Statistic based noise analysis in structural damage detection

Since measurement noise plays a very important role on vibration-based structural damage detection methods, a newly statistic-based noise analysis method is proposed in this paper to investigate the influence of experimental noise of modal data (frequencies and mode shapes) on sensitivity-based damage detection methods, which is verified by Monte Carlo technique. Instead of commonly used perturbation technique, this new technique is directly deduced by the Moore-Penrose generalized inverse of the sensitivity under the differential quotient rule of composite function, this technique can analyze the zero-mean random noise on both frequencies and mode shapes in a similar way. A one-story portal frame is adopted to evaluate the efficiency of the proposed noise analysis technique, and some related issues are also investigated at the same time.