Multiscale analysis of multi-directional composite laminates to predict stiffness and strength in the presence of micro-defects

Abstract

A methodology is proposed to incorporate spatial distributions of micro-defects within a three dimensional multiscale modeling framework for Carbon Fiber Reinforced Polymer (CFRP) composites to predict stiffness and strength properties of laminates for use in structural design and in establishing nondestructive inspection protocols. Defects in both the fiber and matrix due to variations in the manufacturing process for fiber reinforced composite laminates are considered. Stochastic sampling of micro-defects based on experimental data using Ripley’s K-function is proposed. Three different laminate void volume fractions are used. Spatial statistical analysis is performed for the micro-defects, and then a methodology is presented for predicting stiffness and ultimate strength of three multi-directional composite laminates with distributions of microstructural defects. The results indicate that, in addition to the volume of micro-defects that are present, the spatial distribution of the micro-defects plays a significant role in mechanical properties of laminates, especially the ultimate strength for which a 3.62% reduction is seen in the simulations of laminates containing distributions of micro-defects. Based on these results, it is recommended that the spatial statistics of micro-defects in manufactured composites be incorporated into both design criteria and nondestructive inspection criteria for composite laminates.

Keywords

• Multiscale modeling
• Micro-defects
• Spatial statistics
• Composite laminate