Two-Stage covariance-based multisensing damage detection method

Different types of sensors in a structural health monitoring (SHM) system installed in a structure enable various types of structural responses to be measured. However, their distinct properties and limitations considerably complicate multisensing structural condition assessment. As a result, the information from these sensors is often used separately, and the potential advantage of multisensing information has not been used effectively. This paper first proposes a covariance-based multisensing (CBMS) damage detection method in the time domain in terms of a CBMS vector as a new damage index and a sensitivity study for damage detection. The proposed method has the merit of assimilating heterogeneous data and reducing the adverse effect of measurement noise. The CBMS damage detection method is then used in two stages for detecting damage location and severity consecutively. Numerical studies are finally performed to investigate the feasibility and accuracy of the proposed framework using an overhanging beam with two damage scenarios. The results show that the two-stage CBMS damage detection method improves the accuracy of damage detection and that the proposed method can be effectively used to combine multisensing information for better damage detection. © 2016 American Society of Civil Engineers.