Nuclear fuel assembly abrasion test load standard determination method

Abstract

The invention provides a nuclear fuel assembly abrasion test load standard determination method, and relates to the technical field of nuclear fuel assembly performance testing, and the method comprises the following steps: based on a finite element simulation method or an accumulated displacement loading test method and an initial deformation amount of a clamping structure of a positioning grillwork; determining the maximum deformation of the clamping structure without failure in the test, and constructing a loading displacement range required to be applied by the test; in the loading displacement range, carrying out quasi-static loading test on the positioning grillwork to obtain a load-displacement relation curve; determining a target agent model according to the change trend of the load-displacement relation curve; extracting a sample set in a loading displacement range, training a target agent model based on the sample set to obtain a displacement-load mapping model, and determining test load standard data sets corresponding to different types of positioning grillwork; the comparability and test consistency of test data of different research institutions are guaranteed, and the reliability of the nuclear fuel assembly abrasion test is improved.

Description

Technical Field

[0001] This application generally relates to the field of nuclear fuel assembly performance testing technology, and specifically to a method for determining the load standard for nuclear fuel assembly erosion tests. Background Technology

[0002] Nuclear power, as an important component of the world's energy structure, is a development direction suited to my country's national conditions. During reactor operation, fuel assemblies are subjected to axial and lateral flows from the high-speed coolant, causing fluid-induced vibrations in the fuel rods, known as flow-induced vibrations. These flow-induced vibrations cause fretting between the fuel rods and the locating grid structure, known as vibration erosion, which can lead to fuel rod cladding failure in severe cases. According to IAEA surveys, locating grid and fuel rod erosion is currently the leading cause of fuel failure in pressurized water reactors. Current research on vibration erosion of locating grids and fuel rods mainly focuses on simulating this erosion behavior, specifically conducting vibration erosion tests on locating grids and fuel rods by setting parameters such as load magnitude, displacement amplitude, vibration frequency, and vibration duration.

[0003] However, due to the lack of a systematic and universally applicable load standard development process, different research institutions or companies often determine load parameters based on their own experience or local data when conducting abrasion tests, resulting in a lack of consistency in test schemes. On the one hand, this makes abrasion test data from different sources incomparable, making it impossible to form a unified performance evaluation benchmark; on the other hand, the lack of a unified standard leads to some tests failing to consider the geometric constraints of the clamping structure and the yield characteristics of the material, which may result in the applied load exceeding its fatigue limit. In long-term tests, the clamping structure is prone to plastic deformation or fatigue cracks due to overload, which not only damages the specimen but also leads to premature termination of the test, making it impossible to obtain complete abrasion life data, while wasting test resources and increasing research and development costs. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a standardized and universally applicable method for determining the load standard for nuclear fuel assembly erosion tests.

[0005] This application provides a method for determining the load standard for nuclear fuel assembly erosion tests, including the following steps: Obtain the initial deformation of the clamping structure of the positioning grid; Based on the finite element simulation method or the cumulative displacement loading test method and the initial deformation, the maximum deformation of the clamping structure without failure during the test is determined; and the initial deformation is used as the lower limit and the maximum deformation is used as the upper limit to construct the range of loading displacement to be applied in the test. Within the specified loading displacement range, a quasi-static loading test is performed on the positioning grid to obtain the load-displacement relationship curve; Based on the changing trend of the load-displacement relationship curve, the target proxy model is determined; and a sample set is extracted within the loading displacement range, and the target proxy model is trained based on the sample set to obtain the displacement-load mapping model. Based on the displacement-load mapping model, a set of standard test load data corresponding to different types of positioning grids is determined; the set of standard test load data is used to guide different types of positioning grids and fuel rods in abrasion tests.

[0006] According to the technical solution provided in this application, the initial deformation of the clamping structure of the positioning grid is obtained, specifically including the following steps: Obtain the manufacturing dimensions and tolerances of the positioning grid, strips, clamping structures, and fuel rods; Based on the calculation method of maximum and minimum entity size or the variance calculation method based on size deviation, the pre-deformation amount generated by the clamping structure when the fuel rod is inserted into the positioning grid cell is obtained, and the pre-deformation amount is used as the initial deformation amount of the clamping structure of the positioning grid.

[0007] According to the technical solution provided in this application, based on the finite element simulation method and the initial deformation, the maximum deformation of the clamping structure to prevent failure during the test is determined, specifically including the following steps: Starting from the initial deformation, a finite element simulation of the clamping structure is performed, and the displacement loading of the clamping structure is gradually increased to obtain the simulation results. Based on the stress distribution of the clamping structure in the simulation results, determine the additional deformation when the maximum stress approaches the yield strength or fatigue limit of the clamping structure material. The sum of the initial deformation and the additional deformation is taken as the maximum deformation in which the clamping structure will not fail during the test.

[0008] According to the technical solution provided in this application, based on the cumulative displacement loading test method and the initial deformation, the maximum deformation in which the clamping structure will not fail during the test is determined, specifically including the following steps: Starting from the initial deformation, the clamping structure is subjected to staged incremental displacement loading under experimental conditions to obtain the actual deformation state of the clamping structure. Based on the actual deformation state of the clamping structure, determine the additional deformation amount corresponding to the clamping structure before it reaches the geometric constraint limit with the strip it is located; The sum of the initial deformation and the additional deformation is taken as the maximum deformation in which the clamping structure will not fail during the test.

[0009] According to the technical solution provided in this application, the target proxy model is determined based on the changing trend of the load-displacement relationship curve, specifically including the following steps: If the load-displacement relationship curve shows a linear or weakly nonlinear trend, then a polynomial response surface model is selected as the target surrogate model. If the load-displacement relationship curve shows a strong nonlinear or piecewise change, and local features need to be accurately captured, then the Kriging model is selected as the target surrogate model. If the load-displacement relationship curve shows a highly nonlinear trend, then an artificial neural network model is selected as the target surrogate model.

[0010] According to the technical solution provided in this application, a sample set is extracted within the loading displacement range, and the target proxy model is trained based on the sample set to obtain a displacement-load mapping model. Specifically, the steps include: Within the specified loading displacement range, sample points are extracted using the Latin hypercube sampling method to obtain a sample set; Based on the load-displacement relationship curve, the load value corresponding to the displacement value of each sample point in the sample set is extracted to obtain a dataset; the dataset includes the displacement value and load value of multiple sample points. The target proxy model is trained using the dataset to obtain a displacement-load mapping model.

[0011] According to the technical solution provided in this application, the target proxy model is trained using the dataset to obtain a displacement-load mapping model, specifically including the following steps: The dataset is divided into a training set and a validation set; The target agent model is trained using the training set to obtain an initial model; The initial model is validated using the validation set to obtain validation results. The initial model is then adjusted based on the validation results until the model accuracy meets the preset requirements, thus obtaining the displacement-load mapping model.

[0012] According to the technical solution provided in this application, the test load standard data set includes at least multiple displacement points and load values ​​corresponding to the displacement points, as well as load-displacement relationship curves.

[0013] According to the technical solution provided in this application, the quasi-static loading test is performed on experimental equipment; The test equipment includes at least: A mechanical testing mechanism has a load-bearing section and a test section connected to each other. A movable guide groove is provided on the test section, and the extension direction of the movable guide groove is perpendicular to the surface of the load-bearing section. A drive cylinder is provided at the end of the movable guide groove away from the load-bearing section. A grid fixing structure is disposed on the bearing section and is used to clamp and position the grid; the positioning grid has multiple grid elements, and each grid element is provided with a clamping structure. A pressure head is connected to the drive end of the drive cylinder and located above the grid fixing structure; the pressure head is plugged into the grid element and is used to load the clamping structure of the positioning grid from its free state to the upper limit of the loading displacement range.

[0014] According to the technical solution provided in this application, the pressure head includes a fixed section and an installation section that are interconnected and vertically arranged; The fixed section is connected to the drive end of the drive cylinder, and the mounting section has a contact surface for contacting the inner wall surface of the grid element.

[0015] As can be seen from the above technical solution, this application has at least the following beneficial effects: This application provides a method for determining the load standard for nuclear fuel assembly erosion tests, comprising the following steps: obtaining the initial deformation of the clamping structure of the positioning grid; determining the maximum deformation of the clamping structure that will not fail during the test based on finite element simulation or cumulative displacement loading test methods and the initial deformation; constructing the loading displacement range required for the test by using the initial deformation as the lower limit and the maximum deformation as the upper limit; performing a quasi-static loading test on the positioning grid within the loading displacement range to obtain the load-displacement relationship curve; determining the target surrogate model based on the changing trend of the load-displacement relationship curve; extracting a sample set within the loading displacement range, training the target surrogate model based on the sample set to obtain a displacement-load mapping model; determining the test load standard data set corresponding to different types of positioning grids based on the displacement-load mapping model; and using the test load standard data set to guide different types of positioning grids in erosion tests with fuel rods.

[0016] This application utilizes the initial deformation of the quantified clamping structure and the maximum deformation without failure to construct the loading displacement range. Combined with quasi-static loading tests, the load-displacement relationship curve is obtained. Then, based on the changing trend of the load-displacement relationship curve, a suitable target proxy model is selected and a reliable displacement-load mapping model is formed through sample training. Finally, based on the displacement-load mapping model, a standard data set of test loads suitable for different types of positioning grids is formulated. This load determination scheme not only eliminates the dependence on traditional empirical data and ensures that the test load is highly matched with the real load under actual flow-induced vibration in the reactor, effectively avoiding the problems of specimen damage and data distortion caused by underestimation of the degree of abrasion or test overload, but also eliminates the need for customized special equipment, reducing the threshold for conducting tests and R&D costs. At the same time, it establishes a systematic and universal load standard formulation process, ensuring the comparability and consistency of test data from different research institutions and improving the reliability of nuclear fuel assembly abrasion tests. Attached Figure Description

[0017] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0018] Figure 1 A flowchart for determining the load standard for nuclear fuel assembly erosion tests.

[0019] Figure 2 This is a structural diagram of the experimental equipment.

[0020] Figure 3 for Figure 2 A magnified view of a portion of the image.

[0021] Figure 4 for Figure 3 Side view.

[0022] Figure 5 This is a structural diagram of the pressure head.

[0023] The diagram is labeled as follows: 1. Mechanical testing mechanism; 2. Indenter; 3. Grid fixing structure. Detailed Implementation

[0024] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0025] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0026] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first: Nuclear fuel assemblies are the core components of nuclear reactors, mainly composed of fuel rods and positioning grids. The positioning grids secure the fuel rods using clamping structures (such as springs and rigid protrusions). During reactor operation, fuel assemblies are subject to coolant flow-induced vibrations, leading to fretting abrasion between the fuel rods and the clamping structures. Long-term effects may cause cladding damage, threatening reactor safety. Currently, nuclear fuel assembly erosion tests suffer from several drawbacks: the determination of test loads relies excessively on empirical data or traditional parameters, making it difficult to match the mechanical characteristics and contact behavior of novel clamping structures; the test equipment is complex and costly, and there is a lack of unified and systematic load standard development methods, resulting in inconsistent test schemes among different institutions, poor data comparability, and even the potential for overload damage to the clamping structures due to unreasonable load design, affecting the effectiveness and reliability assessment of the tests.

[0027] In view of this, this application utilizes the initial deformation of the quantified clamping structure and the maximum deformation without failure to construct the loading displacement range. Combined with quasi-static loading tests, the load-displacement relationship curve is obtained. Then, based on the changing trend of the load-displacement relationship curve, a suitable target proxy model is selected and a reliable displacement-load mapping model is formed through sample training. Finally, based on the displacement-load mapping model, a standard data set of test loads suitable for different types of positioning grids is formulated. This load determination scheme not only eliminates the dependence on traditional empirical data and ensures that the test load is highly matched with the real load under actual flow-induced vibration in the reactor, effectively avoiding the problems of specimen damage and data distortion caused by underestimation of the degree of abrasion or test overload, but also eliminates the need for customized special equipment, reducing the threshold for conducting tests and R&D costs. At the same time, it establishes a systematic and universal load standard formulation process, ensuring the comparability and consistency of test data from different research institutions and improving the reliability of nuclear fuel assembly abrasion tests.

[0028] To make the method for determining the nuclear fuel assembly erosion test load standard provided in this application clearer and easier to understand, the method is described below with reference to the accompanying drawings. Figure 1 As shown in the figure, this is a method for determining the load standard for nuclear fuel assembly erosion tests, which includes the following steps: S100: Obtain the initial deformation of the clamping structure of the positioning grid.

[0029] Here, obtaining the initial deformation of the clamping structure of the positioning grid specifically includes the following steps: Obtain the manufacturing dimensions and tolerances of the positioning grid, strips, clamping structures, and fuel rods; Based on the calculation method of maximum and minimum entity size or the variance calculation method based on size deviation, the pre-deformation of the clamping structure when the fuel rod is inserted into the positioning grid cell is obtained, and the pre-deformation is used as the initial deformation of the clamping structure of the positioning grid.

[0030] The positioning grid is constructed by interlacing and welding strips together. The manufacturing dimension of the positioning grid refers to the grid pitch (i.e., the side length of the grid cell). The manufacturing dimension of the strip refers to the thickness of a single strip. The manufacturing dimension of the clamping structure refers to its height; for example, if the clamping structure is a spring, the manufacturing dimension is the free height of a single spring; if it is a rigid convex structure, the manufacturing dimension is the height of the rigid convex structure. The manufacturing dimension of the fuel rod refers to the diameter of the fuel rod. The tolerances of each component can be set according to actual requirements.

[0031] The basic calculation formulas for the maximum and minimum material size calculations and the variance calculation method based on size deviations are the same. Specifically, the calculation formulas are as follows: ; In the formula, This is the initial deformation amount. The diameter of the fuel rod. The grid spacing is the grid spacing. The height of the rigid convex part. For strip thickness, The free height of a single spring.

[0032] The specific calculation method for the maximum and minimum material dimensions is to take the maximum material dimension (i.e., the upper limit of the manufacturing dimensions of each component, such as...) as the maximum material dimension. Take the largest diameter, Substituting the minimum grid pitch into the above formula, the maximum pre-deformation is calculated; the minimum solid dimension (i.e., the lower limit of the manufacturing dimension of each component, such as...) is then used to calculate the maximum pre-deformation. Take the smallest diameter Substitute the maximum grid pitch into the above formula to calculate the minimum pre-deformation amount; use the maximum and minimum pre-deformation amounts to form a pre-deformation amount interval, and then select the middle value of the interval or select data according to the test requirements within the pre-deformation amount interval as the initial deformation amount.

[0033] The specific calculation method based on the variance calculation of size deviation is to extract the deviation between the actual size and the manufacturing size of each component, substitute the deviation value of each component into the above formula, and quantify the cumulative effect of error through variance calculation to obtain the pre-deformation fluctuation range. The initial deformation amount is selected within this pre-deformation fluctuation range. The selection method is not limited, as long as it can cover the pre-deformation state in most actual assembly scenarios.

[0034] The initial deformation refers to the initial compression amount when the clamping structure just comes into contact with the fuel rod after it has been inserted into the positioning grid. The initial deformation amount is determined by any of the above methods and serves as the starting point for subsequent test loading. The test loading must start from this deformation amount, rather than from the free state of the clamping structure, to avoid underestimating the degree of abrasion due to an excessively low load starting point, while also ensuring that it closely matches the initial contact state after actual assembly within the reactor.

[0035] S200. Based on the finite element simulation method or the cumulative displacement loading test method and the initial deformation, determine the maximum deformation of the clamping structure that will not fail in the test; and use the initial deformation as the lower limit and the maximum deformation as the upper limit to construct the range of loading displacement to be applied in the test.

[0036] The first method, based on finite element simulation and initial deformation, determines the maximum deformation that will prevent the clamping structure from failing during the experiment. This includes the following steps: Starting with the initial deformation, a finite element simulation of the clamping structure was performed, and the displacement loading of the clamping structure was gradually increased to obtain the simulation results. Based on the stress distribution of the clamping structure in the simulation results, determine the additional deformation when the maximum stress approaches the yield strength or fatigue limit of the clamping structure material. The sum of the initial deformation and the additional deformation is taken as the maximum deformation in which the clamping structure will not fail during the test.

[0037] The initial deformation was used as the initial displacement condition in the finite element simulation model. The installation boundary of the clamping structure was fixed to recreate its actual installation constraints within the positioning grid. A displacement-controlled incremental loading method was employed, starting from the initial deformation and increasing the displacement increment by a fixed amount each time. This continuously applied displacement load simulated the cumulative effect of fretting abrasion caused by in-core flow-induced vibration. By gradually loading, the stress distribution changes of the clamping structure at different deformation stages were observed, avoiding misjudgments due to sudden stress changes caused by a single loading. After each loading, stress cloud diagrams and stress value lists of the clamping structure were obtained, with a focus on stress concentration areas (such as the spring root and rigid contact points). The displacement increment was, for example, 0.05 mm / cycle. The stress cloud diagram was used to reflect the stress magnitude at various locations.

[0038] The critical failure state of a clamping structure is based on the material's yield strength or fatigue limit. When the maximum stress approaches the material's yield strength or fatigue limit, the clamping structure is deemed to be about to undergo plastic deformation or fatigue damage, and the corresponding deformation is the additional deformation. The material's yield strength or fatigue limit can be set according to actual conditions.

[0039] The sum of the initial deformation and the additional deformation is taken as the maximum deformation that will prevent the clamping structure from failing during the test. This maximum deformation is the upper limit of the safe deformation of the clamping structure during the test. The loaded displacement must not exceed the maximum deformation, otherwise it will lead to fracture, plastic instability, or fatigue failure of the clamping structure. By using the maximum deformation and the initial deformation to form the range of loading displacement required for the test, it is ensured that the test can cover the cumulative effect of fretting erosion in actual service, while avoiding damage to the specimen and premature termination of the test due to excessive load, thus ensuring the integrity and reliability of the test data.

[0040] The second method, based on the cumulative displacement loading test method and the initial deformation, determines the maximum deformation in which the clamping structure will not fail during the test. This includes the following steps: Starting from the initial deformation, the clamping structure is subjected to staged incremental displacement loading under experimental conditions to obtain the actual deformation state of the clamping structure. Based on the actual deformation state of the clamping structure, determine the additional deformation amount corresponding to the clamping structure before it reaches the geometric constraint limit between itself and the strip it is located; The sum of the initial deformation and the additional deformation is taken as the maximum deformation in which the clamping structure will not fail during the test.

[0041] The process involves first preloading the clamping structure according to the initial deformation, fixing the installation boundary of the positioning grid element sample (including the target clamping structure), and simulating its actual installation constraint within the positioning grid. Then, a staged incremental cumulative loading method is used. Starting from the initial deformation, a fixed small displacement increment is applied each time, for example, 0.1 mm / time, with a loading rate of for example, 0.05 mm / min. This simulates the cumulative effect of fretting abrasion caused by long-term flow-induced vibration within the stack. Through gradual loading, the actual deformation of the clamping structure is visually observed, capturing the entire process from normal deformation to near the constraint limit. Furthermore, after each stage of loading is completed, the loading is paused, and the actual deformation state of the clamping structure is obtained through visual observation or with the aid of simple measuring tools. Special attention is paid to any abnormal bending, local bulges, or changes in distance from the strip, while simultaneously recording the displacement values ​​at each stage.

[0042] When loading reaches a certain stage, if the clamping structure is about to contact the strip, that is, before the clamping structure reaches the geometric constraint limit with the strip, loading is stopped. The total displacement corresponding to this stage is the additional deformation. The geometric constraint limit refers to the vertical distance between the clamping structure and the strip, for example, 1.66 mm. This is the physical boundary at which the clamping structure can deform. Once it contacts the strip, subsequent deformation will cause a sharp increase in stress due to the abrupt change in constraint, which can easily lead to failure.

[0043] The sum of the initial deformation and the additional deformation is taken as the maximum deformation that the clamping structure will not fail in the test. This maximum deformation is the safe upper limit of the test load to ensure that the clamping structure will not break or become plastically unstable due to exceeding geometric constraints or overload in the test.

[0044] This method obtains actual deformation data through physical experiments, which is consistent with actual working conditions and is suitable for complex clamping structures that are difficult to model accurately. At the same time, the operation does not require complex customized equipment, which lowers the threshold for testing.

[0045] S300. Within the loading displacement range, a quasi-static loading test is performed on the positioning grid to obtain the load-displacement relationship curve.

[0046] Using experimental equipment in displacement control mode, the grid element specimen (including the target clamping structure) of the positioning grid is loaded from a free state to the upper limit of the loading displacement range. The loading speed is, for example, 0.05 mm / min. The load data corresponding to each displacement is recorded, and a complete load-displacement relationship curve is plotted. This curve can directly reflect the stiffness characteristics of the clamping structure, and the upper limit of the load corresponding to the maximum displacement can be read from the curve.

[0047] Specifically, the quasi-static loading test is performed on the test equipment, such as... Figure 2 , Figure 3 and Figure 4 As shown, the test equipment includes at least: Mechanical testing mechanism 1 has a load-bearing section and a test section connected to each other. A movable guide groove is provided on the test section. The extension direction of the movable guide groove is perpendicular to the surface of the load-bearing section. A drive cylinder is provided at the end of the movable guide groove away from the load-bearing section. The grid fixing structure 3 is installed on the bearing section and is used to clamp and position the grid; the positioning grid has multiple grid elements and the grid elements are equipped with clamping structures. The pressure head 2 is connected to the drive end of the drive cylinder and is located above the grid fixing structure 3; the pressure head 2 is plugged into the grid element and is used to load the clamping structure of the positioning grid from its free state to the upper limit of the loading displacement range.

[0048] The mechanical testing mechanism 1 provides a stable loading reference, guiding constraints, and power output. Specifically, the mechanical testing mechanism 1 has an interconnected load-bearing section and a test section. The load-bearing section is used to install the grid fixing structure 3 and forms a stable test reference surface, preventing data accuracy from being affected by base swaying during loading. The test section is connected to the load-bearing section, and the guide groove of the test section extends perpendicularly to the surface of the load-bearing section, providing guidance for the movement of the indenter 2. This ensures that the indenter 2 can move vertically up and down, preventing horizontal offset or tilting during loading and ensuring that the loading force is consistent with the force direction of the clamping structure. The drive cylinder is located at the end of the moving guide groove away from the load-bearing section (i.e., the upper part of the equipment), providing driving force to drive the indenter 2 to apply a displacement load downward. Here, the mechanical testing mechanism 1 is, for example, a universal testing machine, which is a well-known and mature device in the field and will not be described in detail here.

[0049] The grid fixing structure 3 is fixed to the reference surface of the bearing section, corresponding vertically to the pressure head 2, ensuring that the pressure head can accurately align with the positioning grid element. The grid fixing structure 3 is used to simulate the actual installation state of the positioning grid in the reactor, fix the sample position, and prevent the overall displacement or shaking of the positioning grid during loading. Here, there are no restrictions on the specific structure of the grid fixing structure 3; the positioning grid sample can be fixed by clamping, locking, or other methods.

[0050] Additionally, the pressure head 2 includes a fixed section and a mounting section that are interconnected and vertically arranged; the fixed section is connected to the drive end of the drive cylinder, and the mounting section has a contact surface for contacting the inner wall surface of the grid cell. The pressure head 2 is made of, for example, Q235 material, and its diameter is, for example, 7mm.

[0051] Specifically, such as Figure 5 As shown, B is the fixed section and C is the mounting section, forming an integrated structure perpendicular to each other, resembling an L-shape. The fixed section is columnar or plate-shaped, and its dimensions must match the interface of the drive cylinder's drive end (e.g., threaded connection, snap-fit ​​fixation) to ensure sufficient connection strength to withstand the maximum load during loading. The mounting section adapts to the internal space of the positioning grid element, achieving stable loading through precise contact between its bottom contact surface and the inner wall / clamping structure of the positioning grid element. Here, the contact surface is designed as a plane, and its dimensions must match the inner wall space of the positioning grid element to increase the contact area, avoid local stress concentration leading to damage to the clamping structure, and simultaneously form stable contact with the inner wall of the positioning grid element, ensuring uniform transmission of loading force.

[0052] The specific testing procedure of this experimental equipment is as follows: the positioning grid sample containing the target clamping structure is fixed on the grid fixing structure 3, ensuring that the grid element to be tested is aligned with the pressure head 2; the pressure head 2 is lowered by the drive cylinder, so that the lower end of the pressure head 2 is inserted into the positioning grid element and contacts the clamping structure in a free state (no load applied); then, a quasi-static loading test is performed, controlling the drive cylinder to slowly drive the pressure head downward at a preset speed (e.g., 0.05 mm / min) to apply displacement load to the clamping structure, gradually loading from the initial deformation to the maximum deformation; during the loading process, the equipment synchronously records the load data corresponding to each displacement, providing data support for subsequent plotting of the load-displacement relationship curve.

[0053] S400. Determine the target surrogate model based on the changing trend of the load-displacement relationship curve; and extract a sample set within the loading displacement range, train the target surrogate model based on the sample set, and obtain the displacement-load mapping model.

[0054] The process of determining the target surrogate model based on the changing trend of the load-displacement relationship curve includes the following steps: If the load-displacement relationship curve shows a linear or weakly nonlinear trend, then a polynomial response surface model should be selected as the target surrogate model. If the load-displacement relationship curve shows a strong nonlinear or piecewise change, and local features need to be accurately captured, then the Kriging model should be selected as the target surrogate model. If the load-displacement relationship curve shows a highly nonlinear trend, then an artificial neural network model should be selected as the target surrogate model.

[0055] It should be noted that the purpose of determining the target surrogate model is to accurately fit the displacement-load correspondence. Different models have different mathematical characteristics and are suitable for curves of varying complexity. This avoids the problem of using a simple model to fit a complex curve, resulting in large errors, or using a complex model to fit a simple curve, leading to wasted resources. Specifically, if the load-displacement relationship curve shows a linear or weakly nonlinear trend, it indicates that the displacement and load are approximately proportional, without obvious segmentation or fluctuations, and the overall trend is stable. Therefore, the polynomial response surface model is chosen as the target surrogate model. It has the advantages of simple principle and high computational efficiency, and can achieve ideal fitting results without a large number of samples, making it suitable for scenarios where the fitting requirements for local details are not high. If the load-displacement relationship curve shows a strongly nonlinear or segmented trend, it indicates that the displacement and load are not proportional, with obvious segmented trends or local fluctuations and inflection points. Accurate reproduction of detailed changes is required. Therefore, the Kriging model is chosen as the target surrogate model. The Kriging model is good at handling nonlinear and non-stationary data. Through kernel functions, it can flexibly fit trends of different segments, capture local fluctuation characteristics, and provide assessment of prediction uncertainty, resulting in high fitting accuracy. The load-displacement relationship curve exhibits an extremely strong nonlinear trend, indicating that the relationship between displacement and load is extremely complex and unpredictable. It may involve multiple abrupt changes, violent fluctuations, or irregular curve shapes (such as large oscillations or nonlinearity far exceeding piecewise changes). Therefore, an artificial neural network model is chosen as the target proxy model because it has a strong nonlinear fitting capability and can learn the intrinsic correlation of complex data through a multi-layer network structure, thus handling extremely strong nonlinear relationships with no obvious patterns.

[0056] By selecting a matching surrogate model based on the curve trend, it is ensured that the displacement-load mapping model constructed subsequently can truly reflect the stiffness characteristics and nonlinear behavior of the clamping structure. This ensures both the accuracy of the model prediction (such as accurately calculating the corresponding load for a specified displacement) and the computational efficiency and practicality, laying the foundation for the final formulation of accurate test load standards.

[0057] Furthermore, a sample set is extracted within the loaded displacement range, and a target surrogate model is trained based on the sample set to obtain a displacement-load mapping model. This process specifically includes the following steps: Within the loaded displacement range, sample points are extracted using the Latin hypercube sampling method to obtain a sample set; Based on the load-displacement relationship curve, the load value corresponding to the displacement value of each sample point in the sample set is extracted to obtain the dataset; the dataset includes the displacement value and load value of multiple sample points. The target surrogate model is trained using the dataset to obtain the displacement-load mapping model.

[0058] Here, Latin hypercube sampling is an efficient spatial sampling technique that extracts sample points within the loading displacement range that can fully cover the entire design space, resulting in a series of discrete displacement sample points that form a sample set. Each sample point corresponds to a specific displacement value within the loading displacement range.

[0059] Based on the aforementioned load-displacement relationship curve, the load value corresponding to each displacement sample point in the sample set is extracted, forming a one-to-one displacement-load data pair. For each displacement value in the sample set, the corresponding load value is found on the load-displacement relationship curve, forming a dataset containing multiple displacement values ​​and their corresponding load values.

[0060] By using a dataset, the target surrogate model learns the intrinsic relationship between displacement and load. Through iterative optimization, the model can accurately predict the load corresponding to any displacement, resulting in a displacement-load mapping model that has been verified for accuracy. This displacement-load mapping model can accurately reflect the stiffness characteristics and nonlinear behavior of the clamping structure and can be directly used for subsequent load standard formulation.

[0061] The process of training the target proxy model using the dataset to obtain the displacement-load mapping model includes the following steps: The dataset is divided into a training set and a validation set; The target agent model is trained using the training set to obtain the initial model; The initial model is validated using a validation set to obtain validation results. The initial model is then adjusted based on the validation results until the model accuracy meets the preset requirements, thus obtaining the displacement-load mapping model.

[0062] Here, the dataset is split into a training set and a validation set according to a preset ratio, for example, 8:2. The target proxy model adjusts the internal parameters of the rice noodle using the training set data to fit the intrinsic relationship between displacement and load, forming an initial prediction model. Furthermore, for different proxy models, the corresponding core parameters are adjusted; for example, the polynomial coefficients are optimized for the polynomial response surface model, the kernel function parameters are adjusted for the Kriging model, and the weights and biases are optimized for the artificial neural network model. Through iterative calculations, the model minimizes the error (such as mean square error) between the predicted load value and the actual load value in the training set, gradually locking in the optimal parameter combination, and finally obtaining an initial model that can initially reflect the displacement-load relationship.

[0063] Then, the predictive reliability of the initial model is tested using a validation set. If the error exceeds the preset requirement, the model parameters are adjusted in reverse until the accuracy meets the standard. Here, to avoid misjudgments caused by accidental sampling of the validation set, cross-validation, such as 5-fold validation, is required. The dataset is split into training and validation sets multiple times, and training, validation, and adjustment are repeated to ensure that the model can stably meet the standard on different data subsets. When the model's error on the validation set meets the preset requirement and the overall error of cross-validation is stable, a reliable displacement-load mapping model can be obtained. The preset requirement is, for example, a root mean square error < 1N. S500. Based on the displacement-load mapping model, determine the standard data set of test loads corresponding to different types of positioning grids; the standard data set of test loads is used to guide the abrasion test of different types of positioning grids and fuel rods.

[0064] Here, based on the load-displacement mapping model, a standard load data set adapted to different types of positioning grid element specimens is generated. This standard load data set includes at least multiple displacement points, corresponding load values, and load-displacement relationship curves. Test personnel can directly query the load value corresponding to a specified displacement from the test load data set without additional calculations, quickly determining loading parameters and ensuring loading accuracy. Simultaneously, test personnel can also quickly predict load change trends in different displacement ranges based on the load-displacement relationship curve, and use it as a benchmark for comparing test data to verify whether the actual test loading deviates from the standard curve. This solution provides unified and accurate load guidance for abrasion tests of various positioning grids and fuel rods, ensuring comparable test data and reliable results.

[0065] To facilitate understanding, the method for determining the load standard for nuclear fuel assembly erosion tests in this application will be introduced below with specific examples.

[0066] A single spring was chosen as the research object, and the diameter of the fuel rod was obtained. Grid spacing Rigid convex height strip thickness Single spring free height The initial deformation of a single spring is calculated using a variance calculation method based on dimensional deviations. ; That is, the initial deformation amount .

[0067] Initial deformation Based on this, cumulative displacement loading was applied to a single spring through preliminary experiments. Considering the vertical distance of 1.66 mm between the highest point of the single spring and the strip it was located on, the maximum deformation was determined. .

[0068] Using the initial deformation With maximum deformation Establish a loading displacement range, and determine the required loading displacement for the test based on the loading displacement range. ,in This experiment was conducted over the entire displacement range, i.e. .

[0069] Taking a single spring within a positioning grid cell and its adjacent cells as the research object, a universal testing machine was used to load the single spring from its free state to the displacement required for the test using an indenter. The loading speed was 0.05 mm / min, and the displacement and corresponding load data were recorded. A complete load-displacement curve was plotted, and the maximum loading displacement was read from the load-displacement curve. The load value corresponding to the time That is, the upper limit of the load of the single spring within the test displacement range.

[0070] Based on the load-displacement curve data mentioned above, 200 sample points were generated in the displacement interval [0, 1.5] using Latin hypercube sampling to ensure that the samples could fully cover the entire design space. The load-displacement curve was analyzed to show a trend similar to linear, yielding, and strengthening. Considering that the surrogate model needs to describe the nonlinearity and stage changes of the curve, capture local features, and provide prediction uncertainty, the Kriging model was selected as the target surrogate model and trained on the 200 sample points to construct the displacement-load mapping relationship model. By applying a displacement-load mapping model, a set of standard load data is developed to guide abrasion tests. For example, a single spring with a stiffness of 52.58 requires a load of [0, 54.51] N within a displacement of [0, 1.5] mm. Yielding occurs when the displacement is increased to approximately 0.7 mm, at which point the load fluctuates around 40 N. The strengthening stage begins when the displacement is increased to approximately 1.1 mm, and the load corresponding to a displacement of 1.5 mm is approximately 54.51 N. This set of standard load data is compiled into a queryable table, allowing users to look up the clamping force of the single spring and the vibration amplitude of the fuel rod at different stages of the abrasion test.

[0071] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for determining the load standard for nuclear fuel assembly erosion tests, characterized in that, Includes the following steps: Obtain the initial deformation of the clamping structure of the positioning grid; Based on the finite element simulation method or the cumulative displacement loading test method and the initial deformation, determine the maximum deformation of the clamping structure that will not fail during the test; The initial deformation amount is used as the lower limit value, and the maximum deformation amount is used as the upper limit value to construct the range of loading displacement required for the test; Within the specified loading displacement range, a quasi-static loading test is performed on the positioning grid to obtain the load-displacement relationship curve; Based on the changing trend of the load-displacement relationship curve, the target proxy model is determined; and a sample set is extracted within the loading displacement range, and the target proxy model is trained based on the sample set to obtain the displacement-load mapping model. Based on the displacement-load mapping model, a set of standard test load data corresponding to different types of positioning grids is determined; the set of standard test load data is used to guide different types of positioning grids and fuel rods in abrasion tests.

2. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, Obtaining the initial deformation of the clamping structure of the positioning grid includes the following steps: Obtain the manufacturing dimensions and tolerances of the positioning grid, strips, clamping structures, and fuel rods; Based on the calculation method of maximum and minimum entity size or the variance calculation method based on size deviation, the pre-deformation amount generated by the clamping structure when the fuel rod is inserted into the positioning grid cell is obtained, and the pre-deformation amount is used as the initial deformation amount of the clamping structure of the positioning grid.

3. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, Based on the finite element simulation method and the initial deformation, the maximum deformation of the clamping structure to prevent failure during the test is determined, specifically including the following steps: Starting from the initial deformation, a finite element simulation of the clamping structure is performed, and the displacement loading of the clamping structure is gradually increased to obtain the simulation results. Based on the stress distribution of the clamping structure in the simulation results, determine the additional deformation when the maximum stress approaches the yield strength or fatigue limit of the clamping structure material. The sum of the initial deformation and the additional deformation is taken as the maximum deformation in which the clamping structure will not fail during the test.

4. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, Based on the cumulative displacement loading test method and the initial deformation, the maximum deformation that prevents the clamping structure from failing during the test is determined, specifically including the following steps: Starting from the initial deformation, the clamping structure is subjected to staged incremental displacement loading under experimental conditions to obtain the actual deformation state of the clamping structure. Based on the actual deformation state of the clamping structure, determine the additional deformation amount corresponding to the clamping structure before it reaches the geometric constraint limit with the strip it is located; The sum of the initial deformation and the additional deformation is taken as the maximum deformation in which the clamping structure will not fail during the test.

5. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, Based on the changing trend of the load-displacement relationship curve, the target surrogate model is determined, specifically including the following steps: If the load-displacement relationship curve shows a linear or weakly nonlinear trend, then a polynomial response surface model is selected as the target surrogate model. If the load-displacement relationship curve shows a strong nonlinear or piecewise change, and local features need to be accurately captured, then the Kriging model is selected as the target surrogate model. If the load-displacement relationship curve shows a highly nonlinear trend, then an artificial neural network model is selected as the target surrogate model.

6. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, Extracting a sample set within the loaded displacement range, and training the target proxy model based on the sample set to obtain a displacement-load mapping model, specifically includes the following steps: Within the specified loading displacement range, sample points are extracted using the Latin hypercube sampling method to obtain a sample set; Based on the load-displacement relationship curve, the load value corresponding to the displacement value of each sample point in the sample set is extracted to obtain a dataset; the dataset includes the displacement value and load value of multiple sample points. The target proxy model is trained using the dataset to obtain a displacement-load mapping model.

7. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 6, characterized in that, The target proxy model is trained using the dataset to obtain a displacement-load mapping model, specifically including the following steps: The dataset is divided into a training set and a validation set; The target agent model is trained using the training set to obtain an initial model; The initial model is validated using the validation set to obtain validation results. The initial model is then adjusted based on the validation results until the model accuracy meets the preset requirements, thus obtaining the displacement-load mapping model.

8. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, The test load standard data set includes at least multiple displacement points and the load values ​​corresponding to the displacement points, as well as load-displacement relationship curves.

9. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 1, characterized in that, The quasi-static loading test was performed on the test equipment; The test equipment includes at least: A mechanical testing mechanism (1) has a load-bearing section and a test section connected to each other. A movable guide groove is provided on the test section. The extension direction of the movable guide groove is perpendicular to the surface of the load-bearing section. A drive cylinder is provided at the end of the movable guide groove away from the load-bearing section. A grid fixing structure (3) is provided on the bearing section and is used to clamp and position the grid; the positioning grid has multiple grid elements and a clamping structure is provided in each grid element. The pressure head (2) is connected to the drive end of the drive cylinder and is located above the grid fixing structure (3); the pressure head (2) is plugged into the grid element and is used to load the clamping structure of the positioning grid from its free state to the upper limit of the loading displacement range.

10. The method for determining the load standard for nuclear fuel assembly erosion tests according to claim 9, characterized in that, The pressure head (2) includes a fixed section and an installation section that are interconnected and arranged vertically; The fixed section is connected to the drive end of the drive cylinder, and the mounting section has a contact surface for contacting the inner wall surface of the grid element.

Images