Statistic Strategy of Damage Detection for Composite Structure Using the Correlation Function Amplitude Vector

A statistic method of damage detection based on the Correlation function amplitude Vector (CorV) is proposed to detect local delamination damage of composite structures. In the numerical simulation, finite element models of cantilevered composite laminates for the pre-damage state and post-damage state are firstly established and local delamination damage is simulated, which aims to fetching the acceleration responses signals under concentrated steady random excitation imposed on each of them. Then different nodes are used as the reference points in each CorV calculation for statistic evaluation. The Correlation function amplitude Vector Assurance Criterion (CVAC) is adopted as the damage extent indicator. And the relative changes between the CorVs of the intact and damaged laminate are utilized to determine the delamination damage area roughly. Combined with a statistic evaluation formula, the precise damage can be finally located. The effectiveness and robustness of the proposed statistic detection approach are verified by the damage detection examples for single-damage and multiple-damage detection as well as an experiment on a composite honeycomb sandwich beam with prefab debonding damage, respectively. (C) 2015 Published by Elsevier Ltd.