From
Timetable to
Place Amphithéâtre 1Z14, ENS
How can the computational complexity of non-linear solid mechanics be reduced ?
Finite element computations in nonlinear solid mechanics are computationally demanding, particularly for industrial-scale models with a large number of spatial degrees of freedom. Furthermore, computational complexity increases drastically when parametrized investigations are performed, for example for design optimisation, uncertainty quantification, or multi-scale problems in the spatial and temporal domains.
In this presentation, the speaker will share their experience of model-order reduction methods developed within their working group over the past few decades, in close collaboration with LMPS colleagues within the framework of common research training groups. First, there will be a brief overview of different applications, such as stochastic computations and multi-scale methods in the time domain. Then, a systematic approach to rolling tyre mechanics will be presented in detail. In addition to the fundamental modelling approach using the Arbitrary Lagrangian-Eulerian description and the treatment of inelastic material behaviour, the efficient modelling of random excitation from rough-road contact will be discussed.