A method for designing a ribbed cylindrical shell based on data-driven technology

Abstract

The present invention relates to the field of equipment design, and in particular to a method for designing a cylinder with ribbed plates based on data-driven technology. By establishing a basic database on the basis of existing technologies, an optimization basis can be provided for structural optimization design quickly and efficiently. At the same time, a heat transfer efficiency function based on the thickness and height of the ribbed plates is established by adopting a polynomial regression analysis method in machine learning, and an optimization design of the geometric parameters of the ribbed plates is performed. This method can achieve optimized geometric parameters of the ribbed plates that can improve the heat transfer efficiency of the ribbed plates while ensuring that the mechanical strength of the cylinder with ribbed plates meets the requirements. The method belongs to a technical solution for designing the ribbed plates of a cylinder with ribbed plates based on data-driven design. Compared with the empirical design method based on industry standards or enterprise standards in the existing technology, the method has stronger scientificity and can make up for the defects of the empirical design.

Description

Technical Field

[0001] The present invention relates to the field of equipment design, and in particular to a method for designing a ribbed plate cylinder based on data-driven technology. Background Art

[0002] Existing ribbed plate cylinder components primarily consist of a cylindrical inner cylinder, an outer cylinder, and several ribs radially connecting the inner and outer cylinders. These ribbed plate cylinder components are key structures for high-end equipment such as aerospace and weapons, and are also components of high-burnup spent fuel transport containers. While ensuring the structural strength of the components, the ribbed plate cylinders sometimes also need to dissipate heat from internal components. For example, when a high-burnup spent fuel transport container is loaded with high-temperature spent fuel, its external ribbed plate cylinder dissipates heat from the spent fuel rods to the environment. However, when designing these ribbed plate cylinders, the relevant geometric parameters are generally determined or verified using empirical formulas in industry or enterprise standards. The determination of design parameters can only be completed based on the designer's experience, and then verified through experiments, simulations, and other methods. This does not fully achieve an effective match between structural strength and heat transfer performance, affecting design efficiency and potentially leading to design defects that affect the performance of the ribbed plate cylinder components. Summary of the Invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art in which the design method of the ribbed plate cylinder components is based on expert experience, which affects the design efficiency and performance, and to provide a method for designing the ribbed plate cylinder based on data-driven technology.

[0004] The present invention provides a method for designing a ribbed plate cylinder based on data-driven technology, comprising the following steps:

[0005] S1. Establish a basic database of cylinder with ribbed plates; the basic database includes N corresponding cylinder geometric parameters, ribbed plate geometric parameters and ribbed plate heat transfer efficiency. A data set of , N is an integer greater than or equal to 100;

[0006] S2. Calculate the heat transfer efficiency difference ω of the rib plate for each data group;

[0007] ;

[0008] Where, is the maximum heat transfer efficiency of the ribbed plate in the basic database, The optimized heat transfer efficiency is calculated by using a heat transfer efficiency function based on rib thickness and rib height established by using polynomial regression analysis in machine learning;

[0009] S3. Build and optimize the database;

[0010] When ω≤m, the corresponding rib thickness, rib height and optimized heat transfer efficiency data are put into the optimization database, 0<m≤0.01;

[0011] When ω>m, adjust the rib thickness and rib height and recalculate the optimized heat transfer efficiency. and the difference between the rib plate heat transfer efficiency ω, until ω ≤ m, the corresponding rib plate thickness, rib plate height and optimized heat transfer efficiency data are put into the optimization database;

[0012] S4. Filter out the maximum value of the optimized heat transfer efficiency from the optimization database, and determine the corresponding rib thickness and rib height data as the geometric parameters of the optimized ribbed cylinder.

[0013] The method of the present invention is a method for designing a cylinder with ribbed plates based on data-driven technology. By establishing a basic database on the basis of existing technology, it can quickly and efficiently provide an optimization basis for structural optimization design. At the same time, by adopting the polynomial regression analysis method in machine learning to establish a heat transfer efficiency function based on the thickness and height of the ribbed plates, the optimization design of the geometric parameters of the ribbed plates is carried out. It can achieve the optimized geometric parameters of the ribbed plates that can improve the heat transfer efficiency of the ribbed plates on the basis of ensuring that the mechanical strength of the cylinder with ribbed plates meets the requirements. It is a technical solution for the ribbed plates of a cylinder with ribbed plates based on data-driven design. Compared with the empirical design method based on industry standards or enterprise standards in the existing technology, it is more scientific and can make up for the defects of empirical design.

[0014] Preferably, in step S2, the heat transfer efficiency function based on the rib thickness and rib height is:

[0015] ;

[0016] Where B is the thickness of the rib plate, L is the height of the rib plate, and is the coefficient given by the random forest algorithm in machine learning.

[0017] Preferably, in step S3, the method for adjusting the thickness of the rib plate includes adding u based on the last calculated parameter, and the method for adjusting the height of the rib plate includes adding u based on the last calculated parameter, 0<u≤0.01.

[0018] Preferably, in step S1, any two data groups have the same cylinder geometric parameter series and different rib geometric parameter series, the cylinder geometric parameters include cylinder height, outer cylinder thickness, inner cylinder radius and inner cylinder thickness, and the rib geometric parameters include rib thickness, rib height and adjacent rib distance; the rib thickness B, rib height L and adjacent rib distance d satisfy: 、 、 , is the thickness of the outer cylinder. Ensure that the mechanical strength of the cylinder with ribbed plates meets the application requirements.

[0019] Preferably, in step S1, a basic database is established by numerical calculation method, which specifically includes the following steps:

[0020] S1.1. Determine the geometric parameters of the cylinder of N cylinders with ribbed plates by using the existing design method, and determine the range of the distance between adjacent ribs.

[0021] S1.2. Keep the geometric parameters of the cylinder unchanged, randomly design the corresponding rib thickness and rib height, and randomly select the corresponding distance between adjacent ribs within the range of the distance between adjacent ribs.

[0022] S1.3. Establish a geometric model of the cylinder with ribbed plates, and determine the heat transfer efficiency of the ribs through numerical calculation.

[0023] S1.4. Organize the geometric parameters of the cylinder, the geometric parameters of the ribs and the heat transfer efficiency of the ribs correspondingly to obtain the basic database.

[0024] Collecting geometric parameters by numerical calculation method and calculating the heat transfer efficiency of the ribs is beneficial to improving the construction efficiency of the basic database and reducing the construction cost.

[0025] Preferably, in step S1.3, the heat transfer efficiency of the ribs satisfies:

[0026] ;

[0027] In the formula, is the heat transfer efficiency of the ribs, q is the actual heat transfer flux of the ribs, and Q is the theoretical heat transfer flux of the ribs.

[0028] Preferably, step S1.3 specifically includes the following steps:

[0029] S1.3.1. Establish a geometric model of the cylinder with ribbed plates;

[0030] S1.3.2. Divide the mesh of the geometric model in ANSYS-ICEM software;

[0031] S1.3.3. Import the geometric model with the divided mesh into ANSYS-FLUENT software, and calculate the actual heat transfer flux q of the ribs through the heat conduction model, heat radiation model and energy equation in ANSYS-FLUENT software;

[0032] S1.3.4. Calculate the theoretical heat transfer flux Q of the ribs;

[0033] ;

[0034] In the formula, is the thermal conductivity, A is the heat transfer area, is the temperature difference between the inner and outer cylinders of the cylinder with ribs;

[0035] S1.3.5. Calculate and determine the heat transfer efficiency of the ribbed plate.

[0036] Preferably, in step S1.4, the corresponding sorting includes removing duplicates from the acquired data, eliminating outliers, and filling missing values.

[0037] Preferably, the deduplication process includes removing adjacent duplicate data using a proximity sorting algorithm, the outlier removal process includes removing data with η≥1 and η≤0 using a maximum and minimum value method, and the missing value filling process includes supplementing missing values using a regression filling method.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] 1. The present invention provides a method for designing a ribbed cylinder based on data-driven technology. Building on existing technologies, a basic database is established to quickly and efficiently provide a foundation for structural optimization design. By using polynomial regression analysis in machine learning to establish a heat transfer efficiency function based on rib thickness and rib height, the rib geometric parameters are optimized. This method ensures that the mechanical strength of the ribbed cylinder meets requirements while obtaining optimized rib geometric parameters that enhance the rib heat transfer efficiency. This method represents a technical solution for data-driven design of ribbed cylinders.

[0040] 2. The present invention provides a method for designing a ribbed plate cylinder based on data-driven technology. Compared with the empirical design method based on industry standards or enterprise standards in the existing technology, the method is more scientific and can make up for the defects of empirical design. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The present invention is a flowchart of a method for designing a ribbed plate cylinder based on data-driven technology.

[0042] Figure 2 This is a schematic structural diagram of the ribbed plate cylinder described in Example 1.

[0043] Figure 3 This is a schematic diagram of the cross-sectional structure of the ribbed plate cylinder described in Example 1.

[0044] Figure 4 This is a schematic diagram of the cross-sectional structure of the ribbed plate cylinder described in Example 1.

[0045] Figure 5 Schematic diagram of the meshed geometric model in Example 1.

[0046] Markings in the figure:

[0047] 1 - Inner cylinder, 2 - Outer cylinder, 3 - Rib plate. Detailed implementation manners

[0048] The present invention will be further described in detail below in conjunction with specific embodiments. However, this should not be construed as limiting the scope of the above-mentioned subject matter of the present invention to the following embodiments. Any technology implemented based on the content of the present invention belongs to the scope of the present invention.

[0049] In the description of the specific embodiments of the present invention, without special instructions, the expression terms indicating the orientation or positional relationship such as "upper", "lower", "left", "right", "center", "inner", "outer", etc. are all based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship when the invention product / device / device is usually used and placed. These orientation or positional relationship terms are only for the convenience of describing the present invention solution or simplifying the description in the specific embodiments, so as to facilitate technicians to quickly understand the solution, rather than indicating or implying that a specific device / component / element must have a specific orientation, or be constructed and operated in a specific positional relationship. Therefore, it should not be construed as a limitation to the present invention.

[0050] In addition, if terms such as "horizontal", "vertical", "hanging", "parallel" appear, it does not mean that the corresponding device / component / element is required to be absolutely horizontal or vertical or hanging or parallel, but can be slightly inclined or have a deviation. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined. Or, it can be simply understood that the corresponding device / component / element is arranged in the directions of "horizontal", "vertical", "hanging", "parallel", etc., and can have an error / deviation of ±10% relative to the corresponding direction setting, more preferably within ±8%, more preferably within ±6%, more preferably within ±5%, more preferably within ±4%. As long as the corresponding device / component / element is within the error / deviation range, it can still achieve its function in the present invention solution.

[0051] In addition, the expressions such as "first", "second", "third", etc. in the terms are only used to distinguish the description of the same or similar components, and should not be construed as emphasizing or implying the relative importance of a specific component.

[0052] In addition, in the description of the embodiments of the present invention, "several", "multiple", "a number of" represent at least 2. It can be any situation such as 2, 3, 4, 5, 6, 7, 8, 9, etc., or even more than 9.

[0053] In addition, in the description of the technical solution of the present invention, unless otherwise clearly specified / defined / restricted, where the terms "set", "install", "connect", "couple", "provided with", "lay", "arrange" appear, they should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection, and can be welding, riveting, bolting, threaded connection, etc., which are common connection means in the art. Such a connection can be a mechanical connection, an electrical connection or a communication connection; it can be directly connected, or indirectly connected through an intermediate medium, and can be the communication inside two components.

[0054] Embodiment 1

[0055] As Figures 2 - 4 shown, the ribbed cylinder includes an inner cylinder, an outer cylinder, and ribs connected therebetween. The ribs play a role in connecting the inner cylinder and the outer cylinder, and at the same time have the function of bearing the heat dissipation of the internal components of the inner cylinder. The thickness and height of the ribs have an important impact on the heat dissipation efficiency of the ribs and the structural strength of the ribbed cylinder. A method for designing a ribbed cylinder based on data-driven technology in this embodiment is used to optimize the geometric parameters of the thickness and height of the ribs, so as to improve the heat transfer efficiency of the ribbed cylinder.

[0056] As Figure 1 shown, a method for designing a ribbed cylinder based on data-driven technology in this embodiment specifically includes the following steps S1 - S4.

[0057] S1. Establish a basic database of the ribbed cylinder; the basic database includes N data groups containing corresponding cylinder geometric parameters, rib geometric parameters, and rib heat transfer efficiency data, where N is an integer greater than or equal to 100. That is, the basic database includes N data groups, and each data group includes cylinder geometric parameters, rib geometric parameters, and rib heat transfer efficiency , and the geometric parameters and heat transfer efficiency correspond to each other. By establishing the basic database based on the existing technology in step S1, it can quickly and efficiently provide an optimization basis for the structural optimization design.

[0058] In an optional implementation, in step S1, relevant geometric parameter data and heat transfer efficiency data of the ribbed cylinder can be collected through sensors, and then the basic database can be constructed; or the cylinder geometric parameters of the ribbed cylinder can be designed based on the empirical design method of industry standards or enterprise standards, and then the rib geometric parameters and rib heat transfer efficiency can be optimized based on the cylinder geometric parameters. It is also possible to use numerical calculation methods to calculate and determine the rib heat transfer efficiency based on the known cylinder geometric parameters, and then establish the basic database.

[0059] In an alternative embodiment, in step S1, it is preferable to establish a basic database by using a numerical calculation method, collect the geometric parameters of the cylinder body by using a numerical calculation method, calculate the heat transfer efficiency of the rib plates, and form the basic database, which is beneficial to improving the construction efficiency of the basic database and reducing the construction cost.

[0060] In an alternative embodiment, step S1 establishes a basic database by using a numerical calculation method, which specifically may include the following steps S1.1 - S1.4.

[0061] S1.1. Determine the geometric parameters of the cylinder body with rib plates in N groups by using an existing design method, and determine the range of the distance between adjacent rib plates. d of.

[0062] Specifically, the cylinder height, the outer cylinder thickness , the inner cylinder radius R2, and the inner cylinder thickness of the cylinder body with rib plates can be determined based on any existing design method such as sensor collection, empirical design, etc., and the range of the distance d between adjacent rib plates is determined.

[0063] As Figure 4 shown, in the cross-section of the cylinder body with rib plates, the outer cylinder radius R1, the outer cylinder thickness , the inner cylinder radius R2, the inner cylinder thickness , the rib plate thickness B, the rib plate height L, and the distance d between adjacent rib plates are shown.

[0064] S1.2. Keep the geometric parameters of the cylinder body unchanged, randomly design the corresponding rib plate thickness B and rib plate height L, and randomly select the corresponding distance d between adjacent rib plates within the range of the distance between adjacent rib plates.

[0065] Specifically, it is possible to randomly design N rib plate thickness data and rib plate height data while keeping the cylinder height, the outer cylinder thickness , the inner cylinder radius R2, and the inner cylinder thickness unchanged, and randomly select N distances between adjacent rib plates within the range of the distance between adjacent rib plates determined in step S1.1, and ensure that the N data groups composed of the rib plate thickness, the rib plate height, and the distance between adjacent rib plates satisfy: , , , so that the mechanical strength of the designed cylinder body with rib plates meets the application requirements.

[0066] S1.3. For each data group determined in step S1.2, establish a geometric model of the cylinder body with rib plates, and determine the heat transfer efficiency of the rib plates through numerical calculation. The heat transfer efficiency of the rib plates satisfies:

[0067] -- Equation (1);

[0068] In the formula, η is the heat transfer efficiency of the rib plate, q is the actual heat transfer flux of the rib plate, and Q is the theoretical heat transfer flux of the rib plate.

[0069] Among them, the actual heat transfer flux can be obtained by establishing a geometric model of the cylinder with rib plates, and the theoretical heat transfer flux can be obtained by calculating with reference to the existing technology and combining structural parameters.

[0070] In an optional implementation manner, step S1.3 may specifically include the following steps:

[0071] S1.3.1: The geometric model of the cylinder with rib plates can be established by using Unigraphics NX software;

[0072] S1.3.2: The established geometric model can be imported into ANSYS-ICEM software, and the mesh of the geometric model is divided in ANSYS-ICEM software. The divided mesh needs to capture the important geometric features of the geometric model to ensure that the dimensional error between the model obtained by ANSYS-ICEM and the geometric model drawn by Unigraphics NX software is controlled within 5%, as Figure 5 shown;

[0073] S1.3.3: The geometric model with the divided mesh is imported into ANSYS-FLUENT software. Through the heat conduction model, heat radiation model and energy equation in ANSYS-FLUENT software, the actual heat transfer flux q of the rib plate of the cylinder with rib plates can be calculated;

[0074] S1.3.4: Calculate and determine the theoretical heat transfer flux Q of the rib plate by referring to the existing technology;

[0075] -- Equation 2;

[0076] In the formula, λ is the thermal conductivity, A is the heat transfer area, ΔT is the temperature difference between the inner cylinder and the outer cylinder of the cylinder with rib plates;

[0077] S1.3.5: Calculate and determine the heat transfer efficiency of the rib plate corresponding to 100 data groups.

[0078] S1.4: Based on the data groups obtained in step S1.2 and the heat transfer efficiency obtained in step S1.3, the geometric parameters of the cylinder, the geometric parameters of the rib plate and the heat transfer efficiency of the rib plate are sorted correspondingly, so that 100 groups of geometric parameter data and 100 heat transfer efficiencies of the rib plate correspond to each other, and are entered into an EXCEL table, and a basic database including 100 data groups can be obtained.

[0079] In an optional implementation manner, in step S1.4, the corresponding sorting may include deduplication processing, outlier removal processing, and missing value filling processing of the obtained data.

[0080] Specifically, the deduplication processing may include removing adjacent duplicate data by using the proximity sorting algorithm, the outlier removal processing includes removing data where η≥1 and η≤0 by using the maximum-minimum method, and the missing value filling processing includes supplementing missing values by using the regression filling method.

[0081] S2. Calculate the difference ω in the heat transfer efficiency of each rib plate. Step S2 is used to perform an optimization design based on the heat transfer efficiency for each set of rib plate thickness and rib plate height data in the basic database obtained in step S1.

[0082] Specifically, in step S2, first, a heat transfer efficiency function based on the rib plate thickness and rib plate height is established by using polynomial regression analysis in machine learning:

[0083] -- Equation Three;

[0084] In the formula, B is the rib plate thickness, L is the rib plate height, and are coefficients given by the random forest algorithm in machine learning; through this heat transfer efficiency function, the optimized heat transfer efficiency ;

[0085] Then, in combination with the calculation formula of the difference ω in the heat transfer efficiency of the rib plate:

[0086] -- Equation Four;

[0087] In the formula, is the maximum value of the heat transfer efficiency of the rib plate in the basic database, is the optimized heat transfer efficiency obtained by calculating through Equation Three; through this Equation Four, the corresponding difference ω in the heat transfer efficiency of the rib plate can be calculated, and by performing calculations for each data group in a loop, the difference ω in the heat transfer efficiency of each data group can be obtained.

[0088] S3. Build an optimized database. The optimized database contains optimized and corresponding rib plate thickness, rib plate height, and optimized heat transfer efficiency. The building of the optimized database is judged by analyzing the difference between the difference ω in the heat transfer efficiency of the rib plate and the set comparison value m.

[0089] In an optional implementation manner, 0 < m ≤ 0.01.

[0090] Specifically, when ω ≤ m, the rib thickness, rib height, and corresponding optimized heat transfer efficiency data as calculation parameters can be used as parameters for constructing an optimization database. Therefore, the corresponding rib thickness, rib height, and optimized heat transfer efficiency data can be put into the optimization database.

[0091] Specifically, when ω > m, adjust the rib thickness and rib height, and recalculate the optimized heat transfer efficiency and the rib heat transfer efficiency difference ω until ω ≤ m, and then put the corresponding rib thickness, rib height, and optimized heat transfer efficiency data into the optimization database.

[0092] In an optional implementation, the rib thickness can be increased by u based on the previous calculation parameters, and the rib height can be increased by u based on the previous calculation parameters. Then, substitute the adjusted rib thickness and adjusted rib height data into Equation 3 to recalculate the optimized heat transfer efficiency and substitute the optimized heat transfer efficiency obtained from the recalculation into Equation 4 to recalculate the rib heat transfer efficiency difference ω, and then compare it with the set comparison value m. Repeat the operations of adjustment, calculation, and comparison until ω ≤ m is satisfied, and then put the corresponding rib thickness, rib height, and optimized heat transfer efficiency data into the optimization database.

[0093] In an optional implementation, 0 < u ≤ 0.01.

[0094] S4. Screen out the maximum value of the optimized heat transfer efficiency from the optimization database, and determine the corresponding rib thickness and rib height data as the geometric parameters of the ribbed cylinder after optimization.

[0095] Specifically, first screen out the maximum value of the optimized heat transfer efficiency from the optimization database. This maximum heat transfer efficiency is the optimized heat transfer efficiency data. Then, find the corresponding rib thickness and rib height data for this maximum heat transfer efficiency, which are the optimized rib thickness and rib height data. After determining the optimized rib thickness and rib height data, combined with the known data of the inner cylinder radius, inner cylinder thickness, and adjacent rib distance, the parameters such as the outer cylinder radius can be determined, and the geometric parameter data of the ribbed cylinder after optimization can be obtained.

[0096] Taking a certain ribbed cylinder as an example, based on Step S1, using the existing design method to determine the geometric parameters of a certain ribbed cylinder including: cylinder height 50 cm, outer cylinder thickness cm, inner cylinder radius R2 = 447 cm, inner cylinder thickness cm. Based on the geometric parameters of the cylinder, 100 groups of data sets including the thickness of the rib plate, the height of the rib plate, and the distance between adjacent rib plates are randomly designed. After combination, 100 groups of data sets containing the corresponding geometric parameters of the cylinder and the rib plate are obtained. Each data set specifically includes the height of the cylinder, the thickness of the outer cylinder , the inner cylinder radius R2, the thickness of the inner cylinder , the thickness B of the rib plate, the height L of the rib plate, and the parameter of the distance d between adjacent rib plates. Based on step S2, the difference ω in the heat transfer efficiency of the rib plate corresponding to each data set is calculated. Based on step S3, with the comparison value m being 0.01 and the adjusted dimension u being 0.01, after checking each data set, an optimization database is obtained. From the optimization database, the optimized thickness of the rib plate, the height of the rib plate, and the optimized heat transfer efficiency can be selected. The comparison of the data before and after optimization is shown in Table 1 below:

[0097] Table 1 Optimization results of the geometric parameters of the rib plate of the cylinder with rib plates

[0098]

[0099] As can be seen from Table 1, by using the method for designing a cylinder with rib plates based on data-driven technology in this embodiment, the heat transfer efficiency of the rib plate of the cylinder with rib plates can be increased from 0.22 to 0.54, significantly improving the heat transfer efficiency of the rib plate of the cylinder with rib plates, and thus realizing the overall heat transfer efficiency of the cylinder with rib plates.

[0100] The method for designing a cylinder with rib plates based on data-driven technology in this embodiment establishes a heat transfer efficiency function based on the thickness and height of the rib plate by using polynomial regression analysis in machine learning, and conducts optimized design of the geometric parameters of the rib plate. It can obtain optimized geometric parameters of the rib plate that can improve the heat transfer efficiency of the rib plate on the basis of ensuring that the mechanical strength of the cylinder with rib plates meets the requirements. It belongs to the technical solution for designing the rib plate of the cylinder with rib plates based on data-driven design. Compared with the existing methods such as empirical design based on industry standards or enterprise standards, it has stronger scientificity, can make up for the defects of empirical design, and obtain the structural parameters of the cylinder with rib plates with higher heat transfer efficiency.

[0101] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for designing a ribbed plate cylinder based on data-driven technology, characterized in that: The following steps are involved: S1. Establish a basic database of cylinder with ribbed plates; the basic database includes N corresponding cylinder geometric parameters, ribbed plate geometric parameters and ribbed plate heat transfer efficiency. A data set of , N is an integer greater than or equal to 100; S2. Calculate the heat transfer efficiency difference ω of the rib plate for each data group; ； Where, is the maximum heat transfer efficiency of the ribbed plate in the basic database, The optimized heat transfer efficiency is calculated by using a heat transfer efficiency function based on rib thickness and rib height established by using polynomial regression analysis in machine learning; S3. Build and optimize the database; When ω≤m, the corresponding rib thickness, rib height and optimized heat transfer efficiency data are put into the optimization database, 0<m≤0.01; When ω>m, adjust the rib thickness and rib height and recalculate the optimized heat transfer efficiency. and the difference between the rib plate heat transfer efficiency ω, until ω ≤ m, the corresponding rib plate thickness, rib plate height and optimized heat transfer efficiency data are put into the optimization database; S4. Filter out the maximum value of the optimized heat transfer efficiency from the optimization database, and determine the corresponding rib thickness and rib height data as the geometric parameters of the optimized ribbed cylinder.

2. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 1, characterized in that: In step S2, the heat transfer efficiency function based on the rib thickness and rib height is: ； Where B is the thickness of the rib plate, L is the height of the rib plate, and is the coefficient given by the random forest algorithm in machine learning.

3. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 1, characterized in that: In step S3, the method for adjusting the thickness of the rib plate includes adding u based on the previous calculation parameters, and the method for adjusting the height of the rib plate includes adding u based on the previous calculation parameters, 0<u≤0.

01.

4. A method for designing a ribbed plate cylinder based on data-driven technology according to any one of claims 1 to 3, characterized in that: In step S1, any two data groups have the same cylinder geometric parameter series and different rib geometric parameter series. The cylinder geometric parameters include cylinder height, outer cylinder thickness, inner cylinder radius and inner cylinder thickness. The rib geometric parameters include rib thickness, rib height and adjacent rib distance. The rib thickness B, rib height L and adjacent rib distance d satisfy: 、 、 , is the thickness of the outer cylinder.

5. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 4, characterized in that: Step S1 uses a numerical calculation method to establish a basic database, which specifically includes the following steps: S1.

1. Use existing design methods to determine the geometric parameters of N cylinders with ribs and the range of distances between adjacent ribs; S1.

2. Keeping the cylinder geometry parameters unchanged, randomly design the corresponding rib thickness and rib height, and randomly select the corresponding adjacent rib distance within the range of adjacent rib distance; S1.

3. Establish a geometric model of the cylinder with ribbed plates and determine the heat transfer efficiency of the ribbed plates through numerical calculations. S1.

4. Arrange the cylinder geometric parameters, rib plate geometric parameters, and rib plate heat transfer efficiency to obtain a basic database.

6. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 5, characterized in that: In step S1.3, the heat transfer efficiency of the rib plate satisfies: ； Where, is the heat transfer efficiency of the rib plate, q is the actual heat transfer flux of the rib plate, and Q is the theoretical heat transfer flux of the rib plate.

7. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 6, characterized in that: Step S1.3 specifically includes the following steps: S1.3.

1. Establish the geometric model of the cylinder with ribbed plates; S1.3.

2. Mesh the geometric model in ANSYS-ICEM software; S1.3.

3. Import the meshed geometric model into ANSYS-FLUENT software and calculate the actual heat transfer flux q of the ribbed plate using the thermal conduction model, thermal radiation model, and energy equation in ANSYS-FLUENT software. S1.3.

4. Calculate the theoretical heat transfer flux Q of the ribbed plate; ； Where, is the thermal conductivity, A is the heat transfer area, is the temperature difference between the inner and outer cylinders of the cylinder with ribs; S1.3.

5. Calculate and determine the heat transfer efficiency of the ribbed plate.

8. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 6, characterized in that: In step S1.4, the corresponding sorting includes removing duplicates from the acquired data, eliminating outliers, and filling in missing values.

9. The method for designing a ribbed plate cylinder based on data-driven technology according to claim 8, characterized in that: The deduplication process includes using the neighbor sorting algorithm to remove adjacent duplicate data, the outlier removal process includes using the maximum and minimum value method to remove data with η≥1 and η≤0, and the missing value filling process includes using the regression filling method to supplement the missing values.

Images