Thermal protection structure design method based on thermal radiation directional regulation and control and thermal protection structure

Abstract

The invention discloses a thermal protection structure design method based on thermal radiation directional regulation and control and a thermal protection structure, and relates to the technical field of thermal protection of high-speed aircrafts. The method comprises the steps that photonic crystal structure design parameters are received and preprocessed, and establishment of a data set is completed; a bidirectional neural network model is constructed, the bidirectional neural network model is trained by using the established data set until the loss function is minimized, the trained bidirectional neural network model is obtained, and the bidirectional neural network model takes the target reflectivity of the photonic crystal as input and takes the predicted reflectivity of the photonic crystal as output; and evaluating the output predicted reflectivity by adopting a mean square error. The directional design of the reflectivity of the photonic crystal can be achieved through the bidirectional neural network, high reflectivity is achieved for thermal radiation of a specific wave band, directional radiation regulation and control of a thermal protection structure in a target application scene are achieved, and effective protection of an aircraft in a complex aerodynamic thermal environment in the wide-speed-domain and long-endurance service process is met.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal protection for high-speed aircraft, and in particular to a thermal protection structure design method and a thermal protection structure based on directional control of thermal radiation. Background Art

[0002] The aerodynamic heating environment of high-speed aircraft increases significantly, making existing thermal protection solutions difficult to meet. It is necessary to explore more effective design solutions. In high-temperature environments, multi-band thermal radiation is the primary mode of energy transfer. Radiation suppression can improve the thermal insulation capacity of thermal protection structures, necessitating the development of effective radiation control and targeted design methods.

[0003] At present, for the radiation control and suppression of thermal protection structures, infrared sunshades are generally introduced during the preparation of thermal insulation materials to increase the specific extinction coefficient, thereby enhancing the radiation performance of the structure. However, there is still a lack of effective reflection capability for the radiation energy entering the interior of the thermal protection structure from the outside, and the improvement of the infrared reflectivity is still insufficient. In addition, the design of the thermal protection structure lacks wavelength-selective modulation of the thermal radiation, and there is a lack of effective directional control mechanism for thermal radiation in complex multi-band ranges. Therefore, the present invention proposes a thermal protection structure design method and a thermal protection structure based on directional control of thermal radiation. Summary of the Invention

[0004] The purpose of the present invention is to provide a thermal protection structure design method and a thermal protection structure based on directional control of thermal radiation, which can realize the directional design of photonic crystal reflectivity through a bidirectional neural network, realize the directional radiation control of the thermal protection structure in the target application scenario, and effectively improve the thermal insulation capacity of the structure.

[0005] According to a first aspect of the present invention, in order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for designing a thermal protection structure based on directional control of thermal radiation, comprising the following steps:

[0006] Receive photonic crystal structure design parameters and perform preprocessing to complete the establishment of the data set;

[0007] Constructing a bidirectional neural network model, and training the bidirectional neural network model using the established data set until the loss function is minimized, thereby obtaining a trained bidirectional neural network model, wherein the bidirectional neural network model takes the target reflectivity of the photonic crystal as input and the predicted reflectivity of the photonic crystal as output;

[0008] The output predicted reflectivity is evaluated using the mean square error. If it meets the error requirements, the photonic crystal structure is designed based on the output predicted reflectivity, thereby obtaining a thermal protection structure design scheme. Otherwise, the output predicted reflectivity is iterated again until it meets the requirements.

[0009] Furthermore, the photonic crystal structure design parameters include material, the designed number of dielectric layers, and the designed thickness of the dielectric layers.

[0010] Furthermore, the photonic crystal structure design parameters are received and preprocessed to complete the establishment of the data set, as follows:

[0011] (31) Determine the value range of the design parameters of the photonic crystal structure;

[0012] (32) Latin hypercube sampling is used to select sample points within the value range:

[0013] The value range of each design parameter is divided into several equally probable subintervals, random sampling is performed in each subinterval to obtain the sampling points of each design parameter, and all the sampling points of the design parameters are randomly arranged and combined into a sample space;

[0014] (33) The reflectivity of the photonic crystal structure corresponding to the sample point is obtained using the transfer matrix method:

[0015] Establish the characteristic matrix of each layer of the photonic crystal medium:

[0016]

[0017] Where A, B, C, and D all represent matrices, d represents the thickness of the dielectric layer, the subscript l represents the dielectric layer, ch and sh represent the hyperbolic sine function and the hyperbolic cosine function, j is the imaginary unit, and β is expressed as:

[0018]

[0019] Where k represents the wave number, ε r represents the real part of the dielectric constant, θ represents the incident angle, tanδ represents the loss tangent, and ε i / ε r We get, where ε i represents the imaginary part of the dielectric constant;

[0020] Z is the intermediate representation for solving the A, B, C, and D matrices. The specific expressions are as follows:

[0021]

[0022] The total characteristic matrix of the photonic crystal structure is expressed as the product of the characteristic matrices of each dielectric layer:

[0023]

[0024] The characteristic matrix of the overall structure is obtained by the characteristic matrix of each dielectric layer using the transfer matrix method. The reflectivity R of the photonic crystal is expressed as:

[0025]

[0026] Furthermore, a bidirectional neural network model is constructed as follows:

[0027] (41) establishing a reverse design neural network model, wherein the reverse design neural network takes the target reflectivity of the photonic crystal as input and the photonic crystal structure design parameters as output;

[0028] (42) establishing a forward prediction neural network model, wherein the forward prediction neural network takes the photonic crystal structure design parameters as input and the photonic crystal predicted reflectivity as output;

[0029] (43) By connecting the reverse design neural network and the forward prediction neural network in series, a bidirectional neural network model is established.

[0030] Furthermore, the reverse design neural network model specifically includes an input layer, a hidden layer and an output layer. The input layer represents the target reflectivity and has a total of 199 units; the hidden layer has two layers, including 500 and 200 units; the output layer represents the structural design parameters and has a total of 5 units;

[0031] The forward prediction neural network specifically includes an input layer, a hidden layer and an output layer. The input layer represents the structural design parameters, with a total of 5 units; the hidden layer has 2 layers, including 200 and 500 units; the output layer represents the target reflectivity, with a total of 199 units.

[0032] Furthermore, the loss function is specifically as follows:

[0033] Loss=MSE(r,r') (6)

[0034] Where Loss represents the loss function, and MSE represents the mean square error between the input reflectivity and output reflectivity of the neural network model.

[0035] Furthermore, the mean square error is used to evaluate the output predicted reflectivity, as follows:

[0036]

[0037] Where MSE represents the mean square error between the target reflectivity and the output predicted reflectivity, r represents the target reflectivity, r' represents the output predicted reflectivity, and N represents the number of points obtained by discretizing the reflectivity.

[0038] According to a second aspect of the present invention, the present invention provides a high-speed aircraft thermal protection structure, which is designed using the above-mentioned thermal protection structure design method based on directional control of thermal radiation, and includes a substrate, the top of the substrate is connected to a thermal insulation layer 1, the top of the thermal insulation layer 1 is connected to a photonic crystal layer, the top of the photonic crystal layer is connected to a thermal insulation layer 2, the top of the thermal insulation layer 2 is connected to a composite material panel, and the top surface of the composite material panel is coated with a high-emissivity coating.

[0039] According to a third aspect of the present invention, a thermal protection structure design system based on directional thermal radiation regulation is provided, which is used to implement the above-mentioned thermal protection structure design method based on directional thermal radiation regulation, comprising:

[0040] A data set building module is used to receive the photonic crystal structure design parameters and perform preprocessing to complete the establishment of the data set;

[0041] A model building module is used to build a bidirectional neural network model, and train the bidirectional neural network model using the established data set until the loss function is minimized, thereby obtaining a trained bidirectional neural network model, wherein the bidirectional neural network model takes the target reflectivity of the photonic crystal as input and the predicted reflectivity of the photonic crystal as output;

[0042] The evaluation output module is used to evaluate the output predicted reflectivity using the mean square error. If the error requirements are met, the photonic crystal structure is designed based on the output predicted reflectivity, thereby obtaining a thermal protection structure design scheme. Otherwise, the predicted reflectivity is re-iterated until it meets the requirements.

[0043] According to a fourth aspect of the present invention, the present invention provides a terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that when the processor loads and executes the computer program, the above-mentioned thermal protection structure design method based on directional control of thermal radiation is adopted.

[0044] The present invention has at least the following beneficial effects:

[0045] 1. The thermal protection structure designed by the present invention can carry out targeted optimization of parameters such as the bandgap position, width and reflection band range of the photonic crystal, achieve high reflectivity for thermal radiation in a specific band, and can effectively protect the aircraft from complex aerodynamic thermal environments during service in a wide speed range and long flight time.

[0046] 2. The present invention adopts a bidirectional neural network framework to establish a directional design method for photonic crystal structures, taking the target reflectivity of the photonic crystal structure as input and the reflectivity predicted based on the directional design method as output. This can effectively solve the problem of multiple solutions for the design parameters corresponding to the target reflectivity of the photonic crystal, improve the design accuracy of the photonic crystal structure, obtain better structural design parameters, realize the directional design of the target bandgap characteristics, reduce the heat entering the interior of the thermal protection structure, and effectively improve the thermal insulation performance of the thermal protection structure.

[0047] 3. The directional design method proposed in this invention is combined with a data-driven framework, which can achieve enhanced red-band radiation heat dissipation at high temperatures or adjust reflection efficiency in specific flight speed ranges and airspaces, and has dynamic control capabilities.

[0048] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the design method of the present invention;

[0050] Figure 2 This is a schematic diagram of the framework of the design method of the present invention;

[0051] Figure 3 Schematic diagram of the structure of the bidirectional neural network model in Example 1 of the present invention;

[0052] Figure 4 This is a schematic diagram of a thermal protection structure for a high-speed aircraft in a second embodiment of the present invention;

[0053] Figure 5 This is a schematic diagram of the prediction results of the photonic crystal reflectivity of the present invention;

[0054] Figure 6 Schematic diagram of the comparison of the bottom surface temperature of the radiation-controlled thermal protection structure of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the accompanying drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the embodiments described are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present disclosure.

[0056] Example 1:

[0057] See also Figure 1-Figure 2 The present invention provides a technical solution: a thermal protection structure design method based on directional control of thermal radiation, comprising the following steps:

[0058] S1. Receive photonic crystal structure design parameters and perform preprocessing to complete the establishment of a data set;

[0059] Photonic crystal structure design parameters include material, number of designed layers, and designed thickness;

[0060] Preprocess the photonic crystal structure design parameters and complete the data set establishment, as follows:

[0061] (S11) determining the value range of the design parameters of the photonic crystal structure;

[0062] (S12) using Latin hypercube sampling to select sample points within the value range, dividing the value range of each design parameter into a number of equally probable subintervals, randomly sampling within each subinterval to obtain a sampling point for each design parameter, and combining the sampling points of all design parameters into a sample space by random permutation;

[0063] (S13) Using the transfer matrix method to obtain the reflectivity of the photonic crystal structure corresponding to the sample point, a characteristic matrix of each layer of the photonic crystal medium is established:

[0064]

[0065] Where d represents the thickness of the dielectric layer, the subscript l represents the dielectric layer, ch and sh represent the hyperbolic sine function and the hyperbolic cosine function, and j is the imaginary unit. β can be expressed as:

[0066]

[0067] Where k represents the wave number, ε r represents the real part of the dielectric constant, θ represents the incident angle, tanδ represents the loss tangent, and ε i / ε r We get, where ε i represents the imaginary part of the dielectric constant;

[0068] Z is the intermediate representation for solving the A, B, C, and D matrices. The specific expressions are as follows:

[0069]

[0070] The total characteristic matrix of the photonic crystal structure can be expressed as the product of the characteristic matrices of each dielectric layer:

[0071]

[0072] The characteristic matrix of the overall structure is obtained by the characteristic matrix of each dielectric layer using the transfer matrix method. The reflectivity of the photonic crystal can be expressed as:

[0073]

[0074] S2. Construct a bidirectional neural network model and train the bidirectional neural network model using the established data set until the loss function is minimized, thereby obtaining a trained bidirectional neural network model. The bidirectional neural network model takes the target reflectivity of the photonic crystal as input and the predicted reflectivity of the photonic crystal as output, as follows:

[0075] (41) establishing a reverse design neural network model, wherein the reverse design neural network takes the target reflectivity of the photonic crystal as input and the photonic crystal structure design parameters as output;

[0076] (42) establishing a forward prediction neural network model, wherein the forward prediction neural network takes the photonic crystal structure design parameters as input and the photonic crystal predicted reflectivity as output;

[0077] (43) By connecting the reverse design neural network and the forward prediction neural network in series, a bidirectional neural network model is established;

[0078] like Figure 3 As shown in the figure, the inverse design neural network model specifically includes an input layer, a hidden layer, and an output layer. The input layer represents the target reflectivity, with a total of 199 units; the hidden layer has two layers, including 500 and 200 units; the output layer represents the structural design parameters, with a total of 5 units;

[0079] The forward prediction neural network specifically includes an input layer, a hidden layer, and an output layer. The input layer represents the structural design parameters, with a total of 5 units; the hidden layer has 2 layers, including 200 and 500 units; the output layer represents the target reflectivity, with a total of 199 units;

[0080] Furthermore, the loss function is as follows:

[0081] Loss=MSE(r,r') (13)

[0082] Where Loss represents the loss function, MSE represents the mean square error between the input reflectivity and output reflectivity of the neural network model, r represents the target reflectivity, and r' represents the output predicted reflectivity;

[0083] S3. Evaluate the output predicted reflectivity using mean square error. If the output meets the error requirements, design a photonic crystal structure based on the output predicted reflectivity, thereby obtaining a thermal protection structure design. Otherwise, iterate the output predicted reflectivity again until it meets the requirements.

[0084] The mean square error is calculated as follows:

[0085]

[0086] Where MSE represents the mean square error between the target reflectivity and the output predicted reflectivity, r represents the target reflectivity, r' represents the output predicted reflectivity, and N represents the number of points obtained by discretizing the reflectivity.

[0087] Next, the technical solution of the present invention will be further described with reference to specific embodiments:

[0088] This embodiment uses Figure 4 The high-speed aircraft thermal protection structure shown in FIG. 1 is used as an example to illustrate the present invention. Figure 4 The radiation control thermal protection structure scheme shown in FIG uses a bidirectional neural network model to perform directional design of the photonic crystal layer. The established bidirectional neural network model is as follows: Figure 3 shown.

[0089] This embodiment realizes the directional design of the target reflectivity and obtains a photonic crystal structure consisting of 5 layers of dielectric materials. The reflectivity prediction results are as follows: Figure 5 As shown, the structural directional design carried out in this embodiment is aimed at the application of thermal protection structure under 1000K radiation load. Figure 6 Compared with the temperature of the traditional thermal protection structure, it can be seen that the method of the present invention can achieve directional design of target reflectivity and effectively improve the thermal insulation capacity of the traditional thermal protection structure.

[0090] In summary, the present invention adopts a bidirectional neural network framework to establish a directional design method for photonic crystal structures, which can effectively solve the problem of multiple solutions of design parameters corresponding to the target reflectivity, improve the design accuracy of the photonic crystal structure, obtain more optimal structural design parameters, and facilitate the directional optimization of parameters such as the band gap position, width and reflection band range of the photonic crystal, thereby achieving enhanced radiation heat dissipation in the red band under high temperature or adjusting the reflection efficiency in a specific flight speed domain and airspace, and having dynamic regulation capabilities.

[0091] Example 2:

[0092] like Figure 4 As shown, this embodiment provides a high-speed aircraft thermal protection structure, which is designed using the thermal protection structure design method based on directional control of thermal radiation described in Example 1, including a substrate, the top of the substrate is connected to a thermal insulation layer 1, the top of the thermal insulation layer 1 is connected to a photonic crystal layer, the top of the photonic crystal layer is connected to a thermal insulation layer 2, the top of the thermal insulation layer 2 is connected to a composite material panel, and the top surface of the composite material panel is coated with a high-emissivity coating.

[0093] Example 3:

[0094] This embodiment provides a thermal protection structure design system based on thermal radiation directional control, which is used to implement the above-mentioned thermal protection structure design method based on thermal radiation directional control, including:

[0095] A data set building module is used to receive the photonic crystal structure design parameters and perform preprocessing to complete the establishment of the data set;

[0096] A model building module is used to build a bidirectional neural network model, and train the bidirectional neural network model using the established data set until the loss function is minimized, thereby obtaining a trained bidirectional neural network model, wherein the bidirectional neural network model takes the target reflectivity of the photonic crystal as input and the predicted reflectivity of the photonic crystal as output;

[0097] The evaluation output module is used to evaluate the output predicted reflectivity using the mean square error. If the error requirements are met, the photonic crystal structure is designed based on the output predicted reflectivity, thereby obtaining a thermal protection structure design scheme. Otherwise, the predicted reflectivity is re-iterated until it meets the requirements.

[0098] Specifically, the above-mentioned data set construction module, model construction module and evaluation output module can be embedded in a computer processing system. The computer calls the above-mentioned modules to complete the task of designing a scheme for the thermal protection structure based on the above-mentioned thermal protection structure design method based on directional control of thermal radiation; the above-mentioned data set construction module, model construction module and evaluation output module can perform operations according to the specific steps given in the thermal protection structure design method based on directional control of thermal radiation.

[0099] It should be noted that it should be understood that the division of the various modules of the above system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. Moreover, these modules can all be implemented in the form of software called by processing elements; they can also all be implemented in the form of hardware; some modules can also be implemented in the form of processing elements calling software, and some modules can be implemented in the form of hardware. For example, the data set construction module can be a separately established processing element, or it can be integrated into a chip of the above-mentioned device. In addition, it can also be stored in the memory of the above-mentioned device in the form of program code, and called and executed by a processing element of the above-mentioned device to perform the functions of the above-mentioned signal processing module. The implementation of other modules is similar. In addition, these modules can all or partly be integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the hardware integrated logic circuit in the processor element or software instructions.

[0100] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module is implemented by scheduling program code through a processing element, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0101] Example 4:

[0102] According to a fourth aspect of the present invention, the present invention provides a terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that when the processor loads and executes the computer program, the above-mentioned thermal protection structure design method based on directional control of thermal radiation is adopted.

[0103] It should be noted that the terminal device can be a computer device such as a desktop computer, a laptop computer or a cloud server, and the terminal device includes but is not limited to a processor and a memory. For example, the terminal device can also include input and output devices, network access devices and buses, etc.

[0104] Furthermore, the processor may adopt a central processing unit (CPU). Of course, depending on the actual usage, other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0105] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0106] For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. When an element is referred to as being "assembled on", "installed on", "fixed on" or "set on" another element, it can be directly on the other element or there can be a central element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there can be a central element at the same time. The terms "vertical", "horizontal", "up", "down", "left", "right" and similar expressions used herein are for illustrative purposes only and are not intended to be the only embodiment.

[0107] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

[0108] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Claims

1. A thermal protection structure design method based on directional control of thermal radiation, applied to photonic crystal structure design, characterized by: The following steps are involved: Receive photonic crystal structure design parameters and perform preprocessing to complete the establishment of the data set; Constructing a bidirectional neural network model, and training the bidirectional neural network model using the established data set until the loss function is minimized, thereby obtaining a trained bidirectional neural network model, wherein the bidirectional neural network model takes the target reflectivity of the photonic crystal as input and the predicted reflectivity of the photonic crystal as output; The output predicted reflectivity is evaluated using the mean square error. If it meets the error requirements, the photonic crystal structure is designed based on the output predicted reflectivity, thereby obtaining a thermal protection structure design scheme. Otherwise, the output predicted reflectivity is iterated again until it meets the requirements.

2. The thermal protection structure design method based on thermal radiation directional control according to claim 1 is characterized in that: The photonic crystal structure design parameters include material, the designed number of dielectric layers, and the designed thickness of the dielectric layers.

3. The thermal protection structure design method based on thermal radiation directional control according to claim 2 is characterized in that: Receive the photonic crystal structure design parameters and perform preprocessing to complete the establishment of the data set, as follows: (31) Determine the value range of the design parameters of the photonic crystal structure; (32) Latin hypercube sampling is used to select sample points within the value range: The value range of each design parameter is divided into several equally probable subintervals, random sampling is performed in each subinterval to obtain the sampling points of each design parameter, and all the sampling points of the design parameters are randomly arranged and combined into a sample space; (33) The reflectivity of the photonic crystal structure corresponding to the sample space is obtained using the transfer matrix method: Establish the characteristic matrix of each layer of the photonic crystal medium: Where A, B, C, and D all represent characteristic matrices, d represents the thickness of the dielectric layer, the subscript l represents the dielectric layer, ch and sh represent the hyperbolic sine function and the hyperbolic cosine function, j is the imaginary unit, and β is expressed as: Where k represents the wave number, ε r represents the real part of the dielectric constant, θ represents the incident angle, tanδ represents the loss tangent, and ε i / ε r We get, where ε i represents the imaginary part of the dielectric constant; Z is the intermediate representation for solving the A, B, C, and D matrices. The specific expressions are as follows: The total characteristic matrix of the photonic crystal structure is expressed as the product of the characteristic matrices of each dielectric layer: The characteristic matrix of the overall structure is obtained by the characteristic matrix of each dielectric layer using the transfer matrix method. The reflectivity R of the photonic crystal is expressed as:

4. The thermal protection structure design method based on thermal radiation directional control according to claim 3 is characterized in that: Construct a bidirectional neural network model as follows: (41) establishing a reverse design neural network model, wherein the reverse design neural network takes the target reflectivity of the photonic crystal as input and the photonic crystal structure design parameters as output; (42) establishing a forward prediction neural network model, wherein the forward prediction neural network takes the photonic crystal structure design parameters as input and the photonic crystal predicted reflectivity as output; (43) By connecting the reverse design neural network and the forward prediction neural network in series, a bidirectional neural network model is established.

5. The thermal protection structure design method based on thermal radiation directional control according to claim 4 is characterized in that: The reverse design neural network model specifically includes an input layer, a hidden layer and an output layer. The input layer represents the target reflectivity and has a total of 199 units; the hidden layer has two layers, including 500 and 200 units; the output layer represents the structural design parameters and has a total of 5 units; The forward prediction neural network specifically includes an input layer, a hidden layer and an output layer. The input layer represents the structural design parameters, with a total of 5 units; the hidden layer has 2 layers, including 200 and 500 units; the output layer represents the target reflectivity, with a total of 199 units.

6. The thermal protection structure design method based on thermal radiation directional control according to claim 5 is characterized in that: The loss function is as follows: Loss=MSE(r,r') (6) Where Loss represents the loss function, MSE represents the mean square error between the input reflectivity and output reflectivity of the neural network model, r represents the target reflectivity, and r' represents the output predicted reflectivity.

7. The thermal protection structure design method based on thermal radiation directional control according to claim 6 is characterized in that: The mean square error is used to evaluate the output predicted reflectivity as follows: Where MSE represents the mean square error between the target reflectivity and the output predicted reflectivity, r represents the target reflectivity, r' represents the output predicted reflectivity, and N represents the number of points obtained by discretizing the reflectivity.

8. A high-speed aircraft thermal protection structure designed using the thermal protection structure design method based on thermal radiation directional control according to any one of claims 1 to 7, characterized in that: It includes a substrate, the top of which is connected to a first insulation layer, the top of which is connected to a photonic crystal layer, the top of which is connected to a second insulation layer, the top of which is connected to a composite material panel, and the top surface of the composite material panel is coated with a high-emissivity coating.

9. A thermal protection structure design system based on thermal radiation directional control, for implementing the thermal protection structure design method based on thermal radiation directional control according to any one of claims 1 to 7, characterized in that: include: A data set building module is used to receive the photonic crystal structure design parameters and perform preprocessing to complete the establishment of the data set; A model building module is used to build a bidirectional neural network model, and train the bidirectional neural network model using the established data set until the loss function is minimized, thereby obtaining a trained bidirectional neural network model, wherein the bidirectional neural network model takes the target reflectivity of the photonic crystal as input and the predicted reflectivity of the photonic crystal as output; The evaluation output module is used to evaluate the output predicted reflectivity using the mean square error. If the error requirements are met, the photonic crystal structure is designed based on the output predicted reflectivity, thereby obtaining a thermal protection structure design scheme. Otherwise, the predicted reflectivity is re-iterated until it meets the requirements.

10. A terminal device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor loads and executes the computer program, the thermal protection structure design method based on thermal radiation directional control according to any one of claims 1 to 7 is adopted.

Images