13:00 - Modelling and Controlling Turbulent Flows Through Deep Learning
Dr Ricardo Vinuesa, KTH Royal Institute of Technology
The advent of new powerful deep neural networks (DNNs) has fostered their application in a wide range of research areas, including more recently in fluid mechanics.
In this presentation, we will cover some of the fundamentals of deep learning applied to computational fluid dynamics (CFD). Furthermore, we explore the capabilities of DNNs to perform various predictions in turbulent flows: we will use convolutional neural networks (CNNs) for non-intrusive sensing, i.e. to predict the flow in a turbulent open channel based on quantities measured at the wall.
We show that it is possible to obtain very good flow predictions, outperforming traditional linear models, and we showcase the potential of transfer learning between friction Reynolds numbers of 180 and 550.
We also discuss other modelling methods based on autoencoders (AEs) and generative adversarial networks (GANs), and we present results of deep-reinforcement-learning-based flow control.
About the speaker
Ricardo is an Associate Professor at the Department of Engineering Mechanics, at KTH Royal Institute of Technology in Stockholm.
He is also a Researcher at the KTH Climate Action Centre and Vice Director of the KTH Digitalization Platform. He studied Mechanical Engineering at the Polytechnic University of Valencia (Spain), and he received his PhD in Mechanical and Aerospace Engineering from the Illinois Institute of Technology in Chicago.
His research combines numerical simulations and data-driven methods to understand and model complex wall-bounded turbulent flows, such as the boundary layers developing around wings and urban environments.
Dr. Vinuesa has received, among others, an ERC Consolidator Grant, the Göran Gustafsson Award for Young Researchers and he is coordinating a number of EU-funded projects.
13:30 - Challenges and Opportunities for Machine Learning in Fluid Mechanics
Dr Miguel A. Mendez, von Karman Institute for Fluid Mechanics
Big data and machine learning are driving comprehensive economic and social transformations and rapidly reshaping applied scientists' toolbox and methodologies. Machine learning tools are designed to learn functions from data with little to no need for prior knowledge.
As continuous developments in experimental and numerical methods improve our ability to collect high-quality data, machine learning tools become increasingly viable and promising in disciplines rooted in physical principles.
This talk explores how machine learning can be integrated and combined with more classic methods in fluid dynamics. After a brief review of the machine learning landscape, we show how many problems in fluid mechanics can be framed as machine learning problems and explore challenges and opportunities.
We consider several relevant applications: aeroacoustic noise prediction, turbulence modelling, reduced-order modelling and forecasting, meshless integration of (partial) differential equations, super-resolution and flow control. While this list is by no means exhaustive, the presentation will provide enough concrete examples to offer perspectives on how machine learning might impact the way we do research and learn from data.
About the speaker
Miguel received his PhD in engineering science from “Université Libre de Bruxelles” in 2018.
He is currently an Assistant Professor at the von Karman Institute for Fluid Dynamics, where he teaches courses on modelling and control of fluid flows, measurement techniques, signal processing and machine learning.
He has extensively used data-driven methods for post-processing numerical and experimental data. His main research activities include experimental fluid mechanics (particularly image-based velocimetry and image processing), reduced-order modelling, flow control, and machine learning.
14:00 – Physics-constrained Machine Learning in Extreme Fluids
Dr Luca Magri, Imperial College London
The ability of fluid mechanics modelling to predict the evolution of a flow is enabled by physical principles and empirical approaches. Physical principles, for example conservation laws, are extrapolative (until the assumptions upon which they hinge break down): they provide predictions on phenomena that have not been observed.
Human beings are excellent at extrapolating knowledge because we are excellent at finding physical principles. Empirical modelling provides correlation functions within data. Artificial intelligence and machine learning are excellent at empirical modelling.
In this talk, the complementary capabilities of both approaches will be exploited to achieve adaptive modelling and optimization of nonlinear, unsteady and uncertain flows. The focus of the talk is on computational methodologies for modelling and optimization of turbulent flows: (i) prediction and control of extreme events with reservoir computers, and (ii) reduced-order modelling of turbulent flows with auto-encoders, which generalise POD/DMD methods to nonlinear dynamics. Robustness, generalisability, and interpretability will be discussed.
About the speaker
Luca is a Reader (equivalent to US Full Professor) in data-driven fluid mechanics at Imperial College London, Aeronautics Department; Fellow of The Alan Turing Institute, Hans Fischer Fellow of the Institute for Advanced Study (TU Munich) and Affiliated Faculty at Cambridge University Engineering Department.
Prior to joining Imperial, Luca was a Lecturer at Cambridge University Engineering Department, Royal Academy of Engineering (RAEng) Research Fellow, and Fellow of Pembroke College.
Prior to becoming a lecturer and RAEng Research Fellow at Cambridge, he was a postdoctoral Fellow at Stanford University Center for Turbulence Research. He obtained his PhD in Engineering at the University of Cambridge. His research is currently funded by an ERC Starting Grant; UKRI; and industry.
At Imperial, Luca is leader and found of the Research Centre of Data-Driven Engineering, and The Alan Turing Institute Luca is group leader of Physics-Informed Machine Learning and Data Assimilation under the Data-Centric Engineering Programme.
14:30 – Discussion
See the poster for more information