Time series analysis has been applied to structural monitoring signals for systemdamage identification in a number of research literatures. Among various time seriesanalysis tools, univariate autoregressive modeling (AR) is one of the most commonlyused methods because of its innate computational efficiency. In this paper, threeautoregressive damage features extracted directly from the ambient vibration data andfrom the vibration signal autocorrelation will be presented. Two of the features aredistance functions of AR model parameters and the third feature is a function of ARresiduals. These features are then applied to acceleration measurements collectedfrom a member of a truss bridge to detect a structural change, and their performancesare compared and commented.
展开▼
