Structural damage detection with random excitation and correlation function amplitude vector

The principle of structural damage detection based on Correlation Function Amplitude Vector (CorV) is presented in this paper. It is verified that under random excitation with specific frequency spectrum, the CorV of structure only depends on the frequency response matrix of the structure, and the normalized CorV has a specific shape. Therefore with the correlativing and the relative change between the CorVs obtained from the intact and the damaged structure under the same random excitation, the damage in the structure can be detected with the concept of Correlation function amplitude Vector Assurance Criterion (CVAC). Furthermore, based on this damage detecting method, a damage locating method is proposed. To locate damages in the structure, second order difference handling of CorVs are used and it is found that the damaged location will emerge and easy to be identified. This approach is demonstrated with the damage locating examples of a 4 side-clamped plate and a benchmark model of ASCE. The damage detecting and locating results show the feasibility of this method and can be used in the structural health monitoring with ambient excitation.