Automatic edge detection of ply cracks in glass fiber composite laminates under quasi-static and fatigue loading using multi-scale Digital Image Correlation

Abstract

Digital image correlation is utilized to detect automatically crack density in glass/epoxy laminates with different lay-ups under quasi-static and fatigue loading conditions. The DIC setups used in the present work are 2D-DIC and 3D-DIC which are considered to be at meso-scale and macro-scale, respectively. The main objective of this work is to study whether the DIC technique is reliable in automatic crack detection and functional in calculation of crack density. Therefore, an optimization analysis is implemented to evaluate the resolution and the standard uncertainty in strain and displacement measurements. An optimized compromise is attained between the resolution and the uncertainty of both measurements to decide about the final configuration of the DIC setup. The setup is executed by acquiring in-situ images under quasi-static and fatigue experiments so that the images are synchronized with the data acquired from the testing machine. The crack detection analysis is attained by DIC analysis using strain and displacement fields in order to obtain optimum results. Then a semi-automatic program is created to evaluate the crack density and weighted crack density from the edge and top of the specimen, respectively. Moreover, a comparison is established between both strain and displacement fields in crack characterization. Consequently, the displacement field has proved to deliver better and more accurate results than strain fields in crack density calculation and individual crack characterization. Accordingly, the results obtained from DIC analysis are compared and validated by microscopic images which are acquired after performing each test.