Generative engineering: design optimization of energy-absorbing lattice structures
While the impact of a projectile causes a very localized material stress, the impact of an explosion usually results in a planar material load. In high-speed dynamics, it is generally of great importance to be able to design materials and structures that are capable of absorbing energy while still offering a high residual load-carrying capacity. Metallic 3D printing allows a high degree of design freedom, and especially locally adaptable lattice structures appear to be particularly suitable due to their lightweight construction. To find an optimal design of these complex structures, the use of machine learning and evolutionary algorithms in combination with numerical simulations is being explored.
Potential and challenge of additively manufactured lattice structures for energy absorption
Lightweight cellular lattice structures can exhibit remarkable mechanical properties such as increased specific strength or energy absorption. This makes them particularly attractive for use in crash, impact and blast loading applications. Due to major advances in additive manufacturing techniques, such as laser powder bed fusion, it is now possible to fabricate filigree and complex structures with new materials and tailored material properties. This offers countless opportunities to use function-driven design to improve the mechanical properties of components and structures. However, the possibilities are numerous and difficult to oversee. To exploit the full potential, this requires new optimization methods on the meso- and macroscopic scales.
New design optimization methods for energy-absorbing lattice structures
In this context, Fraunhofer EMI is researching the development of new optimization methods for the design of lattice structures, which can be used for energy absorption under dynamic load scenarios. Thereby, neural networks are used to predict the complex structural behavior and to reduce the number of costly FE simulations. Together with evolutionary algorithms, the structural behavior can be optimized. Furthermore, the optimized structures are investigated in laboratory scale. The results show that a significant improvement of the properties can be achieved.