Jung-Wuk Hong

Urban canopy flows exhibit complex interactions between geometric heterogeneity and turbulent organization, posing challenges for air-quality, ventilation, and low-altitude urban-flow prediction. While canonical studies have characterized idealized building arrays, realistic urban morphologies often involve building-height variability (H/3≤h≤H) and variable orientations relative to prevailing winds, whose combined effects remain insufficiently characterized. This study investigates turbulent flow over rotated building arrays with randomly sampled heights at 0°, 15°, 30°, and 45° using large-eddy simulations, with single-cube validation and array-level grid-refinement checks. A balanced rotation–height matrix retains the same sampled height layouts across orientations, allowing trends associated with the rotation-induced projected-blockage change to be distinguished from realization-dependent variability. Mean velocity and wake-deficit fields show that local height contrast influences passage acceleration and channeling flows between buildings, whereas array rotation induces wake deflection, skewness, and wake-deficit organization. Across the 16 rotation–height cases, array rotation increases projected frontal blockage, domain-averaged turbulent kinetic energy (TKE), and low-to-intermediate spanwise spectral energy. The transition from θ = 30° to 45° separates wake-deficit and turbulence responses: wake-deficit metrics decrease or saturate in several realizations, whereas TKE and transport-related quantities continue to increase. A projected-blockage-adjusted comparison relates the remaining TKE response to the spanwise gradient of turbulent streamwise-momentum transport. Local analyses show that height realization modulates wake asymmetry and turbulence redistribution. These results support a multiscale interpretation in which the rotation-induced projected-blockage response is associated with canopy-scale wake organization and turbulence response, whereas height arrangement and spanwise exchange remain important for local recovery.

Twin impinging jets are utilized in various engineering fields such as electronics cooling and aerospace vehicles. The fundamental physics of nonlinear load transitions remains unclear despite the wide use of these jets. High-resolution steady Reynolds-averaged Navier–Stokes simulations are performed using the k−ω shear stress transport model. A macro-scale diameter of 1.2 m is selected to ensure high numerical resolution. The numerical model is validated through a comparison with standard experimental data. Nozzle spacing and altitude are found to be the primary parameters of nonlinear load transitions. The localized decrease in ground load at specific geometric conditions is defined as the low-wall shear stress (WSS) valley. This transition is caused by a flow regime shift from individual jet dominance to an upwash fountain structure. A viscous scaling relation is identified where WSS is proportional to velocity raised to the power of 1.82 within the investigated low-Mach conditions. Critical load regions are identified through geometric parameters on a 2D regime map. The results provide practical guidance for interpreting load trends within the investigated range.