Big data in turbulence: unravelling disorder, one vortex at a time

A lunchtime seminar series showing how big data is used, and debated, on campus.

Location: Speakers Corner, Stauffer Library, Queen's University

Ugo Piomelli, Department of Mechanical and Materials Engineering, Queen’s University

Turbulence is everywhere: it affects aircraft drag and fuel consumption, blood flow in arteries, the dispersion of pollution in the air, the formation of weather patterns. Yet no theory has been developed that can explain it, except in the simplest scenarios. Numerical models have become one of the most useful tools to analyze its effects in engineering devices and in the natural environment.

In a numerical model, the history of the fluid velocity, temperature and pressure, is calculated on a grid of very closely spaced points that cover the geometry of interest (a pipe, a turbine blade, a heart valve). Hundreds of millions of points may be necessary, and the history may consist of thousands of snapshots of the flow. Heavyweight computer power is required, and very large amounts of data are generated, that allow the researcher to examine in detail the flow development. Too much detail sometimes: the life of each whorl, vortex or eddy is an open book and hunting for answers in such datasets is as challenging as generating the data itself. Examples of such hunting expeditions will be presented, to show how big data can aid in the understanding of turbulence, and how this understanding can be applied to the real life problems that motivate this work.

Ugo Piomelli holds a Tier 1 Canada Research Chair in Turbulence Simulations and Modelling and the HPCVL-Sun Microsystems Chair in Computational Science and Engineering. He is a Fellow of the Royal Society of Canada, the American Physical Society and the American Society of Mechanical Engineers. 

Everyone welcome!

Part of the BD175 series.