Performance enhancement of a C-shaped printed circuit heat exchanger in supercritical CO2 Brayton cycle: A machine learning-based optimization study

The present work is focused on enhancing the overall thermo-hydraulic performance of a previously proposed C-shaped printed circuit heat exchanger (PCHEs) using Machine Learning (ML) Algorithms. In this context, CFD analysis carried out 81 different channel configurations geometry, and computed data used to train three ML algorithms. Later, geometry optimized by coupling trained model with multi-objective genetic algorithm (MOGA). Finally, (called optimizedML) investigated numerically for wide range Reynolds numbers. Its compared zigzag base C-shape response surface methodology (RSM). findings showed that multilayered approach combining MOGA, CFD, machine learning techniques beneficial accomplish robust realistic solution. Comparing characteristics optimizedML shows former up 1.24 times better than latter based evaluation criteria (PEC). Furthermore, optimizeML was found 21% 16% higher RSM cold hot sides, respectively.