High-sensitivity damage detection based on enhanced nonlinear dynamics

Abstract

One of the most important aspects of detecting damage in the framework of structural health monitoring is increasing the sensitivity of the monitored feature to the presence, location, and extent of damage. Distinct from previous techniques of obtaining information about the monitored structure - such as measuring frequency response functions-the approach proposed herein is based on an active interrogation of the system. This interrogation approach allows for the embedding of the monitored system within a larger system by means of a nonlinear feedback excitation. The dynamics of the larger system is then analyzed in state space, and the shape of the attractor of its dynamics is used as a complex geometric feature which is very sensitive to damage. The proposed approach is implemented for monitoring the structural integrity of a panel forced by transverse loads and undergoing limit cycle oscillations and chaos. The nonlinear von Karman plate theory is used to obtain a model for the panel combined with a nonlinear feedback excitation. The presence of damage is modeled as loss of stiffness of various levels in a portion of the plate at various locations. The sensitivity of the proposed approach to parametric changes is shown to be an effective tool in detecting damages. © 2005 IOP Publishing Ltd.