Lectura de Tesi doctoral - Sr. Alireza Taherzadeh Fard

Composite materials, particularly fiber-reinforced polymers and fiber metal laminates have gained significant attention in industries such as aerospace, automotive, and marine engineering due to their high strength-to-weight ratio, excellent fatigue resistance, and corrosion durability. Despite these advantages, their structural performance is often limited by various damage mechanisms, including fiber failure, matrix cracking, and delamination. Among these, delamination is one of the most critical failure modes, as it can significantly compromise structural integrity. The accurate prediction and modeling of delamination and its interaction with other damage mechanisms remain a major challenge in the field of computational mechanics. This thesis aims to develop a robust constitutive model capable of predicting these failure mechanisms within a homogenized framework, enabling computationally efficient and physically realistic simulations.

The research begins with an extensive evaluation of existing models for intra-laminar and inter-laminar damage mechanisms. Conventional modeling approaches, including stress-strength-based criteria, fracture mechanics methods, and cohesive zone models, are examined to determine their suitability for delamination damage prediction. By utilizing previously developed numerical frameworks within Kratos Multiphysics, the fundamental aspects of damage modeling are explored and benchmarked against experimental data. The key challenge addressed in this thesis is the integration of these theories into a homogenized computational approach, which eliminates the need for explicit layer-wise modeling while preserving accuracy in capturing damage initiation and propagation.

To achieve this, a delamination homogenization theory is developed, allowing for the efficient representation of inter-laminar failure without the computational burden of explicitly modeling each interface. This framework is further extended to cyclic loading conditions, where fatigue-induced damage plays a crucial role in long-term structural degradation. The fatigue model incorporates a damage evolution law calibrated against experimental S-N curves, enabling the accurate prediction of crack growth rates (da/dN) under various loading conditions. The proposed methodology is validated through standardized benchmark test cases, demonstrating its capability in capturing both intra-laminar and inter-laminar damage progression.

The applicability of the developed model is assessed through a series of component-level simulations to ensure its real-world usability. One of the primary case studies involves the open-hole problem, a well-known benchmark in composite mechanics, which is analyzed under both tensile and bending conditions. These simulations help evaluate stress concentration effects and the initiation and growth of delamination cracks. Additionally, a cross-beam member—previously studied with a steel material—is reanalyzed using a composite material, incorporating both delamination and intra-layer damage mechanisms. These case studies illustrate the efficiency and accuracy of the homogenized approach in predicting failure mechanisms under complex loading and geometrical conditions.

The findings of this research contribute to the advancement of composite damage modeling by offering a computationally efficient framework for predicting delamination and intra-laminar damage. The homogenized approach significantly reduces preprocessing complexity while maintaining a high level of accuracy in failure prediction, making it particularly suitable for large-scale structural applications. Future research directions include incorporating fiber bridging effects, extending the framework to dynamic and thermal loading conditions, and utilizing mixed strain/displacement finite elements. Additionally, the application of machine learning techniques to enhance predictive capabilities presents an exciting avenue for further investigation.