Local Damage Detection with the Global Fitting Method Using Mode Shape Data in Notched Beams

Abstract

Damage in a structure alters its dynamic characteristics such as frequency response functions and modal parameters. The present study extends the ‘gapped smoothing method’ for identifying the location of structural damage in a beam by introducing the ‘global fitting method.’ The procedure uses only the mode shape data obtained from a damaged structure with an assumption that the undamaged structure is homogeneous and uniform and the damage size is small. The sensitivity of damage detection algorithm is evaluated using a finite element analysis (FEA) of a few beams having a notch. Structural irregularity index (SSI) was used to identify the locations and size of damage. The ability to detect damage was enhanced by averaging SSI over a few modes. A statistical procedure was applied to identify damage with respect to background noise. A methodology and quantitative criteria was developed to select the optimum excitation grid spacing. Numerical results showed that the present method can detect both narrow (13 mm width) and wide damage (126 mm width) associated with less than 3% local thickness reductions. Experimental results validated the numerical results and detected the depth to thickness ratio about 41% and 35% for the wide and narrow notch beams, respectively. The present method showed improved resolution on detecting the location and size of damage in a beam over the previous methods using mode shape data reported in literature.