Dr John Shrimpton "Adventures in MultiScale Mechanics"
Date: Thu 16 Feb 2012, 11:00 - 11:45
Location: SEMS Seminar Room
Dr John Shrimpton, University of Southampton, School of Eng Sciences
By definition, accurate prediction of non-analytical problems becomes impossible when the scale range of one or more dimensions exceeds the capacity of the computational hardware available. Unfortunately for society, fortunately for us, this is the reality for a depressingly large number of engineering science problems. The solution is usually to investigate a similar problem with a reduced range of scales, or to accommodate the full range only approximately through some sort of model. I tend to use reduced scale results to validate models and make use of experimental data wherever available. The scope for research in this area is vast, and this presentation introduces the three specific areas of expertise I have within this field and the tools I use to undertake such study.
The first area is the study of the interactions between a dispersed phase suspended in and interacting with a turbulent flow. Here the main challenge is the interactions of particles of a certain momentum relaxation timescale with turbulent eddies of similar lifetime. The aim is to develop closures for an Eulerian two fluid model that does not bin the particle diameter dimension and is based on Langevin equations defined for both fluid and particle. Numerical studies of fully resolved spheres, point particles suspended in various isotropic and homogenous turbulence configurations and also experimental data all contribute to model development. The second area is the study of the electrohydrodynamics (EHD) of electrically insulating liquids and here the challenge is the number to electrical and hydrodynamic timescales to consider. The research is primarily empirical and is directed towards developing a range of EHD assisted atomizers for liquid food and combustion oils, though recently some large scale direct numerical simulations of the instability mechanisms have been completed. The third and final area is that of pharmaceutical fluid mechanics, primarily related to powder inhaler design and powder mechanics. One challenge here is to model accurately particle-particle contact forces within a problem space comprised of many thousands of particles. Here the research effort is a mix between empirical studies, often experimental, and also the development and deployment of in house computer codes. These codes are very flexible and I am keen to develop the application of these to other health related areas.
The aim of the talk that this abstract introduces is to give an overview of my research interests, expertise and the tools I deploy to do the job with a view to finding collaborators at Queen Mary. Both this abstract and the presentation will be necessarily brief and if you find the above interesting there is more information on my website, www.pnume.org which I hope you will take the time to have a look at.