Use this URL to cite or link to this record in EThOS:
Title: Velocity distribution prediction in rectangular and compound channels under smooth and rough flow conditions
Author: Alawadi, W. A. A. K.
ISNI:       0000 0004 7960 2804
Awarding Body: University of Salford
Current Institution: University of Salford
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Many practical problems in river engineering require predictions of velocity distributions in single and compound open channels. In this study, velocity distributions in rectangular and compound open channels under different flow regimes, in terms of roughness and turbulence conditions, were investigated experimentally and numerically. The detailed results from the experiments and numerical simulations were then used to assess and improve two velocity distribution models, which are widely used for rectangular and compound channel flows. The focus was mainly on the flow in narrow open channels. Two sets of experiments were conducted on rectangular and compound channels that had different hydraulic and roughness characteristics. In these experiments, three flow regimes in terms of the roughness (smooth, transitional and rough regimes) were established by changing roughness of the bed. Detailed measurements of velocity distributions were carried out to study the effect of roughness on the non-uniformity of velocity distributions in such channels. The channel flows tested in the experiments were also simulated using Computational Fluid Dynamic (CFD) models. The CFD models were developed and run using CFX package (v.15). The effect of the roughness on the velocity distributions in outer region was found to be significant. The non-uniformity of the velocity distribution, which can be described by the energy coefficient (α) and momentum coefficient (β), was considerably high in fully rough flow regime than in smooth and transitionally rough flows. The values of α are ranging from 1.07 to 1.16 in fully rough regimes while α values are varying from 1.05 to 1.13 in the corresponding smooth flow cases. The similar trend can be seen for β values, where β lies within a range of (1.013 - 1.032) in smooth flows and (1.022 - 1.045) in rough flows. It was also found that the velocity distribution coefficients (α and β) can be related to roughness Reynolds number (Re*) and aspect ratio (Ar) in rectangular channels and to the relative depth (Dr) in compound channels. The relationships for the velocity distribution coefficients (α or β) proposed in the present study can be used to describe the non-uniformity of the velocity distributions and solve engineering flow problems that depend on the velocity distribution such as sediments and pollution transports. The detailed CFD results were also used as part of an investigation in the significance of secondary flows and turbulent eddy viscosity in the calculations of primary velocity distributions. Two analytical models for velocity distribution were used in this study. For rectangular channel flows, the dip-modified log wake law (DMLW-law) was used. The DMLW-law accounts for the effects of the secondary flow and turbulence by two representative parameters. These two parameters are the dip correction factor (μ) and the wake strength parameter (Π). The analytical model developed by Shiono and Knight (1988, 1991), which called SKM model, was applied for depth-averaged velocity calculations in compound channels. The SKM model relies on three parameters, namely friction factor (f), dimensionless eddy viscosity (λ), and secondary flow term (Γ). For velocity prediction in rectangular channels, expressions for estimating the values of μ and Π has been proposed in this study. Comparison with the experimental results indicated that the proposed expressions to calibrate the parameters (Π and μ) can provide efficiency to the application of the analytical model (DMLW-law) for rectangular channel flows under different flow regimes. For compound channel flows, it was found that the application of analytical SKM model with the existing expressions for the three parameters (f, λ, Γ) does not fit to the compound channel with narrow floodplains such as the one used in the present study. Therefore, the detailed Computational Fluid Dynamics (CFD) results were used to modify the traditional expressions. The modified expressions for λ and Г parameters proposed in the present study were proved to give better predictions for the narrow compound channel flows than the traditional calibration expressions.
Supervisor: Not available Sponsor: Iraqi Government
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available