Undergrad Research

From top to bottom: RK4 time stepping scheme, particles flowing in velocity field, and the underlying stream function

Fluid Mechanics with Machine Learning Research

  • Problem and Background

    • Oceanic researchers track floating "drifters" as a window into flow patterns

    • With the limited drifter data it is often difficult to reconstruct the underlying ocean velocity fields

    • A new machine learning system could be adopted to see patterns in drifter data

    • The Bickley Jet is a mathematically described flow field which has gyres similar to ocean currents

  • Plan of attack

    • Use MATLAB and a 4th order Runge-Kutta time stepping method to generate simulated drifter data from a Bickley Jet stream function

    • Run machine learning on the simulated data

    • Compare velocity field estimated by the machine learning system to known Bickley Jet velocity field to validate technique and refine the machine learning

    • Use the machine learning on oceanic data to extract velocity field information

Confocal microscope images and height data of GaN film printed onto a silicon wafer

Scanning electron microscope image of a gold pad deposited on top of the GaN layer. The problem in the image is the low density of GaN particles at the gold-GaN interface

"Printed" GaN Field Effect Transistor at CHN

  • Problem and Background

    • Silicon/silicon oxide based field effect transistors (FETS) currently have significant energy losses at high powers and frequencies

    • Gallium nitride based FETs offer an alternative for high power high frequency applications

    • Unfortunately, current GaNFETs are only made with an expensive deposition based process

    • "Printing," or deposition of GaN particles on a chip through fluidic assembly, could dramatically drive down the cost of production

  • Solution

    • Developed method for suspending particles in solution for long enough to print

    • Determined method for printing even 200nm layer of GaN particles on silicon

  • Results

    • Successfully created even GaN layers on silicon chips

    • Unfortunately, the project lost funding because of trouble forcing the GaN layer to sinter into a coherent layer instead of a collection of particles