Impact of turbulence and secondary flow on the water surface in partially filled pipes

Abstract

Large eddy simulations (LESs) of turbulent flow in partially filled pipes at various filling degrees are conducted to investigate the response of the water surface to the turbulence and the secondary flow below it. LESs are validated first using experimental and direct numerical simulation data. At increasing water depth, the magnitude of water surface fluctuations increases with increasing strength of the main secondary flow. Visualizations of the instantaneous water surface and the turbulent flow underneath reveal that thin surface waves are generated by flow meandering in the shallower case, whereas surface waves in the deeper cases are influenced by the vertical velocity fluctuation. Pre-multiplied spectra of the water surface fluctuation, h′, provide further evidence of the origin of the surface waves. In the shallow flow, the peak frequency of the h′ spectra is consistent with the peak frequency of the u′ and v′ spectra, while for deeper flows, it agrees more with the w′ spectra. Furthermore, the transport patterns of the surface waves are investigated by the wavenumber-frequency spectra. Three types of surface waves are observed in the wavenumber-frequency spectra, i.e., (1) convective waves with phase velocity equaling the surface velocity, (2) irrotational dispersive gravity-capillary waves, and (3) stationary waves caused by secondary currents.

Reynolds stress modeling of supercritical narrow channel flows using OpenFOAM: Secondary currents and turbulent flow characteristics

“…Furthermore, the maximum resultant secondary velocity varies from 1.7% to 2.13% of U max or from 1.92% to 2.44% of U , which agrees well with previous experimental and numerical results: Broglia et al (2003); Naot and Rodi (1982); Nezu and Nakagawa (1986); Nezu and Rodi (1985); Tominaga et al (1989) but is lower than the DNS values obtained by Nikitin (2021). Comparatively higher values are also reported in partial filled pipes (Brosda and Manhart, 2022;Liu et al, 2022aLiu et al, , 2022bNg et al, 2018). Interestingly, a decrease in the a r reduces the maximum normalized lateral velocity and the maximum normalized resultant secondary velocity for the tested rectangular channel flows.…”

“…Broglia et al (2003), Grega et al (2002), Kang and Choi (2006), and Nikitin (2021) observed similar vortex in their LES (large-eddy simulation), experiments, RSM (Reynolds stress model), and DNS, respectively. These are also reported in partial filled pipes especially with 52% pipe filling (Brosda and Manhart, 2022;Liu et al, 2022aLiu et al, , 2022bNg et al, 2018;Wu et al, 2018). However, the bottom vortex is not found in those studies due to the absence of a solid corner between the walls.…”

“…Rodr ıguez and Garc ıa (2008) observed similar magnitude for roughness induced secondary currents in channels with a r > 5. In the case of partial filled circular pipes, the strength of the secondary current increases with the filling ratio, and magnitudes up to 7.1% of U are reported (Brosda and Manhart, 2022;Liu et al, 2022aLiu et al, , 2022bNg et al, 2018;Wu et al, 2018).…”

“…For a r > 5, the turbulence-driven secondary currents were observed experimentally and numerically throughout the channel width due to difference in the roughness between the bed and the sidewall (Albayrak and Lemmin, 2011;Rodr ıguez and Garc ıa, 2008;Talebpour and Liu, 2019) and because of alternate bed roughness (Kundu et al, 2022;Stoesser et al, 2015;Talebpour and Liu, 2019;Wang and Cheng, 2005). In partial filled pipes, the secondary currents are observed due to the mixed corner between the free surface and the solid boundary, and they have been studied more in recent times using experiments and simulations (Brosda and Manhart, 2022;Liu et al, 2022aLiu et al, , 2022bNg et al, 2018Ng et al, , 2021. The maximum secondary velocity for narrow open channels is typically around 1.5%-3% of either the maximum longitudinal velocity U max or the bulk velocity U (Albayrak and Lemmin, 2011;Broglia et al, 2003;Naot and Rodi, 1982;Nezu and Nakagawa, 1986;Nezu and Rodi, 1985;Tominaga et al, 1989).…”

Roles of Tidal Cycling, Hyporheic Exchange and Bioirrigation on Metal Release From Estuary Sediments

“…Large-eddy simulation (LES) method, using the code Hydro3D (Liu et al, 2022a(Liu et al, , 2022b, is applied in solving the filtered N-S equations for incompressible, unsteady and viscous flow:…”