Marco Graffiedi

Marco Graffiedi is a graduate student in nuclear science and engineering, researching boiling heat transfer for applications in space exploration and electronic cooling. He integrates experimental data with computational modeling to develop and validate microscopic heat transfer mechanisms using cryogenic and dielectric fluids. As a MathWorks Fellow, he will design a framework that combines image-based detection of liquid-vapor interfaces with finite element solvers to estimate heat flux mechanisms in scenarios where direct measurement of heat flux, such as infrared thermography, is impossible. Marco uses MATLAB’s deep learning models and other tools, such as temporal U-Net architectures and Kalman filtering, to improve the segmentation of high-speed video recordings of boiling surfaces. Finite element thermal modeling is then used to reconstruct the heat flux at the surface based on the segmented video data. Marco’s work could reshape how scientists and engineers characterize thermal behavior in extreme environments, opening new possibilities for efficient heat management in both aerospace systems and high-performance electronics.