Damage Identification in Large-scale Double-layer Truss Structures Via a Two-stage Approach

Abstract

In this study, a two-stage damage identification approach based on modal flexibility differences and whale optimization algorithm (WOA) was applied to localize and quantify damages in large-scale double-layer truss structures. In first stage, damage locating vector (DLV) method using EDS (exponential decreased stress) was employed to find the real damaged elements of structure; then, WOA algorithm was used to determine the severity of suspected damaged elements obtained from the first stage. To evaluate the reliability of the proposed approach, two large-scale double-layer truss structures were studied. Furthermore, to assess the effect of noise on the accuracy of damage detection, the article compares the results of EDS with NCE. Calculation results demonstrate that the combination of DLV method using EDS and WOA algorithm provides an effective tool to carefully determine the location and the severity of structural damages in noisy condition directly. Moreover, the approach determines damages even though there are the low number of used mode shapes and a high number of structural elements.