Defects characterization, damage mapping, and property evaluation of composites
Lua, Jim Y.
Karuppiah, Anand Vijay
Seneviratne, Waruna P.
MetadataShow full item record
Lua, Jim Y.; Sadeghirad, Alireza; Cui, Xiaodong; Karuppiah, Anand Vijay; Saathoff, Caleb; Seneviratne, Waruna P. 2018. Defects characterization, damage mapping, and property evaluation of composites. 33rd Technical Conference of the American Society for Composites 2018, vol. 4:pp 2943-2956
The presence of voids in the composite laminates can reduce the mechanical properties in composites and these voids can become precursors in the activation of other failure modes such as delamination. Knowing the volume fraction of voids is not enough since their criticality is highly reliant on their location, size, and geometric configuration. While the use of X-ray computed tomography (CT) combined with an advanced image processing and segmentation procedure can extract geometric configurations of voids, the introduction of these voids into a finite element model has posed a great challenge due to the presence of distinct length scales in the description of these voids and evaluation of their effects on the performance of composite components. In order to map an arbitrary distribution of voids into a 3D finite element model, two modeling approaches are developed based on an immersed method for an implicit representation of voids and a computational geometry method to explicitly determine the intersection of a discretized void with its surrounding elements. Based on the intersection of an arbitrarily oriented void with elements and a ply interface, the resulting volume fraction of voids within each element and the surface area at the ply interface is determined and used to compute the initial stiffness degradation and initial delamination area for the subsequent progressive failure analysis. Modeling efficiency and accuracy of both methods are demonstrated using a void with an analytical geometry and a discretized geometry generated from the X-ray CT data.
Click on the DOI link to access the article (may not be free).