Flow boiling is important in industries such as power generation, air-conditioning, aerospace, defense, and thermal management. However, some of the fundamental dynamics of flow boiling are not well understood when analyzed using traditional approaches. Machine learning methods have shown potential in revealing fundamental behaviors in complex flows by combining innovative data modeling tools with physical knowledge of the underlying system. This project will use a new machine learning-guided investigation approach to perform advanced experiments and modeling of thermal transport during flow boiling. The project outcomes will help improve the design and control of future thermal systems that rely on flow boiling. The project will support development of an educational module “Phase of Matter” for Case Western Reserve University’s NSF-supported Introduction to Innovation Teaching program. In addition, a new internship opportunity will be provided for high-school students in the Cleveland district to link classroom science to technologies in thermal sciences. Phase-change configurations like flow boiling can significantly reduce the size and weight of thermal management systems, thereby directly reducing energy costs. However, the barrier to more widespread implementation of flow boiling phase-change configurations stem from a lack of good understanding of complex flow dynamics and thermal transport physics. This project will use a novel machine learning-guided investigat