Inversion method and device of nuclear leakage radioactive substance diffusion situation

Abstract

The invention discloses a nuclear leakage radioactive substance diffusion situation inversion method and device, and relates to the technical field of nuclear emergency response. The method comprises the following steps: acquiring initial source item parameters and meteorological data, and obtaining a nuclide concentration predicted value through a Gaussian puff model according to the initial source item parameters and the meteorological data; acquiring a nuclide concentration observation value corresponding to the nuclear leakage area, and performing Kalman filtering fusion processing according to the nuclide concentration prediction value and the nuclide concentration observation value to obtain a nuclide optimal concentration field and analysis error distribution; adjusting parameters of a Gaussian diffusion mode according to the analysis error distribution to obtain an adjusted Gaussian diffusion mode, and obtaining target source item parameters based on the adjusted Gaussian diffusion mode and the optimal concentration field; and performing iteration by taking the target source item parameter as an initial source item parameter until a preset requirement is met, and taking the source item parameter obtained when iteration is ended as a final source item parameter. The method can dynamically respond to the sudden change of atmospheric stability, and improves the prediction precision of the nuclear hazard area range.

Description

Technical Field

[0001] This application relates to the field of nuclear emergency response technology, and in particular to a method and apparatus for inverting the diffusion pattern of radioactive materials in a nuclear leak. Background Technology

[0002] In the event of a nuclear accident, rapidly and accurately obtaining real-time information on the accident's progress and the spread of contamination is crucial for organizing effective emergency response. However, traditional reconnaissance methods are inadequate for close-range investigations in the face of the highly radioactive environment surrounding the accident source and the heavily contaminated area. Inversion and analysis systems, on the other hand, can collect, fuse, and display information on radioactive materials, and utilize GIS combined with environmental information to conduct numerical simulations of radioactive material diffusion in a virtual geographic environment for sudden nuclear facility leaks.

[0003] However, in current technologies, radiation monitoring data (such as mobile robot measurements and fixed stations), meteorological data (wind speed, wind direction, and stability), and geographic information (topography and buildings) are usually processed independently, lacking a spatiotemporal alignment mechanism. This leads to inconsistent prediction inputs and consequently, high source strength inversion errors. Furthermore, traditional Gaussian diffusion models use fixed parameters and cannot dynamically respond to abrupt changes in atmospheric stability, resulting in significant boundary prediction biases.

[0004] Therefore, there is an urgent need for a new inversion method to investigate the spread of radioactive materials from nuclear leaks in order to solve the problems in existing technologies. Summary of the Invention

[0005] The purpose of this application is to address at least one of the aforementioned technical deficiencies.

[0006] On the one hand, embodiments of this application provide an inversion method for the diffusion pattern of radioactive materials in a nuclear leak, the method comprising: The initial source term parameters and meteorological data corresponding to the nuclear leakage area are obtained, and the predicted values ​​of nuclide concentration are obtained through the Gaussian plume model based on the initial source term parameters and meteorological data. The initial source term parameters are determined based on the Gaussian diffusion model. The observed values ​​of nuclide concentrations corresponding to the nuclear leakage area are obtained, and the predicted and observed values ​​of nuclide concentrations are fused using Kalman filtering to obtain the optimal nuclide concentration field and the distribution of analysis errors. The parameters of the Gaussian diffusion model are adjusted according to the analysis error distribution to obtain the adjusted Gaussian diffusion model. Based on the adjusted Gaussian diffusion model and the optimal concentration field, the target source term parameters are obtained. The target source term parameter is used as the initial source term parameter for iteration until the preset requirements are met, and the source term parameter obtained at the end of the iteration is used as the final source term parameter.

[0007] Optionally, obtain the initial source term parameters corresponding to the nuclear leakage area, including: Obtain radiation data and historical meteorological data corresponding to the nuclear leak area. The radiation data includes dose rate, latitude and longitude of the collected dose rate, and timestamp. The radiation data is input into the preset nuclide immersion irradiation model to obtain the target nuclide concentration observation value. The initial source term parameters are obtained by inversion calculation based on historical meteorological data and the target nuclide concentration observation value.

[0008] Optionally, initial source term parameters can be obtained by inversion calculation based on historical meteorological data and observed concentrations of the target nuclide, including: The inversion calculation parameters are determined based on historical meteorological data and target nuclide concentration observations. The inversion calculation parameters include the differential gradient of equipment detection data and the Gaussian diffusion model. Inversion calculations are performed based on the inversion calculation parameters to obtain the initial source term parameters.

[0009] Optionally, the inversion calculation parameters are determined based on historical meteorological data and observed concentrations of the target nuclide, including: Based on historical meteorological data and atmospheric stability level tables, the diffusion coefficient was determined, and a Gaussian diffusion model was constructed based on the diffusion coefficient. Obtain the guessed source term parameters, and based on the guessed source term parameters and the Gaussian diffusion mode, obtain the predicted concentration of each spatial point, and determine the residual value according to the observed value of the target nuclide concentration and the predicted concentration of each spatial point; Spatial gradient calculation of the residuals is performed based on the central difference method to obtain the difference gradient of the equipment detection data.

[0010] Optionally, inversion calculations are performed based on the inversion calculation parameters to obtain the initial source term parameters, including: Based on the device detection data, the guessed source term parameters are iteratively optimized using the gradient descent method until the convergence condition is met, thus obtaining the initial source term parameters.

[0011] Optionally, obtain the observed values ​​of nuclide concentrations corresponding to the nuclear leakage area, including: Obtain the latest radiation and meteorological data from the nuclear leak area. The latest radiation data includes the latest dose rate, the latitude and longitude of the latest dose rate, and the timestamp. The latest collected radiation data is input into the preset nuclide immersion irradiation model to obtain the observed nuclide concentration values.

[0012] Optionally, Kalman filtering is performed to fuse the predicted and observed nuclide concentrations to obtain the optimal nuclide concentration field and the analysis error distribution, including: Obtain the predefined observation operator matrix, prediction error covariance matrix, and observation error covariance matrix; The Kalman gain matrix is ​​determined based on the observation operator matrix, the prediction error covariance matrix, and the observation error covariance matrix. Based on the Kalman gain matrix, predicted nuclide concentrations, and observed nuclide concentrations, the optimal nuclide concentration field is obtained. The analysis error distribution is obtained based on the Kalman gain matrix, the observation operator matrix, and the prediction error covariance.

[0013] Optionally, the parameters of the Gaussian diffusion model are adjusted according to the analysis error distribution to obtain an adjusted Gaussian diffusion model, including: The gradient of the analysis error distribution is determined, and the parameters of the Gaussian diffusion mode are iteratively adjusted using the gradient descent method based on the gradient to obtain the adjusted Gaussian diffusion mode.

[0014] Optional meteorological data include wind speed and direction, and source term parameters include the source strength, release location, and effective altitude of the radioactive material.

[0015] On the other hand, embodiments of this application provide an inversion device for the diffusion pattern of radioactive materials in a nuclear leak, comprising: The nuclide concentration prediction module is used to acquire the initial source term parameters and meteorological data corresponding to the nuclear leakage area, and to obtain the predicted nuclide concentration value based on the initial source term parameters and meteorological data through the Gaussian plume model. The initial source term parameters are determined based on the Gaussian diffusion model. The data processing module is used to acquire the observed values ​​of nuclide concentrations corresponding to the nuclear leakage area, and to perform Kalman filtering fusion processing on the predicted values ​​and observed values ​​of nuclide concentrations to obtain the optimal nuclide concentration field and the distribution of analysis errors. The parameter adjustment module is used to adjust the parameters of the Gaussian diffusion model according to the analysis error distribution to obtain the adjusted Gaussian diffusion model, and to obtain the target source term parameters based on the adjusted Gaussian diffusion model and the optimal concentration field. The model iteration module is used to iterate the target source term parameters as initial source term parameters until the preset requirements are met, and the source term parameters obtained at the end of the iteration are used as the final source term parameters.

[0016] In another aspect, embodiments of this application provide an electronic device, including a processor and a memory: The memory is configured to store machine-readable instructions that, when executed by the processor, cause the processor to perform any of the methods in a method for reversing the spread of radioactive material in a nuclear leak.

[0017] The beneficial effects of the technical solutions provided in this application include at least the following: In this application, radioactive material data and meteorological data can be collected. A Kalman-based data assimilation technique and a Gaussian plume model are employed. By fusing observed nuclide concentrations, source term inversion is achieved to obtain the final source term parameters. This process ensures consistency of the input prediction data, avoiding the problem of high source intensity inversion errors. Furthermore, the parameters of the Gaussian diffusion model are dynamically adjusted, thus enabling dynamic responses to abrupt changes in atmospheric stability. This improves the accuracy of predicting the extent of nuclear hazard zones, providing strong support for proposing scientific emergency protection measures and recommendations, and ultimately minimizing the harm caused by accidents.

[0018] In this application, the precise location of model error sources through gradient information can improve inversion accuracy and avoid global search, thereby increasing iteration efficiency, reducing computational load, and being compatible with discrete and non-uniform observation point distributions, making it more suitable for robot reconnaissance characteristics. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a method for retrieving the diffusion pattern of radioactive materials in a nuclear leak, provided in an embodiment of this application; Figure 2 A flowchart illustrating another method for inverting the diffusion pattern of radioactive materials in a nuclear leak, provided in an embodiment of this application. Figure 3 A schematic diagram of the structure of an inversion device for the diffusion pattern of radioactive materials in a nuclear leak, provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting the invention.

[0022] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0023] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0024] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0025] Specifically, such as Figure 1 As shown, the method may include: Step S101: Obtain the initial source term parameters and meteorological data corresponding to the nuclear leakage area, and obtain the predicted value of nuclide concentration based on the initial source term parameters and meteorological data through the Gaussian plume model. The initial source term parameters are determined based on the Gaussian diffusion model.

[0026] Optionally, when determining the hazard range of a nuclear leak area, the initial source term parameters corresponding to the nuclear leak area can be obtained first based on the Gaussian diffusion model. Then, based on the obtained meteorological data and the initial source term parameters, the predicted values ​​of nuclide concentrations corresponding to the nuclear leak area can be obtained through the Gaussian plume model.

[0027] In optional embodiments of this application, meteorological data includes wind speed and wind direction, and source term parameters include the source strength, release location, and effective altitude of the radioactive material.

[0028] Optionally, meteorological data refers to parameters such as wind speed, wind direction, atmospheric stability (AF level), and temperature and humidity at the nuclear leak site. In practical applications, meteorological data can be collected using miniature meteorological sensors (anemometers, thermometers, hygrometers, and barometers, etc.) integrated on the top of the robotic vehicle. Source term parameters refer to the source strength, release location, and effective altitude of the radioactive material. The source strength specifically refers to the release rate of the radioactive material, and the release location can be expressed using latitude and longitude. In practical applications, the hazard range corresponding to the nuclear leak area can be predicted based on the source term parameters, and subsequent work can then be carried out based on the obtained hazard range.

[0029] The formula for the Gaussian smoke model is expressed as follows:

[0030] in, These represent the coordinates of the calculation point relative to the leak point. Indicates in t At any point in time and space （ ） The predicted nuclide concentration at the release location (in the source term parameters) is given, with the leak point at (0, 0, 0). Q This represents the source strength in the initial source term parameters. u This represents the average wind speed at the source elevation. This represents the atmospheric lateral diffusion coefficient. σ z Indicates the atmospheric vertical diffusion coefficient. H This indicates the effective height of the source strength in the initial term parameters. This is a radioactive decay term. f q For deposition attenuation term, Λ The wet deposition coefficient is... The settling velocity is the dry deposition velocity of aerosol particles.

[0031] Optional, It satisfies the radionuclide decay formula. f q The deposition attenuation term satisfies the following formula:

[0032] Λ The wet deposition coefficient satisfies the following formula:

[0033] in, I The value represents the rainfall intensity (mm / h). a and b are empirical coefficients, assigned values ​​based on whether the released substance contains iodine or not. For example, for iodine-containing substances, a can be taken as a = 8 × 10⁻⁶. -5b = 0.6, while for substances without iodine, a = 1.2 × 10⁻⁶. -4 b=0.5.

[0034] Optional, due to σ x , σ y and σ z Determined by atmospheric stability, which is categorized into six classes (AF) from highly unstable to stable. The specific atmospheric stability can be determined by referring to the table below based on the wind speed data obtained from the meteorological data. Then, based on the corresponding atmospheric stability... σ x , σ y (s y ) and σ z (s z ) The calculation formula in Table 2 is used to obtain the result.

[0035] Table 1. Correspondence of Atmospheric Stability

[0036] Table 2 Calculation Formulas

[0037] In an optional embodiment of this application, obtaining the initial source term parameters corresponding to the nuclear leakage region includes: Obtain radiation data and historical meteorological data corresponding to the nuclear leak area. The radiation data includes dose rate, latitude and longitude of the collected dose rate, and timestamp. The radiation data is input into the preset nuclide immersion irradiation model to obtain the target nuclide concentration observation value. The initial source term parameters are obtained by inversion calculation based on historical meteorological data and the target nuclide concentration observation value.

[0038] Optionally, when determining the initial source term parameters corresponding to the nuclear leak area, radiation data corresponding to the nuclear leak area can be obtained. This radiation data specifically includes the dose rate collected by the gamma dose rate monitor mounted on the robot, as well as information such as latitude, longitude, and timestamp when the dose rate was collected.

[0039] Furthermore, the obtained radiation data can be input into a preset nuclide immersion irradiation model to obtain the observed value of the target nuclide concentration. This nuclide immersion irradiation model describes the quantitative relationship between the γ dose rate and the nuclide activity concentration, and its core formula is:

[0040] in, For γ dose rate, This refers to the activity concentration of the radionuclide. For a specific γ dose constant of a radionuclide, The decay constant of a nuclide. This is the dielectric attenuation factor.

[0041] Correspondingly, after obtaining the observed values ​​of the target nuclide concentration, the initial source term parameters can be obtained by inversion calculation based on historical meteorological data and the observed values ​​of the target nuclide concentration.

[0042] In an optional embodiment of this application, an inversion calculation is performed based on historical meteorological data and observed concentrations of the target nuclide to obtain initial source term parameters, including: The inversion calculation parameters are determined based on historical meteorological data and target nuclide concentration observations. The inversion calculation parameters include the differential gradient of equipment detection data and the Gaussian diffusion model. Inversion calculations are performed based on the inversion calculation parameters to obtain the initial source term parameters.

[0043] Optionally, the historical meteorological data refers to the meteorological data acquired in the nuclear leakage area within a set period, such as meteorological data acquired within the past three days. Accordingly, the differential gradient and Gaussian diffusion model (i.e., inversion calculation parameters) of the equipment detection data can be determined based on the determined historical meteorological data, and then inversion calculations can be performed based on the differential gradient and Gaussian diffusion model of the equipment detection data to obtain the initial source term parameters.

[0044] In an optional embodiment of this application, the inversion calculation parameters are determined based on historical meteorological data and observed concentrations of the target nuclide, including: Based on historical meteorological data and atmospheric stability level tables, the diffusion coefficient was determined, and a Gaussian diffusion model was constructed based on the diffusion coefficient. Obtain the guessed source term parameters, and based on the guessed source term parameters and the Gaussian diffusion mode, obtain the predicted concentration of each spatial point, and determine the residual value according to the observed value of the target nuclide concentration and the predicted concentration of each spatial point; Spatial gradient calculation of the residuals is performed based on the central difference method to obtain the difference gradient of the equipment detection data.

[0045] Optionally, since the diffusion coefficient is required when constructing the Gaussian diffusion model, it can be determined by referring to the atmospheric stability level table based on historical wind speed data from historical meteorological data. Then, the Gaussian diffusion model can be constructed based on the determined diffusion coefficient, specifically expressed by the following formula:

[0046] in, These are predicted values ​​for nuclide concentration. The diffusion coefficient is... To release the position, Q Indicates source strength, u This represents the average wind speed at the source elevation. H This indicates the effective height of the source strength in the initial term parameters. This represents the terrain occlusion coefficient.

[0047] Furthermore, after constructing the Gaussian diffusion model, source term parameters can be assumed (i.e., guessed source term parameters). These guessed parameters are then input into the constructed Gaussian diffusion model to obtain the predicted concentration at each spatial point. The difference between the previously determined observed concentration of the target nuclide and the predicted concentration at each spatial point is then calculated to obtain the residual value. Finally, the spatial gradient of the residual is calculated using the central difference method to obtain the differential gradient of the device detection data. If the differential gradient of the device detection data is greater than 0, its direction indicates an area underestimated by the model, suggesting that the source strength needs to be increased or the source position moved to that direction. The magnitude of the differential gradient of the device detection data represents the severity of the error change. The central difference method is expressed by the following formula:

[0048]

[0049]

[0050] in, For the device to detect the gradient of data differences, Δ x and Δ y Take the average distance between adjacent observation points of the robot. For the first i The position of each point in space.

[0051] In an optional embodiment of this application, inversion calculation is performed based on inversion calculation parameters to obtain initial source term parameters, including: Based on the device detection data, the guessed source term parameters are iteratively optimized using the gradient descent method until the convergence condition is met, thus obtaining the initial source term parameters.

[0052] Optionally, the objective function can be defined to minimize the magnitude of the residual gradient:

[0053] Furthermore, based on the defined objective function, the guessed source term parameters are iteratively optimized using the following formula until the convergence condition is met:

[0054]

[0055]

[0056] in, , , For adaptive step size, adjustments can be made based on Armijo rules.

[0057] Correspondingly, the source term parameters that satisfy the convergence condition are the initial source term parameters. The location information in the initial source term parameters is taken as the upwind position of the point of highest concentration, and the source strength is estimated using the empirical formula for maximum concentration. The convergence condition can be set according to the actual situation, such as setting... Less than the threshold or reached the maximum number of iterations, etc.

[0058] In this application, the precise location of model error sources through gradient information can improve inversion accuracy and avoid global search, thereby increasing iteration efficiency, reducing computational load, and being compatible with discrete and non-uniform observation point distributions, making it more suitable for robot reconnaissance characteristics.

[0059] Step S102: Obtain the observed values ​​of nuclide concentrations corresponding to the nuclear leakage area, and perform Kalman filtering fusion processing based on the predicted values ​​of nuclide concentrations and the observed values ​​of nuclide concentrations to obtain the optimal nuclide concentration field and the analysis error distribution.

[0060] In practical applications, due to the complexity and variability of surface conditions and nuclear contaminants themselves, dose rate measurements at nuclear facilities often result in significant errors at certain locations due to numerous surrounding interference factors, while at other locations with fewer interference factors, the observed values ​​exhibit a non-uniform error distribution. Furthermore, the predicted values ​​calculated using the Gaussian model also exhibit a non-uniform error distribution. Based on this, this application obtains the observed nuclide concentration values ​​corresponding to the nuclear leak area, then performs Kalman filtering fusion processing on the predicted and observed nuclide concentration values ​​to obtain the optimal nuclide concentration field and analytical error distribution. Based on the optimal nuclide concentration field and analytical error distribution, the diffusion process of the nuclide is corrected accordingly, thereby reducing errors.

[0061] In this application, in optional embodiments, obtaining the observed nuclide concentration values ​​corresponding to the nuclear leakage area includes: Obtain the latest radiation and meteorological data from the nuclear leak area. The latest radiation data includes the latest dose rate, the latitude and longitude of the latest dose rate, and the timestamp. The latest collected radiation data is input into the preset nuclide immersion irradiation model to obtain the observed nuclide concentration values.

[0062] Optionally, the latest radiation and meteorological data for the nuclear leak area can be collected, and then the latest collected radiation data can be input into a preset nuclide immersion irradiation model to obtain the observed nuclide concentration values ​​corresponding to the nuclear leak area. The specific method for obtaining the observed nuclide concentration values ​​based on the nuclide immersion irradiation model has been described above and will not be repeated here.

[0063] In an optional embodiment of this application, Kalman filtering fusion processing is performed based on the predicted and observed nuclide concentrations to obtain the optimal nuclide concentration field and analysis error distribution, including: Obtain the predefined observation operator matrix, prediction error covariance matrix, and observation error covariance matrix; The Kalman gain matrix is ​​determined based on the observation operator matrix, the prediction error covariance matrix, and the observation error covariance matrix. Based on the Kalman gain matrix, predicted nuclide concentrations, and observed nuclide concentrations, the optimal nuclide concentration field is obtained. The analysis error distribution is obtained based on the Kalman gain matrix, the observation operator matrix, and the prediction error covariance.

[0064] Optionally, the Kalman filter fusion process is divided into two stages: prediction and update. The posterior estimate of the state variable is obtained by weighted averaging the background estimate and the current observation. Then, based on minimizing the mean square error of the posterior estimate, the weight coefficients or matrix are derived, and the prediction and update process is performed iteratively. Each time step can be divided into two steps: time update and state update, which are performed iteratively.

[0065] In practical applications, predefined observation operator matrix, prediction error covariance matrix, and observation error covariance matrix can be obtained. Then, based on the observation operator matrix, prediction error covariance matrix, and observation error covariance matrix, the Kalman gain matrix can be determined using the following formula.

[0066]

[0067] in, Here is the Kalman gain matrix. For the prediction error covariance matrix, To observe the operator matrix, Let be the observation error covariance matrix.

[0068] Accordingly, by substituting the Kalman gain matrix, the predicted nuclide concentration, and the observed nuclide concentration into the following formula, the optimal nuclide concentration field is obtained.

[0069]

[0070] in, This represents the optimal concentration field for nuclides. These are predicted values ​​for nuclide concentration. These are observed values ​​for nuclide concentration. To observe the operator matrix, This is the Kalman gain matrix.

[0071] Meanwhile, the Kalman gain matrix, prediction error covariance matrix, and observation operator matrix values ​​can be substituted into the following formula to obtain the analysis error distribution.

[0072]

[0073] in, To analyze the error distribution, It is the identity matrix. To observe the operator matrix, Here is the Kalman gain matrix. This is the prediction error covariance matrix.

[0074] Step S103: Adjust the parameters of the Gaussian diffusion model according to the analysis error distribution to obtain the adjusted Gaussian diffusion model, and obtain the target source term parameters based on the adjusted Gaussian diffusion model and the optimal concentration field.

[0075] Optionally, the analysis error distribution characterizes the uncertainty of the concentration field after quantization and assimilation. In this case, the parameters of the Gaussian diffusion mode can be adjusted according to the analysis error distribution to obtain the adjusted Gaussian diffusion mode. Then, the optimal concentration field is input into the adjusted Gaussian diffusion mode and the optimal concentration field to obtain the target source term parameters.

[0076] In an optional embodiment of this application, the parameters of the Gaussian diffusion mode are adjusted according to the analysis error distribution to obtain an adjusted Gaussian diffusion mode, including: The gradient of the analysis error distribution is determined, and the parameters of the Gaussian diffusion mode are iteratively adjusted using the gradient descent method based on the gradient to obtain the adjusted Gaussian diffusion mode.

[0077] Optionally, the following objective function can be defined to minimize the trace of the analysis error covariance matrix.

[0078]

[0079] in, To analyze the error distribution, For diffusion parameters, For the source's effective height, The settling velocity is the dry deposition velocity of aerosol particles.

[0080] Furthermore, the gradient of the objective function with respect to the parameters of the Gaussian diffusion model is calculated using the adjoint mode or the finite difference method. The parameters of the Gaussian diffusion model are then iteratively adjusted based on the update rule to obtain the adjusted parameters. Finally, the Gaussian diffusion model is constructed based on these adjusted parameters. The update rule is expressed as follows:

[0081] in, For adaptive step size, For the first The parameter values ​​obtained in the next iteration For the first The parameter values ​​obtained in the next iteration The gradient of the objective function with respect to the parameters of the Gaussian diffusion mode.

[0082] Step S104: Iterate the target source term parameters as initial source term parameters until the preset requirements are met, and use the source term parameters obtained at the end of the iteration as the final source term parameters.

[0083] Optionally, the target source term parameter can be used as the initial source term parameter mentioned above to execute the method provided in this application again until the source term parameter meets the preset requirements. At this point, the source term parameter obtained when the iteration ends is the final source term parameter.

[0084] In this application, radioactive material data and meteorological data can be collected. A Kalman-based data assimilation technique and a Gaussian plume model are employed. By fusing observed nuclide concentrations, source term inversion is achieved to obtain the final source term parameters. This process ensures consistency of the input prediction data, avoiding the problem of high source intensity inversion errors. Furthermore, the parameters of the Gaussian diffusion model are dynamically adjusted, thus enabling dynamic responses to abrupt changes in atmospheric stability. This improves the accuracy of predicting the extent of nuclear hazard zones, providing strong support for proposing scientific emergency protection measures and recommendations, and ultimately minimizing the harm caused by accidents.

[0085] Optionally, to better understand the method provided in this application, the following will be combined with... Figure 2 The method provided in this application will be explained again. Specifically: The reconnaissance robot collects data at nuclear monitoring nodes. This data includes information such as dose rate, latitude and longitude, and timestamps. Based on the known nuclide immersion irradiation model (i.e., the nuclide analysis types / known types in the figure), the nuclide concentration is deduced from the detected dose rate data. Then, source term inversion is achieved using source term observation data and historical meteorological data (using the difference gradient of equipment detection data (the dose rate difference gradient in the figure) and Gaussian model back-calculation). Based on the source term inversion results and meteorological data, the nuclide concentration at the next moment is positively predicted using a multi-Gaussian plume model. This prediction is then fused with the observed nuclide concentration at the next moment, and the data is assimilated using the Kalman filter method. By analyzing the error distribution pattern, the Gaussian model parameters are adjusted to calculate the source term results for the next round. The process is iterated until the optimal source term parameters are obtained. At this point, the nuclear radiation situation and range can be determined based on the optimal source term parameters.

[0086] Optionally, this application also provides a situation inversion and analysis system, which can collect, fuse, and display information on radioactive materials. Utilizing GIS combined with environmental information, it can conduct numerical simulations of radioactive material diffusion based on a virtual geographic environment for sudden nuclear facility leakage accidents. This situation inversion and analysis system mainly consists of the following parts: a) Data layer: The situation fusion and control system requires various data support, including three-dimensional terrain data, building data, road data, radioactive monitoring data, meteorological data (wind speed, wind direction), etc. These data are the basis for simulating the spread of nuclear materials in sudden events.

[0087] b) Interface Layer: The interface layer organizes and manages the data provided by the data layer. The GIS provides typical terrain features and data rendering interfaces, the database interface manages nuclide features, etc., the wireless data transmission interface collects real-time monitoring data (including nuclear radiation monitoring data and meteorological data) from each wireless monitoring point, and the Gaussian multi-puff model diffusion calculation library provides calculation methods for airborne radioactive diffusion models.

[0088] c) Logical Layer: The computational domain of the Gaussian diffusion model is meshed. Using the Gaussian diffusion model calculation formula and library, the sudden leakage and diffusion of nuclear materials are simulated. The nuclear radiation dose rate (concentration) is calculated and visualized on a 2D / 3D GIS (contour lines, isosurfaces, etc.). Data assimilation is used to integrate front-end observation data, reducing the prediction error of nuclear hazard diffusion.

[0089] d) Application layer: Using interactive customization or automatic data loading to start up, it realizes the integration of nuclear emergency monitoring information and regional hazard assessment, and realizes application function modules such as nuclear radiation dose rate (concentration) information display, prediction-observation data assimilation, information fusion and situation display, nuclear pollution source inversion, and nuclear hazard assessment.

[0090] This application provides an embodiment of a device for inverting the diffusion pattern of radioactive materials from a nuclear leak, such as... Figure 3 As shown, the device may include: a nuclide concentration prediction module 301, a data processing module 302, a parameter adjustment module 303, and a model iteration module 304, wherein, The nuclide concentration prediction module is used to acquire the initial source term parameters and meteorological data corresponding to the nuclear leakage area, and to obtain the predicted nuclide concentration value based on the initial source term parameters and meteorological data through the Gaussian plume model. The initial source term parameters are determined based on the Gaussian diffusion model. The data processing module is used to acquire the observed values ​​of nuclide concentrations corresponding to the nuclear leakage area, and to perform Kalman filtering fusion processing on the predicted values ​​and observed values ​​of nuclide concentrations to obtain the optimal nuclide concentration field and the distribution of analysis errors. The parameter adjustment module is used to adjust the parameters of the Gaussian diffusion model according to the analysis error distribution to obtain the adjusted Gaussian diffusion model, and to obtain the target source term parameters based on the adjusted Gaussian diffusion model and the optimal concentration field. The model iteration module is used to iterate the target source term parameters as initial source term parameters until the preset requirements are met, and the source term parameters obtained at the end of the iteration are used as the final source term parameters.

[0091] Optionally, when acquiring the initial source term parameters corresponding to the nuclear leakage region, the nuclide concentration prediction module is specifically used for: Obtain radiation data and historical meteorological data corresponding to the nuclear leak area. The radiation data includes dose rate, latitude and longitude of the collected dose rate, and timestamp. The radiation data is input into the preset nuclide immersion irradiation model to obtain the target nuclide concentration observation value. The initial source term parameters are obtained by inversion calculation based on historical meteorological data and the target nuclide concentration observation value.

[0092] Optionally, when the nuclide concentration prediction module performs inversion calculations based on historical meteorological data and observed target nuclide concentrations to obtain initial source term parameters, it is specifically used for: The inversion calculation parameters are determined based on historical meteorological data and target nuclide concentration observations. The inversion calculation parameters include the differential gradient of equipment detection data and the Gaussian diffusion model. Inversion calculations are performed based on the inversion calculation parameters to obtain the initial source term parameters.

[0093] Optionally, the nuclide concentration prediction module, when determining the inversion calculation parameters based on historical meteorological data and observed target nuclide concentrations, is specifically used for: Based on historical meteorological data and atmospheric stability level tables, the diffusion coefficient was determined, and a Gaussian diffusion model was constructed based on the diffusion coefficient. Obtain the guessed source term parameters, and based on the guessed source term parameters and the Gaussian diffusion mode, obtain the predicted concentration of each spatial point, and determine the residual value according to the observed value of the target nuclide concentration and the predicted concentration of each spatial point; Spatial gradient calculation of the residuals is performed based on the central difference method to obtain the difference gradient of the equipment detection data.

[0094] Optionally, when the nuclide concentration prediction module performs inversion calculations based on the inversion calculation parameters to obtain the initial source term parameters, it is specifically used for: Based on the device detection data, the guessed source term parameters are iteratively optimized using the gradient descent method until the convergence condition is met, thus obtaining the initial source term parameters.

[0095] Optionally, when acquiring the observed nuclide concentration values ​​corresponding to the nuclear leakage area, the data processing module is specifically used for: Obtain the latest radiation and meteorological data from the nuclear leak area. The latest radiation data includes the latest dose rate, the latitude and longitude of the latest dose rate, and the timestamp. The latest collected radiation data is input into the preset nuclide immersion irradiation model to obtain the observed nuclide concentration values.

[0096] Optionally, when the data processing module performs Kalman filtering fusion processing on the predicted and observed nuclide concentrations to obtain the optimal nuclide concentration field and analysis error distribution, it is specifically used for: Obtain the predefined observation operator matrix, prediction error covariance matrix, and observation error covariance matrix; The Kalman gain matrix is ​​determined based on the observation operator matrix, the prediction error covariance matrix, and the observation error covariance matrix. Based on the Kalman gain matrix, predicted nuclide concentrations, and observed nuclide concentrations, the optimal nuclide concentration field is obtained. The analysis error distribution is obtained based on the Kalman gain matrix, the observation operator matrix, and the prediction error covariance.

[0097] Optionally, when the parameter adjustment module adjusts the parameters of the Gaussian diffusion mode according to the analysis error distribution to obtain the adjusted Gaussian diffusion mode, it is specifically used for: The gradient of the analysis error distribution is determined, and the parameters of the Gaussian diffusion mode are iteratively adjusted using the gradient descent method based on the gradient to obtain the adjusted Gaussian diffusion mode.

[0098] Optional meteorological data include wind speed and direction, and source term parameters include the source strength, release location, and effective altitude of the radioactive material.

[0099] The inversion device for the diffusion status of radioactive materials in a nuclear leak according to this embodiment can execute the inversion method for the diffusion status of radioactive materials in a nuclear leak as shown in the embodiment of this application. The implementation principle is similar and will not be described again here.

[0100] This application provides an electronic device, which includes: a processor; and a memory configured to store machine-readable instructions that, when executed by the processor, cause the processor to perform a method for reversing the spread of radioactive materials in a nuclear leak.

[0101] This application provides an electronic device, such as... Figure 4 As shown, Figure 4 The illustrated electronic device includes a processor 2001 and a memory 2003. The processor 2001 and the memory 2003 are connected, for example, via a bus 2002. Optionally, the electronic device 2000 may further include a transceiver 2004. It should be noted that in practical applications, the transceiver 2004 is not limited to one type, and the structure of this electronic device 2000 does not constitute a limitation on the embodiments of this application.

[0102] Processor 2001 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 2001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0103] Bus 2002 may include a pathway for transmitting information between the aforementioned components. Bus 2002 may be a PCI bus or an EISA bus, etc. Bus 2002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0104] The memory 2003 may be ROM or other type of static storage device capable of storing static information and instructions, RAM or other type of dynamic storage device capable of storing information and instructions, or EEPROM, CD-ROM or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0105] The memory 2003 stores the application code that executes the scheme of this application, and its execution is controlled by the processor 2001. The processor 2001 executes the application code stored in the memory 2003 to implement... Figure 3 The illustrated embodiment provides the operation of an inversion device for the diffusion pattern of radioactive materials in a nuclear leak.

[0106] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0107] The above description is only a partial embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for inverting the diffusion pattern of radioactive materials from a nuclear leak, characterized in that, include: The initial source term parameters and meteorological data corresponding to the nuclear leakage area are obtained, and the predicted values ​​of nuclide concentration are obtained by using the Gaussian plume model based on the initial source term parameters and the meteorological data. The initial source term parameters are determined based on the Gaussian diffusion model. The observed values ​​of nuclide concentrations corresponding to the nuclear leakage area are obtained, and Kalman filtering fusion processing is performed on the predicted values ​​of nuclide concentrations and the observed values ​​of nuclide concentrations to obtain the optimal nuclide concentration field and the analysis error distribution. The parameters of the Gaussian diffusion mode are adjusted according to the analysis error distribution to obtain the adjusted Gaussian diffusion mode, and the target source term parameters are obtained based on the adjusted Gaussian diffusion mode and the optimal concentration field. The target source term parameter is used as the initial source term parameter and iterated until the preset requirement is met. The source term parameter obtained at the end of the iteration is used as the final source term parameter.

2. The method according to claim 1, characterized in that, The acquisition of the initial source term parameters corresponding to the nuclear leakage area includes: The radiation data and historical meteorological data corresponding to the nuclear leak area are obtained. The radiation data includes the dose rate, the latitude and longitude of the dose rate collection, and the timestamp. The radiation data is input into a preset nuclide immersion irradiation model to obtain the target nuclide concentration observation value. The initial source term parameters are obtained by inversion calculation based on the historical meteorological data and the target nuclide concentration observation value.

3. The method according to claim 2, characterized in that, The initial source term parameters are obtained by inversion calculation based on the historical meteorological data and the observed concentration of the target nuclide, including: The inversion calculation parameters are determined based on the historical meteorological data and the observed concentration of the target nuclide. The inversion calculation parameters include the differential gradient of the equipment detection data and the Gaussian diffusion mode. The initial source term parameters are obtained by performing inversion calculations based on the inversion calculation parameters.

4. The method according to claim 3, characterized in that, The process of determining the inversion calculation parameters based on the historical meteorological data and the observed concentrations of the target nuclide includes: Based on the historical meteorological data and atmospheric stability level table, the diffusion coefficient is determined, and a Gaussian diffusion model is constructed based on the diffusion coefficient. Obtain the guessed source term parameters, and based on the guessed source term parameters and the Gaussian diffusion mode, obtain the predicted concentration of each spatial point, and determine the residual value according to the observed value of the target nuclide concentration and the predicted concentration of each spatial point; The spatial gradient of the residual is calculated based on the central difference method to obtain the difference gradient of the device detection data.

5. The method according to claim 4, characterized in that, The step of performing inversion calculations based on the inversion calculation parameters to obtain the initial source term parameters includes: Based on the device detection data, the guessed source term parameters are iteratively optimized using the gradient descent method until the convergence condition is met, thus obtaining the initial source term parameters.

6. The method according to claim 1, characterized in that, The process of obtaining the observed nuclide concentration values ​​corresponding to the nuclear leakage area includes: The latest radiation and meteorological data collected in the nuclear leak area are obtained. The latest radiation data includes the latest dose rate, the latitude and longitude of the latest dose rate, and the timestamp. The newly acquired radiation data is input into a preset nuclide immersion irradiation model to obtain the observed nuclide concentration.

7. The method according to claim 1, characterized in that, The step of performing Kalman filtering fusion processing based on the predicted and observed nuclide concentrations to obtain the optimal nuclide concentration field and analysis error distribution includes: Obtain the predefined observation operator matrix, prediction error covariance matrix, and observation error covariance matrix; The Kalman gain matrix is ​​determined based on the observation operator matrix, the prediction error covariance matrix, and the observation error covariance matrix. Based on the Kalman gain matrix, the predicted nuclide concentration, and the observed nuclide concentration, the optimal nuclide concentration field is obtained; The analysis error distribution is obtained based on the Kalman gain matrix, the observation operator matrix, and the prediction error covariance.

8. The method according to claim 1, characterized in that, The step of adjusting the parameters of the Gaussian diffusion mode according to the analysis error distribution to obtain the adjusted Gaussian diffusion mode includes: The gradient of the analysis error distribution is determined, and the parameters of the Gaussian diffusion mode are iteratively adjusted based on the gradient using the gradient descent method to obtain the adjusted Gaussian diffusion mode.

9. The method according to claim 1, characterized in that, The meteorological data includes wind speed and wind direction, and the source term parameters include the source strength, release location, and effective altitude of the radioactive material.

10. An inversion device for the diffusion pattern of radioactive materials in a nuclear leak, characterized in that, include: The nuclide concentration prediction module is used to acquire the initial source term parameters and meteorological data corresponding to the nuclear leakage area, and to obtain the predicted nuclide concentration value based on the initial source term parameters and the meteorological data through a Gaussian plume model. The initial source term parameters are determined based on the Gaussian diffusion model. The data processing module is used to acquire the observed values ​​of nuclide concentrations corresponding to the nuclear leakage area, and to perform Kalman filtering fusion processing on the predicted values ​​of nuclide concentrations and the observed values ​​of nuclide concentrations to obtain the optimal nuclide concentration field and the analysis error distribution. The parameter adjustment module is used to adjust the parameters of the Gaussian diffusion mode according to the analysis error distribution to obtain the adjusted Gaussian diffusion mode, and to obtain the target source term parameters based on the adjusted Gaussian diffusion mode and the optimal concentration field. The model iteration module is used to iterate the target source term parameters as the initial source term parameters until the preset requirements are met, and to use the source term parameters obtained at the end of the iteration as the final source term parameters.

Images