An optimization design method of a broadband electromagnetic wave absorbing coded metamaterial and a related device

Abstract

The application belongs to the technical field of electromagnetic metamaterial optimization design, and discloses a kind of optimization design method and related device of broadband electromagnetic wave-absorbing coding metamaterial, comprising: generating grid weighting-based metamaterial structure, using coding matrix to characterize, calculating the fitness value of coded metamaterial structure;According to the fitness value of the metamaterial structure, the weight disturbance ratio and learning rate are adaptively adjusted based on the adaptive gradient descent algorithm to generate the optimal weight;According to the optimal weight, generate initial population, establish the complementary adjustment mechanism of crossover rate and mutation rate by adaptive genetic algorithm, regulate and control the metamaterial structure, obtain the optimal topological structure and optimal parameter of electromagnetic wave-absorbing coding metamaterial.The application combines the local search of gradient algorithm and the global optimization ability of genetic algorithm, greatly reduces the early invalid search and improves the global convergence speed, improves the wave-absorbing bandwidth of metamaterial, and does not need to rely on complex physical modeling and artificial scheduling.

Description

Technical Field

[0001] This invention belongs to the field of electromagnetic metamaterial optimization design technology, and relates to an optimization design method and related device for broadband electromagnetic wave-absorbing coded metamaterials. Background Technology

[0002] Electromagnetic absorbing metamaterials can achieve efficient absorption of incident electromagnetic waves through the resonant response of their subwavelength structure. These metamaterials possess wide bandwidth, wide-angle incidence, and polarization insensitivity characteristics, while also being lightweight and thin, making them widely applicable in fields such as radar stealth, non-destructive testing, high-sensitivity sensing, and wireless communication.

[0003] Currently, the design of electromagnetic absorbing metamaterials mainly relies on empirical methods. Designers typically propose initial unit configurations based on existing experience and achieve design goals by repeatedly adjusting the shape and geometric parameters of the unit structures. However, due to the diverse shapes, geometric parameters, and material types of metamaterials, this approach is inefficient and lacks the ability to handle complex modeling, making it difficult to improve the absorption bandwidth of metamaterials. In recent years, topology optimization methods based on surface structure discretization have been introduced into the design of electromagnetic absorbing metamaterials. This method discretizes the resonant structure of the metamaterial into a coded sequence, overcoming the over-reliance on experience and the limitations of simple models in traditional design. However, existing optimization designs of coded metamaterials often employ non-gradient evolutionary algorithms such as genetic algorithms. The randomization of the initial population makes it difficult to directly generate topological structures with excellent resonant characteristics, which can easily mislead the direction of subsequent structural optimization. Secondly, traditional genetic algorithms use fixed crossover and mutation rates. If the parameter values ​​are set too low, it can easily lead to homogenization of the population topology; if the parameter values ​​are set too high, it can easily destroy excellent structures, making it difficult to design high-performance electromagnetic absorbing structures. Therefore, there is an urgent need for an optimization design method that can effectively guide the direction of structural optimization and automatically adjust the degree of structural changes according to the population evolution state, so as to achieve rapid and high-performance design of electromagnetic wave absorbing coded metamaterials. Summary of the Invention

[0004] The purpose of this invention is to provide an optimization design method and related device for broadband electromagnetic absorbing coded metamaterials, which solves the problem that existing genetic algorithms have difficulty converging to the global optimum due to the randomization of the initial population and fixed parameters, thus limiting the improvement of absorption performance.

[0005] To achieve the above objectives, the present invention employs the following technical solution: An optimized design method for broadband electromagnetic wave-absorbing coded metamaterials includes: Generate mesh-weighted metamaterial structures, characterize them using an encoding matrix, and calculate the fitness value of the encoded metamaterial structures. Based on the fitness value of the metamaterial structure, the optimal weights are generated by adaptively adjusting the weight perturbation ratio and learning rate using an adaptive gradient descent algorithm. An initial population is generated based on the optimal weights. An adaptive genetic algorithm is used to establish a complementary regulation mechanism between crossover rate and mutation rate to control the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

[0006] Furthermore, the method for calculating the fitness value of metamaterial structures is as follows:

[0007] in, This represents the fitness value of the metamaterial structure. This represents the average absorption rate of the metamaterial within the target frequency band. This represents the proportion of frequency points in the target frequency band where the metamaterial's absorption rate exceeds a preset target value, out of the total number of frequency points in the entire frequency band. This represents the maximum absorption rate of the metamaterial at the target frequency band. , and These represent the average weighting factor, bandwidth weighting factor, and peak weighting factor, respectively.

[0008] Furthermore, the weight perturbation ratio is:

[0009] The learning rate is:

[0010] Among them, Indicates the weight perturbation ratio. Indicates the learning rate. , , ... They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. The weight perturbation ratio for each fitness value segment. , , ... They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. The learning rate is segmented into fitness values. , , ... , They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. Each fitness value segment boundary.

[0011] Furthermore, methods for generating optimal weights include: Based on the fitness value of the metamaterial structure, the initial weights are positively perturbed by a selected weight perturbation ratio to obtain the first new weights; Calculate the gradient of the difference between the old and new fitness values ​​with respect to the difference between the old and new weights. The gradient guides the update direction of the initial weights, resulting in the second new weights. Based on the second new weights, construct the metamaterial structure and calculate its fitness value. Repeat the above steps until the termination condition is met, and output the final optimal weight.

[0012] Furthermore, the first new weight is:

[0013]

[0014] in, Indicates the first new weight. This represents the weighted perturbation. Indicates the initial weights. Indicates the weight perturbation ratio; The second new weight is:

[0015]

[0016]

[0017]

[0018] in, This indicates the second new weight. This represents the gradient of the difference between the old and new fitness values ​​relative to the difference between the old and new weights. This represents the fitness value of the metamaterial structure established based on the first new weight. This represents the fitness value of the metamaterial structure.

[0019] Furthermore, methods for regulating metamaterial structures through complementary modulation mechanisms of crossover rate and variation rate include: The crossover rate, mutation rate, crossover distribution index, and mutation distribution index are adaptively adjusted based on the current population fitness value, while preventing the parameters from exceeding the extreme value boundary. The best individuals are selected based on the elite retention ratio and directly enter the offspring population; A tournament selection mechanism is used to select the parent from the current individual; Perform two-point crossover and bit mutation operations on the encoding matrix of the selected parent individuals, and perform simulated binary crossover and polynomial mutation operations on the real matrix to generate a new offspring population. The fitness value of each individual in the new population is calculated, and through iteration, the optimal solution is output as the optimal topology and optimal parameters of the electromagnetic wave absorbing encoded metamaterial.

[0020] Furthermore, the crossover rate is:

[0021]

[0022] The mutation rate is:

[0023]

[0024] The cross-distribution index is:

[0025]

[0026] The variation distribution index is:

[0027]

[0028] in, Indicates the crossover rate. This indicates the upper limit of the cross rate. This indicates the lower bound of the crossover rate. , , Both represent nonlinear parameters. This represents the average fitness value of the population. The variance of the population fitness values ​​is represented by the following: Indicates the rate of variation. This indicates the upper limit of the mutation rate. This indicates the lower limit of the mutation rate. Indicates the cross-distribution index. This indicates the upper limit of the cross-distribution index. This indicates the lower bound of the cross-distribution index. Indicates the distribution index of variation. This represents the upper limit of the variation distribution index. This indicates the lower bound of the variation distribution index.

[0029] An optimized design system for broadband electromagnetic absorbing coded metamaterials includes: The encoding module is used to generate mesh-weighted metamaterial structures, which are characterized using an encoding matrix. The fitness value of the encoded metamaterial structure is then calculated. The adjustment module is used to adaptively adjust the weight perturbation ratio and learning rate based on the fitness value of the metamaterial structure and the adaptive gradient descent algorithm to generate the optimal weights. The regulation module is used to generate an initial population based on the optimal weights, and to establish a complementary regulation mechanism between crossover rate and mutation rate through an adaptive genetic algorithm to regulate the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

[0030] A terminal device includes 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 steps of the method.

[0031] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method.

[0032] Compared with the prior art, the present invention has the following beneficial effects: This invention provides an optimization design method for broadband electromagnetic wave-absorbing coded metamaterials. By uniformly representing the generated grid-weighted metamaterial structure using an encoding matrix, and adaptively adjusting the weight perturbation ratio and learning rate based on an adaptive gradient descent algorithm, a high-performance solution space is quickly located, generating an initial structure with excellent resonance characteristics, providing a high-quality initial population for subsequent genetic algorithms. Subsequently, a complementary adjustment mechanism between crossover and mutation rates is established through an adaptive genetic algorithm to finely control the metamaterial structure, obtaining the optimal topology and parameters of the electromagnetic wave-absorbing coded metamaterial, thus improving the metamaterial's wave-absorbing performance. This invention combines optimization algorithms and metamaterial design, integrating the local search capability of gradient algorithms and the global optimization capability of genetic algorithms. Gradient algorithms are used to quickly find high-quality metamaterial structures, and then genetic algorithms are used to perform more refined optimization design of the metamaterial structure, resulting in a higher-performance electromagnetic wave-absorbing coded metamaterial. Improving from the root of initial population generation, the gradient algorithm generates a high-quality initial structure to guide the algorithm's optimization, significantly reducing later invalid searches, significantly improving global convergence speed, significantly reducing the design cycle, improving wave-absorbing performance, and eliminating the need for complex physical modeling.

[0033] Furthermore, by employing gradient decoupling and adaptive weight perturbation ratios and learning rates, the method increases the magnitude of structural changes when structural performance is poor to seek a better structure, and decreases the magnitude of structural changes when structural performance is good to provide protection, thereby enhancing the stability of structural optimization. This method can quickly generate metamaterial structures with excellent wave-absorbing properties in the early stages of optimization, avoiding the problem that using completely random initial structures in large-scale designs may mislead subsequent optimization directions, and significantly improving optimization efficiency.

[0034] Furthermore, through the complementary regulation mechanism of crossover rate and mutation rate, the mutation rate is reduced when the crossover rate increases to prevent excessive damage to the structure, and the mutation rate is increased when the crossover rate decreases to supplement the regulation of the structure. This achieves a dynamic balance between development and protection, and provides more precise structural regulation capabilities and stronger structural diversity maintenance capabilities.

[0035] Furthermore, the adaptive genetic algorithm establishes a nonlinear mapping relationship between fitness values ​​and algorithm parameters, encouraging the algorithm to generate diverse topologies in the early stages of iteration to increase optimization capabilities, and protecting the superior structure from destruction in the later stages of iteration. This adaptive mechanism can continuously adjust the structural design accuracy in multiple iterations, avoiding the problems of structural homogenization and the destruction of excellent structures, thus enabling the final designed metamaterial to have better wave absorption performance. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of the optimized design method for the broadband electromagnetic absorbing coded metamaterial of the present invention.

[0038] Figure 2 This is a schematic diagram of the unit structure of the electromagnetic wave-absorbing encoded metamaterial in Embodiment 1 of the present invention.

[0039] Figure 3 This is a comparison chart of the convergence curves of the optimal fitness values ​​of the optimization method and the genetic algorithm in Embodiment 1 of the present invention.

[0040] Figure 4 This is a comparison chart of the optimization results of the optimization method and the genetic algorithm in Embodiment 1 of the present invention.

[0041] Figure 5 A schematic diagram of the optimal topological coding configuration obtained by the genetic algorithm.

[0042] Figure 6 This is a schematic diagram of the optimal topology coding configuration obtained by the optimization method in Embodiment 1 of the present invention.

[0043] Figure 7 This is a schematic diagram of the optimized design system structure of a broadband electromagnetic wave-absorbing coded metamaterial according to a preferred embodiment of the present invention.

[0044] Figure 8 This is a schematic diagram of the electronic device structure according to a preferred embodiment of the present invention. Detailed Implementation

[0045] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0046] Obviously, the described embodiments are only some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0047] It should be noted that the terminals involved in the embodiments of this application may include, but are not limited to, mobile phones, personal digital assistants (PDAs), wireless handheld devices, tablet computers, personal computers (PCs), MP3 players, MP4 players, wearable devices (e.g., smart glasses, smartwatches, smart bracelets), smart home devices, and other smart devices.

[0048] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0049] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 1 This invention provides an optimized design method for broadband electromagnetic absorbing coded metamaterials, specifically including the following steps: Construct a grid-weighted topology generation method: using 0 and total weight Given a lower bound and an upper bound, generate an array with a number of elements. An equally spaced sequence, where the interval between adjacent elements is... for: (1) Randomly fill a number with the elements of this sequence. In the matrix, each element is then binarized to obtain a final matrix of size . The binary matrix of 0 / 1 is represented by the binarization function shown in equation (2): (2) in, Represents matrix elements, As the binarization threshold, when hour, At this time, the matrix elements take the value 1; when hour, At this point, the matrix elements take the value 0.

[0050] The result Perform rotational symmetry operations on the submatrix to obtain a complete one. The binary matrix is ​​used as the encoding matrix for modeling, where This allows the designed metamaterial to have a rotationally symmetric configuration, ensuring that it possesses polarization insensitivity properties.

[0051] The fitness function used to represent the properties of electromagnetic absorbing encoded metamaterials is: (3) (4) (5) in, This represents the fitness value of the metamaterial structure. This represents the average absorption rate of the metamaterial within the target frequency band. This indicates that the metamaterial is at angular frequency Absorption rate at the location, Indicates the total number of frequency points within the target frequency band; This represents the proportion of frequency points in the target frequency band where the metamaterial's absorption rate exceeds a preset target value, out of the total number of frequency points in the entire frequency band. When the absorption rate at the target frequency point exceeds the preset target value... Select 1, otherwise select 0; This represents the maximum absorption rate of the metamaterial at the target frequency band. , and These represent the average weighting factor, bandwidth weighting factor, and peak weighting factor, respectively.

[0052] Phase 1: Complete the optimization of the grid-weighted adaptive gradient descent algorithm.

[0053] Based on the initial weights An initial configuration of the metamaterial is generated, and the generated initial configuration is discretized and encoded using real numbers. The encoded design variables are denoted as... , It includes the topological configuration and geometric parameters of the metamaterial, based on design variables. Construct metamaterial structures and calculate their fitness values ; The weight perturbation ratio and learning rate are adaptively adjusted based on the calculated fitness value, as shown in equations (6) and (7): (6) (7) in, Indicates the weight perturbation ratio. Indicates the learning rate. , , ... They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. The weight perturbation ratio for each fitness value segment. , , ... They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. The learning rate is segmented into fitness values. , , ... , They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. Each fitness value segment boundary.

[0054] With the selected weight perturbation ratio For initial weights A positive perturbation is performed to obtain the first new weight. As shown in equations (8) and (9): (8) (9) in, This represents the weighted perturbation. Indicates the initial weights. This represents the first new weight. Based on the first new weight, a metamaterial structure is constructed and its fitness value is calculated. .

[0055] Calculate the gradient of the difference between the old and new fitness values ​​relative to the difference between the old and new weights. Gradient-guided initial weights The update direction yields the second new weight. As shown in equations (10) to (13): (10) (11) (12) (13) Based on the new weights, a metamaterial structure is constructed and its fitness value is calculated. .

[0056] Repeat the above steps until the fitness value is reached. Once the termination condition is met or the algorithm reaches its maximum number of iterations, the final optimal weights are output.

[0057] Phase 2: Complete the optimization based on the nonlinear adaptive genetic algorithm.

[0058] Maximum number of iterations for the initialization algorithm Population size Dimension of the encoding submatrix Nonlinear parameters , , Cross rate limit Crossover rate lower limit Upper limit of mutation rate Lower limit of variation Upper limit of cross-distribution index Lower limit of cross-distribution index Upper limit of the variation distribution index Lower limit of the distribution index of variation and the proportion of elites retained The optimal weights obtained in the first stage of optimization are used to guide the generation of the initial population. .

[0059] Calculate the fitness value of each individual in the current population and the average fitness value of the population. and variance Calculate the adaptive crossover rate. Variation rate Cross-distribution index and variation distribution index At the same time, it prevents the parameters from exceeding the extreme value boundary, as shown in equations (14) to (21): (14) (15) (16) (17) (18) (19) (20) (twenty one) With the proportion of elites retained The best individual is selected and directly enters the offspring population. A tournament selection mechanism is used to select a parent from the current individuals. Two-point crossover and bit mutation operations are performed on the encoding matrix of the selected parent individual. Simulated binary crossover and polynomial mutation operations are performed on the real number matrix to generate a new offspring population. The fitness value of each individual in the new offspring population is calculated. The above steps are repeated until the fitness value reaches the termination condition or the algorithm reaches the maximum number of iterations, at which point the final best individual is output. , It includes the optimal topological configuration and optimal geometric parameters of metamaterials.

[0060] The present invention will be further described in detail below through specific embodiments: Example 1: In this embodiment, the metamaterial structure coding method based on a 6-layer structure includes: To facilitate the mathematical characterization of the electromagnetic wave-absorbing coded metamaterial design, a 0 / 1 encoding matrix is ​​used to represent the topological configuration of the two-layer resistive film of the metamaterial, and real numbers are used to encode the geometric parameters of the metamaterial. During the design process, the 0 / 1 encoding matrix can effectively represent different structures and simplifies the optimization of the metamaterial configuration.

[0061] The electromagnetic absorbing coded metamaterial designed in this embodiment adopts a 6-layer structure, such as... Figure 2 As shown. To increase the degree of freedom in the optimization design, the topological configurations of the upper and lower resistive films are represented by two independent encoding matrices. To ensure the polarization insensitivity of the metamaterial absorber and enhance the absorption performance optimization effect, the topological configuration of the resistive films is designed with rotational symmetry. The upper and lower resistive film layers are generated by three rotational symmetries from two specific patterns.

[0062] Specifically, through the encoding of a 6-layer metamaterial unit structure, the interaction between the resistive film layer and the dielectric layer can effectively regulate the propagation characteristics of electromagnetic waves, reduce reflection and enhance absorption performance. The rotational symmetry of the resistive film topology enables the electromagnetic wave absorption encoded metamaterial to maintain good wave absorption effect under different polarizations and incident angles.

[0063] In this embodiment, the automated modeling and calculation of an individual's absorption rate in the operating frequency band includes: A simulation environment for the metamaterial absorber was set up, with simulation configurations including operating frequency, background material, resistive film material, dielectric material, boundary conditions, global units, and solver type. Specifically, the designed metamaterial operates in the frequency band of 0.1–14 THz. Axial direction and All axial directions are set to unit cell boundaries. The positive axis direction is set to the open boundary. The negative axis direction is set as the electric boundary, the global unit is μm, and the solver type is set as frequency domain solver.

[0064] In the simulation environment described above, a three-dimensional structural model of the metamaterial is established, and a model is created on each of the two required two-dimensional planes. A matrix, where each element is either 0 or 1, corresponding to... Each block has two possible states: filled or unfilled with resistive film material. 1 represents filled, and 0 represents unfilled.

[0065] Specifically, the electromagnetic wave absorbing encoded metamaterial structure was established and simulation calculations were performed. Figure 2 This is a schematic diagram of the unit structure of an electromagnetic wave-absorbing coded metamaterial. The metamaterial unit structure consists of six layers: the top layer is the first dielectric, the upper resistive film topology coding layer is located between the second and third dielectrics, the lower resistive film topology is located between the fourth and fifth dielectrics, and the bottom layer is a metal reflector. The relative permittivity of the first, fourth, and fifth dielectrics is 4.3, and the loss tangent is 0.025°. The relative permittivity of the second and third dielectrics is 2.65, and the loss tangent is 0.015°. Let the thickness of the first dielectric be... The thicknesses of the second and third media are The thicknesses of the fourth and fifth media are The sheet resistance of the upper resistive film is The sheet resistance of the lower resistive film is Two 7×7 sub-matrices are created, and rotational symmetry is performed on these two sub-matrices to generate two 14×14 rotational symmetric matrices, which serve as the encoding matrices corresponding to the topology of the upper and lower resistive films, respectively. Each encoding matrix has 14×14 blocks with a total of 196 square units. Each unit has two possibilities: filled or unfilled with resistive film material. Code 0 represents unfilled, and code 1 represents filled.

[0066] Full-wave simulation calculations were performed on the metamaterial structures to obtain the three-dimensional structural diagrams of each metamaterial and the absorption rate at the set operating frequency band.

[0067] Specifically, the target frequency band was set to 0.1~14 THz. After the calculation, the absorption rate of the electromagnetic absorbing coded metamaterial was obtained, as shown in the following formula: (twenty two) in, For reflectivity, Transmittance, The reflection coefficient, is the transmission coefficient.

[0068] In the optimization process, the fitness function is the standard for evaluating the performance of an individual. The optimization objective of this embodiment is to maximize the absorption rate and broaden the absorption bandwidth within the target frequency band. Therefore, the fitness function in this embodiment is: (twenty three) (twenty four) (25) in, Indicates the individual's fitness value. This represents the average absorption rate of the metamaterial within the target frequency band. This indicates that the metamaterial is at angular frequency Absorption rate at the location, This indicates the total number of frequency points within the target frequency band. This represents the proportion of frequency points in the target frequency band where the metamaterial's absorption rate exceeds a preset target value, out of the total number of frequency points in the entire frequency band. When the absorption rate at the target frequency point exceeds the preset target value... Select 1, otherwise select 0. This represents the maximum absorption rate of the metamaterial at the target frequency band. , and These are the average weighting factor, bandwidth weighting factor, and peak weighting factor, respectively, with values ​​of 0.4, 0.4, and 0.2.

[0069] The optimized design method for broadband electromagnetic absorbing coded metamaterials in this embodiment specifically includes the following steps: Phase 1: Complete the optimization of the grid-weighted adaptive gradient descent algorithm.

[0070] Set the maximum number of optimizations for the gradient algorithm to 20, and the initial weights... Set the weights to 51 to initialize the absorber topology. Generate an initial absorber topology based on the initial weights. The specific process is as follows: With 0 and total weight Given a lower bound and an upper bound, generate an array with a number of elements. A sequence of equally spaced numbers, where the interval between adjacent elements is: (26) (27) Next, randomly fill the elements in this sequence into a... In the matrix, each element is then binarized to obtain a final matrix of size . A 0 / 1 binary matrix: (28) Equation (28) is a binary function, where Represents matrix elements, This represents a threshold value of 50. hour, At this time, the matrix elements take the value 1; when hour, At this point, the matrix elements are all 0. There is now a total weight. A size of was obtained The 0 / 1 binary matrix is ​​then subjected to a symmetry operation according to the rotational symmetry principle to finally obtain a complete symmetric matrix. Binary matrix.

[0071] Fitness value calculation: Next, the absorption rate of the initial individual in the operating frequency band is calculated through full-wave simulation, and the individual fitness value is further calculated. The specific fitness value is a standard used to measure how well an individual performs in solving a specific problem; Set the adaptive weight perturbation ratio and learning rate as shown in equations (29) and (30): (29) (30) in, This represents the percentage of disturbance to the weights. For learning rate, , , ... They represent, , , ... They represent, , , ... , They represent respectively.

[0072] Using fitness values Selected weight perturbation ratio For initial weights Perform a positive perturbation to obtain a perturbed weight. The specific settings are as follows: (31) (32) in, It is the weighted perturbation. These are the initial weights. These are the updated weights. A new topology is generated based on the perturbed weights, and the fitness value of the new topology is obtained through full-wave simulation. .

[0073] Based on the fitness values ​​obtained above The gradient is calculated from the results, and the initial weights are then adjusted. Update to obtain a new weight. The specific settings are as follows: (33) (34) (35) (36) Among them, This is the weight update factor, which takes the value -1 when the gradient is greater than or equal to 0, and takes the value 1 when the gradient is less than 0. This ensures that the weight update direction is in the opposite direction of the current gradient. This represents the weight update magnitude.

[0074] Generate new individuals based on the new weights obtained in the previous step, and calculate the fitness value of the new individuals. Determine if the iteration termination condition has been met. If yes, end the iteration and output the optimal weights; otherwise, repeat the above operations of generating the structure, calculating the gradient, updating the parameters, and calculating the fitness value until the fitness value is reached. Once the termination condition is met or the algorithm reaches its maximum number of iterations, the final optimal weights are output.

[0075] Phase 2: Complete the optimization based on the nonlinear adaptive genetic algorithm.

[0076] Algorithm parameter initialization: Set the population size to 5, the maximum number of iterations to 100, and the nonlinear parameters... , , Set to 9, maximum cross rate. The value is 0.8, which is the lower limit of the cross rate. The value is 0.5, which is the upper limit of the mutation rate. The value is 0.1, which is the lower limit of the mutation rate. The value is 0.001, which is the upper limit of the cross-distribution index. The value is 2, which is the lower limit of the cross-distribution index. The value is 0.01, which is the upper limit of the variation distribution exponent. The value is 10, which is the lower limit of the variation distribution index. The value is 1, representing the elite retention rate. Set it to 0.2.

[0077] The optimal weights obtained in the first stage of optimization are used to guide the generation of the initial population. The fitness value of each individual in the population is calculated.

[0078] Calculate the average fitness value of the population and variance The crossover rate is adaptively adjusted based on the current population fitness value. Variation rate Cross-distribution index and variation distribution index To prevent parameters from exceeding their maximum / minimum boundaries, the specific settings are as follows: (37) (38) (39) (40) (41) (42) (43) (44) With the proportion of elites retained Select the best individuals to directly enter the offspring population; A tournament selection mechanism is used to select the parent from the current individual; Perform two-point crossover and bit mutation operations on the encoding matrix of the selected parent individuals, and perform simulated binary crossover and polynomial mutation operations on the real matrix to generate a new offspring population. Calculate the fitness value of each individual in the new population, and determine if the iteration termination condition has been met. If yes, end the iteration and output the optimal solution as the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial. If not, repeat the above operations of selection, crossover, mutation, and fitness value calculation until the fitness value reaches the termination condition or the algorithm reaches the maximum number of iterations, and output the final best individual. , It includes the optimal topological configuration and optimal geometric parameters of metamaterials.

[0079] In this embodiment, firstly, to address the ineffectiveness of completely random coding, the optimal weights found using gradient information guide the generation of a topology with excellent resonance characteristics, avoiding the problem that using a completely random initial structure in large-scale design might mislead subsequent optimization directions. Secondly, the proposed gradient decoupling and adaptive strategy can dynamically adjust the algorithm parameters according to the population evolution status, continuously adjusting the structural design accuracy in multiple iterations, avoiding the problems of structural homogenization and the destruction of excellent structures, thus enabling the final designed metamaterial to have better wave absorption performance.

[0080] Table 1 shows the range of values ​​for the metamaterial geometric parameters in Example 1. Using the exact same initial population, optimization was performed using the present invention and a genetic algorithm. The optimal parameters and performance obtained by the genetic algorithm were compared with those obtained by the genetic algorithm. Tables 2 and 3 show the comparison of the optimal geometric parameter values ​​of the metamaterial and the performance of the electromagnetic wave absorbing encoded metamaterial. The convergence curve of the optimal fitness value and the optimization results are compared as follows. Figures 3-4 As shown, the optimal topological coding configuration is as follows: Figures 5-6 As shown.

[0081] Table 1. Range of values ​​for the geometric parameters of the metamaterial in Example 1

[0082] Table 2. Optimal geometric parameter values ​​of metamaterials obtained by genetic algorithm and this invention.

[0083] Table 3. Performance of the electromagnetic absorbing coded metamaterial obtained by the genetic algorithm and this invention.

[0084] From Table 2, Figure 4 , Figure 5 and Figure 6 It can be seen that the electromagnetic absorption performance of metamaterials corresponding to the combination of geometric parameters and coding structures obtained by the genetic algorithm is relatively limited. However, the algorithm of this invention, through gradient information to guide the generation of the initial population and the adaptive complementary adjustment mechanism of algorithm parameters, can more finely control the metamaterial structure and obtain a better combination of geometric parameters and coding structures. It shows better performance in broadband impedance matching and multi-resonance mode excitation of metamaterials and significantly expands the absorption bandwidth of metamaterials.

[0085] As shown in Table 3, the genetic algorithm, after 600 evaluations, yielded an optimal fitness value of 0.781 and a relative absorption bandwidth of 101.31% for the electromagnetic wave absorbing encoded metamaterial. The optimization method proposed in this invention, after 540 evaluations, yielded an optimal fitness value of 0.867 and a relative absorption bandwidth of 139.89% for the electromagnetic wave absorbing encoded metamaterial. The optimization method proposed in this invention obtained individuals with higher fitness values ​​using fewer evaluations, and the relative absorption bandwidth of the optimized electromagnetic wave absorbing encoded metamaterial was improved by 38.08% compared to the genetic algorithm, demonstrating the superiority of the proposed optimization method.

[0086] Depend on Figure 3 It can be seen that in the first 20 iterations, the optimization method proposed in this invention can quickly find structures with excellent resonance characteristics using gradient information. In contrast, the genetic algorithm, due to the ineffectiveness of random encoding, requires a larger computational load in the early stages to find a better structure. Therefore, the optimization method proposed in this invention achieves faster fitness improvement and requires less computation in the first 20 iterations. In the iterations from generation 20 to generation 120, the genetic algorithm is prone to destroying excellent structures when the population fitness value is high, leading to local optima problems. Specifically, the convergence curve stagnates for a long time in generations 29 to 69 and 70 to 109. The optimization method proposed in this invention can automatically adjust algorithm parameters according to the population evolution status, reducing the crossover rate and increasing the mutation rate when the population fitness value is high, avoiding destruction of excellent structures and improving the accuracy of structural design. Therefore, it can achieve stable optimization of the metamaterial structure from generation 30 to generation 120, specifically showing a stable increase in the convergence curve and reaching a high level in generation 50. In summary, the optimization method proposed in this invention has a faster convergence speed and stronger optimization ability compared to the genetic algorithm, and the final designed metamaterial has better wave absorption performance.

[0087] Example 2: This invention also provides an optimized design system for broadband electromagnetic absorbing coded metamaterials, such as... Figure 7 As shown, the system includes: an encoding module, an adjustment module, and a control module.

[0088] The encoding module is used to generate mesh-weighted metamaterial structures, which are characterized using an encoding matrix. The fitness value of the encoded metamaterial structure is then calculated. The adjustment module is used to adaptively adjust the weight perturbation ratio and learning rate based on the fitness value of the metamaterial structure and the adaptive gradient descent algorithm to generate the optimal weights. The regulation module is used to generate an initial population based on the optimal weights, and to establish a complementary regulation mechanism between crossover rate and mutation rate through an adaptive genetic algorithm to regulate the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

[0089] It is understood that the optimization design system for broadband electromagnetic absorbing coded metamaterials provided by this invention corresponds to the optimization design method for broadband electromagnetic absorbing coded metamaterials provided in the foregoing embodiments. The relevant technical features of the optimization design system for broadband electromagnetic absorbing coded metamaterials can be referred to the relevant technical features of the optimization design method for broadband electromagnetic absorbing coded metamaterials, and will not be repeated here.

[0090] Another object of the present invention is to provide an electronic device, such as... Figure 8 As shown, it includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor performing the steps of the optimized design method for the broadband electromagnetic wave-absorbing coded metamaterial.

[0091] The optimized design method for the broadband electromagnetic absorbing coded metamaterial includes the following steps: Generate mesh-weighted metamaterial structures, characterize them using an encoding matrix, and calculate the fitness value of the encoded metamaterial structures. Based on the fitness value of the metamaterial structure, the optimal weights are generated by adaptively adjusting the weight perturbation ratio and learning rate using an adaptive gradient descent algorithm. An initial population is generated based on the optimal weights. An adaptive genetic algorithm is used to establish a complementary regulation mechanism between crossover rate and mutation rate to control the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

[0092] A fourth objective of this invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the optimized design method for the broadband electromagnetic wave-absorbing coded metamaterial.

[0093] The optimized design method for the broadband electromagnetic absorbing coded metamaterial includes the following steps: Generate mesh-weighted metamaterial structures, characterize them using an encoding matrix, and calculate the fitness value of the encoded metamaterial structures. Based on the fitness value of the metamaterial structure, the optimal weights are generated by adaptively adjusting the weight perturbation ratio and learning rate using an adaptive gradient descent algorithm. An initial population is generated based on the optimal weights. An adaptive genetic algorithm is used to establish a complementary regulation mechanism between crossover rate and mutation rate to control the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

[0094] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0096] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0097] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An optimization design method of a broadband electromagnetic wave absorbing coded metamaterial, characterized in that, include: Generate mesh-weighted metamaterial structures, characterize them using an encoding matrix, and calculate the fitness value of the encoded metamaterial structures. Based on the fitness value of the metamaterial structure, the optimal weights are generated by adaptively adjusting the weight perturbation ratio and learning rate using an adaptive gradient descent algorithm. An initial population is generated based on the optimal weights. An adaptive genetic algorithm is used to establish a complementary regulation mechanism between crossover rate and mutation rate to control the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

2. The method of claim 1, wherein the optimization design of the broadband electromagnetic wave absorbing coded metamaterial is characterized in that, The method for calculating the fitness value of metamaterial structures is as follows: in, This represents the fitness value of the metamaterial structure. This represents the average absorption rate of the metamaterial within the target frequency band. This represents the proportion of frequency points in which the metamaterial's absorption rate exceeds a preset target value within the target frequency band, out of the total number of frequency points in the entire frequency band. This represents the maximum absorption rate of the metamaterial at the target frequency band. , and These represent the average weighting factor, bandwidth weighting factor, and peak weighting factor, respectively.

3. The optimized design method for broadband electromagnetic absorbing coded metamaterials according to claim 1, characterized in that, The weighted perturbation ratio is: The learning rate is: in, Indicates the weight perturbation ratio. Indicates the learning rate. , , ... They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. The weight perturbation ratio for each fitness value segment. , , ... They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. The learning rate is segmented into fitness values. , , ... , They represent the 1st, 2nd, 3rd, ..., 1st, 2nd, 3rd, ..., 1st. Each fitness value segment boundary.

4. The optimized design method for broadband electromagnetic absorbing coded metamaterials according to claim 1, characterized in that, Methods for generating optimal weights include: Based on the fitness value of the metamaterial structure, the initial weights are positively perturbed by a selected weight perturbation ratio to obtain the first new weights; Calculate the gradient of the difference between the old and new fitness values ​​with respect to the difference between the old and new weights. The gradient guides the update direction of the initial weights, resulting in the second new weights. Based on the second new weights, construct the metamaterial structure and calculate its fitness value. Repeat the above steps until the termination condition is met, and output the final optimal weight.

5. The optimized design method for broadband electromagnetic absorbing coded metamaterials according to claim 4, characterized in that, The first new weight is: in, Indicates the first new weight. This represents the weighted perturbation. Indicates the initial weights. Indicates the weight perturbation ratio; The second new weight is: in, This indicates the second new weight. This represents the gradient of the difference between the old and new fitness values ​​relative to the difference between the old and new weights. This represents the fitness value of the metamaterial structure established based on the first new weight. This represents the fitness value of the metamaterial structure.

6. The optimized design method for broadband electromagnetic absorbing coded metamaterials according to claim 1, characterized in that, Methods for regulating metamaterial structures through complementary regulation mechanisms of crossover rate and variation rate include: The crossover rate, mutation rate, crossover distribution index, and mutation distribution index are adaptively adjusted based on the current population fitness value, while preventing the parameters from exceeding the extreme value boundary. The best individuals are selected based on the elite retention ratio and directly enter the offspring population; A tournament selection mechanism is used to select the parent from the current individual; Perform two-point crossover and bit mutation operations on the encoding matrix of the selected parent individuals, and perform simulated binary crossover and polynomial mutation operations on the real matrix to generate a new offspring population. The fitness value of each individual in the new population is calculated, and through iteration, the optimal solution is output as the optimal topology and optimal parameters of the electromagnetic wave absorbing encoded metamaterial.

7. The optimized design method for broadband electromagnetic absorbing coded metamaterials according to claim 6, characterized in that, Crossover rate: The mutation rate is: The cross-distribution index is: The variation distribution index is: in, Indicates the crossover rate. This indicates the upper limit of the cross rate. This indicates the lower bound of the crossover rate. , , Both represent nonlinear parameters. This represents the average fitness value of the population. The variance of the population fitness values ​​is represented by the following: Indicates the rate of variation. This indicates the upper limit of the mutation rate. This indicates the lower limit of the mutation rate. Indicates the cross-distribution index. This indicates the upper limit of the cross-distribution index. This indicates the lower bound of the cross-distribution index. Indicates the distribution index of variation. This represents the upper limit of the variation distribution index. This indicates the lower bound of the variation distribution index.

8. An optimized design system for a broadband electromagnetic wave-absorbing coded metamaterial, characterized in that, include: The encoding module is used to generate mesh-weighted metamaterial structures, which are characterized using an encoding matrix. The fitness value of the encoded metamaterial structure is then calculated. The adjustment module is used to adaptively adjust the weight perturbation ratio and learning rate based on the fitness value of the metamaterial structure and the adaptive gradient descent algorithm to generate the optimal weights. The regulation module is used to generate an initial population based on the optimal weights, and to establish a complementary regulation mechanism between crossover rate and mutation rate through an adaptive genetic algorithm to regulate the metamaterial structure and obtain the optimal topology and parameters of the electromagnetic wave absorbing encoded metamaterial.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images