Optimization of 3D print strategy for aeronautical components produced by Laser Powder Bed Fusion

Topology Optimization is a numerical procedure that can be used for different purposes. This Master Thesis shows that it can be applied in a novel approach that permits to obtain final design of support structures for models to be 3D printed with AM technologies and in particular with Selective Laser Melting (SLM). Although different commercial software are already available to immediately produce support structures, these cannot be easily used by industries because the absence of a complete and tracked logic under their presence. In fact, mass production must respect certain prerequisites in terms of quality and control of the process, especially when operating in the aeronautic field. The structure of this document is split into three parts. The first one is intended to provide a general overview of the socioeconomic context in which the researches discussed in the subsequent chapters were curried on. The concepts of S-curve and Hype cycle try to explain the innovation, while the comparison between the technologies of Computer Numerical Control (CNC) and Additive Manufacturing (AM) clarify why year by year the latter is becoming an industrial reality establishing its own technological paradigma. It will never substitute the other conventional manufacturing processes but it will grow in parallel with them, in such a way that, in the future, both will be able exploit their competitive advantages. In the second part more details about AM are provided: the author wants to make easily understandable the different needs that AM processes require and that designers have to manage in order to exploit AM competitiveness. In the third part, a novel methodology regarding the use of the numerical approach known as Topology Optimization is described, underlining in particular the strengths and the threats of the mentioned approach. An iterative design process results to be the core of the entire work. It gives the designer the possibility to avoid the use of auto-generate support structures for SLM given as output from commercial software, also reducing the material used and component deformations due to thermal stress with the result of an improved process execution in terms of timing and material wasting. Every execution point of the methodological algorithm described is crucial to find optimal solutions. A practical problem on the manufacturing side is finally revealed, and a preliminary study about a possible solution is proposed.