Journal of Hydraulic Engineering

Entrapped air blocking the flow in pipeline systems is a common cause of increased pumping costs. At present, air is generally removed via valves or pipeline excavation and drilling. This becomes inefficient in large networks where the precise location of the air is unknown. Fluid transients are a potential tool for detecting and locating air in pipelines. The effect of a stationary air pocket part of the way along the pipe, which occupies the main flow path and acts as a blockage without causing a hydraulic jump or column separation, has not previously been studied experimentally. This paper presents experimental results for a transient pulse interacting with an in-line air pocket for a range of pocket sizes and system pressures. In accordance with the impedance theory, the reflective power of the air increases with pocket size. Other notable characteristics of the interaction include frequency-dependent transmissivity, an out-of-phase reflection, and a substantial reflection under zero base flow. These effects set air pockets apart from solid blockages, allowing a transient detection methodology to differentiate between the two cases, although they have similar effects at steady-state.

Plastic/viscoelastic pipes such as polyvinyl chloride (PVC), polyethylene (PE), and high-density polyethylene (HDPE) pipes have been increasingly applied in fluid piping systems. The understanding of the hydrodynamic behavior and features of such pipe materials are important in their practical applications. This paper develops an effective frequency-domain transient-based method (FDTBM) for the efficient and accurate identification of viscoelastic parameters of plastic pipes. The analytical expression of transient frequency response in a typical viscoelastic pipeline system is first derived to describe the dependence relationship among different factors and coefficients in the system (e.g., pipe and fluid properties). The obtained result is then applied to inversely identify the viscoelastic parameters of plastic pipes under different flow and operational conditions. A multistage analysis framework is proposed to enhance the robustness and effectiveness of the proposed FDTBM to obtain accurate and unique solutions of viscoelastic parameters for the plastic pipes used in this study. The proposed method and analysis framework have been validated and evaluated through various experimental tests and numerical simulations.

Geysers can occur in storm sewer systems during heavy storm events, leading to public safety concerns. In this study, a physical model as well as a transient three-dimensional computational fluid dynamic model were used to investigate potential mitigation methods. Two types of geysers, caused by the rapid filling and release of air pockets, were simulated numerically and validated with physical measurements. For geyser mitigation, benching, water recirculation chambers, and orifice plates were proposed and assessed numerically, and the effect of an inflow control method was also studied. The numerically tested geyser mitigation methods showed positive effects on reducing the amount of water erupting out of the manhole, but they also increased the amount of air transported downstream, which could cause issues further downstream. Water recirculation chambers mitigated the water erupting out of manholes without increasing the pressure in the pipes significantly. The strategy of sealing the riser was able to completely mitigate geyser events, but it also generated higher pressure in the upstream pipe and transported a higher amount of air downstream. It was found that using orifice plates to mitigate the second type geyser can generate a water hammer–like pressure pattern in the riser, with a peak pressure reaching about seven times the driving pressure when the orifice diameter is 0.2 times the riser diameter. With the inflow control method, the physically measured pressure variation was reduced, and the onset of geysers was visually reduced.

Swirling jets induced by propellers have a significant impact on riverine and marine environments. Erosion around quay structures and contaminant dispersion in harbors or along navigation channels are examples of these effects. Simulating propeller-induced jet velocities that extend from near to far fields is the first step in elucidating this phenomenon and seeking solutions. Most previous investigations have proposed semiempirical equations for estimating velocity. Different formulas have been derived for velocity in both the zone of flow establishment and in the zone of established flow. In this study, the mean and turbulent flow fields downstream from a rotating propeller were investigated using an acoustic Doppler velocity profiler (ADVP). The results showed that the propeller-induced swirling effect was directly proportional to the propeller rotational speed, but decreased as the distance from the efflux plane increased. However, this effect has little effect on the radial distribution of the axial mean velocity and the decay of the maximum velocity. A point-source method was employed, along with the adjustment of the jet axial momentum spreading coefficient, to provide an analytical solution for the axial mean velocity within the propeller jet. Comparisons with measured data from the published literature and from the present study for both flat and equilibrium scour beds showed good agreement. Based on results of this study, the axial velocity profile calculated using the point-source method may be applied to investigate the scour and sediment resuspension induced by a propeller jet.

A rotary hydrostatic pressure machine (RHPM) is a hydropower converter for very low head applications (less than 2.5 m). An RHPM can be installed in straight canals, because its wheel can generate a hydraulic head as a result of the created dam effect. Some new experimental results on an RHPM are presented in this paper. Because of the asymmetric shape of the wheel, the effects of the canal width and the lateral position of the wheel inside the canal have been studied. The water entrance process has been found to be very sensitive to the presence of the canal walls: the results pertaining to the side of the wheel where the blades touch the water first show that the minimum distance between the canal wall and the side of the wheel should be 0.3 times the width of the wheel in order to maintain optimal wheel performance. Upstream and downstream water level variations have also been studied. Higher upstream water levels than the water levels adopted in previous studies (coincident with the upper edge of the hub) can lead to benefits. The results shown in this manuscript may be useful for practical applications in order to understand the optimal position of an RHPM inside a canal, the minimum canal width, and the optimal water levels.

Under the umbrella of one-dimensional (1D) transient flow analysis, unsteady friction models are popular tools to better represent the decay in pressure signal associated with high wall shear stresses. There are generally two families of unsteady friction models, namely those based on instantaneous acceleration (IAB) and those based on convolution (CB). This paper considers a classic laminar water hammer and compares two unsteady friction models with both experimental results and those obtained using a high-resolution computation fluid dynamics (CFD) model, with CFD having a more native representation of frictional losses. Because dissipation is fundamentally an energetic phenomenon, an integrated total energy (ITE) approach is used here as a lens to view and compare results. The capabilities and limitations of the ITE approach are discussed. More specifically, the ITE method is used not only to quantify dissipation but also to spatially resolve when, where, and how energy is dissipated in each model. Most notably, CFD and CB models are found to differ significantly for positive flows, with CFD results exhibiting little difference between positive and zero flows.

The present study develops a refined water hammer model by postulating the presence of a water jet within the reservoir as the water hammer wave is reflected at the pipe inlet. The waterjet leads to a new boundary expression that induces a smoothing effect on the pressure wave front by introducing a smoothing factor, which is determined by minimizing the difference between pressure histories predicted numerically and those measured experimentally. The proposed boundary expression is applied in conjunction with the quasi-steady, the Brunone, and the Zielke friction water hammer models and is shown to more accurately replicate the peak pressure magnitudes, pressure wave shape, and the phase shifting when compared to experimental results for a variety of pipe systems and steady flow conditions. Using the quasi-steady and the Brunone models, the study shows that both the friction stresses and the proposed boundary expression attenuate the wave amplitude. However, only the proposed boundary expression smoothens the pressure distribution and delays the wave reflection in a manner consistent with experimental results. While the conventional Zielke model is shown to attenuate and smoothen the pressure distribution, the proposed boundary expression is shown to introduce additional damping and further phase shifting.

Properly modeling the sediment transport under water waves is critical to understanding morphodynamic processes such as, e.g., sandbar migration at beaches. In the last two decades, the influence of the wave asymmetries (mainly velocity and acceleration skewnesses) has been shown to be a key issue for explaining onshore sandbar migrations under mild wave conditions when wave-induced currents are small. However, there is still a lack of knowledge, and the expressions that relate the sediment transport to the velocity and/or acceleration skewnesses are site-specific. In this work, an intrawave process-based sediment transport model was applied to provide a general expression for bedload sediment transport under wave action within a proper dimensional framework. The new expression is not site-specific, and it is based on the process-based model which was checked against experimental data. Furthermore, the obtained equation has some similarities with expressions in the literature that are discussed.

The local scour and sediment deposition at a bridge pier during flood waves is analyzed to investigate the effects of different flow and sediment regimes (regulated and unregulated discharges with or without excess sediment supply). Concurrent field measurements of scour and streamflow were performed during 6 days at the Rapel Bridge, over the Rapel River (mean annual discharge, $\mathrm{QMA}=172{\mathrm{m}}^3/\mathrm{s}$), located in Central Chile (71°44’9” W, 33°56’22” S). During the measurements, river discharge was regulated by the operation of a hydropower plant, located 24 km upstream of the bridge, which follows a daily hydropeaking scheme. A simple mathematical model of scour and deposition is proposed, and field measurements are used to estimate optimal model parameters and to evaluate model performance. The model was applied to pre- and postdam scenarios to compare expected scour caused by a natural flow regime and by hydropeaking considering different excess sediment supply. Results show that the ultrasonic scour sensor is reliable for real scour monitoring under the presented field conditions. A single and easy-to-perform measurement of scour evolution during one flood was enough for estimation of optimal model parameters. The calibrated model reproduced measured scour and deposition in a verification case with high precision, i.e., root mean square error, $\mathrm{RMSE}=0.023\mathrm{m}$ and Nash-Sutcliffe efficiency, $\mathrm{NSE}=0.96$. The model application showed that scour and deposition are very sensitive to the excess sediment supply. After two years, scour in the predam scenario resulted higher than in the postdam scenario when equilibrium conditions or no sediment deposition occurred. In case of equilibrium conditions and excess sediment supply deposition occurred during the falling limb of floods producing important refilling of the scour hole. However, floods’ high peak-discharges and excess sediment supply produced high scour depths of comparable magnitude as those after the two years hydrograph, which occurred only briefly around the peak discharges before sediment deposition, illustrating the complex interactions between flow and sediment in time, with important consequences for monitoring of bridge pier scour in the field and for forensic analyses.

In a previous paper, the authors proposed a transient-based area reconstruction method to identify multiple blockages in pipe systems without prior knowledge of their geometry and number, resulting in very accurate numerical results. In this paper, the accuracy and robustness of the method is verified in a real laboratory system in which the experimental setup considers mild ($\approx 35\%$ reduction in area) and severe ($\approx 85\%$ reduction in area) blockage cases. It is found that both blockage cases can be successfully identified; however, there are some differences between the actual and estimated pipe areas. These differences are shown to be mainly due to the limited frequency bandwidth (FBW) of the generated transient wave, causing three major errors: (1) distortion of the image of the step changes in area, (2) uneven distortion of the image of the two sharp changes of the area, and (3) underestimation of the blockage severity. These limitations and errors are discussed in detail in the paper.

Stochastic Galerkin methods can quantify uncertainty at a fraction of the computational expense of conventional Monte Carlo techniques, but such methods have rarely been studied for modeling shallow water flows. Existing stochastic shallow flow models are not well-balanced, and their assessment has been limited to stochastic flows with smooth probability distributions. This paper addresses these limitations by formulating a one-dimensional stochastic Galerkin shallow flow model using a low-order Wiener-Hermite polynomial chaos expansion with a finite volume Godunov-type approach, incorporating the surface gradient method to guarantee well-balancing. Preservation of a lake at rest over uncertain topography is verified analytically and numerically. The model is also assessed using flows with discontinuous and highly non-Gaussian probability distributions. Prescribing constant inflow over uncertain topography, the model converges on a steady-state flow that is subcritical or transcritical depending on the topography elevation. Using only four Wiener-Hermite basis functions, the model produces probability distributions comparable to those from a Monte Carlo reference simulation with 2,000 iterations while executing about 100 times faster. Accompanying model software and simulation data are openly available online.

Education provides the foundation for a future generation to advance the field of hydraulic engineering. This paper discussed lessons learned and techniques for educating students and new civil engineers in the field of hydraulic engineering from both academic and industry perspectives. Instances from the last 5 years were presented in which applied hydraulic engineering and research were used for education, in the classroom (lectures, experiments, and projects) and in industry (analyses, studies, design, and construction). This paper highlighted several methods that teach hydraulic design via anticipatory learning, collaborative innovation, and real-world connectivity between the classroom, workplace, and hydraulic infrastructure systems. Additionally, this paper discussed observed challenges facing the current generation of young engineers in ever-evolving classrooms and workplaces. The goals of the paper were to foster discussion regarding innovative teaching activities that deepen understanding and application, foster creativity, and strengthen the bridge between academia and industry such that education is viewed as a mutually beneficial, ongoing process.

Accurate estimates of continental-scale channel flows are needed to understand spatiotemporal variability in water supplies and the water balance. At regional scales, models of connecting channel flows are commonly used to understand how variability in the water cycle propagates into engineering-oriented decisions related to water quantity and water quality management. Since 1958, deterministic monthly flows have been calculated for all of the connecting channels of the Great Lakes–St. Lawrence River system through a binational, multiagency coordination process. This article provides a review of these historical estimates, most of which have never appeared (or appeared decades ago) in the peer-reviewed literature, and compares them to new estimates from a novel statistical water balance model. This new model was developed using a variety of water balance component estimates across the entire Great Lakes system and includes an explicit expression of uncertainty. The findings of this research indicate that the historical range of deterministic channel flow estimates is similar to the range of uncertainty represented by the authors’ statistical water balance model. Findings also indicate that historical internationally coordinated flows for this massive lake and river system from the late 1990s through 2009 appear to be negatively biased and may need to be revised. The proposed statistical water balance model provides an ideal platform for implementing this revision and other future updates to regional water balance information.

Equilibrium clear-water scour depths predicted by existing equations based on the peak flow rate of a flood hydrograph can be overly conservative for submerged weirs in mild-slope sandy rivers, because the equilibrium timescale normally is much larger than the timescale of a flood event. Therefore it is important to understand the clear-water scour development at submerged weirs in order to improve the accuracy of scour estimation. This paper presents new data on the temporal evolution of the downstream clear-water scour depth at a submerged weir under steady flow conditions and analyzed the impacts of flow intensity, weir height, sand coarseness parameter, and weir slope on the development of the downstream clear-water scour depth. This study defined a dimensionless equilibrium time scale $t_e^{*}$ and evaluated its dependencies on the aforementioned parameters. Predictors for $t_e^{*}$ and the temporal downstream clear-water scour depth at submerged weirs are proposed.