Structural Damage Identification With Performance-Based Uncertainty Quantification

This paper presents an investigation of damage detection of a structure with significant uncertainty. The proposed approach integrates the performance-based uncertainty quantification and feedback control techniques for damage identification. A singular value decomposition (SVD) technique is used to decompose parameter variations into principal components that are weighted with the sensitivity of the performance metric for damage detection. This SVD technique can be used to differentiate between dominant uncertainties and minor uncertainties corresponding to damage identification. To improve the performance of damage identification under uncertainty, a feedback controller is incorporated into the structure to reduce the effect of the quantified uncertainty. The probability framework, based on Monte Carlo simulation, is used to predict the false identification of damage. Examples based on the structural dynamics challenge problem issued by Sandia National Laboratories are used to demonstrate the proposed techniques.