Application of a novel time series based method for damage detection, localization and quantification using output only acceleration data

This paper presents preliminary results from an ongoing study for developing a novel approach for damage detection based time series analysis and sensor clusters. The basic idea of this approach is that localized changes in the structural properties of a system can be related to the dynamic response of the structure under ambient vibration if a sensor clustering based approach is used. By creating time series models for these sensor clusters and extracting the damage features using the coefficients of these models, the damage can be detected, located and quantified. In this paper, the theoretical development of the approach and its application to a 4-DOF mass-spring system is presented. The results show that the induced damage can be identified, located and quantified accurately by using the approach. In addition, this approach can also distinguish between the changes caused due to mass variations from the structural damage, which is a very promising property for eliminating the operational effects, e. g. traffic crossing the bridge.