Software Toolset for Rapid Finite Element (FE) Mesh Generation of As-Built Large Laminated Composite Structural Components

Global Engineering and Materials, Inc. (GEM), along with its team members National Institute for Aviation Research (NIAR) at Wichita State University and Sikorsky, proposes to build on the current success of the Composite Bolted and Bonded Analysis Toolkit for Abaqus (CB2ATA) tool for rapid finite element mesh generation of as-built laminated composite structural components. Two NDI techniques based on the X-ray computed tomography (XCT) and optical microscopy will be used for the characterization of volumetric defects (e.g., voids, resin rich zones, ply drops) and interface anomalies (e.g., ply layers, wavy ply interfaces), respectively. Given the low contrast variation and the resulting challenge in determining interface anomalies via XCT, photomicrograph images will be processed to automatically generate the ply interface information. A multi-zoning approach coupled with the classification of gray values will be implemented in conjunction with a machine learning driven pattern recognition algorithm. Defects characterization and NDI data collection will be performed for fabricated composite beams with voids, ply drop, resin rich zones, and ply waviness. Void and ply waviness information are extracted from XCT and optical microscopy, and an efficient image-to-simulation approach will be developed to map the detected voids and ply waviness into the FE based progressive failure analysis model. The static failure prediction of as-build tapered composite beams with measured defects will be compared with testing data under axial tension loading and mixed tension and bending loading