A data-driven simulation method for flow characteristics and fluid excitation in-pile flow fields

Abstract

The invention discloses a method for simulating flow characteristics and fluid excitation of an in-pile flow field based on data-driven, comprising the following steps: establishing a static reduced-order model of the flow field of a lower head of a pressure vessel between structural design parameters and flow velocity distribution; establishing a fuel assembly channel reduced-order model for predicting the flow field distribution inside a fuel assembly, wherein the input of the fuel assembly channel reduced-order model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduced-order model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel; establishing a fuel assembly sub-channel flow field reduced-order model, wherein the input is the flow velocity distribution characteristics of a sub-channel and the structural characteristics inside a sub-channel, and the output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a sub-channel, and the fuel assembly sub-channel flow field reduced-order models of different sub-channels in the fuel assembly are connected in parallel to form a fuel assembly channel reduced-order model; and performing transient CFD simulation to calculate the turbulent force and flow elastic force based on the combined fuel assembly sub-channel flow field reduced-order model.

Description

Technical Field

[0001] The present invention relates to the field of nuclear reactor engineering technology, and more specifically to a data-driven in-core flow field flow characteristics and fluid excitation simulation method. Background Art

[0002] The core of a large pressurized water reactor is composed of tens of thousands of fuel rods, which are supported by multiple layers of grids covered with fins. In addition to the lower internal components such as the flow distribution device, the flow field in the reactor is extremely complex, and it is difficult to accurately obtain the flow characteristics and fluid excitation at each position in the flow field. Experimental research and CFD simulation are the two main means of obtaining in-core flow data. High-precision flow data can be obtained through experiments, but the experimental cost is high, and the number and location of measurement points are limited. Parallel and supercomputing technologies make CFD simulations useful for large-scale flow field analysis, but their computational cost is still not suitable for real-time health monitoring for digital twins. Summary of the invention

[0003] In order to solve the above problems of the shortcomings and defects in the prior art, the present invention provides a data-driven in-pile flow field flow characteristics and fluid excitation simulation method.

[0004] In order to achieve the above-mentioned purpose of the present invention, the technical scheme adopted is as follows:

[0005] A data-driven method for simulating flow characteristics and fluid excitation of an in-pile flow field, the method comprising the following steps:

[0006] S1: Establish a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity distribution, so as to output the core inlet flow velocity distribution characteristics under any given structural design parameters;

[0007] S2: Establishing a fuel assembly channel reduction model for predicting the flow field distribution inside the fuel assembly, wherein the input of the fuel assembly channel reduction model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduction model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel;

[0008] S3: Establish a fuel assembly subchannel flow field reduction model whose input is the velocity distribution characteristics of a subchannel and the structural characteristics in a subchannel, and whose output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a subchannel. The fuel assembly subchannel flow field reduction models of different subchannels in the fuel assembly are connected in parallel to form a fuel assembly channel reduction model;

[0009] S4: Perform transient CFD simulation to calculate turbulent force and flow elastic force based on the reduced-order model of the fuel assembly sub-channel flow field.

[0010] Preferably, establishing a static reduced-order model of the flow field of the lower head of a pressure vessel comprises the following steps:

[0011] S101: First find a set of orthogonal bases in the space Ψ spanned by the snapshots Make the collection The projection of the elements in on this set of bases is the largest, and the mathematical formula is:

[0012]

[0013] Where φ is the basis vector, U (i) is the snapshot, k is the count, i is the number of vectors;

[0014] S102: for CFD calculation results under different structural design parameters, the main flow modes are extracted by using intrinsic orthogonal decomposition and spectral intrinsic orthogonal decomposition in combination with formula (1);

[0015] S103: Based on the existing engineering experimental data as a benchmark, a data fusion method is used to calibrate the simulation data and correct the main flow mode;

[0016] S104: using intrinsic orthogonal decomposition and spectral eigenorthogonal decomposition to perform secondary decomposition on the main modes of the flow under different parameters, and calculating the weight corresponding to each main mode;

[0017] S105: according to the core inlet flow velocity under different structural design parameters, a flow velocity index at the core inlet is obtained by using statistical analysis;

[0018] S106: Based on the system identification method, a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity index is constructed to output the core inlet flow velocity distribution under any given design parameters.

[0019] Furthermore, when establishing the static reduced-order model of the flow field of the lower head of the pressure vessel, the calculation domain includes the water inlet pipe, the downcomer and the core; the core porous medium model is used to simulate the flow field of the lower head downstream;

[0020] The upstream core porous medium model and the static reduced-order model of the pressure vessel lower head flow field are connected in series to form the overall in-core flow field analysis model.

[0021] Preferably, the sample library of the fuel assembly channel reduced-order model is derived from CFD simulation calculations and visualization experiments.

[0022] Furthermore, the visualization experiment adopts an experiment body to conduct the experiment, and the experiment body includes an inlet section, an experiment section and an outlet section;

[0023] A grid plate is installed in the inlet section to reduce flow disturbance upstream of the experimental body;

[0024] An experimental piece is provided in the experimental section, and the experimental piece includes 1 row of 3 boxes of small components, each box of small components is a 4×4 rod bundle; the rod bundle is composed of a number of fuel rods arranged at equal intervals; a positioning grid with stirring wings is provided in the middle of the rod bundle.

[0025] Furthermore, the structural features within the sub-channels include the number and position of the grids, and the size, shape and inclination of the stirring vanes.

[0026] Furthermore, when establishing the fuel assembly sub-channel flow field reduction model, the sample library of the fuel assembly sub-channel flow field reduction model is derived from transient CFD simulation calculation results, sub-channel-level flow field observation results of fusion SPIV experiments, and existing experimental data and empirical data on fluid forces acting on fuel rods.

[0027] Furthermore, the specific steps of establishing the fuel assembly sub-channel flow field reduced-order model are as follows:

[0028] S301: Based on the finite propagation model and in accordance with the experimental body of the visualization experiment, transient CFD calculations of 3×3 sub-channels in a single group of 4×4 rod bundles are carried out to verify the finite propagation in the rod bundle structure and determine the turbulent pulsation propagation distance. Based on the calculated turbulent pulsation propagation distance, the number of sub-channels is minimized.

[0029] S302: According to the position and flow path of the sub-channel, the sub-channel is divided into two types: an intermediate sub-channel and a component peripheral sub-channel. For different types of sub-channels, turbulent force and flow elastic force are calculated through transient CFD simulation;

[0030] S303: according to the inlet velocity distribution of different sub-channels, the time-average value distribution, the pulsation amplitude distribution, and the pulsation frequency band distribution of the turbulent force are obtained by using statistical analysis and frequency domain method;

[0031] S304: Constructing a fuel assembly subchannel flow field reduced-order model based on a system identification method to output turbulent force characteristics of any given subchannel inlet flow velocity.

[0032] Furthermore, for the middle sub-channel, the turbulent force is calculated by transient CFD simulation, as follows:

[0033] The interface between the fuel rod and the flow field is set as a fixed boundary, and the LES model is used to carry out transient flow field simulation; the calculation time step is set to 1ms; during the simulation process, Euler measurement points are arranged in the middle sub-channel, flow snapshots are extracted, and typical working conditions are selected for comparison with the visualized experimental result data;

[0034] Several turbulent pulsation pressure monitoring points are set on the fuel rods and guide tube walls in the middle sub-channel to obtain the distribution law of turbulent force in the axial direction of the component, thereby completing the transient CFD simulation calculation of turbulent force.

[0035] Furthermore, for the peripheral sub-channels of the component, the flow elastic force is calculated through transient CFD simulation, as follows:

[0036] The fluid-structure coupling effect is characterized by two regions, namely, the first inter-component flow channel which is approximately the flow between the plates, and the second inter-component flow channel which is the flow between the rod bundles;

[0037] When calculating the flow elastic force through transient CFD simulation, the motion displacement and velocity of the structural interface in the peripheral sub-channel of the component are given, and the flow elastic force is calculated using dynamic mesh technology and converted into three fluid-solid coupling effects: additional stiffness, additional damping and additional mass. The fluid-solid coupling effect is determined through transient CFD simulation calculation and verified through experimental results of measuring the vibration characteristics of the rod bundle in air, still water and moving water.

[0038] The beneficial effects of the present invention are as follows:

[0039] The present invention proposes a data-driven method for studying the flow characteristics and fluid excitation of in-pile flow fields. Based on CFD simulation data and visualized experimental data, a static reduced-order model of the flow field of the lower head of a pressure vessel is constructed. Through the static reduced-order model of the flow field of the lower head of the pressure vessel, a fuel assembly channel reduced-order model and a fuel assembly sub-channel flow field reduced-order model are successively established. Based on the fuel assembly sub-channel flow field reduced-order model, transient CFD simulation is performed to calculate turbulent force and flow elastic force, which greatly reduces the calculation cost of complex flow fields in the pile and solves the problem that the flow characteristics and fluid excitation characteristics of complex flow fields are difficult to characterize. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of a data-driven in-pile flow field reduction model provided by an embodiment of the present invention.

[0041] Figure 2 It is a schematic diagram of SPIV optical imaging provided by an embodiment of the present invention.

[0042] Figure 3 It is a schematic diagram of the experimental entity provided by the embodiment of the present invention.

[0043] Figure 4 Schematic diagram of a 4×4 rod bundle test piece provided in an embodiment of the present invention.

[0044] Figure 5 It is a top view of a 4×4 rod bundle test piece provided in an embodiment of the present invention.

[0045] Figure 6is a top view of a grid provided in an embodiment of the present invention.

[0046] Figure 7 It is a schematic diagram of sub-channel flow field calculation provided by an embodiment of the present invention.

[0047] Figure 8 yes Figure 7 A partial enlarged view of point C in the middle.

[0048] Fig. 9 It is a schematic diagram of flow channels between adjacent components provided in an embodiment of the present invention.

[0049] Fig.10 yes Fig. 9 A partial enlarged view of point A in the middle.

[0050] Fig.11 yes Fig. 9 A partial enlarged view of point B in the middle. DETAILED DESCRIPTION

[0051] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Example 1

[0053] like Figure 1 As shown, a data-driven in-pile flow field flow characteristics and fluid excitation simulation method, the method comprising the following steps:

[0054] S1: Establish a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity distribution, so as to output the core inlet flow velocity distribution characteristics under any given structural design parameters;

[0055] S2: Establishing a fuel assembly channel reduction model for predicting the flow field distribution inside the fuel assembly, wherein the input of the fuel assembly channel reduction model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduction model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel;

[0056] S3: Establish a fuel assembly subchannel flow field reduction model whose input is the velocity distribution characteristics of a subchannel and the structural characteristics in a subchannel, and whose output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a subchannel. The fuel assembly subchannel flow field reduction models of different subchannels in the fuel assembly are connected in parallel to form a fuel assembly channel reduction model;

[0057] S4: Perform transient CFD simulation to calculate turbulent force and flow elastic force based on the reduced-order model of the fuel assembly sub-channel flow field.

[0058] Example 2

[0059] like Figure 1 As shown, a data-driven in-pile flow field flow characteristics and fluid excitation simulation method, the method comprising the following steps:

[0060] S1: Establish a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity distribution, so as to output the core inlet flow velocity distribution characteristics under any given structural design parameters;

[0061] S2: Establishing a fuel assembly channel reduction model for predicting the flow field distribution inside the fuel assembly, wherein the input of the fuel assembly channel reduction model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduction model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel;

[0062] S3: Establish a fuel assembly subchannel flow field reduction model whose input is the velocity distribution characteristics of a subchannel and the structural characteristics in a subchannel, and whose output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a subchannel. The fuel assembly subchannel flow field reduction models of different subchannels in the fuel assembly are connected in parallel to form a fuel assembly channel reduction model;

[0063] S4: Perform transient CFD simulation to calculate turbulent force and flow elastic force based on the reduced-order model of the fuel assembly sub-channel flow field.

[0064] In a specific embodiment, establishing a static reduced-order model of the flow field of the lower head of a pressure vessel includes the following steps:

[0065] S101: First find a set of orthogonal bases in the space Ψ spanned by the snapshots Make the collection The projection of the elements in on this set of bases is the largest, and the mathematical formula is

[0066]

[0067] Where φ is the basis vector, U (i) is the snapshot, k is the count, i is the number of vectors;

[0068] S102: Based on the CFD calculation results under different structural design parameters, the main flow modes are extracted by using intrinsic orthogonal decomposition and spectral intrinsic orthogonal decomposition in combination with formula (1); according to the CFD simulation data, the working condition with the largest flow rate is selected for comparison with the existing engineering experimental result data, and the calculation is repeated iteratively to obtain high-precision flow field calculation data;

[0069] S103: Based on the existing engineering experimental data as a benchmark, the CFD simulation data is calibrated using a data fusion method to correct the main flow mode;

[0070] S104: Use intrinsic orthogonal decomposition and spectral eigenorthogonal decomposition to perform secondary decomposition on the main modes of the flow under different parameters to obtain a more unified flow basic mode, and calculate the weight corresponding to each main mode; the weight refers to the proportion of a single main mode to the total modes.

[0071] S105: Statistical analysis is used to obtain core inlet flow velocity indicators for core inlet flow velocities under different structural design parameters; representative flow velocity indicators include time-averaged flow velocity distribution, flow velocity, and flow rate.

[0072] S106: Based on the system identification method, a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity index is constructed to output the core inlet flow velocity distribution under any given design parameters.

[0073] Furthermore, when establishing the static reduced-order model of the flow field of the lower head of the pressure vessel, in order to ensure that the flow inside the lower head has accurate upstream and downstream boundary conditions, the calculation domain includes the water inlet pipe, the downcomer and the core; the core flow field itself is very complex and is not the focus of the research on the reduced-order model of the lower head flow field. The core porous medium model is used to simulate the downstream lower head flow field;

[0074] The upstream core porous medium model and the static reduced-order model of the pressure vessel lower head flow field are connected in series to form the overall in-core flow field analysis model.

[0075] In this embodiment, the upstream core porous medium model and the static reduced-order model of the pressure vessel lower head flow field are connected in series, and the output of the core porous medium model is used as the input of the static reduced-order model of the pressure vessel lower head flow field, and they are combined together to form an overall in-core flow field analysis model.

[0076] Example 3

[0077] like Figure 1 As shown, a data-driven in-pile flow field flow characteristics and fluid excitation simulation method, the method comprising the following steps:

[0078] S1: Establish a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity distribution, so as to output the core inlet flow velocity distribution characteristics under any given structural design parameters;

[0079] S2: Establishing a fuel assembly channel reduction model for predicting the flow field distribution inside the fuel assembly, wherein the input of the fuel assembly channel reduction model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduction model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel;

[0080] S3: Establish a fuel assembly subchannel flow field reduction model whose input is the velocity distribution characteristics of a subchannel and the structural characteristics in a subchannel, and whose output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a subchannel. The fuel assembly subchannel flow field reduction models of different subchannels in the fuel assembly are connected in parallel to form a fuel assembly channel reduction model;

[0081] S4: Perform transient CFD simulation to calculate turbulent force and flow elastic force based on the reduced-order model of the fuel assembly sub-channel flow field.

[0082] The specific description of step S1 is shown in Example 2. This example specifically describes step S2:

[0083] The sample library of the fuel assembly channel reduced-order model is derived from CFD simulation calculations and visualization experiments (Stereoscopic Particle Image Velocimetry, SPIV).

[0084] In a specific embodiment, Figure 2 As shown, a visualization experiment SPIV is used to measure the flow field in a row of three rod bundle channels. The visualization experiment is conducted using an experimental body, which includes an inlet section, an experimental section, and an outlet section. Figure 3 , 4 , 5, and 6.

[0085] A grid plate is installed in the inlet section to reduce flow disturbance upstream of the experimental body;

[0086] The experimental section is provided with an experimental piece, which includes 1 row of 3 boxes of small components, each box of small components is a 4×4 rod bundle; the rod bundle is composed of a number of fuel rods arranged at equal intervals; the middle of the rod bundle is supported by a positioning grid with a turbulent wing. The height of the rod bundle is 1100mm, of which 520mm is upstream of the grid, 540mm is downstream of the grid, and the height of the grid is 40mm. The upper part of the rod bundle uses perfluoroethylene propylene copolymer (FEP) as the tube wall material, and the tube is filled with water. The refractive index compensation technology is used to solve the problem of the difference in refractive index between the rod bundle and water and the rod bundle scale effect. The visualization experiment does not consider the influence of structural movement on the flow field. The actuator connected to the middle rod bundle grid in the experimental body does not move, but plays the role of fixing the grid to keep the flow channel boundary stable as much as possible during the experiment. In a specific embodiment, the structural features in the subchannel include the number and position of the grids, and the size, shape and inclination of the turbulent wing.

[0087] Example 4

[0088] like Figure 1 As shown, a data-driven in-pile flow field flow characteristics and fluid excitation simulation method, the method comprising the following steps:

[0089] S1: Establish a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity distribution, so as to output the core inlet flow velocity distribution characteristics under any given structural design parameters;

[0090] S2: Establishing a fuel assembly channel reduction model for predicting the flow field distribution inside the fuel assembly, wherein the input of the fuel assembly channel reduction model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduction model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel;

[0091] S3: Establish a fuel assembly subchannel flow field reduction model whose input is the velocity distribution characteristics of a subchannel and the structural characteristics in a subchannel, and whose output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a subchannel. The fuel assembly subchannel flow field reduction models of different subchannels in the fuel assembly are connected in parallel to form a fuel assembly channel reduction model;

[0092] S4: Perform transient CFD simulation to calculate turbulent force and flow elastic force based on the reduced-order model of the fuel assembly sub-channel flow field.

[0093] The specific description of step S1 is shown in Example 2, and the specific description of step S2 is shown in Example 3. This example specifically describes step S2:

[0094] When establishing the fuel assembly sub-channel flow field reduction model, the sample library of the fuel assembly sub-channel flow field reduction model comes from transient CFD simulation calculation results, sub-channel level flow field observation results of SPIV experiments, and existing experimental data and empirical data on fluid forces acting on fuel rods.

[0095] In a specific embodiment, Figure 7 , 8 , 9, 10, 11, the specific steps of establishing the fuel assembly sub-channel flow field reduced-order model are as follows:

[0096] S301: Based on the finite propagation model and in accordance with the experimental body of the visualization experiment, transient CFD calculations of 3×3 sub-channels in a single group of 4×4 rod bundles are carried out to verify the finite propagation in the rod bundle structure and determine the turbulent pulsation propagation distance. Based on the calculated turbulent pulsation propagation distance, the number of sub-channels is minimized.

[0097] S302: According to the position and flow path of the sub-channel, the sub-channel is divided into two types: intermediate sub-channel and component peripheral sub-channel. For different types of sub-channels, the turbulent force and flow elastic force are calculated through transient CFD simulation. Through trial calculation and combining the existing experimental measured data and turbulent force engineering experience data, further correction is made on the basis of the turbulent force calculated by considering only the time-averaged flow velocity at the core inlet.

[0098] S303: For the inlet velocity distribution of different sub-channels, statistical analysis and frequency domain method are used to obtain the time-average value distribution, pulsation amplitude distribution, and pulsation frequency band distribution of turbulent force;

[0099] S304: Based on the system identification method, a fuel assembly subchannel flow field reduction model is constructed to achieve the output of the turbulent force characteristics of any given subchannel inlet flow velocity. The transient CFD simulation calculation results of the turbulent force are integrated with the observation data of the SPIV experimental subchannel level, and the existing fuel rod turbulent force PSD in the engineering project is used as data supplement.

[0100] In a specific embodiment, for the middle sub-channel, the turbulent force is calculated by transient CFD simulation as follows:

[0101] The interface between the fuel rod and the flow field is set as a fixed boundary, and the LES model is used to carry out transient flow field simulation; the calculation time step is set to 1ms; during the simulation process, Euler measurement points are arranged in the middle sub-channel, flow snapshots are extracted, and typical working conditions are selected for comparison with the visualized experimental result data to obtain high-precision flow data;

[0102] Several turbulent pulsation pressure monitoring points are set on the fuel rods and guide tube walls in the middle sub-channel to obtain the distribution law of turbulent force in the axial direction of the component, thereby completing the transient CFD simulation calculation of turbulent force.

[0103] In a specific embodiment, the flow elastic force is calculated for the peripheral sub-channel of the component through transient CFD simulation, as follows:

[0104] The fluid-structure coupling effect is characterized by two regions, namely, the first inter-component flow channel which is approximately the flow between the plates, and the second inter-component flow channel which is the flow between the rod bundles;

[0105] When calculating the flow elastic force through transient CFD simulation, the motion displacement and velocity of the structural interface in the peripheral sub-channel of the component are given, and the flow elastic force is calculated using dynamic mesh technology and converted into three fluid-solid coupling effects: additional stiffness, additional damping and additional mass. The fluid-solid coupling effect is determined through transient CFD simulation calculation and verified through experimental results of measuring the vibration characteristics of the rod bundle in air, still water and moving water.

[0106] Obviously, the above embodiments of the present invention are only examples for clearly explaining the present invention, and are not intended to limit the implementation methods of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. A data-driven method for simulating flow characteristics and fluid excitation in a pile, characterized by: The methods described include the following: S1: Establish a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the velocity distribution, so as to output the velocity distribution characteristics of the core inlet under any given structural design parameters; S2: Establishing a fuel assembly channel reduction model for predicting the flow field distribution inside the fuel assembly, wherein the input of the fuel assembly channel reduction model includes the flow velocity distribution characteristics on the fluid boundary of the fuel assembly channel and the structural characteristics inside the flow channel, and the output of the fuel assembly channel reduction model is the flow velocity distribution characteristics of each sub-channel inside the assembly channel; S3: Establish a fuel assembly subchannel flow field reduction model whose input is the velocity distribution characteristics of a subchannel and the structural characteristics in a subchannel, and whose output is the turbulent force and flow elastic force distribution characteristics acting on the boundary of a subchannel. The fuel assembly subchannel flow field reduction models of different subchannels in the fuel assembly are connected in parallel to form a fuel assembly channel reduction model; The specific steps to establish the reduced-order model of the fuel assembly sub-channel flow field are as follows: S301: Based on the finite propagation model and in accordance with the experimental body of the visualization experiment, transient CFD calculations of 3×3 sub-channels in a single group of 4×4 rod bundles are carried out to verify the finite propagation in the rod bundle structure and determine the turbulent pulsation propagation distance. Based on the calculated turbulent pulsation propagation distance, the number of sub-channels is minimized. S302: According to the position and flow path of the sub-channel, the sub-channel is divided into two types: an intermediate sub-channel and a component peripheral sub-channel. For different types of sub-channels, turbulent force and flow elastic force are calculated through transient CFD simulation; S303: For the inlet velocity distribution of different sub-channels, statistical analysis and frequency domain method are used to obtain the time-average value distribution, pulsation amplitude distribution, and pulsation frequency band distribution of turbulent force; S304: constructing a fuel assembly sub-channel flow field reduced-order model based on a system identification method to output turbulent force characteristics of any given sub-channel inlet flow velocity; S4: Perform transient CFD simulation to calculate turbulent force and flow elastic force based on the reduced-order model of the fuel assembly sub-channel flow field.

2. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 1 is characterized in that: Establishing a static reduced-order model of the flow field at the lower head of a pressure vessel includes the following steps: S101: First find a set of orthogonal bases in the space Ψ spanned by the snapshots Make the collection The projection of the elements in on this set of bases is the largest, and the mathematical formula is: Where φ is the basis vector, U (i) is the snapshot, k is the count, i is the number of vectors; S102: Based on the CFD calculation results under different structural design parameters, the main flow modes are extracted by using intrinsic orthogonal decomposition and spectral intrinsic orthogonal decomposition in combination with formula (1); S103: Based on the existing engineering experimental data as a benchmark, a data fusion method is used to calibrate the simulation data and correct the main flow mode; S104: using intrinsic orthogonal decomposition and spectral eigenorthogonal decomposition to perform secondary decomposition on the main modes of the flow under different parameters, and calculating the weight corresponding to each main mode; S105: according to the core inlet flow velocity under different structural design parameters, a flow velocity index at the core inlet is obtained by using statistical analysis; S106: Based on the system identification method, a static reduced-order model of the flow field at the lower head of the pressure vessel between the structural design parameters and the flow velocity index is constructed to output the core inlet flow velocity distribution under any given design parameters.

3. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 2 is characterized in that: When establishing the static reduced-order model of the flow field at the lower head of the pressure vessel, the computational domain includes the water inlet pipe, the downcomer and the core; the core porous medium model is used to simulate the flow field at the lower head downstream; The upstream core porous medium model and the static reduced-order model of the pressure vessel lower head flow field are connected in series to form the overall in-core flow field analysis model.

4. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 1 is characterized in that: The sample library of the fuel assembly channel reduced-order model is derived from CFD simulation calculations and visualization experiments.

5. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 4 is characterized in that: The visualization experiment adopts an experiment body to conduct the experiment, and the experiment body includes an inlet section, an experiment section and an outlet section; A grid plate is installed in the inlet section to reduce flow disturbance upstream of the experimental body; An experimental piece is provided in the experimental section, and the experimental piece includes 1 row of 3 boxes of small components, each box of small components is a 4×4 rod bundle; the rod bundle is composed of a number of fuel rods arranged at equal intervals; a positioning grid with stirring wings is provided in the middle of the rod bundle.

6. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 5 is characterized by: The structural features in the sub-channels include the number and position of the grids, and the size, shape and inclination of the stirring vanes.

7. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 6 is characterized in that: When establishing the fuel assembly sub-channel flow field reduction model, the sample library of the fuel assembly sub-channel flow field reduction model comes from transient CFD simulation calculation results, sub-channel level flow field observation results of SPIV experiments, and existing experimental data and empirical data on fluid forces acting on fuel rods.

8. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 7 is characterized in that: For the middle sub-channel, the turbulence force is calculated by transient CFD simulation as follows: The interface between the fuel rod and the flow field is set as a fixed boundary, and the LES model is used to carry out transient flow field simulation; the calculation time step is set to 1ms; during the simulation process, Euler measurement points are arranged in the middle sub-channel, flow snapshots are extracted, and typical working conditions are selected for comparison with the visualized experimental result data; A number of turbulent pulsation pressure monitoring points are set on the fuel rods and guide tube walls in the middle sub-channel to obtain the distribution law of turbulent force in the axial direction of the component; thereby completing the transient CFD simulation calculation of turbulent force.

9. The data-driven in-pile flow field flow characteristics and fluid excitation simulation method according to claim 7 is characterized in that: For the peripheral sub-channels of the components, the flow elastic force is calculated through transient CFD simulation, as follows: The fluid-structure coupling effect is characterized by two regions, namely, the first inter-component flow channel which is approximately the flow between the plates, and the second inter-component flow channel which is the flow between the rod bundles; When calculating the flow elastic force through transient CFD simulation, the motion displacement and velocity of the structural interface in the peripheral sub-channel of the component are given, and the flow elastic force is calculated using dynamic mesh technology and converted into three fluid-solid coupling effects: additional stiffness, additional damping and additional mass. The fluid-solid coupling effect is determined through transient CFD simulation calculation and verified through experimental results of measuring the vibration characteristics of the rod bundle in air, still water and moving water.

Images