Machine learning for transmission spectra prediction on gradient seismic metastructure

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

Organisationseinheiten

Externe Organisationen

  • Tongji University
  • Fudan University
  • Bauhaus-Universität Weimar
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Aufsatznummer109750
FachzeitschriftComputer physics communications
Jahrgang315
Frühes Online-Datum5 Juli 2025
PublikationsstatusVeröffentlicht - Okt. 2025

Abstract

Seismic metastructure based on phononic crystal theory provides a possible solution to accurately manipulating surface acoustic waves. However, the prediction of gradient seismic metastructure for transmission spectra in clay remains a significant challenge due to the damping characteristics of actual soils and practical engineering factors, which become a research hotspot in recent years. Based on finite element analyses and machine learning techniques, this work proposed a data-driven method for building a general prediction model of embedded pillar seismic metastructure with different multi-resonator gradients in the clayed soil. We employed a multilayer perceptron (MLP) model, with the multi-resonator gradients of the metastructure as the input, to predict the transmission spectra. To achieve input standardization, we applied an Autoencoder (AE) to construct a unified representation of the inputs. Due to the inherent non-linearity and variability in soil-structure interactions, the attenuation zones prediction results can only offer approximate engineering applications under specific conditions. By utilizing machine learning, our method achieves better generalization and can be adapted to a wider range of metastructure configurations. This research not only advances the gradient seismic metastructure design framework but also opens new avenues for practical applications in surface acoustic wave management.

ASJC Scopus Sachgebiete

Zitieren

Machine learning for transmission spectra prediction on gradient seismic metastructure. / Rao, Yilin; He, Liangshu; Jin, Yabin et al.
in: Computer physics communications, Jahrgang 315, 109750, 10.2025.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Rao Y, He L, Jin Y, Zhu H, Rabczuk T, Zhuang X. Machine learning for transmission spectra prediction on gradient seismic metastructure. Computer physics communications. 2025 Okt;315:109750. Epub 2025 Jul 5. doi: 10.1016/j.cpc.2025.109750
Download
@article{497bcb934ec44b0ba73a4bd471705720,
title = "Machine learning for transmission spectra prediction on gradient seismic metastructure",
abstract = "Seismic metastructure based on phononic crystal theory provides a possible solution to accurately manipulating surface acoustic waves. However, the prediction of gradient seismic metastructure for transmission spectra in clay remains a significant challenge due to the damping characteristics of actual soils and practical engineering factors, which become a research hotspot in recent years. Based on finite element analyses and machine learning techniques, this work proposed a data-driven method for building a general prediction model of embedded pillar seismic metastructure with different multi-resonator gradients in the clayed soil. We employed a multilayer perceptron (MLP) model, with the multi-resonator gradients of the metastructure as the input, to predict the transmission spectra. To achieve input standardization, we applied an Autoencoder (AE) to construct a unified representation of the inputs. Due to the inherent non-linearity and variability in soil-structure interactions, the attenuation zones prediction results can only offer approximate engineering applications under specific conditions. By utilizing machine learning, our method achieves better generalization and can be adapted to a wider range of metastructure configurations. This research not only advances the gradient seismic metastructure design framework but also opens new avenues for practical applications in surface acoustic wave management.",
keywords = "Gradient metamaterial, Machine learning, Non-hermitian system, Seismic metastructure, Surface acoustic wave",
author = "Yilin Rao and Liangshu He and Yabin Jin and Hehua Zhu and Timon Rabczuk and Xiaoying Zhuang",
note = "Publisher Copyright: {\textcopyright} 2025 Elsevier B.V.",
year = "2025",
month = oct,
doi = "10.1016/j.cpc.2025.109750",
language = "English",
volume = "315",
journal = "Computer physics communications",
issn = "0010-4655",
publisher = "Elsevier BV",

}

Download

TY - JOUR

T1 - Machine learning for transmission spectra prediction on gradient seismic metastructure

AU - Rao, Yilin

AU - He, Liangshu

AU - Jin, Yabin

AU - Zhu, Hehua

AU - Rabczuk, Timon

AU - Zhuang, Xiaoying

N1 - Publisher Copyright: © 2025 Elsevier B.V.

PY - 2025/10

Y1 - 2025/10

N2 - Seismic metastructure based on phononic crystal theory provides a possible solution to accurately manipulating surface acoustic waves. However, the prediction of gradient seismic metastructure for transmission spectra in clay remains a significant challenge due to the damping characteristics of actual soils and practical engineering factors, which become a research hotspot in recent years. Based on finite element analyses and machine learning techniques, this work proposed a data-driven method for building a general prediction model of embedded pillar seismic metastructure with different multi-resonator gradients in the clayed soil. We employed a multilayer perceptron (MLP) model, with the multi-resonator gradients of the metastructure as the input, to predict the transmission spectra. To achieve input standardization, we applied an Autoencoder (AE) to construct a unified representation of the inputs. Due to the inherent non-linearity and variability in soil-structure interactions, the attenuation zones prediction results can only offer approximate engineering applications under specific conditions. By utilizing machine learning, our method achieves better generalization and can be adapted to a wider range of metastructure configurations. This research not only advances the gradient seismic metastructure design framework but also opens new avenues for practical applications in surface acoustic wave management.

AB - Seismic metastructure based on phononic crystal theory provides a possible solution to accurately manipulating surface acoustic waves. However, the prediction of gradient seismic metastructure for transmission spectra in clay remains a significant challenge due to the damping characteristics of actual soils and practical engineering factors, which become a research hotspot in recent years. Based on finite element analyses and machine learning techniques, this work proposed a data-driven method for building a general prediction model of embedded pillar seismic metastructure with different multi-resonator gradients in the clayed soil. We employed a multilayer perceptron (MLP) model, with the multi-resonator gradients of the metastructure as the input, to predict the transmission spectra. To achieve input standardization, we applied an Autoencoder (AE) to construct a unified representation of the inputs. Due to the inherent non-linearity and variability in soil-structure interactions, the attenuation zones prediction results can only offer approximate engineering applications under specific conditions. By utilizing machine learning, our method achieves better generalization and can be adapted to a wider range of metastructure configurations. This research not only advances the gradient seismic metastructure design framework but also opens new avenues for practical applications in surface acoustic wave management.

KW - Gradient metamaterial

KW - Machine learning

KW - Non-hermitian system

KW - Seismic metastructure

KW - Surface acoustic wave

UR - http://www.scopus.com/inward/record.url?scp=105010204918&partnerID=8YFLogxK

U2 - 10.1016/j.cpc.2025.109750

DO - 10.1016/j.cpc.2025.109750

M3 - Article

AN - SCOPUS:105010204918

VL - 315

JO - Computer physics communications

JF - Computer physics communications

SN - 0010-4655

M1 - 109750

ER -

Von denselben Autoren