Research Assistant (f/m/d, No. 186-25)
up to German pay grade E14 TV-L

The ERC StG BeyondRVE focuses on novel simulation technology for computing the material behavior of materials with a random microstructure. The goal is to build up a groundbreaking multiscale methodology, taking into account the latest neural-network technology and screening the spurious boundary layers which arise for digital volume images of microstructures. Upon completion of the project, a significant boost for the nonlinear mechanics of heterogeneous materials and lightweight design is expected, providing multiscale methods with more expressive results in shorter time for larger classes of materials with higher complexity.

For non-periodic microstructures or when working with non-periodic boundary conditions for the displacement, boundary-layer effects pollute the computed effective properties of random heterogeneous materials. Only low accuracy can be reached, unless gigantic cells are considered.

The aim of the first Post-Doc project is to establish a novel computation methodology to mitigate these problems by screening boundary-layer effects appropriately.

Usually, only the expected mechanical response of a random material is computed in multiscale methods of mechanics. Within the BeyondRVE projects, higher-order statistical quantities will be computed as part of the computational strategy and made available on a macroscopic scale.

The aim of the second Post-Doc project is to establish a novel technology to quantify the microstructure-induced uncertainty using the framework of Deep Material Networks or an extension thereof.