20 may 2021. 13.00 CEST
Prof. Javier Jiménez Sendín from Technical University of Madrid (UPM)
We explore how far the scientific discovery process can be automated. Using the identification of causally significant flow structures in two-dimensional turbulence as an example, we probe how far the usual procedure of planning experiments to test hypotheses can be substituted by `blind’ randomised experiments, and note that the increased efficiency of computers is beginning to make such a `Monte-Carlo’ approach practical in fluid mechanics. We briefly describe the process of data generation, classification and model creation, and, although the emphasis of the talk is to explore the procedure rather than to model turbulence, we will show that that the Monte Carlo process naturally leads to the consideration of building blocks of the flow which are fairly different from the conventional individual vortex cores. This `spontaneous’ discovery supports the claim that an important advantage of randomised experiments is to bypass researcher prejudice and alleviate paradigm lock. We will remark that the method can be extended to three-dimensional flows in practical times.
Javier Jiménez is a Doctor in Aeronautical Engineering by the U. Politécnica de Madrid, and holds a Master in Aeronautics and a PhD in Applied Mathematics by the California Institute of Technology in Pasadena, CA. He is currently distinguished research professor of Fluid Mechanics at the U. Politécnica of Madrid, and has been professor or long-time visitor at the Ec. Polytechnique in Palaiseau, Paris, and the Centre for Turbulence Research at Stanford University, at Caltech, the U. Cambridge, UK, the U. California San Diego, the KITP at the U. California Santa Barbara, and the UNAM in Mexico, among others. He previously worked as research scientist at the IBM Scientific Centre in Madrid.
He is a member of the Spanish Royal Academy of Sciences and the Royal Academy of Engineering, and serves, or has served, in random committees of academic journals and similar things. His research activity is centered on the study of fluid mechanics and turbulence, and on the use of large scale computing techniques for the analysis of turbulent flows and of experimental data. He is currently looking for research students and postdocs.