Evaluation of implicit design methods and optimization tools for aeronautical applications

The increasing quest for lightweight, high-performance components in the aerospace industry is driving the adoption of advanced structural design and manufacturing techniques. This thesis investigates the integration of topology optimization and additive manufacturing (AM) with implicit modeling methods, focusing on their potential to improve the design process of aeronautical components. Conducted in collaboration with Avio Aero – a GE Aerospace business – the study compares two distinct optimization workflows: a traditional explicit modeling approach using Altair HyperWorks (OptiStruct), and an implicit modeling approach using nTopology. Emphasis is placed not only on the structural performance and weight reduction achieved in the optimized components, but also on the efficiency of the design process, including workflow automation, geometric smoothness, and post-processing effort. The research includes two test cases: a simplified academic benchmark (a loaded and constrained plate with holes) and a real industrial component (GE9X engine inlet gearbox housing). Both are topology optimized using the explicit and implicit modeling approaches under identical boundary conditions and optimization problem settings (objective function and constraints). Preliminary results suggest that the implicit modeling approach, carried out with nTopology, offers significant advantages in terms of geometric smoothness and workflow efficiency, enabling reductions in design time and cost by leveraging automated smoothing of the raw topology-optimized geometry. However, while the optimization of a single-component case (such as the academic benchmark) provides similar results between the two approaches, the same does not fully apply to the multi-component case (like the GE9X inlet gearbox housing). In fact, in this scenario, although the topology optimization results are comparable, the finite element analysis performed in nTopology appears to overestimate the component stiffness. Overall, implicit modeling proves to be a promising alternative for the development of next-generation aerospace components. The final evaluations will help determine whether adopting an implicit design environment can lead to improved performance and productivity in the industrial context of additive manufacturing.