Numerical studies on using deep sparse autoencoders for damage identification of structures

摘要：

　　This paper proposes using deep sparse autoencoders for the damage identification of structures.Modal information,such as frequencies and mode shapes,are used as input to the deep sparse autoencoders,and the output will be the structural stiffness parameters.Numerical studies on a steel frame structure are conducted to verify the accuracy and performance of using the proposed approach for structural damage identification.Measurement noise effect and uncertainty in the finite element modelling are considered in the identification analysis.Results demonstrate that the proposed approach has a good and robust performance in identifying both the damage locations and severities in structures.

关键词： Sparse Deep neural networks Autoencoders Damage identification Structures

作者: Ling Li Hong Hao Jun Li C.S.N.Nadith Wanquan Liu Ruhua Wang

作者单位: School of Electrical Engineering,Computing and Mathematical Science,Curtin University,Perth,Australi Centre for Infrastructural Monitoring and Protection,School of Civil and Mechanical Engineering,Curt Centre for Infrastructural Monitoring and Protection,School of Civil and Mechanical Engineering,Curt

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: 149-157

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）