Active sensing acousto-ultrasound SHM via stochastic time series models

In this work, a statistical damage diagnosis scheme using stochastic time series models in the context of acousto ultrasound guided wave-based structural health monitoring (SHM) has been proposed and its performance has been assessed experimentally. Three different methods of damage diagnosis were employed, namely: i) standard autoregressive (AR)-based method, ii) singular value decomposition (SVD)-based method, and iii) principal component analysis-based method. For estimating the AR model parameters, the asymptotically efficient weighted least squares (WLS) method was used. The estimated model parameters were then used to estimate a statistical characteristic quantity that follows a chi-squared distribution. A statistical threshold derived from the chi-squared distribution that depends on the number of degrees of freedom was used instead of a user-defined margin to facilitate automatic damage detection. The method’s effectiveness is assessed via multiple experiments under various damage scenarios using damage intersecting as well as non-intersecting paths.