CN113761749B - Nuclear reactor probability safety margin analysis method, system, terminal and storage medium - Google Patents
Nuclear reactor probability safety margin analysis method, system, terminal and storage medium Download PDFInfo
- Publication number
- CN113761749B CN113761749B CN202111063589.6A CN202111063589A CN113761749B CN 113761749 B CN113761749 B CN 113761749B CN 202111063589 A CN202111063589 A CN 202111063589A CN 113761749 B CN113761749 B CN 113761749B
- Authority
- CN
- China
- Prior art keywords
- analysis
- probability
- psm
- sequence
- nuclear reactor
- 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.)
- Expired - Fee Related
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 106
- 238000003860 storage Methods 0.000 title claims abstract description 11
- 238000000034 method Methods 0.000 claims abstract description 95
- 238000004364 calculation method Methods 0.000 claims abstract description 61
- 238000005070 sampling Methods 0.000 claims abstract description 36
- 238000012216 screening Methods 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims description 28
- 238000011160 research Methods 0.000 claims description 11
- 238000012284 sample analysis method Methods 0.000 claims description 8
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 230000035945 sensitivity Effects 0.000 claims description 2
- 238000002347 injection Methods 0.000 description 30
- 239000007924 injection Substances 0.000 description 30
- 238000005253 cladding Methods 0.000 description 17
- 238000009826 distribution Methods 0.000 description 15
- 238000004088 simulation Methods 0.000 description 13
- 239000002826 coolant Substances 0.000 description 11
- 238000000342 Monte Carlo simulation Methods 0.000 description 10
- 238000010206 sensitivity analysis Methods 0.000 description 9
- 239000000446 fuel Substances 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 239000000243 solution Substances 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005507 spraying Methods 0.000 description 4
- 238000013076 uncertainty analysis Methods 0.000 description 4
- 101000691787 Homo sapiens Phosphoinositide-interacting protein Proteins 0.000 description 3
- 102100026124 Phosphoinositide-interacting protein Human genes 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 238000012614 Monte-Carlo sampling Methods 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Monitoring And Testing Of Nuclear Reactors (AREA)
Abstract
The invention discloses a nuclear reactor probability safety margin analysis method, a system, a terminal and a storage medium, which are applied to the technical field of nuclear reactor safety analysis and comprise the following steps: the method comprises the steps of data acquisition, model building and analyzing, probability calculation and PSM solving. The invention establishes a reduced-order model by screening event tree branches and combining small samples with an advanced sampling method, and improves PSM analysis efficiency from three levels of reducing branch sequence number, reducing single-branch calculation number and improving single-case calculation efficiency.
Description
Technical Field
The invention relates to the technical field of nuclear reaction safety analysis, in particular to a nuclear reactor probability safety margin analysis method, a nuclear reactor probability safety margin analysis system, a nuclear reactor probability safety margin analysis terminal and a storage medium.
Background
As an important clean energy source, how to prolong the service life of a nuclear power plant and properly increase the power so as to improve the economic benefit of the nuclear power plant is an important means for achieving the goals of carbon peak reaching and carbon neutralization. The safety margin is an important safety index when the nuclear power plant is used for prolonging the service life and evaluating a proper power increasing scheme. The safety margin obtained by the traditional determinism safety analysis method is the difference value between the safety limit value and the actual operation value of the key parameters of the nuclear power plant. As shown in FIG. 1, the safety margin is always determined by the difference between two values, whether it is the traditional conservative calculation method or the most advanced best estimate plus uncertainty analysis (BEPU) method. However, during operation of the nuclear power plant, the safety limits of the nuclear power plant and the actual operating values of the key parameters are dynamically changing, not two determined values, but two distributions. The safety margin obtained by the two distributions is called a Probabilistic Safety Margin (PSM). As shown in fig. 2, a key parameter of interest of the nuclear power plant is a cladding peak temperature (PCT), and a corresponding safety limit of the nuclear power plant is a fuel cladding performance, wherein a shadow overlapping part is a probability safety margin.
The risk guidance safety margin characteristic analysis method (RISMC) combining the determinism and the probability theory can effectively analyze the nuclear power plant probability safety margin. The main analysis steps are shown in FIG. 3. In order to obtain load distribution, the traditional RISMC analysis method needs to perform optimal estimation and uncertainty analysis on all branches of an event tree; and each branch is subjected to a large number of calculations using monte carlo sampling and system simulation programs to ensure calculation confidence. The RISMC analysis method needs to consume a large amount of time resources and calculation resources, and has low analysis efficiency.
Therefore, it is an urgent need to solve the above technical problems by providing a method, a system, a terminal and a storage medium for analyzing a nuclear reactor probability safety margin.
Disclosure of Invention
In view of the above, the invention provides a nuclear reactor probability safety margin analysis method, a nuclear reactor probability safety margin analysis system, a nuclear reactor probability safety margin terminal and a nuclear reactor probability safety margin storage medium.
In order to achieve the purpose, the invention adopts the following technical scheme:
an efficient probabilistic safety margin analysis method for a nuclear reactor, comprising the steps of:
a data acquisition step: acquiring target parameters, modeling requirements and accident sequences of a research object;
model building and analyzing steps: performing key sequence screening and BEPU analysis after the deterministic theory modeling;
and a probability calculation step: adopting a multi-branch event tree/fault tree model to establish a probability safety analysis model, carrying out sequence probability calculation and sequencing, wherein a sequence containing an overall probability value as a set value is a PSM analysis object;
PSM solving step: and calculating the PSM according to the PSM analysis object by combining a small sample with an efficient sampling analysis method and a reduced-order model method.
Optionally, in the data obtaining step, the target parameter of the research object is a parameter representing a safety state of the nuclear power plant corresponding to the specific nuclear power plant and the accident type;
the modeling requirement is specifically that the special safety facilities in the accident process are modeled according to the accident process;
the accident sequence is specifically determined according to an accident process and a key phenomenon, and comprises the following steps: determining a failure sequence, a probability overrun sequence and a safety sequence.
Optionally, the specific content of the model building and analyzing step includes: completing determinism modeling and verifying model accuracy; determining uncertainty parameters in the BEPU analysis process; determining parameter sampling methods and numbers; analyzing and screening key parameters of sensitivity; and (4) screening a key sequence by a limit surface method.
Optionally, the overall probability set value in the probability calculating step is 99.9%.
Optionally, the PSM solving step specifically includes: screening out a key sequence by integrating the deterministic theory screening result and the probabilistic theory screening result; performing BEPU analysis on each sequence by a small sample analysis method; calculating RELAP example by a reduced order model; core damage frequency and PSM solution.
A nuclear reactor efficient probabilistic safety margin analysis system comprising: the device comprises a data acquisition module, a model establishing and analyzing module, a probability calculation module and a PSM solving module;
the data acquisition module is used for acquiring target parameters, modeling requirements and accident sequences of the research object;
the input end of the model building and analyzing module, the input end of the probability calculating module and the output end of the data acquiring module share an endpoint;
the model establishing and analyzing module is used for performing key sequence screening and BEPU analysis after the deterministic theory modeling;
the probability calculation module is used for establishing a probability safety analysis model by adopting a multi-branch event tree/fault tree model, performing sequence probability calculation and sequencing, wherein a sequence containing the overall probability value as a set value is a PSM analysis object;
the input end of the PSM solving module, the output end of the model establishing and analyzing module and the output end of the probability calculating module share a terminal point;
the PSM solving module is used for calculating the PSM according to the PSM analysis object through a small sample combined with an efficient sampling analysis method and a reduced-order model method.
A terminal, comprising: the nuclear reactor safety margin analysis system includes at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform a nuclear reactor efficient probabilistic safety margin analysis method.
A computer readable storage medium having stored thereon computer instructions for causing a computer to perform a nuclear reactor efficient probabilistic safety margin analysis method.
Compared with the prior art, the invention provides a nuclear reactor probability safety margin analysis method, a system, a terminal and a storage medium by the technical scheme: by screening event tree branches and combining small samples with an advanced sampling method, a reduced-order model is established, and PSM analysis efficiency is improved from three levels of reducing the number of branch sequences, reducing the number of single-branch calculation and improving the calculation efficiency of a single case.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic diagram of the concept of safety margin in deterministic theory;
FIG. 2 is a schematic diagram of a nuclear power plant probability safety margin;
FIG. 3 is a flow chart of a conventional RISMC analysis method;
FIG. 4 is a flow chart of a nuclear reactor efficient probabilistic safety margin analysis method of the present invention;
FIG. 5 is a branch screening flow chart of the present invention;
FIG. 6 is a schematic diagram of three sequence types of the present invention, wherein 6.1 is a determined failure sequence, 6.2 is a probability overrun sequence, and 6.3 is a determined safety sequence;
FIG. 7 is a process of the reduced order model construction of the present invention;
FIG. 8 is a block diagram of an efficient probabilistic safety margin analysis system for a nuclear reactor according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 4, the invention discloses a method for analyzing a nuclear reactor high-efficiency probability safety margin, which comprises the following steps:
a data acquisition step: acquiring target parameters, modeling requirements and accident sequences of a research object;
model building and analyzing steps: performing key sequence screening and BEPU analysis after the deterministic theory modeling;
and a probability calculation step: adopting a multi-branch event tree/fault tree model to establish a probability safety analysis model, carrying out sequence probability calculation and sequencing, wherein a sequence containing an overall probability value as a set value is a PSM analysis object;
PSM solving step: and calculating the PSM according to the PSM analysis object by combining a small sample with an efficient sampling analysis method and a reduced-order model method.
In a specific embodiment, in the data acquisition step, the target parameter of the research object is a parameter representing the safety state of the nuclear power plant corresponding to a specific nuclear power plant and an accident type;
the modeling requirement is specifically that the special safety facilities in the accident process are modeled according to the accident process;
the accident sequence is specifically determined according to an accident process and a key phenomenon, and comprises the following steps: determining a failure sequence, a probability overrun sequence and a safety sequence.
In a specific embodiment, the data acquisition step comprises the following steps: and selecting a proper nuclear power plant and a corresponding accident according to research requirements. The key safety parameters involved in different accidents are different, and after a specific nuclear power plant and an accident type are determined, key target parameters need to be determined so as to represent the safety state of the nuclear power plant. For example, for a small break loss of coolant accident in a nuclear power plant, the integrity of the fuel cladding, the first safety barrier in the nuclear power plant, is of primary concern, and thus the key safety parameter is the Peak Cladding Temperature (PCT).
Determining modeling requirements: accident processes of different accidents of the nuclear power plant are different, and related key phenomena and special safety facilities are different, so that targeted modeling is needed according to the accident processes. Only the special safety facilities involved in the accident process need to be modeled, so that the modeling accuracy is ensured, and the modeling efficiency can be improved. For example, the special safety facilities mainly involved in the small-breach loss of coolant accident include a high-pressure safety injection system, a safety injection box system, a low-pressure safety injection system and a containment spraying system. These ad hoc security facilities must be involved in the process of modeling determinism. After all the systems and devices involved in modeling are specified, the granularity of modeling needs to be determined. The simulation is performed by using a system simulation program RELAP5, a reasonable node division scheme needs to be determined for different systems and devices, and the complexity of the system is reduced under the condition of ensuring accurate calculation.
Determining an accident sequence: and (3) researching relevant power plant files, phenomenon identification and classification tables (PIRT) and experience judgment according to different accident types, and determining key phenomena in the accident process. And determining an event sequence according to the accident process and the key phenomenon, and preparing for modeling of a determinism and a probability theory.
In one embodiment, the step of model building and analyzing comprises:
and (3) completing determinism modeling and verifying model accuracy: after the modeling requirements are clarified, a nuclear power plant model is built using the system simulation program RELAP 5. Comparing the steady-state simulation value obtained by the model calculation with the design value of the nuclear power plant, ensuring that each key parameter for modeling is within a reasonable error range, and verifying the accuracy of the model. For transient simulation, the development trend of key parameters and key phenomena are mainly compared. Wherein the error range of the key parameters is within 1 percent. By carrying out uncertainty analysis and calculating multiple groups of initial conditions, the model can be stably calculated, and the stability of the model is verified.
Determining uncertainty parameters in the BEPU analysis process: in the process of BEPU analysis, key input parameters, distribution types of the key parameters and uncertainty ranges of the key parameters need to be determined. The main bases for determining the above data are three: designing and operating files of nuclear power plants. PIRT documents and references. And thirdly, experience judgment in the absence of data. Wherein, the key parameters include: the method comprises the following steps of (1) core power, primary coolant initial flow, primary coolant initial temperature, primary coolant initial pressure, high-pressure safety injection flow, low-pressure safety injection flow, safety injection tank initial pressure, safety injection tank initial safety injection temperature, safety injection tank safety injection pipeline friction coefficient, core thermal channel friction coefficient, breach area, breach friction coefficient, fuel cladding thermal conductivity and fuel cladding heat capacity.
Determining parameter sampling method and number: to perform sensitivity analysis, it is necessary to sample input parameters, and to improve analysis efficiency, it is necessary to select an appropriate sampling method. Different sampling methods have different requirements on the number of sampling samples, and the appropriate number of samples needs to be selected for different sampling methods. The Latin hypercube sampling method is adopted in the text. When the Latin hypercube sampling is used for sensitivity analysis, the number of the sampling samples is more than 4/3 times of the number of the uncertain parameters.
Sensitivity analysis screening key parameters: and carrying out sensitivity analysis according to the determined key input parameters. And selecting the spearman coefficient as a correlation coefficient, further screening out key parameters in the correlation coefficient, and reducing the number of uncertain parameters in subsequent analysis. The method comprises the following specific processes: carrying out sensitivity analysis on each input parameter and each output parameter, and calculating a spearman coefficient of each input parameter and each output parameter; the strong correlation is obtained when the absolute value of the spearman coefficient is equal to or greater than 0.7 and equal to or less than 1, the moderate correlation is obtained when the absolute value of the spearman coefficient is equal to or greater than 0.3 and less than 0.7, and the weak correlation is obtained when the absolute value of the spearman coefficient is equal to or greater than 0 and less than 0.3. The medium and strong correlation input parameters were screened for further analysis by the sperman coefficient calculation.
Screening a key sequence by a limit surface method: the traditional RISMC method carries out BEPU analysis on all sequences, and the number of calculation cases is huge. Therefore, in order to reduce the number of calculation branches, when the critical sequences are screened by the limit surface method, only the BEPU analysis is carried out on the critical sequences, so that the calculation amount can be effectively reduced, and the calculation efficiency is improved. All sequence nominal values need to be calculated before screening. On the basis, all sequences are divided into three types of sequences by using a limit surface method, wherein the three types of sequences are respectively as follows: determining the overrun probability. Probability overrun sequence. And determining a safety sequence. And (4) carrying out further BEPU analysis on the probability overrun sequence by screening to solve the conditional failure probability. The extreme surface method is to perform deep analysis on the key input parameters screened by the sensitivity analysis. Different combinations of each group of key parameters are calculated in an optimization algorithm mode, and the optimal or worst nuclear power plant state is found. Example (c): each group of input parameters is combined by 6 input parameters, and the parameters have mutual influence, so that the optimal or worst state of the nuclear power plant is determined by comprehensively considering the random combination of the 6 parameters. The exhaustion method wastes a large amount of computing resources and time, so the state capture is carried out by adopting the idea of an optimization algorithm.
Referring to fig. 5, a branch screening process is disclosed, in the conventional analysis method, the total number of branches depends on the number of event headers and the branch rule, and the invention adopts a polymorphic branch rule, that is, branches are performed according to the number of special safety facilities which can be put into operation. Suppose that each head event has m branches, the number of the head events is n, and the total number of the branches is mnAnd (4) respectively. When the event tree is complex and the number of header events is large, the number of branches is large, and a large amount of computing time and computing resources are occupied. Therefore, there is a need to screen sequences to reduce the number of branches for further analysis.
After all branch sequences are determined, the nominal values of all sequences are calculated, and a critical sequence is screened by using a limit surface method. All sequences can be divided into three categories, as shown with particular reference to FIG. 6:
1) determining a failure sequence. There are two types of sequences that can be considered as sequence 1: the first is that the cladding temperature is raised continuously during the calculation until the limit is exceeded. The second is that PCT is above the limit during the calculation even when the nuclear plant is in an optimal state.
2) And (4) probability overrun sequence. A peak occurs during the simulation of such a sequence and is relatively close to the safety limit. It cannot be determined by simple analysis whether the safety limit has to be exceeded.
3) A security sequence is determined. There are two types of sequences that can be considered as 3: the first is that there is no significant peak throughout the simulation and the cladding temperature is far from the safety limit. The second is that PCT is below the limit even when the nuclear plant is in the worst state during the calculation.
In one embodiment, the overall probability set value in the probability calculation step is 99.9%.
In one embodiment, the PSM solving step includes:
and (3) screening out a key sequence by integrating the determinacy screening result and the probability screening result: and the determinacy screening is to classify accident sequences by a limit surface method, only carry out further uncertainty analysis on probability overrun sequences in the accident sequences, and not carry out further analysis on the other two sequences. And the probability theory screening is to perform descending order sorting on the occurrence probability of all the sequences, and the sequences with the overall probability of 99.9 percent are taken as analysis objects from the large probability to the small probability.
BEPU analysis was performed on each sequence by small sample analysis: the conventional PSM solving method uses the monte carlo Method (MC) to solve the load distribution curve of each sequence. However, solving the load distribution curve of each sequence by using the MC method requires hundreds of calculations, which wastes a large amount of computation time and resources. A small sample analysis method is adopted to replace the traditional MC analysis method. The calculation amount of hundreds of groups per sequence is reduced to tens of groups. And then, the small sample analysis sampling process is optimized through an efficient sampling algorithm, the calculated amount of dozens of groups is further reduced, and the overall analysis efficiency is improved.
The RELAP example was calculated by reduced order model simulation: in the conventional PSM solution, a single example calculation of a single sequence is performed by the system simulator RELAP 5. The 5000s simulation calculation with the device i7-7700K CPU 4.2GHz RAM 32G requires 6min50 s. And single calculation can be controlled within dozens of seconds by utilizing the reduced-order model simulation, so that the calculation efficiency is greatly improved.
Core Damage Frequency (CDF) and PSM solution: and rapidly solving the conditional overrun probability of each key sequence by using a small sample analysis method and a reduced order model, solving the CDF by integrating the conditional overrun probabilities of each sequence, and drawing a load probability distribution curve according to the solved CDF. At present, when a safety limit is used as a load curve, the solved CDF is the PSM.
For loss of coolant accidents, for example, the safety state of a nuclear power plant is primarily characterized by the integrity of the fuel clad. The actual bearing capacity curve of the nuclear power plant is the cladding melting temperature distribution. The operation unit, the supervision unit and the historical operation data of the nuclear power plant need to be comprehensively analyzed, so that the solving difficulty is too high. Therefore, a simplified method commonly used in the RISMC analysis procedure is used: safety limits or a triangular distribution replace the nuclear power plant capacity distribution curve.
The invention realizes the reduction of the number of single-branch calculation:
in the PSM solving process of the traditional RISMC method, a load distribution curve is obtained by adopting a traditional Monte Carlo method. This approach requires hundreds of sets of calculations per event branch. In order to reduce the number of single-branch calculation times required by obtaining the load curve, the invention adopts an efficient sampling method and combines a small sample analysis method to obtain the high-reliability load curve. Firstly, a chi-square fitting method and a fitting conversion method are adopted as the fitting method of the small sample, wherein the chi-square fitting method is adopted when the data type conforms to normal distribution, and the fitting conversion method is adopted when the data type does not conform to the normal distribution. The number of single-branch calculations can be reduced from several hundred sets to 50 sets by using a small sample analysis method. Secondly, the conventional small sample fitting method adopts the traditional MC sampling method in the sampling process, so that the sampling efficiency is low. In order to further reduce the number of single-branch calculation, the method adopts a deterministic sampling and self-adaptive sampling method to replace the traditional MC method, thereby improving the sampling calculation efficiency.
By adopting the efficient sampling method and combining the small sample analysis method, the calculation quantity of a single sequence can be reduced from hundreds of groups to dozens of groups or even a plurality of groups, the calculation quantity is greatly reduced, and the analysis efficiency is effectively improved.
The invention improves the simulation calculation efficiency of a single case:
the computational efficiency of a single example depends primarily on the computer configuration and system simulation software. Currently, the system emulation software RELAP5 widely used in RISMC analysis is mature commercial software, but the modification difficulty of the code is large. Therefore, the invention adopts a machine learning method, and improves the calculation efficiency while ensuring the calculation accuracy by establishing the reduced-order model. The invention adopts a Singular Value Decomposition (SVD) method to establish a reduced order model, establishes a data matrix according to input and output data of a system simulation program, and screens main vectors in the matrix for data fitting. The method can consider the change of the parameters along with the time, and meets the requirement of transient data fitting. On the basis, in order to improve the modeling efficiency, a reduced-order model is quickly established by adopting a method of combining adaptive sampling and SVD (singular value decomposition). FIG. 7 illustrates the entire reduced order model modeling process.
By combining the SVD method and the self-adaptive method, the reduced-order model is quickly trained, the simulation calculation time of a single system is reduced from several minutes to dozens of seconds, and the analysis efficiency is effectively improved.
Referring to fig. 8, the present invention also discloses a nuclear reactor efficient probabilistic safety margin analysis system, including: the device comprises a data acquisition module, a model establishing and analyzing module, a probability calculation module and a PSM solving module;
the data acquisition module is used for acquiring target parameters, modeling requirements and accident sequences of the research object;
the input end of the model building and analyzing module, the input end of the probability calculating module and the output end of the data acquiring module share an endpoint;
the model establishing and analyzing module is used for performing key sequence screening and BEPU analysis after the deterministic theory modeling;
the probability calculation module is used for establishing a probability safety analysis model by adopting a multi-branch event tree/fault tree model, performing sequence probability calculation and sequencing, wherein a sequence containing the overall probability value as a set value is a PSM analysis object;
the input end of the PSM solving module, the output end of the model establishing and analyzing module and the output end of the probability calculating module share a terminal point;
the PSM solving module is used for calculating the PSM according to the PSM analysis object through a small sample combined with an efficient sampling analysis method and a reduced-order model method.
In one particular embodiment, a terminal includes: the nuclear reactor safety margin analysis system includes at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform a nuclear reactor efficient probabilistic safety margin analysis method.
In a particular embodiment, a computer-readable storage medium has computer instructions stored thereon for causing a computer to perform a nuclear reactor efficient probabilistic safety margin analysis method.
In another embodiment, the analysis method is used for carrying out corresponding research on small-break water loss accidents of the Qinshan nuclear power plant. Since nuclear power plant loss of coolant accidents are studied, the Peak Cladding Temperature (PCT) was chosen as the key safety parameter. The cladding failure is considered when the cladding peak temperature exceeds 1204 ℃.
The safety injection system and the spraying system are mainly used as specially-designed safety facilities related to small-break water loss accidents. The safety injection system comprises a high-pressure safety system, a safety injection box system and a low-pressure safety injection system. The spraying system is a containment spraying system. Therefore, the modeling of the determinism RELAP5 is carried out by including the above-mentioned several special security facilities in addition to the basic primary and secondary loop systems. The main special safety facilities are made clear, unnecessary system modeling is avoided, and the modeling complexity is reduced.
The accident sequence of the system obtained by modeling according to the determinism is mainly as follows: (1) the method comprises the steps of (1) breaking a main coolant cooling pipe section of a nuclear power plant, (2) triggering shutdown signals of a reactor to stop the reactor, (3) stopping the main pump, (4) isolating main steam, (5) isolating a main water supply system, (6) starting a high-pressure safety injection system, (7) starting a safety injection box system, (8) starting a low-pressure safety injection system.
And (3) simulating the standard parameters and the multiple groups of input parameters by utilizing modeling of a determinism RELAP5, and determining the accuracy and stability of the model. The error between the standard parameter simulation value of the RELAP5 model and the key parameter of the operation nominal value of the nuclear power plant is kept within 1 percent, and the requirement is met. Correct results can be obtained by using the RELAP5 model to carry out multiple sets of input parameter simulation, and the stability of the model is verified.
The method comprises the following steps of designing a nuclear power plant, researching a PIRT file and a reference document, and judging experience when data is lacked. The main key input parameters in this incident are: the method comprises the following steps of initial power of a reactor core, initial flow of main coolant, initial temperature of the main coolant, initial pressure of the main coolant, high-pressure safety injection flow, low-pressure safety injection flow, initial flow of a safety injection tank, initial pressure of the safety injection tank, initial temperature of the safety injection tank, friction coefficient of a safety injection pipeline of the safety injection tank, friction coefficient of a thermal channel of the reactor core, crevasse area, crevasse friction coefficient, thermal conductivity of fuel cladding and thermal capacity of the fuel cladding.
The traditional sampling method, namely the Monte Carlo sampling method, has high requirements on the number of calculated samples, and in order to reduce the number of sampled samples and improve the analysis efficiency, the Latin hypercube sampling method is adopted for sampling. When the Latin hypercube sampling is used for sensitivity analysis, the number of the sampling samples is more than 4/3 times of the number of the uncertain parameters. For 15 key input parameters, theoretically at least 20 samples are needed for sensitivity analysis.
The sperman coefficient calculation was screened for moderately relevant and strongly relevant input parameters for further analysis by sensitivity analysis.
Aiming at small-break water loss accidents of the Qinshan nuclear power plant, all sequences are divided into three types of sequences, which are respectively as follows: firstly, determining an overrun sequence, secondly, determining a probability overrun sequence, and thirdly, determining a safety sequence. When all the special safety facilities cannot normally operate, the peak temperature of the cladding after the occurrence of the breach is always increased suddenly and finally exceeds the safety limit value, and the sequence is a determined overrun sequence because no relief measures are taken after the occurrence of the accident. When all the specially-arranged safety facilities normally operate, the safety injection flow is sufficient and timely, the temperature rise of the cladding can be well relieved, and the temperature of the cladding is not obviously increased in the whole accident process, so that the sequence is a determined safety sequence. For the probability overrun sequence, due to the fact that the operation state of the nuclear power plant is complex, complex correlation exists between key parameters, and the correlation of each parameter is difficult to consider in sampling. Therefore, an optimization algorithm, namely a genetic algorithm, is adopted to calculate the probability overrun sequence and screen out the optimal/worst state.
The probability calculation for each incident sequence is solved by building an event tree/fault tree analysis model and combining the relevant reliability data. The event with the head of the event tree is a high-pressure safety injection system, a safety injection box safety injection system and a low-pressure safety injection system. The Qinshan nuclear power plant is a two-loop pressurized water reactor, so that each safety injection system has two sets of 1 and 2 loops. There are four branches at each topic event: (1) the method comprises the following steps of (1) normally operating two rows 1 and 2, (2) normally operating the two rows 1, (3) normally operating the two rows 2, (4) abnormally operating the two rows 1 and 2. Thus there are 64 accident sequences in the case.
And (4) screening out the key sequence by combining the deterministic theory screening result and the probabilistic theory screening result. And (3) determining theoretical screening, namely screening probability overrun sequences in the small-break water loss accidents in the Qinshan mountains, wherein 10 probability overrun sequences are provided in total. A total of 4 sequences containing an overall probability of 99.9% were screened by probability theory. A total of 4 sequences were screened for further BEPU analysis.
BEPU analysis was performed on each sequence by a small sample analysis method. The conventional PSM solving method uses the monte carlo Method (MC) to solve the load distribution curve of each sequence. However, solving the load distribution curve of each sequence by using the MC method requires hundreds of sets of calculation examples, which wastes a large amount of calculation time and resources. In the scheme, the load curve can be solved by adopting a fitting conversion method and a chi-square fitting analysis method and utilizing 50 groups of samples.
A reduced-order model is established through a singular value decomposition method and a self-adaptive sampling method, the calculation process is simplified through screening key vectors in a matrix, and the calculation efficiency is improved.
The method reduces the original 64 sequence analyses to 4 sequences, reduces the analysis times of BEPU of each sequence from hundreds to 50, and reduces the calculation time of a single set of calculation example from about 7 minutes to tens of seconds. The overall analysis efficiency is greatly improved.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention in a progressive manner. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. An efficient probabilistic safety margin analysis method for a nuclear reactor, comprising the steps of:
a data acquisition step: acquiring target parameters, modeling requirements and accident sequences of a research object;
model building and analyzing steps: performing key sequence screening and BEPU analysis after the deterministic theory modeling;
and a probability calculation step: performing probability calculation and sequencing on the accident sequences, wherein the sequence containing the overall probability value as a set value is a PSM analysis object;
PSM solving step: and calculating the PSM according to the PSM analysis object by combining a small sample with an efficient sampling analysis method and a reduced-order model method.
2. The nuclear reactor-efficient probabilistic safety margin analysis method of claim 1,
in the data acquisition step, target parameters of a research object are parameters representing the safety state of the nuclear power plant corresponding to a specific nuclear power plant and an accident type;
the modeling requirement is specifically that the special safety facilities in the accident process are modeled according to the accident process;
the accident sequence is specifically determined according to an accident process and a key phenomenon, and comprises the following steps: determining a failure sequence, a probability overrun sequence and a safety sequence.
3. The nuclear reactor-efficient probabilistic safety margin analysis method of claim 1,
the specific contents of the model establishing and analyzing steps comprise: completing determinism modeling and verifying model accuracy; determining uncertainty parameters in the BEPU analysis process; determining parameter sampling methods and numbers; analyzing and screening key parameters of sensitivity; and (4) screening a key sequence by a limit surface method.
4. The nuclear reactor-efficient probabilistic safety margin analysis method of claim 1,
the overall probability set value in the probability calculation step is 99.9%.
5. The nuclear reactor-efficient probabilistic safety margin analysis method of claim 1,
the PSM solving step specifically comprises the following steps: screening out a key sequence by integrating the deterministic theory screening result and the probabilistic theory screening result; performing BEPU analysis on each sequence by a small sample analysis method; calculating RELAP example by a reduced order model; core damage frequency and PSM solution.
6. A nuclear reactor-efficient probabilistic safety margin analysis system for implementing a nuclear reactor-efficient probabilistic safety margin analysis method according to any one of claims 1 to 5, comprising: the device comprises a data acquisition module, a model establishing and analyzing module, a probability calculation module and a PSM solving module;
the data acquisition module is used for acquiring target parameters, modeling requirements and accident sequences of the research object;
the input end of the model building and analyzing module, the input end of the probability calculating module and the output end of the data acquiring module share an endpoint;
the model establishing and analyzing module is used for performing key sequence screening and BEPU analysis after the deterministic theory modeling;
the probability calculation module is used for establishing a probability safety analysis model by adopting a multi-branch event tree/fault tree model, performing sequence probability calculation and sequencing, wherein a sequence containing the overall probability value as a set value is a PSM analysis object;
the input end of the PSM solving module, the output end of the model establishing and analyzing module and the output end of the probability calculating module share a terminal point;
the PSM solving module is used for calculating the PSM according to the PSM analysis object through a small sample combined with an efficient sampling analysis method and a reduced-order model method.
7. A terminal, characterized in that,
the method comprises the following steps: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the nuclear reactor efficient probabilistic safety margin analysis method of any of claims 1 to 5.
8. A computer-readable storage medium, characterized in that,
the computer readable storage medium stores computer instructions for causing a computer to perform the nuclear reactor efficient probabilistic safety margin analysis method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111063589.6A CN113761749B (en) | 2021-09-10 | 2021-09-10 | Nuclear reactor probability safety margin analysis method, system, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111063589.6A CN113761749B (en) | 2021-09-10 | 2021-09-10 | Nuclear reactor probability safety margin analysis method, system, terminal and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113761749A CN113761749A (en) | 2021-12-07 |
| CN113761749B true CN113761749B (en) | 2022-03-22 |
Family
ID=78794824
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111063589.6A Expired - Fee Related CN113761749B (en) | 2021-09-10 | 2021-09-10 | Nuclear reactor probability safety margin analysis method, system, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113761749B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114970324A (en) * | 2022-04-29 | 2022-08-30 | 哈尔滨工程大学 | Accident sequence screening method based on combination of FCNN and PSO |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103548093A (en) * | 2010-11-23 | 2014-01-29 | 西屋电气有限责任公司 | Full spectrum LOCA evaluation model and analysis methodology |
| CN106548279A (en) * | 2016-11-02 | 2017-03-29 | 中广核研究院有限公司 | A kind of safety analytical method for the advanced reactor master-plan stage |
| CN107644694A (en) * | 2017-09-20 | 2018-01-30 | 岭东核电有限公司 | A kind of nuclear power plant's large break crash analysis method |
| CN107767025A (en) * | 2017-09-20 | 2018-03-06 | 中广核研究院有限公司 | A kind of nuclear power plant Large break LOCA method of risk-informed |
| CN111967130A (en) * | 2020-07-07 | 2020-11-20 | 中国核电工程有限公司 | Analysis method for supercooling margin fixed value under accident condition of pressurized water reactor nuclear power plant |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100681487B1 (en) * | 2002-10-30 | 2007-02-09 | 한국전력공사 | Best Estimated Evaluation System for Safety Analysis at Nuclear Power Plant |
-
2021
- 2021-09-10 CN CN202111063589.6A patent/CN113761749B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103548093A (en) * | 2010-11-23 | 2014-01-29 | 西屋电气有限责任公司 | Full spectrum LOCA evaluation model and analysis methodology |
| CN106548279A (en) * | 2016-11-02 | 2017-03-29 | 中广核研究院有限公司 | A kind of safety analytical method for the advanced reactor master-plan stage |
| CN107644694A (en) * | 2017-09-20 | 2018-01-30 | 岭东核电有限公司 | A kind of nuclear power plant's large break crash analysis method |
| CN107767025A (en) * | 2017-09-20 | 2018-03-06 | 中广核研究院有限公司 | A kind of nuclear power plant Large break LOCA method of risk-informed |
| CN111967130A (en) * | 2020-07-07 | 2020-11-20 | 中国核电工程有限公司 | Analysis method for supercooling margin fixed value under accident condition of pressurized water reactor nuclear power plant |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113761749A (en) | 2021-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9208911B2 (en) | Full spectrum LOCA evaluation model and analysis methodology | |
| KR100681487B1 (en) | Best Estimated Evaluation System for Safety Analysis at Nuclear Power Plant | |
| Kang et al. | A combined deterministic and probabilistic procedure for safety assessment of beyond design basis accidents in nuclear power plant: application to ECCS performance assessment for design basis LOCA redefinition | |
| CN113761749B (en) | Nuclear reactor probability safety margin analysis method, system, terminal and storage medium | |
| Gonzalez-Cadelo et al. | Analysis of cold leg LOCA with failed HPSI by means of integrated safety assessment methodology | |
| Vorobyev et al. | Development and application of a genetic algorithm based dynamic pra methodology to plant vulnerability search | |
| Sun et al. | An advanced probability safety margin analysis approach combined deterministic and probabilistic safety assessment | |
| Pourgol-Mohamad et al. | Integrated methodology for thermal-hydraulic code uncertainty analysis with application | |
| Kim et al. | Quantitative evaluation of change in core damage frequency by postulated power uprate: Medium-break loss-of-coolant-accidents | |
| CN110021933B (en) | Power information system control function reliability assessment method considering component faults | |
| Zio et al. | Bootstrap and Order Statistics for Quantifying Thermal‐Hydraulic Code Uncertainties in the Estimation of Safety Margins | |
| Lee et al. | Application of dynamic probabilistic safety assessment approach for accident sequence precursor analysis: case study for steam generator tube rupture | |
| Ma et al. | A probabilistic cascading failure model for dynamic operating conditions | |
| Mandelli et al. | Improved Sampling Algorithms in the Risk-Informed Safety Margin Characterization Toolkit | |
| Johst et al. | Extension of a Level 2 PSA event tree based on results of a probabilistic dynamic safety analysis of induced Steam Generator Tube Rupture | |
| Groudev et al. | Validation of a VVER 1000 Trace Model | |
| Galushin et al. | An approach to grouping and classification of scenarios in integrated deterministic-probabilistic safety analysis | |
| De Caro et al. | Review of Recent Trends in Power System Resilience-Oriented Decision-Making Methods | |
| Yue et al. | Development of a risk-informed decision-making capability using standard electric power industry planning tools | |
| Xu et al. | Risk-Informed Safety Margin Evaluation Approach of Nuclear Power Plant Integrated with Adaptive Dynamic Event Tree Algorithm | |
| Martorell et al. | Safety margins estimation method considering uncertainties within risk-informed decision making framework | |
| Yu et al. | Safety margin quantification by integrating probabilistic and deterministic safety assessments—Application to design extension conditions | |
| Sun et al. | Extreme risk assessment in power system considering cyber attacks | |
| Xu et al. | RISK-INFORMED SYSTEMS ANALYSIS OPTIMIZATION WITH ADAPTIVE DYNAMIC EVENT TREE AND SUPPORT VECTOR MACHINE | |
| Sun et al. | Research on simplification of branches method of accident sequences based on expert knowledge and heuristic optimization algorithm |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220322 |