Deep learning for reduced order modelling of wall bounded, turbulent flows

Simulation (and detailed experiments) of turbulent fluid may be defined as a "data-rich, knowledge-poor" activity: the complexity of underlying physics is so high that it is almost impossible to learn and extract new knowledge from a large amount of the observed data directly. This thesis will focus on data accumulated with high fidelity simulations

to build effective reduced order models set up using a feed forward neural network.

Once a reduced order model is available, one can have rapid estimates of flow field changes as the parameters are changed. This can be used as an environment for reinforcement learning algorithms.