Thursday, May 17, 2007

My graduation project

Hi everybody,

In a few months I will graduate at the faculty of Aerospace Engineering at the TUDELFT.
But only a few people know what I am doing. So here's some info. I'm writing in english so everybody can understand. Imagine this:

In a few years from now...., you feel tired very fast. Every time you go up the stairs, you feel awfully tired. So you decide to see a doctor. He decides to send you to a hospital for a scan. After a few days, you hear the results and the doctor concludes one of you major arteries has narrowed extensively (called artherosclerosis). So you need treatment. The doctors in the hospital suggest to do surgery and show you the scan and what kind of surgery is needed (for example a bypass) . Now wouldn't it be nice, before you go into surgery, to have computer model telling you what complete influence the bypass has on your vascular system and for the doctor to see if the bypass will have the desired effect?.... This is where my graduation project starts..

This computer model is quite complicated. We need to model the behaviour of blood, the blood vessel and the interaction with eachother. This will result in a so-called fluid-structure-interaction problem which shows to be very challenging. Since both blood and blood vessel vary under certain circumstances (like age, percentage of cholesterol in your blood and many other parameters) these models are extremenly complicated. And even when we have defined proper models, we need a solution technique that is efficient enough to run on a single computer within small amount of time, say a few hours. ( It becomes a bit mathematical now..) . Although in real life your arteries are three-dimensional "tubes", I am looking at a simplified two-dimensional problem, to reduce the complexity a little bit. In the last two decades, a lot of research has been done on the Finite Element Method. This method approximates the solution to the difficult equations we need to solve ( like the Incompressible Navier-Stokes equations for the fluid ) by a serie of known simple functions. So the more simple function we include in our approximation, the better the approximation but the larger the calculation. Now, using this FEM method, I try to solve the fluid-structure-interaction problem and look at how to couple the fluid and the structure and do it in a efficient manner.

Although the explanation is a bit short, I hope it gives an idea..

grt Eliam
( One of the most famous examples of fluid-structure-interaction is the Tacoma bridge see picture at the top)

1 comment:

Anonymous said...

Hey Eliam,
wat leuk om wat meer over je afstudeerproject te lezen. Echt interessant, omdat ik dus de patienten verpleeg die voor een bypass gaan.

Liefs Nicole