Progressive Damage in Fibre-Reinforced Composites: Towards More Accurate and Efficient Computational Modelling and Analysis

Abstract

A cursory examination of fracture surfaces of fibre-reinforced composite laminates should convince most observers that failure of composites is generally a complicated event. Progressive damage culminating in final failure or separation of parts is, in fact, a sequence of multi-scale events in composite structures. Statistically sensitive features, such as fibre alignment, arrangements, sizes, voids, defects, etc., may only be explicitly interrogated at the appropriate length scales. However, modelling and predicting progressive failure in composite structures for engineering applications require a balance between computational efficiency and sufficient fidelity to the mechanics and physics of damage and fracture. Current models, even at the coupon level and lower length scales, already contain a large amount of detail. Presently, it is still computationally unrealistic to port all information to the next higher scale of modelling actual structures. Effective and efficient bridging between multiple scales to sift relevant information, from molecular mechanics to micromechanics to coupon-level models, is required. In this chapter, we review recent developments of approaches such as the smeared crack model and more novel ones such as the integrated XFEM-cohesive element (XFEM-CE) and floating node method (FNM), which could potentially be applied to virtual testing of composite structures. Today, challenges of technical nature as well as availability of computational muscle remain, although a tremendous wealth of research has been published and much significant progress has been made.