Journal of Hydraulic Engineering

Hydraulic jumps are characterized by strong flow turbulence, flow aeration, and three-dimensional flow motions. Whereas comprehensive research into hydraulic jumps on smooth bed has improved the understanding of flow aeration and turbulence, limited research has been done of hydraulic jumps on rough beds. Herein novel experiments were conducted in hydraulic jumps on uniformly-distributed bed macroroughness. Both air-water flow patterns and basic air-water flow properties were investigated. The hydraulic jumps on the rough bed exhibited some remarkable differences compared with smooth bed jumps including some preaeration of the flow upstream of the jump, an upwards shift of the jump roller and a clear water-flow region underneath the jump. Air-water flow measurements were conducted with a phase-detection probe, showing similar distributions of air-water flow properties for the rough and smooth bed jumps. Comparative analyses highlighted some distinctive effects of the bed roughness including an upwards shift of the hydraulic jump and an increase in bubble count rate and void fractions in the region close to the jump toe. In the second half of the hydraulic jumps, the rough bed led to a clear-water region with large-scale vortices which were advected downstream. The present study highlighted the potential that improved and nonstandard invert designs may have for flow manipulations and design enhancements.

Sediment transport capacity is one of the key parameters in river sediment dynamics. New formulas for fractional and total transport capacity for nonuniform sediment were established and are presented in this paper, based on energy equilibrium and the fact that flow resistance decreases as sediment concentration increases. As this study shows, the transport capacities for a single fraction and total sediment are logarithmic functions of the sediment transport parameter (STP). The proposed formulas based on logarithmic law have many advantages: (1) the critical velocity for suspension of nonuniform sediment is innately embodied with simple form in the logarithmic formulas; (2) the interaction between different size particles was considered in settling velocity for any fraction of nonuniform sediment and the mean settling velocity for nonuniform sediment is equal to the geometric average of settling velocities for a sediment fraction; and (3) the formulas have the same form for fractional and total transport capacity and self-adaptability to uniform sediment. The logarithmic law was verified against laboratory and field data, and was found to perform better than power law. The derived critical velocity for suspension with interaction between nonuniform particles taken into account in the fractional and total critical velocity for suspension was also tested by using laboratory data, and it performs quite well for a wide range of particle diameters despite its simple form. The logarithmic law appropriately depicts the whole suspension process from initiation to transport and thus provides insight into the suspended motion of nonuniform sediment.

Floods in rivers can be modeled with one dimensional flood routing methods, which include an approximate solution of the one-dimensional (1D) Saint-Venant momentum equation and a solution of the corresponding mass conservation equation for a storage element representing the river reach. The accuracy of a method applied to a natural river can be improved by introducing an averaged compound section along the reach. But it is shown analytically and numerically that the irregular geometry and hydraulic roughness contribute to the section dimensions. These contributions can be introduced either by the concept of storage areas in the model section or by routing flows along a randomly irregular model of the reach. By compartmentalizing the model section and adopting analytical, parameterized functions for the terrain in each compartment, an optimal section can be obtained by comparing the predicted with the observed hydrographs downstream. The accuracy of the routing method normally increases as the number of parameters is increased for a sequence of standard cross sections. This new method is applied to the River Wye, United Kingdom.

Geysers are explosive eruptions of air-water mixture from manholes in drainage systems. When the design capacity of urban storm water drainage systems is exceeded during extreme rainfall, rapid inflows into the drainage network can lead to air-water interactions that give rise to geysers—causing damage to the water infrastructure and threatening human lives. Although extensive research has revealed the role of entrapped air in causing large pressure transients in drainage tunnels, the mechanism of geyser formation remains elusive. In this study, an unsteady three-dimensional (3D) computational fluid dynamics (CFD) model is developed to simulate the pressure transients and air-water interactions during geyser events using the volume-of-fluid (VOF) technique. Extensive numerical simulations are conducted to study the air-pocket dynamics caused by release of trapped air from a horizontal tunnel into a vertical riser. Model predictions of the air-water interface in the vertical shaft are in good agreement with laboratory measurements by a high-speed camera; the mechanism for the formation of geysers is elucidated. It is found that during a geyser event, compression of the air pocket in the riser can lead to rapid acceleration of the overlying water column and its expulsion from the riser; the air-pocket pressure is significantly higher than the hydrostatic pressure of the overlying water column, and the velocity is greater than that of a Taylor bubble. Comprehensive numerical modeling has been conducted to study the effect of the vertical shaft diameter, the upstream pressure head and the air-pocket volume on the air-pocket dynamics; the results show that geyser formation is primarily controlled by the riser to tunnel diameter ratio, $D_r/D$. 3D CFD simulations have also been carried out for an idealized prototype drainage system; it is shown that the geyser behavior can be characterized by extremely large vertical air velocities that result in dispersed air-water mixtures often observed in the field.

The resistance of an open-channel flow under the influence of bending deformation of submerged flexible vegetation canopy was experimentally studied in an indoor water flume. Altogether, 27 tests under the cross-combinations of three bulk flow velocities ($V$), three vegetation elastic moduli ($E$), and three vegetation stem thicknesses ($b$) were conducted. The total drag force experienced by the vegetation stem was accurately calculated based on the measured vertical profiles of flow velocity and the second order turbulent momentum in front of an individual plant, and this enabled the estimation of the overall drag coefficient ($C_D$) for bulk flow. The results show that the bending deformation of vegetation helps to transfer momentum downward toward the channel bed, thus helping to reduce the overall resistance to flow. The Cauchy number ($C_Y$) is an important parameter representing the bending property of the vegetation in open-channel flow; it’s functionally related to the resistance of flow. For small deformation configurations, the bending angle ($\theta$) and the relative bending displacement (${\delta }_z/h_p$) are linearly dependent on $C_Y$ and ${(C_Y)}^2$, respectively. The $C_D$ and Manning’s $n$ for the bulk flow are both power law functions of $C_Y$. The semiempirical formulas of $C_D$ and $n$ proposed in this study are expected to apply to further experimental studies and river engineering.

Under relatively low heads, the occurrence of nappe oscillation, also known as nappe vibration, may be observed on hydraulic structures with a free overfall, such as weirs, crest gates, and fountains. This phenomenon is characterized by oscillations of the thin flow nappe cascading downstream of the crest and results in a disturbing noise production that increases the environmental and societal impacts of the hydraulic structure. Given limited information available regarding the physical processes involved in this phenomenon, a detailed investigation has been undertaken to characterize the flow for free-overfall structures where nappe oscillation may be of concern. The research is conducted on a prototype-scale linear weir model (weir length of 3.5 m and a fall height of 3 m) using high-speed cameras and audio equipment to characterize the nappe oscillation. The paper presents the quantitative characteristics of the nappe oscillation gained from images and sound analysis, especially in terms of frequency, for aerated and confined nappe configurations.

The variation of upstream slope of a submerged weir could significantly alter the local flow pattern and sediment transport. This paper presents an experimental investigation into local scour at submerged weirs with different upstream weir slope angles. In this study, 62 tests are carried out in a tilting recirculating flume, including 38 tests using coarse sand and 24 tests using fine sand, to investigate the effects of upstream weir slope on the local scour at a submerged weir. The tests of each sand use four weirs with different upstream weir slopes. For clear-water scour, no upstream scour depth is observed at the sloped weirs, and downstream scour depth is independent of the upstream weir slope. For live-bed scour, a flatter upstream weir slope reduces upstream scour depth for both sands. The results for fine sand show that when the slope is gentler than a critical angle ${\alpha }_0$, the effects of upstream weir slope on upstream scour depth are minimized, such that the upstream scour depth no longer decreases with decreasing weir slope angle. The existence of upstream weir slope slightly reduces the downstream scour depth when the sediment is transported as bedload. When the sediment transport mode is suspension load, the downstream scour depth is independent of the weir slope. New predictors for the average scour depth including the effects of upstream weir slope are developed. A relationship between the average and maximum scour depth is also proposed.

Accurate predictions of sediment loads released by sewer overflow discharges are important for being able to provide protection to vulnerable receiving waters. These predictions are sensitive to the estimated sediment characteristics and on the site conditions of in-pipe deposit formation. Their application without a detailed analysis and understanding of the initial conditions under which in-sewer deposits were formed normally results in very poor estimations. In this study, in-sewer sediment samples deposited during dry periods in a combined sewer system were collected, and their properties assessed. Parameters in a sediment transport relationship for in-pipe deposits were estimated by simulating the in-pipe deposit formation conditions in laboratory erosion tests. The measured parameters were then used to simulate sediment transport through a small combined sewer network for a number of rain events for which rainfall, hydraulic, and water quality data were available. Results showed that the model of Skipworth can provide good predictions of the sediment loads released from such in-sewer deposits. The experimentally derived calibration parameters used with Skipworth’s model allowed for a realistic simulation of the in-sewer sediment behavior, and so can be used to accurately estimate the sediment load released from combined sewer systems during rainfall events.

This article addresses the issues, related to transitioning junction flow patterns, that afflict the hydraulic simulation of pipe network. At transition points, the values of head change caused by a junction are not unique, and thus the head change versus flow rate function is not differentiable. Nonlinear solution algorithms including Newton’s method are likely to fail when faced with such discontinuities, or at best produce solutions that depend on the initial guess for starting the solution. The difficulties can be removed by replacing dual junction head changes with unique values that can be determined based on the transitioning flow patterns and then incorporating the unique values into smooth interpolating functions that replace the step functions. Splines in tension serve as the interpolation functions. Existing flow solvers can be retrofitted with the proposed procedure. Implementation of this procedure is demonstrated for T junctions by solving several steady-state and time-accurate network flows. In all cases the strong convergence property of Newton’s method is restored.

This study examines the generation of warning waves with prescribed characteristics in a mountain stream. The authors determine which dam release will generate the desired warning wave. They solve this inverse problem following a two-model approach. An analytical kinematic model is used for a preliminary design of the dam release, and a detailed two-dimensional (2D) fully dynamic model is used to converge to the final solution. Although the presented case study is far from an idealized academic case, the analytical model performs well and, beyond its role for preliminary design, turns out to be of prime interest for both understanding and discussing the results of the detailed 2D model. The complex interactions between the release hydrograph, geometry of the river, and the friction formula are brought to light by the analytical model, which highlights the complementarity of both models and the usefulness of such a two-model approach.

Dam foundation seepage is a major concern for the stability of a hydropower project in southwest China (referred to here as Project X). The complicated geological structure, multiple grouting curtains, and adjustable drainage system of Project X significantly increase the complexity of the dam foundation seepage field. In this work, a three-dimensional (3D) air-water two-phase seepage mathematical model, coupled with the volume of fluid (VOF) method, is developed based on the computational fluid dynamics (CFD) approach. The 3D unified model of Project X, including strata, unfavorable geological bodies, grouting curtains, adjustable drainage system, consolidation grouting, and gravity dam body, is later established in an integrated manner for the entire dam area. Furthermore, an extended “fissure to represent line of holes” method is proposed to model the drainage holes with adjustable opening degrees. Subsequently, the distributions of the water head, hydraulic gradient, and uplift pressure, both before and after adjusting the drainage system, are analyzed. The overall behavior of the seepage control system and the influence of the opening degrees of the drainage holes on the seepage field are comprehensively discussed, and an optimized operating state is recommended for the drainage system. The discussion reveals that the maximum hydraulic gradient at the grouting curtains decreases and the average uplift pressure on the dam foundation surface increases with decreasing water discharge of the drainage system. Comparisons with in situ monitoring data demonstrate the reliability of the proposed methodology.

Previous hydraulic studies of Archimedes screw power generators (ASGs) have been mostly at laboratory scale. The validity of scaling up models based on these studies for application in field-scale ASGs has been a major research gap. This study developed a nondimensional artificial neural networks (ANN) model to predict shaft power of an ASG using extensive multiscale data sets. The model was trained using 583 experimental observations from laboratory-scale and field-scale Archimedes screws over a wide range of volume flow rates, operating speeds, and outlet water levels. The input training data was nondimensionalized to allow for scaling between different size screws. The trained ANN model was used to predict the power output of a different ASG with an average error of 6%. It was found that an ANN can be trained to provide reasonably accurate predictions of ASG power if the training data includes a range of ASG sizes.

Incipient motion of seashells under a unidirectional flow of water in a horizontal open channel over surfaces with differing roughnesses and shell placements is investigated experimentally. Drag and lift forces are measured with strain gauges and, from these measurements, the lift and drag coefficients are inferred. The various shell shapes that are investigated in this study are the Patellogastropoda, commonly known as the limpet shell, and the Mytilus galloprovincialis, commonly known as the blue mussel, as well as fossilized shells and shell fragments obtained through a dredging campaign. The bed shear stresses at incipient motion are measured and determined from an existing model for shells and the results are compared to the measurements and published data. It is indicated that a critical Shields parameter for shells may not always be an appropriate measure to use to estimate the incipient motion of shells.

An experimental study has been carried out on the injection of a single piston into a confined cross-flow of a round pipe. Five liters per minute have been used for the cross-flow with various values of injection flows. The study concentrates on the performance of the variable displacement radial injection pump (VDRIP) with a single piston driven by a cam action. The displacement action could be divided into three volumetric efficiency zones: high, moderate, and low. The recommended operating zone ranges between 30 and 100% pump displacement, indicating maximum volumetric efficiencies of 95–99.8%, respectively. The results also indicate that increasing the pump displacement enhances the volumetric efficiency while reducing the mechanical efficiency especially at low pressure ranges of up to 4 MPa (40 bar).

Many water intakes and diversions are prone to blockage by drifting material, notably vegetation and ice. A way to avert blockage is by means of a deflector-wall and floating debris boom. However, the wall may be subject to local scour and cause bed sediment to accumulate at the intake or diversion entrance. This paper presents data and observations from live-bed, laboratory flume tests showing that a porous deflector wall reduces scour depth and height of sediment accumulation. The flume tests used deflector walls angled at 90° and 30° to the bank, as well as varying levels of wall porosity. A minimum value deflector-wall porosity of 50–75% is recommended for design. This range ensures minimal scour or accumulation of sediment at the wall and suitable wall performance as a deflector.