Assessment of Different RANS Turbulence Models in TPMS Pipes through a Multivariate Metric Evaluation
Please login to view abstract download link
Several engineering applications in energy and non-energy sectors require for their correct functioning the removal of extremely large heat loads, of the order of some MW/m2. Thus, innovative cooling systems are necessary to improve their performance, or merely allow them to operate. Additive Manufacturing is an important breakthrough in the production of such enhanced cooling systems, which enables the design of very complex structures, such as Triply Periodic Minimal Surfaces (TPMS). These are porous structures generated from 3D trigonometric functions. TPMS are smooth structures that can significantly increase the heat transfer surfaces while still maintaining low pressure drops. A very important issue in the study of these novel structures is the validation of the numerical models to predict their thermo-fluid-dynamic behavior and possibly optimize it. Recently, circular pipes equipped with different TPMS were tested in turbulent subcooled water flows at the Texas A&M University of Galveston, allowing the experimental determination of the hydraulic characteristic of the samples. The aim of this work is to explore the suitability of different RANS turbulence models, eventually with different CFD tools, in reproducing the experimental results. More in detail, it is shown to what extent the difference between the experimental and computed results can be explained by the experimental uncertainty of the test measurements and by the input and numerical uncertainties that affect the simulations. The analysis has been carried out by means of a multivariate metric that makes a global evaluation for all the hydraulic characteristics measured for each sample.
