A Generative Machine Learning-Based Approach for Inverse Design of Multilayer Metasurfaces
Abstract : The synthesis of a metasurface exhibiting a specific set of desired scattering properties is a time-consuming and resource-demanding process, which conventionally relies on many cycles of full-wave simulations.It requires an experienced designer to choose the number of the metallic layers, the scatterer shapes and dimensions, and the type and the thickness of the separating substrates.Here, we propose a generative machine learning (ML)-based approach to solve this one-tomany mapping and automate the inverse design of dual- and triple-layer metasurfaces.Using this approach, it is possible to solve multiobjective optimization problems by synthesizing thin structures composed of potentially brand-new scatterer designs, in cases where the inter-layer coupling between the layers is non-negligible and synthesis by traditional methods becomes cumbersome.Various examples to provide specific magnitude and phase responses of x- and y-polarized scattering coefficients across a frequency range as well as mask-based responses for different metasurface applications are presented to verify the practicality of the proposed method.
? Existing methods to perform inverse freeform design, ranging from heuristic to gradient-based topology optimization, are not able to effectively solve for the global optimum because the design space for photonic devices is vast and non-convex.
? In all neural-network-based inverse-design methods discussed thus far, which rely on a training set, global optimization is only possible if devices near or at the global optimum are included in the training set.
? Discriminative and generative neural networks can be effective at fitting training data, but they cannot perform meaningful extrapolation tasks beyond the training set
? Due to their increased degrees of freedom, multilayer EMMSs are both promising solutions and difficult to optimize for multiobjective electromagnetic problems.
? Placing the layers closely using thin substrates offers additional degrees of freedom in terms of the order of behavior, but at the same time, it complicates the design further.
? Conventionally, designing and optimizing structures with high interlayer coupling is difficult. It either requires sophisticated equivalent circuit models (ECMs) (and corresponding simulations to find the values of the ECMs), or brute-forcing the problem.
? In general, the solution space to design a multilayer EMMS includes different categories of scatterer shapes.
The proposed framework performs inverse design and optimization mechanism for the generation of meta-atoms and meta-molecules as metasurface units based on DL models and genetic algorithm.
The framework includes consecutive DL models that emulate both numerical electromagnetic simulation and iterative processes of optimization, and generate optimized structural designs while simultaneously performing forward and inverse design tasks.
A selection and evaluation of generated structural designs is performed by the genetic algorithm to construct a desired optical response and design space that mimics real world responses. Importantly, our cyclical generation framework also explores the space of new metasurface topologies.
• The VAE has been previously employed to represent single- and duallayer metasurfaces with negligible interlayer coupling using a continuous latent space .
• However, the optimum latent variable is found through ad hoc random selection instead of an optimization technique.
• Hence, this approach may miss an opportunity to exploit one of the important advantages of the VAE to accelerate optimization.
• we employ a generative model to produce new structures based on the target scattering coefficients in case exploring the gaps in the latent space becomes advantageous.
|