Searcharxiv⌕ Search

arXiv · 2611.15422

Reinforcement learning to design near-perfect acoustic absorbers

Abstract

Reinforcement learning (RL) is proposed as a strategy for the multi-objective design of Helmholtz resonators with near-perfect acoustic absorption. The approach formulates resonator design as a sequential decision-making problem in which geometrical parameters are iteratively updated to balance competing objectives, including resonator volume, resonance frequency, and acoustic reflection. A Q-learning algorithm based on a neural network is developed to explore the design space and identify non-dominated solutions, providing a Pareto front of optimal configurations. The acoustic response is evaluated using a transfer matrix method that incorporates viscothermal losses and acoustic end corrections. Three representative designs are subsequently investigated using full-wave finite element simulations, showing good agreement in their absorption and reflection spectra and validating the accuracy of the reduced-order model. Finally, one optimized resonator is fabricated by three-dimensional printing and experimentally characterized in an impedance tube. The measured absorption coefficient reaches approximately 95\% near the resonance frequency, in agreement with the numerical predictions. These results demonstrate the potential of reinforcement learning for the automated design of compact acoustic absorbers and establish a computationally efficient framework combining multi-objective optimization, numerical modelling, and experimental validation.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

M. Novelli, R. Arnau, L. M. Garcia-Raffi, V. Romero-García. 2026-11-11. Reinforcement learning to design near-perfect acoustic absorbers. https://searcharxiv.com/abs/2611.15422

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

Terahertz emission from $α$-W/CoFe epitaxial spintronic emitters

We report efficient terahertz (THz) generation in epitaxial $α$-W/Co$_{60}$Fe$_{40}$ spintronic emitters. Two types of emitters have been investigated; epitaxial $α$-W$(110)$/Co$_{60}$Fe$_{40}(110)$ and $α$-W$(001)$/Co$_{60}$Fe$_{40}(001)$ deposited on single crystalline Al$_{2}$O$_{3}$($11\bar{2}0$) and MgO($001$) substrates, respectively. First principle calculations of the electronic band structure at the W$(001)$ surface reveal Dirac-type surface states, similar to that reported previously for the W$(110)$ surface. The generated THz radiation is about $10\%$ larger for $α$-W$(110)$/Co$_{60}$Fe$_{40}(110)$ grown on single crystalline Al$_{2}$O$_{3}$($11\bar{2}0$), which is explained by the fact that the $α$-W$(110)$/Co$_{60}$Fe$_{40}(110)$ interface for this emitter is more transparent to the spin current due to the presence of Ångstr\" om-scale interface intermixing at the W/CoFe interface. Our results also reveal that the generation of THz radiation is larger when pumping with the laser light from the substrate side, which is explained by a larger part of the laser light due to interference effects in the film stack being absorbed in the ferromagnetic Co$_{60}$Fe$_{40}$ layer in this measurement configuration.

physics.app-ph↗

Measurement and Modeling of Structure-Induced Surface Scattering on Terahertz Channel

As terahertz (THz) frequencies emerge as promising candidates for next-generation wireless networks, accurate characterization of propagation mechanisms in indoor/outdoor environments becomes essential for system design and performance optimization. This article presents an experimental and theoretical investigation of structure-induced indoor surface scattering on THz channels, examining how material properties and structural configurations jointly govern channel power and angular distribution. Six representative indoor surfaces are characterized, revealing that intrinsic structural inhomogeneity - particularly the quasi-periodic earlywood-latewood arrangement in pine wood - induces measurable angular scattering whose dominant lobes and angular shifts are reproduced by a beam-propagation modeling (BPM) framework. Material-covered surface configurations are further investigated, demonstrating that thin dielectric covering layers can substantially modify reflection characteristics through thickness- and frequency- dependent thin-film interference effects. Wide-angle bistatic measurements in a conference room show that folded curtains broaden the measured angular distribution of detectable received power. These findings suggest that structure-induced surface scattering mechanisms may provide scattering-assisted propagation opportunities for indoor non-line-of-sight THz scenarios.

physics.app-ph↗

Development and Validation of a preCICE Based OpenMOC OpenFOAM Coupling Framework for Nuclear Reactor Analysis

To address the long computation time required by stochastic solvers, which are applied in the coupled neutronics-thermal hydraulics (N-TH) calculations, a faster N-TH coupling framework is developed based on the multiphysics coupling library preCICE, which enables data exchange between OpenMOC and OpenFOAM. As an alternative to commonly used stochastic solvers, like OpenMC, OpenMOC is used as a neutronics solver providing much faster steady-state solution in a Picard iteration. To realize this coupling strategy, OpenMC is used for an initial one-time computation cost for multi-group cross sections data generation that is used throughout OpenMOC calculations. A preCICE library adapter for OpenMOC is also developed, allowing data exchange between neutronic power distributions and thermal hydraulic temperature fields. Our proposed coupling framework is validated through simulations of a simplified 2 times 2 fuel assembly, and a quartered geometry of 17 times 17 fuel assembly. The comparison shows good agreement with established referenced solutions, highlighting our proposed coupling framework for the steady-state N-TH coupling at a fraction of the time required by stochastic solver codes. This advantages could be extended for future applications where fast approximated solutions are required to address the change of N-TH coupling due to small parametric changes like in digital twin of nuclear reactor.

physics.app-ph↗