Vibration based structural damage detection with residual network

摘要：

　　Feature-extraction is the most important process in structural health monitoring(SHM).Due to the great ability of automatic feature-extraction,deep learning based method has attracted increasing attention in SHM field.Although partial success has been obtained by these methods,the potential to achieve better performance still exists.This work use a residual neural network,a powerful deep learning algorithm,to localize damage on a simple supported beam model.Excellent localization accuracy is obtained by directly inputting timedomain signal.

关键词： structural health monitoring residual neural network deep learning

作者: Yi-zhou Lin Hong-wei Ma

作者单位: School of Mechanics and Construction Engineering,Jinan University & Key Lab of 1Disaster Forecast an Dongguan University of Technology,Dongguan,China

会议类型: 国际会议

会议名称: The 7th World Conference on Structural Control and Monitoring(7WCSCM)(第七届结构控制与监测世界大会)

会议地点: 青岛

会议语种:英文

页码: 2397-2404

在线出版日期: 2018-07-22（万方平台首次上网日期，不代表论文的发表时间）