PhD Oral Exam - Ajeesh Nair, Mechanical Engineering

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

The design and performance of composite structures is significantly influenced by uncertainties in material properties and geometrical features, necessitating reliability quantification and reliability-based design. Traditional structural reliability evaluation methods, however, often involve high computational costs, limiting their practical engineering use. To overcome this shortcoming, a novel methodology that integrates Artificial Intelligence (AI) modeling, Monte Carlo Simulation (MCS), and the Finite Element Method (FEM) to efficiently evaluate the structural reliability of tapered composite tubes under axial and torsional loadings, explicitly accounting for uncertainties in material properties, ply thickness, and taper angle of the tube is presented. An approximate-analytical solution based on the Donnell-Mushtari-Vlasov shell theory is developed to predict axial deformation and angle of twist of the tapered composite tube. An analytical model based on the total potential energy principle and solved using the Rayleigh-Ritz method is developed to compute the critical buckling load of the tapered composite tube. The analytical models developed are used as a reference for validating the finite element model. Additionally, the finite element model and analytical solution are validated against benchmark analytical solutions and experimental results available in the literature, ensuring the accuracy and reliability of the approach. A novel physics-informed dimensionality reduction methodology is developed and implemented to simplify the AI-model architecture and to limit the computational effort while maintaining the accuracy in the structural reliability evaluation. The proposed structural reliability evaluation methodology demonstrates accuracy and computational efficiency in reliability evaluation through comparisons with the direct Monte Carlo Simulation method. Reliability analysis quantifies the influence of random variables on structural response, revealing that designs based solely on mean material properties result in approximately 50% reliability, indicating a 50% probability of failure. The effect of taper angle and stacking sequence on the structural reliability of the composite tubes is characterized. This integrated framework provides a computationally efficient and validated tool for the design-for-reliability of tapered composite tubes, enabling broader applications in composite structural engineering.