Won Best Poster at the International Conference on Thermoelectrics for project “Numerically Resolved Radiation View Factors for Single and Multi-Junction Thermoelectric Devices” that focused on developing methods to quickly and accurately determine radiation view factors within thermoelectric devices (TEDs).
Thermoelectric devices take thermal heat and convert it into electrical energy. Through this project, conducted as part of an independent Special Projects course at the University of Pittsburgh, I modeled heat exchange between the surfaces of a TED considering surface orientation and geometry since these factors influence heat transfer. Performing calculations for TEGs that incorporate multiple junctions requires parallelization and high computing power. The modeling I undertook demonstrates how distribution of computing power to the GPU accelerates calculations of radiation view factors and how variation of geometry controls radiation heat transfer.
This project is important for providing engineers and scientists with capability to predict heat transfer to accurately capture the physics within their systems. By computing and understanding the limitations of maximizing temperature differences across emitting and receiving TED surfaces, space probes can become lighter with improved thermal efficient and power output.