Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Electrical and Electronic Engineering | en_US |
| dc.contributor.advisor | Lin, Wei (EEE) | en_US |
| dc.creator | Huang, Jieming | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/13884 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | AI-assisted design of Huygens dipole antenna | en_US |
| dcterms.abstract | Aiming at the innovative topic of artificial intelligence-assisted Huygens dipole antenna design, this paper proposes an intelligent optimization method for antenna structure parameters based on deep learning. | en_US |
| dcterms.abstract | Traditional Huygens antenna design mainly relies on engineers' experience and repeated trials, which is not only time-consuming but also difficult to obtain the optimal solution. To overcome this limitation, this study constructed an innovative artificial intelligence-assisted design neural network framework. The complex mapping relationship between the key structural parameters of the antenna and the geometric scaling relationship is learned through a deep neural network. | en_US |
| dcterms.abstract | This paper first uses HFSS electromagnetic simulation software to construct a training data set of a certain scale, which contains multiple sets of antenna structure parameters and their corresponding performance indicators. On this basis, an improved residual neural network structure with attention mechanism is designed, which can predict the influence of antenna parameters on radiation characteristics and the nonlinear relationship between geometric scaling and key parameters. The study shows that the deep learning-assisted design method proposed in this paper constructs a nonlinear mapping relationship between the structural parameters of the Huygens dipole antenna and its electromagnetic performance after geometric scaling. Through the learning ability of the neural network, the study found that there is a predictable correlation between the geometric scaling of the antenna structure and its optimal size. This intelligent prediction model can quickly locate the parameter estimation value close to the optimal solution, greatly reducing the computational complexity of parameter optimization. Especially in the process of structural scaling, the neural network accurately captures the parameter change trend, provides designers with a clearer optimization direction, and effectively avoids the blind search in traditional methods. | en_US |
| dcterms.abstract | This intelligent prediction mechanism based on deep learning not only establishes an efficient mapping channel and framework from the initial design parameters to the optimal solution, but also provides a universal intelligent design paradigm for the structural optimization of Huygens antennas. This study provides an efficient and feasible solution for the engineering design of Huygens antennas. | en_US |
| dcterms.extent | 55 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2024 | en_US |
| dcterms.educationalLevel | M.Sc. | en_US |
| dcterms.educationalLevel | All Master | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 8291.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 2.69 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/13884