Dr Ati Sharma is Senior Lecturer within Engineering and the Environment at the University of Southampton.
Ati Sharma is currently part of the Aerodynamics and Flight Mechanics group at the University of Southampton.
Biography:
Dr Sharma's research lies at the confluence of control theory and fluid mechanics. He is particularly interested in finding low-order models of turbulent flows, and then finding ways to interact with them to achieve a desired effect. This includes resolvent-based modelling of turbulence, adjoint-based optimisation, nonlinear robust control and probabilistic control of chaotic systems.
Associate Professor, University of Southampton, 10/2014 - Present
Senior Lecturer, University of Southampton, 3/2013 - 10/2014
Lecturer, University of Sheffield, 7/2011 - 3/2013
Junior Research Fellow, Imperial College London, 10/2009 - 7/2011
Research Associate, Imperial College London, 8/2004 - 10/2009
Index Options Flow Trader, JP Morgan Chase, 4/2003 - 5/2004
Quantitative Analyst, EA Capital, 5/2002 – 3/2003
PhD, Imperial College London, 1998 - 2002
MSci (Physics), University College London, 1994 - 1998
Research
Responsibilities
Publications
Teaching
Contact
Research Interests
Turbulence is the chaotic movement of fluids, which is all around us in the form of fast flowing streams, the wind and flow of water past a ship’s hull. A full understanding of turbulence has been described as one of the last unsolved problems of classical physics.
Learning how to eliminate turbulence would will give great benefits. For instance, reducing the turbulence-induced drag on a plane's wing by 30% could save billions of pounds in fuel costs worldwide and associated emissions every year.
More efficient tools for predicting features in turbulent flow could also help with weather prediction and climate modelling.
Low-order models of turbulence
On the modelling side, Dr Sharma has recently published a simple model that successfully predicts persistent structure in turbulent fluid flows. This work is in collaboration with Professor McKeon's group at Caltech.
Hairpin-like structures
The new work describes how wall turbulence can be broken down into constituent blocks that can be simply pieced together, lego-like, to approach and eventually get back to the full equations. When a few blocks, or sub-equations, are added together the results reproduce important features found in laboratory experiments but the calculations can be made on a laptop instead of a supercomputer.
Turbulence control
Dr Sharma is also developing a methodology to controller for turbulent flows. Alongside the widely addressed question of actuation and sensing technology, are the questions of performance specification and control logic.
The generalisation of this theoretical work is currently being pursued in a joint project with Dr Bryn Jones at the University of Sheffield, and tests at higher Reynolds number are under way.
Compliant metamaterial surfaces for turbulence control
Previously, materials with “negative density” would have been thought impossible, however such material properties are now being seriously considered, in applications such as acoustic and visual cloaking.
One possible design
Such metamaterials can be surprisingly simple to implement: one material providing negative effective density consists of metal beads coated in silicone, embedded in an epoxy medium. The obvious benefit of such a passive surface is that it requires no power source and has the potential to be durable. Active metamaterials however can have their response tuned over a wide range of frequencies and wavelengths Dr Sharma is actively researching the suitability of such materials for realistic implementation by metamaterials in flows of practical interest.
Tokamak nuclear fusion reactors
The RZIP tokamak model is suitable for modern control design for tokamaks. The RZIP tokamak model (documented in Dr Sharma’s PhD thesis) is freely available on GitHub.
Grid-based Bayesian estimation
Other research interests
Dr Sharma has published work on the state estimation of non-linear systems with non-Gaussian uncertainty, using grid-based Bayesian filtering methods.
