Central goal of the proposed project is to reformulate and apply the previously developed space-time discretization scheme for non-linear, deformable continuum bodies to the field of structural mechanics using continuum degenerated rigid bodies, beams and shells. Most industry related research on dynamical systems originates from the field of multibody dynamics, including classical forward simulation, acoustic investigation or inverse and optimal control problems, among many other.
Due to the widespread use of such industrial simulations, a significant amount of energy and computational resources are consumed, which we can reduce in future. The considered space-time formulation is extremely fast and efficient, allowing for adaptive refinements at certain events and for the use of established parallel solution technologies. Hence, this methodology benefits from state-of-the-art parallel solver and optimization in modern finite element technologies along with the rising core count in modern cpu architectures, reducing the economical and energetic impact of dynamic simulations dramatically.
Initial configuration of the two-ring impact problem (left hand side). Solution without contact (middle) and orthogonal projections (right hand side).
Final configuration with a converged active set, 350 seconds using 100 blocks with each 10 elements in temporal direction, i.e. 0.2 seconds per temporal element.
Group: Prof. Dr. Christian Hesch
More Detail: