Federal Project Grant Award Summary

The University of Florida received a $300,000 EAGER (Early-Concept Grants for Exploratory Research) award from the National Science Foundation's Division of Chemical, Bioengineering, Environmental, and Transport Systems under the Engineering program (CFDA 47.041) to develop predictive machine learning models for microchannel flow boiling systems. The project, which commenced May 1, 2025 and concludes April 30, 2027, will deliver a Parametrized Hybrid Physics-Informed Neural Network (PH-PINN) designed to predict the transition of bubble flow regimes in microchannels from small, separated bubbles to elongated, merged bubbles. This research integrates high-speed experimental data, computational simulations, and machine learning techniques to create rapid predictive capabilities across varying operating conditions, addressing critical thermal management challenges in high-performance computing, power electronics, and renewable energy applications.

The research deliverables will advance microchannel cooling system design by moving beyond trial-and-error methodologies to enable more efficient and reliable thermal management solutions. Through combining fundamental thermodynamics with advanced computational techniques, the project will produce novel predictive models that capture complex multiphase interactions in bubbly flow without requiring extensive computing resources. The research will also support student training in interdisciplinary areas spanning fluid dynamics, artificial intelligence, and thermal management, contributing to workforce development in emerging engineering fields critical to next-generation thermal systems.