Design of high-performance water-in-glass evacuated tube solar water heaters by a high-throughput screening based on machine learning: A combined modeling and experimental study

How to design water-in-glass evacuated tube solar water heater (WGET-SWH) with high heat collection rates has long been a question. Here, we propose a high-throughput screening (HTS) method based on machine learning to design and screen 3.538125 × 108 possible combinations of extrinsic properties of WGET-SWH, to discover promising WGET-SWHs by comparing their predicted heat collection rates. Two new-designed WGET-SWHs were installed experimentally and showed higher heat collection rates (11.32 and 11.44 MJ/m2, respectively) than all the 915 measured samples in our previous database. This study shows that we can use the HTS method to modify the design of WGET-SWH with just few knowledge about the highly complicated correlations between the extrinsic properties and heat collection rates of solar water heaters.