Loading [MathJax]/extensions/tex2jax.js

Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

  • Wenhui Chu
  • Zhuojia Fu
  • S. S. Nanthakumar
  • Wenzhi Xu
  • Xiaoying Zhuang

Organisationseinheiten

Externe Organisationen

  • Hohai University
  • Tongji University

Details

OriginalspracheEnglisch
Seiten (von - bis)547-576
Seitenumfang30
FachzeitschriftInternational Journal of Mechanics and Materials in Design
Jahrgang21
Ausgabenummer3
Frühes Online-Datum17 Apr. 2025
PublikationsstatusVeröffentlicht - Juni 2025

Abstract

The development of phononic crystals provides a possible solution for the precise control of acoustic/elastic waves. Designing phononic crystals with a target characteristic has become a research hotspot in recent years. Nevertheless, the precision with which the acoustic and mechanical waves can be altered remains a major challenge for existing inverse design methods. The rapidly growing machine learning methods revolutionize the design of these materials. As an important branch of machine learning, reinforcement learning is being attempted to solve mechanical problems more intelligently through the interaction of environment and agent. In this paper, we adopt machine learning to successfully design 2D phononic crystals with expected band structure. We firstly applied the meshless generalized finite difference method in solving the dispersion equation for a periodic structure. Then, in order to widen the first-order bandgap width over a desired frequency range, we employ the reinforcement learning algorithm modified by particle swarm optimization to effectively estimate the shape parameters. The parallel technology saves computational costs remains independent of the initial state and target, in addition to being effective and stable. This improved reinforcement learning based interaction design scheme can easily accommodate several other reverse engineering problems.

ASJC Scopus Sachgebiete

Zitieren

Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method. / Chu, Wenhui; Fu, Zhuojia; Nanthakumar, S. S. et al.
in: International Journal of Mechanics and Materials in Design, Jahrgang 21, Nr. 3, 06.2025, S. 547-576.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Chu W, Fu Z, Nanthakumar SS, Xu W, Zhuang X. Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method. International Journal of Mechanics and Materials in Design. 2025 Jun;21(3):547-576. Epub 2025 Apr 17. doi: 10.1007/s10999-025-09749-5
Download
@article{ae364e4cce0644b68e72f09edbf963a8,
title = "Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method",
abstract = "The development of phononic crystals provides a possible solution for the precise control of acoustic/elastic waves. Designing phononic crystals with a target characteristic has become a research hotspot in recent years. Nevertheless, the precision with which the acoustic and mechanical waves can be altered remains a major challenge for existing inverse design methods. The rapidly growing machine learning methods revolutionize the design of these materials. As an important branch of machine learning, reinforcement learning is being attempted to solve mechanical problems more intelligently through the interaction of environment and agent. In this paper, we adopt machine learning to successfully design 2D phononic crystals with expected band structure. We firstly applied the meshless generalized finite difference method in solving the dispersion equation for a periodic structure. Then, in order to widen the first-order bandgap width over a desired frequency range, we employ the reinforcement learning algorithm modified by particle swarm optimization to effectively estimate the shape parameters. The parallel technology saves computational costs remains independent of the initial state and target, in addition to being effective and stable. This improved reinforcement learning based interaction design scheme can easily accommodate several other reverse engineering problems.",
keywords = "Generalized finite difference method, Inverse design, Parallel technology, Particle swarm optimization, Phononic crystal, Reinforcement learning",
author = "Wenhui Chu and Zhuojia Fu and Nanthakumar, {S. S.} and Wenzhi Xu and Xiaoying Zhuang",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive licence to Springer Nature B.V. 2025.",
year = "2025",
month = jun,
doi = "10.1007/s10999-025-09749-5",
language = "English",
volume = "21",
pages = "547--576",
journal = "International Journal of Mechanics and Materials in Design",
issn = "1569-1713",
publisher = "Springer Netherlands",
number = "3",

}

Download

TY - JOUR

T1 - Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method

AU - Chu, Wenhui

AU - Fu, Zhuojia

AU - Nanthakumar, S. S.

AU - Xu, Wenzhi

AU - Zhuang, Xiaoying

N1 - Publisher Copyright: © The Author(s), under exclusive licence to Springer Nature B.V. 2025.

PY - 2025/6

Y1 - 2025/6

N2 - The development of phononic crystals provides a possible solution for the precise control of acoustic/elastic waves. Designing phononic crystals with a target characteristic has become a research hotspot in recent years. Nevertheless, the precision with which the acoustic and mechanical waves can be altered remains a major challenge for existing inverse design methods. The rapidly growing machine learning methods revolutionize the design of these materials. As an important branch of machine learning, reinforcement learning is being attempted to solve mechanical problems more intelligently through the interaction of environment and agent. In this paper, we adopt machine learning to successfully design 2D phononic crystals with expected band structure. We firstly applied the meshless generalized finite difference method in solving the dispersion equation for a periodic structure. Then, in order to widen the first-order bandgap width over a desired frequency range, we employ the reinforcement learning algorithm modified by particle swarm optimization to effectively estimate the shape parameters. The parallel technology saves computational costs remains independent of the initial state and target, in addition to being effective and stable. This improved reinforcement learning based interaction design scheme can easily accommodate several other reverse engineering problems.

AB - The development of phononic crystals provides a possible solution for the precise control of acoustic/elastic waves. Designing phononic crystals with a target characteristic has become a research hotspot in recent years. Nevertheless, the precision with which the acoustic and mechanical waves can be altered remains a major challenge for existing inverse design methods. The rapidly growing machine learning methods revolutionize the design of these materials. As an important branch of machine learning, reinforcement learning is being attempted to solve mechanical problems more intelligently through the interaction of environment and agent. In this paper, we adopt machine learning to successfully design 2D phononic crystals with expected band structure. We firstly applied the meshless generalized finite difference method in solving the dispersion equation for a periodic structure. Then, in order to widen the first-order bandgap width over a desired frequency range, we employ the reinforcement learning algorithm modified by particle swarm optimization to effectively estimate the shape parameters. The parallel technology saves computational costs remains independent of the initial state and target, in addition to being effective and stable. This improved reinforcement learning based interaction design scheme can easily accommodate several other reverse engineering problems.

KW - Generalized finite difference method

KW - Inverse design

KW - Parallel technology

KW - Particle swarm optimization

KW - Phononic crystal

KW - Reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=105002737213&partnerID=8YFLogxK

U2 - 10.1007/s10999-025-09749-5

DO - 10.1007/s10999-025-09749-5

M3 - Article

AN - SCOPUS:105002737213

VL - 21

SP - 547

EP - 576

JO - International Journal of Mechanics and Materials in Design

JF - International Journal of Mechanics and Materials in Design

SN - 1569-1713

IS - 3

ER -