A design method and application of a multi-component electromagnetic wave absorbing material based on machine learning inverse design

Abstract

The application relates to a design method and application of a multi-component electromagnetic wave absorbing material based on machine learning reverse design, and aims to solve the problem that intrinsic electromagnetic parameters of a plating layer cannot be obtained at present, and a design method for quickly and effectively designing different microwave absorption targets is obtained. Intrinsic electromagnetic parameters are obtained, a genetic algorithm in machine learning is combined with the guidance of Maxwell Garnett and transmission line principle calculation, and a multi-component composite wave absorbing material meeting different targets is prepared. The multi-component composite wave absorbing material prepared by the application has the innovation that intrinsic electromagnetic parameters of a plating layer are obtained, the formula of the material is optimized under the guidance of the algorithm, and the electromagnetic characteristics of the material itself are fully utilized. The application is applied to the field of electromagnetic wave absorbing materials.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic wave absorbing materials preparation and application, specifically to a design method and application of multi-component electromagnetic wave absorbing materials based on machine learning reverse design. Background Technology

[0002] The rapid development of electromagnetic wave technology has greatly promoted social progress, making significant contributions in both civilian and military fields such as 5G communication and radar detection. However, the use of electromagnetic waves has also brought serious problems, including electromagnetic radiation / leakage and weapon exposure. Microwave absorbing materials capable of dissipating electromagnetic wave energy and solving electromagnetic pollution problems have become extremely important in modern society. In recent years, the application scenarios of electromagnetic wave technology have become increasingly frequent; therefore, there is an urgent need for strategies that can design electromagnetic parameters according to a range of requirements, especially for broadband and low-frequency absorption.

[0003] Therefore, the research and application of full-band microwave absorbing materials are particularly important. To address the bottleneck issues of "broadband absorption" and "low-frequency absorption" in electromagnetic wave absorbing materials, the balance between the complex permittivity and complex permeability in the system is the first key area to explore. It is well known that considering only the electromagnetic parameters corresponding to each integer frequency (2.0-18.0 GHz), this spans at least 16^4 design matrix spaces, a space so large that it cannot be managed using traditional microwave absorbing material design methods. Even if the complex permittivity and permeability required for the target electromagnetic wave absorption performance are known, they are still difficult to achieve due to limitations in intrinsic electromagnetic properties.

[0004] Patent publication CN114016182A, entitled "Preparation Method and Application of a Multifunctional, High-Temperature Resistant, Broadband Absorption Periodically Braided Electromagnetic Wave Absorbing Material," discloses a method for preparing a broadband electromagnetic wave absorbing periodically woven fiber material using a braiding machine by obtaining intrinsic electromagnetic parameters and employing a genetic algorithm combined with CST software guidance. The innovation of this invention lies in obtaining the intrinsic electromagnetic parameters of the fiber material and optimizing the periodic macroscopic structure of the material under the guidance of algorithms and simulations, thus fully utilizing the material's electromagnetic losses while simultaneously coordinating with the resonant losses of the periodic structure. However, this invention is limited by the woven fiber material and the braiding process, thus restricting its electromagnetic absorption performance and preventing the achievement of effective broadband absorption at low frequencies. Summary of the Invention

[0005] This invention aims to address the current inability to obtain the intrinsic electromagnetic parameters of the coating, thus hindering the development of microwave absorbing materials that simultaneously possess adaptive electromagnetic parameters and dynamically achieve the target electromagnetic wave absorption performance. To this end, this invention utilizes machine learning algorithms combined with the Maxwell-Garnett equivalent dielectric equation and transmission line principle to develop a design method and application for multi-component electromagnetic wave absorbing materials, accelerating the design process and expanding the material performance gaps.

[0006] This invention obtains the intrinsic electromagnetic parameters of a single phase using the Maxwell-Garnett equation and extends this to obtain the intrinsic electromagnetic parameters of the coating. Utilizing a genetic algorithm from machine learning combined with the transmission line principle, a multi-component microwave absorbing material is designed and fabricated. The innovation of this invention lies in obtaining the intrinsic electromagnetic parameters of the coating material, optimizing the material formulation under the guidance of the algorithm, fully utilizing the intrinsic electromagnetic properties of the material, expanding the performance gaps, and ultimately achieving ultra-wideband absorption, low-frequency absorption, and target frequency band absorption. This results in a multi-component microwave absorbing material and its design method.

[0007] This invention discloses a design method for multi-component electromagnetic wave absorbing materials based on machine learning reverse design, which specifically comprises the following steps:

[0008] I. Preparation of Magnetic Metal Electroplating on Carbon Fibers

[0009] ① Pretreatment before electroplating: First, immerse the carbon fiber in acetone at room temperature to remove impurities, then wash it with distilled water, dry it, heat the carbon fiber to remove organic binders, and then cool it to room temperature.

[0010] ② Electroplating of magnetic metal coatings onto carbon fibers;

[0011] ③ After electroplating, the coated fibers are washed with distilled water, dried, and then cut into short fibers of a fixed length.

[0012] II. Obtaining the intrinsic electromagnetic parameters of the coating absorber material:

[0013] ① Take the electroplated magnetic metal carbon fiber cut in step one, calculate and weigh the mass of the short electroplated magnetic metal carbon fiber with a volume fraction of 0.5 to 2 vol.% and the mass of the liquid paraffin.

[0014] ② Following the calculation results in step 2①, weigh and mix the chopped electroplated magnetic metal fibers and liquid paraffin to ensure thorough dispersion of the chopped electroplated magnetic metal carbon fibers and liquid paraffin, with the orientation of the chopped electroplated magnetic metal carbon fibers randomly distributed. Obtain multiple samples of chopped electroplated magnetic metal carbon fiber composite paraffin material with different thicknesses. Name the sample with the largest thickness among the multiple samples of chopped electroplated magnetic metal carbon fiber composite paraffin material with different thicknesses as NIFA and the sample with the smallest thickness as NIFB. Then measure the electromagnetic parameters of the NIFA and NIFB samples. Input the characteristic dimensions of the electroplated magnetic metal chopped carbon fibers and calculate the depolarization factor of the microwave absorber.

[0015] ③ By comparing and analyzing the electromagnetic parameters, Cole-Cole, and C0 of NIFA and NIFB, if the electromagnetic parameters do not change abruptly, the number of semicircles representing polarization in the Cole-Cole curve is consistent, and the number of peaks in C0 remains unchanged, then it can be determined that the increase in coating thickness does not introduce a new loss mechanism.

[0016] ④ Substitute the electromagnetic parameter test results of the NIFA and NIFB samples from step 2 ② into the inverse formula of the Maxwell-Garnett equivalent equation, and substitute the complex permittivity ε of NIFB. m and the NIFA complex permittivity ε of the microwave absorbing agent composite material eff To obtain the intrinsic electromagnetic parameters ε of the electroplated magnetic metal layer. a ;

[0017]

[0018] III. Obtaining the intrinsic electromagnetic parameters of a single-phase microwave absorbing material:

[0019] ① Mix the single-phase microwave absorber with liquid paraffin and disperse it fully to obtain composite samples of microwave absorbers with different volume fractions. Press the obtained composite samples into coaxial ring samples and then measure the dielectric parameters (including 2D material tungsten disulfide, 3D material Co spheres, etc.).

[0020] ② Input the characteristic size, i.e., the dispersion state, of a single-phase absorber and calculate the depolarization factor of the absorber;

[0021] ③ Substitute the dielectric parameters and depolarization factor of the absorber obtained in steps ① and ② into the Maxwell-Garnett equivalent equation to obtain the intrinsic dielectric parameters of a single-phase absorber;

[0022]

[0023] IV. Optimization Design Based on Genetic Algorithm and Transmission Line Principle:

[0024] ① Based on the intrinsic dielectric parameters of the absorbing agent and the intrinsic electromagnetic parameters of other absorbing agents obtained in steps two and three, establish an intrinsic electromagnetic parameter library, and randomly select different absorbing agents from the intrinsic electromagnetic parameter library to form multi-component materials.

[0025] ②. Using all material parameters in the multi-component composite material from step 4① as independent variables, input them into the Maxwell–Garnett equivalent medium model to calculate the equivalent electromagnetic parameters of the multi-component composite material:

[0026]

[0027] ③ Set the evaluation function, including broadband absorption, low-frequency absorption and special frequency bands under different thicknesses as absorption targets, input the electromagnetic parameters obtained in step ①, set the target thickness, and use the transmission line principle to calculate the reflection loss of the multi-component composite material; when the reflection loss is ≤-10dB, it is determined as effective absorption and all effective bandwidths are counted. Finally, save the parameters and reflection loss curves that are close to the target.

[0028] V. Preparation of multi-component electromagnetic wave absorbing materials:

[0029] The microwave absorber obtained according to the optimized parameters was mixed with liquid paraffin, and the mixed material sample was finally pressed into a coaxial ring standard sample.

[0030] Furthermore, the microwave absorbing agent is a one-dimensional, two-dimensional, or three-dimensional material; the one-dimensional material is a polymer fiber, metal fiber, oxide fiber, ceramic fiber, or carbon fiber; the two-dimensional material is graphite nanosheets or graphene; and the three-dimensional material is nanospheres or hollow microspheres.

[0031] Furthermore, in step two ④, the complex permittivity ε of the NiFB composite paraffin matrix is ​​utilized. m and the complex permittivity ε of NiFA composite materials eff To obtain the intrinsic electromagnetic parameters ε of the electroplated magnetic metal layer. i The formula for obtaining it is as follows:

[0032]

[0033] Where, ε m ε eff ε a These represent the complex dielectric constants of the matrix, the composite material of the microwave absorber, and the intrinsic dielectric constant of the carbon fiber, respectively; f is the volume fraction of the amount added, N j The depolarization factor is defined along the triaxial axis of the fiber inclusions; where the volume fraction of the microwave absorber in the microwave absorber composite material is 1 vol.%.

[0034] Furthermore, the intrinsic electromagnetic parameters of a single phase were obtained using the inverse formula of the Maxwell–Garnett equivalent medium model, and the electromagnetic parameter ε of a single-phase composite material with a volume fraction of 1 vol.% was obtained. eff The electromagnetic parameters of the matrix are those of pure paraffin, specifically ε. m ; Calculate the electromagnetic parameters ε of a single-phase absorber. i .

[0035] Furthermore, the formula for calculating the depolarization factor of the absorber described in step 3② is as follows:

[0036]

[0037] Where a x a y a z These are the semi-major axis and two semi-minor axis lengths of the short-cut electroplated magnetic metal carbon fiber, respectively.

[0038] Furthermore, the equivalent electromagnetic parameters of the multi-component composite material are obtained using the Maxwell–Garnett equivalent medium model, with the intrinsic electromagnetic parameters ε of each absorber in the multi-phase composite material input. i The electromagnetic parameters of the matrix are those of pure paraffin, specifically ε. m ; Calculate the equivalent electromagnetic parameters ε of the multi-component microwave absorber. eff :

[0039]

[0040] Further, in step 3, the multi-component microwave absorbing agent described in ① is mixed with liquid paraffin at 60°C and fully dispersed to obtain a composite material sample, and a pure paraffin sample is prepared.

[0041] Furthermore, the evaluation function described in step four ③ is a piecewise function.

[0042] Furthermore, the evaluation function described in step four ③ is a piecewise function, specifically if RL frq >-10dB,set V frq =0; if RL frq <-10dB, V frq =-1, the piecewise function is changed according to the target absorption frequency band; where RL frq V represents the reflection loss at each frequency point. frq This is a set value that indicates whether the absorption is effective.

[0043] The present invention relates to an application of a multi-component electromagnetic wave absorbing material based on machine learning reverse design. It is used as an electromagnetic wave absorbing material in various waveband scenarios; the various wavebands are ultraviolet waveband, infrared waveband, or microwave waveband.

[0044] The beneficial effects of this invention are:

[0045] This invention innovatively proposes using the electromagnetic parameters of magnetic metal coating materials with the same composition as the matrix and equivalent electromagnetic parameters, respectively. Guided by the Maxwell-Garnett equivalent equation, it solves the current problem of not being able to obtain the intrinsic electromagnetic parameters of coating materials. The method of this invention achieves increased coating thickness without introducing new loss mechanisms, thus obtaining more accurate intrinsic electromagnetic parameters of the coating. Based on an intrinsic electromagnetic parameter library, algorithms can be designed to optimize the formulation of multi-component composite materials. Without extensive experimental exploration, it is possible to obtain electromagnetic wave absorbing materials with different requirements for effective absorption frequency bands in different application scenarios.

[0046] I. Electroplated magnetic metal short-cut carbon fiber materials are prepared using traditional electroplating processes. The preparation process of this invention is low-cost, simple, and can be mass-produced.

[0047] II. A method for obtaining intrinsic electromagnetic parameters of coatings on electroplated magnetic metal short-cut carbon fiber materials is provided.

[0048] Third, this invention uses calculation and optimization to replace the cumbersome experimental trial and error process, saving a great deal of experimental costs and time, and greatly expanding the performance gap range of materials.

[0049] Fifth, multi-component composite absorbing materials can be optimized with different formulations according to the requirements of different wavebands, thereby achieving multi-functionality.

[0050] VI. The innovation of the design method for multi-component electromagnetic wave absorbing materials based on machine learning reverse design proposed in this invention lies in obtaining the intrinsic electromagnetic parameters of the coating material, optimizing the material formulation under the algorithm, fully utilizing the material's electromagnetic parameters, and ultimately achieving directional design. This results in a multifunctional multi-component electromagnetic wave absorbing material with specified frequency band absorption, obtained efficiently and at low cost. This invention is applied in the field of electromagnetic wave absorbing materials. Attached Figure Description

[0051] Figure 1 The images shown are SEM images of the nickel-plated carbon fiber described in Example 1; where image a shows the morphology of NIF04 and image b shows the morphology of NIF12.

[0052] Figure 2 The XRD pattern described in Example 1 shows that line A represents carbon fiber, line B represents NIF04, and line C represents NIF12.

[0053] Figure 3The electromagnetic parameters of the carbon fiber, NIF04, and NIF12 composite material samples described in Example 1 are shown in the figure; where a is the complex permittivity and b is the complex permeability; in the figure, A is carbon fiber, B is NIF04, and C is NIF12.

[0054] Figure 4 The image shows the Cole-Cole diagram of the carbon fiber, NIF04, and NIF12 composite material sample described in Example 1; in the figure, A represents carbon fiber, B represents NIF04, and C represents NIF12.

[0055] Figure 5 The image shows the CO diagram of the carbon fiber, NIF04, and NIF12 composite material sample described in Example 1; in the image, A represents carbon fiber, B represents NIF04, and C represents NIF12.

[0056] Figure 6 The image shows the RL diagram of the carbon fiber, NIF04, and NIF12 composite material sample described in Example 1; in the image, A represents carbon fiber, B represents NIF04, and C represents NIF12.

[0057] Figure 7 The diagram shows the intrinsic electromagnetic parameters of the nickel plating layer described in Example 1; where A is ε”, B is ε’, C is μ”, and D is μ’.

[0058] Figure 8 This is a flowchart of the machine learning process described in Example 1;

[0059] Figure 9 This is a morphology image of Ni@CF with a coating thickness of 8 micrometers as described in Example 1;

[0060] Figure 10 When the thickness is limited to 2.0 mm as described in Example 1, the electromagnetic parameters of Ni@CF are optimized using a genetic algorithm for the Ku and X bands (the left figure shows the dielectric parameter comparison, and the right figure shows the magnetic permeability comparison); in the figure, Measurement represents the actual test results; calculation represents the results predicted by this method.

[0061] Figure 11 When the thickness is limited to 2.0 mm as described in Example 1, a genetic algorithm is used to optimize the Ni@CF reflection loss RL diagram for the Ku and X bands; in the diagram, Measurement represents the length; calculation represents the prediction result.

[0062] Figure 12As described in Example 1, the thickness is limited to less than 3.0 mm. For the Ku and X bands (the left figure shows the dielectric parameter comparison, and the right figure shows the magnetic permeability comparison), the electromagnetic parameter diagram of Ni@CF is optimized using a genetic algorithm; in the figure, Measurement represents the length; calculation represents the calculated prediction result.

[0063] Figure 13 As described in Example 1, the thickness is limited to less than 3.0 mm. For the Ku and X bands, the reflection loss RL diagram of Ni@CF is optimized using a genetic algorithm; in the diagram, Measurement represents the length; calculation represents the calculated prediction result.

[0064] Figure 14 The images shown are SEM images of the iron-plated carbon fiber described in Example 2; where image a shows the morphology of F02 and image b shows the morphology of F08.

[0065] Figure 15 The XRD pattern of the iron-plated carbon fiber described in Example 2; where A is F02 and b is F08;

[0066] Figure 16 The figures show the electromagnetic parameters of the F02 and F06 composite material samples described in Example 2; the left figure shows the dielectric parameters of the matrix F02 and F08 samples, and the right figure shows the magnetic permeability; in the figures, A is ε', B is ε”; C is μ', and D is μ”.

[0067] Figure 17 The above are Cole-Cole diagrams of the F02 and F08 composite material samples described in Example 2; where A represents F04 and B represents F12.

[0068] Figure 18 The image shows the CO diagram of the F02 and F08 composite material samples described in Example 2; where A represents F04 and B represents F12.

[0069] Figure 19 The diagram shows the intrinsic electromagnetic parameters of the iron coating described in Example 2; where B is ε”, A is ε’, C is μ’, and D is μ”.

[0070] Figure 20 This is a diagram of the intrinsic electromagnetic parameters of the carbon fiber described in Example 2; in the diagram, B represents ε” and A represents ε’.

[0071] Figure 21 Figure 2 shows the electromagnetic parameters of the Co spheres described in Example 2; where Figure 1a is the electromagnetic parameter diagram of the paraffin matrix, where A is ε', C is ε', B is μ', and D is μ”; Figure 2b is the electromagnetic parameter diagram of the Co sphere composite paraffin sample, where A is ε', C is ε', B is μ', and D is μ”.

[0072] Figure 22This is a diagram of the intrinsic electromagnetic parameters of the Co sphere described in Example 1; in the diagram, C is ε”, A is ε’, B is μ’, and D is μ”.

[0073] Figure 23 The figure shows the electromagnetic parameters of the tungsten disulfide sheet described in Example 1; wherein, the figure shows the electromagnetic parameters of the 1 vol% tungsten disulfide sheet composite paraffin sample; C is ε”, A is ε’, B is μ’, and D is μ”;

[0074] Figure 24 This is a diagram showing the intrinsic electromagnetic parameters of the tungsten disulfide sheet described in Example 1; in the diagram, B represents ε” and A represents ε’.

[0075] Figure 25 When the thickness is limited to within 2.0 mm as described in Example 2, the electromagnetic parameters of the multi-component composite material are optimized using a genetic algorithm for the Ku and X bands; the upper figure is a comparison of dielectric constants, and the lower figure is a comparison of magnetic permeability.

[0076] Figure 26 When the thickness is limited to within 2.0 mm as described in Example 2, the RL diagram of the multi-component composite material is optimized using a genetic algorithm for the Ku and X bands.

[0077] Figure 27 When the thickness is limited to within 5.0 mm as described in Example 2, the electromagnetic parameter diagram of the multi-component composite material is optimized using a genetic algorithm for the S-band.

[0078] Figure 28 When the thickness is limited to within 5.0 mm as described in Example 2, the RL diagram of the multi-component composite material is optimized using a genetic algorithm for the S-band.

[0079] Figure 29 When the thickness is limited to within 6.0 mm as described in Example 2, the electromagnetic parameter diagram of the multi-component composite material is optimized using a genetic algorithm for the C-band.

[0080] Figure 30 When the thickness is limited to within 6.0 mm as described in Example 2, the RL diagram of the multi-component composite material is optimized using a genetic algorithm for the C-band.

[0081] Figure 31 When the thickness is limited to within 8.0 mm as described in Example 2, the electromagnetic parameter diagram of the multi-component composite material is optimized using a genetic algorithm for the entire wavelength range.

[0082] Figure 32 When the thickness is limited to within 8.0 mm as described in Example 2, the electromagnetic parameter diagram of the multi-component composite material is optimized using a genetic algorithm for the entire wavelength range. Detailed Implementation

[0083] Those skilled in the art will understand that the above embodiments are specific examples of implementing the present invention, and in practical applications, various changes in form and detail may be made without departing from the spirit and scope of the present invention.

[0084] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the spirit of the contents disclosed in the present invention will be described in detail below. After understanding the embodiments of the present invention, any person skilled in the art can make changes and modifications based on the technology taught in the present invention without departing from the spirit and scope of the present invention.

[0085] The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but are not intended to limit the present invention.

[0086] The beneficial effects of the present invention are verified using the following embodiments:

[0087] Example 1:

[0088] This embodiment presents a design method for multi-component electromagnetic wave absorbing materials based on machine learning reverse design, characterized by the following steps:

[0089] I. Preparation of Magnetic Metal Electroplating on Carbon Fibers

[0090] ① Pretreatment before electroplating: First, immerse the carbon fiber in acetone at room temperature for 24 hours to remove impurities, then wash it three times with distilled water, dry it in an oven at 60℃ for 4 hours, then heat the carbon fiber in air at 450℃ for 30 minutes to burn off the organic binder, and then cool it to room temperature.

[0091] ② In this experiment, a magnetic metallic coating was prepared on carbon fiber using electroplating. Taking nickel as an example, the surface-treated carbon fiber was electroplated in an electroplating solution with the following composition: 140g L -1 Ni2SO4·6H2O, 0.3 g·L -1 C6H5NaO2S, 2.0 g·L - 1 C7H5NO3S, 40g·L -1 H3BO3, 15g·L -1 C6H5Na3O7·2H2O and 30g·L -1 The pH of the KCl solution was controlled at 3.0 ± 0.5. During the electroplating process, 10 cm long carbon fibers were cut and used as the cathode, while a 65 × 60 × 5 mm pure nickel plate was polished and used as the anode. Electroplating was carried out at 60°C for 3–20 minutes at a current density of 0.2 A dm³. -2 It is controlled by a DC regulated power supply, and the coating thickness is between 0.4 micrometers and 1.5 micrometers.

[0092] ③ After electroplating, the coated fibers are washed three times with distilled water and dried in an oven at 60℃ for 4 hours. Then, they are cut into short fibers of a fixed length.

[0093] II. Obtaining the intrinsic electromagnetic parameters of the coating absorber material:

[0094] ① Electroplated nickel carbon fibers with different coating thicknesses were prepared using the above process and cut into short-cut nickel-plated fibers (1.0 mm) with uniform length. Two short-cut nickel-plated fibers with stable coating composition and uniform thickness (0.4 μm and 1.2 μm) were taken, and the mass of the short-cut nickel-plated carbon fibers with a filling amount of 1 vol.% (volume fraction of carbon fiber, not volume fraction of electroplated carbon fiber) and the corresponding mass of paraffin wax were calculated and weighed.

[0095] ② The weighed short-cut nickel-plated fibers were mixed with liquid paraffin to ensure thorough dispersion of the fibers and random orientation of the fibers. This yielded short-cut nickel-plated fiber composite paraffin material samples with different coating thicknesses, named NIF04 and NIF12, respectively. Figure 1 Then, the electromagnetic parameters are measured; the characteristic dimensions of the electroplated nickel short-cut carbon fiber are input, and the depolarization factor of the microwave absorber is calculated;

[0096] ③ By analyzing the electromagnetic parameters of the two composite materials ( Figure 2 ), C0 ( Figure 3 ), Cole-Cole ( Figure 4 Comparative analysis showed that increasing the coating thickness and content did not introduce a new loss mechanism.

[0097] ④ Using ③ as the basis for calculation, expand the application scope of the Maxwell-Garnett equivalent medium model by substituting the electromagnetic parameter test results of NiF04 and NiF12 samples into the Maxwell-Garnett equivalent medium model. m ε eff To obtain the intrinsic electromagnetic parameters ε of the nickel plating. i ;( Figure 5 )

[0098]

[0099] Where, ε m ε eff , respectively, are the complex permittivity of the matrix and the complex permittivity of the microwave absorber composite material; f is the volume fraction of the amount added, N j The depolarization factor is defined along the triaxial axis of the fiber inclusions; where the volume fraction of the microwave absorber in the microwave absorber composite material is 1 vol.%.

[0100] ⑤ The intrinsic electromagnetic parameters of the coating thickness calculated in ④ are verified by the electromagnetic parameters of electroplated nickel chopped fiber composites with other coating thicknesses.

[0101] III. Optimization Design Based on Genetic Algorithm and Transmission Line Principle:

[0102] ① Combining the intrinsic electromagnetic parameters of the coating obtained through the above methods ( Figure 6 ) and intrinsic electromagnetic parameters of NIF04 material ( Figure 7 ), where A is ε”, B is ε’; C is μ”, D is μ’;.

[0103] ② Input all parameters of Ni@CF as independent variables into the Maxwell–Garnett equivalent media model to calculate the multi-Ni@CF composite matrix. Independent variables include Ni@CF thickness, Ni@CF volume fraction (total integral range 0–1.5 vol.%), and dimensions (length within 2 mm), etc. Figure 8 As can be seen, this strategy integrates multiple algorithms to optimize the designability factor of the entire absorbing material system, selecting the formulation that best meets the target from a large number of formulations, thus saving experimental costs and accelerating the design process; Figure 8 )

[0104] The formula for calculating the depolarization factor of the absorber is as follows:

[0105]

[0106] Where ax, ay, and az are the semi-major axis and the two semi-minor axis lengths of the short-cut electroplated magnetic metal carbon fiber, respectively.

[0107] ③ Set the evaluation function, including targets such as broadband absorption, low-frequency absorption and special frequency band absorption under different thicknesses. Input the electromagnetic parameters obtained from the Maxwell-Garnett equivalent medium model and set the target thickness. Use the transmission line principle to obtain the reflection loss of the multi-component composite material. When the reflection loss is <-10dB, it is determined to be effective absorption and all effective bandwidths are counted. Finally, save the parameters and reflection loss curves that are close to the target.

[0108] The equivalent electromagnetic parameters of the multi-component composite material are obtained using the Maxwell–Garnett equivalent medium model, with the intrinsic electromagnetic parameters ε of each absorber in the multi-phase composite material as input. i The electromagnetic parameters of the matrix are those of pure paraffin, specifically ε. m ; Calculate the equivalent electromagnetic parameters ε of the multi-component microwave absorber. eff :

[0109]

[0110] εeff is the complex permittivity of the composite material of the microwave absorber; f is the volume fraction of the amount added, N j is the depolarization factor along the triaxial axis of fiber inclusions.

[0111] IV. Preparation of Ni@CF fundamental wave absorbing materials:

[0112] Ni@CF-10 obtained according to the optimized parameters Figure 9 Simultaneously, it is mixed with liquid paraffin, and finally the Ni@CF-based sample is pressed into a coaxial ring standard sample. For the widest absorption in the X and Ku bands within 2.0 mm, the electromagnetic parameters and RL values ​​show that this method is rapid and effective. Figure 10 , 11). Regarding the widest absorption in the X and Ku bands within 3.0 mm, testing its electromagnetic parameters and RL values ​​shows that this method is quick and effective. Figure 12 ,13)

[0113] Example 2

[0114] 1. A design method for multi-component electromagnetic wave absorbing materials based on machine learning reverse design, characterized in that the method is carried out in the following steps:

[0115] I. Preparation of Magnetic Metal Electroplating on Carbon Fibers

[0116] ① Pretreatment before electroplating: First, immerse the carbon fiber in acetone at room temperature for 24 hours to remove impurities, then wash it three times with distilled water, dry it in an oven at 60℃ for 4 hours, then heat the carbon fiber in air at 450℃ for 30 minutes to burn off the organic binder, and then cool it to room temperature.

[0117] ② In this experiment, a magnetic metallic coating was prepared on carbon fiber using electroplating. Taking nickel as an example, the surface-treated carbon fiber was electroplated in an electroplating solution with the following composition: 130g L -1 FeSO4·7H2O, 0.3 g·L -1 C6H5NaO2S, 2.0 g·L - 1 C7H5NO3S, 40g·L -1 H3BO3, 15g·L -1 C6H5Na3O7·2H2O and 30g·L -1 The pH of the KCl solution was controlled at 3.0 ± 0.5. During the electroplating process, 10 cm long carbon fibers were cut and used as the cathode, while a 65 × 60 × 5 mm pure nickel plate was polished and used as the anode. Electroplating was carried out at 60°C for 3–20 minutes at a current density of 0.2 A dm³. -2 It is controlled by a DC regulated power supply, and the coating thickness is between 0.2 micrometers and 0.8 micrometers.

[0118] ③ After electroplating, the coated fibers are washed three times with distilled water and dried in an oven at 60℃ for 4 hours. Then, they are cut into short fibers of a fixed length.

[0119] ④ Wasted water is recycled into a special container to prevent water resources from being contaminated.

[0120] II. Obtaining the intrinsic electromagnetic parameters of the coating absorber material:

[0121] ① Electroplated iron carbon fibers with different coating thicknesses were prepared using the above process and cut into short-cut nickel-plated fibers (1.0 mm) with uniform length. Two short-cut nickel-plated fibers with stable coating composition and uniform thickness (0.4 μm and 1.2 μm) were taken, and the mass of the short-cut iron-plated carbon fibers and the corresponding paraffin wax mass at a filling amount of 1 vol.% (volume fraction of carbon fiber, not volume fraction of electroplated carbon fiber) were calculated and weighed.

[0122] ② Mix the weighed short-cut iron-coated carbon fibers with liquid paraffin to fully disperse the short-cut iron-coated carbon fibers and paraffin, and to obtain short-cut iron-coated carbon fiber composite paraffin material samples with different coating thicknesses, named F02 and F08 respectively. Then, measure the electromagnetic parameters; input the characteristic dimensions of the short-cut iron-coated carbon fibers and calculate the depolarization factor of the microwave absorber.

[0123] ③ By comparing and analyzing the electromagnetic parameters, C0, and Cole-Cole of the two composite materials, it was determined that the increase in coating thickness and content does not introduce a new loss mechanism. Figure 15-18 ).

[0124] ④ Using ③ as the basis for calculation, expand the application scope of the Maxwell-Garnett equivalent medium model by substituting the electromagnetic parameter test results of NIF04 and NIF12 samples into the Maxwell-Garnett equivalent medium model. m ε eff To obtain the intrinsic electromagnetic parameters ε of the nickel plating. i ;( Figure 19 )

[0125]

[0126] Where, ε m ε eff , respectively, are the complex permittivity of the matrix and the complex permittivity of the microwave absorber composite material; f is the volume fraction of the amount added, N j The depolarization factor is defined along the triaxial axis of the fiber inclusions; where the volume fraction of the microwave absorber in the microwave absorber composite material is 1 vol.%.

[0127] ⑤ The intrinsic electromagnetic parameters of the coating thickness calculated in ④ are verified by the electromagnetic parameters of electroplated nickel chopped fiber composites with other coating thicknesses.

[0128] III. Obtaining the intrinsic electromagnetic parameters of a single-phase microwave absorbing material:

[0129] ① Mix the single-phase microwave absorber with liquid paraffin and disperse thoroughly to obtain composite samples with different volume fractions of microwave absorber (0 vol.%, 1 vol.%). Press the obtained composite samples into standard coaxial ring samples, and then measure the dielectric parameters (including 2D materials: tungsten disulfide, 3D materials: Co spheres, etc.); Figure 20-24 )

[0130] ② Input the characteristic dimensions of the material and calculate the depolarization factor of the absorber;

[0131] The formula for calculating the depolarization factor of the absorber is as follows:

[0132]

[0133] Where a x a y a z These are the semi-major axis and two semi-minor axis lengths of the short-cut electroplated magnetic metal carbon fiber, respectively.

[0134] ③ Substitute the parameters obtained in ① and ② into the Maxwell-Garnett equivalent equation to obtain the intrinsic dielectric parameters of the absorber;

[0135] IV. Optimization Design Based on Genetic Algorithm and Transmission Line Principle:

[0136] ① Based on the intrinsic electromagnetic parameters obtained by the above methods and those obtained from the literature, an intrinsic electromagnetic parameter library is established, and the composition of multi-component materials is defined as materials randomly selected from the intrinsic electromagnetic parameter library.

[0137] ② Input all material parameters of the multi-component composite material as independent variables into the Maxwell–Garnett equivalent medium model to calculate the equivalent electromagnetic parameters of the multi-component composite material. Independent variables include the type and thickness of the magnetic material coating, the volume fraction of the absorbing agent (the total integral of all absorbing agents ranges from 0 to 2 vol.%), and the size (the size of the absorbing agent is within 2 mm).

[0138] ③ Set the evaluation function, including targets such as broadband absorption, low-frequency absorption and special frequency band absorption under different thicknesses, input the Maxwell-Garnett equivalent medium model to obtain electromagnetic parameters, set the target thickness, and use the transmission line principle to obtain the reflection loss of multi-component composite materials; when the reflection loss is <-10dB, it is determined as effective absorption and all effective bandwidths are counted. Finally, save the parameters and reflection loss curves that are close to the target.

[0139] The equivalent electromagnetic parameters of the multi-component composite material are obtained using the Maxwell–Garnett equivalent medium model, with the intrinsic electromagnetic parameters ε of each absorber in the multi-phase composite material as input. i The electromagnetic parameters of the matrix are those of pure paraffin, specifically ε. m ; Calculate the equivalent electromagnetic parameters ε of the multi-component microwave absorber. eff :

[0140]

[0141] ε eff is the complex permittivity of the composite material of the microwave absorber; f is the volume fraction of the amount added, N j is the depolarization factor along the triaxial axis of fiber inclusions.

[0142] V. Preparation of multi-component electromagnetic wave absorbing materials:

[0143] The microwave absorber obtained according to the optimized parameters was mixed with liquid paraffin, and the mixed material sample was finally pressed into a coaxial ring standard sample.

[0144] For the widest absorption in the X and Ku bands within 2.0 mm, testing its electromagnetic parameters and RL value shows that this method is quick and effective. Figure 25 ,26).

[0145] For the widest absorption in the S-band within 5.0 mm, testing its electromagnetic parameters and RL value shows that this method is quick and effective. Figure 27 ,28).

[0146] For the widest absorption in the C-band within 6.0 mm, testing its electromagnetic parameters and RL value shows that this method is quick and effective. Figure 29 ,30).

[0147] For the widest absorption across the entire 8.0mm band, testing its electromagnetic parameters and RL value shows that this method is quick and effective. Figure 31 ,32).

Claims

1. A design method for multi-component electromagnetic wave absorbing materials based on machine learning reverse design, characterized in that... This method is specifically carried out according to the following steps: I. Preparation of Magnetic Metal Electroplating on Carbon Fibers ① Pretreatment before electroplating: First, immerse the carbon fiber in acetone at room temperature to remove impurities, then wash it with distilled water, dry it, heat the carbon fiber to remove organic binders, and then cool it to room temperature. ② Electroplating of magnetic metal coatings onto carbon fibers; ③ After electroplating, the coated fibers are washed with distilled water, dried, and then cut into short fibers of a fixed length. II. Obtaining the intrinsic electromagnetic parameters of the coating absorber material: ① Take the electroplated magnetic metal carbon fiber cut in step one, calculate and weigh the mass of the short electroplated magnetic metal carbon fiber with a volume fraction of 0.5~2 vol.% and the mass of the liquid paraffin. ② Following the calculation results in step 2①, weigh and mix the chopped electroplated magnetic metal fibers and liquid paraffin to ensure thorough dispersion of the chopped electroplated magnetic metal carbon fibers and liquid paraffin, with the orientation of the chopped electroplated magnetic metal carbon fibers randomly distributed. Obtain multiple samples of chopped electroplated magnetic metal carbon fiber composite paraffin material with different thicknesses. Select the sample with the largest thickness from the multiple samples of chopped electroplated magnetic metal carbon fiber composite paraffin material with different thicknesses and name it NIFA, and the sample with the smallest thickness and name it NIFB. Then, measure the electromagnetic parameters of the NIFA and NIFB samples. Input the characteristic dimensions of the electroplated magnetic metal chopped carbon fibers and calculate the depolarization factor of the microwave absorber. ③ By comparing and analyzing the electromagnetic parameters, Cole-Cole, and C0 of NIFA and NIFB, if the electromagnetic parameters do not change abruptly, the number of semicircles representing polarization in the Cole-Cole curve is consistent, and the number of peaks in C0 remains unchanged, then it can be determined that the increase in coating thickness does not introduce a new loss mechanism. ④ Substitute the electromagnetic parameter test results of the NIFA and NIFB samples from step 2 ② into the inverse formula of the Maxwell-Garnett equivalent equation, and substitute the complex permittivity of NIFB. ε m and the NIFA complex permittivity ε of the microwave absorbing agent composite material eff To obtain the intrinsic electromagnetic parameters ε of the electroplated magnetic metal layer. a ; ,in, f It is the volume fraction; N j It is the depolarization factor in all directions; III. Obtaining the intrinsic electromagnetic parameters of a single-phase microwave absorbing material: ① Mix the single-phase microwave absorber with liquid paraffin and disperse it fully to obtain composite samples of microwave absorbers with different volume fractions. Press the obtained composite samples into coaxial ring samples and then measure the dielectric parameters. ② Input the characteristic size, i.e., the dispersion state, of a single-phase absorber and calculate the depolarization factor of the absorber; ③ Substitute the dielectric parameters and depolarization factor of the absorber obtained in steps ① and ② into the Maxwell-Garnett equivalent equation to obtain the intrinsic dielectric parameters of a single-phase absorber; ,in, f It is the volume fraction; N j These are the depolarization factors in various directions; among them, f It is the volume fraction; N j It is the depolarization factor in all directions; IV. Optimization Design Based on Genetic Algorithm and Transmission Line Principle: ① Based on the intrinsic dielectric parameters of the absorbing agent and the intrinsic electromagnetic parameters of other absorbing agents obtained in steps two and three, establish an intrinsic electromagnetic parameter library, and randomly select different absorbing agents from the intrinsic electromagnetic parameter library to form multi-component materials. ②. Using all material parameters in the multi-component composite material from step 4① as independent variables, input them into the Maxwell–Garnett equivalent medium model to calculate the equivalent electromagnetic parameters of the multi-component composite material: ③ Set the evaluation function, including broadband absorption, low-frequency absorption and special frequency bands under different thicknesses as absorption targets, input the electromagnetic parameters obtained in step ①, set the target thickness, and use the transmission line principle to calculate the reflection loss of the multi-component composite material; when the reflection loss is ≤-10dB, it is determined as effective absorption and all effective bandwidths are counted. Finally, save the parameters and reflection loss curves that are close to the target. V. Preparation of multi-component electromagnetic wave absorbing materials: The microwave absorber obtained according to the optimized parameters was mixed with liquid paraffin, and the mixed material sample was finally pressed into a coaxial ring standard sample.

2. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 1, characterized in that... The microwave absorbing agent is a one-dimensional, two-dimensional, or three-dimensional material; the one-dimensional material is a polymer fiber, metal fiber, oxide fiber, ceramic fiber, or carbon fiber; the two-dimensional material is graphite nanosheets or graphene; and the three-dimensional material is nanospheres or hollow microspheres.

3. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 1, characterized in that... Step 2, section 4 describes utilizing the complex permittivity ε of the NiFB composite paraffin matrix. m and the complex permittivity ε of NiFA composite materials eff To obtain the intrinsic electromagnetic parameters ε of the electroplated magnetic metal layer. i The formula for obtaining it is as follows: Where, ε m ε eff ε a These represent the complex dielectric constants of the matrix, the composite material of the microwave absorber, and the intrinsic dielectric constant of the carbon fiber, respectively; f is the volume fraction of the amount added, N j , where is the depolarization factor along the triaxial axis of the fiber inclusion; wherein, the volume fraction of the microwave absorber in the microwave absorber composite material is 1 vol.%.

4. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 3, characterized in that... The intrinsic electromagnetic parameters of a single phase composite material with a volume fraction of 1 vol.% were obtained using the inverse formula of the Maxwell–Garnett equivalent medium model. eff The electromagnetic parameters of the matrix are those of pure paraffin, specifically ε. m ; Calculate the electromagnetic parameters ε of a single-phase absorber. i .

5. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 1, characterized in that... The formula for calculating the depolarization factor of the absorber mentioned in step 3② is as follows: Where a x a y a z These are the semi-major axis and two semi-minor axis lengths of the short-cut electroplated magnetic metal carbon fiber, respectively.

6. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 3, characterized in that... The equivalent electromagnetic parameters of the multi-component composite material are obtained using the Maxwell–Garnett equivalent medium model, with the intrinsic electromagnetic parameters ε of each absorber in the multi-phase composite material as input. i The electromagnetic parameters of the matrix are those of pure paraffin, specifically ε. m ; Calculate the equivalent electromagnetic parameters ε of the multi-component microwave absorber. eff : 。 7. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 1, characterized in that... In step three, the multi-component microwave absorbing agent described in step ① is mixed with liquid paraffin at 60°C and fully dispersed to obtain a composite material sample, and a pure paraffin sample is prepared.

8. The design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 1, characterized in that... The evaluation function described in step 4③ is a piecewise function.

9. A design method for a multi-component electromagnetic wave absorbing material based on machine learning reverse design according to claim 1 or 6, characterized in that... The evaluation function mentioned in step 4.③ is a piecewise function, specifically if RL frq >-10 dB, set V frq = 0; if RL frq < -10 dB, V frq = -1, the piecewise function is changed according to the target absorption frequency band; where, RL frq V represents the reflection loss at each frequency point. frq This is a set value that indicates whether the absorption is effective.

10. The application of a multi-component electromagnetic wave absorbing material based on machine learning reverse design as described in claim 1, characterized in that... It is used as an electromagnetic wave absorbing material in various waveband scenarios; the various wavebands are ultraviolet waveband, infrared waveband or microwave waveband.

Images