Session: K8-01: FUNDAMENTALS OF MACHINE LEARNING IN HEAT TRANSFER

Paper Number: 139237

139237 - Fundamentals of Machine Learning for Phase Change Heat Transfer 

Abstract: 

Multiphase phenomena are observed in our everyday life in nature and many industrial applications, ranging from dew condensation on insects, water harvesting, electronics cooling, climatology prediction, hydrogen generations, and manufacturing. While the fundamentals of multiphase processes have been studied for over a century, key scientific questions remain regarding the fundamental mechanisms governing complex phenomena. The intricate interplay between the evolution of phase boundaries and mass transport results in nonlinear behavior, where subtle changes in one parameter can have profound and unexpected effects on others. The multimodal, multidimensional, and transient nature of these processes presents challenges for investigation and comprehension. Additionally, interpreting experimental data and predicting multiphase phenomena remain significant challenges. Central to unraveling these complexities is the extraction of interpretable and rich datasets from dynamic multimodal information around phase boundaries, such as bubbles, droplets, or interfaces. To address these challenges, our research group seeks to integrate cutting-edge computer vision and machine learning strategies. 

This presentation highlights the key approaches developed by my group, unveiling previously undefined features and hidden mechanisms. Additionally, I will introduce examples demonstrating how AI technologies facilitate the learning, understanding, and prediction of the dynamic nature of multiphase phenomena through the exploration of various data types. Multiple scenarios will showcase the impact of different combinations of data types, such as fast optical, events, or thermography. In conclusion, this talk will briefly discuss potential game-changing innovations for real-world applications, delving into the exciting opportunities emerging from the convergence of multiphase physics and advanced technologies.

Presenting Author: Yoonjin Won University of California, Irvine

Presenting Author Biography: Yoonjin Won received a B.S. degree in Mechanical and Aerospace Engineering from Seoul National University, and M.S. and Ph.D. degrees in Mechanical Engineering from Stanford University. She is currently an Associate Professor of Mechanical and Aerospace Engineering at the University of California, Irvine. She has courtesy appointments in Electrical Engineering and Computer Science and Materials Science Engineering. Dr. Won's overarching research goal is to gain fundamental insights into multiphase thermal science, centering on keywords—AI for science, graphic-driven physics, data-driven approach, and materials design. She is a recipient of the National Science Foundation CAREER Award, the ASME Electronic & Photonic Packaging Division Early Career Award, the ASME Electronic & Photonic Packaging Division Women Engineer Award, the ASME ICNMM Outstanding Leadership Award, the Emerging Innovation/Early Career Innovator from UCI Beall Innovation Center, Faculty Excellence in Research Awards, Mid-Career from UCI, and numerous best paper and poster awards. The key papers are published in high impact journals including Advanced Science, Advanced Functional Materials, Small, Proceedings of National Academy of Science (PNAS), and American Chemical Society (ACS) journals. Additional details for Dr. Won’s qualifications and research group are available online (won.eng.uci.edu).

Authors: