Assessment of Reynolds-Averaged Navier-Stokes Modeling of Jet Interaction in Fan-Array Wind Generator Flows

TL;DR

This study evaluates the effectiveness of RANS modeling in predicting jet interactions in fan-array wind generators, comparing numerical results with experimental data to assess accuracy and limitations.

Contribution

It demonstrates that RANS modeling with pressure-jump boundary conditions can reasonably predict global flow features of FAWGs, highlighting both capabilities and limitations.

Findings

RANS captures overall jet interaction topology and velocity decay.

Systematic discrepancies occur in near-field and shear layers.

Turbulence intensity predictions show larger deviations due to modeling limitations.

Abstract

Fan-array wind generators (FAWGs) provide controlled turbulent inflow conditions that cannot be reproduced in conventional wind tunnels. Despite their increasing use in experimental studies, numerical modeling of FAWG-generated flows remains largely unexplored. The present study assesses the capability of Reynolds-Averaged Navier-Stokes (RANS) modeling to predict jet interaction in a 10x10 fan-array wind generator. Numerical predictions are compared against experimental measurements of axial velocity and turbulence intensity from a reference configuration. Individual fan units are represented using a pressure-jump boundary condition based on a reconstructed performance curve derived from manufacturer data. Grid convergence is verified, and the influence of fan representation, operating point and inflow turbulence conditions is examined. The results show that RANS modeling captures the global jet interaction topology and downstream velocity decay with reasonable accuracy. However, systematic magnitude discrepancies are observed in the near-field injection region and peripheral shear layers. Turbulence intensity predictions exhibit larger deviations, reflecting limitations of the eddy-viscosity closure in highly mixing-dominated flows. A low-aspect-ratio flat plate is included as a demonstrative application to illustrate the aerodynamic impact of FAWG-generated inflow. Overall, the study shows that RANS modeling, combined with a pressure-jump fan representation, provides a computationally efficient framework for predicting the mean-flow structure of FAWG systems, while exhibiting clear limitations in resolving localized turbulence characteristics.