Segmented smooth inversion method and system for radionuclide release timing

Abstract

The application relates to a segmented smoothing inversion method and system for radionuclide release timing, which comprises the following steps: S1, inputting and normalizing a transport matrix and an observation vector; S2, initializing a correction coefficient vector and a source term leakage rate sigma n ; S3, inputting iteration number N max , a regularization coefficient lambda, an L1 regularization term coefficient alpha and convergence conditions kernel tol and x tol; S4, updating a diagonal correction coefficient matrix W m×m ; S5, updating a source term leakage rate sigma k‑1 of a kth step by using a diagonal correction coefficient matrix W k of a k-1th step; S6, updating a correction coefficient vector of the kth step by using the source term leakage rate sigma k of the kth step; S7, processing a center distance of the correction coefficient vector; S8, judging whether an iteration condition is met; and S9, performing de-normalization on a leakage rate vector sigma n obtained through iteration to obtain a segmented smoothing source term leakage rate. The application can directly perform ideal segmented smoothing source term inversion from sparse monitoring data.

Description

Technical Field

[0001] The present invention relates to the field of nuclear application technology, and in particular to a piecewise smooth inversion method and system for radionuclide release time series. Background Art

[0002] When a radionuclide (source term) leaks, the temporal variation of its leakage rate is crucial information for assessing its consequences. In accident scenarios, such as those seen in the Chernobyl and Fukushima nuclear accidents, due to limited and uncertain temporal and spatial information about meteorological conditions, leak sources, and concentration monitoring, it is impossible to directly reconstruct the time-series continuous source term leakage rate. Instead, a piecewise-smoothed leakage rate estimate can only be approximated. This form of piecewise smoothing assumes that the instantaneous leakage rate is constant within a limited time window, allowing for sharp jumps or drops at the window's boundaries. While piecewise smoothing may lose some detail about source term releases, it better aligns with various environmental monitoring results during accident scenarios and is widely accepted. However, for nuclear accidents, obtaining such piecewise-smoothed leakage rate curves has, to date, only been possible through expert decision-making, requiring extensive manual effort and subjectivity and uncertainty.

[0003] Inversion methods estimate the temporal variation of leakage rates by minimizing the discrepancy between atmospheric diffusion model predictions and environmental monitoring. Their mathematical framework is clearer and their results more objective. However, because the inversion of atmospheric emissions of radionuclides from nuclear accidents is an ill-posed problem, the estimated leakage rates exhibit numerous spurious oscillations, lack physical meaning, and lack piecewise smoothness, presenting a core bottleneck for current inversion methods.

[0004] The inadequacy of environmental monitoring and the uncertainty of the atmospheric diffusion model are the core reasons for the above-mentioned defects, and are also two issues that need to be urgently addressed in the inversion. Existing inversion methods mainly use regularization techniques to incorporate rough guesses of leakage rates as priors into the inversion process to alleviate pathological conditions. However, this prior information is difficult to construct accurately and has large inherent biases, which is insufficient to fill the gaps in environmental monitoring data, resulting in a large amount of oscillation in the solution results. On the other hand, the uncertainty of the atmospheric diffusion model has been recognized as unavoidable, and existing inversion methods are unable to correct it, which will amplify and exacerbate the oscillation phenomenon. The above two reasons make it impossible to obtain a piecewise smooth leakage rate estimate through inversion. Summary of the Invention

[0005] In view of the above problems, the purpose of the present invention is to provide a radionuclide release time series inversion method and system that can make the time series profile of the leakage rate have piecewise smooth characteristics and be more consistent with the actual situation.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present application provides a segmented smoothing inversion method for radionuclide release timing, comprising:

[0008] S1, input and normalize the transport matrix and the observation vector;

[0009] S2, initialize the correction coefficient vector and the source term leakage rate n , n represents the time step of simulation;

[0010] S3, input the iteration number N max , the regularization coefficient λ, the L1 regularization term coefficient α, and the convergence conditions kernel tol and x tol;

[0011] S4, update the diagonal correction coefficient matrix W m×m , m is the data size provided by the measuring point in the complete simulation period;

[0012] S5, update the source term leakage rate k-1 of the kth step using the diagonal correction coefficient matrix W k of the k-1th step;

[0013] S6, update the correction coefficient vector k of the kth step using the source term leakage rate

[0014] S7, normalize the center distance of the correction coefficient vector;

[0015] S8, judge whether the relative error of the correction coefficient vectors of the kth step and the k-1th step reaches the convergence condition kernel tol, if it reaches the convergence condition, enter S9; if it does not reach, judge whether the relative error of the leakage rate estimation value max of the kth step and the k-1th step reaches the convergence condition x tol, if it reaches the convergence condition, enter S9; if it does not reach the convergence condition, further judge whether the iteration number N

[0016] S9, perform anti-normalization on the leakage rate vector n obtained by iteration to obtain the source term inversion result based on the dilution observation input, that is, the segmented smoothing source term leakage rate u.

[0017] Further, the specific process of S1 above comprises:

[0018] The normalized representation of the transport matrix H m×n is H m×n / coef;

[0019] The observation vector yobs m The normalized representation of y obs m / coef;

[0020] Among them, the normalization factor coef takes the observation vector y obs m The largest element in absolute value.

[0021] Furthermore, the specific process of the above S2 includes:

[0022] Initialize the correction coefficient vector That is, all its elements are set to 1;

[0023] Initialize the source term leakage rate vector σ n , that is, setting all elements in the vector to 0.

[0024] Furthermore, the convergence condition kerneltol in the above S3 is determined by the formula:

[0025] Correction coefficient vectors for the kth and k-1th steps The relative error When it is less than kerneltol, the iteration terminates, kerneltol = 1e-5.

[0026] Furthermore, the convergence condition xtol in the above S3 is determined by the formula:

[0027] The relative error of the leakage rate estimation value σ between the kth step and the k-1th step When it is less than xtol, the iteration terminates, xtol = 1e-4.

[0028] Furthermore, the above S5 updates the k-th step source term leakage rate σ k The formula is:

[0029]

[0030] Where ε is the update step size; H is the transport matrix; f is the observation value; is the derivative operator, Corresponding to the L1-regularization term, corresponds to the total variation regularization term.

[0031] Furthermore, the above S7 normalizes the center distance of the correction coefficient vector using the MCD method, specifically:

[0032]

[0033]

[0034]

[0035] In a second aspect, the present invention further provides a piecewise smooth inversion system for radionuclide release time series, the system comprising:

[0036] The parameter input unit is configured to input the transport matrix, observation vector, and number of iterations N max , regularization coefficient λ, L1 regularization term coefficient α and convergence conditions kerneltol and xtol;

[0037] A parameter initialization unit configured to initialize the correction coefficient vector and the source term leakage rate σ n , n represents the time step of simulation;

[0038] A diagonal correction coefficient updating unit configured to update the diagonal correction coefficient matrix W m×m , m is the amount of data provided by the measurement points during the complete simulation period;

[0039] The source term leakage rate update unit is configured to use the k-1 step diagonal correction coefficient matrix W k-1 Update the k-th step source leakage rate σ k ;

[0040] The correction coefficient vector updating unit is configured to use the k-th step source term leakage rate σ k Update the k-th step correction coefficient vector

[0041] a correction coefficient vector processing unit configured to normalize the center distance of the correction coefficient vector;

[0042] An iteration judgment unit, configured to judge whether a set iteration condition is met;

[0043] The source term leakage rate curve u calculation unit is configured to iterate the leakage rate vector σ n After performing inverse normalization, the source term inversion result based on the diluted observation input is obtained, that is, the piecewise smoothed source term leakage rate u is obtained.

[0044] In a third aspect, the present invention also provides a processing device, which includes at least a processor and a memory, wherein a computer program is stored in the memory, and when the processor runs the computer program, it executes the segmented smooth inversion method of the radionuclide release time series.

[0045] In a fourth aspect, the present invention further provides a computer storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions can be executed by a processor to implement the piecewise smooth inversion method of the radionuclide release time series.

[0046] The present invention has the following advantages due to the adoption of the above technical solution:

[0047] 1. Due to the use of dual sparsity constraints, the present invention can compensate for the information loss problem caused by insufficient monitoring data and improve the piecewise smoothness characteristics of the estimated source term, which includes the L1 regularization term and the total variation regularization term. If the vector σ is used to represent the source term leakage rate of the entire period, in the case of instantaneous leakage, the leakage rate has non-zero values ​​only at a few moments; in the case of smooth leakage, the time gradient of the leakage rate has only a few non-zero values; the above-mentioned sparsity characteristics correspond to the 1-norm ||σ||1 of the leakage rate vector and its time gradient 1-norm of Therefore, when inverting the leakage rate vector σ, the ||σ||1 and It can improve the sparsity of the leakage rate vector and compensate for the incompleteness of the data;

[0048] 2. This paper uses a joint approach to reduce the oscillation of the estimation results caused by the uncertainty of the atmospheric diffusion model itself, using a diagonal correction coefficient matrix. Where W ≥ 0 and its central moment is a constant, this diagonal correction coefficient matrix does not require additional prior input and is only updated by observations and subsequent iterative processes, effectively reducing the complexity of the problem.

[0049] 3. The present invention uses a projection alternating minimization algorithm to stably solve the cost function by alternately solving the transport matrix and the leakage rate vector, and updates the two sub-problems with a small step size each time to reconstruct the information of the non-steady source leakage;

[0050] In summary, the present invention can obtain piecewise-smoothed source term leakage rate estimation results from sparse observation data. This is verified using the European tracer experiment scenario. The obtained total leakage amount has a small relative deviation from the true value, and the time series profile of the leakage rate is consistent with the actual situation. In particular, the initial leakage time is accurately estimated, and a relatively ideal piecewise-smoothed source term inversion can be directly performed from sparse monitoring data. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. Throughout the drawings, the same reference numerals are used to denote the same components. In the drawings:

[0052] Figure 1 4 is a flow chart of a piecewise smooth inversion method for a radionuclide release time series according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0054] It should be understood that the terms used herein are for the purpose of describing specific example embodiments only and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "comprise", "include", "contain" and "have" are inclusive and therefore specify the presence of stated features, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring them to be performed in the specific order described or illustrated, unless the order of execution is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0055] For ease of description, spatially relative terms may be used herein to describe the relationship of one element or feature relative to another element or feature as shown in the figures, such as "inside," "outside," "inner side," "outer side," "lower," "upper," etc. Such spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures.

[0056] The present invention combines the dual sparsity constraint and the joint method to modify the regularization term of the cost function in the traditional inversion method, and uses the L1-regularization term (α||σ||1, α is the L1 regularization term coefficient) and the total variation regularization term As a replacement. The L1-regularization term makes the non-zero values ​​of the leakage rate concentrated only in a limited time window, ensuring the inversion effect of instantaneous release; the total variation regularization term limits the frequent changes of the leakage rate in time series, thereby ensuring the inversion effect of the smooth release situation. In the problem solving process, the uncertainty of the atmospheric diffusion model is corrected iteratively, and the correction coefficient matrix and the leakage rate vector are updated in turn. The correction process utilizes a joint method, using the diagonal correction coefficient matrix W m×m To reduce the oscillation amplification phenomenon caused by the uncertainty of the atmospheric diffusion model itself, that is, to reduce the transport matrix H m×n The update process of the joint method uses the projection alternating minimization method to alternately update the source leakage rate σ n and the correction coefficient vector To approach the optimal solution to the problem, each update process uses a small step size (for example, 10 -3 ) is used to perform the iteration. After completing the iteration, the piecewise smoothed source leakage rate estimation result can be obtained.

[0057] Example 1

[0058] Based on the above implementation principle, such as Figure 1 As shown, the segmented smoothing inversion method for the radionuclide release time series provided in this embodiment includes the following steps:

[0059] S1. Input and normalize the transport matrix H m×n and the observation vector y obs m ;

[0060] Specifically, the transport matrix H m×n Produced using an atmospheric diffusion model, m represents the amount of data provided by a measurement point during a complete simulation period, and n represents the simulation time step. If the number of measurement points is num, and the number of measurements at any measurement point during a complete simulation period is f, then m = num*f.

[0061] Observation vector y obs m It is composed of real environment monitoring data. It should be noted that the arrangement order of monitoring data in the observation vector should be consistent with the arrangement order of data in any column of the transport matrix.

[0062] Normalization: The normalization factor coef takes the observation vector y obs m The largest element in absolute value.

[0063] Transport matrix H m×n The normalization of can be expressed as H m×n / coef;

[0064] Observation vector y obs m The normalization of can be expressed as y obs m / coef.

[0065] S2. Initialize the correction coefficient vector and the source term leakage rate σ n ;

[0066] Specifically, the correction coefficient vector is a parameter in the joint method, using the formula To calculate the diagonal correction coefficient matrix, the diagonal correction coefficient matrix W of the k-1 step will be used to update the source leakage rate σ of the k-th step n , and the source term leakage rate σ of the kth step nCan be used to update the k-th step

[0067] Initialize the correction coefficient vector That is, all its elements are set to 1;

[0068] Initialize the source term leakage rate vector σ n , that is, setting all elements in the vector to 0.

[0069] S3. Input the number of iterations N max , regularization coefficient λ, L1 regularization term coefficient α and convergence conditions kerneltol and xtol

[0070] Specifically, the number of iterations N max Used to limit the maximum number of iterations; the regularization coefficient λ and the L1 regularization term coefficient α are the parameters required for the required iterations.

[0071] Specifically, the number of iterations N max is set to 50000; the update step size ε is set to 1e-3; the regularization coefficient λ is calculated using the generalized cross-validation method; the L1 regularization term coefficient α is set to 0.1.

[0072] The convergence condition kerneltol indicates the correction coefficient vector of the adjacent two steps The relative error threshold of xtol indicates the relative error threshold of the leak rate estimates between two adjacent steps. The iteration terminates when any of the above errors is less than the given threshold.

[0073] In some implementations, the two convergence conditions are determined by the following formulas:

[0074] Correction coefficient vectors for the kth and k-1th steps The relative error When it is less than kerneltol, the iteration terminates, where kerneltol = 1e-5;

[0075] The relative error of the leakage rate estimation value σ between the kth step and the k-1th step When it is less than xtol, the iteration terminates, where xtol = 1e-4.

[0076] S4. Update the diagonal correction coefficient matrix W m×m

[0077] Specifically, through the formula To update, that is, use the correction coefficient vector of step k The diagonal correction coefficient matrix W of the kth step is updated by using the elements of , which is used to update the source item leakage rate.

[0078] S5, using the k-1 step diagonal correction coefficient matrix Wk-1 Update the k-th step source leakage rate σ k , specifically:

[0079]

[0080] Where ε is the update step defined in S3; H is the transport matrix defined in S1; f is the observation value, i.e., y defined in S1 obs ;λ and α are the regularization coefficient and L1 regularization term coefficient defined in S3 respectively; is the derivative operator, Corresponding to the L1-regularization term, corresponds to the total variation regularization term.

[0081] S6. Using the k-th step source term leakage rate σ k Update the k-th step correction coefficient vector Specifically:

[0082]

[0083] Where ε is the update step defined in S3; H is the transport matrix defined in S1; f is the observation value, i.e., y defined in S1 obs .

[0084] S7. Process the correction coefficient vector and normalize its center distance to 1, thereby improving the accuracy of subsequent iterations.

[0085] Specifically, keep the k-th step correction coefficient vector The non-negativity of each element. Use the MCD method to calculate its central moment and use this central moment to normalize it, that is, perform the following operations:

[0086]

[0087]

[0088]

[0089] S8. Determine the correction coefficient vectors of step k and step k-1 Whether the relative error of the estimated value σ of the leakage rate in the kth step and the k-1th step reaches the convergence condition kerneltol, if so, enter S9. If not, determine whether the relative error of the estimated value σ of the leakage rate in the kth step and the k-1th step reaches the convergence condition xtol, if so, enter S9, if not, further determine whether the number of iterations N is reached max , if the number of iterations is reached, go to S9, otherwise go to S4.

[0090] S9, the leakage rate vector σ obtained by iterationn The normalized factor (σ n ·coef) in step S1 is multiplied, that is, the inverse normalization is performed, to obtain the source term inversion result based on the diluted observation input, that is, the result vector u, which is the piecewise smoothed source term leakage rate curve and can be used as the prior leakage rate input for subsequent research and provides more accurate estimation.

[0091] Embodiment Two

[0092] The above embodiment one provides a piecewise smoothing inversion method for radionuclide release timing, and correspondingly, the present embodiment provides a piecewise smoothing source term inversion system. The piecewise smoothing source term inversion system provided by the present embodiment can implement the piecewise smoothing inversion for radionuclide release timing of embodiment one. The system can be realized by software, hardware or a combination of software and hardware. For example, the system can include integrated or separate functional modules or functional units to perform the corresponding steps in the methods of embodiment one. Since the piecewise smoothing source term inversion system of the present embodiment is basically similar to the method embodiment, the description process of the present embodiment is relatively simple, and the relevant parts can be referred to the part of the description of embodiment one. The piecewise smoothing source term inversion system of the present embodiment is only illustrative.

[0093] The piecewise smoothing source term inversion system provided by the present embodiment comprises:

[0094] A parameter input unit configured to input a transport matrix, an observation vector, an iteration number N max , a regularization coefficient λ, an L1 regularization term coefficient α, and convergence conditions kernel tol and x tol;

[0095] A parameter initialization unit configured to initialize a correction coefficient vector and a source term leakage rate σ n , wherein n represents a time step of simulation;

[0096] A diagonal correction coefficient updating unit configured to update a diagonal correction coefficient matrix W m×m , wherein m is the size of the data provided by the measuring point in the complete simulation period;

[0097] An updated source term leakage rate unit configured to update the source term leakage rate σ k-1 in the kth step by using the diagonal correction coefficient matrix W k in the k-1th step;

[0098] A correction coefficient vector updating unit configured to update the correction coefficient vector k in the kth step by using the source term leakage rate σ

[0099] a correction coefficient vector processing unit configured to normalize the center distance of the correction coefficient vector;

[0100] An iteration judgment unit, configured to judge whether a set iteration condition is met;

[0101] The source term leakage rate curve u calculation unit is configured to iterate the leakage rate vector σ n After performing inverse normalization, the source term inversion result based on the diluted observation input is obtained, that is, the piecewise smooth source term leakage rate curve u is obtained.

[0102] Example 3

[0103] This embodiment provides a processing device for implementing the piecewise smooth inversion method of the radionuclide release sequence provided in the first embodiment. The processing device can be a processing device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the piecewise smooth inversion method of the radionuclide release sequence provided in the first embodiment.

[0104] The processing device includes a processor, a memory, a communication interface, and a bus. The processor, memory, and communication interface are connected via the bus to facilitate communication between them. The memory stores a computer program executable on the processor. When the processor executes the computer program, it executes the piecewise smoothed inversion method for radionuclide release time series provided in the first embodiment.

[0105] Preferably, the memory may be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0106] Preferably, the processor may be a central processing unit (CPU), a digital signal processor (DSP), or other general-purpose processors of various types, which are not limited here.

[0107] Example 4

[0108] The piecewise smooth inversion method for the radionuclide release time series of the first embodiment is specifically implemented as a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing the piecewise smooth inversion method for the radionuclide release time series described in the first embodiment.

[0109] A computer-readable storage medium may be a tangible device that holds and stores instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features, and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A piecewise smooth inversion method for radionuclide release time series, characterized in that include: S1. Input and normalize the transport matrix and observation vector; S2. Initialize the correction coefficient vector and source term leakage rate , represents the time step of the simulation; S3. Input the number of iterations , regularization coefficient , L1 regularization term coefficient and the convergence conditions kerneltol and xtol; S4. Update the diagonal correction coefficient matrix , The amount of data provided for the measurement points during the complete simulation period; S5, using the k-1 step diagonal correction coefficient matrix Update the leakage rate of the source term at step k , the specific formula is: Where, is the update step size; is the transport matrix; is the observed value; is the derivative operator, Corresponding to the L1-regularization term, Corresponding to the total variation regularization term; S6. Using the k-th step source item leakage rate Update the k-th step correction coefficient vector S7, normalizing the center distance of the correction coefficient vector; S8. Determine the correction coefficient vectors of step k and step k-1 Whether the relative error reaches the convergence condition kerneltol, if it reaches the convergence condition, then enter S9; if not, judge the leakage rate estimation value of step k and step k-1 Whether the relative error reaches the convergence condition xtol, if it does, then go to S9; if not, further judge whether the number of iterations is reached , if the number of iterations is reached, then go to S9, otherwise go to S4; S9, the leakage rate vector obtained by iteration Perform inverse normalization to obtain the source term inversion result based on the diluted observation input, that is, the piecewise smoothed source term leakage rate .

2. The piecewise smooth inversion method for radionuclide release time series according to claim 1, characterized in that: The specific process of S1 above includes: Transport matrix The normalized representation of ; Observation vector The normalized representation of ; Among them, the normalization factor Take the observation vector The largest element in absolute value.

3. The piecewise smooth inversion method for radionuclide release time series according to claim 1, characterized in that: The specific process of the above S2 includes: Initialize the correction coefficient vector , that is, all its elements are set to 1; Initialize the source term leakage rate vector , that is, setting all elements in the vector to 0.

4. The piecewise smooth inversion method for radionuclide release time series according to claim 1, characterized in that: The formula for determining the convergence condition kerneltol in S3 above is: Correction coefficient vectors for the kth and k-1th steps The relative error Less than The iteration terminates when = 1e-5.

5. The method for piecewise smooth inversion of radionuclide release time series according to claim 1, characterized in that: The judgment formula of the convergence condition xtol in S3 above is: Leakage rate estimates at step k and step k-1 The relative error When it is less than xtol, the iteration terminates. = 1e-4.

6. The method for piecewise smooth inversion of radionuclide release time series according to claim 1, characterized in that: The above S7 normalizes the center distance of the correction coefficient vector using the MCD method, specifically: ; ; 。 7. A piecewise smooth inversion system for radionuclide release time series, characterized by The system includes: Parameter input unit, configured to input transport matrix, observation vector, number of iterations , regularization coefficient , L1 regularization term coefficient and the convergence conditions kerneltol and xtol; A parameter initialization unit configured to initialize the correction coefficient vector and source term leakage rate , represents the time step of the simulation; A diagonal correction coefficient updating unit configured to update the diagonal correction coefficient matrix , The amount of data provided for the measurement points during the complete simulation period; Update source term leakage rate unit is configured to use the k-1th step diagonal correction coefficient matrix Update the leakage rate of the source term at step k The specific process is: Where, is the update step size; is the transport matrix; is the observed value; is the derivative operator, Corresponding to the L1-regularization term, Corresponding to the total variation regularization term; The correction coefficient vector updating unit is configured to use the leakage rate of the source term at the kth step Update the k-th step correction coefficient vector a correction coefficient vector processing unit configured to normalize the center distance of the correction coefficient vector; An iteration judgment unit, configured to judge whether a set iteration condition is met; Source term leakage rate curve The calculation unit is configured to convert the leak rate vector obtained by iteration Perform inverse normalization to obtain the source term inversion result based on the diluted observation input, that is, the piecewise smoothed source term leakage rate .

8. A processing device, comprising at least a processor and a memory, wherein a computer program is stored in the memory, wherein: When the processor runs the computer program, the computer program is executed to implement the piecewise smooth inversion method for the radionuclide release time series according to any one of claims 1 to 6.

9. A computer storage medium, characterized in that Computer-readable instructions are stored thereon, and the computer-readable instructions can be executed by a processor to implement the piecewise smooth inversion method for the radionuclide release time series according to any one of claims 1 to 6.

Images