个人信息Personal Information
教授
博士生导师
硕士生导师
性别:男
毕业院校:大连工学院
学位:博士
所在单位:水利工程系
电子邮箱:zhouj@dlut.edu.cn
A novel transmissibility concept based on wavelet transform for structural damage detection
点击次数:
论文类型:期刊论文
发表时间:2013-09-01
发表刊物:SMART STRUCTURES AND SYSTEMS
收录刊物:SCIE、Scopus
卷号:12
期号:3-4,SI
页面范围:291-308
ISSN号:1738-1584
关键字:structural health monitoring; damage detection; transmissibility; wavelet transform; outlier analysis
摘要:A novel concept of transmissibility based on a wavelet transform for structural damage detection is presented in this paper. The main objective of the research was the development of a method for detecting slight damage at the incipient stage. As a vibration-based approach, the concept of transmissibility has attracted considerable interest because of its advantages and effectiveness in damage detection. However, like other vibration-based methods, transmissibility-based approaches suffer from insensitivity to slight local damage because of the regularity of the traditional Fourier transform. Therefore, the powerful signal processing techniques must be found to solve this problem. Wavelet transform that is able to capture subtle information in measured signals has received extensive attention in the field of damage detection in recent decades. In this paper, we first propose a novel transmissibility concept based on the wavelet transform. Outlier analysis was adopted to construct a damage detection algorithm with wavelet-based transmissibility. The feasibility of the proposed method was numerically investigated with a typical six-degrees-of-freedom spring-mass system, and comparative investigations were performed with a conventional transmissibility approach. The results demonstrate that the proposed transmissibility is more sensitive than conventional transmissibility, and the former is a promising tool for structural damage detection at the incipient stage.