DAMAGE DETECTION AND DAMAGE DETECTABILITY - ANALYSIS AND EXPERIMENTS

A technique to identify structural damage in real time using limited i nstrumentation is presented. Contrast maximization is used to find the excitation forces that create maximum differences in the response of the damaged structure and the analytical response of the undamaged str ucture. The optimal excitations for the damage structure are then matc hed against a database of optimal excitations to locate the damage. To increase the reliability of the approach under modeling and measureme nt errors, the contrast maximization approach is combined with an appr oach based on changes in frequency signature. The detectability of any particular damage with the proposed technique depends on the ratio of the magnitude of damage and the magnitude of errors in the measuremen ts, as well as on how much the damage influences the measurements. A d amage detectability prediction measure, that incorporates these effect s, is developed. The technique is first tested numerically on a 36 deg ree-of-freedom space truss. To simulate experimental conditions, an ex tensive study is carried out in the presence of noise. A similar truss is then built and the finite-element method (FEM) model of the struct ure is corrected using experimental data. The technique is applied to locate the damage in several members. The experimental results indicat e that this technique can robustly identify the damaged member with li mited measurements and real-time computation. The effectiveness of the damage detection measure is also demonstrated.