Description of field of research:

Using hydrogen as a form of stored renewable electrical energy opens several pathways to decarbonise the global economy. To enable this, we need to develop ways of converting the hydrogen back into electricity at the point of end use. One option is to employ gas turbines, which have real-world efficiencies that exceed fuel cells, offer exceptional power densities, and are robust to fuel impurities.

Burning hydrogen in a gas turbine is not trivial, however, due to its very different combustion properties compared to hydrocarbon fuels, namely its high flame speed and short quenching distance, which promote undesirable flashback of flames from their design locations into mixing sections.

A way to avoid this is to employ a two-stage combustion system. In these systems, there is an ultra fuel-lean first stage that is stabilised conventionally and a secondary stage in which fuel or a fuel-air mixture is injected into the hot products of the first stage. In this case autoignition is the likely mode of stabilisation, but this has not been confirmed.

In this project, you will use high fidelity simulations conducted on national supercomputers to try to figure out how hydrogen flames stabilise in a model of a 2-stage combustion system. The project will involve first establishing that the simulations correctly predict the non-reacting flows involved, and as a stretch goal to carry out reacting simulations and describe the difference between methane and hydrogen flames.

School

Mechanical and Manufacturing Engineering

Research areas

Combustion, Hydrogen, Engines

Our group currently consists of 5 PhD students and a postdoc. You will be working very closely with a PhD student Hari whose PhD thesis in connected to this topic, with strong input from Evatt. In this project we are collaborating with GE, the world's leading manufacturer of gas turbines and it is anticipated that there should be an opportunity to share the results with them.

You will learn:

  • a lot about combustion and turbulent flows
  • how to build and run an in house CFD simulation code on our national supercomputers
  • how to conduct post-processing of the resulting large datasets
  • how to critically analyse data and communicate findings

On our side, we hope to gain some understanding of:

  • how well we can predict non-reacting and reacting cases with our in-house code
  • how hydrogen flame stabilise differently to methane ones

We also hope to attract students into a follow on honours project and/or PhD project.

Please contact Hari and Evatt.