Training data acquisition method and device applied to electromagnetic device design, medium and equipment

Abstract

The invention discloses a training data acquisition method and device applied to electromagnetic device design, a medium and equipment, and relates to the technical field of electromagnetic device design. According to the method, the inverse mapping relation between the electromagnetic response curve and the topological distribution of the frequency selective surface structure is utilized, topological modeling and full-wave simulation are firstly carried out, a topological structure is expressed by utilizing a binary matrix, in iteration of a target function, the topological binary matrix and the electromagnetic response reserved in each iteration are used as a training data set, and the frequency selective surface structure is obtained. According to the method, more useful design information can be reserved, the electromagnetic response and the design target have more similar characteristics, misleading information is less, and the quality of training data in electromagnetic device design is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electromagnetic device design, in particular to a training data obtaining method and device applied to electromagnetic device design, a medium and equipment. BACKGROUND

[0002] Deep learning neural network (DNN) is gradually applied to solve various problems in the electromagnetic field, such as microwave filter design, antenna design, beam pattern synthesis, and meta-surface design, due to its powerful data fitting capability. In order to make electromagnetic device design have greater freedom, topological modeling is attracting more and more attention. Topological modeling largely contains more modeling variables for constructing the topological structure of the modeling domain, thereby providing more freedom for design. However, due to too many combinations of binary variables and uncertain initial structure, the existing method of collecting training samples based on random collection and design experiments is not applicable. How to construct a high-quality training dataset within limited full-wave simulation time, that is, to obtain more qualified samples to provide useful information for design, is a very meaningful research problem. SUMMARY

[0003] The main purpose of the present application is to provide a training data obtaining method and device applied to electromagnetic device design, a medium and equipment, which aims to solve the problem of poor dirty rendering effect caused by light conditions in the prior art.

[0004] To achieve the above-mentioned purpose, the technical solutions adopted by the embodiments of the present application are as follows: In a first aspect, the embodiments of the present application provide a training data obtaining method applied to electromagnetic device design, comprising the following steps: initializing structure parameters; performing topological modeling and full-wave simulation of the frequency selective surface structure based on the structure parameters to obtain a topological binary matrix and an electromagnetic response curve of the frequency selective surface structure; calculating a target function according to the electromagnetic response curve; wherein the target function is used to represent the optimization direction; judging whether the target function reaches a maximum iteration number; in the case of no, updating the structure parameters according to an updating mechanism, obtaining the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve, and returning to the step of performing topological modeling and full-wave simulation of the frequency selective surface structure based on the structure parameters to obtain the topological binary matrix and the electromagnetic response curve of the frequency selective surface structure, until the result of the judgment is yes; constructing a training dataset according to the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve.

[0005] In an implementation form of the first aspect, before the calculating the objective function according to the electromagnetic response curve, the method further comprises: sine-wave processing the value of the electromagnetic response curve at each sampling frequency point to obtain a target electromagnetic response curve; calculating the objective function according to the electromagnetic response curve, comprising: calculating the objective function according to the target electromagnetic response curve.

[0006] In an implementation form of the first aspect, the sine-wave processing the value of the electromagnetic response curve at each sampling frequency point to obtain a target electromagnetic response curve, comprises: threshold-based binary processing the value of the electromagnetic response curve at each sampling frequency point, setting the value less than a threshold to a first numerical value, and otherwise setting to a second numerical value, to obtain the target electromagnetic response curve.

[0007] In an implementation form of the first aspect, the topology modeling and full-wave simulation of the frequency selective surface structure based on the structure parameters to obtain the topology binary matrix and the electromagnetic response curve of the frequency selective surface structure, comprises: topology modeling of the frequency selective surface structure based on the structure parameters to obtain a topology structure of the frequency selective surface structure; binary representing the topology structure based on a binary representing rule to obtain the topology binary matrix of the frequency selective surface structure.

[0008] In an implementation form of the first aspect, before the binary representing the topology structure based on the binary representing rule to obtain the topology binary matrix of the frequency selective surface structure, the method further comprises: obtaining the binary representing rule according to the metal distribution of the frequency selective surface structure sample.

[0009] In an implementation form of the first aspect, the obtaining the binary representing rule according to the metal distribution of the frequency selective surface structure sample, comprises: confirming a metal layer and discretizing the metal layer into a plurality of pixel-level metal pieces according to the layer structure of the frequency selective surface structure sample; obtaining the binary representing rule according to the metal distribution of the metal piece representing area in the topology modeling.

[0010] In an implementation form of the first aspect, the calculating the objective function according to the electromagnetic response curve, comprises: determining an optimization direction according to the electromagnetic response curve; obtaining a sub-function and a weight thereof according to the optimization direction and the importance; calculating the objective function according to the sub-function and the weight thereof.

[0011] In a second aspect, an embodiment of the present application provides a training data obtaining device applied to electromagnetic device design, comprising: an initialization module configured to initialize a structure parameter; a simulation module configured to perform topology modeling and full-wave simulation on the frequency selective surface structure based on the structure parameter, to obtain a topology binary matrix and an electromagnetic response curve of the frequency selective surface structure; a calculation module configured to calculate a target function according to the electromagnetic response curve, wherein the target function is used to represent an optimization direction; a judgment module configured to judge whether the target function reaches a maximum iteration number; an iteration module configured to, in a case where the judgment result is no, update the structure parameter according to an update mechanism, to obtain the topology binary matrix and the corresponding electromagnetic response curve of the frequency selective surface structure, and to return to the topology modeling and the full-wave simulation on the frequency selective surface structure based on the structure parameter, to obtain the topology binary matrix and the electromagnetic response curve of the frequency selective surface structure, until the judgment result is yes; a construction module configured to construct a training data set according to the topology binary matrix and the corresponding electromagnetic response curve of the frequency selective surface structure.

[0012] In a third aspect, an embodiment of the present application provides a computer readable storage medium storing a computer program, and the computer program is loaded and executed by a processor to implement the training data obtaining method applied to electromagnetic device design provided in any one of the first aspect.

[0013] In a fourth aspect, an embodiment of the present application provides an electronic device comprising a processor and a memory, wherein, the memory is configured to store a computer program; the processor is configured to load and execute the computer program, so that the electronic device performs the training data obtaining method applied to electromagnetic device design provided in any one of the first aspect.

[0014] Compared with the prior art, the present application has the following beneficial effects: The application embodiment provides a training data obtaining method, device, medium and equipment applied to electromagnetic device design. The method comprises the following steps: initializing a structure parameter; performing topological modeling and full-wave simulation on a frequency selective surface structure based on the structure parameter to obtain a topological binary matrix of the frequency selective surface structure and an electromagnetic response curve; calculating a target function according to the electromagnetic response curve; wherein the target function is used to represent an optimization direction; judging whether the target function reaches a maximum iteration number; in the case that the judgment result is no, updating the structure parameter according to an updating mechanism, obtaining the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve, and returning to the step of performing topological modeling and full-wave simulation on the frequency selective surface structure based on the structure parameter to obtain the topological binary matrix of the frequency selective surface structure and the electromagnetic response curve until the judgment result is yes; and constructing a training data set according to the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve. The application utilizes the inverse mapping relationship between the electromagnetic response curve and the topological distribution of the frequency selective surface structure, firstly performs topological modeling and full-wave simulation, and utilizes the binary matrix to express the topological structure. In the iteration of the target function, the topological binary matrix and the electromagnetic response retained in each iteration are used as the training data set, more useful design information can be retained, the electromagnetic response has more similar characteristics with the design target, misleading information is less, and the quality of the training data in the electromagnetic device design is improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 An electronic device structure schematic diagram of a hardware running environment related to the application embodiment; Figure 2 A flowchart of the training data obtaining method applied to electromagnetic device design provided by the application embodiment; Figure 3 A layer stack structure schematic diagram of the frequency selective surface structure in the training data obtaining method applied to electromagnetic device design provided by the application embodiment; Figure 4 Schematic diagrams of three frequency selective surface structures in the training data obtaining method applied to electromagnetic device design provided by the application embodiment, which are structure 1, structure 2 and structure 3 respectively; Figure 5 Schematic diagrams of three frequency selective surface structures in the training data obtaining method applied to electromagnetic device design provided by the application embodiment, which are structure 4, structure 5 and structure 6 respectively; Figure 6 A discrete mode schematic diagram of a metal layer in the training data obtaining method applied to electromagnetic device design provided by the application embodiment; Figure 7 A schematic diagram for testing the sample 1 to verify the effectiveness of the method provided by the application embodiment; Figure 8 A schematic diagram for testing sample 2 to verify the effectiveness of the method provided by the embodiments of the present application is shown in the figure. Figure 9 A schematic diagram for testing sample 3 to verify the effectiveness of the method provided by the embodiments of the present application is shown in the figure. Figure 10 A schematic diagram for testing sample 4 to verify the effectiveness of the method provided by the embodiments of the present application is shown in the figure. Figure 11 A schematic diagram of the training data obtaining device for electromagnetic device design provided by the embodiments of the present application is shown in the figure. Marked in the figure: 101-processor, 102-communication bus, 103-network interface, 104-user interface, 105-memory. DETAILED DESCRIPTION

[0016] It should be understood that the specific embodiments described herein are merely intended to explain the present application and are not intended to limit the present application.

[0017] Referring to the drawings Figure 1 , the drawings Figure 1 A schematic diagram of the electronic device structure of the hardware running environment involved in the embodiments of the present application is shown in the figure, which can include a processor 101, for example, a central processing unit (CPU), a communication bus 102, a user interface 104, a network interface 103, and a memory 105. The communication bus 102 is used to realize the connection and communication between these components. The user interface 104 can include a display screen (Display) and an input unit such as a keyboard (Keyboard). The optional user interface 104 can further include a standard wired interface and a wireless interface. The network interface 103 can optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface). The memory 105 can be a storage device independent of the aforementioned processor 101. The memory 105 can be a high-speed random access memory (RAM) memory or a stable non-volatile memory (NVM), for example, at least one disk memory. The processor 101 can be a general-purpose processor, including a central processing unit, a network processor, etc. It can also be a digital signal processor, an application-specific integrated circuit, a field programmable gate array, or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0018] Those skilled in the art can understand that the drawings Figure 1The structure shown in the figure does not constitute a limitation on the electronic device, and can include more or fewer components than shown, or combine certain components, or different component arrangements.

[0019] As shown in the accompanying drawings Figure 1 The memory 105 as a storage medium can include an operating system, a network communication module, a user interface module, and a training data obtaining apparatus for electromagnetic device design.

[0020] In the electronic device shown in the accompanying drawings Figure 1 In the electronic device shown in the accompanying drawings, the network interface 103 is mainly used for data communication with the network server; the user interface 104 is mainly used for data interaction with the user; the processor 101 and the memory 105 in the present application can be arranged in the electronic device, and the electronic device calls the training data obtaining apparatus for electromagnetic device design stored in the memory 105 through the processor 101, and executes the training data obtaining method for electromagnetic device design provided in the present application.

[0021] Deep learning neural network (DNN) has been gradually applied to solve various problems in the electromagnetic field, such as microwave filter design, antenna design, beam pattern synthesis, and meta-surface design, due to its powerful data fitting capability. In order to make electromagnetic device design have greater freedom, topological modeling is attracting more and more attention. However, unlike the main type of design method based on machine learning, parameterized modeling, topological modeling contains more modeling variables to a large extent for constructing the topological structure of the modeling domain, thereby providing more freedom for design.

[0022] In the training process, the training data set plays an important role, and its quantity and quality directly determine the training convergence and training performance of the DNN model. For parameterized modeling, random collection and design of experiment (DoE) are usually used to collect training samples. However, for topological modeling, due to too many combinations of binary variables and uncertain initial structure, the above method is not applicable. Therefore, how to effectively construct the training data set is still a challenging task.

[0023] In terms of the number of training samples, a large number (sometimes thousands of samples) can bring better mapping performance and avoid overfitting to a certain extent. However, collecting a large number of sample training data sets from full-wave electromagnetic simulation will consume a large amount of computing cost. Therefore, the method of increasing the number of training samples to improve the performance of the machine learning model is high-cost and inefficient.

[0024] In terms of the quality of the training data set, high-quality samples can also improve the mapping performance of the DNN model. High quality means that the electromagnetic response of the sample has some important features similar to the design target, and almost no misleading information. A high-quality training data set is helpful for design. For two training data sets of the same number, using a higher quality data set to train the DNN is more likely to obtain better performance without increasing additional training samples.

[0025] Therefore, how to construct a high-quality training data set within a limited full-wave simulation time, that is, to obtain more qualified samples, is a very meaningful research problem. However, in the machine learning-based electromagnetic device design, there is currently no research on this problem; for this purpose, the embodiment of the present application provides a scheme, which uses the inverse mapping relationship between the electromagnetic response curve and the topological distribution of the frequency selective surface structure, first performs topological modeling and full-wave simulation, and uses a binary matrix to express the topological structure. In the iteration of the objective function, the topological binary matrix and the electromagnetic response retained in each iteration are used as the training data set, which can retain more useful information for design, and the electromagnetic response has more similar features to the design target, and less misleading information, thereby improving the quality of the training data in electromagnetic device design.

[0026] Referring to the accompanying Figure 2 , based on the hardware device of the foregoing embodiment, the embodiment of the present application provides a training data obtaining method applied to electromagnetic device design, comprising the following steps: S10: initializing the structure parameters.

[0027] S20: topological modeling and full-wave simulation of the frequency selective surface structure based on the structure parameters, to obtain the topological binary matrix of the frequency selective surface structure and the electromagnetic response curve.

[0028] In the specific implementation process, since it is an iterative process, the geometric structure parameters need to be initialized first, and the frequency selective surface (FSS) structure is modeled and full-wave simulated in the MATLAB-CST joint simulation program based on the structure parameters. The topological modeling and electromagnetic response curve |S 11 | of the frequency selective surface structure are obtained respectively. The topological modeling is represented in the form of a binary matrix. Specifically, the topological modeling and full-wave simulation of the frequency selective surface structure based on the structure parameters are performed to obtain the topological binary matrix of the frequency selective surface structure and the electromagnetic response curve, including: topological modeling of the frequency selective surface structure based on the structure parameters, to obtain the topological structure of the frequency selective surface structure; The topological structure is binarized based on the binary representation rule to obtain the topological binary matrix of the frequency-selective surface structure.

[0029] In practical implementation, binary matrices can simplify the representation of the topological structure of frequency-selective surface structures. By establishing rules for binary representation in advance, topological modeling can directly output the topological structure in the form of a binary matrix. That is, before obtaining the topological binary matrix of the frequency-selective surface structure by performing binary representation of the topological structure based on the binary representation rules, the method also includes: Based on the frequency of the selected surface structure samples, the metal distribution is analyzed to obtain a binary characterization rule.

[0030] The binary distribution is characterized by its metal distribution, thus obtaining binary representation rules. Specifically, the binary representation rules are obtained by selecting surface structure samples based on their metal distribution according to frequency, including: The stacked structure of the surface structure sample is selected based on frequency, the metal layer is identified, and the metal layer is discretized into multiple pixel-level metal sheets; Based on the metal distribution in the metal sheet representation region in topological modeling, binary representation rules are obtained.

[0031] In the specific implementation process, the stacked structure of the frequency-selective surface structure is as shown in the attached figure. Figure 3 As shown, the basic structure consists of three metal layers and two dielectric layers. The top and bottom metal layers have the same metal patch pattern, and both substrates have the same dimensions. D is the dielectric layer thickness, and P is the width of the frequency-selective surface structure. (See attached diagram.) Figure 4 Appendix Figure 5 Taking the six different types of FSS structures shown as examples, L1-L5 respectively identify some parameter dimensions on the metal layers. The two images in each column represent different metal layers of the same FSS structure. To obtain the topological distribution of the FSS structure, each metal layer of the FSS structure is discretized into q×q square pixel metal sheets, as shown in the attached figure. Figure 6 As shown, using a discrete method with q=160, the topological distribution of the FSS structure can then be represented by a binary matrix, where "1" represents metal and "0" represents non-metal.

[0032] S30: Calculate the objective function based on the electromagnetic response curve; whereby the objective function is used to characterize the optimization direction.

[0033] In practical implementation, the objective function used for electromagnetic response limiting determines the direction of data optimization. The objective function typically consists of several sub-functions, each with a specific weight to determine its importance. That is, based on the electromagnetic response curve, the objective function is calculated, including: Determine the optimization direction based on the electromagnetic response curve; According to the optimization direction and the importance, a sub-function and a weight are obtained; According to the sub-function and the weight, the target function is calculated.

[0034] Taking the FSS design as an example, the target function in the embodiment is as follows:

[0035] The target function is composed of two sub-functions, and the weights w1 and w2 are the same, which is 0.5, because the importance of the corresponding two targets is equivalent. Among them, the sub-function f1(x) is used to optimize the bandwidth of the passband, because narrow bandwidth is not conducive to the training of the DNN model; the sub-function f2(x) is used to optimize the number of passbands, because when |S 11 | occurs drastic changes, it is also not conducive to the learning of the deep learning model. It is worth noting that the target function is not for strict electromagnetic response limits, but to make the sample contain some important features similar to the design target.

[0036] In an embodiment, before calculating the target function according to the electromagnetic response curve, the method further includes: The value of the electromagnetic response curve at each sampling frequency point is square wave processed to obtain a target electromagnetic response curve.

[0037] In the specific implementation process, in order to facilitate the judgment during iterative optimization, square wave processing can be performed before calculating the target function. The ideal square wave only has two values of high and low, which is actually a kind of binary processing method. Specifically, the value of the electromagnetic response curve at each sampling frequency point is binary processed based on a threshold value, and the value less than the threshold value is set to a first value, otherwise it is set to a second value, to obtain a target electromagnetic response curve. For example, the value of |S 11 | at each sampling frequency point is processed into two values of 0dB and-10dB, and-10dB is set as the threshold value. If the value is less than-10dB, it is set to-10dB, otherwise it is set to 0dB. Such square wave processing makes the |S 11 | curve easy to identify the number and bandwidth of the passband.

[0038] Based on the foregoing steps, the target function is calculated according to the electromagnetic response curve, including: The target function is calculated according to the target electromagnetic response curve.

[0039] S40: Determine whether the target function reaches the maximum number of iterations.

[0040] S50: In the case of the result of the judgment is no, the structure parameters are updated according to the update mechanism, the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve are obtained, and the topological modeling and full-wave simulation of the frequency selective surface structure based on the structure parameters are returned to obtain the topological binary matrix of the frequency selective surface structure and the electromagnetic response curve until the result of the judgment is yes.

[0041] S60: According to the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve, a training data set is constructed.

[0042] In the specific implementation process, after the target function calculation is completed, it is necessary to judge whether the maximum iteration number is reached, if not, the structure parameters are updated by using the evolutionary algorithm update mechanism, and the topological modeling and full-wave simulation are performed again. In each iteration, the binary matrix of the FSS structure topological distribution and the simulated electromagnetic response are retained. The iteration will continue until the maximum iteration number, at this time, the training data set can be constructed, wherein the data set is composed of the FSS structure binary matrix obtained by optimization processing and the corresponding electromagnetic response.

[0043] In the embodiment, the inverse mapping relationship between the electromagnetic response curve and the topological distribution of the frequency selective surface structure is utilized, first, the topological modeling and full-wave simulation are performed, and the binary matrix is used to express the topological structure, in the iteration of the target function, the topological binary matrix and the electromagnetic response retained in each iteration are used as the training data set, more useful information of the design can be retained, the electromagnetic response has more similar characteristics with the design target, the misleading information is less, and the quality of the training data in the electromagnetic device design is improved.

[0044] After obtaining the training data set, a deep learning model is constructed, in the embodiment, taking the deep learning model MLP-mixer as an example, |S 11 is mapped to the structure topological distribution of the FSS. The MLP-Mixer mainly consists of three parts, which are full connection module, mixed architecture module and structure output module.

[0045] The full connection module sends the segmented input electromagnetic response parameters into the network, and encodes each segment of data through a full connection layer of different parameters to obtain multiple encoding results. The multiple encoding results are combined and input into the mixed architecture module. The mixed architecture module is composed of multiple mixed layers. Each mixed layer uses two types of MLP blocks: spatial mixing MLP and channel mixing MLP. The spatial mixing MLP realizes the mixing between data blocks at different spatial positions, and the channel MLP realizes the mixing between data blocks at different channels, and each data block is independently operated. The spatial mixing MLP and the channel mixing MLP are connected in the mixed layer in an interleaved manner. The structure prediction module is composed of a global average pooling layer, a full connection layer and a binary layer. The global average pooling layer reduces the dimensionality of the electromagnetic response parameter feature information extracted by the mixed architecture module, and then obtains the initial frequency selective surface structure through the full connection layer.

[0046] Specifically, when the MLP-mixer is applied to the FSS design, the input |S 11 is first segmented into 250 non-overlapping small blocks. Then, by setting the hidden layer dimension to 1024, a two-dimensional table is derived. The spatial mixing MLP and the channel mixing MLP will respectively complete the mapping of Finally, the structure parameters of the FSS will be output by the structure prediction layer. In order to model, it is necessary to quantize the discrete values of the topology distribution binary matrix of the FSS to 0 or 1. Therefore, the output topology distribution parameters are utilized in the following formula:

[0047] where x is the output value of the classifier module, and the binary matrix of the FSS can be obtained by inputting the target |S 11 into the trained MLP-mixer.

[0048] In order to fully prove the effectiveness of the training data collection method based on the evolutionary algorithm proposed in the present application, two methods based on CS and PSO-GA and a random training data collection method are used to collect the same number of samples, and then the MLP-mixer is trained and its performance is evaluated.

[0049] In the present embodiment, the number of training data is set to be proportional to the number of geometric variables, because more data is needed in a larger sampling space. Therefore, the amount of data for each FSS structure is set to 100xp, where p is the number of variables for each FSS structure. The training data collected by the CS method and the PSO-GA method are shown in Table 1 and Table 2, respectively. Figure 4 -Table 1 Figure 5It can be seen that each FSS structure has [3, 2, 4, 5, 5] variables, and 2400 samples are collected for training by using three data collection methods. 240 samples (1 / 10 of the training data) are randomly selected for testing, and all calculations are performed on the same computer.

[0050] In training, the learning rate is set to 5x10 -4 and the mean absolute percentage error (MAPE) is used as the loss function, which can be expressed as follows: MAPE =

[0051] wherein, represents the predicted FSS structure binary matrix of the nth sample, represents the actual FSS structure binary matrix of the nth sample in the training data set, and N is the total number of samples in the training data set.

[0052] For testing, the output is first input to CST through MATLAB-CST joint simulation, 11 and the derived electromagnetic response is |S APE = m∈

[0053] wherein, represents the set of passband frequency points obtained by simulation of the mth predicted FSS structure, represents the set of passband frequency points obtained by simulation of the mth actual FSS structure, and M is the total number of samples in the test data set.

[0054] In order to compare the performance, the MLP-Mixer is trained using the above three different data sets, and the same 240 test samples are used for testing. The trained MLP-Mixer is evaluated for performance using APE, and the performance comparison table as shown in Table 1 is obtained: Table 1 - Performance Comparison Table

[0055] For the MLP-Mixer trained using the CS-based training data collection method, 54.17% of the samples in the test dataset had an error of ≤5%, and 62.50% had an error of ≤10%. For the MLP-Mixer trained using the PSO-GA-based training data collection method, 50.00% of the test data samples had an error of ≤5%, and 62.92% had an error of ≤10%. However, for the MLP-Mixer trained using the random training data collection method, only 35.00% of the samples in the test dataset had an error of ≤5%, and only 47.08% had an error of ≤10%. With an error of ≤5%, the MLP-Mixer test accuracy using the CS-based and PSO-GA-based training data collection methods was 25% and 19.17% higher than that using the random training data collection method, respectively. Meanwhile, with an error ≤10%, the test accuracy of the CS-based training data collection method and the PSO-GA-based training data collection method improved by 15.42% and 15.84%, respectively. If the goal is to achieve the same test error level as the proposed training data collection method using a randomized training data collection method, the number of full-wave simulations required for the former will be approximately twice that of the latter.

[0056] To further demonstrate the test performance, see attached Figure 7 -Appendix Figure 10 The effectiveness of the method provided in the embodiments of this application is verified, and attached... Figure 7 -Appendix Figure 8 A CS-based training data collection method is presented, along with... Figure 9 -Appendix Figure 10 Some test examples of the training data collection method based on PSO-GA are presented. The left side of the examples compares the actual and predicted structures, while the right side compares the electromagnetic responses of the actual and predicted structures. The APE values ​​of the four examples are 1.64%, 1.81%, 0.49%, and 1.92%, respectively. Due to the non-uniqueness of the inverse problem, some discrepancies between the predicted and actual structures may be observed in the test dataset. These results demonstrate that the training data acquisition method proposed in this application is effective and can improve the performance of DNN models.

[0057] See attached document Figure 11 Based on the same inventive concept as in the foregoing embodiments, this application also provides a training data acquisition device for electromagnetic device design, comprising: The initialization module is used to initialize structure parameters; The simulation module is configured to perform topology modeling and full-wave simulation on the frequency selective surface structure based on the structure parameters, to obtain a topology binary matrix and an electromagnetic response curve of the frequency selective surface structure. The calculation module is configured to calculate a target function according to the electromagnetic response curve; the target function is used to represent an optimization direction. The judgment module is configured to judge whether the target function reaches a maximum iteration number. The iteration module is configured to, in a case where the judgment result is no, update the structure parameters according to an update mechanism, to obtain a topology binary matrix and a corresponding electromagnetic response curve of the frequency selective surface structure, and to return to the topology modeling and full-wave simulation on the frequency selective surface structure based on the structure parameters, to obtain the topology binary matrix and the electromagnetic response curve of the frequency selective surface structure, until the judgment result is yes. The construction module is configured to construct a training data set according to the topology binary matrix and the corresponding electromagnetic response curve of the frequency selective surface structure.

[0058] Those skilled in the art should understand that the division of each module in the embodiments is only a logical division of functions, and all or part of the modules can be integrated onto one or more actual carriers in actual applications, and the modules can all be implemented in the form of software calling a processing unit, or all be implemented in the form of hardware, or be implemented in the form of software and hardware combination. It should be noted that the modules in the training data obtaining apparatus for electromagnetic device design in the embodiments are one-to-one corresponding to the steps in the training data obtaining method for electromagnetic device design in the foregoing embodiments, and therefore, the specific embodiments of the present embodiments can refer to the embodiments of the foregoing training data obtaining method for electromagnetic device design, which will not be described here.

[0059] Based on the same inventive concept as in the foregoing embodiments, the embodiments of the present application also provide a computer-readable storage medium storing a computer program, which is loaded and executed by a processor to implement the training data obtaining method for electromagnetic device design provided by the embodiments of the present application.

[0060] Based on the same inventive concept as in the foregoing embodiments, the embodiments of the present application also provide an electronic device including a processor and a memory, wherein, The memory is configured to store a computer program; The processor is configured to load and execute the computer program to enable the electronic device to perform the training data obtaining method for electromagnetic device design provided by the embodiments of the present application.

[0061] In some embodiments, the computer readable storage medium can be a memory such as a FRAM, ROM, PROM, EPROM, EEPROM, flash memory, a magnetic surface memory, an optical disk, or a CD-ROM, etc.; or can be various devices including one or any combination of the above memories. The computer can be various computing devices including a smart terminal and a server.

[0062] In some embodiments, the executable instructions can take the form of a program, software, software modules, scripts, or code, written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages, and can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0063] By way of example, the executable instructions can, but need not, reside in a file system's files, can be stored in a part of a file that holds other programs or data, for example, in one or more scripts stored in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files, for example, files that store one or more modules, sub programs, or portions of code.

[0064] By way of example, the executable instructions can be deployed to be executed on one computing device, or on multiple computing devices that are located at one site, or that are distributed across multiple sites and that are interconnected by a communication network.

[0065] It should be noted that, in this document, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or system. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or system that comprises the element.

[0066] The above-mentioned sequence numbers of the embodiments of the present application are only for description, and do not represent advantages or disadvantages of the embodiments.

[0067] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned example method can be realized by means of software and a necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, an optical disk), and includes a plurality of instructions for causing a multimedia terminal device (which can be a mobile phone, a computer, a television receiver, or a network device, etc.) to execute the method described in each embodiment of the present application.

[0068] In summary, the method, device, medium and equipment for obtaining training data applied to electromagnetic device design provided by the embodiments of the present application, the method comprises: initializing a structure parameter; performing topological modeling and full-wave simulation of a frequency selective surface structure based on the structure parameter to obtain a topological binary matrix of the frequency selective surface structure and an electromagnetic response curve; calculating a target function according to the electromagnetic response curve; wherein the target function is used to represent an optimization direction; determining whether the target function reaches a maximum iteration number; in the case of no, updating the structure parameter according to an updating mechanism, obtaining the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve, and returning to the step of performing topological modeling and full-wave simulation of the frequency selective surface structure based on the structure parameter to obtain the topological binary matrix of the frequency selective surface structure and the electromagnetic response curve until the result of the determination is yes; and constructing a training data set according to the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve. The present application utilizes the inverse mapping relationship between the electromagnetic response curve and the topological distribution of the frequency selective surface structure, firstly performs topological modeling and full-wave simulation, and utilizes the binary matrix to express the topological structure, and in the iteration of the target function, the topological binary matrix and the electromagnetic response retained in each iteration are used as the training data set, which can retain more useful information of the design, and the electromagnetic response has more similar characteristics with the design target, and has less misleading information, thereby improving the quality of the training data in the electromagnetic device design.

[0069] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for obtaining training data applied to the design of electromagnetic devices, characterized in that, Includes the following steps: Initialize structure parameters; Based on the structural parameters, topological modeling and full-wave simulation of the frequency selective surface structure are performed to obtain the topological binary matrix and electromagnetic response curve of the frequency selective surface structure. Based on the electromagnetic response curve, the objective function is calculated; wherein, the objective function is used to characterize the optimization direction; Determine whether the objective function has reached the maximum number of iterations; If the result of the judgment is negative, the structural parameters are updated according to the update mechanism to obtain the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve, and the steps of performing topological modeling and full-wave simulation of the frequency selective surface structure based on the structural parameters to obtain the topological binary matrix and electromagnetic response curve of the frequency selective surface structure are returned until the result of the judgment is positive. A training dataset is constructed by selecting the topological binary matrix of the surface structure based on the frequency and the corresponding electromagnetic response curve.

2. The method for obtaining training data for electromagnetic device design according to claim 1, characterized in that, Before calculating the objective function based on the electromagnetic response curve, the method further includes: The electromagnetic response curve is processed into a square wave at each sampling frequency point to obtain the target electromagnetic response curve. The step of calculating the objective function based on the electromagnetic response curve includes: Calculate the objective function based on the target electromagnetic response curve.

3. The method for obtaining training data for electromagnetic device design according to claim 2, characterized in that, The step of processing the electromagnetic response curve at each sampling frequency point using square wave processing to obtain the target electromagnetic response curve includes: The electromagnetic response curve is binarized at each sampling frequency point based on a threshold. Values ​​less than the threshold are set as the first value, and otherwise as the second value, to obtain the target electromagnetic response curve.

4. The method for obtaining training data for electromagnetic device design according to claim 1, characterized in that, The process of performing topological modeling and full-wave simulation of the frequency-selective surface structure based on the structural parameters to obtain the topological binary matrix and electromagnetic response curve of the frequency-selective surface structure includes: Based on the structural parameters, a topological model of the frequency-selective surface structure is performed to obtain the topological structure of the frequency-selective surface structure. The topological structure is binarized based on the binary representation rules to obtain the topological binary matrix of the frequency-selective surface structure.

5. The method for obtaining training data for electromagnetic device design according to claim 4, characterized in that, Before performing binary representation of the topology based on binary representation rules to obtain the topological binary matrix of the frequency-selective surface structure, the method further includes: The binary characterization rule is obtained by selecting the metal distribution of the surface structure sample based on the frequency.

6. The method for obtaining training data for electromagnetic device design according to claim 5, characterized in that, The process of selecting the metal distribution of surface structure samples based on frequency to obtain the binary characterization rule includes: The stacked structure of the surface structure sample is selected based on frequency, the metal layer is identified, and the metal layer is discretized into multiple pixel-level metal sheets; The binary representation rule is obtained based on the metal distribution of the metal sheet representation region described in the topology modeling.

7. The method for obtaining training data for electromagnetic device design according to claim 1, characterized in that, The step of calculating the objective function based on the electromagnetic response curve includes: Based on the electromagnetic response curve, determine the optimization direction; Based on the optimization direction and importance, the sub-functions and their weights are obtained; The objective function is calculated based on the sub-functions and their weights.

8. A training data acquisition device for electromagnetic device design, characterized in that, include: The initialization module is used to initialize structure parameters; The simulation module is used to perform topological modeling and full-wave simulation of the frequency selective surface structure based on the structural parameters, and to obtain the topological binary matrix and electromagnetic response curve of the frequency selective surface structure. The calculation module is used to calculate the objective function based on the electromagnetic response curve; wherein the objective function is used to characterize the optimization direction. The judgment module is used to determine whether the objective function has reached the maximum number of iterations; The iterative module is used to update the structural parameters according to the update mechanism when the judgment result is negative, obtain the topological binary matrix of the frequency selective surface structure and the corresponding electromagnetic response curve, and return the topological modeling and full-wave simulation of the frequency selective surface structure based on the structural parameters to obtain the topological binary matrix and electromagnetic response curve of the frequency selective surface structure, until the judgment result is positive. A construction module is used to select the topological binary matrix of the surface structure and the corresponding electromagnetic response curve according to the frequency, and construct a training dataset.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it implements the training data acquisition method for electromagnetic device design as described in any one of claims 1-7.

10. An electronic device, characterized in that, Including processor and memory, among which, The memory is used to store computer programs; The processor is used to load and execute the computer program to cause the electronic device to perform the training data acquisition method for electromagnetic device design as described in any one of claims 1-7.

Images