Bridging data-driven modeling and CFD
Jiaqing Kou （寇家庆）
I'm an Alexander von Humboldt Postdoctoral Fellow at Institute of Aerodynamics, RWTH Aachen University. I was a former postdoctoral researcher at Universidad Politécnica de Madrid in 2022. From 2019 to 2022, I was a Marie-Curie Early Stage Researcher (PhD student) at Numeca International and Universidad Politécnica de Madrid, under the ASIMIA program (Advanced High-Order Simulation Methods for Industrial Applications). My supervisors are Prof. Esteban Ferrer and Prof. Charles Hirsch. I obtained my Bachelor and Master degrees in 2015 and 2018, respectively, from the School of Aeronautics, Northwestern Polytechnical University, under the supervision of Professor Weiwei Zhang.
My research interests include data-driven modeling and numerical methods, with applications to unsteady aerodynamics and aeroelasticity. (My CV)
Background
System identification learns the intrinsic relationship between input-output data and establishes a low-order model for dynamic systems. We utilize and improve the nonlinear neural network model to identify the unsteady aerodynamics, as well as develop hybrid reduced-order models for both aerodynamic linear and nonlinear behaviors.
Mode decomposition techniques are commonly used in model order reduction of full-order, complex and unsteady flows. Our aim is to make these approaches more tractable to different modeling tasks, and use them to discover the intrinsic, low-order, dominant system characteristics based on a huge amount of available samples.