Efficient 3D truss topology optimization for aeronautical structures

Truss lattices are potential candidates for the design of innovative aerostructures, thanks to their high stiffness-to-weight ratio, modularity, and aeroelastic properties. However, when designing ultralight structures, multiple mechanical constraints, such as maximum internal stress or local buckling constraints, must be taken into account since the early design phase. In response to this, a volume minimization problem for 3D structures, subject to multiple load cases, maximum stress, and topological buckling constraints, is formulated in this work. The optimization is solved using a two-step optimization strategy. First, a relaxed formulation is solved by a Sequential Linear Programming algorithm and is used to explore the vast design space of the optimization. During this phase, a heuristic is proposed to reduce the influence of the starting point on the optimized structure. The solution is refined in a second optimization step in which the full non-linear problem is solved using IPOPT, making sure that all the mechanical constraints are respected. The proposed method is validated on multiple two-dimensional classical benchmarks, showing robust behavior with respect to random starting point initializations. Later, the three-dimensional wingbox of the Common Research Model subject to multiple load cases is optimized. The results show that the proposed method can deal with real-sized structures with thousands of candidate members, all while being computationally efficient, optimizing the structure in minutes on a consumer notebook.