| Summary: | Additive Layer Manufacturing (ALM) is growing rapidly due to the unprecedented design freedom. Thus, the structures' complexity can be drastically increased without significant raises in costs. However, the economic viability of ALM is strongly dependent on the full exploration of the referred design freedom. In fact, the ALM is only cost-effective in highly customized parts. Moreover, the mechanical behavior of materials processed via ALM is an ongoing challenge due to defects, uncertainties in material characterization, and verification methods. Thus, the goal of the present work is the development of a robust methodology for the mechanical optimum design of metallic space structures obtained from additive manufacturing. Thus, two main tasks were established. The first task is related to the mechanical characterization of a Ti6Al4V alloy, processed via Selective Laser Melting (SLM). Therefore, an experimental testing campaign of Ti6Al4V samples is presented using homogeneous macroscopic testing (tensile, compression, density, hardness, and fatigue) and microscopic testing (defects detection via microcomputed tomography). These samples show better static properties than the other counterparts, obtained by traditional manufacturing processes. However, the repeatability of the SLM samples is still a challenge (particularly in its fatigue behavior) and more testing is needed. Furthermore, these campaigns are expensive and, consequently, more information per test is required. With the development of full-field measurement methods, material model calibration strategies call upon the use of heterogeneous testing specimens. In the scope of this work, an indirect TO methodology is presented, being capable of designing a wide range of different heterogeneous specimens. Then, a stress states performance indicator is also presented to help the selection of the most promising geometry. The second task is related to the definition of the engineering cycle for ALM structures in its mains phases: (i) design for ALM, (ii) bridging between Topology Optimization (TO) and ALM, (iii) process simulation and structural verification, and (iv) manufacturing. Concerning the first phase, ALM provides great geometric freedom however, there are some design limitations. Therefore, a systematic design methodology is presented, being based on a topology optimization algorithm capable of incorporating the main ALM design limitations (minimum member size and overhang angle). Furthermore, the non-trivial task of bridging between TO and the final smooth geometry is also studied (second phase). The referred task uses a Laplacian smoothing algorithm, which is based on the new concept of mutable diffusion. This new concept shows better properties than the classic algorithms, giving promising results. Furthermore, a new volume constraint is presented, which exhibits a less detrimental impact on the chosen structural indicators. Regarding the remaining phases, these were analyzed via industrial case studies. For instance, process simulation can provide crucial insight into the optimum manufacturing direction and might dictate the difference between success and failure upon manufacturing. The impact of this Ph.D. is related with some improvements in (i) the characterization of ALM-produced materials as well as the geometry of the specimens used for their characterization; and in (ii) the engineering cycle of ALM structures, allowing higher efficiency in the structural solutions for the space industry with lower costs.
|
|---|