Detection of Fatigue Cracks Using Random Decrement Signatures

Abstract

Damage in structural members usually causes stiffness changes that could be linear or nonlinear. Opening and closing fatigue cracks are usually represented by a bilinear stiffness characteristic. This work describes damage identification approach that is based on the changes in statistical properties of randomdec signatures caused by the onset of nonlinearity. The approach is applied to acceleration data collected from a cracked beam. The results suggest that the method allows to detect the crack if excitation level is high enough. Practical difficulties encountered during implementation are also discussed.