EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Advances in 3D-printed acoustic panels: experimental investigations, numerical modelling, and machine learning applications

Ganesh Kekan, Virendra Bhojwani, Sachin S. Pawar, Tanmay Deshpande

Frontiers in Mechanical Engineering · 04.11.2026 · 12

DOI: 10.3389/fmech.2026.1975352 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Mechanical Engineering, online erschienen am 04.11.2026. Verfasst von Ganesh Kekan und 3 weiteren Autor:innen. Die Arbeit nennt 190 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Additive manufacturing provides a flexible route for fabricating acoustic panels with controlled porous, lattice, micro-perforated, resonant, and metamaterial architectures that are difficult to realize using conventional manufacturing. This review combines a bibliometric analysis of Scopus-indexed literature spanning 1980–2026 with a critical synthesis of relevant experimental, analytical, numerical, and machine-learning studies addressing the design, fabrication, modelling, and acoustic performance of 3D-printed structures. Across the reviewed literature, porosity, cell topology, perforation geometry, cavity depth, layer thickness, lattice orientation, airflow resistance, and printing parameters repeatedly emerge as key factors governing sound absorption, particularly in the low- and mid-frequency ranges. Empirical models remain useful for rapid assessment when limited material information, especially airflow resistivity, is available, whereas JCA-type equivalent-fluid models provide a more physically representative description when reliable microstructural transport parameters are known. Transfer-matrix and resonator-based analytical methods are computationally efficient for layered and geometrically regular systems, while finite-element and other numerical approaches are better suited to complex geometries, manufacturing imperfections, and coupled vibro-acoustic behaviour, although at higher computational cost. Machine-learning methods show potential for rapid forward prediction, optimization, and inverse design, but their wider use is currently constrained by limited standardized datasets and experimental validation. The reviewed evidence also shows that additive manufacturing can improve geometric tunability and, in selected designs, reduce material usage or structural mass; however, broader claims of superior performance or sustainability remain application-dependent and should consider fabrication energy, material durability, recyclability, and end-of-life treatment. Major barriers to practical implementation include manufacturing-induced dimensional deviations, inconsistent testing protocols, limited diffuse-field and full-scale validation, material durability, multifunctional design requirements, production scalability, and cost. Future progress therefore requires integrated experimental–numerical–data-driven frameworks, standardized validation procedures, sustainable material selection, and scalable manufacturing strategies.

Zitieren

APA

Kekan, G., Bhojwani, V., Pawar, S. S., & Deshpande, T. (2026). Advances in 3D-printed acoustic panels: experimental investigations, numerical modelling, and machine learning applications. Frontiers in Mechanical Engineering, 12. https://doi.org/10.3389/fmech.2026.1975352

Vancouver

Kekan G, Bhojwani V, Pawar SS, Deshpande T. Advances in 3D-printed acoustic panels: experimental investigations, numerical modelling, and machine learning applications. Frontiers in Mechanical Engineering. 2026;12. doi:10.3389/fmech.2026.1975352

BibTeX
@article{Kekan2026,
  author = {Kekan, Ganesh and Bhojwani, Virendra and Pawar, Sachin S. and Deshpande, Tanmay},
  title = {Advances in 3D-printed acoustic panels: experimental investigations, numerical modelling, and machine learning applications},
  journal = {Frontiers in Mechanical Engineering},
  year = {2026},
  volume = {12},
  doi = {10.3389/fmech.2026.1975352}
}
Kostenlos starten →