An experimental study for damage detection using a hexagonal truss

This paper describes the damage detection study results using a partially instrumented laboratory truss that was tested in several damage conditions. After evaluating several existing model refinement and model reduction/expansion techniques, a two-step damage detection approach was developed for large structures with limited instrumentation. A general area of structural damage is first identified and, subsequently, a specific damaged structural component is located. This approach takes advantage of two different model refinement methods (optimal-update and design sensitivity) and two different model size matching methods (model reduction and eigenvector expansion). Performance of the proposed approach was demonstrated with testing and analysis of a ten-bay hexagonal truss structure. This technology can be applied to structural inspection or health monitoring of spacecraft, aircraft, offshore platforms, oil tankers, bridges, and buildings.