World Environmental and Water Resources Congress 2018: Hydraulics and Waterways, Water Distribution Systems Analysis, and Smart Water

Stormwater geysers are an undesired operational issue that occurs when runoff in stormwater systems returns to the grade through vertical shafts when these systems are undergoing rapid filling. Among other mechanisms, the release of large entrapped air pockets through ventilation shafts is a likely cause of many recently reported geyser events. The existing investigations upon air pocket release induced stormwater geyser indicate that the water level displacement and velocity in the shafts, representing the geyser intensity, is related to the size of released air pocket, size of the ventilation shaft, and the initial standing water depth in the shaft. However, there has been fewer studies considering the release of various discrete air pockets through vertical shafts. Considering that various reported geysers are linked to a sequence of air-water releases, it is possible that these are linked to the release of various air pockets through shafts. The present research implements a series of computational fluid dynamics (CFD) simulations on the stormwater geyser episodes with the release of varying number and size of air pockets. It is expected to qualitatively understanding the process of multiple air pocket release and quantitatively investigate the flow characteristics of the geyser with a sequence of breakthroughs.

In thermal power plants, the function of a condensate system is to remove and recycle condensed steam from the steam turbine and to deliver condensate to the feedwater system and/or to other equipment depending on the configuration of the power plant. Hydraulic transients are created when sudden changes in flow rates occur due to operations of pumps and valves in the pumping systems. The pressures generated due to these transient events may be high enough to damage or even cause catastrophic failure of pipelines. Therefore, it is very important to perform the hydraulic transient analysis to identify problem areas and provide mitigation measures to avoid potential piping over-pressure or damage to pipes. The current study simulates numerically the transient events and calculate transient pressures and dynamic loads associated with condensate pump operations (startup, trip, switch-over) and valve operations (opening, closing) under normal (pump startup, shutdown) and accidental operating conditions (power failure, accidental valve closure, etc.) of a condensate system. Different operating conditions are analyzed and transient mitigations are evaluated and recommended to avoid extreme transient pressures and associated high dynamic loads to the piping segments. Furthermore, in order to minimize check valve slamming during a pump trip or switch-over, different kinds of check valves (ball, swing, resilient hinge, tilted disc, dual plate, and center guided silent) with different dynamic characteristics are evaluated, and the resulting dynamic loads are estimated. The magnitude of the dynamic loadings for different kinds of check valves are compared and a recommendation for the correct selection of check valves is proposed.

Three-dimensional (3D) flow and sediment transport modelling, with non-hydrostatic assumption, is rarely carried out in hydraulic applications. The use is often limited to academic research to gain physical understanding of complex flows or sediment transport processes. Such a modelling is labor intensive and demands prohibitive computing resources. Preparation of 3D meshes is a tedious process that requires the use of commercial software. Model execution often requires access to super computers unavailable to most engineers. In this study, we report the research and development effort at the Bureau of Reclamation in developing a 3D model for future general-purpose and license-free modelling of practical river applications. Several new techniques are proposed to achieve the objectives. First, an unstructured physical-coordinate sigma mesh is proposed and developed. With this mesh, only a two-dimensional (2D) horizontal mesh is needed, which may assume arbitrary polygonal shapes. The vertical mesh is automatically generated conforming to the bed and free surface elevations. Second, the mesh is allowed to move in the vertical direction by adopting the Lagrangian-Eulerian formulation; this way, the mesh points move adaptively according to the free surface and bed changes. Third and final, a special 3D sediment transport module is developed and coupled to the 3D flow solver. The outcome is a general 3D model that may be used to simulate local scours in practical rivers. In the paper, we will highlight the numerical methods developed, and present model test, and verification results in comparison with available experimental data.

The hydraulics unit in MnDOT’s bridge office applied for a research grant to develop in-house underwater acoustic 3D imaging capabilities. This research report presents both stationary and mobile scanning techniques, outlines the setup of both systems, discusses field operations, summarizes the data analysis and post-processing of images, and reviews lessons learned. Several case studies will frame a discussion of the capabilities and limitations of 3D acoustic imaging for underwater bridge inspection. The case studies provide examples of different applications of this technology. Underwater acoustic imaging has been shown to have real value to bridge inspectors and owners for a variety of applications.

Fox River is known as a major supplier of total suspended sediments (TSS) into Green Bay (GB). Therefore, understanding the concentration of TSS transported by Fox River to the GB is crucial in restoration and coastal management plans, and for any computational efforts. Currently, there is not a continuous record of in situ TSS observations in Fox River. However, United States Geological Survey (USGS) has a continuous record of turbidity (TU) measurements at the Fox River mouth. In addition, there are point measurements of TSS and TU by GB NEW Water, in the bay. In this study, two approaches were examined to predict TSS from TU using NEW Water and USGS data. First, various regression methods were tested to find an empirical relation between TSS and TU from the measured data in Southern GB, and best model was selected based on a random sampling technique. Comparison of the observed and estimated TSS—based on empirical equation—shows the normalized root mean squared error (NRMSE), and Nash-Sutcliffe efficiency (NSE) coefficient are 0.26 and 0.93, respectively. In addition, cumulative distribution function was used to estimate TSS based on USGS records of TU. Lognormal distribution was selected among best-fit distributions for probability approach. Estimated data were evaluated against measured TSS, and NRMSE and NSE are 0.30, and 0.91, respectively. Based on error criteria, both models can successfully estimate TSS; however, the lognormal distribution shows a slightly better estimate. A study is underway to check the ability of both methods in estimation of TSS, using recently measured data.

Preventing shoreline erosion is critically important as coastal hazards intensify. Living shoreline stabilization protects vulnerable coastal assets while promoting biodiversity. However, increased understanding of impacts to physical processes could improve restoration designs. This before-after-control-impact study quantifies hydrodynamic change resulting from restoring a degraded estuarine shoreline using living shoreline techniques. The eroding shoreline, located in Florida’s Mosquito Lagoon, was restored using oyster shell bags and emergent vegetation. Hydrodynamic data were collected from three shoreline sites (the Restored shoreline, a Control degraded shoreline, and a control Reference shoreline stabilized by mature mangroves) before and after living shoreline implementation. We find immediate and pronounced hydrodynamic effects attributed to restoration. Incoming velocities were reduced in the Restored site soon after restoration and significant wave heights were attenuated by the living breakwaters. Notably, onshore-offshore gradients of mean velocities and turbulence statistics within the restored site differed from the reference shorelines.

Sedimentation plays a key role in determining the morphology of a river. The river changes its bed because of hydrodynamics resulting from variabilities of the flow. Sedimentation in Mississippi River and its tributaries have been a topic of major concern. Sedimentation phenomenon is followed by erosion and deposition in the river and changes the river geometry continuously with the changing discharge and sediment properties. The erosion and deposition makes the design and operation of a physical infrastructure such as dams, locks, and levees a challenging task. The study of sediment movement in the Mississippi River can help in better operation of infrastructure and planning for appropriate mitigating measures. The current model utilizes the National Center for Computation Hydroscience and Engineering 2D tool, also known as CCHE2D, for modelling the transportation of sediments at the upstream location of United States Geological Survey streamflow station 07022000 Mississippi River at Thebes, Illinois. Although the 2D model requires more computation time but is effective in incorporating spatial river features as they impact the hydrodynamics and sedimentation. The model results evaluated the river morphology variations for maximum long term mean monthly flow due to sediment transport. The results also include the hydraulic flow variables estimations by numerical modelling. The results obtained from the study can help hydraulic engineers for sediment and flow management.

Scour at hydraulic structures such as bridge piers and abutments, stilling basins, and river bends can cause loss of life, structural collapse, and environmental damage to the ecosystem. Although many passive (requiring no greenhouse emitting use of electricity) scour countermeasures exist, such as riprap, submerged vanes, collars, spur dikes, sills, and sacrificial piles, these are subject to local scour and destruction themselves. An alternative flow-altering scour countermeasure consisting of air injection into the scour-inducing flow pattern can have benefits. Air injection cannot cause local scour due to the flow-adjusting nature of the bubbles since they are not rigid. Although injecting air does require an air compressor, it can be powered by solar panels and set up to only be activated during high-flows. Air injection is not needed in low-flows since the water velocity is slow and below the scour incipient motion conditions. Air injection has been proven successful in laboratory experiments for stilling basins and cylindrical bridge piers. Scour is reduced by the uplifting of the flow by the rising bubbles and the lowering of turbulence by the bubbles.

Existing empirical methods to predict reservoir sediment distribution are largely based on the original reservoir shape. To describe reservoir sediment distribution, one usually compares the original reservoir capacity and surface area as functions of the water depth to a future reservoir capacity and surface area. Examination of survey data of 23 reservoirs of Bureau of Reclamation found that area reduction method (ARM) produces the best results with existing Bureau of Reclamation data. Further examinations show that the reservoir type determined only from the reservoir shape usually does not always produce the best sediment distribution fitting compared to reservoir survey data. Reservoir sedimentation pattern depends not only on the reservoir shape, but also on reservoir operation rules, water surface levels, and sediment size distribution. Several reservoir operation parameters are presented in an attempt to relate sediment distribution with reservoir operations. Initial results show that sediment distribution is more complex than what simple parameters can present. Reservoir shape, defined as slope of the depth-capacity curve on log-log coordinates, presents the best correlation with the sediment distribution. Poor correlations between proposed reservoir operation parameters and sediment distribution were found. Sediment size distribution would likely improve the results, but this data is not available for the majority of the reservoirs analyzed in this study.

The construction of hardscape features in hydraulic physical models requires a high degree of precision to ensure proper results and feature integrity. Small shifts in cross sectional area, degradation of the feature material, or changes in material shape during testing are challenges faced when constructing unique prototype design features at model scale. Furthermore, the physical modeling design phase may be governed entirely by the availability of standard sized construction materials. The use of 3D printing offers a cost and time efficient solution to construction of hardscape features that must be precise, non-standard size, or sturdy enough to conduct testing. This presentation is the third of three presentations documenting advancements made by partnership between the University of Missouri Kansas City (UMKC), and Water Resources Solutions, LLC to efficiently validate two and three-dimensional numerical models using modest sized (mesoscale) physical models. These physical models, are smaller than the models typically used in the major labs to study large river projects, and larger than the micro-models which provide limited complex hydrodynamic specificity. The authors will demonstrate the validity and advantages of a physical/numerical modeling strategy using physical models followed by multi-dimensional numeric modeling. This paper discusses advances and advantages of 3D printing and associated rapid prototyping methods to construct structural hardscape and design options for the physical models. The first paper explores the technical aspects of the approach. The second paper in this series explores the project economics of the Physical First modeling approach and clearly demonstrates a significant reduction on overall project costs.

Four frequently used turbulence models, namely k-ε, realizable k-ε, k-ω, and SST k-ω have been used for simulating violent geysers in vertical pipes. The simulations have been performed using OpenFOAM and the assessment of the adopted models has been carried out by comparing the results with a set of experimental data from the literature. The quantities compared are pressure traces at certain locations, geyser height, and snapshots in the horizontal and vertical pipes at various stages during the geysering. The comparison of turbulence models suggests that all of these models predict almost the same geyser height with a difference of less than 2%. Pressure fluctuations predicted by realizable k-ε and SST k-ω models, however, are in better agreement with the data. Additionally, comparison of experimental and numerical snapshots reveals that the flow patterns obtained using the realizable k-ε model matches those of the experiments more closely. Regarding the computation time, k-ε model outperformed other models, as CPU times of k-ω, SST k-ω, and realizable k-ε models were 1.5, 1.2, and 1.1 times that of k-ε model, respectively.

Gas ebullition (formation of biogenic bubbles in sediment) is a complex interplay of biological, physical, and chemical processes. Research in our lab and by others has demonstrated the importance of gas ebullition in mobilizing sediment pollutants to the water column. In this study we developed a model using fundamental biological, chemical, and geomechanical principles to better understand the simultaneous impact of non-aqueous phase liquids (NAPLs) and methanogenesis on gas ebullition in sediments. Methanogenic gas production was predicted using our previously published dynamic model with sediment organic content, temperature, and sediment mechanical properties as key forcing variables. The model was validated using gelatin as a model elastic material with properties similar to those of cohesive sediment and mineral oil as a model NAPL. Sediment fracture results with gelatin were similar to those in model predictions, and the presence of NAPL was shown to decrease the sediment strength. These simulation and experimental results demonstrate the importance of temperature and NAPL presence in predicting gas ebullition rate.

There are various equations for predicting general scour that are presented in technical papers, design guidelines, and design manuals. Three such equations are the regime equations which were originally published by Gerald Lacey in 1930, Thomas Blench in 1969, and C.R. Neill in 1973. These regime equations are included in a United States Bureau of Reclamation (USBR) technical guideline titled, “Computing Degradation and Local Scour”, 1984; National Resources Conservation Service (NRCS), National Engineering Handbook (NEH) Part 654-Technical Supplement 14B, Scour Calculations, 2007; and in various local agency design manuals. However, the regime equations as presented in these publications are different in form than the equations in the original reference documents and as a result provide different calculated scour depths. The paper examines the differences between the equations as reported in the design guidelines/manuals and the equations as originally documented, summarizes limitations of the equations, and provides examples to compare the variation in scour depth when applying different forms of the equations.

A stream called English Coulee with a catchment area of 347 km² passes through the city of Grand Forks in North Dakota before emptying into the Red River of the North. The coulee has a low head dam located in the upper basin area. Midway through the watershed, the coulee is routed into a diversion channel system with two entry points that bypass excess flood discharge around the city. The stretch of English Coulee that passes through the city has stagnation of flow, overgrowth of cattails, poor odor, and deposition of debris and generally poor water quality. A design project to restore the coulee was presented as a challenge to a group of students. The project adapted a hydraulic modeling approach which enabled gauging the effectiveness of multiple design options. The coulee basin’s hydrologic runoff and continuous inflow simulation were developed using HEC-HMS. HEC-RAS was used to model the hydraulic influence of the diversion channel, diversion intakes, weirs, bridges, culverts, and urban reaches. Pertinent survey data for existing control structures was collected. Simulations were carried out to evaluate the effectiveness of four design options consisting of structural controls and channel modifications. Modeling runs showed control modification as the more cost effective solution.

Plunging flow dropshafts in urban drainage systems have been reported to cause downstream air pressurization and subsequent sewer odor issues. Retrofitting dropshafts with air circulation pipes has been attempted to reduce sewer odour complaints in Edmonton, Alberta, Canada. In the current work, laboratory model study, field monitoring, and model prediction were conducted to assess the effectiveness of this retrofit and explore its design optimization. This retrofitting was found to be effective in reducing air entrainment of the dropshaft and depressurizing the downstream sewers. However, the study showed that horizontal air circulation pipes at various elevations contributed differently in reducing the air entrainment. Based on the air flow model, increasing the size of the horizontal pipes had a more significant effect on reducing the downstream air pressure compared to increasing the number of the pipes. It was found that the retrofitting can be optimized by removing the bottom horizontal air circulation pipes while increasing the pipe size.

Bridge scouring is a complex phenomenon, responsible for bridge damage and failure in many cases. Numerous studies have given empirical models for the bridge scour components, whereas, some investigators have studied the interactive bridge scour relationship. However; the location of the bridge scour has not been given due attention which affects the bridge stability and design and also governs the design of downstream structures. A detailed investigation has been conducted to study the location of point of deepest scour in the floodplain in a compound channel, which includes free, submerged orifice flow, and overtopping flow conditions. Experiments were conducted for five different abutment ratios in clear water scour condition for erodible spill-through abutments. Location of the point of deepest scour, both in flow and transverse direction were observed and model has been suggested to predict the location of point of deepest scour. This will help improve the bridge safety and economy along with the effect of scour on river training works in the vicinity of bridge.

This paper presents the numerical simulation of flow past a bottom rack using CFD software, Flow-3D. The governing equations for fluid flow are the Reynolds-averaged Navier–Stokes equations (RANS) along with the volume of fluid (VOF) equation for surface tracking. In addition, the renormalized group (RNG) turbulence model is also used. The numerical model is validated by comparing with experimental results for pressure and velocity profile available in the literature. Results for free surface profile and velocity profile are also presented. Five different fluids are considered to study the effect of rheology on discharge characteristics of the bottom rack. As indicated in the simulations, viscosity is the dominating factor influencing the discharge through the bottom rack. Density also has an impact on the discharge characteristics. Based on the cases studied, recommendations for the size of the bottom rack, considering rheological characteristics of the fluid, are made.

In the design of stream restorations, boundary shear stress (shear stress) during high flow events is one of the key parameters in the assessment of the risk of morphological failure associated with channel bed and bank erosion and sediment transport. The use of two-dimensional hydrodynamic models (2D models) is becoming more common to estimate shear stress for stream restorations. These models can provide detailed distribution of shear stress over channels and floodplain surfaces. Obtaining accurate and reliable estimates of stress requires an accurate digital terrain model, estimates of input flows, and surface roughness coefficients. Impediments to the use of these models include the cost associated with extensively detailed terrain surveys, distributed information about the roughness coefficients, and the determination of appropriate flow conditions that must be modeled to identify erosion susceptible components. In the present research, the two-dimensional hydrodynamic model TUFLOW is employed to assess the sensitivity of shear stress to flow magnitude, upstream extent of the model, and roughness coefficients. Topography and hydraulic data obtained from monitoring efforts of the restored channel and floodplain of Slabcamp Creek located in Rowan County, Kentucky, was used to develop and calibrate the 2D model. Flow magnitude, the location of the upstream inflow boundary condition, and the roughness coefficients were varied to determine the sensitivity of the shear stress at a critical location in the restoration. The shear stress analysis was conducted along cross sections located in areas identified as being potentially at risk for high shear stress. The shear stress distribution was found to be relatively insensitive to the location of the upstream boundary condition and most sensitive to the flow and roughness coefficients of the floodplain. An interesting observation was that the shear stress in the channel decreased with the increased roughness of the floodplain.

This paper compares the simulated storm surge levels near Miami, Florida, resulting from Hurricane Andrew in 1992 and Hurricane Irma in 2017, using two different computational models: ADCIRC (advanced circulation model) and SLOSH (sea, lake, and overland surges from hurricanes). When Hurricane Andrew made landfall in the Bahamas and Florida in August 1992, it was a category 5 hurricane on the Saffir/Simpson Hurricane Scale. It brought devastations along its path including the Bahamas, southern Florida, and south-central Louisiana. Hurricane Irma, while was categorized as a category 5 hurricane along its path, made landfall on the Key West coast as a category 4 hurricane. Hurricane Irma devastated areas along Leeward Islands, Puerto Rico, Dominican Republic, Cuba, and southern Florida in September 2017. This study investigates the storm surge levels from these two major hurricanes along the coastal area of Miami, the largest metropolis and growth center in southern Florida. The performance of two storm surge models are examined. The ADCIRC model is a two-dimensional, hydrodynamic circulation model developed by ADCIRC Development Group which included universities, consultants, and the United States Army Corps of Engineers. The SLOSH model is a two-dimensional operational storm surge model developed by National Oceanic and Atmospheric Administration. The ADCIRC and SLOSH models results for storm surges caused by Hurricanes Andrew and Irma, at Miami coastlines are presented and compared against observations. The comparisons of model results and observations showed that model results should be interpreted with an understanding of the model capabilities and limitations. Utilizing the model results without understanding the limitations might lead to erroneous conclusions.

The South Florida Water Management District (SFWMD) operates and maintains over 650 water control structures. Flow through these structures are estimated with rating equations as a function of the headwater, tailwater stages, and operating conditions. Typically, field flow measurements are used for calibration, validation, and verification of the site-specific rating formulas. However, these measurements can be expensive and time consuming, but more importantly inadequate to cover all flow ranges. In recent years, the SFWMD has successfully developed and used a CFD-based flow rating framework. The application of this framework to large number of structures can be complex, and requires background knowledge of CFD. The focus of this paper is to demonstrate the feasibility of automating the application of developed flow rating framework using Visual Basic (VBA) and IronPython scripting. The parallel, pressure-based, non-hydrostatic commercial code ANSYS FLUENT was employed as the Reynolds-Averaged Navier-Stokes (RANS) solver for demonstration. The developed automation process includes planning CFD data generation for boundary conditions, feeding the required data inputs to ANSYS Workbench from an Excel environment, executing the CFD simulations using IronPython script, passing back the simulation results to the Excel session, and conducting flow rating for multiple flow conditions at multiple structures using Excel VBA. This study demonstrates the feasibility of automating the entire structure rating process. The developed automation opens new frontiers in cost-effective flow ratings for large amount of water control structures.

This study aims to predict probabilistic shoreline evolution in response to sea level changes, land subsidence, and onshore sand transport by using Monte Carlo simulation approach. A shoreline evolution model, GenCade, is used to simulate longshore sediment transport and long-term shoreline changes induced by random offshore waves. To determine landward sediment transport and shoreline responses to sea level changes, a modified Bruun rule is implemented into the GenCade model. The other two terms for long-term shoreline changes are also implemented into the GenCade model. A probability density function is used to capture stochastic features of wave heights under both fair weather and extreme weather conditions. It produces time series of wave heights including small and extreme waves based on their probabilities (or frequencies of appearance). A long-term shoreline changes in a hypothetical beach are simulated by considering wave climates, sea level change scenarios, land subsidence, and onshore transport. Probabilistic shoreline changes are computed by using the GenCade and Monte Carlo simulations.

In this study, computational fluid dynamics (CFD) was used to complement field flow measurements to study discharge capacity of water control structures having complex geometries. The CFD approach described herein is based on solving Reynolds-Averaged Navier-Stokes equations using κ-ε turbulence model with the free-surface resolved by the volume of fluid method. Two groups of complex water control structures have been investigated as case studies with this hybrid flow rating approach. The first group are the Golden Gate structures which are hinged crest gated structures, also known as pivot weirs. In this paper, a new broad crest weir flow equation which includes a parameterized pivot angle and a modified Villemonte submergence coefficient was proposed. CFD data, in conjunction with field measurement, was used to calibrate the proposed flow rating equations. The methodology proved to be very accurate with an overall error between measured and estimated flows of less than 6% for all three structures. The same approach was applied to Kissimmee River ship locks. In this case, a Villemonte-equation based submergence correction was added to a standard Parshall Flume flow equation for discharge estimation. CFD simulation was then applied to further calibrate the coefficient of this rating formula. Despite limited number of field measurements, a reasonably accurate flow rating equation was obtained. The above cases showed that CFD can be a cost effective, and powerful complementary tool to field measurements to accurately rate water control structures of complex geometries.

The influence of air phase on the flow behavior in urban stormwater systems was experimentally assessed in a physical model, which reproduces a long section of a stormwater system implemented in 2008, in Porto Alegre (Brazil). A severe structural failure has occurred in this stormwater system in 2013 and has motivated the physical model set up. All geometric singularities of the actual stormwater system (e.g. sharp bends, drop sections) were represented in the physical model. The geometric scale factor adopted is 1:15, resulting in a model length of 30 m and variable height (16 cm–42 cm) and width (20 cm–50 cm). The experimental procedure consisted of establishing steady flows and then mimicking the unsteady stages of the filling process by quickly opening a ball valve to allow further inflow. Varying degrees of ventilation were enabled by using various orifices opened in the upstream section of the apparatus. During the experimental runs, the inflow discharge was measured by electromagnetic flowmeters, pressure data was acquired using ten piezo-resistive transducers spread in various point of the model. Air pressurization is noticed immediately in the whole system, and ventilation geometry is linked to air pressurization magnitude.

The USACE Hydrologic Engineering Center (HEC) reservoir simulation software, HEC-ResSim, is being used by reservoir modelers world-wide. HEC-ResSim can be used to simulate reservoir operations for multiple objectives including flood risk reduction, navigation, hydropower, and environmental support. ResSim uses a map-based schematic to represent the river and reservoir system. Reservoir operation is modeled through a set of operation rules that describe the operating goals and constraints. The Tioga-Hammond reservoir system, located in the Chemung River watershed at the headwaters of the Susquehanna River Basin, contains two parallel reservoirs, Tioga and Hammond, whose pools are joined by a connecting channel. The connecting channel contains a control structure with two gates and an overflow weir. Under normal operations, Hammond’s pool is kept 5 feet higher than Tioga’s and Hammond’s excess inflows are released into Tioga by utilizing the gates in the connecting channel. During flood operations, the pools may rise and the 5 foot elevation difference may not be maintained. Under extreme events, one or both pools may exceed the elevation of the connecting channel’s weir resulting in an uncontrolled exchange of flow between the reservoirs. HEC-ResSim was designed to represent only dendritic river systems so modeling the bi-directional flow of the Tioga-Hammond connecting channel posed a significant challenge. The solution involved the use of some special modeling techniques and the powerful scripting capability within ResSim. This presentation will describe the intricacies of this modeling effort within ResSim including how the network was constructed and how scripting was used to make it work.

Experiments of air-water flows were conducted in a long straight rectangular pipe with the longitudinal length L=10 m to investigate the air-water flow properties including bubble shapes and bubble distributions for a varied range of discharges in cross-sections of different locations. The present investigation in hydraulic engineering mostly focused on the air concentration. Though some researchers noticed the importance of the size and shape for every single bubble in bubbly flows, solely relying on the measurement of the conductivity probe could only get the chord length of the assuming equivalent sphere of the bubble. From the observation of experiments, the spherical hypothesis was not accurate enough and some important parameters like the aspect ratio of the bubble were missing. Therefore, the high-speed camera and image processing techniques were applied in this paper to get more precise information of the bubble under different discharges. Additionally, to get rid of the mutual interference and overlap of the bubbles, the syringe pump that is capable of accurately controlling gas flow was used to generate a succession of disperse bubbles. Improved understanding is gained through the investigation, which reveals the major parameters of the bubble and the development of the bubble along the flow direction.

Due to deterioration of sewer pipes, soils can be lost by water infiltration through pipe defects, which may result in ground collapses and/or the formation of urban sinkholes. In this paper, soil particles without cohesion are simulated using a coupled discrete element method, and the Darcy’s water model is incorporated to account for the soil/water interaction. The parameters in this numerical model are calibrated using previous physical experimental measurements. From the numerical results, soils above the pipe defect will be washed out by water flow in a narrow zone just above the pipe defect, and the width is proportional to the pipe defect size. Based on the analysis of the force chain between soil particles in the erosion process, it is found that soil particles close to the pipe defect are lost under the gravity and drag force, whereas the friction between soil particles has a negligible effect on the soil particle motion. From the variation of soil flow rate in the erosion process, the rate of soil loss is proportional to the water flow rate in the erosion process. An analytical model is developed incorporating Stokes law to account for the water flow, and the model is verified by comparing with physical and numerical results. The mechanism of soil loss due to defective sewer pipe is investigated in this numerical simulation, and the proposed analytical model provides an effective approach for estimating the soil flow rate in the erosion.

In this study, we modified and applied an open source CFD software package, the TELEMAC, to study the sediment transport and scouring processes around Gangjeong Weir on Nakdong River in South Korea. The simulations have been carried out with the real weir and river bathymetry. We particularly have studied the effect of grain size distributions on bed evolutions, and taken into account the flood return periods to evaluate the maximum potential erosion around the hydraulic structures caused by floods. An agreement between numerical result and observation shows the ability of the model to predict maximum potential scoured depths near hydraulic structures.

This paper outlines the engineering controls designed and installed to stabilize a river bank segment of approximately 350 m at Río Gurabo in Puerto Rico using rock vanes. The design strategy consisted on providing a sustainable, environmental friendly, and economically feasible structural control to stabilize the bank and prevent future erosion. Structural controls includes the installation of seven rock vanes to redirect the erosive forces from the river bank toward the center of the channel, and the construction of a rock bench which serves as an adequate structure to tie in the rock vanes, and provide additional protection to the bank toe. This document summarizes the planning, geomorphologic and engineering evaluations, permitting, design, and construction phase of the project. System performance of the flood event associated with Hurricane María on September 20, 2017, is discussed.

Understanding the behavior of air-water interactions in closed conduits is very important for urban water systems that are subject to rapid filling conditions, such as stormwater systems during intense rain events. Among different mechanisms for air pocket appearance in closed conduits, shear flow instability is one that have significant capability to capture large volumes of air. Upon capture, air can impact surging and, upon uncontrolled release, lead to issues such as manhole cover displacement and/or geysering. This work present results from ongoing experimental and numerical research on air pocket entrapment based on shear flow instabilities. A fully-filled horizontal water pipe is partially opened pipe at the downstream end and create a cavity flow. After some advance within the pipe, a second valve is maneuver at the upstream end, enabling pressurized flows from the upstream end. The pipe-filling bore that is created pushes air in high velocity over the air cavity. In some cases, air pocket entrapment follows and lead to interesting peaks of pressure that seem to be a characteristic of this type of these entrapments.

The Potomac Interceptor (PI) system comprises three sewers totaling approximately 60 linear miles. Following major rain events in April and May 2014, DC Water incorporated these interceptors into one comprehensive sewer model to inform decision making on a variety of rehabilitation and O&M improvement initiatives. A portion of this model was built into the Transient Analysis Program (TAP) to conduct surge modeling to gain a better understanding of possible air movement and transients within the PI system. The surge modeling indicated the likelihood that entrapped air contributed to the weakening of two structures that connect two sewers in the PI system. TAP modeling was also utilized to support the rehabilitation design of one of these two structures. Results indicated that the capacity exceedance that occurred during the April 2014 rain event was likely due to accumulated sediment, operational challenges, and trapped air pockets. Additionally, model results indicated that the recommended solutions to mitigate future exceedances along the PI system included preventative maintenance and preparing the sewers to operate under pressurized conditions.

With significant advancement in computational speed, storage, and graphical visualization, there has been a renewed belief that the need to physically observe hydrodynamic processes in the field and laboratory models is obsolete. This hubris is no more valid today as it was when numerical modelers made advances in the 1970s, 1980s, and 1990s. Analysis of complex flow problems must regard these two approaches as complimentary, not exclusionary. Any river project, regardless of size, can have significant hydraulic issues that can complicate the design. Often, since only a simple 1-D hydraulic analysis is conducted, the resulting design is un-necessarily conservative and expensive. Worse yet, critical hydraulic issues may be completely missed. This presentation is the first of three presentations documenting advancements made by partnership between the University of Missouri Kansas City, and Water Resources Solutions, LLC to efficiently validate two and three-dimensional numerical models using modest sized (mesoscale) physical models. These physical models, are smaller than the models typically used in the major labs to study large river projects, and larger than the micro-models which provide limited complex hydrodynamic specificity. The authors will demonstrate the validity and advantages of a physical/numerical modeling strategy using physical models followed by multi-dimensional numeric modeling. This paper explores the technical aspects of this approach. The second paper in this series explores the project economics of the physical first modeling approach and clearly demonstrates a significant reduction on overall project costs. The final paper discusses advances and advantages of 3-D printing and associated rapid prototyping methods to construct structural hardscape and design options for the physical models.

With significant advancement in computational speed, storage, and graphical visualization, there has been a renewed belief that the need to physically observe hydrodynamic processes in the field and laboratory models is obsolete. This attitude is no more valid today as it was when numerical modelers made advances in the 1970s, 1980s, and 1990s. Analysis of complex flow problems must regard these two approaches as complimentary, not exclusionary. Any river project, regardless of size, can have significant hydraulic issues that can complicate the design. Often, since only a simple 1-D hydraulic analysis is conducted, the resulting design is un-necessarily conservative and expensive. Worse yet, as we can demonstrate, critical hydraulic issues may be completely missed. This presentation is the second of three presentations documenting advancements made by partnership between the University of Missouri Kansas City, and Water Resources Solutions, LLC to efficiently validate two and three-dimensional numerical models using modest sized (mesoscale) physical models. These physical models, are smaller than the models typically used in the major labs to study large river projects, and larger than the micro-models which provide limited complex hydrodynamic specificity. In the first presentation the authors describe an efficient approach to conducting hydraulic physical models which are used to validate two-and three-dimensional numeric models used to complete the final design of structures and countermeasures. This paper explores the project economics of the physical first modeling approach and clearly demonstrates a significant reduction on overall project costs. The final paper discusses advances and advantages of 3-D printing and associated rapid prototyping methods to construct structural hardscape and design options for the physical models.

Forested riparian zones moderate stream temperatures, modify sediment transport, and improve stream health. Clearing of riparian areas due to agriculture or urban development can lead to major changes in stream ecosystem structure and function including higher water temperatures. A study was conducted to compare changes in temperature, dissolved oxygen (DO) concentration, pH, total dissolved solids (TDS) concentration, and turbidity in the streamflow along a 540-m stream with transitional changes in the characteristics of the riparian zones. The first 180-m of the reach have a 30-m to 65-m width forested riparian zone, followed by the remaining 360-m of a reach under incipient degradation in which grass grows and is mowed along both sides of the stream. Weekly monitoring has been performed in 8 stations along the study reach including grab water sampling, water quality, and flow velocity. Compared with the forested segment, the open reach has higher temperature (+2 ^oC), pH (+0.1), and total dissolved solids concentration (+0.03 g/l). The DO concentration remains almost constant along the forested reach, but drastically decrease along the clear segment (-2 mg/l). Forested riparian zones can reduce stream temperatures and maintain favorable concentrations of dissolved oxygen with a biological significance for the living organisms in the stream. Consequently, maintenance of forested riparian zones might provide benefits in mitigating some of the ecological effects of climate change on water temperature.

A conjunctive use strategy is described for the 104 year-old Patillas Reservoir on the south coast of Puerto Rico, which has lost 57% of its storage capacity to a combination of sedimentation (39%) plus forced drawdown due to dam safety concerns (18%). A new reservoir operating rule is described which offsets storage loss by allowing the diversion, on average, of 11% of reservoir inflow for storage by aquifer recharge in Salinas Municipality. Salinas is 100% dependent on groundwater for municipal water supply, and experienced water rationing (intermittent service) during the 2015 drought in the face of increasing dissolved solids, attributed to saline intrusion. Of 7.8 mm³/yr diverted to recharge, only 0.6 mm³/yr comes from reduced reservoir firm yield while the remaining 7.2 mm³/yr represents reduced spills to the Caribbean Sea.

Our nation’s reservoirs are filling with sediment that deposits in and upstream from dams. Dam owners, whether be governmental, commercial, or private, will be faced with the issue of managing reservoir sedimentation issues in the not-too-distant future. Six brief case studies of how sediment is managed illustrate the diverse and complex nature of the problem. There is an acute need to update bathymetric surveys of our reservoirs to assess the magnitude of sedimentation at each project and to carry out pilot projects to extend the experiential database for the benefit of all managers and operators.

To date, as information and communication technology enters into water distribution field, smart water system gains potential of worldwide application. Smart water system enables to collect, organize, and analyze real-time data and produce new data, which provides engineers and consumers with real-time water information. In this project, a micro-smart water test bed was established to realize the real-time data acquisition by using flow sensor and Arduino UNO, to achieve data organization by employing observation data model (ODM) database and to accomplish data analysis by using R-studio. Also, the performance of micro-smart water test bed was evaluated by comparing relative errors between the experimental measurements and modeling outputs. System performance analysis shows reasonable flow sensor has reasonable sensitivity to system operation and testing data evaluation illustrates experimental measurements fit modeling outputs reliably. This real-time smart water system project performed as expected. Finally, the future recommendation for this project was made in the last section.

Piezometer tube is an old but still popular device to measure pressure in open-channels, pipes, and groundwater. The device is fairly cheap besides the labor cost of reading the liquid level in it. In order to reduce the labor cost, a new solution with computer vision technology is put forward. In our experiment, a camera is used to monitor several tubes at the same time and the liquid levels are read by an image processing algorithm afterwards. The image processing algorithm is based on pattern recognition. It is robust enough that no special lighting system is needed. The error of the water level is corrected with the relative locations of the pressure tubes and the camera. The whole system is tested with 60 piezometer tubes. The results show that the liquid levels read by the camera are as accurate as those read by naked eyes. Moreover, the performance of the system in weak light is tested and discussed. As a result, the pressure measurement solution is low-cost, portable, accurate, and easy to expend.

A methodology for computation of system resilience for the real-time operation of water distribution systems (WDS) under critical conditions of limited water and limited electrical energy is presented. Critical emergencies in WDS operations can result from extreme drought, electric grid failure, and other severe natural conditions. The basic objective of this study is to apply the concepts of system resilience within an optimization-simulation model for real-time optimal operation of water distribution systems under conditions of limited water and energy availability. The modeling approach interfaces a genetic algorithm optimization procedure with the WDS hydraulic simulator (EPANET) in the framework of an optimal control problem. The methodology is implemented within the framework of MATLAB. An example WDS is used to demonstrate the application of the resilience concepts to assess the performance.

Resilience for water distribution systems is defined as the capability of restoring all operational functions and delivering safe drinking water in a fast manner after major disruptions. The challenge with resilience planning is quantifying it for water networks. Although there are resilience evaluation methods focusing on different dimensions of resilience, there are no metrics to evaluate systems’ operational resilience and recovery in relation to structural properties. The aim of this research is to develop a method that evaluates the rapidity dimension of operational resilience for water networks. The robust operation ranges are incorporated as the measure of performance parameters. The overall systems’ resilience is evaluated simulating disruptions (i.e., excess demands) on critical locations and observing the recovery of water networks. This approach can be used in a decision-making process to assist authorities and decision makers in developing planning and management strategies, and enhancing the resilience of water infrastructure systems.

This study investigates the effects of considering an advanced water quality simulation model for finding the optimal layout and operation of a booster chlorination system. Previous studies have generally relied on EPANET, an advection-based water quality model, to simulate the transport and decay of disinfectant residuals. In this study, an advection-dispersion-reaction (ADR) transport model that considers the spatial distribution of flow demands along the dead-end pipe is considered instead. The results suggest that optimization results generated by EPANET may not produce a booster chlorination system that adequately maintains the minimum residual concentration in the periphery pipes of a real-life water distribution system.

Water distribution systems are particularly vulnerable infrastructure systems as they comprise numerous exposed elements which can be deliberately infiltrated or may malfunction. Faults in pump operation, errors in sensors, power outages, cyber-attacks, contamination intrusions, and other abnormal complications harm water distribution systems operation in various ways. Therefore, implementing reliable events detection mechanisms in drinking water systems is vital for ensuring societal welfare. As such, the ability to rapidly detect such occurrences (and their origin) is a foremost objective. In this study a model for detecting abnormal events in water distribution systems using a minimum volume ellipsoid (MVE), adapted through time, is developed and demonstrated. A preliminary step to the ellipsoid finding is the clustering of different observation groups. It may be inaccurate to apply all observations in the same ellipsoid, as there are routine changes in water parameters. Applying some seasonal adjustments for example, is essential. Pattern recognition can help characterize typical behaviors of a time series. For example, water temperature rises in 2–3 degrees around 10 am, or increases in chloramine concentration don't last more than 10 minutes. Such characteristics may help define more appropriate ellipsoids for each time step thus generate a more sensitive model. The proposed method repeatedly constructs an MVE and modifies its structure according to new measurements and recorded events. The method is demonstrated on an example application through base runs and sensitivity analyses using hourly consumptions, flows, and pressures.

This paper proposes a multi-objective model for the design of a water distribution system (WDS) with numerous elements including pipes, valves, pumps, and storage tanks. The main focus of our decomposition of a WDS to district metered areas is on decreasing water leakage in the system without corrupting the customer service. In other words, the customer demand is to be satisfied and meanwhile the operating pressure and water age will be minimized. The advantages of the proposed model are illustrated through case studies.

Identifying anomalous system behavior is the key to detect any cyber-physical attacks on any water distribution systems. Analyzing system data and understanding the system and the relationships between different components of the system can provide a basis to develop a hybrid method to identify these anomalous behaviors. Lack of in-depth knowledge in system hydraulics and control may fail to identify the anomalous behavior with accuracy. Hence, the implementation of system hydraulics and control in attack detection algorithm is important. This paper presents an algorithm that not only considers the concept of data mining but also the hydraulics and system relationships. Three specific modules namely statistics and pattern recognition module, principal component analysis module, and hydraulics and system relationships module are included in the algorithms to detect any cyber-physical attack on a literature network known as C-Town. The evaluation criteria have been calculated for three of these modules separately. Results show that the hydraulics and system relationship module can perform better than any other modules considered in the study. The performance score for the attack detection and the score for the confusion matrix for this module are respectively 1.0 and 0.86 resulting a combined score of 0.93.

Information related to water distribution systems (WDS) in small and intermediate cities often may be incomplete or unreliable, and data acquisition may be infeasible due to elevated time and economic costs. This lack of information can lead to inaccurate network operation and maintenance, and it can also limit the use of calibrated hydraulic models to support future designs in the system. Virtual case studies can thus be treated and used to overcome the insufficient data problems. The results are not case-specific as the obtained with benchmark networks. This paper aims to evaluate a new methodology by generating virtual networks for different cities in Colombia and then comparing resulting virtual WDS with the real systems, based on data obtained from water utilities. For both sets, network layout similarity and hydraulic factors were analysed. The results of the comparison between real and virtual networks make it possible to conclude about the applicability and limitation of this virtual generation software in small cities or DMAs in large cities, and also allows suggesting alternative solutions.

Population growth is one of the most challenging factors around the design, operation, and renewal of water distribution systems (WDS). Nowadays, cities are subjected to many dramatic changes that may disturb the performance and original approach of urban infrastructure; among these are the informal settlements (slums) and overcrowding phenomena. This paper describes a rehabilitation methodology of real WDS that seeks to identify the proper topology changes, in terms of pipes replacement, under different demand scenarios in order to readjust the network, achieve minimum pressure requirement, and improve the hydraulic performance of the whole system. This research focuses in the assessment of two rehabilitation criteria, tested in two WDSs of different Colombian towns and three district metering area (DMA) of main Colombian cities, with an infrastructure installed at least twenty years ago. Some surrogate measures of networks reliability were taken into account to assess the rehabilitation methodology, and its results were compared with optimal designs. The EPANET toolkit, MATLAB, and REDES software were used in this research. Both criteria proved to be. Both criteria proved to be good benchmarks to improve some performance indicators in each rehabilitated network. Finally, the suitability of each criteria was analysed considering the topologic complexity as well as network capital cost.

In a water distribution system—WDS, every pipe is considered as a potential candidate to locate a valve, regarding that the usage of these accessories are widely applied with different purposes within water supply systems. Framed in the scope of this research, it has been proposed as a previous stage of the optimization process, the implementation of different techniques that leads to the reduction of the solution space based on the hydraulic knowledge of the network, seeking to obtain results of high quality in an efficient way. Because of this, it was decided to analyze the effect of the reduction of the solution space associated with the optimal valve location problem in a WDS, using three hydraulic criteria based on the operational knowledge of the network, obtaining satisfactory results without losing the quality of the reached solution.

In the recent years, many metropolitan cities are facing economic problems with the rehabilitation of their water distribution system. With time, the failure rates and leakage in the pipes will increase even with some of the rehabilitation efforts taking place. Thus, water utilities should consider seeking strategies for water pricing and the system rehabilitation to prevent their existing distribution network from deteriorating. The objective of this study was to understand how different management techniques implemented on a hypothetical water distribution networks would result in a sustainable management of water infrastructure. Using system dynamics approach, a simulation model was developed to study the impacts of management strategies on the distribution network functionality and financial health over a 100-year time horizon. The model considers the interactions among distribution network sector (existing water network differentiated by age), finance sector (user fee, profit/loss account, debt, and available cash), and lastly, consumer demand sector (change in water demand due to population). The study exhibited a comprehensive analysis for a utility on achieving financial sustainability with a comparatively lower user fees with adaptation of rehabilitation measures. Results of the current study can be used by water utilities to analyze different approaches for sustainable financial and infrastructure management.

Water supply is an energy intensive activity, consuming between 2 and 3% of the world's energy. The present work analyzed the feasibility of the use of tanks to minimize electricity costs in the operation of water supply systems, taking into account the hourly variation of the electric energy tariff. A case study was performed using the EPANET software to analyze part of the distribution network in Curitiba, Brazil, with 1476 junctions, 1678 pipes, one reservoir, and two pumps, which in the original configuration operates with 1318 annual low pressure events. The decrease in energy costs is due to the partial operation of pumps during the peak hours, when the energy cost is higher. The installation of tanks also provides better control of pressures, better supply of demand, and reduction of risks of pipe ruptures. It was concluded that there are solutions that lead to the reduction of expenses of the order of 20% and whose return periods are feasible (5 years).