"I love interacting with our industry partners and getting insight into how our fundamental studies help address some of the challenges they face."
Where did you grow up? Were you always interested in your current field?
I was born in Finland but moved to Germany at age four. I was always interested in aeroplanes and rockets and initially wanted to become a pilot. After a while I realized that the most interesting thing to me was the engine that powers the plane. Thus, I decided to study aerospace engineering and specialize in fluid mechanics in order to work on jet engines.
Are you working from home? If so, how’s it going?
Yes, being in Melbourne means that I have worked from home since mid-March. It is going ok, but I miss personal interactions with colleagues and students.
Do you have kids/pets at home that are helping/hindering?
I share space with both my kids, my partner and our cat, Sally. My boys are doing a good job with remote learning and do not need much supervision. Ultimately, they are helping me to keep my sanity. Not sure I can say the same about the cat, though!
What projects are you working on?
My group is involved in a number of projects, all based on exploiting HPC systems in some way. We are working on some very exciting projects that are related to studying the detailed flow physics inside of jet engines. We try to learn from our simulations how jet engines can be made more efficient and reliable. We also use machine learning techniques to help translate the vast amounts of data we generate to models with higher accuracy than current-generation models. This is important as the companies we collaborate with cannot afford to run thousands of design iterations using high-fidelity simulations techniques but have to rely on lower-cost models. We also simulate and predict the noise that flow produces when it moves over aerofoils, and try to reduce that noise by mimicking the silent flyers in the biological world, such as owls.
Underpinning all of the above is further development of our simulation and modelling tools. We use machine learning tools, that are based on evolutionary concepts (survival of the fittest) to create novel models from data. What I really like about this data-driven approach is that the resulting models are symbolic expressions and therefore can easily be interpreted and used in existing simulation codes.
What do you want people to know about the work that you do?
That with the latest generation of supercomputers we are now able to simulate all the complex physical phenomena that occur inside of gas turbines (for aircraft and power generation), at real engine conditions. And that such simulations impact industry in several ways: through insights that enable designers to optimize new technologies and through the generation of gold-standard data sets that can be used to develop better models. Ultimately, our objective is to reduce fuel burn and emissions of gas turbines.
Why do you love what you do?
I love interacting with our industry partners and getting insight into how our fundamental studies help address some of the challenges they face. Fluid flow simulations make for great visualisations: watching movies of the flow we simulate on HPC systems and imagining that this is what happens inside a jet engine when you are hurtling through the atmosphere at 900km/h never fails to excite me. Finally, it is great to see students and other researchers develop new skills and then move on to do fantastic things.
What is one specific thing that you have achieved (research or otherwise) that you are most proud of?
I am definitely proud of having written the code that my team and other researchers internationally are using to perform high-fidelity simulations on some of the world’s fastest supercomputers. It took a long time and had its challenges but seeing it in use and enabling the discovery of phenomena that might help us all fly on more efficient and less polluting planes feels really good.
