کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
250890 | 502912 | 2016 | 8 صفحه PDF | دانلود رایگان |
Typically, numerical simulations of Lamb wave propagation are done using material properties which originate from tensile testing. This approach is well established in relation to isotropic homogenous structures such as aluminium plates. However if this approach is used for woven composites such as carbon fibre reinforced plastics (CFRP), inaccuracies can arise that stem from vastly different stress distributions, strain rates and amplitudes during Lamb wave propagation. In order to account for this, an approach is presented where the elastic properties in a numerical Lamb wave model are optimised to achieve good correlation between model predictions and experimental observations. Since the material properties are determined under a Lamb wave propagation regime, the strain rates and amplitudes are consistent with the intended modelling application. The approach is validated with an experimental case study involving a M18/G939 carbon-epoxy system. The methodology is shown to yield property estimates that furnish simulations that closely match observed behaviours. The optimised properties were significantly different to those supplied by the manufacturer, as much as 52% for the in-plane stiffness. The findings demonstrate that large errors are possible if elastic properties determined using conventional quasi-static testing are used in Lamb wave simulations pertaining to woven composite materials.
Journal: Composite Structures - Volume 141, 1 May 2016, Pages 24–31