Metamaterial structure microwave resonator and resonant frequency prediction method, system and equipment thereof, medium and product

Abstract

The invention discloses a metamaterial structure microwave resonator and a resonant frequency prediction method, system and device thereof, a medium and a product, and relates to the field of metamaterial structure design, and the resonator comprises a coding pattern layer with a two-dimensional coding pattern. The method comprises the following steps: adjusting each region in a region array in the metamaterial structure microwave resonator, randomly generating various two-dimensional coding patterns, and constructing different metamaterial structure microwave resonators; one two-dimensional coding pattern corresponds to one metamaterial structure microwave resonator; reading resonant frequencies of the microwave resonators with different metamaterial structures by using a network analyzer; converting the two-dimensional coding image into a one-dimensional vector, and determining feature information of the one-dimensional vector; training a neural network model by taking the feature information as input and the corresponding resonant frequency as output; and predicting the resonant frequency of the to-be-tested metamaterial structure microwave resonator according to the neural network model. The resonant frequency prediction accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the field of metamaterial structure design, and in particular to a metamaterial structure microwave resonator and a resonant frequency prediction method, system, device, medium and product thereof. Background Art

[0002] Metamaterial structures are artificially manufactured materials whose structures and properties can be carefully designed to achieve properties that do not exist in nature, such as negative refractive index, perfect lenses, invisible cloaks, etc. Due to their unique properties, metamaterial structures have shown great application potential in optics, electronics, biomedicine and other fields. In the field of electronics, metamaterials can be designed to have ultra-thin two-dimensional structures for efficient electromagnetic wave absorption, reflection and manipulation; they can be designed to be antennas with special shapes and sizes to improve the gain, directivity and other performance of antennas; they can be designed to be electronic devices with special functions, such as capacitors, inductors, transistors, etc., to improve the performance and efficiency of electronic devices.

[0003] Traditional artificial metamaterial design methods mainly rely on the designer's electromagnetic knowledge, metamaterial design experience and a large number of attempts. The design cycle is long and the requirements for designers are high, which is not conducive to the mass design of metamaterials and the exploration and research of multifunctional metamaterials. For example, the patent with publication number CN115796046A proposes a sensor in the terahertz frequency band. During the design process, simulation software is needed to simulate the geometric structure parameters of the sensor structure unit to determine the resonant frequency position of the sensor.

[0004] Compared with conventional design methods, machine learning provides new ideas for the design and research of metamaterials. The metamaterial design method based on machine learning has higher design efficiency and provides a method for large-scale and rapid metamaterial design. There are already some solutions for the metamaterial design method based on machine learning, such as the patent with publication number CN113270157A, which discloses a mechanical metamaterial design and optimization method based on machine learning, and the patent with publication number CN117153299A, which discloses a nonlinear response mechanical metamaterial structure inverse design method based on neural network.

[0005] However, the metamaterial design method based on machine learning mainly involves the field of mechanical metamaterials, and solutions in other fields are relatively lacking. Metamaterials have the characteristic of efficiently absorbing reflected electromagnetic waves. Therefore, the use of metamaterials to design microwave resonators is the main research direction at this stage. Since the metamaterial design method based on machine learning cannot meet the requirement of simultaneously predicting the characteristics of metamaterial microwave resonators with multiple resonance peaks, it is impossible to predict the accurate resonance frequency. Summary of the invention

[0006] The purpose of the present application is to provide a metamaterial structure microwave resonator and a resonant frequency prediction method, system, device, medium and product thereof to solve the problem of low resonant frequency prediction accuracy.

[0007] To achieve the above objectives, this application provides the following solutions:

[0008] In a first aspect, the present application provides a metamaterial structure microwave resonator, the metamaterial structure microwave resonator comprising: a coding pattern layer, a first dielectric layer, and a second dielectric layer printed sequentially from top to bottom;

[0009] The coding pattern layer is a metamaterial layer, and the coding pattern layer is divided into a plurality of area arrays; each area in the area array is designed as a blank area or an area for adsorbing nano-silver sheets to form a two-dimensional coding pattern.

[0010] In a second aspect, the present application provides a method for predicting the resonant frequency of a metamaterial structure microwave resonator, comprising:

[0011] Adjusting each region in the region array within the metamaterial structure microwave resonator, randomly generating a variety of two-dimensional coding patterns, and constructing different metamaterial structure microwave resonators; one two-dimensional coding pattern corresponds to one metamaterial structure microwave resonator;

[0012] Use a network analyzer to read the resonant frequencies of microwave resonators with different metamaterial structures;

[0013] Converting the two-dimensional coded image into a one-dimensional vector and determining feature information of the one-dimensional vector;

[0014] Training a neural network model using the characteristic information as input and the corresponding resonant frequency as output;

[0015] The resonant frequency of the microwave resonator of the metamaterial structure to be tested is predicted according to the neural network model.

[0016] In a third aspect, the present application provides a resonant frequency prediction system for a metamaterial structure microwave resonator, comprising:

[0017] A two-dimensional coding pattern random generation module is used to adjust each area in the area array in the metamaterial structure microwave resonator, randomly generate multiple two-dimensional coding patterns, and construct different metamaterial structure microwave resonators; one two-dimensional coding pattern corresponds to one metamaterial structure microwave resonator;

[0018] Network analyzer, used to read the resonant frequency of microwave resonators with different metamaterial structures;

[0019] A feature information determination module, used for converting the two-dimensional coded image into a one-dimensional vector and determining feature information of the one-dimensional vector;

[0020] A neural network model training module, used to train a neural network model using the characteristic information as input and the corresponding resonant frequency as output;

[0021] The resonant frequency prediction module is used to predict the resonant frequency of the microwave resonator of the metamaterial structure to be tested according to the neural network model.

[0022] In a fourth aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-described methods for predicting the resonant frequency of a metamaterial structure microwave resonator.

[0023] In a fifth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the resonant frequency prediction method described in any one of the above is implemented.

[0024] In a sixth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned resonant frequency prediction methods.

[0025] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0026] The present application provides a method for generating different two-dimensional coding patterns by adjusting whether each area in each area array in a coding layer adsorbs nanosilver sheets, designing and generating a metamaterial structure microwave resonator, and applying metamaterials to the field of microwave resonator design; and using a network analyzer to read the resonant frequencies of different metamaterial structure microwave resonators with multiple different two-dimensional coding patterns to construct a training data set to train a neural network model. Since the resonant frequencies of different metamaterial structure microwave resonators serve as data support, it is possible to simultaneously predict multiple resonant peaks of the metamaterial microwave resonator, thereby improving the accuracy of predicting the resonant frequency of the metamaterial structure microwave resonator to be tested. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 A schematic diagram of the structure of a metamaterial microwave resonator provided in one embodiment of the present application;

[0029] Figure 2 A schematic flow chart of a method for predicting the resonant frequency of a microwave resonator with a metamaterial structure provided in one embodiment of the present application;

[0030] Figure 3 A schematic diagram of a Seq2Seq-based neural network model prediction process provided in one embodiment of the present application;

[0031] Figure 4 A schematic diagram of neural network model training and application prediction provided in one embodiment of the present application;

[0032] Figure 5 A schematic diagram of the frequency response of a temperature sensor prepared in one embodiment of the present application at different temperatures;

[0033] Figure 6 A schematic diagram of the frequency response of a humidity sensor prepared in one embodiment of the present application at different humidity levels. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0036] The present application provides a metamaterial structure microwave resonator, such as Figure 1 As shown, the metamaterial structure microwave resonator includes: a coding pattern layer 1, a first dielectric layer 2 and a second dielectric layer 3 printed in sequence from top to bottom; the coding pattern layer 1 is a metamaterial layer, and the coding pattern layer 1 is divided into a plurality of area arrays; each area in the area array is designed as a blank area or an area for adsorbing nanosilver sheets to form a two-dimensional coding pattern.

[0037] In an exemplary embodiment, the first medium layer 2 is PEGDA resin; the second medium layer 3 is EA resin.

[0038] Furthermore, the coding pattern is set as follows: the top 2cm*2cm area is divided into 5*5 small areas. In each small area, there are two options: blank and adsorbed nanosilver flakes. Therefore, there are 2 options for the coding metamaterial structure microwave resonator. 25 Variations of patterns.

[0039] The first dielectric layer 2 is PEGDA resin, which is prepared by mixing phenyl bis (2,4,6-trimethylbenzoyl) phosphine oxide and polyethylene glycol diacrylate.

[0040] The second dielectric layer 3 is EA resin. The first dielectric layer 2 and the second dielectric layer 3 are both used as the substrate of the microwave resonator based on the coding metamaterial structure.

[0041] The coding pattern layer 1, the first medium layer 2 and the second medium layer 3 are printed in sequence by a DLP-3D printer. During the printing process, the resin material in the material box needs to be replaced. The interface strength between different resin materials is reliable and not easy to fall off.

[0042] The present application embodiment provides a method for predicting the resonant frequency of a metamaterial structure microwave resonator. The method is executed by a computer device, and specifically can be executed by a computer device such as a terminal or a server alone, or can be executed by a terminal and a server together. In the present application embodiment, Figure 2 As shown, the method includes the following steps.

[0043] S1: adjusting each region in the region array within the metamaterial structure microwave resonator, randomly generating a variety of two-dimensional coding patterns, and constructing different metamaterial structure microwave resonators; one two-dimensional coding pattern corresponds to one metamaterial structure microwave resonator.

[0044] S2: Use a network analyzer to read the resonant frequencies of microwave resonators with different metamaterial structures.

[0045] S3: Convert the two-dimensional coded image into a one-dimensional vector, and determine feature information of the one-dimensional vector.

[0046] S4: Using the characteristic information as input and the corresponding resonant frequency as output, training a neural network model.

[0047] S5: Predicting the resonant frequency of the microwave resonator of the metamaterial structure to be tested according to the neural network model.

[0048] In an exemplary embodiment, S2 may be replaced by the following steps.

[0049] S21: transmitting electromagnetic waves of different frequency bands to the metamaterial structure microwave resonator through an interrogation antenna, and using a network analyzer to reflect the resonant frequency position of the metamaterial microwave resonator.

[0050] S22: Determine the resonant frequencies of microwave resonators with different metamaterial structures according to the resonant frequency positions.

[0051] In an exemplary embodiment, S3 may be replaced by the following steps.

[0052] S31: Convert the two-dimensional coding image into a two-dimensional coding matrix.

[0053] S32: Flatten the two-dimensional coding matrix and adjust it to a one-dimensional vector.

[0054] S33: performing convolution processing on the one-dimensional vector and adjusting the convolution kernel information to determine the feature information of the one-dimensional vector; the convolution kernel information includes size and stride.

[0055] In an exemplary embodiment, S4 can be replaced by the following steps, where the neural network model includes a long short-term memory neural network and a sequence network connected in sequence.

[0056] S41: Generate a feature information sequence based on all feature information.

[0057] S42: Input the feature information sequence into the long short-term memory neural network, determine the long-term dependency between each feature information, and output a feature value sequence.

[0058] S42: Input the characteristic value sequence into the sequence network, output the resonance frequency corresponding to all characteristic information, and complete the training of the neural network model.

[0059] Taking the generation of 120 types of two-dimensional codes as an example, the technical solution of the present application is further explained.

[0060] Step 1: According to the above microwave resonator structure, 120 coding patterns were randomly generated, and 120 different patterns of coded metamaterial structure microwave resonators were prepared using a DLP-3D printer.

[0061] Step 2: Connect the interrogation antenna to the network analyzer, which reads the resonant frequency of the microwave resonator with different coded pattern metamaterial structures. Electromagnetic waves of different frequency bands are emitted from the interrogation antenna. The microwave resonator can absorb electromagnetic waves of a specific frequency and reflect electromagnetic waves of other bands back. The network analyzer reflects the position of the resonant frequency of the microwave resonator through the S11 parameter, which characterizes the numerical value of the resonant frequency within the measurement frequency range. The resonant frequencies of the microwave resonators of different patterns are recorded, and the recorded resonant frequencies are used as the data set for the training of the long short-term memory neural network.

[0062] In practical applications, the interrogation antenna is located about 10 cm above the sensor; the S11 parameter represents the ratio of the reflected wave to the incident wave at port 1, usually expressed in decibels. The absolute value of the S11 parameter converted to a decibel value is the return loss.

[0063] Step 3: If Figure 3As shown in the figure, the Seq2Seq (Sequence to Sequence) neural network model is used to quickly evaluate the sensor performance parameters of a specified surface shape, where x t is the one-dimensional vector corresponding to the t-th feature information of the input, that is, the feature vector, t = 0, 1, 2, 3, .. n, n is the total number of one-dimensional vectors; h t is the t-th output vector; A is a single recurrent unit; σ is the Sigmoid activation function; and tanh is the activation function.

[0064] like Figure 4 As shown, it includes a model training stage and a model application prediction, wherein the model training stage includes a training data set, a model to be trained and an evaluation index. The model to be trained is trained using the training data set, and the parameters of the model to be trained are optimized using the evaluation index; the model application prediction includes predicting sensor coding, pre-training models and resonant frequency, and the model application prediction stage has the same process as the model training stage.

[0065] During the training phase, the model receives the two-dimensional coding pattern on the surface of the sensor device as input information.

[0066] The two-dimensional coding matrix converted from the two-dimensional coding pattern is flattened and adjusted to a one-dimensional vector. For example, for a matrix of size n×n, it is spliced ​​along the rows to become 1 row×n 2 Then, 1D convolution is used to collect the feature information of the one-dimensional vector by adjusting the size and stride of the convolution kernel.

[0067] The feature information is then used as the input of the neural network model. Considering that each feature information may have an influencing relationship with each other, the long short-term memory neural network is used to more effectively capture and process the long-term dependencies in the feature information. In the sequence network output process, the output vectors are generated one by one, and an output sequence is generated based on the output vector. The output sequence is the corresponding resonant frequency. When the output contains a sequence termination symbol, the model prediction process is terminated, and all results before this sequence termination symbol are the prediction results of the model.

[0068] The predicted resonant frequency is converted into a more accurate resonant frequency through an evaluation index, which is the mean square error (MSE), such as Figure 3 As shown, the predicted results corresponding to the output sequence are 3.32, 6.10, and 7.16, and the corresponding actual results are 3.42807, 6.05124, and 7.2221.

[0069] In this application, considering that the device surface matrix is ​​small, in the convolution process processed by the encoder, the 1D convolution kernel size is set to 25 to handle the possible mutual influence between distant coding units, the stride is set to 1, the padding is set to 'valid' to avoid the impact of additional padding, and the 1-channel input is expanded to 128 channels as output features. The output of the convolution layer is used as the input of the LSTM layer of the decoder, and the LSTM has 128 hidden layer neurons. The model calculates and adjusts the weight parameters in the model through the Adam optimization algorithm provided by the Pytorch framework, and finally achieves the goal of minimizing the mean square error between the model results and the actual results.

[0070] Since the output sequences are of different lengths, the longest length L of the output is obtained first, and then the terminator '0' is added to each sequence in the data set. The length of each group of outputs is adjusted to a uniform L+1. During the model output process, the prediction results are output one by one. When the output character is the terminator '0', the output is terminated. The output result is the prediction result of the model, where the valid number before the terminator is the frequency of each resonance peak.

[0071] The following describes the application of the metamaterial structure microwave resonator as a sensor.

[0072] 1) Using metamaterial structure microwave resonator as temperature sensor:

[0073] Sensing mechanism: The change of dielectric constant of the substrate material at different temperatures causes the resonant frequency of the sensor to shift. Figure 5 As shown, in the range of 40°C to 90°C, the frequency shift range of the sensor is 67.3MHz, and the sensor can observe temperature changes through frequency shift.

[0074] 2) Using metamaterial structure microwave resonator as humidity sensor:

[0075] Preparation process: On the basis of the original nanosilver sheet, a layer of molybdenum disulfide is deposited. First, a 1%wt concentration of molybdenum disulfide dispersion (with deionized water as a carrier) is prepared, and the dispersion is dripped onto the device using a dropper, and heated and dried on a heating table for 2 hours.

[0076] Sensing mechanism: The reason why MoS2 can be used for humidity sensing is the charge transfer between H2O and the MoS2 surface. Due to the active defects generated by Mo atoms at the edges of MoS2 nanosheets, H2O is easily captured by the Mo-O bonds formed by the defects. Then, the H2O physically absorbed by the MoS2 film is ionized into H by the electric field on the metamaterial structure. + and OH - The migration and transfer of H+ under the guidance of electric field leads to the increase of dielectric constant of metamaterial structure, which is reflected in the frequency shift of resonant frequency. Figure 6 As shown, within the range of 40% RH to 90% RH, the frequency shift range of the sensor is 99.0 MHz, and the sensor can observe the change of humidity through frequency shift.

[0077] As a result of randomly selecting two devices with metamaterial coding patterns as sensor tests, it can be seen that the frequency shift is not a fixed value, but changes with the change of the coding pattern.

[0078] Based on the shortcomings of the prior art, this application has the following advantages:

[0079] Advantage 1: The traditional manual metamaterial design method mainly relies on the designer's electromagnetic knowledge, metamaterial design experience and a large number of attempts. It has a long design cycle, high requirements for designers, and is not conducive to the batch design of metamaterials. This application prepares a microwave resonator based on a coding metamaterial structure and uses machine learning to improve design efficiency, aiming to achieve large-scale and rapid metamaterial design.

[0080] Advantage 2: At present, the design of metamaterial structures based on machine learning is mainly aimed at the field of mechanical metamaterials. This application accurately predicts the resonant frequency of the microwave resonator of the metamaterial structure based on the property that metamaterials can efficiently absorb reflected electromagnetic waves.

[0081] Advantage 3: Long short-term memory network has high accuracy. Conventional algorithms can only predict a specified number of resonance peaks, but the number of peaks that may exist in reality is not exactly equal. It can realize numerical prediction of resonance peak frequency and predict the number of resonance peaks according to different patterns on the device surface at the same time.

[0082] Based on the same inventive concept, the embodiment of the present application also provides a resonant frequency prediction device for a metamaterial structure microwave resonator for implementing the resonant frequency prediction method for a metamaterial structure microwave resonator involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of the resonant frequency prediction device for one or more metamaterial structure microwave resonators provided below can refer to the limitations of the resonant frequency prediction method for a metamaterial structure microwave resonator above, and will not be repeated here.

[0083] In an exemplary embodiment, a resonant frequency prediction device of a metamaterial structure microwave resonator is provided, comprising:

[0084] The two-dimensional coding pattern random generation module is used to adjust each area in the area array within the metamaterial structure microwave resonator, randomly generate multiple two-dimensional coding patterns, and construct different metamaterial structure microwave resonators; one two-dimensional coding pattern corresponds to one metamaterial structure microwave resonator.

[0085] Network analyzer, used to read the resonant frequencies of microwave resonators with different metamaterial structures.

[0086] The feature information determination module is used to convert the two-dimensional coded image into a one-dimensional vector and determine the feature information of the one-dimensional vector.

[0087] The neural network model training module is used to train the neural network model using the characteristic information as input and the corresponding resonant frequency as output.

[0088] The resonant frequency prediction module is used to predict the resonant frequency of the microwave resonator of the metamaterial structure to be tested according to the neural network model.

[0089] In an exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the resonant frequency prediction data of the metamaterial structure microwave resonator. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a resonant frequency prediction method of a metamaterial structure microwave resonator is implemented.

[0090] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the above method is implemented when the processor executes the computer program.

[0091] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the above method when executed by a processor.

[0092] In an exemplary embodiment, a computer program product is provided, including a computer program, which implements the above method when executed by a processor.

[0093] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ReadOnlyMemory, ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (Magnetoresistive RandomAccess Memory, MRAM), ferroelectric random access memory (Ferroelectric RandomAccess Memory, FRAM), phase change memory (Phase ChangeMemory, PCM), graphene memory, etc. Volatile memory can include random access memory (RandomAccess Memory, RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0094] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0095] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0096] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0097] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A metamaterial structure microwave resonator, characterized in that: The metamaterial structure microwave resonator comprises: a coding pattern layer, a first dielectric layer and a second dielectric layer printed sequentially from top to bottom; The coding pattern layer is a metamaterial layer, and the coding pattern layer is divided into a plurality of area arrays; each area in the area array is designed as a blank area or an area for adsorbing nano-silver sheets to form a two-dimensional coding pattern.

2. The metamaterial structure microwave resonator according to claim 1, characterized in that: The first medium layer is PEGDA resin; the second medium layer is EA resin.

3. A method for predicting the resonant frequency of a metamaterial structure microwave resonator, characterized in that: include: Adjusting each region in the region array of the metamaterial structure microwave resonator according to claims 1-2, randomly generating a variety of two-dimensional coding patterns, and constructing different metamaterial structure microwave resonators; A two-dimensional coding pattern corresponds to a metamaterial structure microwave resonator; Use a network analyzer to read the resonant frequencies of microwave resonators with different metamaterial structures; Converting the two-dimensional coded image into a one-dimensional vector and determining feature information of the one-dimensional vector; Training a neural network model using the characteristic information as input and the corresponding resonant frequency as output; The resonant frequency of the microwave resonator of the metamaterial structure to be tested is predicted according to the neural network model.

4. The method for predicting the resonant frequency of a metamaterial structure microwave resonator according to claim 3, characterized in that: Use a network analyzer to read the resonant frequencies of microwave resonators with different metamaterial structures, including: Transmitting electromagnetic waves of different frequency bands to the metamaterial structure microwave resonator through an interrogation antenna, and using a network analyzer to reflect the resonant frequency position of the metamaterial microwave resonator; The resonant frequencies of microwave resonators with different metamaterial structures are determined according to the resonant frequency positions.

5. The method for predicting the resonant frequency of a metamaterial structure microwave resonator according to claim 3, characterized in that: Converting the two-dimensional coded image into a one-dimensional vector and determining feature information of the one-dimensional vector specifically includes: Converting the two-dimensional coded image into a two-dimensional coded matrix; Flatten the two-dimensional coding matrix and adjust it to a one-dimensional vector; Convolution processing is performed on the one-dimensional vector, and convolution kernel information is adjusted to determine feature information of the one-dimensional vector; the convolution kernel information includes size and stride.

6. The method for predicting the resonant frequency of a metamaterial structure microwave resonator according to claim 3, characterized in that: Taking the characteristic information as input and the corresponding resonant frequency as output, training a neural network model specifically includes: the neural network model includes a long short-term memory neural network and a sequence network connected in sequence; Generate a feature information sequence according to all feature information; Inputting the feature information sequence into the long short-term memory neural network, determining the long-term dependency between each feature information, and outputting a feature value sequence; The characteristic value sequence is input into the sequence network, and the resonance frequency corresponding to all characteristic information is output to complete the training of the neural network model.

7. A resonant frequency prediction system for a metamaterial structure microwave resonator, characterized in that: include: A two-dimensional coding pattern random generation module, used to adjust each area in the area array in the metamaterial structure microwave resonator according to claims 1-2, randomly generate a variety of two-dimensional coding patterns, and construct different metamaterial structure microwave resonators; A two-dimensional coding pattern corresponds to a metamaterial structure microwave resonator; Network analyzer, used to read the resonant frequency of microwave resonators with different metamaterial structures; A feature information determination module, used for converting the two-dimensional coded image into a one-dimensional vector and determining feature information of the one-dimensional vector; A neural network model training module, used to train a neural network model using the characteristic information as input and the corresponding resonant frequency as output; The resonant frequency prediction module is used to predict the resonant frequency of the microwave resonator of the metamaterial structure to be tested according to the neural network model.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the metamaterial structure microwave resonator according to any one of claims 3 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the metamaterial structure microwave resonator according to any one of claims 3 to 6 is realized.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the metamaterial structure microwave resonator according to any one of claims 3 to 6 is realized.

Images