6533b871fe1ef96bd12d0b92

RESEARCH PRODUCT

Generative design and parametric / topological optimization of cellular structures bio-inspired by additive manufacturing

subject

description

Biomimicry is the practice of learning from nature and imitating its different functionalities. Nature proposes complex forms and objects which inspired designers and engineers to conceive and find solutions for their engineering problems. The fabrication of these complex objects is particularly assured by the different Additive Manufacturing (AM) techniques. Generally, biomimicry can be addressed at different levels, forms, textures, and behaviors, but in AM, it is presented under two main types. The first is the customization of parts (medical prosthesis, implants or custom sport equipments). And the second consists in the optimization for specific properties such as stiffness and lightness (light parts in aerospace or automotive applications). Other types and forms of biomimicry for AM include the incorporation of real biological data, distribution of materials as in cellular and lattice structures, and integrating multifunctionality in design. Particularly, cellular and lattice structures are considered as biomimetic due to their resemblance to biological structures. The utility of their integration in optimization and AM techniques has been proven by several studies mainly in terms of weight reduction, stiffness, and energy absorption rate increase. Taken the fact, that a link exists between biomimicry, cellular structures, optimization, and AM, the following question arises: “What is the utility of biomimicry in optimizating and manufacturing cellular parts that respond to mechanical properties and constraints (lightweight and high stiffness) imposed by AM?”. To be able to answer this question, this research work will focus on highlighting the importance of employing biomimetic algorithms and forms in designing and optimizing lightweight cellular structures with high stiffness. The first contribution focuses on studying the importance of combining topology and parametric optimizations in designing and modeling cellular structures having a good stiffness-to-weight ratio. The study is divided into two cases and special interest is paid to the comparison between the two cases. In both cases, a Design Of Experiments (DOE) and a sensitivity study are conducted. The first case consists of a uniform lattice distribution while the second is a variable-density lattice distribution study. It is shown that the second case provids a better weight and a better strength thus reaching the goal of having a better strength-to-weight ratio. However, the biomimetic aspect is not covered in the first contribution . Thus, in a second part, the Lindenmayer systems (L-systems) – tree growth grammars – are used and a specific distribution is used along the Principal Stress Lines (PSLs) directions imitating the material growth aspect in bone structures. Numerical simulations and parametric optimization schemes based on an L9 DOE sensibility study were conducted. Results show the effectiveness of this method in adapting mechanical structures to various loading cases as well as guaranteeing a good stiffness-to-weight ratio. In the third contribution, a more advanced parametric optimization approach is applied in order to find better results for optimal L-systems beams’ sections: meta-modeling optimization algorithm based on metaheuristics and knowledge.