CN-116187201-B - Method and system for determining structural parameters of annular embedded double-opening resonant ring

CN116187201BCN 116187201 BCN116187201 BCN 116187201BCN-116187201-B

Abstract

The invention discloses a method and a system for determining parameters of an annular embedded double-split resonant ring structure, wherein the method comprises the following steps of generating an experimental data set based on CST simulation, wherein each group of experimental data in the experimental data set comprises a geometric parameter experimental value and an absorption performance parameter experimental value of the annular embedded double-split resonant ring structure, training a neural network model based on the experimental data set to obtain a trained neural network model, wherein the input of the neural network model is an absorption performance parameter, the output of the neural network model is a geometric parameter, and determining a geometric parameter design value of the annular embedded double-split resonant ring structure corresponding to an absorption performance parameter requirement index based on the trained neural network model. The invention realizes the high-efficiency design of the geometric parameters of the ring embedded double-opening resonant ring structure based on the neural network model, improves the absorption performance of the ring embedded double-opening resonant ring structure, and achieves the aims of absorption and high quality factor of the absorption rate approaching 100%.

Inventors

QU WEIWEI
XIE CHAOHUI
LI GUILIN
DENG HU
LIU QUANCHENG
WU ZHIXIANG
SHANG LIPING

Assignees

西南科大四川天府新区创新研究院
西南科技大学

Dates

Publication Date: 20260512
Application Date: 20230322

Claims (9)

1. The method for determining parameters of the ring embedded double-split resonant ring structure is characterized in that the ring embedded double-split resonant ring structure comprises a metal ring and a double-split resonant ring positioned inside the metal ring, and the method comprises the following steps: Generating an experimental data set based on CST simulation, wherein each group of experimental data in the experimental data set comprises a geometric parameter experimental value and an absorption performance parameter experimental value of a ring embedded double-opening resonance ring structure, the absorption performance parameters comprise a quality factor Q and an absorption rate A, and the geometric parameters comprise the inner diameter of a metal ring Inner side length of split resonant ring And an opening width G; Training a neural network model based on the experimental data set to obtain a trained neural network model, wherein the input of the neural network model is an absorption performance parameter, and the output of the neural network model is a geometric parameter; Determining a geometric parameter design value of the ring embedded double-opening resonance ring structure corresponding to the absorption performance parameter requirement index based on the trained neural network model; the method for determining the geometric parameter design value of the ring embedded double-split resonant ring structure corresponding to the absorption performance parameter requirement index based on the trained neural network model specifically comprises the following steps: Inputting the absorption performance parameter demand index into the trained neural network model to obtain a geometric parameter predicted value output by the trained neural network model; Carrying out CST simulation on the geometric parameter predicted value to obtain an absorption performance parameter simulation value; Determining an error rate of the absorption performance parameter demand index and the absorption performance parameter simulation value; judging whether the error rate is smaller than a preset threshold value or not; If yes, outputting the geometric parameter predicted value as a geometric parameter design value of the ring embedded double-split resonant ring structure corresponding to the absorption performance parameter requirement index; if not, the geometric parameter predicted value and the absorption performance parameter simulation value are used as a group of experimental data to be added into the experimental data set, and the step of training the neural network model based on the experimental data set is returned to, so that the trained neural network model is obtained.
2. The method for determining parameters of a ring embedded double-split resonant ring structure according to claim 1, wherein the loss function for training the neural network model is a mean square error between an output of the neural network model and an experimental value of a geometric parameter.
3. The method for determining the structural parameters of the annular embedded double-split resonant ring according to claim 1, wherein the neural network model comprises an input layer, 5 hidden layers and an output layer, and the node numbers of the 5 hidden layers are 6, 9, 12, 9 and 6 respectively.
4. A ring embedded double split resonant ring structural parameter determination system, characterized in that the system is applied to the method of any one of claims 1-3, the system comprising: The simulation module is used for generating an experimental data set based on CST simulation, wherein each group of experimental data in the experimental data set comprises a geometric parameter experimental value and an absorption performance parameter experimental value of the ring embedded double-opening resonance ring structure, the absorption performance parameters comprise a quality factor Q and an absorption rate A, and the geometric parameters comprise the inner diameter of the metal ring Inner side length of split resonant ring And an opening width G; the training module is used for training the neural network model based on the experimental data set to obtain a trained neural network model, wherein the input of the neural network model is an absorption performance parameter, and the output of the neural network model is a geometric parameter; And the geometric parameter determining module is used for determining geometric parameter design values of the annular embedded double-split resonant ring structure corresponding to the absorption performance parameter requirement indexes based on the trained neural network model.
5. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 3 when executing the computer program.
6. A computer readable storage medium, characterized in that the storage medium has stored thereon a computer program which, when executed, implements the method according to any of claims 1 to 3.
7. The ring embedded double-split resonant ring structure is characterized by comprising a metal ring and a double-split resonant ring positioned inside the metal ring; the geometric parameters of the annular embedded double-split resonant ring structure are determined by the method of any one of claims 1-3.
8. The terahertz metamaterial absorber is characterized by comprising a metal bottom layer, an intermediate medium layer and a metal microstructure layer which are sequentially arranged from bottom to top; The metal microstructure layer comprises a plurality of periodically arranged metal microstructures; The metal microstructure is a circular ring embedded double-opening resonant ring structure; the geometric parameters of the annular embedded double-split resonant ring structure are determined by the method of any one of claims 1-3.
9. The terahertz metamaterial absorber according to claim 8, wherein the thicknesses of the metal microstructure layer and the metal underlayer are each 0.2 μm, the thickness of the intermediate medium layer is 50 μm, the period of the metal microstructure is 100 μm, the outer diameter of a metal ring of a ring embedded double-opening resonant ring structure is 50 μm, and the outer length of the double-opening resonant ring is 60 μm.

Description

Method and system for determining structural parameters of annular embedded double-opening resonant ring Technical Field The invention relates to the technical field of terahertz absorber design, in particular to a method and a system for determining structural parameters of a circular ring embedded double-opening resonant ring. Background In recent years, the artificial neural network (ARTIFICIAL NEURAL NETWORK, ANN) exhibits strong learning ability in terms of computer vision, image recognition, natural language processing and the like, and has the advantages that when the non-intuitive problem is solved, the inherent rule can be searched through a large amount of data, and a good prediction effect is achieved, so that the artificial neural network has great potential in MMA (Microwave metamaterial absorber) design. The neural network is used in the design and optimization of MMA, can effectively improve the design efficiency, and is more beneficial to finding out the optimal result. The metamaterial (METAMATERIAL) is a periodically arranged artificial electromagnetic material, and the special design structure can show characteristics which are not possessed by the common material, such as negative refraction, negative magnetic permeability, negative electric conductivity and the like. One attractive field of application for metamaterials is electromagnetic wave "perfect absorbers". Through reasonable design of the geometric dimension and material parameters of the device, the electromagnetic component of the incident electromagnetic wave can be coupled, so that 100% absorption of the electromagnetic wave in a specific frequency band of the absorber is realized. Disclosure of Invention The invention aims to provide a method and a system for determining parameters of a ring embedded double-split resonant ring structure, so as to realize high-efficiency design of geometric parameters of the ring embedded double-split resonant ring structure and improve the absorption performance of the ring embedded double-split resonant ring structure. In order to achieve the above object, the present invention provides the following solutions: the invention provides a method for determining parameters of an annular embedded double-split resonant ring structure, which comprises a metal annular ring and a double-split resonant ring positioned in the metal annular ring, wherein the method comprises the following steps of: Generating an experimental data set based on CST simulation, wherein each group of experimental data in the experimental data set comprises a geometric parameter experimental value and an absorption performance parameter experimental value of an embedded double-opening resonance ring structure of a ring, wherein the absorption performance parameters comprise a quality factor Q and an absorption rate A, and the geometric parameters comprise an inner diameter r 1 of a metal ring, an inner side length L 1 of an opening resonance ring and an opening width G; Training a neural network model based on the experimental data set to obtain a trained neural network model, wherein the input of the neural network model is an absorption performance parameter, and the output of the neural network model is a geometric parameter; And determining a geometric parameter design value of the ring embedded double-split resonant ring structure corresponding to the absorption performance parameter requirement index based on the trained neural network model. Optionally, the loss function used for training the neural network model is a mean square error of an output of the neural network model and a geometric parameter experimental value. Optionally, the determining, based on the trained neural network model, a geometric parameter design value of a ring embedded double-split resonant ring structure corresponding to an absorption performance parameter requirement index specifically includes: Inputting an absorption performance parameter demand index into the trained neural network model, obtaining a geometric parameter predicted value output by the trained neural network model, and performing CST simulation on the geometric parameter predicted value to obtain an absorption performance parameter simulation value; Determining an error rate of the absorption performance parameter demand index and the absorption performance parameter simulation value; judging whether the error rate is smaller than a preset threshold value or not; If yes, outputting the geometric parameter predicted value as a geometric parameter design value of the ring embedded double-split resonant ring structure corresponding to the absorption performance parameter requirement index; if not, the geometric parameter predicted value and the absorption performance parameter simulation value are used as a group of experimental data to be added into the experimental data set, and the step of training the neural network model based on the experimental data set is returned to