Durability Prediction of Structural Composites Using a Continuum Damage Mechanics Approach

Abstract

The main objective of this work is to develop and implement a new analytical model that relies on a continuum damage mechanics approach to predict the evolution of plastic strain and “mechanical” damage until failure in a unidirectional fiber-reinforced composite. The term “damage” is used only in the context of failure mechanisms associated with fracture, which are commonly associated with degradation in stiffness. Plastic strain and damage evolution are related to typical failure mechanisms in composite materials such as fiber, inter-fiber, and intra-fiber fracture.

The plastic strain surface is defined based on the Tsai-Wu failure criterion, while the stiffness degradation damage surface is defined based on the energy release crack growth. The coefficients that characterize the damage and the plastic surfaces are obtained from known material properties. Data obtained from inter-fiber shear load/unload experimental results are used to define the plastic and damage anisotropic associative evolution. The plastic and damage thresholds are obtained by using nonlinear extrapolation. The mathematical equations and physical principles underlying this model are formulated in the tensorial three-dimensional space and tailored to the primary objective of modeling damage evolution.

This model is implemented as a new, user defined material in the commercial finite element analysis software ANSYS. Its results are validated by comparisons with published experimental data from shear load/unload in-plane tests, as well as with published experimental data from load/unload tension tests of a [45°]S composite laminate. The comparison shows a good correlation between the model prediction and the experimental data. Finally, the new material model is implemented in ANSYS to predict the durability of a composite beam subjected to four-point bending, where the evolution of fiber, inter-laminar, and intra-laminar types of damage are quantified.

Research Advisor: Dr. Jacky Prucz
Committee Members: Dr. Ever Barbero, Dr. Bruce Kang, Dr. Samir Shoukry, Dr. Gergis William

Return to event listing.