A method for optimizing the monitoring position of environmental aerosols after a nuclear accident

Abstract

The present application relates to a kind of nuclear accident environmental aerosol monitoring position optimization method, by preliminary estimation of the release source term caused by the accident of nuclear facility, in combination with meteorological parameter and the topographic parameter of the place where accident occurs, preliminary prediction of the concentration distribution of radionuclide in environment based on atmospheric diffusion model;Based on the concentration distribution of radionuclide in environment obtained based on atmospheric diffusion model by environmental aerosol monitoring data, the concentration distribution of radionuclide in environment is corrected;Using space simulated annealing algorithm, generate the emergency monitoring point position arrangement scheme with sampling representative, to optimize aerosol monitoring position sampling layout point.The method disclosed in the present application is systematically integrated to correct the predicted concentration distribution based on environmental aerosol monitoring data, and the environmental monitoring position is optimized, to improve the evaluation accuracy and real-time of the distribution of radionuclide in environment after the accident of nuclear facility, meet the demand of fast and accurate evaluation of environmental impact after the accident of nuclear facility.

Description

Technical Field

[0001] This invention belongs to the field of radiation monitoring and evaluation after a nuclear accident, and specifically relates to a method for optimizing the location of environmental aerosol monitoring after a nuclear accident. Background Technology

[0002] Following an accident at a nuclear facility, radioactive materials may be released into the environment, posing a threat to environmental and human safety. Rapid and appropriate consequence assessments and emergency decisions are crucial for mitigating the consequences of an accident and reducing its harm to people and the environment.

[0003] Rapid and accurate assessment of the spatial distribution of radionuclides in the external environment, and further calculation of their potential radiation doses, are fundamental to making informed decisions. Therefore, extensive research has been conducted both domestically and internationally on estimating personnel radiation doses during nuclear emergencies, resulting in significant progress in areas ranging from source term assessment and contamination monitoring to dose assessment. Current research primarily focuses on the rapid response characteristics of nuclear emergencies; further research is needed to improve the accuracy of assessment results based on rapid evaluation.

[0004] Currently, the main techniques for assessing the spatial distribution of radionuclides include predictive assessment methods based on atmospheric diffusion models and status assessment methods based on environmental monitoring data. Predictive assessment methods based on atmospheric diffusion models use known source terms or source terms estimated by other methods, combined with meteorological and topographical parameters, to predict radionuclide concentrations at different locations and times in the external environment using diffusion models such as Gaussian, Lagrange, and Eulerian models. However, since the source terms of accidents are usually unknown in reality, there is significant uncertainty between the estimated source terms and meteorological parameters, and the concentration distribution predicted by the algorithm may differ from the actual situation. While sampling and analyzing aerosol monitoring points in the external environment can reflect the actual situation, it cannot reflect concentration information at non-monitoring locations. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide a method for optimizing the location of environmental aerosol monitoring after a nuclear accident. This method uses environmental aerosol monitoring data, corrects the predicted aerosol concentration field based on Kriging interpolation, and optimizes the monitoring point locations using a spatial simulated annealing algorithm to obtain an environmental aerosol concentration distribution field that more closely approximates the actual situation.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is: a method for optimizing the location of environmental aerosol monitoring after a nuclear accident, the method comprising the following steps:

[0007] S1. Preliminary estimate of the release source terms caused by an accident at the nuclear facility;

[0008] S2. Using the estimated release source term, meteorological parameters after the accident, and topographic parameters of the accident site, a preliminary prediction of the concentration distribution of radionuclides in the environment is made based on an atmospheric diffusion model.

[0009] S3. Correct the concentration distribution of radionuclides in the environment based on the atmospheric diffusion model based on environmental aerosol monitoring data;

[0010] S4. Use the spatial simulation annealing algorithm to generate a representative emergency monitoring point location layout scheme to optimize the actual aerosol monitoring location sampling point layout.

[0011] Furthermore, after step S4, the method further includes the following steps:

[0012] S5. Use environmental aerosol monitoring data after optimizing the sampling layout of actual aerosol monitoring locations to correct the concentration distribution of radionuclides in the environment.

[0013] Furthermore, after step S4, the method further includes the following steps:

[0014] Based on the corrected concentration distribution of radionuclides in the environment obtained in step S5, a spatial simulation annealing algorithm is used to generate a more representative emergency monitoring point layout scheme to optimize the actual aerosol monitoring sampling point layout.

[0015] Furthermore, the release source item mentioned in step S1 includes the types of radionuclides released by the nuclear facility accident, the total amount released, the amount of each nuclide released, and the duration of the release.

[0016] Furthermore, in step S1, the release source terms caused by an accident at the nuclear facility are estimated based on accident condition monitoring data, expert judgment, or other source term inversion methods.

[0017] Furthermore, the atmospheric diffusion model mentioned in step S2 includes the Gaussian model, the Lagrange model, and the Eulerian model.

[0018] Furthermore, in step S3, based on environmental aerosol monitoring data, after obtaining the Kriging interpolation results, the resulting residual term is input into the predicted concentration field based on the atmospheric diffusion model for correction.

[0019] Furthermore, step S4 includes the following sub-steps:

[0020] S41. Set the regression kriging variance in step S3 as the first part of the cost function, and use the weighted sum of the areas occupied by false positives and false negatives of emergency monitoring points as the second part of the cost function. The two cost functions, the first part of the cost function and the second part of the cost function, are weighted and summed to construct a cost function, so as to optimize the accuracy of the judgment of the scope of the accident impact and the corrected concentration distribution at the same time.

[0021] S42. Use the spatial simulation annealing algorithm to generate a more representative emergency monitoring point location layout scheme.

[0022] Furthermore, step S42 includes the following sub-steps:

[0023] a) Starting from a random initial monitoring location S0, calculate the relevant cost function value C(S0);

[0024] b) Given position S k A new candidate monitoring location S is constructed by moving a randomly selected monitoring location a distance h. k+1 The direction of h is randomly chosen, and its length is a random number between zero and the maximum displacement. The cost function C(S) for calculating the new position is then used. k+1 );

[0025] c) If C(S) k+1 )<C(S k If the new position is accepted, then k is incremented by 1, and the new position S is used. k+1 As the starting point, return to step b); otherwise, use the old position S. k ;

[0026] d) After a certain number of iterations, or when other stopping criteria are met, stop and store the monitoring position that minimizes the cost function value.

[0027] The beneficial technical effects of this invention are as follows: by adopting the method for optimizing the location of environmental aerosol monitoring after a nuclear accident disclosed in this invention, the method of correcting the predicted concentration distribution based on environmental aerosol monitoring data and the method of optimizing the location of environmental monitoring are systematically integrated, thereby improving the accuracy and real-time nature of the assessment of the distribution of radionuclides in the environment after a nuclear facility accident, and meeting the need for rapid and accurate assessment of the environmental impact after a nuclear facility accident. Attached Figure Description

[0028] Figure 1 This is a flowchart of a method for optimizing the location of environmental aerosol monitoring after a nuclear accident, as described in an embodiment of the present invention. Detailed Implementation

[0029] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.

[0030] Example 1

[0031] This invention provides a method for optimizing the location of environmental aerosol monitoring after a nuclear accident, the method comprising the following steps:

[0032] S1. Based on the operating conditions at the time of the nuclear facility accident, make a preliminary estimate of the release sources caused by the accident, mainly including the types of radionuclides released by the nuclear facility accident, the total amount released (Bq), the amount of each nuclide released, and the duration of the release.

[0033] The release source terms caused by the accident can be estimated based on accident condition monitoring data, expert judgment, or other source term inversion methods.

[0034] S2. Using the estimated release source term caused by the accident, meteorological parameters after the accident, and topographic parameters of the accident site, based on the atmospheric diffusion model, make a preliminary prediction of the concentration distribution of radionuclides in the environment, that is, the concentration of radionuclides (Bq / m3) at different locations in the environment.

[0035] Commonly used atmospheric diffusion models include Gaussian models, Lagrange models, and Eulerian models. Mature atmospheric diffusion simulation programs such as AERMOD and CALPUFF are available. Appropriate diffusion models can be selected for calculation based on actual conditions. Specific atmospheric diffusion simulation methods will not be detailed in the embodiments of this invention.

[0036] S3. Correct the concentration distribution of radionuclides in the environment based on the atmospheric diffusion model using environmental aerosol monitoring data.

[0037] Conventional monitoring equipment is usually installed around nuclear facilities to obtain environmental aerosol monitoring data after a nuclear facility accident. Based on the Kriging interpolation method, the preliminary predicted distribution of radionuclide concentrations obtained in step S2 based on the atmospheric diffusion model is corrected.

[0038] Kriging interpolation mathematically provides an optimal linear unbiased estimate (a definite value at a certain point) of the object under study. Its definition is as follows:

[0039]

[0040] Where z(s) is the target environmental variable, i.e., the concentration of radionuclides, x(s) are m environmental covariates (coordinates of monitoring point locations), s = (x, y) represents two-dimensional spatial coordinates, β is the coefficient to be estimated, and ε(s) is the residual term after regression of the target variable and covariates, which follows a normal distribution with a mean of zero. Under the condition of satisfying the second-order stationarity assumption in geostatistics, the spatial autocorrelation characteristics of ε(s) can be quantitatively expressed by the covariance function or the variogram. Formula (1) can be written in matrix form as follows:

[0041] z(s)=x′β+ε(s) (2)

[0042] The target variable z(s) consists of two parts: the first part is the regression term or trend term, and the second part is the residual term. First, it is assumed that the target variable and covariates satisfy a certain regression relationship. Based on n known sample points, the generalized least squares method is used to estimate the regression coefficient β.

[0043]

[0044] Where C is the variance-covariance matrix of the n×n residuals, and X is the covariance matrix of the n×(m+1) sample points. Finally, by modeling the residuals, the optimal linear unbiased estimate at the point s0 to be estimated can be obtained as follows:

[0045]

[0046] Where x0 is a vector of values ​​of the covariates at the point to be estimated, and c0 is a vector of covariances between the sample point and the point to be estimated. Both C and c0 are obtained from the variation plot of ε(s).

[0047] The variance of the regression kriging at point s0 can be obtained as:

[0048] σ 2 (s0)=c(0)-c′0C -1 c0+x′ a (X′C -1 X) -1 x a (4)

[0049] Where, x a =x0-X′C -1 c0. Formula (4) can be decomposed into two parts: the first part (the first two terms) is the variance of the estimation error of the residuals, and the second part (the last term) is the variance of the estimation error of the trend term.

[0050] After obtaining the Kriging interpolation results, the resulting residual terms are input into the predicted concentration field based on the atmospheric diffusion model for correction. The final corrected concentration distribution includes both the concentration distribution data simulated based on the inverted source terms and meteorological parameters, as well as the actual monitoring data from the monitoring points. The corrected concentration distribution will be more consistent with the actual situation.

[0051] S4. Use the spatial simulation annealing algorithm to generate a more representative emergency monitoring point location layout scheme to optimize the actual sampling point layout and obtain representative aerosol monitoring data.

[0052] Optimize the location and layout of emergency monitoring points. After a nuclear facility accident, more emergency monitoring equipment can be deployed to obtain more detailed information on the concentration distribution of radionuclides.

[0053] The arrangement of monitoring points affects the results of concentration distribution correction. Spatial simulation annealing algorithm is used to generate sufficiently representative monitoring locations to meet the concentration distribution correction requirements on a large spatial scale, ensuring that the corrected concentration distribution can describe the actual concentration distribution of all points in space to the greatest extent.

[0054] Step S4 includes the following two sub-steps:

[0055] S41. Constructing the cost function

[0056] One purpose of setting up emergency monitoring points is to obtain the most accurate corrected concentration distribution possible. The first part of the cost function can be set as the regression kriging variance in formula (4) to obtain the optimal interpolation result.

[0057] For post-accident emergency monitoring, after setting an appropriate warning concentration threshold, the location within the target area can be divided into four sub-areas: false positive, false negative, true positive, and true negative.

[0058] Another objective in selecting optimal emergency monitoring sites is to minimize the costs associated with false positive and false negative decisions. False positive decisions occur when a set threshold is exceeded due to incorrect predicted concentration distribution, leading to unnecessary measures, such as evacuating people from a practically safe area. Conversely, the cost associated with false negative decisions is the inaction of necessary measures; the cost of false negatives can be greater than that of false positives because it can seriously impact people's health and lives. Therefore, another optimization objective for emergency monitoring site locations is to minimize the weighted sum of the areas occupied by false positives and false negatives. The second cost function is:

[0059] C = α·area (false positive) + (1-α)·area (false negative) (5)

[0060] Here, α is the weighting factor, and its value should be less than 0.5 because the consequences of false negatives are more serious.

[0061] This patent uses the weighted sum of the two cost functions of formula (4) and formula (5) as the final cost function to simultaneously optimize the accuracy of the judgment of the scope of the accident's impact and the concentration distribution after correction, so as to obtain the best location of the emergency monitoring point.

[0062] S42. Use the spatial simulation annealing algorithm to generate a more representative emergency monitoring point location layout scheme.

[0063] Spatial simulated annealing is a spatial extension of the simulated annealing algorithm, which has five main steps:

[0064] a) Starting from a (random) initial monitoring location S0, calculate the relevant cost function value C(S0);

[0065] b) Given position S k A new candidate monitoring location S is constructed by moving a randomly selected monitoring location a distance h. k+1 The direction of h is randomly chosen, and its length is a random number between zero and the maximum displacement. During the spatial simulation annealing iterations, the maximum displacement gradually decreases.

[0066] c) Calculate the cost function C(S) at the new location. k+1 If C(S) k+1 )<C(S k If a new position is found to be undesirable, then the new position is accepted; otherwise, the new position is accepted with a certain probability (the purpose of which is to ensure that the algorithm can escape local optima). If the new position is accepted, then k is incremented by 1. As the spatial simulated annealing iterations proceed, the probability of accepting an undesirable position gradually decreases.

[0067] d) Return to step b), if the new position is accepted, then use the new position S. k+1 Use it as the starting point; otherwise, use the old position S. k ;

[0068] e) Stop after a certain number of iterations, or when other stopping criteria are met. Store the monitoring position that minimizes the cost function value.

[0069] Since it is impossible to obtain actual values ​​of concentration data for all locations in the external environment, many possible real-world scenarios can be simulated by adding random errors to the predicted concentration distribution based on atmospheric diffusion models. The cost function value for a selected monitoring location can be calculated for a large number of possible real-world scenarios. If a sufficient number of real-world scenarios are simulated, the calculated average cost will be close to the expected cost associated with the monitoring location.

[0070] S5. Use environmental aerosol monitoring data after optimizing the monitoring site layout to correct the concentration distribution of radionuclides in the environment.

[0071] Based on the optimized monitoring site layout scheme obtained in step S4, environmental aerosol monitoring equipment is deployed to acquire environmental aerosol monitoring data. Based on the Kriging interpolation method, the concentration distribution of radionuclides in the environment is corrected using the environmental aerosol monitoring data after the optimized monitoring site layout.

[0072] The process of steps S4 and S5 can be iteratively performed according to actual needs. The environmental monitoring site layout scheme is optimized multiple times based on the space simulation annealing method. Emergency monitoring equipment is deployed to obtain environmental aerosol data. The concentration distribution field of radionuclides in the environment is corrected using the monitoring data based on the Kriging interpolation method to obtain an environmental aerosol concentration distribution field that is more consistent with the actual situation.

[0073] As can be seen from the above embodiments, the environmental aerosol monitoring location optimization method disclosed in this invention is based on Kriging interpolation and uses environmental monitoring data to quickly correct the distribution of radionuclide concentration fields. At the same time, it combines an environmental monitoring location layout optimization method based on spatial simulated annealing algorithm, which improves the accuracy and timeliness of predicting the distribution of radionuclide environmental concentrations after a nuclear facility accident, and achieves a more accurate and real-time post-accident assessment.

[0074] The method described in this invention is not limited to the embodiments described in the specific implementation. Other implementation methods derived by those skilled in the art based on the technical solution of this invention also fall within the scope of technical innovation of this invention.

Claims

1. A method for optimizing the location of environmental aerosol monitoring after a nuclear accident, the method comprising the following steps: S1. Preliminary estimate of release sources caused by an accident at the nuclear facility; S2. Using the estimated release source term, meteorological parameters after the accident, and topographic parameters of the accident site, a preliminary prediction of the concentration distribution of radionuclides in the environment is made based on an atmospheric diffusion model. S3. Based on environmental aerosol monitoring data, after obtaining the Kriging interpolation results, the resulting residual terms are input into the atmospheric diffusion model to correct the concentration distribution of radionuclides in the environment obtained from the atmospheric diffusion model. S4. Use the spatial simulation annealing algorithm to generate a sampling-representative emergency monitoring point layout scheme to optimize the actual aerosol monitoring sampling point layout. Step S4 includes the following sub-steps: S41. Set the regression kriging variance in step S3 as the first part of the cost function, and use the weighted sum of the areas occupied by false positives and false negatives of emergency monitoring points as the second part of the cost function. The two cost functions, the first part of the cost function and the second part of the cost function, are weighted and summed to construct a cost function, so as to optimize the accuracy of the judgment of the scope of the accident impact and the corrected concentration distribution at the same time. S42. Using the spatial simulated annealing algorithm, generate a more representative emergency monitoring point location layout scheme, including the following sub-steps: a) From a random initial monitoring location Begin by calculating the relevant cost function values. ; b) Given location By moving the randomly selected monitoring location a certain distance To construct a candidate new monitoring location , The direction is randomly chosen, and its length is a random number between zero and the maximum displacement. The cost function for calculating the new position is... ; c) If If the new position is accepted, then the new position is accepted; otherwise, the new position is accepted with a certain probability. If the new position is accepted, then... Add 1, use the new position (As a starting point, return to step b); otherwise, use the old position. ; d) After a certain number of iterations, or when other stopping criteria are met, stop and store the monitoring position that minimizes the cost function value.

2. The method for optimizing the location of environmental aerosol monitoring after a nuclear accident as described in claim 1, characterized in that... After step S4, the method further includes the following steps: S5. Use environmental aerosol monitoring data after optimizing the sampling layout of actual aerosol monitoring locations to correct the concentration distribution of radionuclides in the environment.

3. The method for optimizing the location of environmental aerosol monitoring after a nuclear accident as described in claim 2, characterized in that... After step S4, the method further includes the following steps: Based on the corrected concentration distribution of radionuclides in the environment obtained in step S5, a spatial simulation annealing algorithm is used to generate a more representative emergency monitoring point layout scheme to optimize the actual aerosol monitoring sampling point layout.

4. The method for optimizing the location of environmental aerosol monitoring after a nuclear accident as described in claim 3, characterized in that: The release source item mentioned in step S1 includes the types of radionuclides released by the nuclear facility accident, the total amount released, the amount of each nuclide released, and the duration of the release.

5. The method for optimizing the location of environmental aerosol monitoring after a nuclear accident as described in claim 4, characterized in that: In step S1, the release source terms caused by an accident at the nuclear facility are estimated based on accident monitoring data, expert judgment, or other source term inversion methods.

6. The method for optimizing the location of environmental aerosol monitoring after a nuclear accident as described in claim 5, characterized in that: The atmospheric diffusion models mentioned in step S2 include the Gaussian model, the Lagrange model, and the Eulerian model.

Images