Qihong Lorena Li Hu

PhD Thesis

Data-driven modelling of turbulent flows under active flow control

Supervisors

Stefano Discetti (UC3M) & Andrea Ianiro (UC3M)

Abstract

Active Flow control is a paradigm that has shown great potential in the performance enhancement of countless applications/devices in the industry. The complexity of its implementation (actuators positions, control authority, sensoring, definition of control strategy, etc.) makes it an interesting problem that still requires fundamental research to fully understand its potential and limitations. Even when control is successful in the laboratory environment, it is challenging to transfer the lessons learned into the real-world applications. It is necessary to provide complete measurement techniques and analytical tools to interpret the successful control logics to bring their application outside of the lab. The main aim of this thesis is to establish data-driven modeling techniques to distil models from configurations with active flow control. For this purpose, I look forward to developing techniques for nonlinear system identification, and innovative experimental setup configurations capable of obtaining complete information of the flow with limited hardware. A combination of surface measurements, velocimetry, and density-based techniques (for instance Background Oriented Schlieren) will be implemented and tested for the extraction of interpretable control models.