Parameter design method of metal cavity cooling channel based on multi-physics field coupling

Abstract

The present invention relates to the technical field of geometric parameter design optimization. The present invention discloses a parameter design method for a metal cavity cooling channel based on multi-physical field coupling, comprising obtaining basic channel data according to cavity-related data and a detuning factor to provide a basis for the design of the cooling channel, constructing W initial channel data based on the basic channel data by presetting a channel design strategy, and simulating temperature tuning data to ensure that the cooling channel can effectively adjust the temperature to prevent thermal deformation and frequency detuning, and calculating the temperature tuning coefficient and the design influence value by using parameters such as the Prandtl number, the Reynolds number and the Nusselt number to provide a scientific basis for the optimized design, and optimizing the initial channel data by using an intelligent optimization algorithm to obtain the optimal channel data, which can significantly improve the stability and operating life of an RFQ accelerator, ensure the high-power continuous operation requirements, effectively improve the heat dissipation performance, and reduce the risk of equipment failure due to thermal effects.

Description

Technical Field

[0001] The present invention relates to the technical field of geometric parameter design optimization, and more specifically, to a method for designing metal cavity cooling channel parameters based on multi-physical field coupling. Background Art

[0002] Radio frequency quadrupole accelerator (RFQ) is an important device in particle accelerator for focusing and accelerating low-energy particle beams, and has wide applications in materials science, life science, nuclear medicine, etc. In the prior art, although relevant contents on the design of RFQ accelerator are disclosed, there are still some problems.

[0003] In the prior art, for example, a Chinese patent application with publication number CN114528784A provides a method for designing beam dynamics of a radio frequency quadrupole accelerator. The beam dynamics design of the radio frequency quadrupole accelerator adopts a nonlinear programming method to determine decision variables, constraint functions and objective functions. According to the parameters of the previous acceleration unit, the dynamic parameters of the current acceleration unit are calculated step by step to determine whether the outlet energy of the current acceleration unit reaches the synchronization energy required by the design. If not, the calculation of the next acceleration unit is continued. If so, the design is completed.

[0004] Although the prior art has disclosed relevant content about the design of radio frequency quadrupole accelerator (RFQ), the design research on its structural parameters, especially the structural parameters of the cooling channel, is still relatively limited. The cooling channel plays a vital role in RFQ. By regulating the cavity temperature through efficient heat dissipation, it can effectively prevent thermal deformation and frequency detuning, thereby improving the operating stability and service life of the equipment, while meeting the needs of high-power continuous wave operation. The cooling channel is an indispensable and important part of RFQ design. Therefore, optimizing the design of the cooling channel of the radio frequency quadrupole accelerator has become a key issue that needs to be solved urgently.

[0005] In view of this, the present invention proposes a metal cavity cooling channel parameter design method based on multi-physical field coupling to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a metal cavity cooling channel parameter design method based on multi-physical field coupling.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A parameter design method for metal cavity cooling channel based on multi-physics field coupling is provided, including:

[0009] Acquire cavity-related data, determine corresponding detuning factors according to the cavity-related data, input the cavity-related data and the detuning factors into a pre-built basic design model, and obtain channel basic data, wherein the channel basic data is basic data of a cooling channel in a radio frequency quadrupole accelerator;

[0010] Based on the channel basic data and the preset channel design strategy, W channel initial data are constructed, the W channel initial data are traversed, and the temperature tuning data are obtained by simulation according to the channel initial data;

[0011] Obtaining the Prandtl number and the Reynolds number, calculating the Nusselt number according to the Prandtl number and the Reynolds number, calculating the temperature tuning coefficient according to the temperature tuning data, and determining the design influence value according to the Nusselt number and the temperature tuning coefficient;

[0012] Based on the design influence value and intelligent optimization algorithm, the initial data of W channels are optimized to obtain the optimal channel data.

[0013] Furthermore, the cavity-related data at least includes a dipole resonance frequency, an acceleration resonance frequency, and a design resonance frequency. The method for determining a corresponding detuning factor according to the cavity-related data includes:

[0014] The corresponding detuning factor is calculated based on the dipole resonance frequency, the acceleration resonance frequency and the design resonance frequency.

[0015] Furthermore, the construction method of the basic design model includes:

[0016] A preset fully connected neural network is used as a basic model, the input layer in the fully connected neural network receives historical cavity related data and historical detuning factors, and the output layer in the fully connected neural network outputs historical channel basic data;

[0017] When training a fully connected neural network, the cross entropy loss function is selected as the loss function, and the loss function is minimized by the gradient descent method;

[0018] Update the weight parameters of the fully connected neural network and obtain the basic design model through iterative training.

[0019] Further, the method for constructing W channel initial data based on channel basic data and preset channel design strategy includes:

[0020] The variable range of each element in the channel basic data is evenly divided to obtain S variable intervals. Sampling is performed in each variable interval according to the Latin hypercube sampling method to obtain sampling values. The sampling values ​​corresponding to all elements are combined, and the process is repeated W times to construct W channel initial data.

[0021] Furthermore, the method for simulating and obtaining temperature tuning data according to the initial channel data includes:

[0022] A corresponding fluid dynamics model is constructed based on the initial channel data, the preset fluid-related data and the fluid dynamics model are input into the fluid simulation software, and the temperature tuning data is obtained by simulation according to the fluid simulation software.

[0023] Furthermore, the method of obtaining the Reynolds number includes:

[0024] The average velocity of the fluid, the diameter of the cooling channel, and the density of the fluid are obtained, and the Reynolds number is calculated according to the average velocity of the fluid, the diameter of the cooling channel, and the density of the fluid.

[0025] Furthermore, the intelligent optimization algorithm is a gray wolf optimization algorithm. Based on the design influence value and the intelligent optimization algorithm, the W channel initial data are optimized to obtain the optimal channel data. The method includes:

[0026] S401: randomly generate a gray wolf population, the population size is set to W, where each gray wolf represents a channel initial data;

[0027] S402: Calculate the design impact value for each gray wolf and rank the gray wolf populations according to the design impact value:

[0028] Alpha Gray Wolf: The gray wolf with the best design impact value;

[0029] Beta Gray Wolf: The gray wolf with the second best design impact value;

[0030] Delta Gray Wolf: The gray wolf with the third best design impact value;

[0031] The remaining gray wolves are considered ordinary individuals;

[0032] S403: The three alpha wolves, namely Alpha, Beta and Delta, update the position of each gray wolf in the gray wolf population;

[0033] S404: Calculate the design impact value for each updated gray wolf, re-rank according to the design impact value, and update the Alpha gray wolf, the Beta gray wolf, and the Delta gray wolf;

[0034] S405: Repeat the above S403-S404 until the number of iterations reaches a preset value, and use the initial channel data corresponding to the Alpha Gray Wolf as the optimal channel data.

[0035] Furthermore, the method for updating the position of each gray wolf in the gray wolf population by the three alpha gray wolf, beta gray wolf and delta gray wolf includes:

[0036] Calculate the average of the position vectors of the Alpha, Beta, and Delta alpha wolves, subtract the position vector of the t-th generation gray wolf from the average, and obtain the position vector of the t+1-th generation gray wolf.

[0037] Further, obtain The methods include:

[0038] Set the preset dynamic adjustment coefficient Multiply it with the preset prey's position vector, subtract the current position vector of the i-th gray wolf, and finally take its absolute value as , Characterized as The distance between a gray wolf and its prey.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention first obtains channel basic data according to cavity-related data and detuning factors to provide a basis for the design of the cooling channel. By presetting the channel design strategy, W channel initial data are constructed based on the channel basic data. By simulating the temperature tuning data, it is ensured that the cooling channel can effectively adjust the temperature to prevent thermal deformation and frequency detuning. The temperature tuning coefficient and design influence value are calculated using parameters such as Prandtl number, Reynolds number and Nusselt number to provide a scientific basis for the optimized design. The channel initial data are optimized through an intelligent optimization algorithm to obtain the optimal channel data, which can significantly improve the stability and operating life of the RFQ accelerator, ensure high-power continuous operation requirements, effectively improve the heat dissipation performance, and reduce the risk of equipment failure due to thermal effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of the method for designing parameters of metal cavity cooling channel based on multi-physical field coupling in the present invention;

[0042] Figure 2 It is a structural schematic diagram of a metal cavity cooling channel parameter design system based on multi-physical field coupling in the present invention;

[0043] Figure 3 is a schematic cross-sectional view of a cooling channel in the present invention;

[0044] Figure 4 FIG. 4 is a flow chart of a method for constructing initial data of W channels in the present invention.

[0045] Reference numerals:

[0046] 10. Channel body; 20. Blades; 30. Tuner. DETAILED DESCRIPTION

[0047] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings, but it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, it is obvious that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present invention.

[0048] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art. Unless otherwise defined, it should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0049] Embodiment 1:

[0050] See also Figure 1 As shown, this embodiment discloses a method for designing parameters of a metal cavity cooling channel based on multi-physical field coupling, including:

[0051] S10: Acquire cavity-related data, determine a corresponding detuning factor according to the cavity-related data, input the cavity-related data and the detuning factor into a pre-built basic design model, and obtain channel basic data, wherein the channel basic data is basic data of a cooling channel in a radio frequency quadrupole accelerator;

[0052] In this embodiment, the cavity-related data refers to the relevant data of the radio frequency cavity. The radio frequency cavity (RF cavity for short) is a key component in a particle accelerator. Its main function is to accelerate and control particles using a radio frequency electromagnetic field. The radio frequency cavity is a metal cavity of a specific shape, which can form a specific electromagnetic field distribution (such as a resonance mode of an electric field or a magnetic field) inside. By inputting radio frequency power, a high-frequency electromagnetic field will be generated in the cavity, which is used to accelerate, focus or correct the particle beam. It is worth noting that the radio frequency cavity is closely related to the design of the cooling channel behind it. This relationship is mainly reflected in the heat management and frequency stability generated during the operation of the radio frequency cavity. The optimized design of the cooling channel plays a key role in the stable operation of the radio frequency cavity.

[0053] It should be noted that the cavity-related data includes at least a dipole resonance frequency, an accelerating resonance frequency and a designed resonance frequency. The dipole resonance frequency refers to the resonance frequency of the dipole mode (Dipole Mode) in the RF cavity, that is, the natural oscillation frequency when the cavity supports the dipole mode electromagnetic field distribution. The accelerating resonance frequency refers to the resonance frequency of the accelerating mode (Accelerating Mode / TE210) in the RF cavity, that is, the natural oscillation frequency when the cavity supports the accelerating mode electromagnetic field distribution. The RF cavity is an electromagnetic resonator, which can support a variety of electromagnetic field distribution modes inside. Among them, the dipole mode and the accelerating mode are the two main modes inside the RF cavity. Their resonance frequencies and field distributions are determined by the geometric shape and design parameters of the RF cavity. Therefore, in this embodiment, the dipole resonance frequency and the accelerating resonance frequency can be predetermined by the geometric shape and design parameters of the RF cavity.

[0054] It should be added that the design resonant frequency refers to the expected operating frequency of the RF cavity in design and actual operation, that is, the driving frequency of the RF power input into the cavity. The design resonant frequency is the target operating frequency of the accelerator system, which is usually very close to or consistent with the resonant frequency of the acceleration mode (that is, the acceleration resonant frequency, such as the resonant frequency of the TE210 mode).

[0055] Methods for determining the corresponding detuning factor based on cavity related data include:

[0056] The corresponding detuning factor is calculated based on the dipole resonance frequency, the acceleration resonance frequency and the design resonance frequency.

[0057] Specifically include:

[0058] = ;

[0059] In the formula, is the detuning factor, is the quality factor of the RF cavity, is the dipole resonance frequency, To speed up the resonant frequency, To design the resonant frequency.

[0060] It should be noted that the quality factor of the RF cavity is related to the RF cavity material and the geometric design of the RF cavity. For example, highly conductive materials can significantly reduce the current loss on the cavity surface and improve the quality factor. Therefore, this embodiment can pre-detect the quality factor of the RF cavity.

[0061] The construction method of the basic design model includes:

[0062] A preset fully connected neural network is used as a basic model, the input layer in the fully connected neural network receives historical cavity related data and historical detuning factors, and the output layer in the fully connected neural network outputs historical channel basic data;

[0063] When training a fully connected neural network, the cross entropy loss function is selected as the loss function, and the loss function is minimized by the gradient descent method;

[0064] Update the weight parameters of the fully connected neural network and obtain the basic design model through iterative training.

[0065] In this embodiment, the detuning factor characterizes the degree of mode coupling between the dipole mode and the acceleration mode in the RF cavity, and the degree of interference caused by the dipole mode on the acceleration mode. The larger the detuning factor, the stronger the mode coupling between the dipole mode and the acceleration mode in the RF cavity, that is, the greater the interference of the dipole mode on the acceleration mode, and the worse the mode separation effect, which means that the operating stability of the RF cavity is reduced. A smaller detuning factor means that the RF cavity has a low tolerance for frequency drift and is easily affected by external factors such as thermal expansion and mechanical vibration during operation.

[0066] The basic channel data include channel length, channel cross-sectional area, tuner length and number of blades, such as Figure 3 As shown, Figure 3 is the cross-sectional view of the cooling channel, Figure 3The channel body 10, blades 20 and tuner 30, blade rotation degree Δθ, and tuner length PD are shown. The function of blade 20 is to optimize the electric field distribution of the RF cavity and realize the focusing and acceleration of the particle beam. The function of tuner 30 is to adjust the resonant frequency of the RF cavity and compensate for the frequency drift caused by thermal expansion or external interference. The blade rotation degree Δθ refers to the torsion angle of the blade (metal structure used to optimize the electromagnetic field distribution) in the cooling channel relative to the channel axis or the rotation angle difference between adjacent blades; this parameter directly affects the flow characteristics and heat exchange efficiency of the fluid in the channel; since blade 20 is the concentrated area of ​​high-power electromagnetic field in the RF cavity, its surface will carry a large amount of RF current, so the surface of blade 20 will emit a large amount of heat. Similarly, tuner 30 is closely connected to other components of the RF cavity and will absorb part of the heat from the cavity itself, so the tuning The tuner 30 will also emit heat, so it is necessary to have an efficient cooling design for the blades 20 and the tuner 30. The cooling design mainly uses the coolant to quickly take away the heat to avoid local overheating of the surfaces of the blades 20 and the tuner 30. Therefore, the tuner length and the number of blades are crucial. The longer the tuner, the larger its surface area, which helps the coolant to take away the heat more efficiently, thereby reducing the surface temperature of the tuner and ensuring its stable operation in a high temperature environment. Similarly, an increase in the number of blades can further evenly disperse the heat of each blade, ensuring a more uniform surface temperature, while allowing the coolant to more efficiently cover all blade surfaces, thereby improving the overall cooling effect. At the same time, the larger the blade rotation degree Δθ, the more complex the velocity distribution of the fluid in the channel will become, thereby generating a higher shear force, which helps to improve the heat exchange between the coolant and the blade surface and increase the heat conduction efficiency.

[0067] From the above content, it can be seen that the RF cavity is closely related to the design of the cooling channel. Therefore, when the detuning factor is small, it means that the RF cavity is easily affected by thermal expansion factors during operation, and the RF cavity needs to be cooled to ensure the frequency stability and uniformity of the electromagnetic field distribution of the RF cavity. Then for the design of the cooling channel, taking the channel length, channel cross-sectional area, tuner length and number of blades as examples, the channel length should be extended, the channel cross-sectional area should be increased, the tuner length should be increased, and the number of blades should be increased.

[0068] S20: constructing W channel initial data based on the channel basic data and the preset channel design strategy, traversing the W channel initial data, and simulating and obtaining temperature tuning data according to the channel initial data;

[0069] like Figure 4 As shown, the method for constructing W channel initial data based on channel basic data and preset channel design strategy includes:

[0070] The variable range of each element in the channel basic data is evenly divided to obtain S variable intervals. Sampling is performed in each variable interval according to the Latin hypercube sampling method to obtain sampling values. The sampling values ​​corresponding to all elements are combined, and the process is repeated W times to construct W channel initial data.

[0071] In this embodiment, Latin hypercube sampling is a statistical method for efficient sampling from a multidimensional design space. It ensures full coverage of each variable and avoids repeated sampling by dividing the value range of each variable into several uniform intervals and randomly selecting a sample point from each interval. The basic channel data including channel length, channel cross-sectional area, tuner length and number of blades are used as an example for illustration:

[0072] The variable ranges of each element are as follows:

[0073] Channel length: 0.5m-1m, channel cross-sectional area: 20cm²-30cm², tuner length: 6mm-10mm and number of blades: 10-30;

[0074] The range of each element variable is evenly divided into 5 intervals. The five intervals corresponding to the channel length are as follows: [0.5, 0.6], [0.6, 0.7], [0.7, 0.8], [0.8, 0.9], [0.9, 1.0]. The same applies to the channel cross-sectional area, tuner length and number of blades, which will not be described in detail in this embodiment.

[0075] Use Latin hypercube sampling to randomly select a value from each interval:

[0076] Channel length: randomly select values ​​from 5 intervals, for example, 0.54m, 0.63m, 0.78m, 0.87m, 0.92m. Similarly, the channel cross-sectional area, tuner length and number of blades are combined to obtain the initial channel data. The main purpose of constructing the initial channel data is to ensure that the value range of each variable can be fully and representatively sampled by evenly dividing the range of each variable and randomly sampling. This method can effectively avoid the problem of missing some areas of the design space due to uneven sampling. Compared with the traditional full-factor combination sampling (i.e., full combination of each value of each variable), the Latin hypercube sampling method has the advantage of significantly reducing the number of design points, while still being able to efficiently cover the entire design space, thereby reducing the computational complexity while ensuring comprehensiveness.

[0077] The method of simulating temperature tuning data according to the initial channel data includes:

[0078] A corresponding fluid dynamics model is constructed based on the initial channel data, the preset fluid-related data and the fluid dynamics model are input into the fluid simulation software, and the temperature tuning data is obtained by simulation according to the fluid simulation software.

[0079] It should be noted that the corresponding fluid dynamics model based on the initial channel data can be designed through the existing Solid Works software. The fluid-related data refers to parameters or conditions related to fluid flow, heat transfer and physical properties. For example, the fluid-related data include fluid density, fluid viscosity, fluid thermal conductivity and fluid specific heat capacity, etc. The fluid-related data also includes inlet boundary conditions. The inlet boundary conditions are used to describe the flow state and physical properties of the fluid at the inlet. The inlet boundary conditions include fluid velocity, flow rate and pressure values, etc. The fluid simulation software can be ANSYS Fluent. ANSYS Fluent is a professional computational fluid dynamics (CFD) software and is part of the ANSYS software suite. It is widely used in numerical simulation of fluid flow, heat transfer, turbulence and multiphase flow. In this embodiment, the role of ANSYS Fluent is to simulate the velocity field and pressure field of the fluid in the cooling channel, the heat transfer process in the cooling channel, and simulate the fluid temperature distribution, so as to obtain corresponding temperature tuning data.

[0080] It should be added that the fluid can be cooling water, which is the main heat transfer medium in the cooling channel. It carries away the heat generated by the RF cavity or other high-temperature components during operation through its flow, thereby maintaining the stable operation of the RF cavity. The contact surface between the cooling channel and the cooling water is the main area of ​​heat transfer. The cooling water conducts convection heat exchange by contacting the channel wall (such as the blade surface or wall).

[0081] S30: obtaining a Prandtl number and a Reynolds number, calculating a Nusselt number according to the Prandtl number and the Reynolds number, calculating a temperature tuning coefficient according to the temperature tuning data, and determining a design influence value according to the Nusselt number and the temperature tuning coefficient;

[0082] In this embodiment, the Prandtl number represents the relative relationship between the momentum diffusion capacity and the heat diffusion capacity of the fluid. The Prandtl number is a ratio of the physical properties of the fluid. The Reynolds number represents the flow characteristics of the fluid in a pipe or channel and is the ratio of the inertial force to the viscous force.

[0083] Methods for obtaining the Prandtl number include:

[0084] ;

[0085] in, is the Prandtl number, is the specific heat capacity of the fluid, is the thermal conductivity of the fluid, is the dynamic viscosity of the fluid.

[0086] Methods for obtaining the Reynolds number include:

[0087] Obtain the average fluid velocity, the cooling channel diameter and the fluid density value, and calculate the Reynolds number according to the average fluid velocity, the cooling channel diameter and the fluid density value;

[0088] Specifically include:

[0089] ;

[0090] in, is the Reynolds number, is the average velocity of the fluid, is the cooling channel diameter, is the fluid density value.

[0091] Methods for calculating the Nusselt number based on the Prandtl number and the Reynolds number include:

[0092] ;

[0093] in, is the Nusselt number.

[0094] It should be added that the Nusselt number characterizes the intensity of convective heat transfer, the ratio of the heat transferred through convection to the heat transferred through conduction alone, and reflects the relative intensity of the convective heat transfer capacity of the fluid in the cooling channel. In this embodiment, the size of the Nusselt number directly reflects the strength of the convective heat transfer capacity in the cooling channel. Therefore, it can be used as an important indicator to measure the rationality of the design of the channel initial data. The larger the Nusselt number, the stronger the convective heat transfer capacity in the cooling channel relative to the conductive heat transfer capacity. The enhanced convective heat transfer capacity means that the cooling channel design is more conducive to quickly taking away the heat generated during the operation of the RF cavity, thereby ensuring the uniformity of the temperature distribution and the stability of the operation of the cavity. Therefore, it means that the cooling channel design based on the channel initial data is more reasonable. It is worth noting that each of the W channel initial data corresponds to a Nusselt number.

[0095] In this embodiment, the temperature tuning data includes the blade fluid temperature, the wall fluid temperature and the total frequency change value. The blade fluid temperature refers to the fluid temperature of the blade area close to the cooling channel, the wall fluid temperature refers to the fluid temperature of the wall area close to the cooling channel, and the total frequency change value refers to the change in the RF resonance frequency caused by the change in the cooling fluid temperature of the blade and wall areas during the operation of the RF cavity (RF cavity).

[0096] Methods for calculating the temperature tuning coefficient based on the temperature tuning data include:

[0097] ;

[0098] In the formula, is the temperature tuning coefficient, is the total frequency change, is the blade fluid temperature, is the first standard temperature, is the wall fluid temperature, is the second standard temperature.

[0099] From the above content, it can be seen that the first standard temperature and the second standard temperature are reference temperatures for the blade fluid temperature and the wall fluid temperature respectively. These standard temperatures are usually determined according to preset conditions in the design stage or simulation of the cooling channel. In this embodiment, the temperature tuning coefficient characterizes the ability of the cooling fluid temperature change to adjust the stability of the RF cavity resonant frequency, reflecting the sensitivity of the cooling system design to the cavity frequency adjustment. The larger the temperature tuning coefficient, the more obvious the temperature change of the blade and wall area will significantly change the frequency of the RF cavity. Such sensitivity means that there are certain deficiencies in the design. The smaller the temperature tuning coefficient, the more obvious the cooling system can effectively control the impact of the cooling fluid temperature change on the cavity frequency, thereby enabling the RF cavity to maintain higher frequency stability during operation.

[0100] Methods for determining design influence values ​​based on the Nusselt number and temperature tuning coefficient include:

[0101] The reciprocal of the Nusselt number and the temperature tuning coefficient are weighted and summed to obtain the design influence value.

[0102] From the above content, it can be seen that the design impact value represents the comprehensive performance of the cooling channel design between the heat exchange capacity and the frequency regulation stability. When the design impact value is larger, it indicates that there are deficiencies in the cooling channel design and it cannot effectively meet the heat dissipation requirements or frequency stability requirements of the RF cavity. On the contrary, when the design impact value is smaller, it indicates that the cooling channel design is more reasonable and can effectively and quickly take away the heat generated during the operation of the RF cavity while maintaining the stability of the RF cavity frequency.

[0103] S40: Optimize the W channel initial data based on the design influence value and the intelligent optimization algorithm to obtain optimal channel data.

[0104] In this embodiment, the intelligent optimization algorithm is the gray wolf optimization algorithm. The method for optimizing the initial data of W channels based on the design influence value and the intelligent optimization algorithm to obtain the optimal channel data includes:

[0105] S401: Randomly generate a gray wolf population, and the population size is set to W, where each gray wolf represents a channel initial data.

[0106] It should be noted that initializing the entire optimization process and providing enough initial solutions ensures that the optimization process is not limited to a local area of ​​the design space.

[0107] S402: Calculate the design impact value for each gray wolf and rank the gray wolf populations according to the design impact value:

[0108] Alpha Gray Wolf: The gray wolf with the best design impact value;

[0109] Beta Gray Wolf: The gray wolf with the second best design impact value;

[0110] Delta Gray Wolf: The gray wolf with the third best design impact value;

[0111] The remaining gray wolves are considered ordinary individuals.

[0112] In this embodiment, the design impact value of each gray wolf is calculated and the gray wolves are ranked according to the quality of the design impact value. The design impact value measures the comprehensive performance of the heat exchange capacity and frequency regulation capacity of the initial data of the channel. It can be seen from the above content that the smallest design impact value represents the optimal design impact value.

[0113] S403: The three alpha wolves, namely Alpha, Beta and Delta, update the position of each gray wolf in the gray wolf population;

[0114] In this embodiment, the gray wolf gradually approaches the target value (prey) through position updating, realizes the gradual optimization of the channel design parameters, and ensures that the population continues to approach the global optimal solution.

[0115] S404: Calculate the design impact value for each updated gray wolf, re-rank according to the design impact value, and update the Alpha gray wolf, the Beta gray wolf, and the Delta gray wolf;

[0116] In this embodiment, after each position update, the design impact value of each gray wolf is recalculated, and the Alpha gray wolf, Beta gray wolf, and Delta gray wolf are reordered to ensure that the optimal solution in the current population can always correctly guide the population.

[0117] S405: Repeat the above S403-S404 until the number of iterations reaches a preset value, and use the initial channel data corresponding to the Alpha Gray Wolf as the optimal channel data.

[0118] Among them, the method for updating the position of each gray wolf in the gray wolf population around the three alpha gray wolf, Beta gray wolf and Delta gray wolf includes: calculating the average value of the position vectors of the Alpha alpha wolf, the Beta alpha wolf and the Delta alpha wolf, subtracting the position vector of the t-th generation gray wolf from the average value, and obtaining the position vector of the t+1-th generation gray wolf.

[0119] Specifically include:

[0120] ;

[0121] In the formula, Characterized as the In the iteration of the generation The position vector of the gray wolf, Represented as the position vector of Alpha gray wolf, Represented as the position vector of Beta gray wolf, Represented as the position vector of Delta Gray Wolf, Characterized as The distance between a gray wolf and its prey.

[0122] Get The methods include:

[0123] Set the preset dynamic adjustment coefficient Multiply it with the preset prey's position vector, subtract the current position vector of the i-th gray wolf, and finally take its absolute value as , Characterized as The distance between a gray wolf and its prey.

[0124] Specifically include:

[0125] ;

[0126] In the formula, is the dynamic adjustment coefficient, is the position vector of the preset prey, For the The current position vector of the wolf.

[0127] In this embodiment, the two core goals of cooling channel design (efficient heat dissipation and stable frequency) are combined by calculating the Nusselt number (characterizing the heat transfer capacity) and the temperature tuning coefficient (characterizing the frequency regulation sensitivity), ensuring that the channel design not only meets the high-power heat dissipation requirements but also maintains the operating stability of the RF cavity. The combination of the above steps realizes a closed-loop design from data generation to performance evaluation, making the optimization goals more comprehensive and realistic.

[0128] This embodiment optimizes the design of the cooling channel of the radio frequency quadrupole accelerator (RFQ). First, the basic channel data is obtained according to the cavity related data and the detuning factor to provide a basis for the design of the cooling channel. By presetting the channel design strategy, W initial channel data are constructed based on the basic channel data, and by simulating the temperature tuning data, it is ensured that the cooling channel can effectively adjust the temperature to prevent thermal deformation and frequency detuning. The temperature tuning coefficient and the design influence value are calculated using parameters such as the Prandtl number, the Reynolds number and the Nusselt number to provide a scientific basis for the optimized design. The initial channel data is optimized by the intelligent optimization algorithm to obtain the optimal channel data, which can significantly improve the stability and operating life of the RFQ accelerator, ensure the high-power continuous operation requirements, effectively improve the heat dissipation performance, and reduce the risk of equipment failure caused by thermal effects.

[0129] Embodiment 2:

[0130] See also Figure 2 As shown, based on the same inventive concept, this embodiment discloses a metal cavity cooling channel parameter design system based on multi-physical field coupling. For details not provided in this embodiment, please refer to the description of the relevant parts in Embodiment 1. The system includes:

[0131] Basic design module: used to obtain cavity-related data, determine the corresponding detuning factor according to the cavity-related data, input the cavity-related data and the detuning factor into the pre-built basic design model, and obtain channel basic data, wherein the channel basic data is the basic data of the cooling channel in the radio frequency quadrupole accelerator;

[0132] In this embodiment, the cavity-related data refers to the relevant data of the radio frequency cavity. The radio frequency cavity (RF cavity for short) is a key component in a particle accelerator. Its main function is to accelerate and control particles using a radio frequency electromagnetic field. The radio frequency cavity is a metal cavity of a specific shape, which can form a specific electromagnetic field distribution (such as a resonance mode of an electric field or a magnetic field) inside. By inputting radio frequency power, a high-frequency electromagnetic field will be generated in the cavity, which is used to accelerate, focus or correct the particle beam. It is worth noting that the radio frequency cavity is closely related to the design of the cooling channel behind it. This relationship is mainly reflected in the heat management and frequency stability generated during the operation of the radio frequency cavity. The optimized design of the cooling channel plays a key role in the stable operation of the radio frequency cavity.

[0133] Methods for determining the corresponding detuning factor based on cavity related data include:

[0134] The corresponding detuning factor is calculated based on the dipole resonance frequency, the acceleration resonance frequency and the design resonance frequency.

[0135] Specifically include:

[0136] = ;

[0137] In the formula, is the detuning factor, is the quality factor of the RF cavity, is the dipole resonance frequency, To speed up the resonant frequency, To design the resonant frequency.

[0138] The first processing module is used to construct W channel initial data based on the channel basic data and the preset channel design strategy, traverse the W channel initial data, and simulate the temperature tuning data according to the channel initial data;

[0139] The method for constructing W channel initial data based on channel basic data and preset channel design strategy includes:

[0140] The variable range of each element in the channel basic data is evenly divided to obtain S variable intervals. Sampling is performed in each variable interval according to the Latin hypercube sampling method to obtain sampling values. The sampling values ​​corresponding to all elements are combined, and the process is repeated W times to construct W channel initial data.

[0141] The method of simulating temperature tuning data according to the initial channel data includes:

[0142] The corresponding fluid dynamics model is constructed based on the initial channel data, the preset fluid-related data and the fluid dynamics model are input into ANSYS Fluent, and the temperature tuning data is obtained by simulation according to ANSYS Fluent.

[0143] The second processing module is used to obtain the Prandtl number and the Reynolds number, calculate the Nusselt number according to the Prandtl number and the Reynolds number, calculate the temperature tuning coefficient according to the temperature tuning data, and determine the design influence value according to the Nusselt number and the temperature tuning coefficient;

[0144] In this embodiment, the Prandtl number represents the relative relationship between the momentum diffusion capacity and the heat diffusion capacity of the fluid. The Prandtl number is a ratio of the physical properties of the fluid. The Reynolds number represents the flow characteristics of the fluid in a pipe or channel and is the ratio of the inertial force to the viscous force.

[0145] Methods for obtaining the Prandtl number include:

[0146] ;

[0147] in, is the Prandtl number, is the specific heat capacity of the fluid, is the thermal conductivity of the fluid, is the dynamic viscosity of the fluid.

[0148] Methods for obtaining the Reynolds number include:

[0149] Obtain the average fluid velocity, the cooling channel diameter and the fluid density value, and calculate the Reynolds number according to the average fluid velocity, the cooling channel diameter and the fluid density value;

[0150] Specifically include:

[0151] ;

[0152] in, is the Reynolds number, is the average velocity of the fluid, is the cooling channel diameter, is the fluid density value.

[0153] Methods for calculating the Nusselt number based on the Prandtl number and the Reynolds number include:

[0154] ;

[0155] in, is the Nusselt number.

[0156] Methods for calculating the temperature tuning coefficient based on the temperature tuning data include:

[0157] ;

[0158] In the formula, is the temperature tuning coefficient, is the total frequency change, is the blade fluid temperature, is the first standard temperature, is the wall fluid temperature, is the second standard temperature.

[0159] Optimization module: used to optimize the initial data of W channels based on the design influence value and intelligent optimization algorithm to obtain the optimal channel data.

[0160] The flow chart in the accompanying drawings illustrates the possible functions of the method according to various embodiments of the present invention. In this regard, each box in the flow chart can represent a module, a program segment, or a part of a code, and the above-mentioned module, a program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, the boxes represented by two connections can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the flow chart, and the combination of the boxes in the flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0161] Those skilled in the art will appreciate that the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, even if such combinations and / or couplings are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, and all such combinations and / or couplings fall within the scope of the present invention.

[0162] The embodiments of the present invention are described above, but these embodiments are only for illustrative purposes and are not intended to limit the scope of the present invention. Although each embodiment is described above separately, it does not mean that the measures in each embodiment cannot be used in combination to advantage. The scope of the present invention is limited by the attached embodiments and their equivalents. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A parameter design method for metal cavity cooling channel based on multi-physical field coupling, characterized in that: include: Acquire cavity-related data, determine corresponding detuning factors according to the cavity-related data, input the cavity-related data and the detuning factors into a pre-built basic design model, and obtain channel basic data, wherein the channel basic data is basic data of a cooling channel in a radio frequency quadrupole accelerator; Based on the channel basic data and the preset channel design strategy, W channel initial data are constructed, the W channel initial data are traversed, and the temperature tuning data are obtained by simulation according to the channel initial data; Obtaining the Prandtl number and the Reynolds number, calculating the Nusselt number according to the Prandtl number and the Reynolds number, calculating the temperature tuning coefficient according to the temperature tuning data, and determining the design influence value according to the Nusselt number and the temperature tuning coefficient; The reciprocal of the Nusselt number and the temperature tuning coefficient are weighted and summed to obtain the design influence value; Based on the design influence value and intelligent optimization algorithm, the initial data of W channels are optimized to obtain the optimal channel data.

2. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 1, characterized in that: The cavity-related data at least includes a dipole resonance frequency, an acceleration resonance frequency, and a design resonance frequency. The method for determining a corresponding detuning factor according to the cavity-related data includes: The corresponding detuning factor is calculated based on the dipole resonance frequency, the acceleration resonance frequency and the design resonance frequency.

3. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 1, characterized in that: The method for constructing the basic design model includes: A preset fully connected neural network is used as a basic model, the input layer in the fully connected neural network receives historical cavity related data and historical detuning factors, and the output layer in the fully connected neural network outputs historical channel basic data; When training a fully connected neural network, the cross entropy loss function is selected as the loss function, and the loss function is minimized by the gradient descent method; Update the weight parameters of the fully connected neural network and obtain the basic design model through iterative training.

4. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 1, characterized in that: The method for constructing W channel initial data based on channel basic data and preset channel design strategy includes: The variable range of each element in the channel basic data is evenly divided to obtain S variable intervals. Sampling is performed in each variable interval according to the Latin hypercube sampling method to obtain sampling values. The sampling values ​​corresponding to all elements are combined, and the process is repeated W times to construct W channel initial data.

5. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 4, characterized in that: The method for simulating and obtaining temperature tuning data according to initial channel data includes: A corresponding fluid dynamics model is constructed based on the initial channel data, the preset fluid-related data and the fluid dynamics model are input into the fluid simulation software, and the temperature tuning data is obtained by simulation according to the fluid simulation software.

6. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 1, characterized in that: The method for obtaining the Reynolds number includes: The average velocity of the fluid, the diameter of the cooling channel, and the density of the fluid are obtained, and the Reynolds number is calculated according to the average velocity of the fluid, the diameter of the cooling channel, and the density of the fluid.

7. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 1, characterized in that: The intelligent optimization algorithm is the Grey Wolf Optimization Algorithm. The method for optimizing the initial data of W channels based on the design influence value and the intelligent optimization algorithm to obtain the optimal channel data includes: S401: randomly generate a gray wolf population, the population size is set to W, where each gray wolf represents a channel initial data; S402: Calculate the design impact value for each gray wolf and rank the gray wolf populations according to the design impact value: Alpha Gray Wolf: The gray wolf with the best design impact value; Beta Gray Wolf: The gray wolf with the second best design impact value; Delta Gray Wolf: The gray wolf with the third best design impact value; The remaining gray wolves are considered ordinary individuals; S403: The three alpha wolves, namely Alpha, Beta and Delta, update the position of each gray wolf in the gray wolf population; S404: Calculate the design impact value for each updated gray wolf, re-rank according to the design impact value, and update the Alpha gray wolf, the Beta gray wolf, and the Delta gray wolf; S405: Repeat the above S403-S404 until the number of iterations reaches a preset value, and use the initial channel data corresponding to the Alpha Gray Wolf as the optimal channel data.

8. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 7, characterized in that: The method for updating the position of each gray wolf in the gray wolf population around the three alpha gray wolf, the beta gray wolf and the delta gray wolf includes: Calculate the average of the position vectors of the Alpha, Beta, and Delta alpha wolves, subtract the position vector of the t-th generation gray wolf from the average, and obtain the position vector of the t+1-th generation gray wolf.

9. The method for designing metal cavity cooling channel parameters based on multi-physical field coupling according to claim 8, characterized in that: Get The methods include: Set the preset dynamic adjustment coefficient Multiply it with the preset prey's position vector, subtract the current position vector of the i-th gray wolf, and finally take its absolute value as , Characterized as The distance between a gray wolf and its prey.

Images