About us

With cutting-edge numerical approaches, our group looks inside the engines and components central to power generation and transport. Our ultimate aim is to better understand these systems such that we can contribute to their ongoing improvement via increases in efficiency and operability.

The performance and sustainability of these energy systems is dictated by the flow of fluids within them, such as the air flow through a jet engine. These flows are typically turbulent and display a range of complex physical phenomena. This complexity results from harsh operating conditions, including high temperatures, pressures and rotational speeds. Such engine-relevant conditions mean laboratory measurements of these critical fluid flows, upon which society depends for transport and power generation, are difficult, costly and limited. ​

We have therefore developed in-house tools to support the relentless engineering effort to improve the performance, operability and efficiency of these systems. HiPSTAR is our in-house solver for first-principles-based, high-fidelity flow simulation which uncovers previously-unseen flow physics down to the smallest scale. For example, our solver was able to fully resolve engineered micro-textures applied to blades for drag reduction, with dimensions smaller than the width of a human hair. Yet this requires a supercomputer, and is therefore not practical for engineering design iterations. In concert with our high-fidelity studies, our machine-learning tool EVE allows the development of new models via an evolutionary algorithm, which then facilitates analysis suitable for the desktop computer. An advantage of our approach is that resulting models are human-interpretable symbolic expressions, meaning we can deduce key physics from the expressions themselves. ​