Multi-patch hybrid optimization of composite wings

The demand for lightweight and structurally efficient composite aerospace structures continues to stimulate the advancement of refined optimization methodologies. This paper proposes a novel multi-patch, multi-objective optimization framework to minimize mass and strain energy in composite wing-box structures. The optimization problem is inherently mixed-integer, as it involves discrete design variables alongside continuous variables, necessitating tailored solution strategies. Two distinct optimization approaches are compared: the traditional multi-objective genetic algorithm NSGA-II and a hybrid method integrating evolutionary and gradient-based optimization. The gradient computations are directly performed at the fundamental nuclei level derived within the Carrera Unified Formulation (CUF), significantly enhancing computational efficiency and accuracy. We provide a scalable low- to high-fidelity methodology for gradient computation, where the CUF framework enables seamless transitions between different levels of accuracy by adjusting the expansion order. Each wing-box segment, modeled as an independent patch, is optimized separately, enabling tailored fiber angle distribution. The effectiveness of the proposed hybrid optimization strategy is demonstrated through numerical case studies, highlighting superior convergence characteristics and improved optimality compared to the NSGA-II standalone approach. The sensitivity of the optimal design solutions to varying structural modeling approaches within the CUF framework is also briefly discussed, providing valuable insights into the practical implementation of multi-patch gradient-enhanced composite structure optimization.