Multi-scale sensing for damage identification

Abstract

Damage identification is important for the lifetime prediction of any structure. In acomposite structure, damage can occur at several material scales from micro-cracking toglobal buckling or delamination. This makes the identification of damage difficultwith a single sensing device. In this paper, we propose to monitor a structuralvolume with an embedded optical fiber sensor network measuring strain, integratedstrain, and strain gradients. Two methods are also compared for data fusion of themulti-scale data in order to determine damage parameters. The first calculates strainmaps directly from the data; the second method uses a neural network. As anexample, an isotropic, homogeneous structural volume with a localized crack ismodeled. The results demonstrate that (a) the multi-scale sensing approach improvesdamage identification and (b) the neural network is a method well adapted forthe multi-scale data fusion and significantly improves the damage identificationcapability.