Printed circuit heat exchangers (PCHEs) are known for their compact design and efficient heat transfer characteristics; however, optimizing their flow and heat transfer using the finite volume method is computationally inefficient. This study combines Proper orthogonal decomposition (POD), Randomized singular value decomposition (RSVD), and Gaussian process regression (GPR) to develop a reduced-order model with enhanced optimization efficiency. Using Latin hypercube sampling, operational parameters are fed into Fluent to generate sample data fo...
