Abstract
Abstract
The heat exchanger distributor, which connects the tube carrying the medium to the heat transfer geometry, has received less attention in research studies compared to heat transfer geometries. However, the flow distribution is critical to efficiency. Therefore, a model-in-the-loop approach is presented to optimize additively manufactured distributors. A quasi-fractal distributor and all connected geometries, including pipe connections, heat transfer geometry and collector, are meshed and simulated in OpenFOAM using steady-state computational fluid dynamics. The volumetric flow rates determined from the simulation are assigned to the respective flow cross-sections during the distribution process. The flow cross-sections of the geometric model are modified within an optimization process, considering the prevailing volumetric flows. The results indicate a 64 % decrease in the standard deviation of volume flow across all 85 channels with a simple optimizer and few simulation loops. The printable geometry can be extracted directly out of Python using common file formats. The presented workflow has been proven feasible. Future improvements could involve using the automated model-in-the-loop simulation and creating optimizing algorithms considering fluid flow behavior.