Robust Diagnostics for SHM

In Structural Health Monitoring, Kalman filters can be used as means for fusing data and models for the purpose of diagnosis and prognosis. To ensure robust performance, it is necessary to tune the filters' process and measurement noise covariance matrices, which are unknown in practice.

In this paper, which forms part of the academic visit by Stefan Wernitz of LUH to our SMM group, an autocovariance least-squares method is proposed for automate tuning and cast the problem in infinite horizon for rendering estimation robust. We then use the proposed model for damage analysis, by introducing a damage indicator that features a high sensitivity towards localized damage. We demonstrate the efficacy of the proposed schemes for noise covariance estimation and damage analysis in a series of simulations inspired by a laboratory test. We finally offer experimental validation, based on vibration test data of a cantilever beam featuring damages at multiple positions, where high sensitivity towards small local stiffness changes is achieved. In our investigations, we compare the damage detection and localization performance of Kalman and H-infinity  filters as well as differences in mode shape curvatures (MSC). The study shows that a combined application of both estimators can lead to increased robustness and sensitivity regarding damage detection and localization.

read more in external page Volume 170 of MSSP