Smart damage prediction: a distance-to-bifurcation-based approach

Damage detection and prediction is essential for structural health monitoring. Vibration based methods have been used in health monitoring. In this work damage is proposed to be a nonlinear dynamical phenomenon and can be analyzed by utilizing the bifurcation theory. A methodology for predicting failure is proposed which utilizes the concepts of distance to stability boundary as estimated by bifurcation analysis. The proposed methodology is illustrated by developing bifurcation boundary for a two degree of freedom nonlinear mass-spring-damper system. Two damage models are investigated to illustrate the utility the proposed methodology in capturing and estimating the evolution of damage phenomenon.