A computational method for flow and heat transfer in multi-scale sandwich thermal protection structures

Abstract

The present application discloses a method for calculating the flow and heat transfer of a multi-scale sandwich thermal protection structure. Based on the hybrid thermal lattice Boltzmann method (HTLBM), the present application establishes an efficient prediction method for the multi-scale conduction-convection composite heat transfer of a sandwich thermal protection structure containing a complex real microstructure, taking into account the influence of external field aerodynamic heat. The method can quickly predict the thermal protection performance of the sandwich thermal protection structure, and provide technical support for the thermal protection design of aircraft and the reduction of structural redundancy.

Description

Technical Field

[0001] The present application belongs to the field of flow heat transfer calculation, and in particular, relates to a flow heat transfer calculation method for a multi-scale sandwich heat protection structure. Background Art

[0002] As aircraft continue to develop in the direction of ultra-high Mach, cross-airspace and wide speed range, the thermal environment in which they are in service is becoming increasingly severe. During high-speed flight, they will be subjected to extreme heat and pressure, causing serious thermal damage and thermal expansion to the structure and materials of the aircraft, thereby causing serious thermal problems.

[0003] The sandwich heat protection structure not only has a very high cooling efficiency when the coolant is injected, but also has a good heat insulation effect when the coolant is not introduced. It is widely used in thermal protection design. However, for a typical sandwich heat protection structure, it generally consists of a heat protection layer and a cooling layer. The heat protection layer requires the material to have the characteristics of light weight, high strength, and high temperature resistance. The cooling layer is generally filled with porous media. Different shapes of structures have a great influence on both heat insulation and active cooling. Reasonable structural optimization design is of great significance to improving its comprehensive heat transfer performance.

[0004] In addition, the microstructure scales of the heat shield and the cooling layer generally differ by one to two orders of magnitude, which poses a great challenge to the refined modeling and characterization of the structure. Therefore, in order to more accurately describe and evaluate the heat transfer performance of the active / passive heat shield structure, it is necessary to adopt structural multi-scale modeling and efficient refined calculation methods to consider the integrated calculation of the heat shield and the structural layer at different structural scales.

[0005] At the same time, the multi-scale flow and heat transfer performance of the sandwich heat protection structure is also affected by the service environment, including complex flow and heat transfer problems such as external aerodynamic heat-structure heat conduction coupling and internal heat conduction-convection composite heat transfer. These factors further increase the difficulty of accurately evaluating the multi-scale flow and heat transfer characteristics of the sandwich heat protection structure. Summary of the invention

[0006] The purpose of this application is to overcome the problems of the prior art and disclose a method for calculating the flow and heat transfer of a multi-scale sandwich thermal protection structure. Based on the hybrid thermal lattice Boltzmann method (HTLBM), this application establishes an efficient prediction method for the multi-scale conduction-convection composite heat transfer of a sandwich thermal protection structure containing a complex real microstructure under the condition of considering the influence of external field aerodynamic heat. The method can quickly predict the thermal protection performance of the sandwich thermal protection structure, provide technical support for the thermal protection design of aircraft and reduce structural redundancy.

[0007] The purpose of this application is achieved through the following technical solutions:

[0008] A multi-scale sandwich heat protection structure flow heat transfer calculation method, the multi-scale sandwich heat protection structure flow heat transfer calculation method comprising:

[0009] S1: Upscaling modeling of the heat shield layer. The modified deep convolutional generative adversarial network model is used to upscale the heat shield layer of the sandwich heat shield structure, and the REV model is established using the progressive analysis method.

[0010] S2: Downscaling modeling of the cooling layer, using the watershed algorithm of morphological analysis and statistical average analysis to downscale the anisotropic foam porous medium structure of the cooling layer, and using the Kelvin structure for equivalent modeling;

[0011] S3: Multi-scale modeling, based on the heat protection layer model obtained in step S1 and the cooling layer model obtained in step S2, a sandwich heat protection structure model integrating the heat protection layer and the cooling layer is constructed at the REV scale;

[0012] S4: Establish the external field aerodynamic thermal coupling mode, adopt the thermal wall correction formula, that is, the net heat flow entering the outer surface of the sandwich heat protection structure Q in for:

[0013]

[0014] in, Q c represents the ambient heat flow, T inf 、T w and T amb are the adiabatic wall temperature, structural wall temperature and ambient temperature; ε is the surface radiation heat dissipation coefficient; σ is the Stefan–Boltzmann constant;

[0015] S5: Flow heat transfer calculation,

[0016] Based on the sandwich heat protection structure with integrated heat protection layer and cooling layer established by S3, the external field aerodynamic thermal coupling method of S4 was introduced, and HTLBM was used to calculate the flow heat transfer of the sandwich heat protection structure.

[0017] First, the temperature field is iterated internally. If the internal iteration convergence condition is met, the flow field is iterated. When the iteration converges, the cooling efficiency and the outer surface temperature of the heat protection structure are output.

[0018] According to a preferred embodiment, step S1 includes: based on the modified deep convolutional adversarial generative network model, after establishing geometric models of different scales, using the asymptotic analysis method to measure the size of the geometric REV scale with three indicators of structural similarity, peak signal-to-noise ratio and root mean square error, and using the equivalent thermal conductivity as an indicator to measure the size of the heat transfer REV scale, thereby establishing a thermal protection layer REV model.

[0019] According to a preferred embodiment, the modified deep convolutional generative adversarial network model construction process includes: based on the deep convolutional generative adversarial network, introducing volume fraction loss and multi-scale samples, and establishing a modified deep convolutional generative adversarial network model with controllable porosity and structural continuity during upscaling.

[0020] According to a preferred embodiment, for the volume fraction loss, based on the cross entropy loss function of the deep convolution adversarial generative network itself, a volume fraction loss function is introduced:

[0021]

[0022] in, and are the volume fractions of the medium in the generated sample and the original sample, respectively;

[0023] Thus, the total loss of the deep convolutional adversarial network is:

[0024]

[0025] in, is the control parameter of volume fraction loss. Loss function for deep convolutional generative adversarial networks.

[0026] According to a preferred embodiment, in step S2, a watershed algorithm of morphological analysis is used to complete the cooling layer skeleton and pore characterization.

[0027] According to a preferred embodiment, in step S2, the Feret diameter of the pore structure is counted to extract the pore sizes in different directions of the structural parameters, and the Kelvin structure is used for equivalent modeling. By adjusting the structural parameters, a REV model of the same size as in S1 is obtained.

[0028] According to a preferred embodiment, during the flow heat transfer calculation process in step S5, several GPUs are used for parallel calculation.

[0029] The aforementioned main scheme of the present application and its further options can be freely combined to form multiple schemes, all of which are schemes that can be adopted and claimed for protection in the present application. After understanding the scheme of the present application, those skilled in the art can understand that there are multiple combinations based on the prior art and common knowledge, all of which are technical schemes to be protected by the present application, and they are not exhaustively listed here.

[0030] Beneficial effects of this application:

[0031] The multi-scale modeling method adopted in this application is highly versatile, and the materials of the heat protection layer and the cooling layer are not fixed, and can be applied to different types of sandwich heat protection structures.

[0032] When considering the aerodynamic thermal effects, the present application does not need to perform aerodynamic thermal calculations each time, compared with the direct bidirectional coupling method, which greatly reduces the data communication time; combined with HTLBM, compared with other methods, the HTLBM explicit propulsion and thermal wall correction formulas do not require data communication during calculations. Compared with CPU serial and CPU parallel calculations, the single-instruction multi-threaded parallelism of the GPU has more advantages, and is easier to implement efficient large-scale parallelism, greatly reducing the calculation time and improving the design efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the process of this application method;

[0034] Figure 2 It is a schematic diagram of the gap structure of the cooling layer;

[0035] Figure 3 It is a schematic diagram of aerodynamic thermal-structural coupling. DETAILED DESCRIPTION

[0036] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0037] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0038] In the description of the present application, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, or the positions or positional relationships in which the product of the application is usually placed when in use. They are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third", etc. are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance. In addition, the present application would like to point out that in the present application, if the specific structures, connection relationships, positional relationships, power source relationships, etc. involved are not specifically written, then the structures, connection relationships, positional relationships, power source relationships, etc. involved in the present application are all known to those skilled in the art on the basis of the prior art without creative work.

[0039] Example

[0040] A multi-scale sandwich heat protection structure flow heat transfer calculation method, the multi-scale sandwich heat protection structure flow heat transfer calculation method comprises the following steps.

[0041] Step S1: The heat shield layer is upscaled and modeled. The modified deep convolutional generative adversarial network (DCGAN) model is used to upscale the heat shield layer of the sandwich heat shield structure, and the REV model is established using the progressive analysis method.

[0042] Preferably, step S1 includes: based on the modified deep convolutional adversarial generative network model, after establishing geometric models of different scales, using the asymptotic analysis method, using three indicators of structural similarity (SSIM), peak signal to noise ratio (PSNR) and root mean square error (RMSE) to measure the size of the geometric REV scale, using the equivalent thermal conductivity as an indicator to measure the size of the heat transfer REV scale, and finally comprehensively considering the influence of the two to establish a thermal protection layer REV model.

[0043] Furthermore, the modified deep convolutional generative adversarial network model construction process includes: based on the deep convolutional generative adversarial network, introducing volume fraction loss and multi-scale samples, and establishing a modified deep convolutional generative adversarial network model with controllable porosity and structural continuity during upscaling.

[0044] Specifically, based on this modified deep convolutional adversarial generative network model, an efficient upscaling modeling method for heat shield materials is formed. For upscaling interfaces, multi-scale samples are introduced to eliminate the influence of interfaces. For volume fraction loss, based on the cross entropy loss function of the deep convolutional adversarial generative network itself, a volume fraction loss function is introduced:

[0045]

[0046] in, and are the volume fractions of the medium in the generated sample and the original sample, respectively;

[0047] Thus, the total loss of the deep convolutional adversarial network is:

[0048]

[0049] in, is the control parameter of volume fraction loss. Loss function for deep convolutional generative adversarial networks.

[0050] Step S2: Downscaling modeling of the cooling layer. The watershed algorithm of morphological analysis and statistical average analysis are used to downscale the anisotropic foam porous medium structure of the cooling layer, and the Kelvin structure is used for equivalent modeling.

[0051] Specifically, the cooling layer often uses foam porous materials. For this type of structure, the watershed algorithm of morphological analysis can be used to clearly characterize the skeleton and pores, such as Figure 2 This is a pore structure diagram of a typical foam porous material obtained by the watershed algorithm. By counting the Feret diameter of the pore structure in the red line, the pore sizes in different directions of the key structural parameters are extracted, and the Kelvin structure is used for equivalent modeling. By adjusting the structural parameters, a REV model of the same size as in step 1 can be obtained.

[0052] Step S3: multi-scale modeling, based on the heat protection layer model obtained in step S1 and the cooling layer model obtained in step S2, a sandwich heat protection structure model integrating the heat protection layer and the cooling layer is obtained at the REV scale. The multi-scale modeling method used in this application is highly versatile, and the materials of the heat protection layer and the cooling layer are not fixed, and can be applied to different types of sandwich heat protection structures.

[0053] Step S4: Establish the external field aerodynamic thermal coupling mode. For the aerodynamic thermal coupling problem, a simplified two-way coupling boundary is used to simulate the coupling between the external flow field aerodynamic heat and the structural temperature field, that is, a thermal wall correction method is used for two-way coupling calculation. The data flow interaction method adopts CN interaction, and the external field aerodynamic heat provides heat flow, and the structure returns the wall temperature, such as Figure 3 shown.

[0054] Thermal wall correction formula, that is, the net heat flux entering the outer surface of the sandwich heat protection structure Q in for:

[0055]

[0056] in, Q c represents the ambient heat flow, T inf 、T w and T amb are the adiabatic wall temperature, structural wall temperature and ambient temperature; ε is the surface radiation heat dissipation coefficient; σ is the Stefan–Boltzmann constant.

[0057] Step S5: flow heat transfer calculation. Based on the sandwich heat protection structure integrating the heat protection layer and the cooling layer established in S3, the external field aerodynamic thermal coupling method of S4 is introduced, and the HTLBM is used to perform flow heat transfer calculation on the sandwich heat protection structure. The temperature field and flow field of the heat protection layer and the cooling layer are iteratively calculated.

[0058] Furthermore, when the temperature field is internally iterated, if the internal iteration convergence condition is met, the flow field iteration calculation is performed, and when the iteration converges, the cooling efficiency and the outer surface temperature of the heat protection structure are output.

[0059] At the same time, multiple GPUs (Graphics Processing Units) are used to accelerate and improve computing efficiency. The multi-GPU segmentation method uses one-dimensional segmentation. The use of multiple GPUs for parallel acceleration can effectively reflect the aerodynamic thermal structure coupling characteristics and greatly improve computing efficiency.

[0060] When considering the aerodynamic thermal effects, the present application does not need to perform aerodynamic thermal calculations each time, compared with the direct bidirectional coupling method, which greatly reduces the data communication time; combined with HTLBM, compared with other methods, the HTLBM explicit propulsion and thermal wall correction formulas do not require data communication during calculations. Compared with CPU serial and CPU parallel calculations, the single-instruction multi-threaded parallelism of the GPU has more advantages, and is easier to implement efficient large-scale parallelism, greatly reducing the calculation time and improving the design efficiency.

[0061] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A multi-scale sandwich heat protection structure flow heat transfer calculation method, characterized in that: The multi-scale sandwich heat protection structure flow heat transfer calculation method comprises: S1: Upscaling modeling of the heat shield layer. The modified deep convolutional generative adversarial network model is used to upscale the heat shield layer of the sandwich heat shield structure, and the REV model is established using the progressive analysis method. S2: Downscaling modeling of the cooling layer, using the watershed algorithm of morphological analysis and statistical average analysis to downscale the anisotropic foam porous medium structure of the cooling layer, and using the Kelvin structure for equivalent modeling; S3: Multi-scale modeling, based on the heat protection layer model obtained in step S1 and the cooling layer model obtained in step S2, a sandwich heat protection structure model integrating the heat protection layer and the cooling layer is constructed at the REV scale; S4: Establish the external field aerodynamic thermal coupling mode, adopt the thermal wall correction formula, that is, the net heat flow entering the outer surface of the sandwich heat protection structure Q in for: in, Q c represents the ambient heat flow, T inf 、T w and T amb are the adiabatic wall temperature, structural wall temperature and ambient temperature; ε is the surface radiation heat dissipation coefficient; σ is the Stefan–Boltzmann constant; S5: Flow heat transfer calculation, Based on the sandwich heat protection structure with integrated heat protection layer and cooling layer established by S3, the external field aerodynamic thermal coupling method of S4 was introduced, and HTLBM was used to calculate the flow heat transfer of the sandwich heat protection structure. First, the temperature field is iterated internally. If the internal iteration convergence condition is met, the flow field is iterated. When the iteration converges, the cooling efficiency and the outer surface temperature of the heat protection structure are output.

2. The multi-scale sandwich heat protection structure flow heat transfer calculation method according to claim 1, characterized in that: Step S1 includes: Based on the modified deep convolutional generative adversarial network model, after establishing geometric models of different scales, the asymptotic analysis method is used to measure the size of the geometric REV scale with three indicators: structural similarity, peak signal-to-noise ratio and root mean square error. The equivalent thermal conductivity is used as an indicator to measure the size of the heat transfer REV scale, thereby establishing the thermal protection layer REV model.

3. The multi-scale sandwich heat protection structure flow heat transfer calculation method according to claim 1, characterized in that: The modified deep convolutional adversarial network model construction process includes: Based on the deep convolutional generative adversarial network, volume fraction loss and multi-scale samples are introduced to establish a deep convolutional generative adversarial network model with controllable porosity and correction of structural continuity during upscaling.

4. The multi-scale sandwich heat protection structure flow heat transfer calculation method according to claim 3, characterized in that: For the volume fraction loss, based on the cross entropy loss function of the deep convolution adversarial generation network itself, the volume fraction loss function is introduced: in, and are the volume fractions of the medium in the generated sample and the original sample, respectively; Thus, the total loss of the deep convolutional adversarial network is: in, is the control parameter of volume fraction loss, Loss function for deep convolutional generative adversarial networks.

5. The multi-scale sandwich heat protection structure flow heat transfer calculation method according to claim 1, characterized in that: In step S2, the watershed algorithm of morphological analysis is used to complete the cooling layer skeleton and hole characterization.

6. The multi-scale sandwich heat protection structure flow heat transfer calculation method according to claim 5, characterized in that: In step S2, By counting the Feret diameter of the pore structure, the pore sizes in different directions of the structural parameters were extracted, and the Kelvin structure was used for equivalent modeling. By adjusting the structural parameters, the REV model with the same size as in S1 was obtained.

7. The multi-scale sandwich heat protection structure flow heat transfer calculation method according to claim 1, characterized in that: In the process of performing flow heat transfer calculation in step S5, several GPUs are used for parallel calculation.

Images