Multi-field coupled reactor radiation shielding design optimization method and system

Abstract

The invention discloses a multi-field coupled reactor radiation shielding design optimization method and system, and relates to the technical field of nuclear radiation shielding, and the method comprises the steps: collecting nuclear reactor information, determining constraint conditions, and building a nuclear reactor cavity and a shielding model; radiation field distribution parameters are calculated, calculation results are screened, and thermal and structural performance is combined for simulation; and parameters are trained and adjusted through a neural network model, and a shielding design scheme is optimized. According to the multi-field coupling reactor radiation shielding design optimization method, the safety and the reliability of the design scheme are improved according to the dose constraint condition screening result, the screened data are optimized and predicted through the neural network model, the precision and the robustness of the model are improved, the prediction precision and the optimization effect of the design scheme are improved, and the optimization effect of the design scheme is improved. The method has better effects in the aspects of precision, robustness and optimization effect.

Description

Technical Field

[0001] The invention relates to the technical field of nuclear radiation shielding, in particular to a design optimization method for multi-field coupled reactor radiation shielding. Background Art

[0002] The radiation generated by the operation of nuclear reactors can cause ionizing radiation damage to the surrounding environment and personnel, and materials need to be selected for effective shielding. At the same time, the activation reaction and radiation capture of neutrons and shielding materials will also produce secondary products and secondary gamma rays, which have the characteristics of multi-coupling. The safe and reasonable design of reactor radiation shielding mainly involves finding a shielding solution that meets the requirements of small size, light weight, stable structure, low cost and environmental friendliness while ensuring dose safety. It is a complex, multi-variable, multi-objective engineering optimization problem.

[0003] The variables for reactor radiation shielding design optimization include material selection, material sorting, shield size, shield quantity, radiation energy, etc. The optimization objectives include but are not limited to dose limit, duty cycle, total mass, radiation shielding performance, structural performance, heat transfer performance and economy, etc. These optimization objectives usually appear at the same time and influence each other, making it an urgent need to seek a multi-objective, multi-physical field coupled reactor shielding design optimization method for current engineering design.

[0004] In actual projects in the past, the optimization design of nuclear reactor shielding usually provides a series of optional solutions based on experience, or screens them out after manually calculating many solutions. This has the problems of high threshold, low efficiency and poor accuracy. In the past two decades, some scientists at home and abroad have used deterministic methods and Monte Carlo methods to carry out some work on shielding design optimization. However, due to the limitations of deterministic methods and Monte Carlo methods, these studies are usually limited to simple models and simple verifications, and are difficult to be applied to projects with more complex optimization problems. So far, a set of shielding optimization methods with strong robustness, wide applicability and comprehensive considerations has not been formed. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: the existing reactor radiation shielding design method has the problems of insufficient complexity of multi-field coupling, insufficient automation and precision, insufficient comprehensiveness and versatility, and how to adapt to different reactor types and operating conditions and improve the adaptability and robustness of the design scheme.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a multi-field coupled reactor radiation shielding design optimization method, comprising collecting nuclear reactor information and determining constraint conditions, establishing a nuclear reactor cavity and shielding model; calculating radiation field distribution parameters, screening calculation results, and simulating in combination with thermal and structural performance; training and adjusting parameters through a neural network model to optimize the shielding design scheme.

[0008] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization method of the present invention, wherein: the acquisition and determination of nuclear reactor information and determination of constraint conditions include determining nuclear reactor constraint conditions, nuclear reactor dose constraint conditions, heat transfer performance constraint conditions, and structural performance constraint conditions, which are expressed as:

[0009]

[0010] Where D is the dose, Q is the heat transfer performance, σ e is the effective stress of the structural performance, D s1，i and D s2，j are the cumulative doses of the i-th and j-th layers in the primary and secondary shielding, respectively, and D max is the maximum permissible dose, k s1，i and k s2，j are the thermal conductivities of the i-th and j-th layers in the primary and secondary shields, respectively, A s1，i and A s2，j are the areas of the i-th and j-th layers in the primary and secondary shielding, respectively, ΔT s1，i and ΔT s2,j are the temperature differences between the i-th and j-th layers in the primary and secondary shields, respectively, Q max is the maximum allowable heat transfer, σ s1,i and σ s2，j are the stresses of the i-th and j-th layers in the primary and secondary shields, respectively, σ y，s1，i and σ y,s2，j are the yield strength of the material, respectively, to determine the thickness range and height range of the side material layer of the nuclear reactor cavity model, to determine the number of material layers of the shielding model of the primary shielding and secondary shielding models, expressed as:

[0011] t c，min ≤t c ≤t c,max

[0012] h c,min ≤h c ≤h c，max

[0013] 1≤n 1 ≤N s1

[0014] 1≤n2 ≤N s2

[0015] Among them, t c is the thickness of the side material layer, h c is the height of the side material layer, t c,min and t c,max are the minimum and maximum values ​​of the thickness of the side material layer, h c,min and h c，max are the minimum and maximum values ​​of the side material layer height, n 1 and n 2 is the number of shielding model material layers, N s1 and N s2 The maximum number of layers.

[0016] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization method of the present invention, the establishment of the nuclear reactor cavity and shielding model includes model geometric parameters, confirming the geometric shape and geometric size of the nuclear reactor cavity and shielding model, setting the geometry, number of layers, and material of the model as variables, generating input sample 1 by random sampling method and writing it into a file for storage, and the geometric shape and size of the nuclear reactor cavity and shielding model are expressed as:

[0017] V c =πR 2 H.α

[0018]

[0019] Among them, V c is the reactor cavity volume, R is the radius, H is the height, α is the normalization parameter, V s1 and V s2 are the primary shielding volume and the secondary shielding volume, R o and R s are the primary and secondary shielding radii, H o and H s are the primary and secondary shielding heights, t s1，i is the thickness of the i-th layer of material, t s2，j is the thickness of the j-th layer of material, β i and γ j is the normalization factor.

[0020] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization method of the present invention, the calculation of radiation field distribution parameters includes inputting sample file 1 to call nuclear physics calculation, combining the interaction between neutrons and photons, and introducing a coupling coefficient to reflect the attenuation behavior of the interaction between neutrons and photons in the shielding material, which is expressed as:

[0021]

[0022] Among them, Φ n (r) and Φ γ (r) are the neutron flux distribution and photon flux distribution at distance r, Φ n0 and Φ γ0 are the initial neutron flux and initial photon flux respectively, σ n and σ γ are the neutron attenuation constant and the photon attenuation constant, respectively, and κ is the coupling coefficient. The energy spectrum distribution of high-energy neutrons and photons is described by the energy filtering function, which is expressed as:

[0023]

[0024] Among them, φ n (E) and φ γ (E) are the neutron energy spectrum distribution and photon energy spectrum distribution at energy E, φ n0 and φ γ0 is the normalization coefficient, α n , β n , α γ , β γ is the spectral parameter, γ n and γ γ is the energy filtering parameter, E 0 For reference energy, the non-uniformity parameter is introduced to reflect the trend of power attenuation with increasing distance and to describe the non-uniform distribution of power at different distances, which is expressed as:

[0025]

[0026] Where P(r) is the power distribution at distance r, P 0 is the initial power, σ p is the power attenuation constant, δ p is the non-uniformity parameter, R 0 For reference distance, the energy deposition effects of neutrons and photons are combined to describe the energy deposition behavior of different types of radiation in shielding materials, which is expressed as:

[0027]

[0028] Where E(r) is the energy deposition distribution at distance r, ∑ n and∑ γ are the energy deposition coefficients of neutrons and photons respectively, λ is the normalization parameter, and the calculation results are written into output file 1.

[0029] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization method described in the present invention, the screening calculation results includes judging the output file 1 according to the dose constraint condition, if there is a calculation result exceeding the dose constraint condition, the parameter combination containing the result exceeding the dose constraint condition is deleted, the calculation results of all parameter combinations are traversed, and the calculation results and parameter combinations that meet the judgment condition are written into the output file 2, in the multi-physical field coupling calculation or thermal hydraulic analysis, the parameter combination in the output file 2 is read as the setting of the calculation analysis model and regional material, the power distribution or energy deposition distribution is read as the boundary condition of the model, the calculation is performed after the grid is divided, and the calculation result of the parameter combination is output and written into the output file 3, and the output file 3 is judged. If there is a calculation result exceeding one of the heat transfer performance constraint condition or the structural performance constraint condition, the parameter combination containing the result exceeding the constraint condition is deleted, the calculation results of all parameter combinations are traversed, and the calculation results and parameter combinations that meet the judgment condition are written into the output file 4.

[0030] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization method of the present invention, the training and adjustment of parameters includes taking the output file 4 as an input sample, dividing the input sample into a training set, a verification set, and a test set, inputting the training set and the neural network configuration parameters, and initializing the neural network, which is expressed as:

[0031]

[0032] Among them, θ is the parameter set of the neural network, W is the weight, b is the bias term, The mean is 0 and the variance is σ 2 Gaussian distribution, use the training set to train the neural network, adjust the parameters of the neural network model, and adjust the hyperparameters of the neural network model through the validation set, which is expressed as:

[0033]

[0034]

[0035] Among them, L(θ) is the loss function, N is the number of training samples, and y i is the true value, f(x i ; θ) is the predicted value, λ 1 is the regularization parameter, || || 2 is the L2 norm, L v (θ) is the validation set loss function, M is the number of validation samples, η is the learning rate, and T is the number of iterations. The test set is used to determine whether the adjusted neural network model meets the accuracy requirements. If it does not meet the accuracy requirements, the parameters are adjusted repeatedly. If it meets the accuracy requirements, the training ends and the neural network model parameters are output.

[0036] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization method described in the present invention, the optimized shielding design scheme includes taking the neural network model parameters as optimization parameters, initializing the population composed of the output file 4, evaluating the population of the output file 4 according to the dose constraints, heat transfer performance constraints, and structural performance constraints, and obtaining the fitness index of the shielding design performance. According to the fitness of the individuals in the population, selection, crossover, and mutation operations are performed to generate a new population with better distribution, and it is judged whether the termination conditions are met, that is, the dose constraints, heat transfer performance constraints, and structural performance constraints meet the requirements. If not, the iteration is repeated. If the conditions are met, the optimized shielding design scheme is output.

[0037] Another object of the present invention is to provide a multi-field coupled reactor radiation shielding design optimization system, which can perform nuclear physics calculations on the nuclear reactor model through a calculation screening module to generate radiation field distribution parameters, thereby solving the current problem of being unable to fully simulate the complex environment of the reactor.

[0038] As a preferred solution of the multi-field coupled reactor radiation shielding design optimization system described in the present invention, it includes: a confirmation modeling module, a calculation screening module, and a parameter optimization module; the confirmation modeling module is used to collect nuclear reactor information and determine constraints, and establish a nuclear reactor cavity and shielding model; the calculation screening module is used to calculate radiation field distribution parameters, screen calculation results, and simulate in combination with thermal and structural performance; the parameter optimization module is used to train and adjust parameters through a neural network model to optimize the shielding design scheme.

[0039] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that the processor executes the computer program to implement the steps of a multi-field coupled reactor radiation shielding design optimization method.

[0040] A computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of a method for optimizing the design of radiation shielding for a reactor with multi-field coupling are implemented.

[0041] Beneficial effects of the present invention: The multi-field coupled reactor radiation shielding design optimization method provided by the present invention screens the results according to the dose constraint conditions, improves the safety and reliability of the design scheme, optimizes and predicts the screened data through the neural network model, improves the accuracy and robustness of the model, and improves the prediction accuracy and optimization effect of the design scheme. The present invention achieves better results in terms of accuracy, robustness and optimization effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0043] Figure 1 This is an overall flow chart of the multi-field coupled reactor radiation shielding design optimization method provided by the first embodiment of the present invention.

[0044] Figure 2 A schematic diagram of a multi-field coupled reactor radiation shielding design optimization method provided by the first embodiment of the present invention.

[0045] Figure 3 This is an overall module diagram of a multi-field coupled reactor radiation shielding design optimization system provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0047] Example 1

[0048] Reference Figure 1-Figure 2 , which is an embodiment of the present invention, provides a multi-field coupled reactor radiation shielding design optimization method, comprising:

[0049] S1: Collect nuclear reactor information and determine constraints, and establish nuclear reactor cavity and shielding models.

[0050] Furthermore, collecting and determining nuclear reactor information and determining constraints includes determining nuclear reactor constraints.

[0051] It should be noted that the dose constraints, heat transfer performance constraints, and structural performance constraints of the nuclear reactor are expressed as:

[0052]

[0053] Where D is the dose, Q is the heat transfer performance, σ e is the effective stress of the structural performance, D s1，i and D s2，j are the cumulative doses of the i-th and j-th layers in the primary and secondary shielding, respectively, and D maxis the maximum permissible dose, k s1，i and k s2，j are the thermal conductivities of the i-th and j-th layers in the primary and secondary shields, respectively, A s1，i and A s2，j are the areas of the i-th and j-th layers in the primary and secondary shielding, respectively, ΔT s1,i and ΔT s2,j are the temperature differences between the i-th and j-th layers in the primary and secondary shields, respectively, Q max is the maximum allowable heat transfer, σ s1,i and σ s2,j are the stresses of the i-th and j-th layers in the primary and secondary shields, respectively, σ y，s1，i and σ y,s2，j are the yield strength of the material, respectively, to determine the thickness range and height range of the side material layer of the nuclear reactor cavity model, to determine the number of material layers of the shielding model of the primary shielding and secondary shielding models, expressed as:

[0054] t c,min ≤t c ≤t c,max

[0055] h c,min ≤h c ≤h c,max

[0056] 1≤n 1 ≤N s1

[0057] 1≤n 2 ≤N s2

[0058] Among them, t c is the thickness of the side material layer, h c is the height of the side material layer, t c,min and t c,max are the minimum and maximum thickness of the side material layer, h c,min and h c，max are the minimum and maximum values ​​of the side material layer height, n 1 and n 2 is the number of shielding model material layers, N s1 and N s2 The maximum number of layers.

[0059] Furthermore, the nuclear reactor cavity and shielding models are established including model geometry parameters.

[0060] It should be noted that the geometric shape and size of the nuclear reactor cavity and shielding model are confirmed, the geometry, number of layers, and material of the model are set as variables, and the input sample 1 is generated by random sampling method and written into the file for saving. The geometric shape and size of the nuclear reactor cavity and shielding model are expressed as:

[0061] V c =πR 2 H.α

[0062]

[0063] Among them, V c is the reactor cavity volume, R is the radius, H is the height, α is the normalization parameter, V s1 and V s2 are the primary shielding volume and the secondary shielding volume, R o and R s are the primary and secondary shielding radii, H o and H s are the primary and secondary shielding heights, t s1，i is the thickness of the i-th layer of material, t s2，j is the thickness of the j-th layer of material, β i and γ j is the normalization factor.

[0064] S2: Calculate the radiation field distribution parameters, filter the calculation results, and simulate in combination with thermal and structural performance.

[0065] Furthermore, calculating the radiation field distribution parameters includes inputting the sample file 1 to call the nuclear physics calculation.

[0066] It should be noted that, combined with the interaction between neutrons and photons, the coupling coefficient is introduced to reflect the attenuation behavior of the interaction between neutrons and photons in the shielding material, which is expressed as:

[0067]

[0068] Among them, Φ n (r) and Φ γ (r) are the neutron flux distribution and photon flux distribution at distance r, Φ n0 and Φ γ0 are the initial neutron flux and initial photon flux respectively, σ n and σ γ are the neutron attenuation constant and the photon attenuation constant, respectively, and K is the coupling coefficient. The energy spectrum distribution of high-energy neutrons and photons is described by the energy filtering function, which is expressed as:

[0069]

[0070] Among them, φn (E) and φ γ (E) are the neutron energy spectrum distribution and photon energy spectrum distribution at energy E, φ n0 and φ γ0 is the normalization coefficient, α n , β n , α γ , β γ is the spectral parameter, γ n and γ γ is the energy filtering parameter, E 0 For reference energy, the non-uniformity parameter is introduced to reflect the trend of power attenuation with increasing distance and to describe the non-uniform distribution of power at different distances, which is expressed as:

[0071]

[0072] Where P(r) is the power distribution at distance r, P 0 is the initial power, σ p is the power attenuation constant, δ p is the non-uniformity parameter, R 0 For reference distance, the energy deposition effects of neutrons and photons are combined to describe the energy deposition behavior of different types of radiation in shielding materials, which is expressed as:

[0073]

[0074] Where E(r) is the energy deposition distribution at distance r, ∑ n and∑ γ are the energy deposition coefficients of neutrons and photons respectively, λ is the normalization parameter, and the calculation results are written into output file 1.

[0075] Furthermore, screening the calculation results includes based on dose constraints.

[0076] It should be noted that the output file 1 is judged. If there is a calculation result that exceeds the dose constraint, the parameter combination containing the result exceeding the dose constraint is deleted, the calculation results of all parameter combinations are traversed, and the calculation results and parameter combinations that meet the judgment conditions are written into the output file 2. In the multi-physical field coupling calculation or thermal hydraulic analysis, the parameter combination in the output file 2 is read as the calculation analysis model and regional material settings, the power distribution or energy deposition distribution is read as the boundary condition of the model, the grid is divided and calculated, and the calculation results of the parameter combination are output and written into the output file 3. The output file 3 is judged. If there is a calculation result that exceeds one of the heat transfer performance constraint conditions or the structural performance constraint conditions, the parameter combination containing the result exceeding the constraint condition is deleted, the calculation results of all parameter combinations are traversed, and the calculation results and parameter combinations that meet the judgment conditions are written into the output file 4.

[0077] It should also be noted that the neutron flux distribution and the photon flux distribution describe the distribution of neutrons and photons in space in a nuclear reactor, respectively. These distributions are crucial for evaluating the intensity of radiation at different locations. If the neutron or photon flux in a certain area is too high, the radiation dose in that area may exceed the safety limit. Therefore, comparing the neutron and photon flux distributions with the dose constraints can ensure that the radiation dose at each location is within a safe range, thereby ensuring the safety of personnel and the environment.

[0078] It should also be noted that the neutron energy spectrum and photon energy spectrum provide energy distribution information of neutrons and photons. Neutrons and photons of different energies have different penetration capabilities and radiation hazards to materials. High-energy neutrons and photons are more destructive and may result in higher radiation doses. By comparing the calculated results of the neutron energy spectrum and photon energy spectrum with the dose constraints, the shielding effect of the shielding material at different energy levels can be evaluated to ensure that high-energy radiation does not cause excessive doses, thereby optimizing the shielding design and improving safety.

[0079] It should also be noted that the power distribution describes the distribution of power in the reactor in space. The power distribution directly affects the thermal load and radiation dose of the reactor. Areas with excessively high power are often accompanied by high radiation doses, which may exceed safety limits. Comparing the power distribution with the dose constraints can ensure that the radiation dose in high-power areas is within a controllable range and guide the optimization of shielding design to reduce radiation leakage in high-power areas.

[0080] It should also be noted that the energy deposition distribution describes the deposition of radiation energy in the shielding material. The more energy is deposited in an area, the higher the radiation dose. By calculating the energy deposition distribution and comparing it with the dose constraints, high energy deposition areas can be identified, and then the shielding materials and structures can be optimized to ensure that the radiation dose in these areas does not exceed the standard. This comparison process helps to improve the accuracy and reliability of the shielding design and ensure that the shielding effect meets the standards.

[0081] S3: Train and adjust parameters through the neural network model to optimize the shielding design solution.

[0082] Furthermore, training and adjusting the parameters includes taking the output file 4 as an input sample.

[0083] It should be noted that the input samples are divided into training set, validation set and test set, the training set and neural network configuration parameters are input, and the neural network is initialized, which is expressed as:

[0084]

[0085] Among them, θ is the parameter set of the neural network, W is the weight, b is the bias term, The mean is 0 and the variance is σ 2 Gaussian distribution, use the training set to train the neural network, adjust the parameters of the neural network model, and adjust the hyperparameters of the neural network model through the validation set, which is expressed as:

[0086]

[0087] Among them, L(θ) is the loss function, N is the number of training samples, and y i is the true value, f(x i ; θ) is the predicted value, λ 1 is the regularization parameter, || || 2 is the L2 norm, L v (θ) is the validation set loss function, M is the number of validation samples, η is the learning rate, and T is the number of iterations. The test set is used to determine whether the adjusted neural network model meets the accuracy requirements. If it does not meet the accuracy requirements, the parameters are adjusted repeatedly. If it meets the accuracy requirements, the training ends and the neural network model parameters are output.

[0088] Furthermore, the optimized shielding design scheme includes using the neural network model parameters as optimization parameters.

[0089] It should be noted that the population composed of the initialized output file 4 is evaluated according to the dose constraints, heat transfer performance constraints, and structural performance constraints to obtain the fitness index of the shielding design performance. According to the fitness of the individuals in the population, selection, crossover, and mutation operations are performed to generate a new population with a better distribution, and it is judged whether the termination conditions are met, that is, the dose constraints, heat transfer performance constraints, and structural performance constraints meet the requirements. If not, the iteration is repeated. If the conditions are met, the optimized shielding design scheme is output.

[0090] Example 2

[0091] One embodiment of the present invention provides a multi-field coupled reactor radiation shielding design optimization method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0092] The multi-field coupling optimization method of the present invention is compared with the prior art. The test preparation includes collecting specific information of the nuclear reactor, determining various constraints, and verifying the effectiveness of the inventive method through modeling, calculation, screening and optimization. First, the relevant information and constraints of the nuclear reactor are determined, including dose constraints, heat transfer performance constraints, and structural performance constraints. The dose constraints in this experiment are set as the dose around the reactor does not exceed 5mSv / h, the heat transfer performance constraints are set as the shielding layer material temperature does not exceed 300°C, and the structural performance constraints are set as the material stress does not exceed 150MPa. The total mass requirement is as small as possible, and the duty cycle is controlled between 0.2-0.4. In terms of economy, the material and manufacturing costs are reduced as much as possible. Then, a nuclear reactor cavity model, a primary shielding model and a secondary shielding model are established in the nuclear physics calculation program. The cavity model is cylindrical, and its geometric dimensions and material parameters are set as variables. A set of input samples is generated by random sampling, saved as an input file, and the nuclear physics calculation program is called to read the input file for calculation. Calculate and output data such as neutron flux distribution, photon flux distribution, neutron energy spectrum, photon energy spectrum, power distribution and energy deposition distribution, and save them as output file 1. Filter the data in output file 1 according to the dose constraint conditions, eliminate the parameter combination that does not meet the dose constraint, and write the results that meet the requirements into output file 2. Read the data of output file 2 in the multi-physics field coupling calculation software, perform coupled calculation of thermal and structural performance, output data such as material layer temperature and stress, and save them as output file 3. Filter the data in output file 3, eliminate the parameter combination that does not meet the heat transfer performance or structural performance constraint conditions, and write the results that meet the requirements into output file 4. Use the neural network model to train, verify and adjust the filtered data, optimize the neural network model parameters by dividing the training set, verification set and test set, initialize the parameter combination population based on the neural network model parameters, generate the optimized new population through multiple selection, crossover and mutation operations, and perform fitness evaluation to determine whether it meets all constraints, and finally output the optimal shielding design scheme.

[0093] As shown in Table 1, in terms of total mass, the scheme of the present invention is generally lower than the existing technical scheme, which significantly reduces the weight of the structure. In terms of duty cycle, the scheme of the present invention is between 0.20-0.28, which is in line with the reasonable range and is generally lower than the existing technical scheme, which helps to improve the utilization efficiency of shielding materials. In terms of heat transfer performance, the temperature of the scheme of the present invention is generally lower than that of the existing technical scheme, which significantly improves the thermal performance and reduces the thermal load of the material. In terms of structural performance, the stress of the scheme of the present invention is also significantly lower than that of the existing technical scheme, which improves the structural safety and stability. In terms of economy, the cost of the scheme of the present invention is generally lower than that of the existing technical scheme, which significantly reduces the economic cost of the shielding design. The scheme of the present invention is superior to the existing technology in terms of total mass, duty cycle, heat transfer performance, structural performance and economy, reflecting the significant advantages of the present invention in the optimization of radiation shielding design. In particular, through the application of multi-field coupling calculation and optimization algorithm, the scheme of the present invention not only achieves optimization in multiple performance indicators, but also ensures the efficiency and accuracy of the design process.

[0094] Table 1 Experimental data table

[0095] Total mass Duty Cycle temperature stress cost Prior art 1 1500kg 0.35 290℃ 140MPa 1.2 million yuan Prior Art 2 1450kg 0.3 285℃ 135MPa 1.15 million yuan Prior Art 3 1600kg 0.32 295℃ 145MPa 1.25 million yuan The present invention 1 1300kg 0.25 280℃ 130MPa 1.1 million yuan The present invention 2 1350kg 0.28 275℃ 128MPa 1.08 million yuan Present invention 3 1320kg 0.27 278℃ 129MPa 1.09 million yuan The present invention 4 1280kg 0.26 277℃ 127MPa 1.07 million yuan The present invention 5 1310kg 0.25 276℃ 126MPa 1.06 million yuan Present invention 6 1290kg 0.27 279℃ 125MPa 1.05 million yuan The present invention 7 1270kg 0.28 274℃ 124MPa 1.04 million yuan Present invention 8 1330kg 0.26 273℃ 123MPa 1.03 million yuan The present invention 9 1250kg 0.24 272℃ 122MPa 1.02 million yuan The present invention 10 1260kg 0.23 271℃ 121MPa 1.01 million yuan Present invention 11 1240kg 0.22 270℃ 120MPa 1 million yuan Present invention 12 1230kg 0.25 275℃ 119MPa 990,000 yuan Present invention 13 1220kg 0.24 268℃ 118MPa 980,000 yuan Present invention 14 1210kg 0.23 267℃ 117MPa 970,000 yuan Present invention 15 1200kg 0.22 266℃ 116MPa 960,000 yuan Present invention 16 1190kg 0.21 265℃ 115MPa 950,000 yuan The present invention 17 1180kg 0.2 264℃ 114MPa 940,000 yuan Present invention 18 1170kg 0.22 263℃ 113MPa 930,000 yuan The present invention 19 1160kg 0.21 262℃ 112MPa 920,000 yuan The present invention 20 1150kg 0.2 261℃ 111MPa 910,000 yuan

[0096] Example 3

[0097] Reference Figure 3 , which is an embodiment of the present invention, provides a multi-field coupled reactor radiation shielding design optimization system, including: a confirmation modeling module, a calculation screening module, and a parameter adjustment optimization module.

[0098] The confirmation modeling module is used to collect nuclear reactor information and determine constraints, and establish a nuclear reactor cavity and shielding model; the calculation and screening module is used to calculate the radiation field distribution parameters, screen the calculation results, and simulate in combination with thermal and structural performance; the parameter optimization module is used to train and adjust parameters through a neural network model to optimize the shielding design scheme.

[0099] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0100] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0101] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0102] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0103] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-field coupled reactor radiation shielding design optimization method, characterized in that: include: Collect nuclear reactor information and determine constraints to build nuclear reactor cavity and shielding models; Calculate the radiation field distribution parameters, screen the calculation results, and simulate in combination with thermal and structural performance; The parameters are trained and adjusted through the neural network model to optimize the shielding design solution.

2. The multi-field coupled reactor radiation shielding design optimization method according to claim 1, characterized in that: The collecting and determining of nuclear reactor information and determining constraints include determining nuclear reactor constraints, nuclear reactor dose constraints, heat transfer performance constraints, and structural performance constraints, which are expressed as: Where D is the dose, Q is the heat transfer performance, σ e is the effective stress of the structural performance, D s1,i and D s2,j are the cumulative doses of the i-th and j-th layers in the primary and secondary shielding, respectively, and D max is the maximum permissible dose, k s1,i and k s2,j are the thermal conductivities of the i-th and j-th layers in the primary and secondary shields, respectively, A s1,i and A s2,j are the areas of the i-th and j-th layers in the primary and secondary shielding, respectively, ΔT s1,i and ΔT s2,j are the temperature differences between the i-th and j-th layers in the primary and secondary shields, respectively, Q max is the maximum allowable heat transfer, σ s1,i and σ s2,j are the stresses of the i-th and j-th layers in the primary and secondary shields, respectively, σ y,s1,i and σ y,s2,j are the yield strength of the material, respectively, to determine the thickness range and height range of the side material layer of the nuclear reactor cavity model, to determine the number of material layers of the shielding model of the primary shielding and secondary shielding models, expressed as: t c,min ≤t c ≤t c,max h c,min ≤h c ≤h c,max 1≤n1≤N s1 1≤n2≤N s2 Among them, t c is the thickness of the side material layer, h c is the height of the side material layer, tc ,min and tc ,max are the minimum and maximum thickness of the side material layer, h c,min and h c,max are the minimum and maximum values ​​of the side material layer heights, n1 and n2 are the number of shielding model material layers, N s1 and N s2 The maximum number of layers.

3. The multi-field coupled reactor radiation shielding design optimization method according to claim 2, characterized in that: The establishment of the nuclear reactor cavity and shielding model includes model geometry parameters, confirming the geometry and geometry size of the nuclear reactor cavity and shielding model, setting the geometry, number of layers, and material of the model as variables, generating input sample 1 by random sampling method and writing it into a file for storage, and the geometry and size of the nuclear reactor cavity and shielding model are expressed as: V c =πR 2 H·a Among them, V c is the reactor cavity volume, R is the radius, H is the height, α is the normalization parameter, V s1 and V s2 are the primary shielding volume and the secondary shielding volume, R o and R s are the primary and secondary shielding radii, H o and H s are the primary and secondary shielding heights, t s1,i is the thickness of the i-th layer of material, t s2,j is the thickness of the j-th layer of material, β i and γ j is the normalization factor.

4. The multi-field coupled reactor radiation shielding design optimization method according to claim 3, characterized in that: The calculation of the radiation field distribution parameters includes inputting the sample file 1 to call the nuclear physics calculation, combining the interaction between neutrons and photons, and introducing the coupling coefficient to reflect the attenuation behavior of the interaction between neutrons and photons in the shielding material, which is expressed as: Among them, Φ n (r) and Φ γ (r) are the neutron flux distribution and photon flux distribution at distance r, Φ n0 and Φ γ0 are the initial neutron flux and initial photon flux respectively, σ n and σ γ are the neutron attenuation constant and the photon attenuation constant, respectively, and κ is the coupling coefficient. The energy spectrum distribution of high-energy neutrons and photons is described by the energy filtering function, which is expressed as: Among them, φ n (E) and φ γ (E) are the neutron energy spectrum distribution and photon energy spectrum distribution at energy E, φ n0 and φ γ0 is the normalization coefficient, α n , β n , α γ , β γ is the spectral parameter, γ n and γ γ is the energy filtering parameter, E0 is the reference energy, and the non-uniformity parameter is introduced to reflect the trend of power attenuation with increasing distance and describe the non-uniform distribution of power at different distances, which is expressed as: Where P(r) is the power distribution at distance r, P0 is the initial power, σ p is the power attenuation constant, δ p is the non-uniformity parameter, R0 is the reference distance, and the energy deposition effects of neutrons and photons are combined to describe the energy deposition behavior of different types of radiation in the shielding material, which is expressed as: Where E(r) is the energy deposition distribution at distance r, ∑ n and∑ γ are the energy deposition coefficients of neutrons and photons respectively, λ is the normalization parameter, and the calculation results are written into output file 1.

5. The multi-field coupled reactor radiation shielding design optimization method according to claim 4, characterized in that: The screening calculation results includes judging the output file 1 according to the dose constraint conditions, if there is a calculation result exceeding the dose constraint conditions, deleting the parameter combination containing the result exceeding the dose constraint conditions, traversing the calculation results of all parameter combinations, and writing the calculation results and parameter combinations that meet the judgment conditions into the output file 2, reading the parameter combination in the output file 2 as the calculation analysis model and regional material settings in the multi-physical field coupling calculation or thermal hydraulic analysis, reading the power distribution or energy deposition distribution as the boundary conditions of the model, calculating after grid division, outputting the calculation results of the parameter combination, and writing them into the output file 3, judging the output file 3, if there is a calculation result exceeding one of the heat transfer performance constraint conditions or the structural performance constraint conditions, deleting the parameter combination containing the result exceeding the constraint conditions, traversing the calculation results of all parameter combinations, and writing the calculation results and parameter combinations that meet the judgment conditions into the output file 4.

6. The multi-field coupled reactor radiation shielding design optimization method according to claim 5, characterized in that: The training and adjustment of the parameters includes taking the output file 4 as an input sample, dividing the input sample into a training set, a validation set, and a test set, inputting the training set and the neural network configuration parameters, and initializing the neural network, which is expressed as: Among them, θ is the parameter set of the neural network, W is the weight, b is the bias term, The mean is 0 and the variance is σ 2 Gaussian distribution, use the training set to train the neural network, adjust the parameters of the neural network model, and adjust the hyperparameters of the neural network model through the validation set, which is expressed as: Among them, L(θ) is the loss function, N is the number of training samples, and y i is the true value, f(x i ; θ) is the predicted value, λ1 is the regularization parameter, || ||2 is the L2 norm, L v (θ) is the validation set loss function, M is the number of validation samples, η is the learning rate, and T is the number of iterations. The test set is used to determine whether the adjusted neural network model meets the accuracy requirements. If it does not meet the accuracy requirements, the parameters are adjusted repeatedly. If it meets the accuracy requirements, the training ends and the neural network model parameters are output.

7. The multi-field coupled reactor radiation shielding design optimization method according to claim 6, characterized in that: The optimized shielding design scheme includes taking the neural network model parameters as optimization parameters, initializing the population composed of the output file 4, evaluating the population of the output file 4 according to the dose constraint conditions, the heat transfer performance constraint conditions, and the structural performance constraint conditions, and obtaining the fitness index of the shielding design performance. According to the fitness of the individuals in the population, selection, crossover, and mutation operations are performed to generate a new population with a better distribution, and judging whether the termination conditions are met, that is, the dose constraint conditions, the heat transfer performance constraint conditions, and the structural performance constraint conditions meet the requirements. If not, the iteration is repeated. If the conditions are met, the optimized shielding design scheme is output.

8. A system using the reactor radiation shielding design optimization method of multi-field coupling according to any one of claims 1 to 7, characterized in that: Including confirmation modeling module, calculation screening module, and parameter optimization module; The confirmation modeling module is used to collect nuclear reactor information and determine constraint conditions, and establish a nuclear reactor cavity and shielding model; The calculation and screening module is used to calculate the radiation field distribution parameters, screen the calculation results, and simulate in combination with thermal and structural performance; The parameter optimization module is used to train and adjust parameters through a neural network model to optimize the shielding design solution.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multi-field coupled reactor radiation shielding design optimization method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-field coupled reactor radiation shielding design optimization method according to any one of claims 1 to 7 are implemented.

Images