Multilayer wave-absorbing structure design method based on equivalent circuit and genetic optimization algorithm

Abstract

The invention provides a multilayer wave-absorbing structure design method based on an equivalent circuit and a genetic optimization algorithm, and the method comprises the steps: 1, initializing EC-GA hyper-parameters and a physical parameter range [Varmin, Varmax] of a multilayer wave-absorbing structure, and generating a parameter group composed of random binary codes; step 2, decoding the parameter group according to [Varm, Varmax]; step 3, calculating a reflection coefficient S11, a loss function value Loss and individual fitness Fitness of the wave-absorbing structure according to an equivalent circuit and a transmission line theory; 4, comparing the minimum loss value in the current round of group with the minimum value in the previous round, and keeping the minimum loss value as minLoss; 5, selecting individuals to form a new group according to the fitness value by adopting a roulette selection method; step 6, performing crossover and mutation operation on the selected population; and 7, repeating the steps 2-6 until the maximum number of iterations is reached. The physical parameters of the multi-layer wave absorbing structure can be directly obtained from the target reflection coefficient in theory, reverse design is achieved, and optimization efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of metamaterial absorbing structures, and in particular to a multilayer absorbing structure design method based on an equivalent circuit and genetic optimization algorithm (EC-GA). Background Art

[0002] Broadband absorbing structures have broad application prospects in the field of radar stealth because of their ability to dissipate electromagnetic wave energy in a wide frequency range. However, in actual design, the influence of absorption characteristics often involves many factors, which makes the design and optimization of broadband absorbing structures very complicated. Therefore, it is necessary to develop a simple and efficient optimization algorithm to assist the design of broadband absorbing structures.

[0003] In the design of broadband absorbing structures, the most commonly used method is to stack or regularly arrange lossy surfaces based on impedance films or lumped resistance elements in multiple layers. Impedance films have been widely used in the design of broadband absorbing structures because they have good ohmic loss characteristics and can be printed into a variety of shapes to meet different electromagnetic requirements. Traditional absorbing structure design schemes rely on the experience of technicians and usually take a lot of time to adjust various physical parameters. Especially in the field of ultra-wideband absorbing, as the absorbing frequency range is widened, the number of layers of the absorbing structure will increase, and the parameters that need to be designed will also increase, which undoubtedly increases the complexity of the technicians' work. With the development of artificial intelligence technology, a variety of optimization algorithms have been widely used in optimal solution problems. Among them, genetic algorithms can efficiently and stably handle multi-parameter optimization problems by imitating the process of species evolution. However, in order to apply genetic algorithms in the design process of absorbing structures, it is necessary to establish a complete absorbing theoretical calculation system, which requires the establishment of a closed relationship between the lumped parameters of the equivalent circuit and the structural geometric parameters.

[0004] Previous studies have shown that the lumped resistance value is related to the surface impedance and area of ​​the impedance film, and the inductance and capacitance values ​​are related to the geometry and size of the impedance film and the effective dielectric constant of the surrounding medium. The dielectric environment that affects the effective dielectric constant mainly includes the type, quantity, thickness and dielectric constant of the surrounding medium. However, many studies have ignored the influence of the surrounding dielectric environment on the effective dielectric constant, resulting in a large deviation between the theoretical calculation results and the simulation results. Summary of the invention

[0005] In view of this, the present application provides a multi-layer absorbing structure design method based on an equivalent circuit and a genetic optimization algorithm, which can quickly and efficiently optimize the optimal physical parameter combination of the multi-layer absorbing structure according to the target reflection coefficient.

[0006] To achieve the above objectives, the technical solutions adopted in this application are as follows:

[0007] The design method of multilayer absorbing structure based on equivalent circuit and genetic optimization algorithm includes:

[0008] Step 1, initialize the genetic algorithm GA hyperparameters and the physical parameters of the absorbing structure, and generate an initial encoding parameter array consisting of random binary codes of specifications (paranum, popsize, DNAsize). Where paranum is the number of optimization objects, popsize is the population size, and DNAsize is the length of the DNA sequence used for encoding. The genetic algorithm hyperparameters include crossover probability, mutation probability, gene sequence length, number of iterations, and loss function weights. The physical parameters of the absorbing structure include the relative dielectric constant ε of the intermediate dielectric layer. r , the relative dielectric constant ε of the substrate dielectric layer s , the number of layers m and the range of optimization target parameters [Var max ,Var min ];

[0009] Step 2: According to the parameter range [Var max ,Var min ], use formula (1) to decode the parameters:

[0010]

[0011] ennum represents the binary coded parameters before decoding, and denum represents the parameters after decoding.

[0012] Step 3: The equivalent circuit of the multilayer absorbing structure is as follows: Figure 2 As shown, the square ring resistor film is equivalent to a series circuit of resistance, inductance and capacitance. The dielectric layer is represented by its characteristic impedance, and the metal bottom plate is equivalent to a short-circuit line. The reflection coefficient S of the absorbing structure is calculated based on the equivalent circuit and transmission line theory. 11 , loss function value Loss and individual fitness Fitness;

[0013] Step 4: Compare the minimum loss value in this round of the group with the minimum value in the previous round, and retain the minimum loss value as min_Loss; for the first round of iteration, the minimum loss value of the group is used as min_Loss.

[0014] Step 5: Use the roulette wheel selection method to select individuals to form a new group based on their fitness values. The higher the individual fitness, the greater the probability of being selected.

[0015] Step 6: Perform crossover and mutation operations on the selected population. Crossover refers to the crossover recombination of the parent DNA sequence under the control of the crossover probability; mutation refers to the inversion of the binary coding of certain gene bits in the parent DNA sequence under the control of the mutation probability, that is, 0 becomes 1 or 1 becomes 0, thereby forming a new offspring population;

[0016] Step 7: Repeat steps 2 to 6 until the maximum number of iterations is reached.

[0017] Step 8: Output the decoded optimal parameter combination X_best and its corresponding reflection coefficient S 11 and loss function value min_Loss; where X_best refers to the target optimization parameters, including the period p of the absorbing unit, the thickness h of the intermediate dielectric layer r,1 ,h r,2 ...h r,m , the thickness of the substrate dielectric layer h s,1 ,h s,2 ...h s,m 、ITO impedance film square resistance R1, R2...R m 、Square ring side length d1,d2...d m and the widths of the square rings w1, w2...w m , m represents the number of layers of the absorbing structure.

[0018] Furthermore, in step 3, the reflection coefficient S of the absorbing structure is calculated according to the equivalent circuit and transmission line theory. 11 The specific process of loss function value Loss and individual fitness Fitness is as follows:

[0019] Step 3.1, calculate the characteristic impedance Z of the medium around the n-th square ring ITO impedance film g,n and the effective dielectric constant ε eff,n , n=1,2,...,m:

[0020]

[0021] Where ε av,n Represents the average dielectric constant of the surrounding medium:

[0022]

[0023] In the formula, when the impedance film is in the shape of a square ring, the coefficient α = 18, and ε rh,n The expression is as follows,

[0024]

[0025] Among them, h s,n and h r,nare the thickness of the nth layer substrate medium and the intermediate medium respectively.

[0026] The surrounding medium of the last layer of ITO metal film is the substrate medium and free space air, and its average dielectric constant is calculated as follows:

[0027]

[0028] Step 3.2: For each frequency point, calculate the equivalent lumped resistance R of the nth layer of square ring ITO impedance film according to the physical parameters of the nth layer after decoding. n , equivalent capacitance C n and equivalent inductance L n , n=1,2,...,m:

[0029]

[0030]

[0031] In the formula,

[0032]

[0033] In formulas (7)-(12), p represents the unit period of the absorbing structure, d n represents the side length of the ITO impedance film square ring, w n Indicates the width of the square ring; represents the angular frequency, ε eff,n Represents the effective dielectric constant of the surrounding medium, which has an impact on the equivalent capacitance, Z g,n It represents the characteristic impedance of the medium around the square ring of ITO impedance film; F(p,x,λ) and G(p,x,λ) are functions related to the geometric parameters of the impedance film and the wavelength λ, and α and β are two intermediate variables.

[0034] Step 3.3: The input impedance Z of the last layer of the absorbing structure equivalent circuit is in,m And the free space characteristic impedance Z0 to obtain the reflection coefficient Γ:

[0035]

[0036] Step 3.4, calculate the loss value Loss, the weighted mean square error between the reflection coefficient obtained by the equivalent circuit and the target reflection coefficient is used as the value of the loss function, and the weighted mean square error calculation expression is as follows:

[0037]

[0038] R i (f j ) and G i (f j) represent the reflection coefficient and target reflection coefficient calculated based on the transmission line theory, respectively. i (f j )≤G i (f j ), the penalty function E i (f j )=0, otherwise the penalty function is the absolute value of the difference between the two. i represents the weighted order, and j represents the frequency point order in the i-th weighted area.

[0039] Step 3.5, calculate the individual fitness, which is an exponential function of the loss function, and its expression is as follows:

[0040]

[0041] Furthermore, the step 3.3 calculates the input impedance Z of the last layer of the equivalent circuit of the absorbing structure in,m Specifically include:

[0042] Step 3.3.1. Calculate the characteristic impedance and propagation constant of the intermediate dielectric layer and the substrate dielectric layer for each frequency point:

[0043]

[0044] Among them, ε s,n and ε r,n are the dielectric constants of the nth layer substrate medium and the intermediate medium respectively. Secondly, calculate the equivalent impedance of the nth layer ITO impedance film:

[0045] Z ec,n =R n +jωL n +1 / (jωC n ) (20)

[0046] ω=2πf (21)

[0047] Step 3.3.2. Calculate Z' n , which is related to the characteristic impedance Z of the intermediate medium r,n , propagation constant β r,n , thickness h r,n and the input impedance Z of the previous layer in,n-1 Since the metal bottom plate is equivalent to a short-circuit impedance of 0, the input impedance of the previous layer of the first layer is Z in,0 =0:

[0048]

[0049] Step 3.3.3. Calculate Z n , which is related to the characteristic impedance Z of the substrate medium s,n , propagation constant β s,n, thickness h s,n and Z' n related:

[0050]

[0051] Step 3.3.4. Calculate the input impedance Z of the nth layer in,n , which is equal to Z n and Z ec,n in parallel:

[0052] Z in,n =Z n ||Z ec,n (twenty four)

[0053] Step 3.3.5: Repeat steps 3.3.2 to 3.3.4 to calculate the input impedance Z of the last layer. in,m .

[0054] Furthermore, the step 5 specifically includes:

[0055] Step 5.1, calculate the probability of each individual being inherited into the next generation, N is the population size, f(x i ) represents the fitness of the i-th individual:

[0056]

[0057] Step 5.2: Calculate the cumulative probability q of each individual i :

[0058]

[0059] Step 5.3, generate a random number r in the interval [0, 1], when q i-1 ≤r≤q i When , the i-th individual is selected;

[0060] Step 5.4: Repeat steps 5.1 to 5.3 N times, where N is equal to the population size popsize.

[0061] Furthermore, each layer of the absorbing structure is composed of a plurality of absorbing units arranged in the same period, and each absorbing unit includes a metal base plate and n layers of absorbing assemblies from bottom to top; the absorbing assembly includes a material medium layer, an air layer and a loss layer from bottom to top.

[0062] Furthermore, the loss layer is an ITO impedance film in a square ring shape.

[0063] Furthermore, the relative magnetic permeability μ of the dielectric layer is 1, and the relative dielectric constant ε is ≤15.

[0064] The beneficial effects of this application are:

[0065] 1. In theory, the present application can directly obtain the physical parameters of the multi-layer absorbing structure from the target reflection coefficient, realize reverse design, and improve optimization efficiency;

[0066] 2. The optimized object has a simple structure and is easy to manufacture. The optimized result has the advantages of ultra-wide bandwidth, low profile and small period size. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 It is a flow chart of the multi-layer absorbing structure design method of the present application;

[0069] Figure 2 It is a schematic diagram of the multilayer absorbing structure of the present application and its equivalent circuit;

[0070] Figure 3 This is a diagram of a multi-layer wave absorbing structure of an embodiment of the present application;

[0071] Figure 4 is a comparison chart of the simulation and theoretical calculation results of the reflection coefficient of the embodiment of the present application;

[0072] Figure 5 is a comparison chart of the simulation and theoretical calculation results of the wave absorption rate of the embodiment of the present application;

[0073] Figure 6 is a reflection coefficient result diagram of the embodiment of the present application under the condition of oblique incidence of TE polarized electromagnetic wave;

[0074] Figure 7 This is a reflection coefficient result diagram of the implementation example of the present application under the condition of oblique incidence of TM polarized electromagnetic waves. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0076] Genetic Algorithm (GA) was proposed by Holland. It is an optimization algorithm developed under the inspiration of natural selection and genetic mechanism in genetics. Its basic idea is to simulate the method of biological and human evolution to solve complex optimization problems.

[0077] like Figure 2 and 3 As shown, the absorbing structure in the present application is composed of a plurality of absorbing units arranged in the same period, and each absorbing unit includes a metal base plate and n-layer absorbing assembly from bottom to top. The absorbing assembly includes a material dielectric layer, an air layer and a loss layer from bottom to top; wherein n=1,2,...,m; the loss layer is a square ring-shaped ITO impedance film. In the equivalent circuit of the present application, the square ring ITO impedance film is equivalent to a series circuit of resistance, capacitance and inductance, the dielectric layer is equivalent to the characteristic impedance of the transmission line, and the metal base plate is equivalent to a short circuit.

[0078] like Figure 1 As shown, the present application provides a multilayer absorbing structure design method based on an equivalent circuit (EC) and a genetic optimization algorithm (GA), comprising:

[0079] Step 1, initialize the genetic algorithm hyperparameters and the physical parameters of the absorbing structure, and generate an initial encoding parameter array consisting of random binary codes of size (paranum, popsize, DNAsize). Where paranum is the number of optimization objects, popsize is the population size, and DNAsize is the length of the DNA sequence used for encoding. The genetic algorithm hyperparameters include crossover probability, mutation probability, gene sequence length, number of iterations, and loss function weights. The physical parameters of the absorbing structure include the relative dielectric constant ε of the intermediate dielectric layer. r , the relative dielectric constant ε of the substrate dielectric layer s , the number of layers m and the range of optimization target parameters [Var max ,Var min ]; The relationship between the number of physical parameters and the number of layers as the optimization target is paranum = 2 + 4m. The optimization target refers to the mask curve of the target reflection coefficient.

[0080] Step 2: According to the parameter range [Var max ,Var min ], use formula (1) to decode the parameter ennum into denum:

[0081]

[0082] ennum represents the binary coded parameters before decoding, and denum represents the parameters after decoding;

[0083] Step 3: The equivalent circuit of the multilayer absorbing structure is as follows: Figure 2 As shown, each layer of square ring resistor film is equivalent to a series circuit of resistance, inductance and capacitance. The dielectric layer is represented by its characteristic impedance, and the metal bottom plate is equivalent to a short circuit. The reflection coefficient S of the absorbing structure is calculated based on the equivalent circuit and transmission line theory. 11 , loss function value Loss and individual fitness Fitness;

[0084] Furthermore, in step 3, the reflection coefficient S of the absorbing structure is calculated according to the equivalent circuit and transmission line theory. 11 The specific process of loss function value Loss and individual fitness Fitness is as follows:

[0085] Step 3.1, calculate the characteristic impedance Z of the medium around the n-th square ring ITO impedance film g,n and the effective dielectric constant ε eff,n , n=1,2,...,m:

[0086]

[0087] Where ε av,n Represents the average dielectric constant of the surrounding medium:

[0088]

[0089] In the formula, when the impedance film is in the shape of a square ring, the coefficient α = 18, and ε rh,n The expression is as follows,

[0090]

[0091] Among them, h s,n and h r,n are the thickness of the nth layer substrate medium and the intermediate medium respectively.

[0092] The surrounding medium of the last layer of ITO metal film is the substrate medium and free space air, and its average dielectric constant is calculated as follows:

[0093]

[0094] Step 3.2: For each frequency point, calculate the equivalent lumped resistance R of the nth layer of ITO impedance film based on the decoded physical parameters of the nth layer. n , equivalent capacitance C n and equivalent inductance L n , n=1,2,...,m. When the shape of the ITO impedance film is a square ring, the lumped resistance R in the equivalent circuit n The surface resistance R of the ITO resistance film can be s,n , the period area S and the area of ​​the square ring Sd,n Find:

[0095]

[0096] In the formula,

[0097]

[0098] In formulas (7)-(12), p represents the unit period of the absorbing structure, d n represents the side length of the ITO impedance film square ring, w n Indicates the width of the square ring; represents the angular frequency, ε eff,n Represents the effective dielectric constant of the surrounding medium, which has an impact on the equivalent capacitance, Z g,n It represents the characteristic impedance of the medium around the square ring of ITO impedance film; F(p,x,λ) and G(p,x,λ) are functions related to the geometric parameters of the impedance film and the wavelength λ, and α and β are two intermediate variables.

[0099] Step 3.3: The input impedance Z of the last layer of the absorbing structure equivalent circuit is in,m And the free space characteristic impedance Z0 to obtain the reflection coefficient Γ:

[0100]

[0101] Furthermore, the step 3.3 calculates the input impedance Z of the last layer of the equivalent circuit in,m Specifically include:

[0102] Step 3.3.1. Calculate the characteristic impedance and propagation constant of the intermediate dielectric layer and the substrate dielectric layer for each frequency point:

[0103]

[0104] Among them, ε s,n and ε r,n are the dielectric constants of the nth layer substrate medium and the intermediate medium respectively. Secondly, calculate the equivalent impedance of the nth layer ITO impedance film:

[0105] Z ec,n =R n +jωL n +1 / (jωC n ) (17)

[0106] ω=2πf (18)

[0107] Step 3.3.2. Calculate Z′ n , which is related to the characteristic impedance Z of the intermediate medium r,n , propagation constant β r,n , thickness hr,n and the input impedance Z of the previous layer in,n-1 Since the metal bottom plate is equivalent to a short-circuit impedance of 0, the input impedance of the previous layer of the first layer is Z in,0 =0:

[0108]

[0109] Step 3.3.3. Calculate Z n , which is related to the characteristic impedance Z of the substrate medium s,n , propagation constant β s,n , thickness h s,n and Z' n related:

[0110]

[0111] Step 3.3.4. Calculate the input impedance Z of the nth layer in,n , which is equal to Z n and Z ec,n in parallel:

[0112] Z in,n =Z n ||Z ec,n (twenty one)

[0113] Step 3.3.5: Repeat steps 3.3.2 to 3.3.4 to calculate the input impedance Z of the last layer. in,m .

[0114] Step 3.4, calculate the loss value Loss, the weighted mean square error between the reflection coefficient obtained by the equivalent circuit and the target reflection coefficient is used as the value of the loss function, and the weighted mean square error calculation expression is as follows:

[0115]

[0116] R i (f j ) and G i (f j ) represent the reflection coefficient and target reflection coefficient calculated based on the transmission line theory, respectively. When R i (f j )≤G i (f j ), the penalty function E i (f j )=0, otherwise the penalty function is the absolute value of the difference between the two. i represents the weighted order, and j represents the frequency point order in the i-th weighted area.

[0117] Step 3.5, calculate the individual fitness, which is an exponential function of the loss function, and its expression is as follows:

[0118]

[0119] Step 4: Compare the minimum loss value in this round of the group with the minimum value in the previous round, and retain the minimum loss value as min_Loss; for the first round of iteration, the minimum loss value of the group is used as min_Loss.

[0120] Step 5: Use the roulette wheel selection method to select individuals to form a new group based on their fitness values. The higher the individual fitness, the greater the probability of being selected. Specifically, it includes:

[0121] Step 5.1, calculate the probability of each individual being inherited into the next generation, N is the population size, f(x i ) represents the fitness of the i-th individual:

[0122]

[0123] Step 5.2: Calculate the cumulative probability q of each individual i :

[0124]

[0125] Step 5.3, generate a random number r in the interval [0, 1], when q i-1 ≤r≤q i When , the i-th individual is selected;

[0126] Step 5.4: Repeat steps 5.1 to 5.3 N times, where N is equal to the population size popsize.

[0127] Step 6: Perform crossover and mutation operations on the selected population. Crossover refers to the crossover recombination of the parent DNA sequence under the control of the crossover probability; mutation refers to the inversion of the binary coding of certain gene bits in the parent DNA sequence under the control of the mutation probability, that is, 0 becomes 1 or 1 becomes 0, thereby forming a new offspring population;

[0128] Step 7: Repeat steps 2 to 6 until the maximum number of iterations is reached.

[0129] Step 8: Output the decoded optimal parameter combination X_best and its corresponding reflection coefficient S 11 and loss function value min_Loss; where X_best refers to the target optimization parameters, including the period p of the absorbing unit, the thickness h of the intermediate dielectric layer r,1 ,h r,2 ...h r,m , the thickness of the substrate dielectric layer h s,1 ,h s,2 ...hs,m 、ITO impedance film square resistance R1, R2...R m 、Square ring side length d1,d2...d m and the widths of the square rings w1, w2...w m , m represents the number of layers of the absorbing structure.

[0130] To prove the feasibility of the above algorithm, an embodiment is provided below. It is implemented by programming in Python language on the Pycharm editor, and the physical parameters obtained by the EC-GA optimization algorithm are directly modeled and simulated in the CST studio suite 2022 simulation software. The simulation results of the reflection coefficient of the multi-layer absorbing structure are compared with the results calculated according to the equivalent circuit and transmission line theory. It is proved that the proposed calculation method of the multi-layer absorbing structure has a high reflection coefficient calculation accuracy and is feasible in the optimization design of multi-layer ultra-wideband absorbing structures.

[0131] Example:

[0132] The embodiment is a 6-layer ultra-wideband absorbing structure based on an ITO impedance film square ring, such as Figure 3 As shown, the absorbing structural units are arranged periodically, with a metal bottom plate, and the dielectric constant of the substrate medium is ε s =1, the dielectric constant of the intermediate medium ε s =1.05, loss tangent tanδ=0.001. Set the target reflection coefficient to be less than -10dB in the range of 0.1GHz-50GHz, the number of iterations to 1000, the crossover probability to 0.6, and the mutation probability to 0.026. The target parameter values ​​obtained by the EC-GA optimization algorithm iteration are: period p=27.2mm, the substrate medium height of each layer is the same h s =0.46mm; the thickness of the intermediate medium from bottom to top h r,1 =11.9mm,h r,2 =9.58mm,h r,3 =9.54mm,h r,4 =8.24mm,h r,5 =4.97mm,h r,6=2mm; the square resistance of the square ring ITO impedance film from bottom to top is R1 = 265Ω / sqr, R2 = 279Ω / sqr, R3 = 752Ω / sqr, R4 = 766Ω / sqr, R5 = 703Ω / sqr, R6 = 929Ω / sqr; the side length of the square ring from bottom to top is d1 = 27.08mm, d2 = 26.86mm, d3 = 26.18mm, d4 = 26.76mm, d5 = 25.97mm, d6 = 22.77mm; the width of the square ring from bottom to top is w1 = 10mm, w2 = 6.97mm, w3 = 8.47mm, w4 = 8.59mm, w5 = 6.54mm, w6 = 8.25mm; the overall thickness is 48.98mm, and the relative section height is 0.119λ L .

[0133] The reflection coefficient result of the embodiment is shown in FIG. Figure 4 As shown, the dotted line is the reflection coefficient calculated based on the EC-GA algorithm, and the solid line is the reflection coefficient curve obtained by modeling and simulating in CST according to the physical parameters given by the EC-GA algorithm. It can be seen that the calculation results and the simulation results are basically completely consistent, which proves the theoretical accuracy of the design method of this application. In addition, both the calculation results and the simulation results have achieved the reflection coefficient S in the frequency range of 0.729GHz-50GHz (relative bandwidth 194.25%). 11 ≤-10dB(such as Figure 4 As shown), 90% absorption rate (as shown Figure 5 As shown in Figure 2, it is proved that the optimization design method proposed in this application can be well applied to the optimization design of ultra-wideband multilayer absorbing structures. In addition, the oblique incidence performance of the absorbing structure under TE and TM polarized electromagnetic waves is shown in Figure 2. Figure 6 and Figure 7 As shown, it has good wave absorbing stability in the oblique incidence range of θ = 0°-50°.

[0134] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm, characterized in that: include: Step 1, initialize the genetic algorithm hyperparameters and the physical parameters of the absorbing structure, and generate an initial encoding parameter array composed of random binary codes of specifications (paranum, popsize, DNAsize), where paranum is the number of optimization objects, popsize is the population size, and DNAsize is the length of the DNA sequence used for encoding. The genetic algorithm hyperparameters include crossover probability, mutation probability, gene sequence length, number of iterations, and loss function weight. The physical parameters of the absorbing structure include the relative dielectric constant ε of the intermediate dielectric layer. r , the relative dielectric constant ε of the substrate dielectric layer s , the number of layers m and the range of the optimization target parameter [Var max ,Var min ]; Step 2: According to the parameter range [Var max ,Var min ], use formula (1) to decode the parameters: ennum represents the binary coded parameters before decoding, and denum represents the parameters after decoding; Step 3: The square ring resistor film is equivalent to a series circuit of resistance, inductance and capacitance. The dielectric layer is represented by its characteristic impedance, and the metal bottom plate is equivalent to a short-circuit line. The reflection coefficient S of the absorbing structure is calculated according to the equivalent circuit and transmission line theory. 11 , loss function value Loss and individual fitness Fitness; Step 4: Compare the minimum loss value in this round of the group with the minimum value in the previous round, and retain the minimum loss value as min_Loss; for the first round of iteration, the minimum loss value of the group is taken as min_Loss; Step 5: Use the roulette wheel selection method to select individuals to form a new group based on their fitness values. The higher the individual fitness, the greater the probability of being selected. Step 6: Perform crossover and mutation operations on the selected population. Crossover refers to the crossover recombination of the parent DNA sequence under the control of the crossover probability; mutation refers to the inversion of the binary coding of certain gene bits in the parent DNA sequence under the control of the mutation probability, that is, 0 becomes 1 or 1 becomes 0, thereby forming a new offspring population; Step 7: Repeat steps 2 to 6 until the maximum number of iterations is reached. Step 8: Output the decoded optimal parameter combination X_best and its corresponding reflection coefficient S 11 and loss function value min_Loss; where X_best refers to the target optimization parameters, including the period p of the absorbing structure, the thickness h of the intermediate dielectric layer r,1 ,h r,2 ...h r,m , the thickness of the substrate dielectric layer h s,1 ,h s,2 ...h s,m 、ITO impedance film square resistance R1, R2...R m 、Square ring side length d1, d2...d m and the widths of the square rings w1, w2...w m , m represents the number of layers of the absorbing structure.

2. The multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm as claimed in claim 1, characterized in that: In step 3, the reflection coefficient S of the absorbing structure is calculated according to the equivalent circuit and transmission line theory. 11 The specific process of loss function value Loss and individual fitness Fitness is as follows: Step 3.1, calculate the characteristic impedance Z of the medium around the n-th square ring ITO impedance film g,n and the effective dielectric constant ε eff,n , n=1,2,...,m: Where ε av,n Represents the average dielectric constant of the surrounding medium: In the formula, when the impedance film is in the shape of a square ring, the coefficient α = 18, and ε rh,n The expression is as follows, Among them, h s,n and h r,n are the thickness of the nth layer substrate medium and the intermediate medium respectively; The surrounding medium of the last layer of ITO metal film is the substrate medium and free space air, and its average dielectric constant is calculated as follows: Step 3.2: For each frequency point, calculate the equivalent lumped resistance R of the nth layer of square ring ITO impedance film according to the physical parameters of the nth layer after decoding. n , equivalent capacitance C n and equivalent inductance L n , n=1,2,...,m: In the formula, In formulas (7)-(12), p represents the unit period of the absorbing structure, d n represents the side length of the ITO impedance film square ring, w n Indicates the width of the square ring; represents the angular frequency, ε eff,n Represents the effective dielectric constant of the surrounding medium, which has an impact on the equivalent capacitance, Z g,n It represents the characteristic impedance of the medium around the square ring of ITO impedance film; F(p,x,λ) and G(p,x,λ) are functions related to the geometric parameters of the impedance film and the wavelength λ, and α and β are two intermediate variables; Step 3.3: The input impedance Z of the last layer of the absorbing structure equivalent circuit is in,m And the free space characteristic impedance Z0 to obtain the reflection coefficient Γ: Step 3.4, calculate the loss value Loss, the weighted mean square error between the reflection coefficient obtained by the equivalent circuit and the target reflection coefficient is used as the value of the loss function, and the weighted mean square error calculation expression is as follows: R i (f j ) and G i (f j ) represent the reflection coefficient and target reflection coefficient calculated based on the transmission line theory, respectively. i (f j )≤G i (f j ), the penalty function E i (f j )=0, otherwise the penalty function is the absolute value of the difference between the two. i represents the weighted order, j represents the frequency point order in the i-th weighted area; Step 3.5, calculate the individual fitness, which is an exponential function of the loss function, and its expression is as follows:

3. The multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm as claimed in claim 2, characterized in that: The step 3.3 calculates the input impedance Z of the last layer of the equivalent circuit of the absorbing structure in,m Specifically include: Step 3.3.

1. Calculate the characteristic impedance and propagation constant of the intermediate dielectric layer and the substrate dielectric layer for each frequency point: Among them, ε s,n and ε r,n are the dielectric constants of the nth layer substrate medium and the intermediate medium respectively. Secondly, calculate the equivalent impedance of the nth layer ITO impedance film: Z ec,n =R n +jωL n +1 / (jωC n ) (20) ω=2πf (21) Step 3.3.

2. Calculate Z ’ n , which is related to the characteristic impedance Z of the intermediate medium r,n , propagation constant β r,n , thickness h r,n and the input impedance Z of the previous layer in,n-1 Since the metal bottom plate is equivalent to a short-circuit impedance of 0, the input impedance of the previous layer of the first layer is Z in,0 =0: Step 3.3.

3. Calculate Z n , which is related to the characteristic impedance Z of the substrate medium s,n , propagation constant β s,n , thickness h s,n and Z ’ n related: Step 3.3.

4. Calculate the input impedance Z of the nth layer in,n , which is equal to Z n and Z ec,n in parallel: WITH in,n =Z n ||From ec,n (24) Step 3.3.5: Repeat steps 3.3.2 to 3.3.4 to calculate the input impedance Z of the last layer. in,m .

4. The multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm according to any one of claims 1 to 3, characterized in that: The step 5 specifically includes: Step 5.1, calculate the probability of each individual being inherited into the next generation, N is the population size, f(x i ) represents the fitness of the i-th individual: Step 5.2: Calculate the cumulative probability q of each individual i : Step 5.3, generate a random number r in the interval [0, 1], when q i-1 ≤r≤q i When , the i-th individual is selected; Step 5.4: Repeat steps 5.1 to 5.3 N times, where N is equal to the population size popsize.

5. The multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm as claimed in claim 4, characterized in that: Each layer of the absorbing structure is composed of a plurality of absorbing units arranged in the same period, each absorbing unit comprises a metal bottom plate and n-layer absorbing assemblies from bottom to top; the absorbing assemblies comprise a material medium layer, an air layer and a loss layer from bottom to top.

6. The multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm as claimed in claim 5, characterized in that: The loss layer is an ITO impedance film in a square ring shape.

7. The multilayer absorbing structure design method based on equivalent circuit and genetic optimization algorithm as claimed in claim 6, characterized in that: The relative magnetic permeability μ of the dielectric layer is 1, and the relative dielectric constant ε is ≤15.

Images