Metamaterial Reverse Multiple Prediction Method Based on Deep Learning

Abstract

Metamaterials and their related research have had a profound impact on many fields, including optics, but designing metamaterial structures on demand is still a challenging task. In recent years, deep learning has been widely used to guide the design of metamaterials, and has achieved outstanding performance. In this work, a metamaterial structure reverse multiple prediction method based on semisupervised learning was proposed, named the partially Conditional Generative Adversarial Network (pCGAN). It could reversely predict multiple sets of metamaterial structures that can meet the needs by inputting the required target spectrum. This model could reach a mean average error (MAE) of 0.03 and showed good generality. Compared with the previous metamaterial design methods, this method could realize reverse design and multiple design at the same time, which opens up a new method for the design of new metamaterials.

Inverse design of electromagnetically induced transparency metamaterials based on generative adversarial networks

“…mm, L y : [5,7] mm. Meanwhile, a frequency interval of [2,14] GHz is applied to the electromagnetic field, and simulation is performed using unit cell boundaries and the frequency domain solver based on the finite element method. Since the two ring-shaped portions of this EIT structure produce two bright mode transmission curve spectra, a spectral curve containing three polar points is produced when they are coherently superimposed due to the EIT effect.…”

“…The commonly used generative models are variational auto-encoder [9], generative adversarial networks (GANs) [10], and diffusion model [11,12]. Some researchers have also tried applying the first two generative models to material design with excellent results in the one-to-many design [13,14].…”