Executive Summary : | Zero-emission aircraft is a pressing need, driving the development of lightweight high-performance structures and materials. Current constraints on large structural components, such as the upper skin of wing structures, limit their potential for exploitation. Novel manufacturing processes enable variable-stiffness (VS) composites with higher tailoring potential, but the vast design space presents a real optimization challenge. This research aims to develop methods for post-buckling, calculate failure modes, and estimate their impact on stiffened composite panels. Bayesian-based Machine Learning (ML) schemes will be used to map the complex design space and find suitable optimum designs. The proposed framework will solve inverse problems and use a novel finite element Bogner-Fox-Schmit-Castro for efficient modeling of VS panels. The instability problem will be solved using a displacement-based multi-modal formulation of the asymptotic theory. The new failure modes activated by post-buckling will produce additional fatigue damage, modelled according to Socci and Kassapoglou's computationally efficient approach based on crack density. A detailed experimental investigation will validate the developed design framework. 3D printing technology will be used for manufacturing VS composite stiffened panels. |