design Method based on engineering experience | |
| CN112819053B (en) | Model library establishing method and device, diagnosis method and device and prediction method | |
| Lei et al. | Prediction of crucial nuclear power plant parameters using long short‐term memory neural networks | |
| CN109359803B (en) | Reliability evaluation method and device | |
| Dal Vernon et al. | Work domain analysis and sensors I: principles and simple example | |
| CN106250993A (en) | A kind of measurement equipment metering cycle method of adjustment based on warship maintenance section | |
| CN114298226A (en) | Fault diagnosis method, device and medium for nuclear power plant | |
| CN109215821A (en) | Nuclear power plant evaporator main feedwater flow redundant measurement channel comparative approach and system | |
| Wu et al. | Online fault diagnosis of nuclear power plants using signed directed graph and fuzzy theory | |
| CN110688152B (en) | Software reliability quantitative evaluation method combining software development quality information | |
| Pitner et al. | Application of probabilistic fracture mechanics to optimize the maintenance of PWR steam generator tubes | |
| Farber et al. | Using multiple-model adaptive estimation and system identification for fault detection in nuclear power plants | |
| Diaconeasa et al. | Discrete dynamic event tree uncertainty quantification in the ADS-IDAC dynamic PSA software platform | |
| Boring et al. | Human reliability studies with microworld simulators | |
| CN111768125A (en) | Foundation pit deformation safety assessment method and system | |
| CN112259271A (en) | Reactor core thermal power calculation method and device for nuclear power station DCS | |
| Santhosh et al. | Transient Identification & Operator Decision Support System for PHWR (Diagnostic System) | |
| Mercier et al. | A stochastic model for competing degradations | |
| Povarov et al. | Developing and applying modern methods of leakage monitoring and state estimation of fuel at the Novovoronezh nuclear power plant |
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 |