Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Electromagnetic metasurface design method and device based on deep learning

A technology of deep learning and design method, applied in the field of electromagnetic metamaterials, can solve problems such as the actual effect of defect design, theoretical effect, mismatch, etc., to achieve the effect of reducing computational complexity and shortening computational time

Pending Publication Date: 2020-09-01
ZHEJIANG UNIV
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The present invention proposes an intelligent electromagnetic metasurface design method and device based on deep learning, aiming to solve the problem that the actual effect of the electromagnetic metasurface design does not match the theoretical effect caused by the defects of the design method in the prior art

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Electromagnetic metasurface design method and device based on deep learning
  • Electromagnetic metasurface design method and device based on deep learning
  • Electromagnetic metasurface design method and device based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] figure 1 shows a flow chart of the steps of a method for designing an electromagnetic metasurface based on deep learning according to an embodiment of the present application.

[0055] like figure 1 As shown, the deep learning-based electromagnetic metasurface design method of this embodiment specifically includes the following steps:

[0056] Step S101: Construct a deep learning model M1, a deep learning model M2 and a training set.

[0057] In S101, elements of the training set include incident wave information, electromagnetic metasurface structures / arrangements and corresponding optical responses.

[0058] Among them, the samples of the training set are obtained in the following way: First, the electromagnetic simulation software is used to simulate different electromagnetic metasurface structures / arrangements, and the corresponding optical response results are obtained. Then, the optical response results are combined with the electromagnetic metasurface structur...

Embodiment 2

[0072] This embodiment provides an electromagnetic metasurface material. figure 2 A schematic structural diagram of an electromagnetic superunit of an electromagnetic metasurface according to an embodiment of the present application is shown in .

[0073] like figure 2 As shown, the electromagnetic superunit includes an upper dielectric layer 3, an intermediate dielectric layer 2, and a lower dielectric layer 1, specifically:

[0074] The upper dielectric layer 1 is fixed on the upper surface of the intermediate dielectric layer 2, and the upper dielectric layer 3 has a fully polarized resonance structure; the intermediate layer dielectric is a non-metallic layer 2; the lower dielectric layer 1 is a metal layer, and the upper surface of the lower dielectric layer 1 is connected to the The lower surface of the middle dielectric layer 2 is close to and completely matched.

[0075] Wherein, the metal resonant structure may be a metal pattern such as a circle, a ring, an I-sha...

Embodiment 3

[0079] The deep learning-based electromagnetic metasurface design method provided in this application is applicable to various models. This embodiment takes a simple flat model and a flying saucer model with a complex curved structure as examples.

[0080] image 3 A schematic diagram of the structure of a flat plate model covered with an electromagnetic metasurface according to an embodiment of the present application is shown in . Figure 4 shows a schematic structural diagram of a flying saucer model covered with an electromagnetic metasurface according to an embodiment of the present application.

[0081] Figure 5 A flow chart of a method for designing an electromagnetic metasurface based on deep learning according to Embodiment 3 of the present application is shown.

[0082] The present invention is described in further detail with the flying saucer model that is covered with electromagnetic metasurface, as Figure 5 As shown, firstly, step S1 is performed to construc...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention provides an electromagnetic metasurface design method and device based on deep learning. The electromagnetic metasurface design method comprises the steps: carrying outforward prediction of electromagnetic metasurface design according to a deep learning model M1 to obtain a forward mapping relation between electromagnetic metasurface structures / arrangements and corresponding optical responses; and according to the deep learning model M2, carrying out reverse prediction of the electromagnetic super-design to obtain a reverse mapping relationship between the electromagnetic super-surface structure / arrangement and the corresponding optical response. In forward prediction, the defect that a traditional method depends on numerical simulation iteration to solve the problem that a Maxwell equation is complex and time-consuming is overcome, the calculation complexity is greatly reduced, and the calculation time is shortened. In the reverse design, the geometricparameters of the electromagnetic super-unit are effectively discovered and optimized by using the reversely designed deep learning model obtained by training, and the electromagnetic super-surface structure / arrangement under different incident wave information is obtained in real time, so that the electromagnetic super-surface design which is customized by a user and can be strained as requiredis realized.

Description

technical field [0001] The present application belongs to the technical field of electromagnetic metamaterials, and in particular, relates to a deep learning-based electromagnetic metasurface design method and device. Background technique [0002] Electromagnetic metamaterials refer to some composite materials with artificially designed structures and exhibit extraordinary physical properties that natural materials do not have. Electromagnetic metasurfaces refer to artificial layered materials with a thickness smaller than the wavelength. Electromagnetic metasurfaces can be regarded as electromagnetic Two-dimensional counterparts of metamaterials. Electromagnetic metasurfaces, two-dimensional metamaterials with subwavelength structures arranged in a specific way, have many functions that natural materials do not have, and can exhibit many extraordinary physical phenomena that natural materials do not have, such as negative refraction, perfect lenses and invisibility cloaks. ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F30/20G06N3/04G06N3/08G06F111/10
CPCG06F30/20G06N3/08G06F2111/10G06N3/045
Inventor 贾月恬钱超陈红胜
Owner ZHEJIANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com