Analysis method for radioactive source item and consequence of serious accident of pressurized water reactor, and parameter optimization method and device of nuclear reactor

Abstract

The invention discloses a pressurized water reactor serious accident radioactive source item and consequence analysis method and a nuclear reactor parameter optimization method and device.The analysis method comprises the steps that modeling is conducted on a reactor through a target analysis program, benchmark working condition calculation is conducted, and an initial source item result and an initial consequence result are obtained; screening the initial input parameters according to the initial source item result and the initial consequence result to obtain a source item uncertainty parameter set and a consequence uncertainty parameter set; according to the uncertainty parameter set, utilizing a target analysis program to obtain a source item analysis statistical result of the source item uncertainty parameter and a radioactive consequence statistical result of the consequence uncertainty parameter; and on the basis of the source item analysis statistical result, the radioactive consequence statistical result, the source item uncertainty parameter set and the consequence uncertainty parameter set, performing sensitivity analysis on each uncertainty parameter to obtain a sensitivity analysis result of the uncertainty parameters. Parameter uncertainty of source items and consequences can be quantified.

Description

Technical Field

[0001] The present invention belongs to the technical field of reactor severe accident analysis, and in particular relates to a method for analyzing radioactive source items and consequences of severe accidents in pressurized water reactors, and a method and device for optimizing parameters of nuclear reactors. Background Art

[0002] Analysis of severe accident radioactive source terms is an important input for nuclear power plants to conduct on-site and off-site radioactive consequence assessments. In engineering, integrated severe accident analysis procedures or mechanism models are generally used to carry out computational analysis of severe accident source terms and consequences of nuclear power plants.

[0003] However, due to the great uncertainty of severe accident phenomena in nuclear power plants and the release, migration, and deposition of radioactive materials, and the fact that severe accidents involve multiple disciplines including probability theory, neutron physics, thermal hydraulics, high-temperature materials science, chemistry, and structural mechanics, our current understanding of the mechanisms of some key phenomena is still insufficient, and the models of existing representative pressurized water reactor severe accident analysis software have great uncertainty.

[0004] Therefore, the existing technology does not adequately study the uncertainty of parameters in the radioactive source terms and consequence analysis methods for severe accidents, resulting in greater uncertainty in the analysis results. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors, a method and device for optimizing parameters of nuclear reactors in view of the above-mentioned deficiencies in the prior art. The method can be used to quantify the parameter uncertainty and sensitivity of radioactive source terms and consequences after nuclear reactor accidents.

[0006] In a first aspect, an embodiment of the present invention provides a method for analyzing radioactive source terms and consequences of a severe accident in a pressurized water reactor, characterized in that the method comprises:

[0007] Modeling the reactor and performing baseline calculations to obtain initial source term results and initial consequence results through a target analysis program, wherein the target analysis program includes an integrated severe accident analysis program and a consequence mechanism program corresponding to the reactor type and nuclear accident type;

[0008] Screening initial input parameters according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set, wherein the uncertainty parameters include parameters of the source term analysis results or consequence results that affect the target analysis program, and the uncertainty parameter set includes a range and probability distribution of each uncertainty parameter;

[0009] According to the source term uncertainty parameter set and consequence uncertainty parameter set, using the target analysis program to obtain source term analysis statistical results of the source term uncertainty parameters and radioactive consequence statistical results of the consequence uncertainty parameters;

[0010] Based on the statistical results of the source term analysis, the statistical results of the radioactive consequences, the source term uncertainty parameter set and the consequence uncertainty parameter set, a sensitivity analysis is performed on each uncertainty parameter to obtain the sensitivity analysis results of the uncertainty parameters.

[0011] In a second aspect, an embodiment of the present invention provides a method for optimizing parameters of a nuclear reactor, characterized in that the method comprises:

[0012] Determining target adjustment parameters of the nuclear reactor according to the sensitivity analysis results of each uncertainty parameter, wherein the target adjustment parameters are used to guide the optimization direction of severe accident source items and consequence analysis, including parameters such as model settings and design guidance values;

[0013] The target adjustment parameters of the nuclear reactor are adjusted.

[0014] In a third aspect, an embodiment of the present invention further provides an analysis device for radioactive source terms and consequences of a severe accident in a pressurized water reactor, comprising:

[0015] A first calculation module is used to model the reactor and perform baseline condition calculations to obtain initial source term results and initial consequence results through a target analysis program, wherein the target analysis program includes an integrated severe accident analysis program and a consequence mechanism program corresponding to the reactor type and the nuclear accident type;

[0016] A second calculation module, connected to the first calculation module, is used to screen the initial input parameters according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set, wherein the uncertainty parameters include parameters of the source term analysis results or consequence results that affect the target analysis program, and the uncertainty parameter set includes a range and probability distribution of each uncertainty parameter;

[0017] A third calculation module, connected to the second calculation module, is used to obtain source term analysis statistics of source term uncertainty parameters and radioactive consequence statistics of consequence uncertainty parameters using the target analysis program according to the source term uncertainty parameter set and consequence uncertainty parameter set;

[0018] The analysis module is connected to the third calculation module and is used to perform sensitivity analysis on each uncertainty parameter based on the source term analysis statistical results and the radioactive consequence statistical results and the source term uncertainty parameter set and the consequence uncertainty parameter set to obtain the sensitivity analysis results of the uncertainty parameters.

[0019] In a fourth aspect, an embodiment of the present invention further provides a parameter optimization device for a nuclear reactor, comprising:

[0020] The analysis device for radioactive source terms and consequences of severe accidents in pressurized water reactors described in the third aspect is used to obtain sensitivity analysis results of various uncertainty parameters of the nuclear reactor through the analysis method of radioactive source terms and consequences of severe accidents in pressurized water reactors;

[0021] A determination module is connected to the nuclear reactor severe accident radioactive source term and consequence analysis device, and is used to determine the target adjustment parameters of the nuclear reactor according to the sensitivity analysis results of each uncertainty parameter. The target adjustment parameters are used to guide the optimization direction of severe accident source term and consequence analysis, including parameters such as model setting and design guidance values; an adjustment module is connected to the determination module, and is used to adjust the target adjustment parameters of the nuclear reactor.

[0022] The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors of the present invention comprehensively considers various uncertain factors in the nuclear accident process, conducts a full-process analysis of radioactive source terms and consequences, and screens out uncertainty parameters; and then quantifies the sensitivity of the uncertainty parameters, providing a valuable reference for severe accident management and safety improvement of nuclear power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 :A flow chart of a method for analyzing radioactive source terms and consequences of a severe accident in a pressurized water reactor provided by an embodiment of the present invention;

[0024] Figure 2 :A flow chart of a method for analyzing uncertainty and sensitivity of radioactive source terms and consequences of severe accidents provided by an embodiment of the present invention;

[0025] Figure 3 : A schematic diagram of the sampling rules of the Morris method provided in an embodiment of the present invention;

[0026] Figure 4 : A schematic diagram of a loop flow node provided in an embodiment of the present invention;

[0027] Figure 5 : A schematic diagram of a primary circuit water node provided in an embodiment of the present invention;

[0028] Figure 6 : A schematic diagram of the containment control body division provided in an embodiment of the present invention;

[0029] Figure 7 :The release fraction of inert gas provided by the embodiment of the present invention (LLOCA)

[0030] Schematic diagram;

[0031] Figure 8 : A schematic diagram of the release fraction (LLOCA) of CsI provided in an embodiment of the present invention;

[0032] Fig. 9 : A structural diagram of a device for analyzing radioactive source terms and consequences of a severe accident in a pressurized water reactor provided by an embodiment of the present invention;

[0033] Fig.10 : A structural diagram of a nuclear reactor parameter optimization device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0035] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0036] Embodiment 1:

[0037] In order to solve the existing technical problems, such as Figure 1 As shown, this embodiment provides a method for analyzing radioactive source terms and consequences of a severe accident in a pressurized water reactor, including the following steps 101 to 104.

[0038] Step 101, through the target analysis program, the reactor is modeled and the baseline operating condition calculation is performed to obtain the initial source term results and the initial consequence results.

[0039] Among them, the target analysis program is the integrated severe accident analysis program and consequence mechanism program corresponding to the reactor type and nuclear accident type. The integrated severe accident analysis program is used for dynamic simulation of the development from accident conditions to radioactive release after a nuclear accident occurs (i.e., used to obtain source term results); the consequence mechanism program simulates the environmental diffusion and consequence assessment after the source term is released (i.e., used to obtain consequence results).

[0040] Baseline conditions: basic operating events or assumed accident conditions, such as loss of coolant accidents, plant power outage accidents, etc., are calculated under standard initial conditions and default model parameter values ​​to obtain initial analysis results. Initial source term results: radioactive release source terms calculated under baseline conditions, such as the share of inert gas released into the environment. Initial consequence results: the consequences of source terms on the external environment after they are released into the environment, such as dose assessment or public risk.

[0041] Specifically, we first establish corresponding analysis models based on specific reactor types (such as second-generation and third-generation pressurized water reactors) and nuclear accident types (such as loss of coolant accidents, plant power outages, etc.). Under preset accident scenarios (such as randomized parameter changes are not considered), we calculate the dynamic response of the reactor system, and use the severe accident analysis program and consequence mechanism program to obtain the following: source term analysis results (environmental source terms) and consequence analysis results (radioactivity diffusion and dose included in the environment).

[0042] In this step, initial source and consequence results can be obtained as the basis for subsequent uncertainty analysis.

[0043] Step 102, screening the initial input parameters according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set.

[0044] Among them, the uncertainty parameters include the parameters that affect the source term analysis results or consequence results of the target analysis program. For example, all initial input parameters that have a significant impact on the source term or radioactive consequences, such as the in-core damage process (cladding failure temperature, etc.), the core debris collapse model (core support plate failure temperature, etc.) or the fission product release and migration model (decontamination factor, fission product release rate), etc.

[0045] The uncertainty parameter set includes the range and probability distribution of each uncertainty parameter. Each uncertainty parameter needs to be given a range (such as temperature 500-700℃) and a probability distribution (such as uniform distribution, normal distribution) in order to perform sampling analysis on it.

[0046] Specifically, preliminary parameter screening is carried out based on the known model parameters of the traditional process; analysis is carried out on the parameters that affect the initial source results and initial consequence results, and an uncertainty parameter set is constructed; and the variable range and probability distribution are defined for each parameter (based on experimental data, standards, user manuals or scientific research).

[0047] Step 103, according to the source term uncertainty parameter set and the consequence uncertainty parameter set, using the target analysis program to obtain the source term analysis statistical results of the source term uncertainty parameters and the radioactive consequence statistical results of the consequence uncertainty parameters.

[0048] The target analysis program is used to simulate the source term and consequence to obtain the analysis statistical results. For example, based on the source term and consequence uncertainty parameter set, these parameters can be randomly sampled to construct an input matrix; the input matrix is ​​passed to the target analysis program to perform: (1) severe accident source term program calculation to obtain the statistical distribution of the source term results; (2) radioactive consequence mechanism program calculation to obtain the statistical distribution of the consequence results (such as the dose distribution after diffusion).

[0049] Step 104, based on the statistical results of the source term analysis, the statistical results of the radioactive consequences, the source term uncertainty parameter set and the consequence uncertainty parameter set, a sensitivity analysis is performed on each uncertainty parameter to obtain a sensitivity analysis result of the uncertainty parameter.

[0050] Among them, sensitivity analysis: studies the impact of each input parameter on the statistical distribution of the results. Common methods include global sensitivity analysis (such as Morris method) or local sensitivity analysis. Sensitivity analysis results: assign a quantitative sensitivity coefficient (such as impact weight or regression coefficient) to each parameter to evaluate the importance of each parameter.

[0051] Specifically, based on the statistical results of source terms and consequences, a sensitivity analysis of uncertainty parameters is conducted to identify the impact of key parameters.

[0052] In this embodiment, a comprehensive and accurate assessment of the radioactive source term and consequences after a nuclear accident is achieved through modeling, baseline condition calculation, source term and consequence uncertainty analysis, and sensitivity analysis. The introduction of stable statistical methods for source terms and consequences expands the prediction capability of severe accident analysis from a single value to a distribution range, which can more fully estimate the risk of accidents; through sensitivity analysis, the key parameters that affect the results of accidents are accurately located, providing directional basis for subsequent reactor design adjustments (such as safety design optimization or accident mitigation measures). Through multiple calculations and statistical analysis, the uncertainty of each parameter and its impact on the radioactive source term and consequences of the accident can be effectively captured, which improves the reliability of risk assessment and the scientific nature of decision support.

[0053] Optionally, the above step 102 may specifically include:

[0054] Determine initial screening parameters based on reference data, where the reference data includes at least one of experimental data, standards, user manuals, research literature, and expert experience;

[0055] According to the initial source item results, the Morris method is used to perform global sensitivity analysis on each initial screening parameter to determine the source item sensitivity value of each initial screening parameter to the source item results. The source item sensitivity value is used to characterize the degree of influence of the initial screening parameter on the result.

[0056] According to the sensitive value of each initial screening parameter, the initial screening parameters are screened for the second time to obtain the source uncertainty parameters;

[0057] According to the properties of source term uncertainty parameters and reference data, the range and probability distribution of each source term uncertainty parameter are determined to obtain a source term uncertainty parameter set.

[0058] Here, the initial screening parameters refer to the various variables involved in the input parameters in the radioactive source term analysis or consequence analysis of nuclear reactors (such as the cladding failure temperature, core support plate failure temperature, decontamination factor, fission product release rate, etc. in the source term analysis). In order to ensure the efficiency and accuracy of the analysis, it is first necessary to screen out the initial parameters that may have a greater impact on the source term results or consequence results from all the parameters, which are called initial screening parameters.

[0059] Experimental data include measured analysis data from historical experimental studies; standards are design parameters in nuclear safety regulations or generally agreed upon field consensus; user manuals are user manuals for integrated accident analysis procedures; research literature includes parameter knowledge proposed in academic research or technical research; and expert experience includes empirical parameters given based on the knowledge accumulation of field experts.

[0060] The Morris method is a global sensitivity analysis method used for multi-parameter screening. The Morris method is used to calculate the sensitivity value corresponding to each initial screening parameter. The higher the value, the greater its influence on the source term result. The parameter properties can be the physical and mathematical properties of each parameter, for example: whether the parameter is discrete (such as the device state is on / off) or continuous.

[0061] Specifically, first, all possible input parameters (such as reactor physics, thermal hydraulics, material behavior, etc.) are identified, and an initial parameter list is generated by screening reference data; then, each parameter is perturbed and the source term change is calculated using the severe accident analysis program, the parameter sensitivity value is calculated using the Morris method, and the key parameters that have a greater impact on the source term are screened out based on the threshold; subsequently, the value range of the key parameters is determined based on their physical properties, experimental data or standards, and a suitable probability distribution (such as normal, uniform distribution) is assigned, and a parameter set containing important uncertainty parameters and their probability distribution is systematically constructed, laying the foundation for subsequent sampling and statistical analysis, thereby improving the analysis efficiency and the accuracy of the results.

[0062] In this embodiment, the input parameters of uncertainty analysis are optimized by combining reference data, Morris method and multiple screening steps. First, using Morris global sensitivity analysis, two screenings greatly reduce the number of parameters, avoid redundant calculations, and improve program processing performance; secondly, the combination of reference data and quantitative sensitivity values ​​ensures that the selected parameters have sufficient influence and representativeness on the source results, thereby improving the accuracy of the analysis. Finally, by clarifying the range and distribution of uncertainty parameters, subsequent statistical calculations can fully cover the uncertainty of the parameters.

[0063] Optionally, the above step 102 may further include:

[0064] According to the initial consequence results, a global sensitivity analysis is performed on each initial screening parameter using the Morris method to determine the consequence sensitivity value of each initial screening parameter, wherein the consequence sensitivity value is used to characterize the degree of influence of the initial screening parameter on the consequence results;

[0065] According to the consequence sensitivity value of each of the initial screening parameters, the initial screening parameters are screened for a second time to obtain the consequence uncertainty parameter;

[0066] According to the properties of the consequence uncertainty parameters and the reference data, the range and probability distribution of each consequence uncertainty parameter are determined to obtain the consequence uncertainty parameter set.

[0067] The consequence uncertainty parameters here refer to the input variables that have a significant impact on the consequence results after further analysis from the initially screened variable set. They represent all the core elements that may cause the uncertainty of the consequence expansion. Examples: dry deposition rate, dispersion parameters, etc.

[0068] Specifically, in the first step, the initial screening input parameters are sorted; the Morris method is used to perturb each initial parameter, and the analysis program is introduced to simulate the change of consequence results; the global sensitivity value of each parameter is calculated based on the change of consequence results. In the second step, the parameter sensitivity values ​​output by the Morris method are sorted; the limit thresholds are set for parameters with small impact (low sensitivity values) and removed; variables with higher sensitivity values ​​are retained to form consequence uncertainty parameters. In the third step, for the selected consequence uncertainty parameters, their physical and logical properties are confirmed; the variable range is determined; a suitable probability distribution is assigned to each consequence uncertainty parameter to form a structured consequence uncertainty parameter set, and the range, distribution and value characteristics are fully defined.

[0069] In this embodiment, the consequence sensitivity analysis of the initial screening parameters is combined with the global sensitivity calculation of the Morris method to achieve efficient screening of the most critical consequence input parameters from a large data set, and further clarify the main influencing parameters of the consequences through secondary screening. Finally, through the definition of the range and distribution of uncertainty parameters, an accurate and reliable data foundation is established for the entire consequence uncertainty analysis.

[0070] Optionally, the above step 103 may specifically include the following steps:

[0071] The source uncertainty parameter set and the consequence uncertainty parameter set are sampled respectively to generate a source uncertainty parameter input matrix and a consequence uncertainty parameter input matrix;

[0072] According to the source uncertainty parameter input matrix, batch calculation is performed using the integrated severe accident analysis program to obtain the source analysis statistical results of the uncertainty parameters;

[0073] According to the source term results of the benchmark conditions and the consequence uncertainty parameter input matrix, or the source term analysis statistical results of the uncertainty parameters and the consequence uncertainty parameter input matrix, batch calculations are performed using the radioactive consequence mechanism model to obtain the radioactive consequence statistical results of the uncertainty parameters.

[0074] Here, the source uncertainty parameter set refers to the source uncertainty parameter set that has been screened and has a clear range and probability distribution. The consequence uncertainty parameter set refers to the core parameter group related to consequence analysis with a clear range and probability distribution. The source uncertainty parameter input matrix represents a two-dimensional array of multiple source parameter combinations, where each row is a parameter combination generated by a single sampling, and each column corresponds to a random value of a single input parameter. For example, an input matrix may contain hundreds or thousands of combinations. The consequence uncertainty parameter input matrix means that each row represents a set of randomized consequence uncertainty parameter combinations, and each column is a random value of a specific uncertain parameter.

[0075] Source term analysis statistics of uncertainty parameters: Statistical characteristics of multiple groups of source term calculation results output from the accident analysis program, such as the mean, variance, confidence interval and extreme value of source term release. Radiological consequence statistics: Statistical characteristics of a large number of consequence calculation results output by the consequence mechanism model, which may include the mean and extreme value of dose, etc.

[0076] Specifically, in the first step, sampling is carried out on the source term and consequence uncertainty parameter sets respectively: the range and probability distribution of each parameter are taken; multiple parameter groups are generated according to the required sample size; and they are organized into input matrices: the source term input matrix represents the parameter values ​​corresponding to each source term simulation in rows; the consequence input matrix represents each group of parameter combinations for consequence calculation in rows.

[0077] In the second step, the source term input matrix generated by sampling is input row by row into the severe accident analysis program, and the program is automatically run to complete batch calculations; each set of input corresponds to the simulation results of a source term release, and the program outputs the radioactive release amount and key characteristics under this scenario; all simulation results are summarized to obtain the statistical results of the source term analysis.

[0078] The third step is to input the source term results under the baseline conditions or the statistical results of the source term analysis into the consequence mechanism model as the initial conditions; use the parameter combination of the consequence input matrix to run the consequence mechanism model group by group to complete the multi-scenario simulation; obtain the consequence results corresponding to each group of inputs (such as dose distribution, population health risks, etc.), and summarize the results of all simulation scenarios to obtain the statistical characteristics of radioactive consequences.

[0079] In this embodiment, the uncertainty of source and consequence inputs is systematically described through a sampling matrix and a batch calculation program, and statistical methods are combined to clearly quantify the uncertainty range and impact characteristics. The severe accident analysis program and consequence model are used to achieve efficient multi-scenario calculation (batch calculation), which comprehensively covers all possible scenarios in the accident process without missing any key information. The source analysis results are directly input into the consequence model, establishing a full-link connection from source release to consequence assessment, and building an integrated analysis framework for complex accident analysis.

[0080] Optionally, the above-mentioned sampling of the source uncertainty parameter set and the consequence uncertainty parameter set respectively to generate the source uncertainty parameter input matrix and the consequence uncertainty parameter input matrix may specifically include:

[0081] The source uncertainty parameter set is sampled, and values ​​are taken according to the range and probability distribution of the source uncertainty parameters, and various combinations of uncertainty parameters of the source input parameters are simulated to obtain the input matrix of the source uncertainty parameters;

[0082] The consequence uncertainty parameter set is sampled, and values ​​are taken according to the range and probability distribution of the consequence uncertainty parameters. Various uncertainty parameter combinations of consequence input parameters are simulated to obtain the input matrix of consequence uncertainty parameters.

[0083] Among them, the sampling method may include: Monte Carlo sampling method, Latin hypercube sampling method and other methods, which are not limited in this application.

[0084] Specifically, for the source uncertainty parameter set, clarify each parameter in the source parameter set, determine their value range and the corresponding probability distribution type (such as normal or uniform distribution); determine the sampling method (such as Monte Carlo sampling, Latin hypercube sampling) and sample size according to the calculation requirements; perform multiple random sampling on each parameter to generate possible values ​​of the parameter; reshape the sampled values ​​of all parameters into an input matrix, with each row being a complete parameter combination.

[0085] In this embodiment, by quantitatively describing the uncertainty of source term parameters and consequence parameters, a large number of combined scenarios are sampled to comprehensively cover possible accident situations. The input matrix expresses a large number of uncertainty parameter combinations in the form of a two-dimensional array, which is helpful for data input, management and call. The matrix representation is also easy to directly access subsequent calculation programs (such as severe accident analysis programs or consequence mechanism models). By systematically sampling source term uncertainty parameters and consequence uncertainty parameters, input matrices are generated separately, which provides an efficient, reliable and scientific data basis for nuclear accident consequence analysis, and can significantly improve the comprehensiveness of analysis and the reliability of decision-making.

[0086] Optionally, the above-mentioned source term uncertainty parameter input matrix is ​​used to perform batch calculations using an integrated severe accident analysis program to obtain source term analysis statistical results of uncertainty parameters, which may specifically include:

[0087] Input the source term uncertainty input matrix into the integrated severe accident analysis program, perform batch calculations, and obtain the source term results corresponding to each source term uncertainty parameter combination in the source term uncertainty parameter input matrix;

[0088] The source item results are statistically processed to obtain source item analysis statistical results. The source item analysis statistical results include statistical characteristics of the source item results. The statistical characteristics of the source item results are used to characterize the uncertainty of the source item results.

[0089] Specifically, first, the source uncertainty input matrix is ​​loaded row by row into the severe accident analysis program; the program automatically runs the integrated analysis module according to the input parameter combination to simulate various accident scenarios; each group of parameter combinations will generate a corresponding set of source results (for example, a simulation may output: the release share of CsI to the environment); multiple parameter combinations are run in batches to obtain a set of source results representing all scenarios.

[0090] Then, the source result data set of the severe accident analysis program is collected; statistical analysis is performed on each output result, including calculation of mean, variance, probability density distribution, etc.; the uncertainty characteristics of the release amount are presented through data visualization (such as probability distribution diagrams, histograms); based on the source analysis statistical data, the characteristics of uncertainty (such as mean, distribution law) are clarified.

[0091] In this embodiment, by constructing an input-output-analysis link based on a rigorous mathematical model, through multi-scenario combination simulation of input parameters, and combining the data statistical processing process, the uncertainty characteristics of source release are fully quantified, making the prediction information more accurate and scientific.

[0092] Optionally, the above-mentioned source term results according to the benchmark working conditions or the statistical results of the source term analysis of the uncertainty parameters and the consequence uncertainty parameter input matrix are batch calculated using the radioactive consequence mechanism model to obtain the statistical results of the radioactive consequences of the uncertainty parameters, specifically including:

[0093] The source term results of the baseline condition and the consequence uncertainty parameter input matrix are input into the consequence mechanism model to obtain the radioactive consequences corresponding to each consequence uncertainty parameter combination;

[0094] The radioactive consequences are statistically processed to obtain radioactive consequence statistical results, which include statistical characteristics of the radioactive consequences, and the statistical characteristics are used to characterize the uncertainty degree of the radioactive consequences.

[0095] Among them, the source term results of the baseline operating conditions refer to the radioactive source terms obtained through nuclear power plant accident analysis under assumed initial conditions and accident scenarios, which may include the above-mentioned initial source term results.

[0096] Specifically, in the first step, the source term result output of the baseline working condition is used as the initial input of the consequence mechanism model (source term release condition), and the consequence uncertainty parameter input matrix rows (i.e. different environmental condition parameter combinations) are input into the model one by one; under each set of consequence parameter combinations, the mechanism model is used to simulate the diffusion and deposition behavior of radionuclides to generate numerical results of environmental impact (such as dose distribution in different directions); batch calculations are performed on all consequence input parameter combinations to complete the consequence result simulation of all scenarios; the regional doses and contamination characteristics generated by each simulation are the radioactive consequences of a single set of parameters. The second step is to collect the radioactive consequence results of the consequence mechanism model batch calculations; use probability and mathematical statistics methods to extract the trend and uncertainty description of complex consequence data.

[0097] In this embodiment, the volatility of radioactive consequence data is accurately described through the consequence mechanism model and statistical analysis, combining the source term results of the baseline condition and the consequence uncertainty parameter input matrix. The complex results are quantified and simplified through statistical characteristics, making the uncertainty of radioactive consequences not only clear and intuitive, but also significantly enhancing the scientific explanatory power, which is easier for evaluators and decision makers to understand.

[0098] Optionally, the above-mentioned source term results of the benchmark working condition or the statistical results of the source term analysis of the uncertainty parameters and the consequence uncertainty parameter input matrix are used to perform batch calculations using the radioactive consequence mechanism model to obtain the radioactive consequence statistical results of the uncertainty parameters, which may specifically include:

[0099] The statistical results of source term analysis and consequence uncertainty parameter input matrix are input into the consequence mechanism model to obtain the radioactive consequences corresponding to each source term uncertainty parameter combination and consequence uncertainty parameter combination;

[0100] The radioactive consequences are statistically processed to obtain radioactive consequence statistical results, which include statistical characteristics of the radioactive consequences, and the statistical characteristics are used to characterize the uncertainty degree of the radioactive consequences.

[0101] Specifically, in the first step, the statistical results of the source analysis are used as the initial input of the consequence mechanism model (source release conditions), and the consequence uncertainty parameters are input into the model one by one in each row of the matrix (i.e., different environmental condition parameter combinations); batch calculations are performed on all consequence input parameter combinations to complete the consequence result simulation of all scenarios; the regional doses and contamination characteristics generated by each simulation are the radioactive consequences of a single set of parameters. In the second step, the radioactive consequence results of the batch calculation of the consequence mechanism model are collected; and the trend and uncertainty description of the complex consequence data are extracted using probability and mathematical statistics methods.

[0102] In this embodiment, the volatility of radioactive consequence data is accurately described through the consequence mechanism model and statistical analysis, combining the statistical results of source term analysis and the consequence uncertainty parameter input matrix.

[0103] Optionally, the above step 104 specifically includes:

[0104] According to the uncertainty parameters of each source item and the statistical results of source item analysis, the first sensitive reference value of the source item uncertainty parameter to the source item is determined by using the sensitivity analysis method;

[0105] According to the consequence uncertainty parameter and the consequence analysis statistical results, the sensitivity analysis method is used to determine the second sensitive reference value of the consequence uncertainty parameter to the consequence;

[0106] For source and consequence uncertainty parameters, based on the statistical results of consequence analysis, the sensitivity analysis method is used to determine the third sensitive reference value of all uncertainty parameters to the consequence;

[0107] A sensitivity analysis result is obtained according to the first sensitive reference value, the second sensitive reference value and the third sensitive reference value of each uncertainty parameter.

[0108] Among them, the sensitivity analysis results are used to characterize the influence of each source uncertainty parameter and each consequence uncertainty parameter on the source result or consequence result.

[0109] Sensitivity analysis methods refer to the importance analysis methods used to quantify the impact of input parameters on model outputs. They can include: local sensitivity analysis, making small perturbations to input parameters one by one and analyzing output changes; global sensitivity analysis, considering the impact of input parameter changes over the entire range on outputs, such as the Sobol method, sensitivity analysis based on variance decomposition, etc.

[0110] The first sensitive reference value refers to a quantitative indicator of the degree of influence of each source uncertainty parameter on the statistical results of the source term. The higher the sensitive reference value, the greater the impact of the change in the parameter on the release results of the source term. The second sensitive reference value is a quantitative indicator of the impact of each consequence uncertainty parameter on the statistical characteristics of radioactive consequences. Parameters with higher reference values ​​indicate a more significant impact on the consequences of the accident. The third sensitive reference value is a quantitative indicator of the degree of influence of all source terms and consequence uncertainty parameters on the final consequences. The higher the reference value, the more sensitive and important the parameter is to the statistical characteristics of the consequences.

[0111] Specifically, first, the statistical results can be analyzed based on the source term uncertainty parameters and the corresponding source terms; the sensitivity analysis method can be used to perturb each source term parameter (such as changing the release rate) and calculate the degree of influence on the source term results; the sensitivity of each parameter to the source term results can be quantified through sensitivity indicators; the first sensitive reference value set of the source term parameters is obtained, and the parameters that are most important to the source term results are sorted and output.

[0112] Secondly, based on the uncertainty parameters of the consequences and the corresponding statistical results of the consequences; perturb or adjust the distribution of each consequence input parameter and observe the changes in the statistical results; use the sensitivity analysis method to calculate the degree of influence of each parameter on the consequence uncertainty results; obtain the second sensitive reference value set of the consequence uncertainty parameters, and sort them to identify the most sensitive consequence parameters; finally, input the source uncertainty parameters and the consequence uncertainty parameters as a complete parameter set; conduct a comprehensive analysis of the consequence analysis statistical model, and use the global sensitivity analysis method to evaluate the contribution of source terms and consequence parameters to the uncertainty of the consequences; calculate the third sensitive reference value according to the degree of influence of the parameters on the consequence statistical results (such as dose distribution probability); output the sensitivity ranking to highlight the parameters that have a greater impact on the consequences of the accident.

[0113] In this embodiment, the degree of influence of each uncertainty parameter on the source term and radioactive consequences is systematically evaluated. First, based on the source term analysis and radioactive consequence analysis, appropriate sensitivity analysis methods are used to accurately measure the sensitivity of each parameter in different analysis links. Subsequently, by comprehensively considering the sensitive reference values ​​of the source term and the consequences, the influence of the parameters on the overall system is fully reflected. By combining the statistical results of the source term analysis, the statistical results of the radioactive consequences, the uncertainty parameter set and the sensitivity analysis method, a sensitivity analysis is systematically performed on each uncertainty parameter to obtain the first sensitive reference value, the second sensitive reference value and the third sensitive reference value. It can comprehensively cover the key influencing factors in the source term release and radioactive consequence stages, and accurately quantify the sensitivity of each parameter to the consequences of the accident.

[0114] Optionally, the sensitivity analysis method includes at least one of a rank correlation coefficient method, a sensitivity analysis based on a sampling method, and a regression method.

[0115] Rank correlation coefficient method: It is a method based on nonparametric statistics, which is used to measure the correlation between the change of input parameters and the output results. The core of its sensitivity analysis is to quantify the monotonic relationship between variables by calculating the rank correlation between input parameters and output results (such as Spearman rank correlation coefficient). When the absolute value of the rank correlation coefficient is closer to 1, it indicates that the input parameter has a more significant impact on the result (higher sensitivity).

[0116] Sensitivity analysis method based on sampling method: it is a method of randomly or systematically selecting sample points within different ranges to evaluate the sensitivity of the output to various input parameters.

[0117] Regression method: By constructing a mathematical regression model (such as a linear or nonlinear function form) between the input parameters and the output variables, the impact of changes in the input parameters on the output results is evaluated.

[0118] In this embodiment, three sensitivity analysis methods are provided, covering different dimensions from local sensitivity (such as rank correlation coefficient method) to global sensitivity (such as variance decomposition method), which can adapt to different model characteristics and accident analysis scenarios. Users can choose the most suitable analysis method according to actual needs to ensure the scientificity and rationality of model evaluation.

[0119] In order to facilitate the understanding of the analysis method of the radioactive source term and consequences of a severe accident in a pressurized water reactor provided in this embodiment, a practical application description of the above method is provided here. Figure 2 For details, see the following example:

[0120] Step 1: Determine the research object, accident sequence and analysis procedure, and complete the modeling and baseline condition calculation (i.e., through the target analysis procedure, model the reactor and perform baseline condition calculation to obtain the initial source term results and initial consequence results).

[0121] (1) It is necessary to determine the type of pressurized water reactor for the uncertainty analysis of radioactive sources and consequences of severe accidents, and select the appropriate accident sequence. The probabilistic safety analysis results of specific reactor types can be referred to during the accident selection process, such as large break loss of coolant accident (LOCA) and plant blackout accident (SBO). These accidents have a high frequency of occurrence in probabilistic safety analysis (PSA). (2) Based on the design characteristics of the selected reactor type, establish a severe accident radioactive source and consequence analysis model. Select an integrated severe accident analysis program, such as MAAP, which can simulate the main severe accident phenomena, including core melting, radioactive material release and migration. According to the design parameters of the nuclear power plant, establish an accident analysis model in the analysis program. It can include: modeling of the reactor core structure, lower head, reactor coolant system, containment system, safety system, etc. (3) Carry out baseline operating condition calculations. Calculate the selected accident sequence. Provide a reference benchmark for subsequent uncertainty analysis.

[0122] Step 2: Identify and quantify sources of uncertainty.

[0123] In the process of severe accident analysis of nuclear power plants, the calculation results are uncertain due to the uncertainty of program models and power plant status parameters.

[0124] Starting from the release, migration and deposition angles of source terms, we conduct research covering the entire process including in-core processes inside the pressure vessel (such as core melting and migration of molten material), ex-core processes outside the pressure vessel (such as interaction between the molten material and the pressure vessel and reactor cavity), containment behavior (such as containment leakage, etc.), aerosol dynamic behavior (such as generation, agglomeration and sedimentation of radioactive particles), and full-time series of key phenomena of severe accidents, as well as processes such as atmospheric transport, diffusion and sedimentation under the Gaussian plume model, to identify the source terms of severe accidents and sources of uncertainty in their consequences.

[0125] Specifically, set reasonable ranges and probability distributions for each uncertainty factor, for example: parameter uncertainty can be quantified by comparing experimental data, user manuals, and test results in the literature.

[0126] Step 3: Uncertainty parameter database (determine initial screening parameters based on reference data).

[0127] Different severe accident system analysis programs have different simulation models for the same physical phenomena. Based on the phenomenon identification ranking table, relevant separation / integral effect experimental database, public literature, program user manual and severe accident analysis experience, this paper systematically and completely sorts out the models related to severe accident radioactive source terms and consequences for the mainstream integrated severe accident analysis programs, completes the correspondence between "severe accident phenomena and consequences-analysis models-uncertainty parameters", and preliminarily identifies the uncertainty parameters.

[0128] The details are as follows: (1) Collect information.

[0129] Severe accident phenomenon identification: Identify key physical phenomena through phenomenon identification and prioritization (PIRT). Experimental data: Collect relevant data from separation effect experiments and overall effect experiments. Literature and manuals: Consult publicly published research literature and user manuals of analysis programs.

[0130] (2) Parameter identification: Based on the data collected above, the corresponding relationship between physical phenomena and model parameters is established.

[0131] (3) Parameter preliminary selection. List all uncertain parameters that may affect radioactive source terms and consequences to form an initial parameter list. This provides basic information for subsequent uncertainty analysis.

[0132] Step 4: Use the Morris method to complete further screening of the initial screening parameters and determine the value range and probability distribution of the input parameters (based on the source item sensitivity value of each initial screening parameter, perform a second screening of the initial screening parameters).

[0133] The Morris method is a qualitative global sensitivity analysis method based on screening analysis. It uses the "one variable at a time" method to generate a random parameter matrix. Each time, only one parameter is adjusted while keeping other parameters unchanged until all parameters in the matrix are changed. The sampling rules are shown in Figure 3 (That is, based on the initial source item results, the Morris method is used to perform a global sensitivity analysis on each initial screening parameter to determine the source item sensitivity value of each initial screening parameter).

[0134] Specifically, multiple discrete levels are set for each parameter; then an experimental matrix is ​​designed. Model calculation is performed based on the experimental matrix; the average effect and standard deviation of each parameter are calculated; based on the average effect and standard deviation, it is determined which parameters have a significant impact on the output.

[0135] The biggest feature of the Morris method is that it is relatively simple to apply and has a small amount of calculation. It is particularly suitable for complex models with many parameters. A qualitative ranking of model parameters can be obtained through a small amount of model calculation, thereby identifying relatively important parameters to achieve the purpose of simplifying the model. Therefore, the Morris method is used to further screen the initial screening parameters in step 2 to obtain parameters that are more important to the radioactive source terms and consequences of severe accidents (i.e., according to the source term sensitivity value of each initial screening parameter, the initial screening parameters are screened for a second time to obtain the source term uncertainty parameters).

[0136] Refer to public literature, experimental databases, user manuals, etc. to determine the value range and probability distribution of the final screening uncertainty parameters. It should be noted that for a specific severe accident integrated analysis program, there may be correlations between certain parameters, and sampling needs to be performed according to the parameter setting requirements during subsequent sampling. In addition, the default parameter value range in the program may be too large. For example, if the parameters related to the core geometry in the MAAP program are set too large, the core will melt at the beginning of the accident, and the calculation will terminate immediately. The probability distribution of uncertainty parameters can be selected from different types such as discrete distribution, uniform distribution, triangular distribution, logarithmic triangular distribution, normal distribution, lognormal distribution, Beta distribution, linear distribution, etc. (that is, according to the nature of the source uncertainty parameters and reference data, the range and probability distribution of each source uncertainty parameter are determined to obtain the source uncertainty parameter set).

[0137] Step 5: Sample the uncertainty parameters and generate an input matrix (i.e., sample the source uncertainty parameter set and the consequence uncertainty parameter set respectively to generate a source uncertainty parameter input matrix and a consequence uncertainty parameter input matrix).

[0138] Uncertainty analysis methods can be divided into statistical analysis methods and deterministic analysis methods. Statistical analysis methods are widely used in the industrial field due to their advantages such as ease of execution and strong applicability. The Monte Carlo sampling method is simple and direct to implement. It obtains the distribution form of the response parameters by performing a large number of random samplings. Therefore, the Monte Carlo sampling method is selected to sample the selected uncertainty parameters.

[0139] The number of samplings is related to the calculation accuracy and speed of uncertainty and sensitivity analysis of radioactive sources and consequences of severe accidents. Non-parametric statistical methods are used as a means to quantify uncertainty, and the one-sided statistical tolerance interval of the continuous distribution population is solved when the distribution type is unknown.

[0140] For multiple input variables x i Perform N sampling and combine them as function input, and calculate N output variables {y 1 ,y 1 ,…,y N}, let g(y) be y n The probability distribution function of is unknown. The expression for determining the lower tolerance limit L and the upper tolerance limit U is shown in the following formula (1):

[0141]

[0142] Among them, β is the confidence level, γ is the probability level, and [L,U] is the tolerance interval. 1 ,y 1 ,…,y NArrange in ascending order, define y(0) = -∞, y(N+1) = +∞. For a certain confidence level and a certain probability level, the tolerance interval [L,U] can be determined, such that The probability is β, and the calculation formula of β is as follows:

[0143]

[0144] Among them, 0≤r≤s≤N, L=y(r), U=y(s).

[0145] If the minimum and maximum values ​​are taken as the two-sided tolerance interval, the confidence level is:

[0146] β=1-γ N -(N-1)(1-γ)γ N-1 (3)

[0147] If r=0, s=N, the calculation formula for the unilateral tolerance limit can be obtained as follows:

[0148]

[0149] The above formula is the Wilks formula, and the variables in the formula are the confidence level β, the probability level γ, and the number of samplings N. The number of samplings N has nothing to do with the number of input variables and the probability density distribution of each input variable, but only with the confidence level and probability level, and has the characteristics of a black box statistical model.

[0150] In engineering, it is believed that 95% confidence level and 95% probability level are sufficient to meet the high probability criteria. In the analysis of radioactive source items and consequences of severe accidents, it is usually necessary to consider only the single-sided tolerance limit, that is, the tolerance interval [-∞,U]. Table 1 lists the number of sampling times N required for the k-order Wilks formula to meet the single-sided tolerance upper limit at 95% confidence level and 95% probability level. In practical applications, considering that calculations may have problems under certain combinations of uncertain parameters, a certain margin is considered, that is, the number of sampling times is appropriately increased to compensate for possible calculation failures. See Table 1 for details.

[0151] Table 1: 95% confidence level 95% probability level one-sided tolerance interval upper limit sampling number table

[0152] Order k Calculation times N Order k Calculation times N 1 59 6 208 2 93 7 234 3 124 8 260 4 153 9 286 5 181 10 311

[0153] Step 6: Carry out batch calculation of severe accident source items (i.e., based on the source item uncertainty parameter set and consequence uncertainty parameter set, use the target analysis program to obtain the source item analysis statistical results of the source item uncertainty parameters and the radioactive consequence statistical results of the consequence uncertainty parameters).

[0154] The uncertainty input parameter matrix of the source term is written into the severe accident radioactive source term analysis model, batch calculation is carried out, and the source term results are statistically processed, including the maximum value, minimum value, average value, median value, 95% / 95% single-side tolerance upper limit value, and uncertainty band range of the source term. Through a large number of simulations, the changes in output results caused by changes in uncertainty parameters can be captured, the uncertainty of the results can be quantified, and a basis for risk assessment and decision-making can be provided.

[0155] Step 7: Perform batch calculations of radioactive consequences (i.e., based on the source term results of the baseline operating conditions and the consequence uncertainty parameter input matrix or the source term analysis statistical results of the uncertainty parameters and the consequence uncertainty parameter input matrix, use the radioactive consequence mechanism model to perform batch calculations to obtain the statistical results of the radioactive consequences of the uncertainty parameters).

[0156] The baseline source term results obtained in step 1 or the severe accident source term obtained in step 6 are used as the calculation input for radioactive consequences. The consequence uncertainty parameters are written into the radioactive consequence mechanism model. Batch calculations are carried out and statistical processing of the radioactive consequences is performed, including the maximum, minimum, average, median, 95% / 95% one-sided tolerance upper limit of the radioactive dose, and the uncertainty band range.

[0157] Step 8: Conduct sensitivity analysis (based on the statistical results of source term analysis, the statistical results of radioactive consequences, the source term uncertainty parameter set and the consequence uncertainty parameter set, conduct sensitivity analysis on each uncertainty parameter to obtain the sensitivity analysis results of each uncertainty parameter).

[0158] Sensitivity analysis is to find out the sensitive factors that have a significant impact on the phenomenon of interest from many uncertain parameters and rank them in order of importance.

[0159] Specifically, the sensitivity analysis can be completed from three perspectives: a) sensitivity analysis of source term uncertainty parameters to radioactive source terms of severe accidents; b) sensitivity analysis of consequence uncertainty parameters to radioactive consequences; 3) sensitivity analysis of all source term and consequence uncertainty parameters to radioactive consequences.

[0160] During the specific operation, different sensitivity analysis methods can be used to conduct uncertainty analysis on radioactive source terms and consequences, including the following methods:

[0161] (1) Method based on rank correlation coefficient

[0162] The rank correlation coefficient method is a nonparametric statistical method used to measure the relationship between two variables. Common rank correlation coefficient methods include the Spearman rank correlation coefficient and the Person rank correlation coefficient method. Among them, the Spearman rank correlation coefficient is more sensitive to nonlinear relationships.

[0163] Spearman rank correlation coefficient method:

[0164] The Spearman rank correlation coefficient converts the data into corresponding levels and then performs a regression process on the levels. 1 ,X 2 ,…,X n ) and the dependent variable Y(Y 1 ,Y 2 ,…,Y n ) is strong or weak, respectively i and Y i Sort and get R Xi and R Yi , and then the correlation coefficient is calculated according to the calculation formula of Spearman rank correlation coefficient ρ:

[0165]

[0166] The correlation coefficient ρ ranges from -1 to +1. When one variable has a strictly monotonic relationship with another variable, the Sperman rank correlation coefficient between the two variables is +1 or -1, which is called a complete Spearman correlation between the two variables. When the Sperman rank correlation coefficient is greater than 0, it means that the two parameters are positively correlated, less than 0 means negative correlation, and equal to 0 means no correlation.

[0167] Person correlation coefficient method:

[0168] The Person correlation coefficient is a linear correlation coefficient used to reflect the linear correlation between two variables. The specific expression is:

[0169]

[0170] in: X i , Y i The mean of .

[0171] (2) Sensitivity analysis based on sampling method

[0172] The core four items of the variance decomposition method are to decompose the model into single parameters and functions of the combination of parameters, calculate the contribution of the variance of a single input parameter or a set of input parameters to the output variance, and then analyze the importance of the parameters and the interaction between the parameters. This method can simultaneously give the impact of the uncertainty of the parameters themselves on the results, as well as the impact of the uncertainty of the parameters themselves and their interaction parameters on the results. See the following formulas (7) to (18).

[0173] Assume x = (x 1,x 2 ,…,x s ) are input parameters, assuming that the solution domain is Ω, x n It obeys the uniform distribution between [0,1], and the model is y = f(x) and f 2 (x) is integrable, the mean of the function f(x) is f 0 , a random variable with variance D, that is

[0174] f 0 =∫f(x)dx (7)

[0175] D = ∫f(x) 2 dx-f 0 2 (8)

[0176] All integrals are multiple integrals in [0,1] for each dimension. Variance decomposition is based on decomposing D into the effects of individual parameters, the combined effects of pairs of parameters, etc. The function f(x) is decomposed into:

[0177]

[0178] Among them, f 0 is a constant, and the integral of other addends to any variable contained in it is 0, that is,

[0179]

[0180] All addends in the function f(x) are orthogonal and can be expressed as:

[0181]

[0182] Decompose the function f(x) by squaring both sides, integrating and simplifying in the domain, and we get

[0183]

[0184] Integrating the expressions, we get:

[0185]

[0186] Then we can get the total variance D and partial variance of f(x) They are:

[0187] D = ∫f 2 (x)dx-f 0 2 (15)

[0188]

[0189] Corresponding to the corresponding parameter subset, The variance of is called partial variance. The Sobol sensitivity index corresponding to that subset of parameters is defined as:

[0190]

[0191] Where: gives the first-order contribution to the output variance from the ith input parameter, Used to calculate the second-order contribution from the interaction between the ith input parameter and the jth input parameter.

[0192] Finally, the full-order sensitivity index is defined as the sum of all sensitivity indices, which is Quantifies the overall impact of a parameter on the model output.

[0193] The respective sensitivity indices sum to 1 and are defined as:

[0194]

[0195] In summary, the first-order sensitivity index is mainly used to measure the partial contribution of a single parameter to the output variance; the second-order sensitivity index is used to measure the partial contribution of the interaction between parameters to the output variance; the full-order sensitivity index comprehensively considers the main, first-order and second-order effects, covering the evaluation within the full parameter space. The higher the sensitivity index value, the greater the influence of the corresponding model parameters and their related steps. 0.05 is usually used as the value to distinguish important parameters from unimportant parameters.

[0196] (3) Regression method

[0197] Regression sensitivity analysis generally refers to the use of regression analysis techniques to establish a functional relationship between input parameters and output results in order to quantify the impact of parameters on output results. Common regression methods include linear regression, polynomial regression, etc. These regression models require the evaluation of model parameters, such as the slope and intercept of the fitted line, which reflect the relationship between input parameters and output results.

[0198] Use R for analysis 2 , mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), residual analysis and other indicators are used to evaluate the parameters in the regression model.

[0199] Among them, R 2 It indicates the degree of model fitting data, and its value range is between 0 and 1. The closer to 1, the better the model can fit the data, and the closer to 0, the worse the model can fit the data. The specific expression is: R 2 =1-(SSR / SST), SSR is the residual sum of squares, SST is the total sum of squares. MSE is used to measure the square value of the average difference between the model prediction value and the actual value, and the specific expression is: n is the number of samples, y i , are the ith actual value and the model predicted value respectively. RMSE is similar to the MSE indicator.

[0200] One or more sensitivity analysis methods can be selected to conduct sensitivity analysis on uncertainty parameters to determine the parameters that have a greater impact on the radioactive source terms and consequences of severe accidents.

[0201] In this embodiment, in view of the current situation that radioactive source terms and consequences under severe accidents in nuclear power plants have great uncertainty, an uncertainty and sensitivity analysis method for radioactive source terms and consequences of severe accidents in pressurized water reactors is provided to quantify the uncertainty interval of radioactive source terms and consequences, and to sort the important uncertainty parameters to determine the parameters that have a greater impact on radioactive source terms and consequences of severe accidents. This provides a reference for severe accident management and safety improvement of nuclear power plants.

[0202] Overall, this technical solution significantly improves the accuracy and reliability of nuclear reactor accident consequence assessment, optimizes risk management and emergency decision-making processes, and enhances the scientific nature and effectiveness of nuclear safety management.

[0203] In an example, the uncertainty and sensitivity analysis method of severe accident radioactive source terms and consequences is used to analyze the uncertainty and sensitivity of severe accident radioactive source terms in a nuclear power plant. The specific process is as follows:

[0204] 1) The research object of the embodiment is the radioactive source term of a severe accident in a nuclear power plant, and the accident sequence is selected after the large break loss of coolant under the intact release type of the containment. The MAAP5 program is used to model a nuclear power plant, and the modeling mainly includes the core, primary and secondary circuit systems, containment system, severe accident prevention and mitigation system, and source term analysis related models. The primary circuit node and containment control body division diagram are shown in Figure 4 , Figure 5 , Figure 6 .

[0205] The source term results calculated in the program are given in the form of groups, which are divided into 18 groups in total. The nuclides (compounds) mainly included in each group are shown in Table 2.

[0206] Table 2 Grouping of fission products of MAAP5 program

[0207]

[0208] 2) After preliminary screening and further screening, the selected uncertainty parameters, as well as the parameter value range and probability distribution are shown in Table 3.

[0209] Table 3 Source uncertainty parameter list

[0210]

[0211]

[0212] 3) The python script is used to sample the uncertainty parameters. The Monte Carlo method is selected as the sampling method. Referring to the Wilks formula, the number of samples selected is 200 times with a certain margin. After generating the sampling matrix, batch calculations are carried out on the serious accident source items.

[0213] 4) Conduct source uncertainty analysis using noble gases and CsI as examples. Figure 7 , Figure 8 The temporal distribution of the inert gas and CsI fractions released into the environment after a large break LOCA is calculated using 200 sets of source term uncertainty samples.

[0214] Depend on Figure 7 and Figure 8 It can be seen that the two fission product groups are greatly affected by the uncertainty parameters, among which the release share of inert gases is very high; there are some differences in the curve of the change of CsI release share to the environment over time. Under certain conditions, the release share of CsI shows a step phenomenon within a certain period of time after the accident, which may be caused by the secondary vaporization of volatile fission products caused by decay heat.

[0215] Table 4 shows the uncertainty statistics of the release of noble gases and CsI to the environment 72 hours after the accident. The highest environmental source item 72 hours after the accident was 5.73E-3, the lowest was 5.65E-4, the uncertainty band was about 5.16E-3, and the upper limit of the double 95% tolerance was 4.29E-3; the highest environmental source item of CsI 72 hours after the accident was 9.77E-6, the lowest was 1.87E-7, the uncertainty band was about 9.59E-6, and the upper limit of the double 95% tolerance was 5.79E-6;

[0216] Table 4 Comparison of the release fractions of typical fission product groups 72 hours after the accident (LLOCA)

[0217]

[0218]

[0219] The Spearman rank correlation coefficient method was used to analyze the sensitivity of 15 groups of source uncertainty parameters to the release of inert gas and CsI to the environment, and the input parameters that have a greater impact on the calculation results were screened out. Table 5 only lists some of the results that have a greater impact on the environmental source term results. It can be seen from the table that the uncertainty parameters that have a greater impact on the release of inert gas and CsI to the environment mainly include the containment leakage rate, aerosol shape factor, core porosity cutoff value, LMP parameters of core collapse, steam partial pressure of CsI during steam / aerosol equilibrium, and radial friction coefficient of core natural circulation.

[0220] Table 5 Spearman rank correlation coefficients of environmental source terms under LLOCA accidents

[0221]

[0222] Embodiment 2:

[0223] This embodiment provides a method for optimizing parameters of a nuclear reactor, the method comprising:

[0224] By using the parameter analysis method for the source term and consequence of a severe accident in a nuclear reactor provided in any of the above embodiments, a sensitivity analysis result of each uncertainty parameter of the nuclear reactor is obtained;

[0225] According to the sensitivity analysis results of each uncertainty parameter, the target adjustment parameters of the nuclear reactor are determined. The target adjustment parameters are used to guide the optimization direction of severe accident source and consequence analysis, including parameters such as model settings and design guidance values;

[0226] Adjust the target adjustment parameters of the nuclear reactor.

[0227] Specifically, by utilizing the sensitivity analysis results of the analysis method of radioactive source terms and consequences of severe accidents in pressurized water reactors, the target adjustment parameters of the nuclear reactor are scientifically determined and optimized to form a closed-loop process of analysis model and design improvement.

[0228] First, through sensitivity analysis, the key uncertainty parameters that affect source release and consequence results are identified, their sensitivity ranking is quantified, and the most important target adjustment parameters are screened out; secondly, these parameters are used to guide the optimization of the accident analysis model settings (such as adjusting the migration model parameters or the nuclide release rate), and design guidance values ​​related to highly sensitive parameters are proposed (such as optimizing the normal leakage rate of the containment or the risk resistance of the core design); finally, the target adjustment parameters are optimized and adjusted, and the adjusted parameters are re-entered into the model for verification, thereby improving the accuracy of the model analysis and the scientific nature of the design. It can effectively reduce the interference of low-sensitivity parameters, improve the efficiency of accident analysis, and provide a scientific decision-making basis for the design optimization and risk management of nuclear power plants, significantly enhancing their safety and risk resistance.

[0229] In this embodiment, the sensitivity analysis method is used to systematically identify and determine the key uncertainty parameters that have a significant impact on the source items and accident consequences of severe accidents in nuclear reactors, and then optimization strategies are formulated and adjustments are implemented for these parameters, which can effectively improve the safety performance and accident response capabilities of nuclear reactors.

[0230] Embodiment 3:

[0231] like Fig. 9 As shown, this embodiment provides an analysis device 900 for radioactive source terms and consequences of severe accidents in pressurized water reactors, which is used to implement the analysis method for radioactive source terms and consequences of severe accidents in pressurized water reactors provided in any of the above embodiments.

[0232] The analysis device 900 for radioactive source terms and consequences of severe accidents in pressurized water reactors specifically comprises:

[0233] The first calculation module 901 is used to model the reactor and perform baseline condition calculations to obtain initial source term results and initial consequence results through a target analysis program, wherein the target analysis program includes an integrated severe accident analysis program and a consequence mechanism program corresponding to the reactor type and nuclear accident type;

[0234] A second calculation module 902, connected to the first calculation module, is used to screen the initial input parameters according to the initial source item results and the initial consequence results to obtain a source item uncertainty parameter set and a consequence uncertainty parameter set, wherein the uncertainty parameters include parameters that affect the source item analysis results or consequence results of the target analysis program, and the uncertainty parameter set includes a range and probability distribution of each uncertainty parameter;

[0235] A third calculation module 903, connected to the second calculation module, is used to obtain source term analysis statistics of source term uncertainty parameters and radioactive consequence statistics of consequence uncertainty parameters using the target analysis program according to the source term uncertainty parameter set and consequence uncertainty parameter set;

[0236] The analysis module 904 is connected to the third calculation module and is used to perform sensitivity analysis on each uncertainty parameter based on the source term analysis statistical results and the radioactive consequence statistical results and the source term uncertainty parameter set and the consequence uncertainty parameter set to obtain the sensitivity analysis results of the uncertainty parameters.

[0237] Optionally, the second calculation module 902 includes:

[0238] A first determination unit is used to determine initial screening parameters according to reference data, where the reference data includes at least one of experimental data, standards, user manuals, research literature, and expert experience;

[0239] A second determination unit is connected to the first determination unit and is used to perform a global sensitivity analysis on each initial screening parameter using the Morris method according to the initial source item result, and determine a source item sensitivity value of each initial screening parameter, wherein the source item sensitivity value is used to characterize the degree of influence of the initial screening parameter on the source item result;

[0240] A first screening unit, connected to the second determining unit, is used to perform a second screening on the initial screening parameters according to the source item sensitivity value of each initial screening parameter to obtain a source item uncertainty parameter;

[0241] The third determination unit is connected to the screening unit and is used to determine the range and probability distribution of each source item uncertainty parameter according to the properties of the source item uncertainty parameter and reference data to obtain a source item uncertainty parameter set.

[0242] Optionally, the second calculation module 902 further includes:

[0243] The fourth determination unit is used to perform a global sensitivity analysis on each initial screening parameter using the Morris method according to the initial consequence result, and determine the consequence sensitivity value of each initial screening parameter, where the consequence sensitivity value is used to characterize the degree of influence of the initial screening parameter on the consequence result;

[0244] A second screening unit, connected to the fourth determination unit, is used to perform a second screening on the initial screening parameters according to the consequence sensitivity value of each initial screening parameter to obtain a consequence uncertainty parameter;

[0245] The fifth determination unit is connected to the second screening unit and is used to determine the range and probability distribution of each consequence uncertainty parameter according to the properties of the consequence uncertainty parameter and reference data to obtain a consequence uncertainty parameter set.

[0246] In this example, through systematic data collection and initial parameter screening, combined with the Morris method for global sensitivity analysis, key uncertainty parameters are accurately identified and screened. The Monte Carlo method is used to generate a variety of input matrices, and through efficient batch calculations, a comprehensive simulation and statistical analysis of severe accident radioactive source terms and their consequences are performed, quantifying the impact of source term release and timing on radioactive consequences. Finally, through comprehensive sensitivity analysis, the comprehensive impact of each uncertainty parameter on the source term and consequences is clarified.

[0247] Embodiment 4:

[0248] like Fig.10 As shown, this embodiment provides a nuclear reactor parameter optimization device 1010, the device comprising:

[0249] The apparatus 900 for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors provided in any of the above embodiments is used to obtain sensitivity analysis results of various uncertainty parameters of a nuclear reactor through the analysis method of radioactive source terms and consequences of severe accidents in pressurized water reactors;

[0250] The determination module 1001 is connected to the parameter analysis device 900 for the source term and consequence of a severe accident of a nuclear reactor, and is used to determine the target adjustment parameters of the nuclear reactor according to the sensitivity analysis results of each uncertainty parameter. The target adjustment parameters are used to indicate the optimization direction of the severe accident radioactive source term and consequence analysis, including parameters such as model settings and design guidance values;

[0251] The adjustment module 1002 is connected to the determination module 1001 and is used to adjust the target adjustment parameters of the nuclear reactor.

[0252] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors, characterized in that: include: Modeling the reactor and performing baseline calculations to obtain initial source term results and initial consequence results through a target analysis program, wherein the target analysis program includes an integrated severe accident analysis program and a consequence mechanism program corresponding to the reactor type and nuclear accident type; Screening initial input parameters according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set, wherein the uncertainty parameters include parameters of the source term analysis results or consequence results that affect the target analysis program, and the uncertainty parameter set includes a range and probability distribution of each uncertainty parameter; According to the source term uncertainty parameter set and consequence uncertainty parameter set, using the target analysis program to obtain source term analysis statistical results of the source term uncertainty parameters and radioactive consequence statistical results of the consequence uncertainty parameters; Based on the statistical results of the source term analysis, the statistical results of the radioactive consequences, the source term uncertainty parameter set and the consequence uncertainty parameter set, a sensitivity analysis is performed on each uncertainty parameter to obtain the sensitivity analysis results of the uncertainty parameters.

2. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 1, characterized in that: The initial input parameters are screened according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set, specifically including: Determining initial screening parameters based on reference data, wherein the reference data includes at least one of experimental data, standards, user manuals, research literature, and expert experience; According to the initial source item results, a global sensitivity analysis is performed on each initial screening parameter using the Morris method to determine the source item sensitivity value of each initial screening parameter, wherein the source item sensitivity value is used to characterize the degree of influence of the initial screening parameter on the source item results; According to the source item sensitivity value of each of the initial screening parameters, the initial screening parameters are screened for a second time to obtain the source item uncertainty parameter; According to the properties of the source term uncertainty parameters and the reference data, the range and probability distribution of each source term uncertainty parameter are determined to obtain the source term uncertainty parameter set.

3. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 2, characterized in that: The performing uncertainty analysis on the initial input parameters according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set also includes: According to the initial consequence results, a global sensitivity analysis is performed on each initial screening parameter using the Morris method to determine the consequence sensitivity value of each initial screening parameter, wherein the consequence sensitivity value is used to characterize the degree of influence of the initial screening parameter on the consequence results; According to the consequence sensitivity value of each of the initial screening parameters, the initial screening parameters are screened for a second time to obtain the consequence uncertainty parameter; According to the properties of the consequence uncertainty parameters and the reference data, the range and probability distribution of each consequence uncertainty parameter are determined to obtain the consequence uncertainty parameter set.

4. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 1, characterized in that: The obtaining of the source term analysis statistical results of the source term uncertainty parameters and the radioactive consequence statistical results of the consequence uncertainty parameters by using the target analysis program according to the source term uncertainty parameter set and the consequence uncertainty parameter set specifically includes: The source uncertainty parameter set and the consequence uncertainty parameter set are sampled respectively to generate a source uncertainty parameter input matrix and a consequence uncertainty parameter input matrix; According to the source term uncertainty parameter input matrix, batch calculation is performed using an integrated severe accident analysis program to obtain source term analysis statistical results of the uncertainty parameters; According to the source term results of the reference operating conditions and the consequence uncertainty parameter input matrix, or the source term analysis statistical results of the uncertainty parameters and the consequence uncertainty parameter input matrix, batch calculations are performed using a radioactive consequence mechanism model to obtain the radioactive consequence statistical results of the uncertainty parameters.

5. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 4, characterized in that: The sampling of the source uncertainty parameter set and the consequence uncertainty parameter set to generate a source uncertainty parameter input matrix and a consequence uncertainty parameter input matrix specifically includes: Sampling the source term uncertainty parameter set, taking values ​​according to the range and probability distribution of the source term uncertainty parameters, simulating multiple uncertainty parameter combinations of source term input parameters, and obtaining an input matrix of the source term uncertainty parameters; The consequence uncertainty parameter set is sampled, values ​​are taken according to the range and probability distribution of the consequence uncertainty parameters, and multiple uncertainty parameter combinations of consequence input parameters are simulated to obtain an input matrix of the consequence uncertainty parameters.

6. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 4, characterized in that: The step of performing batch calculations based on the source term uncertainty parameter input matrix using an integrated severe accident analysis program to obtain source term analysis statistical results of the uncertainty parameters specifically includes: Inputting the source term uncertainty input matrix into the integrated severe accident analysis program, performing batch calculations, and obtaining source term results corresponding to each source term uncertainty parameter combination; The source item result is statistically processed to obtain the source item analysis statistical result, wherein the source item analysis statistical result includes the statistical characteristics of the source item result, and the statistical characteristics of the source item result are used to characterize the uncertainty degree of the source item result.

7. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 6, characterized in that: The radioactive consequence statistical results of the uncertainty parameters are obtained by batch calculation using the radioactive consequence mechanism model according to the source term results of the reference working condition and the consequence uncertainty parameter input matrix, or the source term analysis statistical results of the uncertainty parameters and the consequence uncertainty parameter input matrix, specifically including: Inputting the source term results of the reference condition and the consequence uncertainty parameter input matrix into the consequence mechanism model to obtain the radioactive consequences corresponding to each combination of consequence uncertainty parameters; The radioactive consequences are statistically processed to obtain the radioactive consequence statistical results, wherein the radioactive consequence statistical results include statistical characteristics of the radioactive consequences, and the statistical characteristics are used to characterize the uncertainty degree of the radioactive consequences.

8. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 6, characterized in that: The method of performing batch calculations based on the statistical results of the source term analysis of the uncertainty parameters and the consequence uncertainty parameter input matrix using the radioactive consequence mechanism model to obtain the statistical results of the radioactive consequences of the uncertainty parameters specifically includes: Inputting the statistical results of the source term analysis and the consequence uncertainty parameter input matrix into the consequence mechanism model to obtain the radioactive consequences corresponding to the source term uncertainty parameter combinations and consequence uncertainty parameter combinations; The radioactive consequences are statistically processed to obtain the radioactive consequence statistical results, wherein the radioactive consequence statistical results include statistical characteristics of the radioactive consequences, and the statistical characteristics are used to characterize the uncertainty degree of the radioactive consequences.

9. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 1, characterized in that: Based on the source term analysis statistical results, radioactive consequence statistical results, source term uncertainty parameter set and consequence uncertainty parameter set, a sensitivity analysis is performed on each uncertainty parameter to obtain a sensitivity analysis result of each uncertainty parameter, specifically including: According to each source term uncertainty parameter and the statistical results of the source term analysis, a sensitivity analysis method is used to determine a first sensitive reference value of the source term uncertainty parameter to the source term; According to the consequence uncertainty parameter and the consequence analysis statistical results, using a sensitivity analysis method, determine a second sensitive reference value of the consequence uncertainty parameter to the consequence; For source items and consequence uncertainty parameters, according to the consequence analysis statistical results, using sensitivity analysis methods, determine the third sensitive reference value of all uncertainty parameters to the consequence; The sensitivity analysis result is obtained according to the first sensitive reference value, the second sensitive reference value and the third sensitive reference value of each uncertainty parameter, and the sensitivity analysis result is used to characterize the influence of each source uncertainty parameter and each consequence uncertainty parameter on the source result or the consequence result.

10. The method for analyzing radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 9, characterized in that: The sensitivity analysis method includes at least one of a rank correlation coefficient method, a sensitivity analysis based on a sampling method, and a regression method.

11. A method for optimizing parameters of a nuclear reactor, characterized in that: The method comprises: By using the analysis method of radioactive source terms and consequences of severe accidents in pressurized water reactors as described in any one of claims 1 to 10, a sensitivity analysis result of each uncertainty parameter of the nuclear reactor is obtained; Determining target adjustment parameters of the nuclear reactor according to the sensitivity analysis results of each uncertainty parameter, wherein the target adjustment parameters are used to guide the optimization direction of severe accident source items and consequence analysis, including parameters such as model settings and design guidance values; The target adjustment parameters of the nuclear reactor are adjusted.

12. An analysis device for radioactive source terms and consequences of severe accidents in pressurized water reactors, characterized in that: The device comprises: A first calculation module is used to model the reactor and perform baseline condition calculations to obtain initial source term results and initial consequence results through a target analysis program, wherein the target analysis program includes an integrated severe accident analysis program and a consequence mechanism program corresponding to the reactor type and the nuclear accident type; A second calculation module, connected to the first calculation module, is used to screen the initial input parameters according to the initial source term results and the initial consequence results to obtain a source term uncertainty parameter set and a consequence uncertainty parameter set, wherein the uncertainty parameters include parameters of the source term analysis results or consequence results that affect the target analysis program, and the uncertainty parameter set includes a range and probability distribution of each uncertainty parameter; A third calculation module, connected to the second calculation module, is used to obtain source term analysis statistics of source term uncertainty parameters and radioactive consequence statistics of consequence uncertainty parameters using the target analysis program according to the source term uncertainty parameter set and consequence uncertainty parameter set; The analysis module is connected to the third calculation module and is used to perform sensitivity analysis on each uncertainty parameter based on the source term analysis statistical results and the radioactive consequence statistical results and the source term uncertainty parameter set and the consequence uncertainty parameter set to obtain the sensitivity analysis results of the uncertainty parameters.

13. The analysis device for radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 12, characterized in that: The second calculation module includes: A first determination unit is used to determine initial screening parameters according to reference data, wherein the reference data includes at least one of experimental data, standards, user manuals, research literature and expert experience; A second determination unit, connected to the first determination unit, is used to perform a global sensitivity analysis on each initial screening parameter using the Morris method according to the initial source item result, and determine a source item sensitivity value of each initial screening parameter, wherein the source item sensitivity value is used to characterize the degree of influence of the initial screening parameter on the source item result; A first screening unit, connected to the second determining unit, is used to perform a second screening on the initial screening parameters according to the source item sensitivity value of each initial screening parameter to obtain the source item uncertainty parameter; The third determination unit is connected to the screening unit and is used to determine the range and probability distribution of each source item uncertainty parameter according to the properties of the source item uncertainty parameter and the reference data to obtain the source item uncertainty parameter set.

14. The analysis device for radioactive source terms and consequences of severe accidents in pressurized water reactors according to claim 13, characterized in that: The second calculation module also includes: A fourth determination unit is used to perform a global sensitivity analysis on each initial screening parameter using the Morris method according to the initial consequence result, and determine a consequence sensitivity value of each initial screening parameter, wherein the consequence sensitivity value is used to characterize the degree of influence of the initial screening parameter on the consequence result; A second screening unit, connected to the fourth determining unit, is used to perform a second screening on the initial screening parameters according to the consequence sensitivity value of each initial screening parameter to obtain the consequence uncertainty parameter; The fifth determination unit is connected to the second screening unit, and is used to determine the range and probability distribution of each consequence uncertainty parameter according to the properties of the consequence uncertainty parameter and the reference data, so as to obtain the consequence uncertainty parameter set.

15. A nuclear reactor parameter optimization device, characterized in that: The device comprises: The analysis device for radioactive source terms and consequences of severe accidents in pressurized water reactors as claimed in any one of claims 12 to 14 is used to obtain sensitivity analysis results of various uncertainty parameters of the nuclear reactor through the analysis method for radioactive source terms and consequences of severe accidents in pressurized water reactors; a determination module, connected to the nuclear reactor severe accident radioactive source term and consequence analysis device, for determining target adjustment parameters of the nuclear reactor according to the sensitivity analysis results of each uncertainty parameter, wherein the target adjustment parameters are used to guide the optimization direction of severe accident source term and consequence analysis, including parameters such as model setting and design guidance value; An adjustment module, connected to the determination module, is used to adjust the target adjustment parameters of the nuclear reactor.

Images