Spectral Element Modeling Of Lamb Wave Propagation And Damage Identification Techniques For Aircraft Structures

Recently, Lamb wave-based structural health monitoring(SHM) technique has been widely studied. This technique makes use of PZT-based sensor array embedded in the structures to transmit/receive Lamb wave signals, which are used to extract damage features and online assess structural health in real time. Based on SHM, the location, level and expand associated to damage are obtained to timely assess the structural integrity, which is significant to enhance flight safety, reduce the maintenance cost and increase aircraft turnout.This dissertion studies modeling techniques of Lamb wave and damage identification strategies. Two novel spectral finite elements are developed to model PZT-induced Lamb wave propagation in structures. An advanced signal processing method and a reference-free method are proposed to image single- and multi-damage. The main contents of this dissertion are as follows:(1) Based on the frequency spectral finite element method(FSEM), the spectral beam elements of both ‘PZT-host structure’ and ‘PZT-adhesive-host structure’ are proposed to model PZT-induced Lamb wave propagation in beam-type structures. Based on the idea of equal displacement on the interface between PZT and host structure, ‘PZT-host structure’ and ‘PZT-adhesive-host structure’ are considered as one hybrid element, respectively, whose dynamic stiffness matrix is derived in the frequency domain. Compared with finite element method(FEM) based on 2D elasticity, the proposed FSEs have been proved to have an accurate result, meanwhile, largely increase the calculation efficiency. In addition, parameters about the PZT length, PZT thickness and adhensive thickness are studied to understand how to influence PZT signals, which may be used for the selection of PZT and adhensive in SHM.(2) Based on the time spectral finite element method(TSEM), 1D spectral beam element of ‘PZT-host beam’ and 2D spectral plate element of ‘PZT-host plate’ are proposed to model PZT-induced Lamb wave propagation in beam-type and plate-type structures, respectively. Based on the idea of equal displacement on the interface between PZT and host structure, ‘PZT-host beam’ and ‘PZT-host plate’ are considered as one hybrid element, respectively, whose dynamic equations are derived upon TSEM. Compared with finite element method(FEM) based on elasticity, the proposed TSEs have been proved to not only be accurate but also increase the calculation efficiency.(3) An advanced signal processing method, fractal dimension(FD), is developed to inspect Lamb wave signals and identify damage in composites. Firstly, box-counting algorithm is used to calculate FDs of signals acquisited by sensor array, and damage indices are obtained by comparing FDs before and after damage. An improved probabilistic reconstruction algorithm of dmage present is developed to intuitively image the damage in composites. Simulation and experimental studies to identify singleand multi-damage in composite plate and stiffened composite panel are carried out to reveal feasibility and effectiveness of FD-based damage identification method.(4) A reference-free method based on similar path is proposed to identify the damage in metal and composite structures. Based on Lamb wave scattering on damage, an idea is about that when one sensing path pass through or near the damge, the first wave pocket of this path will be weakened. Based on this idea, all paths of sensor array are classficated as some similar-path sets. The first wave pocket energy of the paths in one similar-path set are compared with each other to calculate damage indices, which are used to image the damage. The simulation is carried out to identify single- and multi-damage in metal and composite panels, whose results reveal feasibility and effectiveness of similar path-based damage identification method.