Understanding turbulent flow over ripple-shaped random roughness in a channel

Abstract

Direct numerical simulation is used to understand the flow over ripple-shaped random rough elements. The random roughness has been generated using the model for sand ripples consisting of saltation, creep and suspension processes. A set of metrics based on both the geometrical and statistical properties of roughness was derived to characterize random roughness. The eight cases that have been studied with varied asymmetric distribution and “peakedness” as specified by the skewness and kurtosis of the height distribution varying from 0.28 to 0.7, and from 1.8 to 2.2, respectively. Analogous to λ/h for canonical (regular) roughness, λavg/hmax was selected as the geometrical parameter to characterize the surface, ranging from 4 to 26. The results have revealed that roughness significantly alters the mean velocity as well as turbulence in the inner layer. The outer layer is relatively unaffected due to the presence of roughness. The results further revealed that roughness distribution that is symmetric and which is well spread out has a stronger influence on the mean flow in the near-wall flow region. Altering the λavg/hmax affects the skewness and kurtosis of the height distribution, which are the important metrics to quantify the mean flow over a rough wall with random rough elements. The peak turbulence intensities increased with increasing λavg/hmax, however its effect on rms of the velocity fluctuations was confined to a region very close to the rough wall. Similar to mean flow, a symmetrical and a well-spread roughness distribution resulted in more energetic turbulent structures, stronger near-wall vortices, and increased turbulence activity. A similar trend was observed for the small scale features of the turbulent flow as seen from the vorticity fluctuations. To characterize turbulent flow over random rough surface of given maximum height, the geometrical parameter λavg/hmax as well as statistical parameters, the skewness and kurtosis are important metrics to be considered.