Ultrasonic guided wave based structural damage detection and localization using model assisted convolutional and recurrent neural networks

作者：

Highlights：

• A parallel implementation of reduced-order spectral finite element model is used.

• A model assisted deep learning approach is adopted for damage identification.

• A combined damage detection and localization strategy is proposed.

• CNN and RNN are deployed for damage detection, localization and severity.

摘要

•A parallel implementation of reduced-order spectral finite element model is used.•A model assisted deep learning approach is adopted for damage identification.•A combined damage detection and localization strategy is proposed.•CNN and RNN are deployed for damage detection, localization and severity.

论文关键词：Ultrasonic guided waves,Inverse problem,Spectral element method,Model assisted approach,Damage identification,Deep learning

论文评审过程：Received 20 February 2020, Revised 24 August 2020, Accepted 28 October 2020, Available online 8 November 2020, Version of Record 10 February 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114189