Tullio Traverso (Alan Turing institute) - Towards the accurate modelling and control of active fluid systems

In recent years, the study of collective dynamics within suspensions of microscopic swimmers has attracted much attention for its fundamental 
relevance to a variety of applications, from targeted drug delivery, through environmental bioremediation, to the optimization of bioprocesses. These systems, also known as active fluids, are driven by a combination of stochastic forcings and deterministic interactions 
among swimmers and the environment. These interactions can be chemically or hydrodynamically mediated, and lead to fascinating collective 
dynamics. During this talk, I will discuss the results obtained by studying such collective dynamics using a continuum kinetic model, 
performing linear stability analysis, and numerical simulations. Specifically, we will first study the intrinsic emergent dynamics of an 
unconfined suspension, and then characterize its response to an impose flow under confinement. Finally, I will introduce a Bayesian machine 
learning (ML) technique for the data-driven surrogate modelling of drop size distributions within liquid sprays. To conclude, we will explore 
the possibility to use such data-driven ML models to improve our understanding of active fluids and optimally design bioreactors.