Acoustic transmission modelling in locally periodic structures employing machine learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F24%3A00376593" target="_blank" >RIV/68407700:21230/24:00376593 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Acoustic transmission modelling in locally periodic structures employing machine learning

Popis výsledku v původním jazyce

We are dealing with an acoustic analogy of electronic band structure, and we aim to locate the band gaps using the acoustic transmission model. The acoustic transmission model for locally periodic structures with smooth non-trivial geometries has been done mainly numerically. With the approach of data-driven physics, it is possible to obtain an analytical equation relating axis-symmetric geometry with dispersion relation. A dataset of Bloch phases for a given frequency band and structure geometry parameters was designed for this task. By applying Principal Component Analysis (PCA), we transform the Bloch phases from the dataset into a new coordinate system in a lower dimensional space (while keeping as much variance as possible), where the problem is solved with robust regression analysis. This paper provides an insight into the underlying physics and improves the readability of the system features. The results can be used, e.g., to optimize a design for a desired band gap width

Název v anglickém jazyce

Acoustic transmission modelling in locally periodic structures employing machine learning

Popis výsledku anglicky

We are dealing with an acoustic analogy of electronic band structure, and we aim to locate the band gaps using the acoustic transmission model. The acoustic transmission model for locally periodic structures with smooth non-trivial geometries has been done mainly numerically. With the approach of data-driven physics, it is possible to obtain an analytical equation relating axis-symmetric geometry with dispersion relation. A dataset of Bloch phases for a given frequency band and structure geometry parameters was designed for this task. By applying Principal Component Analysis (PCA), we transform the Bloch phases from the dataset into a new coordinate system in a lower dimensional space (while keeping as much variance as possible), where the problem is solved with robust regression analysis. This paper provides an insight into the underlying physics and improves the readability of the system features. The results can be used, e.g., to optimize a design for a desired band gap width

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10307 - Acoustics

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the International Student Scientific Conference Poster – 28/2024

ISBN

978-80-01-07299-8

ISSN

—

e-ISSN

—

Počet stran výsledku

4

Strana od-do

—

Název nakladatele

Fakulta elektrotechnická

Místo vydání

Praha

Místo konání akce

Prague

Datum konání akce

23. 5. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—