A deep learning model for the topological design of 2D periodic wave barriers. (7th July 2021)

• Record Type: Journal Article
• Authors: Liu, Chen‐Xu; Yu, Gui‐Lan
• Is Part Of: Computer-aided civil and infrastructure engineering. Volume 37:Number 4(2022)
• Page Start: 451
• Page End: 469
• Publication Date: 2021-07-07
• Subjects: Civil engineering -- Data processing -- Periodicals; Computer-aided engineering -- Periodicals; 624.0285
• DOI: 10.1111/mice.12743
• Languages: English
• ISSNs: 1093-9687

Abstract

Abstract: This study presents a new deep learning (DL) model to design topological configurations of periodic wave barriers to meet different target frequencies and site conditions. The new DL model is composed of an auto‐encoder (AE) and a conditional variational AE, and 230, 000 sets of data are generated for training, validating, and testing it. The designed results are in good agreement with the targets, and it only takes a few seconds to complete a design. Two barriers are designed by the proposed method for two practical examples, and the vibrations in peak frequency ranges are attenuated. The DL model makes the design of periodic wave barriers smarter and more efficient.