Using computers to simulate the way that fluids move
My name is Joe O'Connor and I am a second year PhD student in Aerospace Engineering here at Manchester. In 2009 I started my first degree at the University of Glasgow in Scotland. During this time I was able to take part in an exchange programme which allowed me to go and study at the University of California, Irvine for one year. Also as part of my undergraduate degree I had the opportunity to work at Rolls-Royce for six months helping to design new aircraft engines. Upon completion of my Masters degree I then moved to Manchester to start my PhD research.
My research is focussed on the field of Computational Fluid Dynamics, or CFD for short. All this means is using computers to simulate the way that fluids move. At this point it is important to understand what exactly a fluid is. When we talk about fluids we usually think of liquids, however gasses are also fluids as well (gasses can flow!). This means the air we breathe, the water we drink, the blood going through our body, and the fuel in our cars are all fluids. Because fluids are literally everywhere it is very important to understand exactly the way fluids behave in certain situations – this allows us to design better aeroplanes, wind turbines, or even artificial hearts. The focus of my research is developing software which will allow us to do this in a better way than what we already are.
Understanding the way that fluids (such as air) move is very important for a number of reasons – Formula 1 teams spend a lot of time and money doing this to make sure their cars are as aerodynamic as possible, as do aeroplane manufactures. However, the really difficult thing about this is that the equations that tell us how fluids move are very long and very complicated – and therefore very difficult to solve. In fact, to this day no one has actually ever been able to solve them exactly and that is why they are one of the 7 Millennium Prize Problems. What that means is that if you find out a way to solve them exactly then someone will give you one million dollars as a reward!
So if no one can actually solve these equations how can we use them to help us simulate the way that fluids move? This is where the field of Computational Fluid Dynamics (CFD) comes in. In CFD we use some very clever mathematical tricks that let us get very very close to the right answer. There are a number of problems in doing it this way though. The first problem is that we don't always get very close to the right answer, in fact sometimes we can get completely the wrong answer (and we don't always know this because we don't know what the actual answer should be in the first place!). Another problem is that to use these mathematical tricks we need very very big computers – there are some people out there running simulations on computers so big they are the equivalent of one million laptops all plugged into each other - and even with these massive computers it can still take months to calculate the answer! The purpose of my research then is to develop new methods and mathematical tricks we can use that allow us to get more reliable results, in a shorter time frame, on smaller computers. This will then allow us to investigate the way that fluids move in more detail and improve the way we design cars, planes and anything else that involves fluids (pretty much everything!).
A typical day for me usually involves being sat at my desk writing code and testing out new ideas. Problem solving plays a large part in programming and software development and the feeling of finally solving that problem you've been stuck on for ages is great. Another great aspect of my research is that, as fluids are involved in nearly all engineering applications, I have the opportunity to work in a range of different industries – from automotive and aerospace engineering to biomedical engineering and biotechnology. There are also examples of researchers in my field who have won Oscars for the fluid models they have made for animated films!
For further updates about my research activities please follow me on Twitter: @joconnor29
The link to the website of the people who will give one million dollars if you solve the fluid equations is here:
For a really good introduction to computers and programming see the 2008 Royal Institution Christmas Lectures:
See these YouTube videos of CFD in action: