The present work addresses the design of structure-preserving numerical methods that emanate from the general equation for non-equilibrium reversible-irreversible coupling (GENERIC) formalism. Novel energy-momentum (EM) consistent time-stepping schemes in the realm of molecular dynamics are proposed. Moreover, the GENERIC-based structure-preserving numerical methods are extended to the context of large-strain thermoelasticity and thermo-viscoelasticity.

This work proposes a new numerical approach for analyzing the behavior of fiber-reinforced materials, which have gained popularity in various applications. The approach combines theories and methods to model the fracture behavior of the polymeric matrix and the embedded fibers separately, and includes a modified plasticity model that considers the temperature-dependent growth of voids. Tests are conducted to explore different types and sequences of failure in long fiber-reinforced polymers.

Proposed in the early 1990s, the enhanced assumed strain (EAS) method is one of the probably most successful mixed finite element methods for solid mechanics. This cumulative dissertation gives a comprehensive overview of previous publications on that method and covers recent improvements for EAS elements. In particular, we describe three key issues of standard EAS elements and develop corresponding solutions.

This work is about the inverse dynamics of underactuated flexible mechanical systems governed by quasi-linear hyperbolic partial differential equations subjected to time-varying Dirichlet boundary conditions that are enforced by unknown, spatially disjunct, hence non-collocated Neumann boundary conditions.

The present work deals with the characterisation and multi-scale modelling of the large-strain response of ternary polymer blends. In a homogenised constitutive modelling approach, particularly the deformation behaviour featuring plastic dilatancy is investigated. Concerning the micromechanical modelling, constitutive models are proposed for the blends' individual phases and compared regarding their capabilities to capture the composition-dependent fracture toughness in unit cell models.