Department of Mechanical and Aerospace Engineering Assistant Professor Qiong Liu, who joined the New Mexico State University College of Engineering faculty in January 2023, focuses her work on advancing fluid mechanics through novel research and development.
Liu leads the Flow Lab team that works to expand the understanding of fluid mechanics in various engineering applications and develop cutting-edge flow control strategies to overcome emerging challenges.
In February, Liu was awarded a grant to research applications of aerodynamics in aviation. The three-year, $457,0777 grant from the Air Force Office of Scientific Research is for the project, “Physics-Informed Reinforcement Learning-based Multiple-Input Multiple-Output Flow Control.” Results of the research could not only impact aviation, but other industries where fluid dynamics are important such as automotive engineering, maritime transport and wind energy. Mechanical and aerospace engineering Associate Professors Andreas Gross and Fangjun Shu are co-principal investigators on the project.
“The aim of this research is to enhance the theoretical understanding and practical applications of aerodynamics in aviation,” Liu said. “Integrating reinforcement learning with flow control opens a new avenue for enhancing the aerodynamic performance of air vehicles. We hope this work will pave the way for advanced machine learning-based flow control to improve the performance of flow-related applications and enrich our physical understanding.”
Liu said the inspiration for this project stems from the perpetual topic of how to improve the aerodynamic control over airfoil to enhance the aerodynamic performance, such as reducing drag and increasing lift of airplanes.
“With the requirement for the maneuverability and agility, the design of an effective flow control becomes more and more challenging,” Liu said. “Traditional flow control design has shown less capable than desirable. Seeking more adaptable and adjustable flow control strategies is therefore crucial. With the development of reinforcement learning, examples have shown that this technique can be used not only for playing games and making decisions but also, inspired by these applications, for integrating into the flow control framework. Deciding on flow control actuation based on real-time analysis of the flow field allows for adjustments in response to timely changes in the flow, addressing the need for adaptive flow control.”
Liu said she believes the biggest challenges for the project could occur when the trained control agent encounters uncertainties in the flow, such as turbulence.
“To address this concern, we will train the control agent using various magnitudes of free stream turbulence to better equip it to cope with uncertainties,” Liu said.
As an AFOSR grant, this project has substantial importance to the U.S. Department of Defense since it investigates advanced active flow control strategies for high-performance aircraft.
“This research has the potential to contribute to the development of aircraft with increased mission effectiveness, survivability, maneuverability and reduced radar cross-section, all of which are crucial factors in modern warfare,” Liu said. “The outcomes of this basic research project will provide the DoD with valuable insights and advancements in aircraft control systems, delivering a technological edge that will enhance military operations and ensure the safety of personnel.”
In addition to the research elements, the project will include educational components and funding for two graduate students. Interested undergraduate students also will have the opportunity to participate in the project.
