Learning to inversely design acoustic metamaterials for enhanced performance

Zhang, Hongjia; Liu, Jiawei; Ma, Weitong; Yang, Hu; Yang, Wang; Yang, Haibin; Zhao, Honggang; Yu, Dianlong; Wen, Jihong

doi:10.1007/s10409-023-22426-x

Cited by 6 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Frequency response modeling of periodic and non-periodic phononic structures was achieved by training an INN model that can forward predict high-fidelity samples and learning the inverse mapping with guaranteed reversibility. Zhang et al's [88] team explored the application of ML in the inverse design of AMs by constructing a framework of forward and backward networks (figure 13(b)), taking the target sound absorption curve as an input and outputting a metamaterial structure that satisfies the absorption curve, and evaluating its performance through the forward network. The trained forward network is able to predict the performance of structures beyond the range of the training set with high accuracy and high generalization performance, while the inverse network is able to autonomously adopt the parameters of structures beyond the range for performance optimization.…”

Section: Am Reverse Design Based On DLmentioning

confidence: 99%

Application of machine learning on the design of acoustic metamaterials and phonon crystals: a review

Chen,

Huang,

et al. 2024

Smart Mater. Struct.

View full text Add to dashboard Cite

This comprehensive review explores the design and applications of machine learning techniques to acoustic metamaterials (AMs) and phononic crystals (PnCs), with a particular focus on deep learning. AMs and PnCs, characterized by artificially designed microstructures and geometries, offer unique acoustic properties for precise control and manipulation of sound waves. Machine learning, including deep learning, in combination with traditional artificial design have promoted the design process, enabling data-driven approaches for feature identification, design optimization, and intelligent parameter search. Machine learning algorithms process extensive acoustic metamaterial data to discover novel structures and properties, enhancing overall acoustic performance. This review presents an in-depth exploration of applications associated with machine learning techniques in AMs and PnCs, highlighting specific advantages, challenges and potential solutions of applying of using machine learning algorithms associated with machine learning techniques. By bridging acoustic engineering and machine learning, this review paves the way for future breakthroughs in acoustic research and engineering.

show abstract

Section: Am Reverse Design Based On DLmentioning

confidence: 99%