Welcome to Gaëtan Cortes

Some questions to Gaëtan ...

What is your field of research ?
My research will focus on a topic called ‘Data-Driven Computational Friction’. The context is as follows: at present, the majority of equations relating to friction have been created empirically, based on experimental results. However, these laws do not necessarily succeed in reproducing certain results accurately and with certainty, which can greatly distort subsequent simulations. The idea behind the DDCF is quite simple: rather than using empirical laws to carry out our simulations, we solve our problem by basing ourselves solely on datasets taken from experiments carried out, and on constitutive laws whose correctness we are certain of (such as the conservation of mass or force, for example). With this new technique, we have every hope of being able to simulate certain friction systems much more accurately, and to improve our simulations in this area! Eventually, we even plan to use Neural Networks to improve our results even further.

Why did you choose this field and what are you particularly enthusiastic about it ?
I had the opportunity to have a first very experimental approach to tribology during a master's project carried out within the Rolex company. As I delved deeper into this field, I realised that the empirical laws currently established are often far removed from the real results, due to the presence of a multitude of factors impacting tribological systems. In my opinion, this is a science that is still relatively poorly understood, despite years and years of research. If we could manage to obtain systems and simulations that are much more accurate than what we can produce at the moment, we would have a much better understanding of this science, and thus be able to improve friction in a multitude of areas! At present, friction accounts for a significant proportion of energy losses, whether in production, in engines or in any transmission system. Wear, meanwhile, is a major source of failure in many mechanical systems (such as watches!). If we could understand a little better how tribology in general works, we could have the tools in hand to improve our friction and wear, and thus make a major contribution to the world of mechanics!

What are the difficulties encountered ?
As I said, tribology is a very complex science, whose results can vary greatly from one test to another, and are highly dependent on the context (what materials are present, what the forces are, the direction of friction, etc.). These are all factors that could hopefully be better understood with the DDCF! Apart from that, the application of this method requires a thorough understanding of the fundamental laws of physics, as well as tribology in general, in order to be able to best combine experimental results and the realm of the possible. Added to this are the complex algorithms and equation systems to be solved, which we need to be able to optimise as much as possible in our codes in order to make them work properly on our computers or calculation servers. These are all challenges that I'm looking forward to tackling!

Outside the lab, what do you enjoy doing most ?
I've been a firefighter and first-aider at EPFL for several years now, and I'm very involved in both areas, always trying to improve my own skills and the quality of these two services. I've had the opportunity to meet some great people who can pull together under any circumstances, and I hope I'll be able to continue working with them for a long time to come! Apart from that, I've been a member of several campus associations (Fréquence Banane, Ludopoly), and I'm now a member of the extended committee of one of the biggest metal festivals in French-speaking Switzerland: Rock The Lakes! Finally, in addition to all that, during my weekends I sometimes take up medieval re-enactment with the Compagnie du Mont-Gibloux. All in all, there's plenty to keep me busy!