Optimisation of additive-manufactured variable-stiffness composite components

Additive manufacturing has represented a breakthrough in the fabrication processes due to its repeatability, accuracy and capacity of processing large components in a relatively short period of time. Indeed, some of the most extended techniques are Automated Fibre Placement (AFP), Automated Tape Laying (ATL), and most recently 3D printing techniques, such as Fused Deposition Modelling (FDM), are attracting the interest of designers. In the present work, AFP-manufactured components are studied. AFP allows the steering of fibres to generate curvilinear paths- the so-called variable stiffness composites(VSC)- thus permitting to tailor the mechanical response by incorporating additional design variables in comparison to classic straight-fibre composites. This study presents a strategy for the optimisation of VSC plates based on surrogate models and Genetic Algorithms (GA) for the optimisation of this sort of structures. Concerning the structural modelling, Carrera Unified Formulation (CUF) is employed thanks to its inherent capabilities of generating low-fidelity up to layer-wise models in a unified and hierarchic manner. This work shows promising outcomes for the optimisation of this sort of composites, for both pristine and flawed structures.