تحقیقات تجربی در مورد امواج هدایت شده در یک ساختار winglet کامپوزیتی
Abstract
1- Introduction
2- Experimental investigations
3- Numerical investigations
4- Results and discussion
5- Conclusions
References
Abstract
The paper deals with numerical and experimental investigations aimed to develop a Finite Element (FE) model for predicting wave propagation in a blended composite winglet. Material anisotropy, inhomogeneity, multi-layered configuration and complex geometries tend to increase the complexity of the wave propagation phenomena and consequently the development of established FE models. Moreover, even if 2D finite elements seem to be not appropriate for modelling guided waves propagation, especially for complex anisotropic structural components, they are more attractive than 3D ones, because of the computational cost saving. For this reason, part of the presented research activity is addressed to investigate the efficiency of shell finite elements in modelling guided waves propagation in a such complex structure as a winglet. The development of an efficient model depends also on the numerical characterization of the medium within which guided waves propagate through. As a consequence, preliminary simple experimental bending and modal tests have been carried out to support the material properties modelling. Subsequently, guided wave propagation FE analyses were performed and the results compared with provided experimental data. A good agreement between numerical and experimental results of the different analyses has been achieved in terms of both signal time of flight and amplitudes.
Introduction
Guided Lamb waves are widely acknowledged as one of the most encouraging tools for quantitative identification of damages in composite structures and relevant researches were intensively conducted since 1980. They were discovered by Horace Lamb in 1917 [1], as waves propagating in thin plates, and only in 1950 it was formulated a comprehensive theory by Mindlin; experimental work on the same topic were conducted by Schoch in 1952 [2] and Frederick in 1962 [3]. Only in 1961, thanks to Worlton [4], Lamb waves were introduced for the first time as a tool for damage detection, establishing their fundamental use as a prominent non-destructive evaluation tool. Due to their high susceptibility to interferences caused by discontinuities on the propagation path (e.g. in case of damage or boundary edge) and their ability to travel over a long distance with a low power consumption, Lamb waves allow quickly inspecting and monitoring a broad area, even in materials characterized by a high attenuation ratio, such as Glass Fibre Reinforced Polymers (GFRP). In particular, the ability to examine the entire cross-sectional area of the structure, by selecting the proper wave modes, allows detecting both internal and surface damages/defects. The potential damage types that a Lamb wave-based system can detect are summarized by Rose [5]. A Lamb wave-based identification, associated with a damage detection method, provides useful information such as qualitative indication of the occurrence of damage [6-7], quantitative assessment of the position of damage [8-10] and quantitative estimation of its severity; all this information can contribute significantly to the prediction of the residual service life of the component [11]. However, the monitoring of the structural health and the propagation of guided waves in complex structures, such as the winglet proposed in this article, is notoriously very complex [12]. With very high speed, waves reflected from boundaries may easily conceal damagescattered components in the signals. In the same way, the different layup for upper and lower winglet surfaces, as well as the foam in the internal region, the spar stiffener and the doublecurvature skin could induce variations in wave propagation. For these reasons, in order to better understand the physical principles of guided wave propagation within complex structures, numerical simulations are essential. The capability to faithfully reproduce what happens during experiments through numerical simulations could be the key issue in developing innovative diagnostic and prognostic techniques for Structural Health Monitoring (SHM) applications [13].