Modeling the progression of damage is required to fully describe the behavior of advanced composite materials in engineering applications. However, damage progression can be complex and is often difficult to determine. Errors in analyses can arise due to uncertainties in the material parameters associated with damage progression models. The commercial software Abaqus uses the Hashin damage criterion that consists of six strength based damage initiation material inputs and four energy based damage propagation inputs for composite lamina. The initiation inputs consist of the tensile and compressive strengths parallel and perpendicular to the fiber direction, longitudinal shear strength, and transverse shear strength. The damage propagation properties consist of the fracture-energies that define the stress-displacement relationship for tension and compression of the fibers and the matrix. To create an accurate finite element model, it is important to understand the effects of the material properties on the outputs of the analysis. The research presented in this study will determine the effect of the ten damage properties under a specific loading case using an Abaqus finite element model, with a focus on determining when the four damage progression properties have a significant effect. Edge-notched panels under mode III loading with 20 and 40 ply layups consisting of 30% zero degree plies were considered in the study. The explicit solver in Abaqus was used for the panel analysis. To evaluate the effects of the properties, fractional factorial sensitivity studies were used. Fractional factorials allow for a broad screening of several factors at relatively small computational cost. The factorial design used the ten Abaqus Hashin properties as factors at levels of ±50% from their nominal values. The maximum load the panel experienced was used as the metric for comparison. The effects were then calculated, weighted to the sum of all effects, and plotted to compare each factor. For both the 20 and 40 ply panels, the tensile strength in the direction of the fibers was shown to have the largest effect. The 20 ply panel showed a very small effect of the fracture energy of the fiber in tension, while the 40 ply panel showed a greater effect of this parameter. This is due to damage propagation mainly occurring after max load for thinner panels. Thicker panels are able to transfer load to more plies as damage occurs and the material softens. This allows the panel to carry an increased load after initial damage and through damage progression. Therefore the damage propagation has more of an effect on max load for the 40 ply panels. This principle is illustrated by differences in the experimental load displacement curve shapes of the 20 and 40 ply panels. In addition, the analysis showed the thicker panels exhibited more damage at the maximum load. These results illustrate where in the mode III loading case the damage progression properties have a major effect. This can be used to inform future analysis and inform further research into measuring the damage progression of composite materials.
Finite Element Analysis and Experimental Characterisation of SMC Composite Car Hood Specimens under Complex Loadings
