I spent 3 years in highschool doing independent research, mostly focusing the application of 3D printing in aviation. This was mostly inspired by the works of 3DLabPrint, as I was fascinated with the idea of flying a 3D printed RC airplane. This was when I learned most of what I know about aeronautics today, in addition to the various computational tools associated with the research. In 12th grade I presented my paper about compliant variable-camber wings to the Regeneron Science Talent Search research competition, and placed in the top 300 semifinalists in the US!

After enrolling in university, I took up research with the Duke Aerolasticity Team, under supervision of Dr. Luisa Serafim and Earl Dowell. I was tasked to collect windtunnel pressure data for an airfoil section under "nonlinear" conditions. This meant studying the airflow as the wing changed its angle of attack, as air has elastic properties and doesn't immediately change in-phase with surrounding objects.

Since March of 2025, I have been working with a fellow undergraduate to study the effects of aeroelastic divergence- a phenomenon where a wing suffers from structural (torsional) instability at a certain critical speed due to an aerodynamic twisting moment. The research paper has recently been submitted to AIAA for review!

Being able to study and conduct both methods of obtaining aerodynamic data (CFD and windtunnel) has vastly expanded my view in aeronautics, as this is what aerospace engineers use in the real world. It is an honor to be able to work with such a talented team on novel research projects, and I hope to continue my involvement in aerodynamic research.

[2025] Researcher at Duke Aeroelasticity

Author/Co-Author of research paper by Harsh Mathur, Eric Huang.

Titled: An Experimental Study of Aeroelastic Divergence

The premise of this research project is to explore aeroelastic divergence, which is a phenomenon where an aircraft wing suffers from structural (torsional) instability due to the moment generated from lift. This occurs at a certain critical speed, which can be predicted with some theory established by Dr. Dowell.

Our job was to test if this divergence theory holds up for a simple flat plate and a 0012 NACA airfoil. My job in this project was to build an apparatus rigid enough to accurately measure the angular deflection in the wing, while my research partner focused on coding to analyze results.

So this is what I made! To the right, you can see the final iteration of the wing mounting apparatus. The torsional component of the wing is simulated as a torsional spring, and the deflection can then be measured on an angle track gauge (accurate to 0.1 degrees!).

What did I learn?

As previously mentioned, the premise of Dr. Serafim's research is to study the elastic behavior of air as an aircraft wing undergoes nonlinear conditions, and to prove the effectiveness of a theoretical model (Volterra series) in this state.

My job was to take physical data using a windtunnel to validate Dr. Serafim's theoretical results. That is, pressure values in the form of voltage readings from ten pressure sensors, each physically piped to a pressure tap on the surface of the wing. The rigid airfoil was mounted to a stepper motor, allowing motor programs to physically simulate nonlinear conditions (rapid changes in angle of attack, and oscillations at different frequencies) at different windtunnel speeds (15-30 m/s).

There were some structural challenges with the 3D printed airfoil base, but I am proud to say that I made the aluminum baseplates for both airfoils entirely by hand (rotary tool, drill press, and a ton of sanding) due to delays with the CNC.

What did I learn?

In an engineering dynamics course we were tasked to do a short research paper on a physical system, and I settled on the infamous "3-body system". Of course I wasn't planning to analytically derive a general formula, but instead create a simulation program capable of plotting out the trajectories of three celestial objects in real time.

There's this neat library called SciPy that allows for numerical solving of first order differential equations. Deriving a series of 18 first-order differential equations allowed the plot of 3 bodies in 3D space. The cool thing about this program is that it doesn't stop at 3 bodies. When two bodies inelastically collide, their combined momentum and mass form a new single body, resulting in a now 2-body system.

What did I learn?

Abstract: A 3D printed aircraft wing was designed so that the camber of the wing can be altered to achieve different aerodynamic characteristics. Finite Element Analysis was used to simulate the deflection on the wing under a specified load, in order to test the compliance of the 3D printed wing. Computational Fluid Dynamics was then used to simulate airflow over the wing at various degrees of camber and angle of attack to determine the optimal wing configuration at different flights situations.

I would like to revisit this project in the future, using composite materials as the compliant material instead of PLA plastic. Using such materials like fiberglass and/or carbon fiber would provide much better strength/weight characteristics, while keeping the favorable compliant properties. Using Duke's windtunnel might also prove useful in collecting/validating the CFD data obtained from the original research project.

What did I learn?