This paper presents a semi-analytical method of suppressing acoustic scattering using reinforcement learning (RL) algorithms. We give a RL agent control over design parameters of a planar configuration of cylindrical scatterers in water. These design parameters control the position and radius of the scatterers. As these cylinders encounter an incident acoustic wave, the scattering pattern is described by a function called total scattering cross section (TSCS). Through evaluating the gradients of TSCS and other information about the state of the configuration, the RL agent perturbatively adjusts design parameters, considering multiple scattering between the scatterers. As each adjustment is made, the RL agent receives a reward negatively proportional to the root mean square of the TSCS across a range of wavenumbers. Through maximizing its reward per episode, the agent discovers designs with low scattering. Specifically, the double deep Q-learning network and the deep deterministic policy gradient algorithms are employed in our models. Designs discovered by the RL algorithms performed well when compared to a state-of-the-art optimization algorithm using fmincon.
Metamaterials are engineered composites that can achieved electromagnetic and mechanical properties that do not exist in natural materials by rearranging their structures. Due to the complexity of the objective functions, it is difficult to find the globally optimized solutions in metameterial design. This talk outlines a gradient-based optimization with generative networks that can search for the globally optimized cloaking devices over a wide range of parameters. The GLO-Net[1] model was developed originally for one-dimensional nano-photonic metagratings is generalized in this work to design two-dimensional broadband acoustic cloaking devices by perturbing positions of each scatterer in planar configuration of cylindrical scatterers. Such optimized cloaking devices can efficiently suppress the total scattering cross section to the minimum at certain parameters over range of wavenumbers. During training each iteration, a generative model generates a batch of metamaterials and compute the total scattering cross section and its gradients using an in-house built multiple scattering MATLAB solver. To evaluate our approach, we compare our obtained results with fmincon in MATLAB. Reference: [1] Jiaqi Jiang and Jonathan A. Fan. Simulator-based training of generative neural networks for the inverse design of metasurfaces. Nanophotonics, 9(5):1059-1069, nov 2019.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.