Wenhui Chu, Zhuojia Fu, S. S. Nanthakumar, Wenzhi Xu, Xiaoying Zhuang
{"title":"基于机器学习与局部配置无网格法的声子晶体智能反设计","authors":"Wenhui Chu, Zhuojia Fu, S. S. Nanthakumar, Wenzhi Xu, Xiaoying Zhuang","doi":"10.1007/s10999-025-09749-5","DOIUrl":null,"url":null,"abstract":"<div><p>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.</p></div>","PeriodicalId":593,"journal":{"name":"International Journal of Mechanics and Materials in Design","volume":"21 3","pages":"547 - 576"},"PeriodicalIF":3.6000,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method\",\"authors\":\"Wenhui Chu, Zhuojia Fu, S. S. Nanthakumar, Wenzhi Xu, Xiaoying Zhuang\",\"doi\":\"10.1007/s10999-025-09749-5\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>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.</p></div>\",\"PeriodicalId\":593,\"journal\":{\"name\":\"International Journal of Mechanics and Materials in Design\",\"volume\":\"21 3\",\"pages\":\"547 - 576\"},\"PeriodicalIF\":3.6000,\"publicationDate\":\"2025-04-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Mechanics and Materials in Design\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s10999-025-09749-5\",\"RegionNum\":3,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, MECHANICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Mechanics and Materials in Design","FirstCategoryId":"88","ListUrlMain":"https://link.springer.com/article/10.1007/s10999-025-09749-5","RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, MECHANICAL","Score":null,"Total":0}
Intelligent inverse design of phononic crystals based on machine learning coupled with localized collocation meshless method
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.
期刊介绍:
It is the objective of this journal to provide an effective medium for the dissemination of recent advances and original works in mechanics and materials'' engineering and their impact on the design process in an integrated, highly focused and coherent format. The goal is to enable mechanical, aeronautical, civil, automotive, biomedical, chemical and nuclear engineers, researchers and scientists to keep abreast of recent developments and exchange ideas on a number of topics relating to the use of mechanics and materials in design.
Analytical synopsis of contents:
The following non-exhaustive list is considered to be within the scope of the International Journal of Mechanics and Materials in Design:
Intelligent Design:
Nano-engineering and Nano-science in Design;
Smart Materials and Adaptive Structures in Design;
Mechanism(s) Design;
Design against Failure;
Design for Manufacturing;
Design of Ultralight Structures;
Design for a Clean Environment;
Impact and Crashworthiness;
Microelectronic Packaging Systems.
Advanced Materials in Design:
Newly Engineered Materials;
Smart Materials and Adaptive Structures;
Micromechanical Modelling of Composites;
Damage Characterisation of Advanced/Traditional Materials;
Alternative Use of Traditional Materials in Design;
Functionally Graded Materials;
Failure Analysis: Fatigue and Fracture;
Multiscale Modelling Concepts and Methodology;
Interfaces, interfacial properties and characterisation.
Design Analysis and Optimisation:
Shape and Topology Optimisation;
Structural Optimisation;
Optimisation Algorithms in Design;
Nonlinear Mechanics in Design;
Novel Numerical Tools in Design;
Geometric Modelling and CAD Tools in Design;
FEM, BEM and Hybrid Methods;
Integrated Computer Aided Design;
Computational Failure Analysis;
Coupled Thermo-Electro-Mechanical Designs.