Experimentally verified dual-scale modelling framework for predicting the strain rate-dependent non-linear anisotropic deformation response of unidirectional non-crimp fabric composites

High-performance fiber-reinforced plastic (FRP) composite materials continue to attract growing attention as promising candidates for replacing load-bearing metallic structures for different applications. For example, FRP composites are used in the automotive industry to build car body panels, bumpers, and structural components. They offer excellent impact resistance and reduce vehicle weight, improving fuel economy and passenger safety. In the aerospace industry, FRP composites are used to build aircraft components such as wings, fuselages, and tails. They provide high strength and stiffness while also being lightweight, which helps reduce fuel consumption and increase efficiency. Unfortunately, FRP composites can be more expensive than traditional materials such as steel and aluminum, particularly for low volume applications.

Recently, automated liquid molded composites comprising low-cost carbon fiber unidirectional non-crimp fabrics (UD-NCFs) and rapid curing resins have proven effective in reducing production cycle times and costs, allowing their integration into high-volume production vehicles. Effective use of UD-NCF composites requires strict compliance with vehicle crashworthiness standards, a process that requires high-fidelity strain rate-dependent material models and accurate material properties. While the strain rate-dependent material properties of UD-NCF composites can be measured experimentally, performing physical experiments is expensive and difficult at high strain rates. Additionally, it is challenging to predict the stain rate-dependent response of UD-NCF composites owing to the multiscale material structure. Therefore, more effective approaches are needed.

In their previous study, PhD candidate Khizar Rouf, Professor Michael Worswick and Professor John Montesano from the University of Waterloo developed a dual-scale modeling framework for efficient prediction of the elastic constants of UD-NCF composite material under quasi-static conditions. Herein, in a new study they extended the hierarchical modeling framework to predict the strain rate-dependent non-linear deformation response of a UD-NCF carbon fiber/snap-cure epoxy composite material. Their work is currently published in the peer-reviewed journal, Composite Structures.

In their approach, a numerical model that captures the material heterogeneity at both the microscale and mesoscale was developed. The microscale scale was characterized by an explicit representation of the microstructure of the impregnated tow, whereas the carbon fibers dispersed nonuniformly were treated as rate-independent linear elastic materials. The elastic–plastic response of the epoxy was modelled using a linear Drucker-Prager model where the tension–compression asymmetry of the surface yield was considered. For the mesoscale, the homogenized tow and surrounding matrix material were modelled as non-linear strain rate-dependent materials. Hill’s anisotropic yield function was used to model the elastic–plastic response of the impregnated tow while the Johnson-Cook model was used to capture the strain rate-dependency of the matrix material. The predicted longitudinal tension, transverse tension, and in-plane shear deformation responses were compared with the available experimental data.

The authors showed that the new computational hierarchical framework accurately predicted the strain rate dependent non-linear stress–strain response of a UD-NCF carbon fiber/snap-cure epoxy composite material. Representation of the fiber architecture at both the microscopic and mesoscopic length scales was critical for the study. The predicted impregnated tow exhibited linear elastic behavior in transverse and longitudinal tension and elastic–plastic in-plane and out-of-plane shear response. The in-plane and out-of-plane shear responses predicted using the present framework increased with an increased strain rate, suggesting they were strongly dependent on the applied strain rate.

The development of better FRP composites is important because it allows for the creation of materials that offer improved performance, increased design flexibility, reduced environmental impact, and innovative applications. In summary, the study demonstrated the practical feasibility of a newly developed computational dual-scale modeling framework for predicting the strain rate-dependent non-linear anisotropic deformation response of unidirectional non-crimp fabric composites. The predicted and experimental results were in good agreement. In a statement to Advances in Engineering, the lead and corresponding author Professor John Montesano stated that the proposed dual-scale modeling framework is a promising tool for the effective prediction of the strain rate-dependent non-linear response for FRP composites.