A frequency selective surface design method and system based on adaptive grey wolf algorithm

Automatically generate units in FSS design through the adaptive gray wolf algorithm, which solves the problems of low efficiency and dependence on experience in traditional design methods, realizes automated optimization and high-precision FSS design, and improves design efficiency and accuracy.

CN114357862BActive Publication Date: 2024-11-29SOUTH CHINA UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111532868.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-15
Publication Date
2024-11-29
Estimated Expiration
2041-12-15

AI Technical Summary

Technical Problem

The traditional frequency selection surface (FSS) design method has the problem of low design efficiency, relying on experience and low efficiency in complex electromagnetic environments, and the optimization model is relatively limited.

Method used

The design method based on the adaptive gray wolf algorithm is adopted, and the target FSS unit is automatically generated without the need for an initial configuration. Through real-number coding, improved strategies such as reverse movement and nonlinear step size, and local search enhancement, the automated optimization process is realized.

Benefits of technology

It realizes automation of FSS design, improves design efficiency and accuracy, gets rid of the dependence on experience, and obtains faster convergence speed and higher solution accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114357862B_ABST
    Figure CN114357862B_ABST
Patent Text Reader

Abstract

The present invention discloses a frequency selective surface design method and system based on an adaptive grey wolf algorithm. The method aims at the frequency selective surface structure optimization problem, takes the filtering effect of the frequency selective surface at the central working frequency as the goal, establishes a node-based frequency selective surface unit optimization model, then adopts the adaptive grey wolf algorithm in combination with the model to solve the problem, and develops a frequency selective surface design system based on the method. Compared with the traditional FSS optimization method, the present invention has a larger optimization range and overcomes the defects of low efficiency, poor quality and reliance on experience of the traditional FSS design. The present invention realizes the automatic design of the FSS unit from design requirements to finished products, thereby improving the design efficiency of the FSS.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of electromagnetic communication and microwave technology, and in particular to a frequency selective surface design method and system based on an adaptive grey wolf algorithm. Background Art

[0002] Frequency selective surface (FSS) is a two-dimensional electromagnetic metamaterial with filtering capabilities, which has the advantages of thinness, convenient design and easy processing. The working performance of FSS is related to factors such as unit structure, arrangement and dielectric substrate. FSS with electromagnetic shielding performance can be obtained through structural design to solve the problem of mutual interference between communication equipment.

[0003] The traditional FSS design method has high requirements for designers and is somewhat blind, which makes the FSS design process inefficient. In order to improve the design efficiency of the FSS structure, the FSS structure can be digitally designed with the help of computers and simulation software. Introducing intelligent optimization algorithms into the design can get rid of the dependence on design experience and improve the design quality of FSS. At present, FSS optimization models are mainly divided into two types. One is the optimization of key dimensions based on the initial configuration, and the other is the topological optimization of the unit shape composed of blocks. Both methods have certain limitations. The first model relies on experience to provide the initial configuration, and the second model has low efficiency and difficulty in solving large-scale design variables in a complex electromagnetic environment. Therefore, it is of great engineering significance to propose a new FSS optimization model and combine it with an intelligent optimization algorithm. Summary of the invention

[0004] In order to overcome the defects and shortcomings of the prior art, the present invention provides a frequency selective surface design method based on an adaptive gray wolf algorithm, which does not require a definite initial configuration and can automatically generate a target FSS unit during the optimization process. This design method has a larger search design domain than the traditional method and ensures that the unit shape is always coherent and complete during the change process; the joint simulation method is used in the optimization process to achieve the automatic completion of the entire optimization process without manual intervention.

[0005] A second object of the present invention is to provide a frequency selective surface design system based on an adaptive grey wolf algorithm.

[0006] A third object of the present invention is to provide a computer-readable storage medium.

[0007] A fourth object of the present invention is to provide a computing device.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A frequency selective surface design method for an adaptive grey wolf algorithm comprises the following steps:

[0010] Analyze the shape characteristics of the FSS unit, and establish an FSS optimization model based on the shape node positions of the FSS unit, wherein the FSS optimization model presets the overall size range and basic deformation rules of the FSS unit;

[0011] The real number coding method is used to store the position information of each node in the FSS optimization model as an individual in the gray wolf algorithm;

[0012] Improved strategies are added to the gray wolf algorithm to form an adaptive gray wolf algorithm. A dispersion strategy is added to the leader of the initial population. Reverse movement and nonlinear step size are introduced when the individual positions of gray wolves are updated, and local search is enhanced in the convergence stage.

[0013] The node information output by the adaptive grey wolf algorithm is simulated and analyzed, and the FSS is set as an infinite periodic plane to obtain the FSS transmission coefficient near the central working frequency;

[0014] The obtained simulation results are returned to the adaptive gray wolf algorithm, the objective function is calculated, and the next iteration is started according to the convergence rule of the adaptive gray wolf algorithm until the algorithm ends;

[0015] Output the entity and obtain the optimized FSS model.

[0016] As a preferred technical solution, the FSS optimization model is established based on the shape node position of the FSS unit, and the specific steps include:

[0017] Analyze the basic structure of the unit, mark the nodes on the outline, and number each node;

[0018] A polar coordinate system is established with the centroid of the FSS unit as the midpoint, and the relative position information of each node is recorded. The position information of all nodes is stored in a real array;

[0019] Preset constraint conditions, which are used to constrain the active area of ​​each node. After adding the constraint, the adaptive gray wolf algorithm is called to optimize the unit shape and update the node position information after each iteration;

[0020] The FSS shape is retrieved based on the node number and position information, and the current node information is stored in the real array for the next iteration.

[0021] As a preferred technical solution, a dispersion strategy is added to the leader of the first-generation population, specifically a leadership dispersion strategy using Euclidean distance as a judgment standard. When the calculated Euclidean distance is less than a limit value, the location information of the gray wolf individual is updated.

[0022] As a preferred technical solution, the calculation formula of the Euclidean distance is expressed as:

[0023]

[0024] Among them, n represents the total number of variables contained in the individual, α, β and δ represent the three gray wolf individuals with the highest fitness in the gray wolf algorithm, and x αi represents the value of the i-th variable of individual α, x pi Represents the value of the i-th variable of individual β or δ.

[0025] As a preferred technical solution, when the calculated Euclidean distance is less than the limit value, the location information of the gray wolf individual is updated. The specific location update method is expressed as follows:

[0026] X(t+1)=U+LX(t) / 2

[0027] Among them, X represents the location information of the gray wolf individual, U represents the upper limit, L represents the lower limit, t represents the generation of the current individual, and t+1 represents the next generation.

[0028] As a preferred technical solution, the reverse movement and nonlinear step length are introduced when updating the position of the individual gray wolf. The specific calculation steps include:

[0029] When an individual updates its position, it moves in the opposite direction of the current potential prey with a set probability. The calculation formula is:

[0030]

[0031] P=0.5+P A

[0032] Among them, α, β and δ represent the three gray wolf individuals with the highest fitness in the gray wolf algorithm, X1, X2, X3 correspond to the displacement of the current individual to α, β and δ respectively, r3 represents a random number, P represents the probability of reverse movement, P A represents the adaptive coefficient;

[0033] The step size reduction process is transformed into a nonlinear process, and the calculation formula is:

[0034]

[0035] A s =2a s *r1-a s

[0036] X1=X α -A s *(D α )

[0037] Among them, a s Indicates the replacement of the gray wolf individual α with the highest fitness in the original gray wolf algorithm, A srepresents the displacement coefficient of the individual moving toward the gray wolf individual α, X α is the position of the gray wolf individual α, D α represents the distance from the individual to the gray wolf individual α, X1 represents the displacement of the individual to the gray wolf individual α, and q and r1 are random numbers.

[0038] As a preferred technical solution, the local search is enhanced in the convergence phase, and the specific steps include: adding an independent iterative calculation to all individuals in the convergence phase so that the individuals have two iterative results, and then forming the next population with the better results.

[0039] In order to achieve the above second purpose, the present invention adopts the following technical solutions:

[0040] A frequency selective surface design system based on an adaptive gray wolf algorithm comprises: a FSS optimization model building module, a position information storage module, a gray wolf algorithm optimization module, a simulation analysis module, a loop iteration module and a physical output module;

[0041] The FSS optimization model building module is used to establish an FSS optimization model, and the FSS optimization model is established based on the shape node position of the FSS unit. The FSS optimization model presets the overall size range and deformation basic rules of the FSS unit;

[0042] The position information storage module is used to store the position information of each node in the FSS optimization model in a real number coding manner as an individual in the gray wolf algorithm;

[0043] The gray wolf algorithm optimization module is used to add improved strategies to the gray wolf algorithm to form an adaptive gray wolf algorithm, add a dispersion strategy to the leader of the initial population, introduce reverse movement and nonlinear step size when the individual gray wolf positions are updated, and enhance local search in the convergence stage;

[0044] The simulation analysis module is used to simulate and analyze the node information output by the adaptive grey wolf algorithm, set the FSS to an infinite periodic plane, and obtain the FSS transmission coefficient near the central working frequency;

[0045] The loop iteration module is used to return the obtained simulation results to the adaptive gray wolf algorithm, calculate the objective function, and start the next iteration according to the convergence rule of the adaptive gray wolf algorithm until the algorithm ends;

[0046] The entity output module is used to output the entity to obtain an optimized FSS model.

[0047] In order to achieve the third objective, the present invention adopts the following technical solutions:

[0048] A computer-readable storage medium stores a program, and when the program is executed by a processor, the frequency selective surface design method based on the adaptive grey wolf algorithm as described above is implemented.

[0049] In order to achieve the fourth objective, the present invention adopts the following technical solutions:

[0050] A computing device comprises a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the frequency selective surface design method based on the adaptive grey wolf algorithm as described above is implemented.

[0051] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0052] (1) The present invention automates the design optimization process of the FSS, gets rid of the disadvantage of the traditional method of relying on experience, and can directly obtain the final FSS unit configuration according to the filtering performance requirements at the center operating frequency.

[0053] (2) Compared with the original grey wolf algorithm, the adaptive grey wolf algorithm obtained by the present invention has a faster convergence speed and higher solution accuracy, and is more in line with the optimization process of FSS. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flow chart of a frequency selective surface design method based on an adaptive grey wolf algorithm of the present invention;

[0055] Figure 2 This is a schematic diagram of the FSS optimization model of the present invention;

[0056] Figure 3 A schematic diagram of searching and updating the Grey Wolf Algorithm of the present invention;

[0057] Figure 4 Schematic diagram of three improved strategies in the adaptive grey wolf algorithm of the present invention;

[0058] Figure 5 This is a schematic diagram of the unoptimized double-layer regular hexadecagonal FSS unit structure of the present invention;

[0059] Figure 6 Schematic diagram of the double-layer FSS unit structure after optimization by the adaptive grey wolf algorithm of the present invention;

[0060] Figure 7 This is a schematic diagram of the final optimized FSS model of the present invention;

[0061] Figure 8 This is a physical sample of the FSS optimization result of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0063] Example 1

[0064] like Figure 1 As shown, this embodiment provides a frequency selective surface design method based on an adaptive gray wolf algorithm, and takes the optimization result of a double-layer sixteen-sided annular FSS unit as an example for explanation, but is not limited to the optimization of the annular FSS unit;

[0065] The specific steps include:

[0066] S1, such as Figure 2 As shown, the shape characteristics of the FSS unit are analyzed, and an optimization model is established based on the shape node position of the FSS unit. The model stipulates the overall size range and basic deformation rules of the FSS unit to ensure that the FSS unit always meets the requirements of filtering performance during the deformation process;

[0067] In step S1, the node FSS optimization model is specifically as follows:

[0068] 1) First, analyze the basic structure of the unit, mark the nodes on the outline, and number each node in a counterclockwise order. Since most FSS units are centrosymmetric structures, a polar coordinate system is established with its centroid as the midpoint to record the relative position information of each node. The information of all nodes is stored in a real array, thereby realizing the parametric expression of the unit shape. The data in the array is changed by the gray wolf algorithm to achieve the optimization of the FSS unit shape.

[0069] 2) In order to maintain the stability of the unit shape during the shape optimization process and keep the overall structure within the size range determined by the original FSS unit, the active area of ​​each node must be constrained. To ensure that the nodes meet the conditions for forming the unit outline, the variables are constrained to avoid the following situations: node agglomeration, node line overlap or node line intersection. After adding constraints, the adaptive gray wolf algorithm is called to optimize the unit shape, and the node position information will be updated after each iteration.

[0070] 3) The FSS shape is retrieved based on the node number and position information, and the current node information will be retained in the real array for the next iteration.

[0071] S2, selecting a real number encoding method to store the position information of each node in the FSS optimization model of step S1, the real number encoding includes the position information of each node as an individual in the gray wolf algorithm;

[0072] S3, such as Figure 3 , Figure 4 As shown, in order to achieve better optimization effect, three improved strategies are added to the gray wolf algorithm, and a dispersion strategy is added to the leader of the initial population; reverse movement and nonlinear step length are introduced when the individual positions of the gray wolves are updated to increase randomness; but in the final convergence stage, local search is enhanced to improve the algorithm accuracy. In this embodiment, the improved gray wolf algorithm is called the adaptive gray wolf algorithm;

[0073] In step S3, the adaptive grey wolf algorithm includes three improved strategies, specifically:

[0074] 1) At the beginning of the algorithm, a leadership dispersion strategy based on the Euclidean distance is added. The leaders in the gray wolf algorithm are the three gray wolf individuals with the highest fitness, represented by α, β, and δ. Specifically, after randomly obtaining the initial population, the Euclidean distance E of α, β, and δ is calculated using the following formula: p .

[0075]

[0076] Where n is the total number of variables included in the individual, x αi represents the value of the i-th variable of individual α, x pi Represents the value of the i-th variable of individual β or δ. Euclidean distance E p Used to judge whether β and δ are too close to α. p When it is less than the limit value, the leader dispersion strategy is used.

[0077] The limit value is determined by the specific optimization problem and is denoted by E L , generally one tenth of the maximum Euclidean distance formed by the upper and lower limits can be used as the limit value, as shown in the following formula;

[0078]

[0079] If the Euclidean distance is less than the limit value in a certain calculation, the individual uses the following formula to determine the moving direction, where U is the upper limit, L represents the lower limit, X is the location information of the gray wolf individual, t represents the generation of the current individual, and t+1 represents the next generation.

[0080] X(t+1)=U+LX(t) / 2

[0081] n is the total number of variables contained in the individual, U i represents the upper limit of the i-th variable, L i Represents the lower limit of the i-th variable.

[0082] The leader dispersion strategy is only applied in the initial population and is not used in subsequent iterations of the algorithm.

[0083] 2) Add a reverse movement strategy to the improved algorithm. The purpose of adding a reverse movement strategy is to provide control capabilities to jump out of the leading wolf in the gray wolf algorithm during the algorithm convergence process. The specific steps are shown in the following formula. When an individual updates its position, it has a probability of moving in the opposite direction of the current potential prey.

[0084]

[0085] P=0.5+P A

[0086] X1, X2, and X3 correspond to the displacements of the current individual to α, β, and δ, respectively, and r3 is a random number between 0 and 1.

[0087] The probability of reverse movement is adjusted by the control optimization situation, P is 0.5 plus the adaptive coefficient P A .P A The initial value is 0. If α stagnates during the exploration phase, it is increased to 0.2, which is conducive to the algorithm to escape from the local optimal solution. On the other hand, the convergence process of the algorithm is driven by a, which decreases linearly from 2 to 0, and the algorithm ends here. However, when dealing with practical problems, a larger step size is required in the early stage of the algorithm to increase exploration and escape from the local optimal solution, and a smaller step size is required in the later stage to increase development accuracy.

[0088] The following formula introduces a randomly changing function in the process of decreasing a, where q is a random number between -10 and 10. s Replace a in the original gray wolf algorithm, so that the step size reduction process is transformed into a nonlinear process. Taking the displacement X1 of the individual moving to α as an example, calculate the displacement coefficient A of the individual moving to α s When using a s , randomness increases, which meets the needs of algorithm improvement. α is the position of α, D α represents the distance from the individual to α, and r1 is a random number between 0 and 1.

[0089]

[0090] A s =2a s *r1-a s

[0091] X1=X α -A s *(D α )

[0092] 3) An enhanced local search strategy is added to the gray wolf algorithm. The specific method is as follows: in order to achieve better optimization results, when the algorithm enters the final convergence stage, an independent iterative calculation is added to all individuals, so that the individuals have two iterative results, and then the better results are used to form the next population, thereby improving the final optimization effect;

[0093] S4. The node information output by the improved grey wolf algorithm is imported into the CAE software for simulation analysis according to the rules set by the FSS optimization model. During the simulation calculation, the FSS is set to an infinite periodic plane. The software simulation results obtain the electromagnetic wave transmission coefficient S from port 1 to port 2 near the central operating frequency. 21 ;

[0094] In step S4, the function used to evaluate the filtering performance of each FSS unit is specifically: the gray wolf algorithm obtains a series of transmission coefficients S near the central operating frequency 21 If the transmission coefficient shows a downward trend near the center point, it reflects that FSS has electromagnetic shielding ability in this frequency band; otherwise, it indicates that FSS has selective permeability in this frequency band.

[0095] S5. The obtained simulation results S 21 Return to the adaptive gray wolf algorithm, calculate the objective function, and then start the next iteration according to the convergence rule of the adaptive gray wolf algorithm until the algorithm ends;

[0096] S6, such as Figure 6 As shown, combined with Figure 5 As shown, the entity is output and finally an optimized FSS model is obtained.

[0097] like Figure 7 , Figure 8 As shown, after the FSS model is obtained, it can be combined into a periodic structure plane FSS in the software and the sample can be manufactured.

[0098] Example 2

[0099] This embodiment provides a frequency selective surface design system based on an adaptive gray wolf algorithm, comprising: a FSS optimization model construction module, a position information storage module, a gray wolf algorithm optimization module, a simulation analysis module, a loop iteration module and a physical output module;

[0100] In this embodiment, the FSS optimization model building module is used to establish the FSS optimization model, and the FSS optimization model is established based on the shape node position of the FSS unit. The FSS optimization model presets the overall size range and deformation basic rules of the FSS unit;

[0101] In this embodiment, the location information storage module is used to store the location information of each node in the FSS optimization model in a real number coding manner as an individual in the gray wolf algorithm;

[0102] In this embodiment, the gray wolf algorithm optimization module is used to add improved strategies to the gray wolf algorithm to form an adaptive gray wolf algorithm, add a dispersion strategy to the leader of the initial population, introduce reverse movement and nonlinear step size when the individual gray wolf positions are updated, and enhance local search in the convergence stage;

[0103] In this embodiment, the simulation analysis module is used to simulate and analyze the node information output by the adaptive grey wolf algorithm, set the FSS to an infinite periodic plane, and obtain the FSS transmission coefficient near the central working frequency;

[0104] In this embodiment, the loop iteration module is used to return the obtained simulation results to the adaptive gray wolf algorithm, calculate the objective function, and start the next iteration according to the convergence rule of the adaptive gray wolf algorithm until the algorithm ends;

[0105] In this embodiment, the entity output module is used to output the entity to obtain an optimized FSS model.

[0106] Example 3

[0107] This embodiment provides a storage medium, which may be a storage medium such as ROM, RAM, disk, or CD. The storage medium stores one or more programs. When the program is executed by a processor, the frequency selective surface design method based on the adaptive gray wolf algorithm of embodiment 1 is implemented.

[0108] Example 4

[0109] This embodiment provides a computing device, which can be a desktop computer, a laptop computer, a smart phone, a PDA handheld terminal, a tablet computer or other terminal device with a display function. The computing device includes a processor and a memory, and the memory stores one or more programs. When the processor executes the program stored in the memory, the frequency selective surface design method based on the adaptive gray wolf algorithm of Example 1 is implemented.

[0110] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A frequency selective surface design method based on an adaptive grey wolf algorithm, characterized in that: The steps include: Analyze the shape characteristics of the FSS unit, and establish an FSS optimization model based on the shape node positions of the FSS unit, wherein the FSS optimization model presets the overall size range and basic deformation rules of the FSS unit; The real number coding method is used to store the position information of each node in the FSS optimization model as an individual in the gray wolf algorithm; Improved strategies are added to the gray wolf algorithm to form an adaptive gray wolf algorithm. A dispersion strategy is added to the leader of the initial population. Reverse movement and nonlinear step size are introduced when the individual positions of gray wolves are updated, and local search is enhanced in the convergence stage. The leader of the first generation population is added with a dispersal strategy, specifically adding a leader dispersal strategy with Euclidean distance as a judgment standard, and when the calculated Euclidean distance is less than a limit value, updating the location information of the gray wolf individual; When the calculated Euclidean distance is less than the limit value, the location information of the gray wolf individual is updated. The specific location update method is expressed as follows: X(t+1)=U+LX(t) / 2 Among them, X represents the location information of the gray wolf individual, U represents the upper limit, L represents the lower limit, t represents the generation of the current individual, and t+1 represents the next generation; The calculation formula of the Euclidean distance is expressed as: Among them, n represents the total number of variables contained in the individual, α, β and δ represent the three gray wolf individuals with the highest fitness in the gray wolf algorithm, and x αi represents the value of the i-th variable of individual α, x pi represents the value of the i-th variable of the β or δ individual; The reverse movement and nonlinear step length are introduced when updating the position of the individual gray wolf. The specific calculation steps include: When an individual updates its position, it moves in the opposite direction of the current potential prey with a set probability. The calculation formula is: P=0.5+P A Among them, α, β and δ represent the three gray wolf individuals with the highest fitness in the gray wolf algorithm, X1, X2, X3 correspond to the displacement of the current individual to α, β and δ respectively, r3 represents a random number, P represents the probability of reverse movement, P A represents the adaptive coefficient; The step size reduction process is transformed into a nonlinear process, and the calculation formula is: A s =2a s *r1-a s X=X α -A s *(D α ) Among them, a s Indicates the replacement of the gray wolf individual α with the highest fitness in the original gray wolf algorithm, A s represents the displacement coefficient of the individual moving toward the gray wolf individual α, X α is the position of the gray wolf individual α, D α represents the distance from the individual to the gray wolf individual α, X1 represents the displacement of the individual to the gray wolf individual α, and q and r1 are random numbers; The node information output by the adaptive grey wolf algorithm is simulated and analyzed, and the FSS is set as an infinite periodic plane to obtain the FSS transmission coefficient near the central working frequency; The obtained simulation results are returned to the adaptive gray wolf algorithm, the objective function is calculated, and the next iteration is started according to the convergence rule of the adaptive gray wolf algorithm until the algorithm ends; Output the entity and obtain the optimized FSS model.

2. The frequency selective surface design method based on the adaptive grey wolf algorithm according to claim 1 is characterized in that: The FSS optimization model is established based on the shape node position of the FSS unit, and the specific steps include: Analyze the basic structure of the unit, mark the nodes on the outline, and number each node; A polar coordinate system is established with the centroid of the FSS unit as the midpoint, and the relative position information of each node is recorded. The position information of all nodes is stored in a real array; Preset constraint conditions, which are used to constrain the active area of ​​each node. After adding the constraint, the adaptive gray wolf algorithm is called to optimize the unit shape and update the node position information after each iteration; The FSS shape is retrieved based on the node number and position information, and the current node information is stored in the real array for the next iteration.

3. The frequency selective surface design method based on the adaptive grey wolf algorithm according to claim 1 is characterized in that: The specific steps of enhancing local search in the convergence stage include: adding an independent iterative calculation to all individuals in the convergence stage, so that the individuals have two iterative results, and then forming the next population with the better results.

4. A frequency selective surface design system based on an adaptive grey wolf algorithm, characterized in that: Used to implement the frequency selective surface design method based on the adaptive gray wolf algorithm as described in any one of claims 1 to 3, the system comprises: an FSS optimization model construction module, a position information storage module, a gray wolf algorithm optimization module, a simulation analysis module, a loop iteration module and a physical output module; The FSS optimization model building module is used to establish an FSS optimization model, and the FSS optimization model is established based on the shape node position of the FSS unit. The FSS optimization model presets the overall size range and deformation basic rules of the FSS unit; The position information storage module is used to store the position information of each node in the FSS optimization model in a real number coding manner as an individual in the gray wolf algorithm; The gray wolf algorithm optimization module is used to add improved strategies to the gray wolf algorithm to form an adaptive gray wolf algorithm, add a dispersion strategy to the leader of the initial population, introduce reverse movement and nonlinear step size when the individual gray wolf positions are updated, and enhance local search in the convergence stage; The simulation analysis module is used to simulate and analyze the node information output by the adaptive grey wolf algorithm, set the FSS to an infinite periodic plane, and obtain the FSS transmission coefficient near the central working frequency; The loop iteration module is used to return the obtained simulation results to the adaptive gray wolf algorithm, calculate the objective function, and start the next iteration according to the convergence rule of the adaptive gray wolf algorithm until the algorithm ends; The entity output module is used to output the entity to obtain an optimized FSS model.

5. A computer-readable storage medium storing a program, characterized in that: When the program is executed by a processor, a frequency selective surface design method based on an adaptive grey wolf algorithm as described in any one of claims 1 to 3 is implemented.

6. A computing device comprising a processor and a memory for storing a program executable by the processor, characterized in that: When the processor executes the program stored in the memory, the frequency selective surface design method based on the adaptive grey wolf algorithm as described in any one of claims 1 to 3 is implemented.

Citation Information

Patent Citations

  • BP neural network optimization method based on improved grey wolf algorithm

    CN111723945A

  • Improved multi-objective grey wolf optimization algorithm

    CN112488283A