Environmentally Opportunistic Computing
My main undergraduate research project was on data center waste heat utilization. Nearly all of the energy used to power servers is eventually converted to heat, and recent estimates have shown that ~40% of energy consumed by data centers is used for thermal management. The concept of Environmentally Opportunistic Computing (EOC) is to harness the excess heat from data centers and use it to provide air and water heating for adjacent buildings. My work was a continuation of the Green Cloud project at Notre Dame in which a small distributed data center was placed next to a greenhouse at the South Bend Botanical Gardens to reduce heating costs in the winter. For my research, I generated a model to estimate the energy savings for an organization that implements EOC.
Woodruff, J. Zachary, Aimee P. C. Buccellato, Paul Brenner, David B. Go, “Environmentally Opportunistic Computing: A distributed waste heat reutilization approach to energy-efficient buildings and data centers.” Energy and Buildings, vol. 69, pp. 41-50, 2014. pdf.
3D Printing with Clay
This was a collaboration between the Chemical Engineering and Ceramics departments at Notre Dame, and was meant to provide art students the ability to generate 3D models and print them in clay. To begin the project I researched numerous closed and open source methods for generating 3D models and creating G-code for running 3D printers. I then built and tested a Bits from Bytes (BfB) RapMan 3.2 3D printer, and began the process of replacing the PLA/ABS printer head with a clay extruder. Electronics issues while integrating the stepper motor for the clay extruder, as well as the acquisition of BfB by 3D systems pushed us to use a different printer. I graduated before the project was complete, but the build blog and outcome of the project can be found here.
Wind Turbine Power Analysis
In this project I investigated the energy production potential of a wind turbine on the Notre Dame campus under varying atmospheric conditions. In collaboration with Professor Robert Nelson, I analyzed months of data from sensors on the turbine and an adjacent meteorological tower to measure the actual versus theoretical performance of the turbines. This data could then be used to improve the energy harvesting of wind turbines by modeling how wind angle, shear, direction, and velocity affect the power output. During this work I learned how to organize large data sets and to manipulate them into useful graphs to compare theoretical and actual performance.
