EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Autonomous design of noise-mitigating structures using deep reinforcement learning

Semere B Gebrekidan iD, Steffen Marburg iD

The Journal of the Acoustical Society of America · 01.07.2024 · 156 (1) · 151-163

DOI: 10.1121/10.0026474 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, online erschienen am 03.07.2024. Verfasst von Semere B Gebrekidan und 1 weiteren Autor:innen. Beteiligte Einrichtung: Chair of Vibroacoustics of Vehicles and Machines, Technical University of Munich, Garching 85748, Germany. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Alexander von Humboldt-Stiftung.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper explores the application of deep reinforcement learning for autonomously designing noise-mitigating structures. Specifically, deep Q- and double deep Q-networks are employed to find material distributions that result in broadband noise mitigation for reflection and transmission problems. Unlike conventional deep learning approaches which require prior knowledge for data labeling, the double deep Q-network algorithm learns configurations that result in broadband noise mitigations without prior knowledge by utilizing pixel-based inputs. By employing unified hyperparameters and network architectures for transmission and reflection problems, the capability of the algorithms to generalize over different environments is demonstrated. In addition, a comparison with a genetic algorithm highlights the potential for generalized design in complex environments, despite the algorithms tending to predict local maxima. Furthermore, we examine the impact of hyperparameters and environment types on agent performance. The autonomous design approach offers generalized learning while avoiding restrictions to specific shapes or prior knowledge of the task.

Autor:innen und Institutionen

  • Chair of Vibroacoustics of Vehicles and Machines, Technical University of Munich, Garching 85748, Germany.

ORCID-Profile: Semere B Gebrekidan, Steffen Marburg

Förderung

  • Alexander von Humboldt-Stiftung

Zitieren

APA

Gebrekidan, S. B., & Marburg, S. (2024). Autonomous design of noise-mitigating structures using deep reinforcement learning. The Journal of the Acoustical Society of America, 156(1), 151-163. https://doi.org/10.1121/10.0026474

Vancouver

Gebrekidan SB, Marburg S. Autonomous design of noise-mitigating structures using deep reinforcement learning. The Journal of the Acoustical Society of America. 2024;156(1):151-163. doi:10.1121/10.0026474

BibTeX
@article{Gebrekidan2024,
  author = {Gebrekidan, Semere B and Marburg, Steffen},
  title = {Autonomous design of noise-mitigating structures using deep reinforcement learning},
  journal = {The Journal of the Acoustical Society of America},
  year = {2024},
  volume = {156},
  number = {1},
  pages = {151--163},
  doi = {10.1121/10.0026474}
}
Kostenlos starten →