Abstract:In traditional metasurface structure design, it heavily relies on electromagnetic simulations to obtain transmission and phase spectral, followed by empirical adjustments. This iterative trial-and-error process, especially when dealing with multi-objective optimization tasks, demands intensive and time-consuming computations, which to a certain extent restricts the development of the metasurface research field. In this paper, a proposed method achieves rapid prediction of spectral responses corresponding to structural units by seeking analytical solutions within the constructed neural network model. The proposed deep learning-based method for predicting transmission and phase spectral of metasurface units consists of metasurface unit dataset construction and a ResNet-based network framework. In the dataset construction approach, an overhead view of the unit structure is extracted and transformed into a binary image, where scaling factors are coupled into the two-dimensional image to increase dimensionality. This enables the representation of different structures such as square pillars, elliptical cylinders, and varying sizes of metasurface units using the same data format, significantly enhancing network generalization. Within the network framework, ResNet is employed to predict the real and imaginary parts of the S-21 parameter, which are then inverted to obtain transmission and phase information. The progressive training method employed in combination with this framework yields high prediction accuracy. The deep learning-based method for predicting transmission and phase spectral of dielectric metasurface units, as revealed in this paper, achieves a 7200-fold increase in prediction speed compared to traditional electromagnetic simulation methods.

On the Application of Deep Learning in Modeling Metasurface

Deep Learning-driven Adaptive Metasurface Transmission Focusing

Symmetry Enhanced Network Architecture Search for Complex Metasurface Design

Leveraging Parallel Speedup for Intelligent Design of Metasurface

Deep Learning-Assisted Design for Metasurfaces with Purified Single Resonant Peak

A Freeform Dielectric Metasurface Modeling Approach Based on Deep Neural Networks

A Deep Learning Approach for Objective-Driven All-Dielectric Metasurface Design

Deep Learning-Based Design Method for Acoustic Metasurface Dual-Feature Fusion

A Deep-Learning Approach for Rapid Prediction of Spectral Responses of Meta-Atoms

Deep Learning: A Rapid and Efficient Route to Automatic Metasurface Design

Meta-Attention Deep Learning for Smart Development of Metasurface Sensors

Deep learning for design metamaterial electromagnetic induction transparent device

Deep learning-based design of broadband GHz complex and random metasurfaces

Physics-Driven Machine-Learning Approach Incorporating Temporal Coupled Mode Theory for Intelligent Design of Metasurfaces

A Novel Modeling Approach for All-Dielectric Metasurfaces Using Deep Neural Networks

Physics-driven unsupervised deep learning network for programmable metasurface-based beamforming

Deep learning for electromagnetically induced transparency (EIT) metasurface optimization design

Deep neural network enabled active metasurface embedded design

Inverse Design of Continuous Domain Bound State All-Dielectric Metasurface Based on Deep Learning

Equivalent Circuit Theory-Assisted Deep Learning for Accelerated Generative Design of Metasurfaces

Pushing the Limits of Multifunctional Metasurface by Deep Learning