Forest Atchison and Christopher Neff won the prestigious Graduate Research Fellowship from the National Science Foundation (NSF) in 2021.
Forest Atchison (left) is currently completing his Bachelor’s degree in Electrical Engineering (BSEE) with a concentration in Power and Energy Systems. He has earned the distinction of being on the Chancellor’s list in every semester that he has been enrolled. He is also an Early Entry student in the Master of Science in Electrical Engineering (MSEE) program and plans to begin his Ph.D. in Electrical Engineering from the fall. Forest was awarded the IEEE PES John W. Estey Outstanding Scholar and the IEEE PES Scholarship Plus awards in 2019 and 2020. As a Graduate Research Fellow, Forest proposes to investigate and formulate system-level analysis tools that couple electrical and thermal characteristics for electric power systems. Tools such as power flow (PF), continuation power flow (CPF), optimal power flow (OPF), and state estimation (SE) are essential for the proper and optimal planning and operation of the electric grid. They are at the basis of all decision making. Increasing the output accuracy of these tools would subsequently enable better-informed decisions. The overarching objective of his proposed research is to use numerical methods to solve computationally intensive load flow-based problems of combined underground/overhead (UG/OH) transmission systems incorporating the coupling of electrical and thermal characteristics. This will develop a greater understanding of the degree to which conventional voltage stability and ampacity limits are too conservative, providing operational flexibility and awareness.
Christopher Neff (right) graduated Magna Cum Laude in the Bachelor of Science in Computer Engineering program in 2019 and continued on to the Ph.D. program in ECE. He received the Outstanding Senior in Computer Engineering award from the ECE Department in 2019 for his outstanding performance in undergraduate studies as well as early success in research. His doctoral research focuses on enabling and constructing real-time artificial intelligence algorithms suitable for real-world computer vision applications. Agile and lightweight algorithms that execute efficiently on available GPU and FPGA resources are critical to enabling real-time execution. Training paradigms that reflect real-world scenarios – such as objects at a distance, noisy data, and robustness to changes in location – are desperately needed in computer vision. To this end, Chris’s aim goes beyond simply designing lightweight algorithms. His goal is to create a methodology for sharing the computation between networks in multi-task systems, enabling higher accuracy and faster execution. He also plans to develop training paradigms that account for objects at a distance, complex non-linearities in movement, and noisy data. This is an area completely ignored by the majority research. Finally, Chris plans to develop an online training method for acutely understanding multi-camera movement in previously unseen environments, lowering the barrier to entry for many multi-camera computer vision applications.