Multi-Scale Adaptive Modelling and Numerical Methods for Reactive Flows

This research project looks to develop computer programs which will enable the study of reactive flows and combustion processes in gas turbine engines. Combustion is inherently a multi-scale process that involves a wide range of complicated physical/chemical phenomena, as well as a wide range of spatial and temporal scales. Due to the limits on available computational resources and the inability to resolve all solution scales for practical configurations, numerical predictions of reactive flows rely heavily on reduced mathematical modelling and sophisticated numerical methods to represent the underlying physics and make the problems of interest more manageable. Unfortunately, current mathematical modelling techniques and numerical solution algorithms are not sufficiently accurate, reliable, nor robust to address the numerous and complex issues associated with high-efficiency and low-emissions combustor design. To remedy this situation, the research team is looking at more accurate and improved multi-scale and adaptive modelling and numerical methods for predicting unsteady turbulent reactive flows in practical combustor geometries. The overarching goal of the research is to use the new and innovative sets of mathematical models and computational tools arising from this research to gain a much better understanding of combustion phenomena and subsequently use this knowledge in the design of more fuel-efficient and green engines.  The industrial partner for this project is Rolls-Royce Canada.

The Globalink student will join and participate in the on-going project outlined above, in which Rolls-Royce Canada is the industrical partner.  The specific focus or role of the Globalink student will involve the study and evaluation of different subfilter-scale models for use in representing the unresolved turbulence-chemistry interactions in large-eddy simulation (LES) of premixed, partially-premixed, and non-premixed turbulent combusting flows.  Several subfilter-scale models will be assessed through comparison of numerical predictions to experimental data for a number of laboratory-scale turbulent flames.  The student will be involved in the development, implementation, and assessment of the LES combustion models.

Pratik Parekh
Faculty Supervisor: 
Dr. Clinton Groth