This document discusses modeling vegetation and roughness in rivers using computational fluid dynamics (CFD). It presents two main approaches: 1) Using drag formulas to model individual vegetation stems or stones, accounting for their flow obstruction. 2) Modeling the vegetation/roughness area as a porous medium, reducing water velocity. The document provides details on implementing these approaches in the SSIIM CFD model and gives two examples of applications to model vegetation in rivers.
Mathematical model ofrcc dam breakbastorarcc dam as a case studyIAEME Publication
This document presents a mathematical model for predicting the failure of roller compacted concrete (RCC) dams due to overstress. As a case study, the model is applied to simulate the hypothetical failure of the Bastora RCC dam in Iraq. The model assumes the failure would occur gradually over a short time period, with the breach geometry increasing in size. Reservoir outflows are calculated using the model and routed downstream to determine flood levels and discharges. The goal is to better understand RCC dam failures and estimate their downstream impacts, which has not been extensively studied previously.
Boundary conditions in groundwater modelingBijit Banik
This document discusses the influence of different boundary conditions on groundwater models using MODFLOW. It describes three main types of boundaries: 1) Dirichlet (prescribed hydraulic head), 2) Neumann (prescribed flux), and 3) Cauchy (semi-permeable or head-dependent flux). Examples are given of each boundary type and how they are commonly applied. The steps to build a simple MODFLOW model in GMS are then outlined, including setting the grid, properties, boundaries, wells, and running the simulation.
This document summarizes the results of a CFD (computational fluid dynamics) analysis of airflow in a data center before and after implementing FlowLogix airflow management solutions. The current environment shows issues like hot air recirculation causing hot spots, unbalanced static pressure leading to inefficient airflow, and lack of temperature monitoring. After FlowLogix installation, the analysis shows improved airflow balance and control directing more cold air to equipment, lower average and maximum inlet temperatures, and elimination of hot spots. Continuous environmental monitoring and annual CFD maintenance are included to optimize efficiencies over time.
The document summarizes a study on hydraulic modeling of a side-channel spillway at Iven Dam in Mongolia. The goals were to determine flow regimes using physical and numerical modeling, study hydraulic modeling methodology, and identify ways to improve standard design methods. Hydraulic modeling methods used included analytical fluid dynamics, experimental fluid dynamics with a physical model, and computational fluid dynamics. Results from physical and CFD models showed significant differences from the standard design method, indicating the need to update standard methods. The study concluded the spillway capacity should be increased and validated all hydraulic structures with physical and numerical models before construction.
One-Click CFD is a CFD framework based on openFoam that allows users to run aerodynamic simulations on car designs with little or no CFD experience. The document outlines the steps to use One-Click CFD, including preparing CAD geometry as STL files, starting a simulation locally or on AWS cloud, and post-processing results using the HTML report or ParaView software. The goal is to make CFD accessible to those interested in racing car aerodynamics.
The Indus Water Treaty was signed in 1960 between India and Pakistan to resolve issues around the distribution of water from the Indus River and its tributaries following the partition of British India and independence of Pakistan in 1947. The treaty divided the rivers between the two countries, with Pakistan getting exclusive rights to the three western rivers of Indus, Jhelum, and Chenab, while India received control over the three eastern rivers of Ravi, Beas, and Sutlej. It also guaranteed Pakistan uninterrupted water supply for 10 years to build dams and irrigation infrastructure financed by World Bank loans and compensation from India. Major projects completed under the treaty included the Warsak, Mangla, and Tarbela dams and several
The document summarizes the key aspects of the Indus Water Treaty between India and Pakistan. It discusses the distribution of the eastern and western rivers between the two countries as outlined in the treaty. It also describes some of the major disputes that have emerged post-treaty, such as Indian dam projects on western rivers. The summary concludes that while the treaty has generally functioned well, both countries should focus on cooperatively developing water resources for the future through modifying the treaty's governing mechanisms.
Mathematical model ofrcc dam breakbastorarcc dam as a case studyIAEME Publication
This document presents a mathematical model for predicting the failure of roller compacted concrete (RCC) dams due to overstress. As a case study, the model is applied to simulate the hypothetical failure of the Bastora RCC dam in Iraq. The model assumes the failure would occur gradually over a short time period, with the breach geometry increasing in size. Reservoir outflows are calculated using the model and routed downstream to determine flood levels and discharges. The goal is to better understand RCC dam failures and estimate their downstream impacts, which has not been extensively studied previously.
Boundary conditions in groundwater modelingBijit Banik
This document discusses the influence of different boundary conditions on groundwater models using MODFLOW. It describes three main types of boundaries: 1) Dirichlet (prescribed hydraulic head), 2) Neumann (prescribed flux), and 3) Cauchy (semi-permeable or head-dependent flux). Examples are given of each boundary type and how they are commonly applied. The steps to build a simple MODFLOW model in GMS are then outlined, including setting the grid, properties, boundaries, wells, and running the simulation.
This document summarizes the results of a CFD (computational fluid dynamics) analysis of airflow in a data center before and after implementing FlowLogix airflow management solutions. The current environment shows issues like hot air recirculation causing hot spots, unbalanced static pressure leading to inefficient airflow, and lack of temperature monitoring. After FlowLogix installation, the analysis shows improved airflow balance and control directing more cold air to equipment, lower average and maximum inlet temperatures, and elimination of hot spots. Continuous environmental monitoring and annual CFD maintenance are included to optimize efficiencies over time.
The document summarizes a study on hydraulic modeling of a side-channel spillway at Iven Dam in Mongolia. The goals were to determine flow regimes using physical and numerical modeling, study hydraulic modeling methodology, and identify ways to improve standard design methods. Hydraulic modeling methods used included analytical fluid dynamics, experimental fluid dynamics with a physical model, and computational fluid dynamics. Results from physical and CFD models showed significant differences from the standard design method, indicating the need to update standard methods. The study concluded the spillway capacity should be increased and validated all hydraulic structures with physical and numerical models before construction.
One-Click CFD is a CFD framework based on openFoam that allows users to run aerodynamic simulations on car designs with little or no CFD experience. The document outlines the steps to use One-Click CFD, including preparing CAD geometry as STL files, starting a simulation locally or on AWS cloud, and post-processing results using the HTML report or ParaView software. The goal is to make CFD accessible to those interested in racing car aerodynamics.
The Indus Water Treaty was signed in 1960 between India and Pakistan to resolve issues around the distribution of water from the Indus River and its tributaries following the partition of British India and independence of Pakistan in 1947. The treaty divided the rivers between the two countries, with Pakistan getting exclusive rights to the three western rivers of Indus, Jhelum, and Chenab, while India received control over the three eastern rivers of Ravi, Beas, and Sutlej. It also guaranteed Pakistan uninterrupted water supply for 10 years to build dams and irrigation infrastructure financed by World Bank loans and compensation from India. Major projects completed under the treaty included the Warsak, Mangla, and Tarbela dams and several
The document summarizes the key aspects of the Indus Water Treaty between India and Pakistan. It discusses the distribution of the eastern and western rivers between the two countries as outlined in the treaty. It also describes some of the major disputes that have emerged post-treaty, such as Indian dam projects on western rivers. The summary concludes that while the treaty has generally functioned well, both countries should focus on cooperatively developing water resources for the future through modifying the treaty's governing mechanisms.
This document summarizes water supply projects being undertaken by the Guadalupe-Blanco River Authority (GBRA) in Central Texas. It discusses the Mid-Basin Water Supply Project (MBWSP) which is part of the regional water planning to supply surface water from the Guadalupe River to areas relying on depleting aquifers. The MBWSP would include an intake on the river, treatment plant, and pipelines to deliver 50,000 acre-feet of water per year. Alternative options discussed include an integrated project sharing facilities with nearby water suppliers or connecting the treated water to the cities of San Antonio or Austin. Estimated costs per acre-foot of water range from $1,467 to $1,
The Indus River originates in Tibet and flows through India and Pakistan before emptying into the Arabian Sea in Pakistan. It has a total length of 3,180 km and drains an area of 1,165,000 square km. The Indus Water Treaty of 1960 allocated the eastern rivers (Sutlej, Beas, Ravi) to India and the western rivers (Jhelum, Chenab, Indus) to Pakistan. However, disputes have arisen over Indian projects on the western rivers, including the Tulbul Navigation Project, Baglihar Dam, and Kishanganga Dam projects. While the treaty helped resolve water disputes, critics say it does not adequately address issues like climate change,
This document summarizes a student project to design and analyze a concrete gravity dam. It includes sections on the project overview, dam types, site selection criteria and investigation, salient dam features, stability and stress calculations using conventional and ANSYS methods, static and hydrostatic analysis using STAAD Pro, and optimization of the dam section using genetic algorithms in MATLAB. The optimized design achieved a reduced cross-sectional area and base width while maintaining safety.
The document summarizes the water dispute between India and Pakistan over the Indus River basin. It provides background on how the dispute arose after partition and India's actions to cutoff water supply. It then discusses the role of the World Bank in brokering negotiations between the two countries, which led to the signing of the landmark Indus Waters Treaty in 1960. The treaty allocated control and usage of the six rivers of the Indus basin between India and Pakistan.
This document provides an overview of the analysis and design of a gravity dam located in seismic zone V. It discusses the project team members and then covers the basic structure and purpose of dams. It reviews the history of dam construction and provides examples of different dam types. The document outlines the necessary investigations and considerations for dam design, including stability, sedimentation, spillways, and energy dissipation structures.
This document provides an overview of spillways and flood control works for dams. It discusses the key components and design considerations for spillways, including approach channels, control structures, discharge carriers, terminal structures, and energy dissipaters. It describes different types of spillways like overflow, trough, siphon, and side channel spillways. Design aspects for spillway crest gates like radial and drum gates are covered. The document also discusses intake and outlet works for reservoirs, including their components and functions.
The document discusses several major rivers within the Indus River System located in India and Pakistan. It mentions the Indus River originates in Tibet near Lake Manasarowar and flows through India's Ladakh region before entering Pakistan. It also discusses tributaries like the Shyok River and Nubra River which feed into the Indus. Other key tributaries mentioned include the Jhelum, Chenab, Ravi, Beas, and Sutlej rivers.
Oldest branch of engineering, next to Military engineering. All engineering works other than for military purposes were grouped in to Civil Engineering. Mechanical, Electrical, Electronics & present day Information technology followed it.
A professional engineering discipline that deals with the analysis, design, construction and maintenance of infrastructural facilities such as buildings, bridges, dams, roads etc.
Civil Engineering is everywhere. Civil Engineering is a composite of many specific disciplines that include structural engineering, water engineering, waste material management and engineering, foundation engineering etc. among many.
Gravity dams are structures designed so that their own weight resists external forces. Concrete is the preferred material. Forces acting on the dam include water pressure, uplift pressure, earthquake forces, silt pressure, wave pressure, and ice pressure. The dam's weight counters these forces. Dams are checked when full and empty, accounting for load combinations. Gravity dams can fail due to overturning, crushing, tension cracks, or sliding along foundation planes. Design aims to prevent failure from these modes.
The document discusses the design and construction of concrete gravity dams. It begins with an introduction of dams and their purposes, then discusses site selection factors, design considerations, foundation investigations, construction procedures, and challenges in construction. The key points are that concrete gravity dams are designed so their own weight resists external forces, and their construction involves dewatering the river, building a cofferdam, removing loose materials, and placing concrete in lifts while controlling the temperature to prevent cracking.
Engineering Geology (Civil Engineering Applications)GAURAV. H .TANDON
This document discusses the important geological factors to consider when selecting sites for dams and reservoirs. Narrow river valleys, shallow bedrock, and competent bedrock foundations are desirable for reducing dam construction costs. Sedimentary rocks like sandstone and limestone can cause water leakage from reservoirs depending on their porosity. Metamorphic rocks like gneiss and quartzite are generally impermeable. Geological structures must also be considered, with horizontal or tilted strata being most suitable and faults or intense fracturing making a site undesirable. The document outlines these considerations in detail.
The Uri Hydel Power Project is a 480 MW hydroelectric power station located on the Jhelum River in Jammu and Kashmir, India. It was constructed between 1989-1997 by the National Hydroelectric Power Corporation. Key features include a 10.64 km head race tunnel, 4 units of 120 MW each located underground, and an annual generation of 2663 million units of power supplied to northern states of India. While it provides clean energy and economic benefits, the project also displaced over 400 families and required extensive afforestation efforts to mitigate environmental impacts. There is ongoing dispute between Jammu and Kashmir and NHPC over sharing of project revenues and control.
Changes in dam break hydrodynamic modelling practice - Suter et alStephen Flood
Abstract: Today, many organisations rely on hydrodynamic modelling to assess the consequences of dam break failure on downstream populations and infrastructure. The availability of finite volume shock-capturing schemes and flexible mesh schematisations in widely used software platforms imply that dam break modelling projects will be carried out differently in the future: Finite volume based platforms allow widespread application of shock-capturing methods and flexible mesh platforms can represent features in the study area more realistically and are more flexible thanks to varying mesh resolutions. Furthermore, the recent adoption of Graphics Processing Unit (GPU) technology in mainstream scientific and engineering computing will also significantly decrease computation times at relatively low cost.
This paper examines the application of finite volume, flexible mesh and GPU technologies to dam break modelling. One-dimensional (1D) modelling results are compared to those from two-dimensional (2D) finite difference and finite volume approaches. The results demonstrate that there are differences between modelling approaches and that the computational speeds of 2D simulations can be significantly reduced by the use of GPU processors.
This document summarizes a study that used computational fluid dynamics (CFD) to model soil erosion during a hole erosion test (HET). The study used the k-epsilon turbulence model in Fluent software to simulate the biphasic turbulent flow and resulting shear stresses at the water-soil interface. This allowed for a three-dimensional analysis of erosion rates along the hole, unlike typical one-dimensional models. Results showed the inlet side of the hole experienced more erosion than the outlet side due to non-uniform shear stresses. The model aims to better understand and predict piping failures in hydraulic structures by quantifying the effects of flow velocity on erosion rates.
This document summarizes a proposed new conceptual model called the Fracturing Impacted Volume (FIV) model as an alternative to the commonly used Discrete Fracture Network (DFN) model for modeling unconventional reservoirs. The FIV model accounts for the pressure-dependent permeability of the reservoir system and fluid loss during fracturing. It views fracturing as pressurizing the reservoir system within a spatial volume around the fracture, including microfractures and pores, rather than solely focusing on discrete fracture propagation. Laboratory tests on tight sandstone and shale cores show that reservoir permeability increases with pressure, supporting this conceptualization. The FIV model aims to provide a more effective way to understand fracturing mechanisms and improve simulation of fract
Design , Analysis and Optimization of Intze Type Water Tank With Sloshing EffectIRJET Journal
This document describes a study on the design, analysis and optimization of an Intze type water tank considering sloshing effects. The study involves designing a circular water tank with a capacity of 300,000 liters and height of 12 meters using design specifications from Indian codes. A 3D model of the tank is created in ANSYS and analyzed using structural and modal analysis to understand stresses and natural frequencies. The results show maximum von mises stresses of 1.61 N/mm2, which is within permissible limits. The natural frequencies help identify ranges requiring additional damping or strength. The study aims to optimize the tank design to minimize stresses and cracking considering seismic loads and sloshing effects.
The document is a final report summarizing the design of a sediment trap for a research flume. It describes the initial designs considered and the final designs selected, including drawings and calculations. The sediment trap will trap and continuously weigh sediment in the flume before removing and transporting it to a dumpster, separating the sediment from the water for recirculation.
Paper 43 - Deep Water Pipeline CT 9_2_15John Grover
This document discusses the use of coiled tubing as down-lines for pre-commissioning deepwater pipelines. It compares coiled tubing to other down-line options like flexible lines. Coiled tubing has advantages over flexible lines in terms of cost, delivery time, deck space requirements, reliability, and ability to provide contingency. The document outlines the latest custom coiled tubing equipment being deployed, which is designed for large diameters of 2 7/8" or 3 1/2" pipe and to operate in water depths up to 3,000m. It was also designed to be road transportable and for flexible installation on vessels through a moonpool or over the side.
Lyapichev. Problems in numerical analysis of CFRDs (ICOLD Bull.155)6 p.)Yury Lyapichev
The document discusses several challenges and developments in numerically analyzing concrete faced rockfill dams (CFRDs). It notes that until recently, CFRDs were designed based on experience rather than analysis. Accurate models have since shown issues like excessive compressibility of downstream rockfill adversely impacting the concrete face. The document also discusses modeling earthquakes, the need for structure-specific models in some cases, and ensuring nonlinear analysis convergence. Overall, it emphasizes the importance of numerical analysis as a tool to supplement—not replace—engineering judgment, especially for extrapolating lessons from incidents at high CFRDs.
This document describes a model developed to simulate the separation of CO2 from multicomponent natural gas mixtures using a spiral wound membrane (SWM) module. The model uses the succession stage method to discretize the separation process within the SWM. The model can evaluate product purity, hydrocarbon loss, stage cut, and permeate acid gas composition for multicomponent mixtures. Results show multicomponent systems provide higher product purity, lower hydrocarbon loss, and higher permeate acid gas content than binary systems. Different multicomponent mixtures yield varying separation performances depending on the acid gas component. The model has potential for designing and optimizing SWM systems for natural gas sweetening.
The document is a final report from a student group (Group B4) evaluating a low pressure chemical vapor deposition (LPCVD) reactor. It includes a study of how temperature, pressure, wafer spacing, inlet mole fraction affect silicon deposition uniformity in the reactor. The group found that temperature and pressure had an inverse relationship with uniformity, while wafer spacing and inlet mole fraction had a direct relationship with uniformity. Temperature was found to have the greatest effect on deposition rate.
Using Half Pipes as Permeable BreakwaterIRJET Journal
This document describes a study that investigated using half pipes as permeable breakwaters to protect coastlines in Egypt. Two types of half pipe breakwaters were tested experimentally and numerically: horizontal half pipes shaped like an H, and vertical half pipes shaped like a C. Physical models were used to identify the hydraulic performance of the barriers under different wave conditions. A numerical model was also developed using FLOW-3D software and validated against the laboratory data. The results showed that increasing the relative water depth decreases the amount of wave transmission through the barrier and increases wave reflection. Permeable breakwaters were found to effectively dissipate wave energy while avoiding issues caused by traditional solid breakwater structures.
This document summarizes water supply projects being undertaken by the Guadalupe-Blanco River Authority (GBRA) in Central Texas. It discusses the Mid-Basin Water Supply Project (MBWSP) which is part of the regional water planning to supply surface water from the Guadalupe River to areas relying on depleting aquifers. The MBWSP would include an intake on the river, treatment plant, and pipelines to deliver 50,000 acre-feet of water per year. Alternative options discussed include an integrated project sharing facilities with nearby water suppliers or connecting the treated water to the cities of San Antonio or Austin. Estimated costs per acre-foot of water range from $1,467 to $1,
The Indus River originates in Tibet and flows through India and Pakistan before emptying into the Arabian Sea in Pakistan. It has a total length of 3,180 km and drains an area of 1,165,000 square km. The Indus Water Treaty of 1960 allocated the eastern rivers (Sutlej, Beas, Ravi) to India and the western rivers (Jhelum, Chenab, Indus) to Pakistan. However, disputes have arisen over Indian projects on the western rivers, including the Tulbul Navigation Project, Baglihar Dam, and Kishanganga Dam projects. While the treaty helped resolve water disputes, critics say it does not adequately address issues like climate change,
This document summarizes a student project to design and analyze a concrete gravity dam. It includes sections on the project overview, dam types, site selection criteria and investigation, salient dam features, stability and stress calculations using conventional and ANSYS methods, static and hydrostatic analysis using STAAD Pro, and optimization of the dam section using genetic algorithms in MATLAB. The optimized design achieved a reduced cross-sectional area and base width while maintaining safety.
The document summarizes the water dispute between India and Pakistan over the Indus River basin. It provides background on how the dispute arose after partition and India's actions to cutoff water supply. It then discusses the role of the World Bank in brokering negotiations between the two countries, which led to the signing of the landmark Indus Waters Treaty in 1960. The treaty allocated control and usage of the six rivers of the Indus basin between India and Pakistan.
This document provides an overview of the analysis and design of a gravity dam located in seismic zone V. It discusses the project team members and then covers the basic structure and purpose of dams. It reviews the history of dam construction and provides examples of different dam types. The document outlines the necessary investigations and considerations for dam design, including stability, sedimentation, spillways, and energy dissipation structures.
This document provides an overview of spillways and flood control works for dams. It discusses the key components and design considerations for spillways, including approach channels, control structures, discharge carriers, terminal structures, and energy dissipaters. It describes different types of spillways like overflow, trough, siphon, and side channel spillways. Design aspects for spillway crest gates like radial and drum gates are covered. The document also discusses intake and outlet works for reservoirs, including their components and functions.
The document discusses several major rivers within the Indus River System located in India and Pakistan. It mentions the Indus River originates in Tibet near Lake Manasarowar and flows through India's Ladakh region before entering Pakistan. It also discusses tributaries like the Shyok River and Nubra River which feed into the Indus. Other key tributaries mentioned include the Jhelum, Chenab, Ravi, Beas, and Sutlej rivers.
Oldest branch of engineering, next to Military engineering. All engineering works other than for military purposes were grouped in to Civil Engineering. Mechanical, Electrical, Electronics & present day Information technology followed it.
A professional engineering discipline that deals with the analysis, design, construction and maintenance of infrastructural facilities such as buildings, bridges, dams, roads etc.
Civil Engineering is everywhere. Civil Engineering is a composite of many specific disciplines that include structural engineering, water engineering, waste material management and engineering, foundation engineering etc. among many.
Gravity dams are structures designed so that their own weight resists external forces. Concrete is the preferred material. Forces acting on the dam include water pressure, uplift pressure, earthquake forces, silt pressure, wave pressure, and ice pressure. The dam's weight counters these forces. Dams are checked when full and empty, accounting for load combinations. Gravity dams can fail due to overturning, crushing, tension cracks, or sliding along foundation planes. Design aims to prevent failure from these modes.
The document discusses the design and construction of concrete gravity dams. It begins with an introduction of dams and their purposes, then discusses site selection factors, design considerations, foundation investigations, construction procedures, and challenges in construction. The key points are that concrete gravity dams are designed so their own weight resists external forces, and their construction involves dewatering the river, building a cofferdam, removing loose materials, and placing concrete in lifts while controlling the temperature to prevent cracking.
Engineering Geology (Civil Engineering Applications)GAURAV. H .TANDON
This document discusses the important geological factors to consider when selecting sites for dams and reservoirs. Narrow river valleys, shallow bedrock, and competent bedrock foundations are desirable for reducing dam construction costs. Sedimentary rocks like sandstone and limestone can cause water leakage from reservoirs depending on their porosity. Metamorphic rocks like gneiss and quartzite are generally impermeable. Geological structures must also be considered, with horizontal or tilted strata being most suitable and faults or intense fracturing making a site undesirable. The document outlines these considerations in detail.
The Uri Hydel Power Project is a 480 MW hydroelectric power station located on the Jhelum River in Jammu and Kashmir, India. It was constructed between 1989-1997 by the National Hydroelectric Power Corporation. Key features include a 10.64 km head race tunnel, 4 units of 120 MW each located underground, and an annual generation of 2663 million units of power supplied to northern states of India. While it provides clean energy and economic benefits, the project also displaced over 400 families and required extensive afforestation efforts to mitigate environmental impacts. There is ongoing dispute between Jammu and Kashmir and NHPC over sharing of project revenues and control.
Changes in dam break hydrodynamic modelling practice - Suter et alStephen Flood
Abstract: Today, many organisations rely on hydrodynamic modelling to assess the consequences of dam break failure on downstream populations and infrastructure. The availability of finite volume shock-capturing schemes and flexible mesh schematisations in widely used software platforms imply that dam break modelling projects will be carried out differently in the future: Finite volume based platforms allow widespread application of shock-capturing methods and flexible mesh platforms can represent features in the study area more realistically and are more flexible thanks to varying mesh resolutions. Furthermore, the recent adoption of Graphics Processing Unit (GPU) technology in mainstream scientific and engineering computing will also significantly decrease computation times at relatively low cost.
This paper examines the application of finite volume, flexible mesh and GPU technologies to dam break modelling. One-dimensional (1D) modelling results are compared to those from two-dimensional (2D) finite difference and finite volume approaches. The results demonstrate that there are differences between modelling approaches and that the computational speeds of 2D simulations can be significantly reduced by the use of GPU processors.
This document summarizes a study that used computational fluid dynamics (CFD) to model soil erosion during a hole erosion test (HET). The study used the k-epsilon turbulence model in Fluent software to simulate the biphasic turbulent flow and resulting shear stresses at the water-soil interface. This allowed for a three-dimensional analysis of erosion rates along the hole, unlike typical one-dimensional models. Results showed the inlet side of the hole experienced more erosion than the outlet side due to non-uniform shear stresses. The model aims to better understand and predict piping failures in hydraulic structures by quantifying the effects of flow velocity on erosion rates.
This document summarizes a proposed new conceptual model called the Fracturing Impacted Volume (FIV) model as an alternative to the commonly used Discrete Fracture Network (DFN) model for modeling unconventional reservoirs. The FIV model accounts for the pressure-dependent permeability of the reservoir system and fluid loss during fracturing. It views fracturing as pressurizing the reservoir system within a spatial volume around the fracture, including microfractures and pores, rather than solely focusing on discrete fracture propagation. Laboratory tests on tight sandstone and shale cores show that reservoir permeability increases with pressure, supporting this conceptualization. The FIV model aims to provide a more effective way to understand fracturing mechanisms and improve simulation of fract
Design , Analysis and Optimization of Intze Type Water Tank With Sloshing EffectIRJET Journal
This document describes a study on the design, analysis and optimization of an Intze type water tank considering sloshing effects. The study involves designing a circular water tank with a capacity of 300,000 liters and height of 12 meters using design specifications from Indian codes. A 3D model of the tank is created in ANSYS and analyzed using structural and modal analysis to understand stresses and natural frequencies. The results show maximum von mises stresses of 1.61 N/mm2, which is within permissible limits. The natural frequencies help identify ranges requiring additional damping or strength. The study aims to optimize the tank design to minimize stresses and cracking considering seismic loads and sloshing effects.
The document is a final report summarizing the design of a sediment trap for a research flume. It describes the initial designs considered and the final designs selected, including drawings and calculations. The sediment trap will trap and continuously weigh sediment in the flume before removing and transporting it to a dumpster, separating the sediment from the water for recirculation.
Paper 43 - Deep Water Pipeline CT 9_2_15John Grover
This document discusses the use of coiled tubing as down-lines for pre-commissioning deepwater pipelines. It compares coiled tubing to other down-line options like flexible lines. Coiled tubing has advantages over flexible lines in terms of cost, delivery time, deck space requirements, reliability, and ability to provide contingency. The document outlines the latest custom coiled tubing equipment being deployed, which is designed for large diameters of 2 7/8" or 3 1/2" pipe and to operate in water depths up to 3,000m. It was also designed to be road transportable and for flexible installation on vessels through a moonpool or over the side.
Lyapichev. Problems in numerical analysis of CFRDs (ICOLD Bull.155)6 p.)Yury Lyapichev
The document discusses several challenges and developments in numerically analyzing concrete faced rockfill dams (CFRDs). It notes that until recently, CFRDs were designed based on experience rather than analysis. Accurate models have since shown issues like excessive compressibility of downstream rockfill adversely impacting the concrete face. The document also discusses modeling earthquakes, the need for structure-specific models in some cases, and ensuring nonlinear analysis convergence. Overall, it emphasizes the importance of numerical analysis as a tool to supplement—not replace—engineering judgment, especially for extrapolating lessons from incidents at high CFRDs.
This document describes a model developed to simulate the separation of CO2 from multicomponent natural gas mixtures using a spiral wound membrane (SWM) module. The model uses the succession stage method to discretize the separation process within the SWM. The model can evaluate product purity, hydrocarbon loss, stage cut, and permeate acid gas composition for multicomponent mixtures. Results show multicomponent systems provide higher product purity, lower hydrocarbon loss, and higher permeate acid gas content than binary systems. Different multicomponent mixtures yield varying separation performances depending on the acid gas component. The model has potential for designing and optimizing SWM systems for natural gas sweetening.
The document is a final report from a student group (Group B4) evaluating a low pressure chemical vapor deposition (LPCVD) reactor. It includes a study of how temperature, pressure, wafer spacing, inlet mole fraction affect silicon deposition uniformity in the reactor. The group found that temperature and pressure had an inverse relationship with uniformity, while wafer spacing and inlet mole fraction had a direct relationship with uniformity. Temperature was found to have the greatest effect on deposition rate.
Using Half Pipes as Permeable BreakwaterIRJET Journal
This document describes a study that investigated using half pipes as permeable breakwaters to protect coastlines in Egypt. Two types of half pipe breakwaters were tested experimentally and numerically: horizontal half pipes shaped like an H, and vertical half pipes shaped like a C. Physical models were used to identify the hydraulic performance of the barriers under different wave conditions. A numerical model was also developed using FLOW-3D software and validated against the laboratory data. The results showed that increasing the relative water depth decreases the amount of wave transmission through the barrier and increases wave reflection. Permeable breakwaters were found to effectively dissipate wave energy while avoiding issues caused by traditional solid breakwater structures.
This document provides guidelines for instrumentation of concrete and masonry dams. It outlines obligatory and optional measurements for dams, including uplift pressure, seepage, temperature, and displacement. Obligatory measurements include uplift pressure, seepage, temperature inside the dam, and displacement measurements using plumb lines or other methods. Optional measurements that may provide additional insights include stress, strain, pore pressure, and seismicity measurements. The document describes different types of measurements in detail and how they can be used to monitor dam performance and safety over time.
This paper discusses the design and analysis of a 250 meter ship floating dry dock with both monohull and twin hull configurations. Hydrostatic and stability analyses were performed using Maxsurf software to compare the ballast water displacement of both hull types under various loading conditions. The monohull design was found to meet project requirements based on the stability analysis results. General arrangements and 3D models were developed using Rhinoceros, Solidworks and AutoCAD. Structural analysis was also conducted to analyze both designs. In conclusion, the monohull configuration was selected based on the results of comparing the ballast water displacement and stability analyses of the two hull designs.
A STUDY ON THE SEISMIC RESPONSE OF ELEVATED WATER TANKIRJET Journal
- The document discusses analyzing the seismic response of an elevated water tank considering soil-structure interaction and sloshing effects.
- A 3m x 3m x 3m reinforced concrete water tank supported by a 6m tall staging is modeled and analyzed using SAP2000 software.
- Fixed base analysis is performed considering empty and full tank conditions for different soil types in seismic zones II and III. Flexible base analysis accounting for soil-structure interaction is also conducted.
- Parameters such as base shear, base moment, displacements, modal periods, and frequencies are calculated and compared between the different analysis methods and soil/tank conditions.
2014 a method for evaluation of water flooding performance in fractured res...AliMukhtar24
This document presents a new mathematical model for evaluating water flooding performance in fractured reservoirs. The model transforms a dual-porosity reservoir into an equivalent single-porosity model using a pseudo relative permeability method. This allows fractures and matrix to each have their own permeability, porosity, saturation, and relative permeability parameters. The model also accounts for imbibition effects by modifying an existing equation. The investigation shows imbibition can impact recovery and lower production rates can improve water flooding by delaying breakthrough and controlling water cut rise. A new chart is proposed to estimate ultimate recovery based on water cut versus recovery curves. The model is shown to estimate recovery within 2% of simulation results for two reservoirs, proving it a reliable evaluation method.
On Modeling Water Transport in Polymer Electrolyte Membrane Fuel Cell_Crimson...Crimson_Biostatistics
This document discusses modeling water transport in polymer electrolyte membrane fuel cells (PEMFCs). It begins by providing background on fuel cells and their components. It then discusses challenges related to water management and flooding in PEMFCs. Several existing continuum and pore-scale models for predicting water transport are reviewed, noting limitations like computational expense. The document proposes a new approach using a volume averaging technique to develop a two-dimensional macroscale model from three-dimensional microscopic equations. This new model represents an improvement over prior reduced continuum models by developing a more accurate closure model for mass exchange between layers. The goal is to use this new model to better understand water flooding and optimize fuel cell performance and durability.
This document compares the design differences between water dams and tailings dams. Some key differences discussed include:
- Tailings dams must safely contain mine tailings and process water in perpetuity after closure, unlike water dams which typically have a 100 year design life.
- Seepage control is more critical for tailings dams due to environmental regulations around containment of contaminants from tailings.
- Tailings properties, such as higher specific gravity, can increase loading stresses on the dam compared to water.
- Tailings can be used advantageously in the design to reduce hydraulic gradients and piping risk, allow use of geosynthetic filters, and provide a seepage barrier, whereas water dams rely
IRJET- Solution for Decrease in Land due to Global Warming by Constructio...IRJET Journal
This document proposes constructing very large floating structures to provide additional land area as global warming causes sea levels to rise. It discusses the causes and impacts of global warming, including melting ice caps and rising ocean temperatures. The document then presents the methodology and results of analyzing a floating pontoon structure model to demonstrate the feasibility of the concept. The pontoon model is used to calculate the depth submerged in water, which represents its self-weight, and the height above water, which represents its weight-carrying capacity. The analysis found the concept viable for supporting conventional building structures and addressing land loss from climate change.
This document proposes an alternative design for constructing the foundations of a new pedestrian bridge across a harbour. It suggests using a temporary sheet pile wall cofferdam that would allow workers to build the pile group and pile cap at the riverbed level, avoiding the need for divers. The cofferdam design is sized at 10x10m and embedded 10m deep. Calculations are presented to check for piping, heaving, and structural failure. A finite element model is also used. It is determined that drains will be needed to reduce water pressures and piping risks. The design of the internal bracing structure and construction sequence are also considered. The cofferdam is concluded to be a feasible alternative construction method for the bridge
Evaluation of Modelling of Flow in Fracturesidescitation
Heat-transport is important for geothermal exploration. The presence of
fractures can have a pronounced effect on groundwater and heat transfer.The inclusion of
fractures into geothermal reservoir models on different scales is often still a difficult task. A
comparison of approaches for flow in fractures has been carried out. A very simple
approach is to simulate fractures with thin but highly conductive layers, for instance by
applying the Cubic-Law. A more sophisticated approach, typically in FEM codes, is the
application of lower dimensional (1D/2D) high permeable discrete elements with specific
flow properties, following e.g. Hagen-Poiseuille or Manning-Strickler. However, such an
approach typically fails while studying only partly saturated fractures. For studying the
applicability of simplified fracture modellingapproaches a comparison with a CFD
(Computational Fluid Dynamics) solution was performed. Furthermore a DEM (Discrete
Element Method)approach has been illuminated. The various methodologies are studied by
varying roughnesses,this way studying the versatility of the approach. The sensitivity of
flow in fractures to various numerical parameters can be studied this way. A detailed
analysis of temperature and flow using Péclet and Reynolds numbers helps to quantify the
contributions of the different transfer processes.
This document discusses the foundation design processes for two major bridge projects - the Vasco da Gama Bridge in Lisbon, Portugal and the Rion-Antirion Bridge in Greece. For the Vasco da Gama Bridge, the foundations consisted of vertical large diameter bored concrete piles due to favorable soil conditions and design requirements. For the Rion-Antirion Bridge, the soil conditions were less favorable so an innovative foundation concept was developed and implemented, which allowed for some permanent displacement under seismic loading. The additional time for design of the Rion-Antirion Bridge was crucial to developing and validating this new foundation solution.
Removing Uninteresting Bytes in Software FuzzingAftab Hussain
Imagine a world where software fuzzing, the process of mutating bytes in test seeds to uncover hidden and erroneous program behaviors, becomes faster and more effective. A lot depends on the initial seeds, which can significantly dictate the trajectory of a fuzzing campaign, particularly in terms of how long it takes to uncover interesting behaviour in your code. We introduce DIAR, a technique designed to speedup fuzzing campaigns by pinpointing and eliminating those uninteresting bytes in the seeds. Picture this: instead of wasting valuable resources on meaningless mutations in large, bloated seeds, DIAR removes the unnecessary bytes, streamlining the entire process.
In this work, we equipped AFL, a popular fuzzer, with DIAR and examined two critical Linux libraries -- Libxml's xmllint, a tool for parsing xml documents, and Binutil's readelf, an essential debugging and security analysis command-line tool used to display detailed information about ELF (Executable and Linkable Format). Our preliminary results show that AFL+DIAR does not only discover new paths more quickly but also achieves higher coverage overall. This work thus showcases how starting with lean and optimized seeds can lead to faster, more comprehensive fuzzing campaigns -- and DIAR helps you find such seeds.
- These are slides of the talk given at IEEE International Conference on Software Testing Verification and Validation Workshop, ICSTW 2022.
Pushing the limits of ePRTC: 100ns holdover for 100 daysAdtran
At WSTS 2024, Alon Stern explored the topic of parametric holdover and explained how recent research findings can be implemented in real-world PNT networks to achieve 100 nanoseconds of accuracy for up to 100 days.
HCL Notes and Domino License Cost Reduction in the World of DLAUpanagenda
Webinar Recording: https://www.panagenda.com/webinars/hcl-notes-and-domino-license-cost-reduction-in-the-world-of-dlau/
The introduction of DLAU and the CCB & CCX licensing model caused quite a stir in the HCL community. As a Notes and Domino customer, you may have faced challenges with unexpected user counts and license costs. You probably have questions on how this new licensing approach works and how to benefit from it. Most importantly, you likely have budget constraints and want to save money where possible. Don’t worry, we can help with all of this!
We’ll show you how to fix common misconfigurations that cause higher-than-expected user counts, and how to identify accounts which you can deactivate to save money. There are also frequent patterns that can cause unnecessary cost, like using a person document instead of a mail-in for shared mailboxes. We’ll provide examples and solutions for those as well. And naturally we’ll explain the new licensing model.
Join HCL Ambassador Marc Thomas in this webinar with a special guest appearance from Franz Walder. It will give you the tools and know-how to stay on top of what is going on with Domino licensing. You will be able lower your cost through an optimized configuration and keep it low going forward.
These topics will be covered
- Reducing license cost by finding and fixing misconfigurations and superfluous accounts
- How do CCB and CCX licenses really work?
- Understanding the DLAU tool and how to best utilize it
- Tips for common problem areas, like team mailboxes, functional/test users, etc
- Practical examples and best practices to implement right away
Essentials of Automations: The Art of Triggers and Actions in FMESafe Software
In this second installment of our Essentials of Automations webinar series, we’ll explore the landscape of triggers and actions, guiding you through the nuances of authoring and adapting workspaces for seamless automations. Gain an understanding of the full spectrum of triggers and actions available in FME, empowering you to enhance your workspaces for efficient automation.
We’ll kick things off by showcasing the most commonly used event-based triggers, introducing you to various automation workflows like manual triggers, schedules, directory watchers, and more. Plus, see how these elements play out in real scenarios.
Whether you’re tweaking your current setup or building from the ground up, this session will arm you with the tools and insights needed to transform your FME usage into a powerhouse of productivity. Join us to discover effective strategies that simplify complex processes, enhancing your productivity and transforming your data management practices with FME. Let’s turn complexity into clarity and make your workspaces work wonders!
Infrastructure Challenges in Scaling RAG with Custom AI modelsZilliz
Building Retrieval-Augmented Generation (RAG) systems with open-source and custom AI models is a complex task. This talk explores the challenges in productionizing RAG systems, including retrieval performance, response synthesis, and evaluation. We’ll discuss how to leverage open-source models like text embeddings, language models, and custom fine-tuned models to enhance RAG performance. Additionally, we’ll cover how BentoML can help orchestrate and scale these AI components efficiently, ensuring seamless deployment and management of RAG systems in the cloud.
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...SOFTTECHHUB
The choice of an operating system plays a pivotal role in shaping our computing experience. For decades, Microsoft's Windows has dominated the market, offering a familiar and widely adopted platform for personal and professional use. However, as technological advancements continue to push the boundaries of innovation, alternative operating systems have emerged, challenging the status quo and offering users a fresh perspective on computing.
One such alternative that has garnered significant attention and acclaim is Nitrux Linux 3.5.0, a sleek, powerful, and user-friendly Linux distribution that promises to redefine the way we interact with our devices. With its focus on performance, security, and customization, Nitrux Linux presents a compelling case for those seeking to break free from the constraints of proprietary software and embrace the freedom and flexibility of open-source computing.
Cosa hanno in comune un mattoncino Lego e la backdoor XZ?Speck&Tech
ABSTRACT: A prima vista, un mattoncino Lego e la backdoor XZ potrebbero avere in comune il fatto di essere entrambi blocchi di costruzione, o dipendenze di progetti creativi e software. La realtà è che un mattoncino Lego e il caso della backdoor XZ hanno molto di più di tutto ciò in comune.
Partecipate alla presentazione per immergervi in una storia di interoperabilità, standard e formati aperti, per poi discutere del ruolo importante che i contributori hanno in una comunità open source sostenibile.
BIO: Sostenitrice del software libero e dei formati standard e aperti. È stata un membro attivo dei progetti Fedora e openSUSE e ha co-fondato l'Associazione LibreItalia dove è stata coinvolta in diversi eventi, migrazioni e formazione relativi a LibreOffice. In precedenza ha lavorato a migrazioni e corsi di formazione su LibreOffice per diverse amministrazioni pubbliche e privati. Da gennaio 2020 lavora in SUSE come Software Release Engineer per Uyuni e SUSE Manager e quando non segue la sua passione per i computer e per Geeko coltiva la sua curiosità per l'astronomia (da cui deriva il suo nickname deneb_alpha).
Building Production Ready Search Pipelines with Spark and MilvusZilliz
Spark is the widely used ETL tool for processing, indexing and ingesting data to serving stack for search. Milvus is the production-ready open-source vector database. In this talk we will show how to use Spark to process unstructured data to extract vector representations, and push the vectors to Milvus vector database for search serving.
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!SOFTTECHHUB
As the digital landscape continually evolves, operating systems play a critical role in shaping user experiences and productivity. The launch of Nitrux Linux 3.5.0 marks a significant milestone, offering a robust alternative to traditional systems such as Windows 11. This article delves into the essence of Nitrux Linux 3.5.0, exploring its unique features, advantages, and how it stands as a compelling choice for both casual users and tech enthusiasts.
Sudheer Mechineni, Head of Application Frameworks, Standard Chartered Bank
Discover how Standard Chartered Bank harnessed the power of Neo4j to transform complex data access challenges into a dynamic, scalable graph database solution. This keynote will cover their journey from initial adoption to deploying a fully automated, enterprise-grade causal cluster, highlighting key strategies for modelling organisational changes and ensuring robust disaster recovery. Learn how these innovations have not only enhanced Standard Chartered Bank’s data infrastructure but also positioned them as pioneers in the banking sector’s adoption of graph technology.
AI 101: An Introduction to the Basics and Impact of Artificial IntelligenceIndexBug
Imagine a world where machines not only perform tasks but also learn, adapt, and make decisions. This is the promise of Artificial Intelligence (AI), a technology that's not just enhancing our lives but revolutionizing entire industries.
1. Department of Hydraulic and Environmental Engineering
The Norwegian University of Science and Technology
CFD modelling for hydraulic structures
Nils Reidar B. Olsen
Preliminary 1st edition, 8. May 2001
ISBN 82-7598-048-8
2. CFD Modelling for Hydraulic Structures 2
Foreword
The present book is written as class notes for an advanced undergrad-
uate course in hydraulic engineering at NTNU. The course will be given
in the fifth and last year of the study. Previously, the students must have
taken the fourth year course SIB 5050 Hydroinformatics for fluvial hy-
draulics and limnology. This course gives the students basic knowledge
and experience in CFD theory and use of CFD models. The present
course builds on this knowledge, and expands further details on the spe-
cifics of modelling hydraulic structures.
In the present text, the implementation of the algorithms are focused on
the SSIIM model. This is partly because I am most familiar with this mod-
el, and partly because this model will be used by my students. The al-
ternative is to use a commercial CFD package, and this is presently too
expensive.
The text sums up experiences we have had for several years with using
SSIIM on modelling flow for hydraulic structures. I therefore I hope it can
also be useful for external SSIIM users.
I want to thank all the people who have given me advice on modelling
hydraulic structures over the years. Hilde Marie Kjellesvig must be men-
tioned especially in this respect. I also want to thank her for permission
to use her input data sets for the Himalayan Intake, in Chapter 5.5. Also
thanks to Thorsten Stösser and Tim Fisher-Antze for using the figures
and data for the Pasche case in Chapter 2.5.
3. 3 Table of content
Table of content
Foreword 2
Table of content 3
1. Introduction 4
2. Vegetation 5
2.1 Roughness 5
2.2 Formulas 6
2.3 Implementation in SSIIM 7
2.4 Example 1. Sokna river 8
2.5 Example 2. Pasche channel 11
3. Spillways 13
3.1 Algorithms for free surface flow 13
3.2 Implementation in SSIIM 15
3.3 Example 1. Standard spillway 15
3.4 Example 2. Spillway with 3D contraction 17
3.5 Example 3. Shock waves 19
4. Local scour 21
4.1 Sediment transport modelling 21
4.2 Implementation in SSIIM 22
4.3 Example: Scour around a circular cylinder 22
5. Intakes 27
5.1 Intake layout and sediment problems 27
5.2 Modelling complex structures with SSIIM 27
5.3 Example 1. Sediment deposition in a sand trap 28
5.4 Example 2. Bed changes in a sand trap 29
5.5 Example 3. Complex geometries - Himalaya intake 32
Literature 35
4. CFD Modelling for Hydraulic Structures 4
1. Introduction
In recent years computers have become fast enough to make CFD com-
putations feasible for engineering purposes. Some guidance is needed
on how to model different types of hydraulic structures. In the present
text, four types of problems are considered:
- Vegetation/stones in rivers
- Spillways
- Intakes
- Local scour
Modelling flow in natural rivers using CFD is commonly used to make
environmental assessment studies, for example habitat for fish, disper-
sion of pollutants etc. This is a relatively straightforward procedure, giv-
en todays CFD models. However, two types of problems often occur:
- The river bed contain very large stones affecting the water flow
- Part of the river is covered with vegetation, reducing the water velocity
Special algorithms has to be included to take these effects into account.
This is discussed in Chapter 2.
Looking at the problems investigated in physical laboratory models, a
large number is a study of spillways. From a point of dam safety, it is
very important to determine the coefficient of discharge with high degree
of accuracy. The spillway capacity depends on the geometry, making
separate studies necessary for each spillway. Modelling spillways is dis-
cussed in Chapter 3.
Sediment transport is a particular topic of hydraulic research. When a
construction is made in a river, for example a bridge pier, a special flow
pattern around the construction is formed. If the construction is made in
a river with erodible bed, the water flow may cause local scour around.
This may again weaken the support for the structure. Local scour is a
typical cause for bridge failures. CFD modelling of local scour is de-
scribed in Chapter 4.
Another hydraulic problem connected to sediment transport is design of
intakes for hydropower or irrigation plants. This requires special care
when the river water contains sediments. If taken into the intake, the
sediments may deposit in the intake channel, clogging it. And if sedi-
ments reach the hydropower machinery, extraordinary wear on hydrau-
lic components may occur. A CFD model may be used to investigate
how much sediment enter the intake, as a function of its design. Often,
intake works includes a desilting basin. A CFD model can be use to
computed its efficiency and the sediment deposition pattern. A discus-
sion of intakes is given in Chapter 5.
5. 5 2. Vegetation
2. Vegetation
Vegetation is probably not what most people think of as a hydraulic
structure. However, in principle a plant stem in water flow is similar to
flow around an obstruction. The same approach for modelling drag can
be applied. In the present chapter, large roughness elements in rivers,
like for example stones, are also included as the same numerical ap-
proach is used.
2.1 Roughness
A natural river will always be classified as hydraulically rough. The ve-
locity profile is then given by Schlicting’s (1979) formula:
------ = -- ln --------
U 1 30y (2.1.1)
- -
U* κ ks
U is the velocity, U* is the shear velocity, y is the distance from the wall
and ks is the wall roughness, where Schlichting used spheres glued to
a flat plate. Later studies have related the roughness to the bed sedi-
ment grain size distribution:
k s = 3d 90 (2.1.2)
where d90 is the grain size fraction of the bed where 90 % of the material
is smaller.
Schlichting’s experiments suggested the wall laws are valid in the range
where y+ is between 30 and 3000, given as:
+ U* y
y = --------- (2.1.3)
ν
where ν is the kinematic viscosity of water. Schlichting’s experiments
suggest that the formula can also be used for larger y+ values, as long
as we can assume Eq. 2.1 holds between the wall and the center of the
cell closest to the wall. For uniform flow in a wide channel, the equation
is used all the way to the water surface, so this may be a valid assump-
tion for many cases. On the other hand, if y+ is very small there may be
other problems. The physical interpretation is that the height of the
roughness elements exceeds the vertical size of the bed cell. Algorithms
with some sort of porosity can then be used (Olsen and Stokseth, 1995;
Fischer-Antze et. al., 2001), described further in the following.
As long as the vegetation height, or roughness height, is small com-
pared with the height of the bed cell, this approach gives good results.
However, for cases with large roughness compared with the cell height,
the approach does not give good results. A solution is to increase the
size of the bed cell (Wright, 2001), introducing an adaptive grid. There
are some disadvantages with this approach:
1. An adaptive grid algorithm is complex and takes computer resources.
The convergence will be slower
6. CFD Modelling for Hydraulic Structures 6
2. For some cases, the roughness is the same size as the water depth.
This can be stones being only partly submerged in the river, or it may be
vegetation where the top is above the water surface. This means that it
is not possible to find a grid cell size that is large enough.
The alternative solution is to introduce a formula for the reduction of the
water velocity inside the area of high roughness. This is further dis-
cussed in the next chapter.
2.2 Formulas
The effect of the roughness on the water is to reduce its velocity. This is
modelled as a force from the roughness on the water in the cell. The Na-
vier-Stokes equations are derived from a force balance, and the force is
then included as a sink in the equations. There are basically two ap-
proaches to computing the sink term:
1. Use formula for drag on a cylinder (vegetation stem) or a sphere
(stone/rock)
2. Use a formula for flow in porous media
Drag formula
The first approach involves a formula from classic hydromechanics, giv-
ing the force, F, on a long cylinder as:
1 2
F = -- C D ρU A
- (2.2.1)
2
F is the drag force on an object, CD is the drag coefficient, ρ is the water
density, U is the velocity and A is the surface area of the object, project-
ed normal to the flow direction. Fischer-Antze et. al. (2001) used this ap-
proach simulate vegetation in a river. Laboratory experiments were
modelled, where vertical circular rods were used to simulate the vege-
tation. Further details are given in Chapter 4.6.
The formula gives good results when the vegetation consist of stems
modelled as a number of cylinders. However, vegetation often also con-
tains leaves. Another problem is that the vegetation may bend when the
water velocity is large. This means that the force between the water and
the vegetation is not proportional to U raised to the power 2, but to a
lower value. Further research on this topic is required.
Porosity
The second approach can be based on equations for groundwater flow
(Engelund, 1953)
3
1 – nU
I = β 0 ----------- -----
3
- - (2.2.2)
n gd
Here, I is the hydraulic gradient, β0 is a constant, n is the porosity, g is
the acceleration of gravity and d is the characteristic particle diameter.
A β0 value of 3.0 was suggested by Engelund, and used by Olsen and
Stokseth.
7. 7 2. Vegetation
An algorithm to estimate the porosity as a function of the bed topogra-
phy was suggested by Olsen and Stokseth (1995):
mt – mk
n k = 1 – c k 1 – ------------------
- (2.2.3)
m t + 0.5
The formula is based on a number of randomly measured points (x,y,z)
in a river. In the 2D depth-averaged cell there are mt points. The poros-
ity, nk, at level k is given by Eq. 2.2.3, where mk is the number of points
in a cell above level k. The empirical parameter ck was varied and a val-
ue of 0.3 was found to produce reasonable results for one particular riv-
er. Olsen and Stokseth used wall laws in the cells above the porous cells
where Eq. 2.1.1 was used, also for the turbulence variables.
Stability considerations
When adding a large negative source to the Navier-Stokes equations,
instabilities may occur. This may make it necessary to use very low re-
laxation coefficients, giving long computational times. The discretization
method for the vegetation force affects the stability and convergence
(Patankar, 1980). Looking at the discretized formula for the velocity U in
cell P:
∑ anb Unb S-
U p = ----------------------- + ----
- (2.2.4)
ap ap
S is the source term, a is a weighting factor and the index nb denotes
the cells surrounding cell P. Further details are given by Olsen (2001).
There are two alternative methods to include the vegetation force:
1. Compute the force, and give this as a negative addition to the S term.
2. Increase the ap term.
Option 2 is sometimes more stable, as large negative S terms can lead
to instabilities. If the drag force is denoted F, the following relation can
be used:
F
F = a p U p or a p = ------ (2.2.5)
Up
If for example Eq. 2.2.1 is used, the addition to ap becomes:
1 2
-- C D ρU p A
-
F 2 1
a p = ------ = --------------------------- = -- C D ρ U p A
- - (2.2.6)
Up Up 2
The increased ap may sometimes also cause instabilities, so it is a still
not certain which method is better.
2.3 Implementation in SSIIM
A large number of algorithms have been implemented in SSIIM, for var-
ious types of roughness.
8. CFD Modelling for Hydraulic Structures 8
Wall laws
The default roughness routine in SSIIM uses wall laws (Eq. 2.1.1). The
roughness then has to be specified in all bed cells. Normally, the Man-
ning-Strickler roughness coefficient given on the W 1 data set is used.
This is converted into a roughness height, ks, by using the following for-
mula:
26 6
k s = -----
-
M
This value can be overrided by giving the value directly on the F 16 data
set.
Sometimes the roughness varies spatially in the geometry. In SSIIM 1,
When doing sediment the roughness can then be given in the bedrough file. Then a roughness
computations, the
is given for each bed cell. The bedrough file can conveniently be gener-
roughness can be
ated by using a spreadsheet.
computed from the
sediment grain size
distribution and the In SSIIM 2, the spatially varying roughness can be given by using the
bed form height. Roughness Editor.
Porosity
The porosity algorithm can be used in SSIIM 1 by giving a P on the F 7
data set. Also, the porosity is given in the porosity file. This can be gen-
erated using a spreadsheet, or generated by SSIIM. The basis for the
SSIIM generation is the values in the geodata file, and an algorithm
based on Eq. 2.2.3. The porosity file gives four values of the porosity at
four different levels for each bed cell. The program then interpolates lin-
early between the values for each cell above the bed.
Drag formulas
The drag formulas are given on the G 11 data set in the control file for
SSIIM 1. The data set contains six indexes, defining the block of cells
where the source is applied. Then there is the source term factor itself,
which is the stem diameter multiplied with the number of stems and the
drag coefficient. The last number is a discretization factor, affecting the
stability. If it is 1, the vegetation force will be subtracted from S in Eq.
2.2.4. If it is 2, the vegetation force will be added to the ap term, as given
in Eq. 2.2.6. If a factor between 1 and 2 is used, the force will partly be
subtracted from the S term and partly added to the ap term. A linear in-
terpolation between 1 and 2 will be used.
2.4 Example 1. Sokna river
The porosity model was tested (Olsen and Stokseth, 1995) in a reach of
the river Sokna, located in Sør-Trøndelag in the middle part of Norway.
The reach is located on a part of the river where the average slope is
approximately 1/50. At this location the mean annual flow is 12,4 m3/s
and annual maximum flow during snow-melt is about 200 m 3/s. The size
of the bed material is up to 2-3 m in diameter. Fine sand is present in the
recirculation zones. Using a theodolite, approximately 2000 points were
9. 9 2. Vegetation
surveyed within an area of 80 times 30 meters (Fig. 2.4.1). The geomet-
ric data were used as input for the geometry of the grid. The grid is
shown in Fig. 3. The geometrical data were also used as input for the
porosity modelling.
Figure 2.4.1 Grid of the
reach, with the measured
points. The different shad-
ings show varying eleva-
tions. The points are used to
make the bed levels of the
grid and the porosity file.
The water is flowing from
left to right.
The water velocities in the most upstream cross-section were measured
using a current meter at two discharges: 5.1 m3/s and 75 m3/s. The di-
rection of the inflowing water was perpendicular to this cross-section. A
map of velocity distribution was made from the measured values. This
gave the upstream boundary condition for the water flow calculation.
The water velocity for the inflowing water was given in the innflow file,
shown in Text Box 2.4.1. Linear interpolation was used for discharges
between 5.1 and 75 m3/s.
Text box 2.4.1 inn- E 2 2 0.149558 0.011505 0
flow file for Sokna, E 2 3 0.448673 0.034514 0
for 5 m3/s. This file E 2 4 0.648083 0.049853 0
gives the upstream E 2 5 0.847493 0.065192 0
boundary condition E 2 6 0.897345 0.069027 0
E 2 7 0.947198 0.072862 0
when the water ve-
E 2 8 1.046903 0.080532 0
locity is not distribut- E 2 9 0.99705 0.076697 0
ed uniformly over E 2 10 1.146608 0.088202 0
the profile. A veloci- E 2 11 1.146608 0.088202 0
ty component in x, y E 3 2 0.099705 0.00767 0
and z direction is E 3 3 0.498525 0.038349 0
given for each up- E 3 4 0.847493 0.065192 0
stream surface cell E 3 5 0.897345 0.069027 0
area. The two first E 3 6 0.867434 0.066726 0
indexes identifies E 3 7 0.837522 0.064425 0
the cell. The first in- E 3 8 0.717876 0.055222 0
dex is the cell index j E 3 9 0.548378 0.042183 0
E 3 10 0.508496 0.039115 0
in the cross-section,
E 3 11 0.478584 0.036815 0
and the second in- E 4 2 0 0 0
dex, k, is in the verti- E 4 3 0.19941 0.015339 0
cal direction. E 4 4 0.498525 0.038349 0
E 4 5 0.648083 0.049853 0
E 4 6 0.707906 0.054455 0
Given the upstream boundary condition for 5.1 and 75 m3/s, intermittent
values for other discharges were calculated by linear interpolation. Ve-
locities close to the surface and the bed are shown in Fig. 2.4.2. For cal-
ibration purposes the velocities were measured in different profiles.
10. CFD Modelling for Hydraulic Structures 10
These profiles are called transects. The velocity vector plots showed
that the flow direction was nearly perpendicular to the transects. Com-
parisons of measured and calculated velocities in the transect are given
in Fig. 2.4.3.
Figure 2.4.2 Velocity vectors in Sokna river close to the bed (left) and close to the water
Level 11
surface (right).
Changes in porosity and roughness have different effect at high and low
water discharges. At 5.1 m3/s, the calibration procedure showed that the
velocity field in the horizontal plane was strongly influenced by the value
of the ci-parameter for the porosity model. The vertical velocity profile
was more affected by variations in the roughness. At 75 m3/s there were
only small changes in the vertical and horizontal velocity profiles when
porosity and roughness were varied. The best total fitting for the three
transect was obtained by a roughness coefficient of 0.1, a minimum po-
rosity of 0.4 and ci=0.3 for both i=1 and i=2 (Equation 2.2.3).
Figure 2.4.3. Measured (x) and computed (lines) velocity profiles in three
cross-sections (transects) of the river Sokna.
Velocity measurements were also taken for verification purposes. These
measurements were taken at 10.0 m3/s for transect 4 and 6.28 m3/s for
transect 2 and 3. In Fig. 6 the calculated and observed velocities are
11. 11 2. Vegetation
shown. The figure shows that the differences between modelled and
measured velocities are nearly the same as for the calibrated data.
Discussion
Fig. 2.4.3 indicates that there is fairly good agreement between meas-
ured and calculated velocities. In areas with rocks and boulders the ve-
locity is small. The velocity pattern shown in Fig. 2.4.2 also coincides
fairly well with the measurements of the flow pattern.
One reason for the deviation between measured and calculated veloci-
ties can be uncertainties in the measurements of the velocities and the
geometry. The vertical location of the bed surface is estimated from lin-
ear interpolation from measured point values. If very few points are lo-
cated within a bed cell, the geometry will be inaccurately modelled.
The largest deviations between the measured and modelled velocities
are found for the velocity distribution at low discharges. This indicates
that the main source of the discrepancies is in the modelling of porosity.
At low discharges the roughness elements are large compared to the
flow depth and the velocity decreases where boulders and stones are
located. This can cause the velocity to vary from 0.0 to 1.5 m/s within a
horizontal distance of only 1-2 meters. The size of the grid cells are typ-
ically 2 times 2 meter in the horizontal plane. The porosity is modelled
as an average for each grid cell, while the roughness element distribu-
tion within one grid cell of the prototype can vary greatly. Reducing the
size of the grid cells in ares of high porosity will therefore probably in-
crease the accuracy in these areas.
The porosity model presented in this study includes one parameter, ci
(Equation 2.3.4), which is calibrated. The parameter can vary between
0.0 and 1.0, and it is found that a value of 0.3 works well. The same pa-
rameter is used for all different discharges and physical locations of the
river. This may be an indication that calculations with the same value of
this parameter can give reasonable results also for other rivers. Further
testing is needed to investigate this.
2.5 Example 2. Pasche channel
Fisher-Antze et. al. (2001) tested the drag formula approach to three dif-
ferent flow cases involving artificial vegetation in straight laboratory
channels. One case was a model by Pasche (1984), involving a com-
pound channel with vegetation on the overbanks. Different vegetation
densities λ, different vegetative element diameters and different vegeta-
tive heights h were used. The geometry and vegetation arrangements
of the laboratory experiments are shown in Fig. 2.5.1.
The vegetation elements were modelled with the G 11 data set:
G 11 2 337 11 20 2 7 0.012 1
The cells in the grid has a size of 8.93 x 5 cm. The vegetation elements
were vertical circular cylinders. With one cylinder in each cell, a CD fac-
12. CFD Modelling for Hydraulic Structures 12
tor of 1.0 and a cylinder diameter of 1.2 cm, the second-last parameter
on the G 11 data set becomes:
C D nd = 1.0x1x0.012 = 0.012
Figure 2.5.1 Cross-
section of Pasche’s
channel, showing an
overbank with vege-
tation rods and a
main channel with-
out.
Figure 2.5.2 shows the comparison of simulated results against ob-
served velocity. The reduction of flow velocities through the vegetation
elements can be simulated fairly well.
Figure 2.5.2 Ve- a.
locity profile in 0.5
the channel. The 0.4
direction is op-
u [m/s]
0.3
posite of Fig.
2.5.1. 0.2
0.1
0.0
0 0.2 0.4 0.6 0.8 1
W idth [m]
calculated measured
Fischer-Antze et al. (2001) also tested the model with varying number of
grid densities, and obtained very good results for both fine and coarse
grids. The water flow direction was fairly perpendicular to the grid for this
case, so there were very little false diffusion.
13. 13 3. Spillways
3. Spillways
The hydraulic design of a spillway involves two problems:
1. Determination of coefficient of discharge
2. Dissipation of the energy downstream.
Problem 2 is often more complex, as air entrainment causes two-phase
flow. The algorithms described in the following is mainly focused on one-
phase flow, and are therefore limited when modelling problem 2.
Early work on modelling spillways include a demonstration case from
FLOW3D modelling a Parshall flume. This was presented as a demo at
the IAHR Biennial Conference in London in 1995 and later by Richard-
son (1997). FLOW3D uses a volume of fluid method (described below)
on an orthogonal grid. Later work was given by Kjellesvig (1996) and
Olsen and Kjellesvig (1998b), computing some of the examples given in
the following chapters. Spaliviero and May (1998) used FLOW3D to
computed coefficient of discharge for a spillway. Yang and Johansson
(1998) computed coefficient of discharge for the same spillway using
three CFD models (CFX, FLUENT and FLOW3D), and found the CFX
model to give best results. The suggested reason was that the adaptive
grid gave a more accurate location of the water surface.
3.1 Algorithms for free surface flow
There are a number of different methods to compute the location of the
free surface in a CFD model:
1. 1D backwater computation
2. 2D depth-averaged approach,
3. Using the 3D pressure field
4. Using water continuity in the cell closest to the water surface
5. The volume of fluid (VOF) method
The methods are described in more details in the following.
1D backwater computation
The 1D backwater computation is based on a friction loss, If, computed
by Mannings formula:
nU
I f = -----------
1
(3.1.1)
--
-
3
r
U is the average velocity, n is Manning’s friction coefficient and r is the
hydraulic radius of the flow. The 1D backwater computation uses the fol-
lowing formula to compute the water elevation difference, ∆z, between
two cross-sections 1 and 2:
2 2
U1 U2
∆z = I f ∆x + -------- – --------
- - (3.1.2)
2g 2g
14. CFD Modelling for Hydraulic Structures 14
The horizontal distance between the cross-sections is denoted ∆x, and
g is the acceleration of gravity.
2D Depth-averaged approach
In a depth-averaged approach, it is assumed that the pressure distribu-
tion in the vertical direction is hydrostatic. This gives a direct proportion-
ality between the pressure, P, and the water depth, H:
P = ρgH (3.1.3)
The water density is denoted ρ. When the 2D CFD model computes the
water depth, Eq. 3.1.3 is used, and after convergence, the water depth
is also found.
Using the 3D pressure field
This method is often used when the water surface slope is not very
large, but still has a spatial variation significant to be of interest. The
principle is first to compute the pressure at the water surface.This is
done by linear extrapolation from the surface cell and the cell below. The
pressure is then interpolated from the cells to the grid intersections.
Then a reference point is chosen, which does not move. This can typi-
cally be a downstream boundary for subcritical flow. Then the pressure
difference between the each grid intersection and the reference point is
computed. This is assumed to be linearly proportional to the elevation
difference between the two points, according to:
( P ij – P r )
z ij – z r = ----------------------
- (3.1.4)
ρg
z is the level of the water surface, P is the pressure at the water surface,
ρ is the water density and g is acceleration of gravity. The index of the
variables are r for the reference point and i,j for the grid intersections.
The water surface is then moved for every n’th iteration, where n is de-
termined by the user. After each movement, the grid is regenerated be-
fore the computation proceeds.
Using water continuity in the cell closest to the water surface
This method is used by SSIIM in all computations of coefficient of dis-
charge for spillways.
Most often, the gravity term is not included in the solution of the Navier-
Stokes equations for computing river flow. This is because the gravity
term is a very large source term and introduces instabilities. However,
when the water surface is complex and an unknown part of the solution,
the gravity term have to be included.
Normally, the SIMPLE method is used to compute the pressure in the
whole domain. The pressure correction is based on the water continuity
defect in each cell. The current method is based on using the SIMPLE
method to compute the pressure in each cell except the cells bordering
the water surface. Instead, the pressure in these cells are interpolated
15. 15 3. Spillways
from the pressure at the surface and the pressure in the cell below the
surface cell.
Volume of fluid method
This method is not used by SSIIM, but it is often used by other CFD
models, computing free surfaces with complex shapes.
The main principle is to do a two-phase flow calculation, where the grid
is filled with water and air. The fraction of water in each cell is computed.
The location of the water surface is then computed to be in the cells with
partial fraction of water. This allows for a very complex water surface,
with air pocket inside the water and parts of water in the air. Such com-
plexity is often encountered in hydraulic jumps downstream spillways.
Note that the grid is kept fixed trough out the computation. This gives a
more stable solution, but some reduction in the accuracy may follow.
3.2 Implementation in SSIIM
A spillway has fairly complex water surface, so it is necessary to use a
fully 3D approach. Initially, it was tested if the pressure method could be
used to model the spillway. This did not give any physically realistic re-
sult. In all successful cases using SSIIM, the method of using water con-
tinuity defect in the cell closest to the water surface has been used.
This method is invoked by specifying F 36 1 in the control file. The meth-
od adds gravity to the Navier-Stokes equations in the vertical direction.
Since this is a very large source term, the solution becomes unstable. A
transient computation must be used, with very short time step. The time
step is given on the F 33 data set. The first parameter is the time step,
and the second parameter is the inner iterations for each time step. Ex-
perience shows that it is advisable to use a shorter time step and keep
a low number of inner iterations instead of vice versa. To improve stabil-
ity, very low relaxation coefficients should also be used. A typical data
set in the control file can be: K 3 0.1 0.1 0.1 0.02 0.05 0.05
Often, the spillway itself is vertical on the upstream side. This is usually
modelled by blocking out cells in the spillway. The G 13 data set is then
used.
The principle is to use a submerged spillway as the initial grid. Then the
water surface adjusts itself to the flow field. Experience shows that using
the default zero-gradient boundary condition gives better stability than
using the G 7 data sets. If G 7 data sets are used, they must be used for
all inflow and outflow boundaries.
3.3 Example 1. Standard spillway
The standard type spillway has a longitudinal bed profile corresponding
to the theoretical shape of a free-fall spillway (US Bureau of Reclama-
tion, 1973). The computed velocity field and water level at different times
are given in Fig. 3.3.1. The computation starts with a horizontal water
16. CFD Modelling for Hydraulic Structures 16
surface. This is then updated based on the algorithm using a gravity
term and water continuity for the cells closest to the water surface.
5 ms 50 ms
0.5 sec. 93 sec.
Figure 3.3.1 Longitudinal profile of the water level and velocities for computation of coefficient
of discharge for a spillway. The numbers show the computed time.
The spillway itself is modelled as an outblocked region, using the G 13
data set. A more detailed example on how this can be done is given in
the next chapter. The computation is very unstable, so a very small time
step is used, together with low relaxation coefficients. This is given on
the following data sets, taken from the original control file:
F 33 0.0005 1
K 3 0.1 0.1 0.1 0.02 0.1 0.1
The specification of the boundary condition for the inflow/outflow and the
water levels upstream and downstream is given in the timei file. The fol-
lowing file was used for this case:
I 0.0 1.0 -1.0 -1.5 -0.5
I 120 1.0 -1.0 -1.5 -0.5
Each line in the timei file starts with a character. The I data set specifies
a data set at a certain time. The time is given as the first number on the
data set. In this case, there are only two data sets, at 0 and 120 sec-
onds. The data sets that follow, are similar. A linear interpolation be-
tween the data sets are used, for times between 0 and 120 seconds.
The data set after the time step is the inflowing water discharge: 1 m3/s.
Then the outflowing water discharge is given. A negative number means
that a zero gradient boundary condition is used, and the outflow water
discharge is not specified. This is the only option that gives sufficient sta-
bility for the computation to converge. A fixed downstream water dis-
charge will not be correct anyway, as it will vary over time as the water
level varies.
17. 17 3. Spillways
The two last numbers are the upstream and downstream water level.
Negative numbers are given, which means SSIIM tries to compute the
values. The upstream water level has to be computed by the program,
as this is used to find the coefficient of discharge. It is also very difficult
to know the downstream water level, so the solution is to let the program
compute it.
The deviation between the computed coefficient of discharge and the
one given by US Bureau of Reclamation (1973) was 0.5 %.
3.4 Example 2. Spillway with 3D contraction
The 3D spillway was made to investigate the ability of the CFD model to
compute three-dimensional effects in the spillway. A physical model was
made of plywood, and inserted into a 0.5 m wide and 0.6 m deep flume.
To facilitate the construction of the model, no surfaces were made
curved. Fig. 3.4.1 shows the grid seen from above.
Figure 3.4.1. 3D spillway seen from above. The right figure shows the grid, and
the left figure shows the velocity field close to the bed of the grid. The contrac-
tion and the downstream part of the grid is blocked out.
Fig. 3.4.2 shows a longitudinal profile of the spillway with velocity vec-
tors and the final location of the free surface. The water discharge in the
physical model was 60 l/s. The difference between the computed and
measured coefficient of discharge was 0.5 %.
Figure 3.4.2 Velocity vectors in
a longitudinal profile
18. CFD Modelling for Hydraulic Structures 18
Defining the outblocked region
The spillway has a vertical upstream side. The only way to model such
a structure in SSIIM is to block out some cells. Comparing Fig. 3.4.2 and
Fig. 3.4.3, it can be seen which cells are blocked out. The outblocking is
defined in the initial grid, shown in Fig. 3.4.4.
Blocked out
Figure 3.4.3 Longitudinal profile of
the grid for the converged solution
k=11
k=6
i=2 i=13
Figure 3.4.4 Longitudinal grid at start of com- k=2
putation. The cell numbers are indicated. The
block starts at i=13 in the horizontal direction.
It ends at k=6 in the vertical direction. i=35
Looking at Fig. 3.4.4, the outblocked region is from i=13 to i=35 in the
streamwise direction. In the vertical direction, the outblocked region is
from k=2 to k=6. The transverse direction has 10 cells, numbered from
2 to 11. The data set in the control file to block out the cells becomes:
G 13 2 13 35 2 11 2 6
The first number, 2, indicates that wall laws are to be used at the top of
the outblocked region. The six following numbers indicate the first and
last cell number in the three directions.
For the spillway case, the water level will move. The size of the grid cells
below the water level will then change in magnitude. However, it is im-
portant that the top of the outblocked region does not move. Another
19. 19 3. Spillways
problem is to specify the slope of the downstream part of the spillway
bed, on top of the outblocked region.
SSIIM distributes the grid cells in the vertical direction according to a ta-
ble given on the G 3 data set. The table gives the vertical distance for
each grid line above the bed as a percentage of the water depth. For the
current case, the data set is given as:
G 3 0.0 10 20 30 40 50 60 70 80 90 100
There are 11 lines and 10 cells in the vertical direction. The first line is
located at 0.0 % of the depth, the second line at 10 % of the depth etc.
The top of the outblocked region is at the fifth grid line, or at 50 % of the
water depth. Looking at Fig. 3.4.4, this means that the slope of the spill-
way is 50 % of the slope of the grid at the lowest level. Since the lowest
level is inside the block, there is no water there. The level can then be
chosen so that the top of the block corresponds with the correct slope.
The level at the bed or bottom of the block is given in the koordina file,
which was generated by a spreadsheet.
An alternative approach would have been to use several G 16 data sets.
These data sets specify a local grid distribution, similar to the G 3 data
set. But before the distribution itself, four integers are read. This gives
the location of the part of the grid where the data set is applied. The G
16 data sets are further described in Chapter 5.5.
3.5 Example 3. Shock waves
The shock wave is a classical test case for CFD modelling of supercrit-
ical flow. Flow with high Froude number in a channel encounters a con-
traction. This causes a standing wave to form, also called a shock wave.
The engineering problem of a shock wave is related to the design of
spillways. Downstream a spillway the shock waves may form, and the
structure needs to be designed to handle the increase in water depth.
The size and shape of the wave can be analysed using simplified theory,
assuming hydrostatic pressure in the vertical direction etc. Measure-
ments of shock waves in physical models have shown that 3D effects
influence the results.
In the following, a physical model test case done by Reinauer (1984)
was used. A 1 meter wide horizontal channel has an inlet with water
depth 5 cm, and a Froude number of 6. The channel contracts, as shown
in the grid in Fig. 3.5.1.
The case was computed by Krüger and Olsen (2001). The water velocity
was relatively uniform throughout the domain, but the water depth var-
ied, forming shock waves. The result (Fig. 3.5.2) is according to classi-
cal theory and Reinauers experiments, but the maximum water depth is
underpredicted. This may be due to air entrainment in the physical mod-
el, which is not included in the CFD model.
20. CFD Modelling for Hydraulic Structures 20
Figure 3.5.1 Grid for shock wave computation. The water is flowing from left to right.
7 9 11 13
7
7 9 9 7
5
7
9 7 7 9 11 13
15
Figure 3.5.2 Water depth (cm) for shock wave computation, for Fr=6.
The case was modelled similar to the spillway cases. A timei file was
made, fixing the upstream water discharge and the upstream water lev-
el. The downstream water discharge and water level was not specified,
and computed by the program.
Very low relaxation coefficients were used, similar to the other spillway
cases. A time step of 0.0001 seconds was used, with 3 inner iterations/
time step (F 33 data set).
Initial water velocity in the grid was set to 4.2 m/s, using the G 8 data set.
This was similar to the inflow water velocity. The minimum water depth
was also given, on the F 108 data set to 0.04 meters. This prevented in-
stabilities.
21. 21 4. Local scour
4. Local scour
Modelling local scour firstly requires the modelling of the water flow field
around an obstacle. The bed shear stress is then found, and it is possi-
ble to assess the potential for erosion. If movement of the bed is predict-
ed, it is possible to try to predict the shape and magnitude of the scour
hole. Then a computation with the new geometry can be carried out. Af-
ter some iteration, it is possible to estimate the size of the scour hole.
This approach was used by Richardson and Panchang (1998).
Olsen and Melaaen (1993) used a slightly different approach. The bed
shear stress was used to compute the bed changes assuming a long
time step with steady flow. This was iterated 10 time steps, giving a
scour hole shape very similar to what was obtained in a physical model
study. But neighter the CFD case nor the physical model test was run to
equilibrium scour hole depth occurred. But this was later done by Olsen
(1996) and Olsen and Kjellesvig (1998). This example is explained in
more detail in Chapter 4.3.
Later, Roulund (2000) also computed maximum local scour hole depth
using a similar approach as Olsen and Kjellesvig. Roulund, however,
used a finer grid and the k-ω model instead of the k-ε model. Roulund
also carried out a detailed physical modelling experiment to evaluate the
CFD results.
4.1 Sediment transport modelling
Sediment transport is computed by solving the convection-diffusion
equation for suspended sediment transport in addition to computing the
bed load transport. Details are given by Olsen (1999).
Special algorithms for erosion on a sloping bed
The local scour case showed that the algorithms for erosion on a sloping
bed is very important. There are two main algorithms:
1. Reduction in the critical shear stress
2. Sand slides
If the bed slopes upwards or sideways compared to the velocity vector,
the critical shear stress for the particle will change. The decrease factor,
K, as a function of the sloping bed was given by Brooks (1963):
sin φ sin α sin φ sin α tan φ 2
K = – ---------------------- + ---------------------- – cos φ 1 – ----------
2
- - -
tan θ tan θ tan θ
(4.1.1)
The angle between the flow direction and a line normal to bed plane is
denoted α. The slope angle is denoted φ and θ is a kind of angle of re-
pose for the sediments. θ is actually an empirical parameter based on
flume studies. The factor K is multiplied with the critical shear stress for
22. CFD Modelling for Hydraulic Structures 22
a horizontal surface to give the effective critical shear stress for a sedi-
ment particle.
4.2 Implementation in SSIIM
The grid around the obstacle needs to be made. Often, an outblocking
of the grid is used to model the geometry. The outblocking of the cells
are given on the G 13 data set in the control file.
Data on the sediment type need to be given. This is done on the S data
set in the control file, where sediment particle size and diameter is given.
For non-uniform sediment grain size distribution, several particle sizes
can be modelled simultaneously. Each size is then given on one S data
set. The data sets are numbered, so that the coarsest size is given on
the S 1 data set, the second coarsest size given on the S 2 data set etc.
The fraction of each sediment size in the bed material is given on the N
data sets. If there are n number of sediment sizes, there must also be n
number of N data sets.
The algorithm for reduction of the shear stress as a function of the slop-
ing bed is invoked by specifying F 7 B in the control file. The slope pa-
rameter in Brook’s formula can be given on the F 109 data set. Default
values are:
F 109 1.23 0.78 0.2
The two first floats are tan(θ) for uphill and downhill slopes, where θ is
given in Eq. 4.1.1. The third float is a minimum value for the reduction
factor. The default values were hard-coded in earlier versions of SSIIM.
The sand slide algorithm is invoked by using the F 56 data set. Two
numbers are given, first an integer and then a float. The float is tan (an-
gle of repose). When a grid line steeper then specified is encountered,
this is corrected by the sand slide algorithm. However, after the correc-
tion, the neighbouring grid lines may be steeper then specified, even if
they were not before the correction. To handle the problem, al the lines
are checked several times. The number of times is given on the first in-
teger on the data set.
4.3 Example: Scour around a circular cylinder
The circular cylinder is the most commonly used case when studying lo-
cal scour. There exist a number of empirical formulas for the scour
depth. Also, studies have been made describing the scour process and
evolution of the scour hole.The case presented here was done by Olsen
and Kjellesvig (1998a). A cylinder with diameter 1.5 meter was used in
a 8.5 meter wide and 23 meter long flume. The water depth was 2 me-
ters, and the upstream average water velocity was 1.5 m/s. The sedi-
ments on the bed had a uniform distribution with average diameter 20
mm.
The grid had 70x56x20 cells in the streamwise, lateral and vertical direc-
tion, respectively. The grid seen from above is given in Fig. 4.3.1. The
23. 23 4. Local scour
time step for the computation was 100 seconds, and a total of 1.5 million
seconds were computed. The scour depth in front of the cylinder as a
function of the computed time decreased similarly as given in Fig. 4.3.2.
The steady state scour hole then had a geometry shown in Fig. 4.3.3.
The maximum scour depth compared reasonably well with empirical for-
mulas (Olsen and Kjellesvig, 1998a).
The sediments were modelled on the S data set, with the following val-
ues: S 1 0.02 1.75. The last number is the fall velocity of the sediments.
A value of 1.75 m/s was used. The current case was clear-water scour,
so now sediment inflow was given the I data set therefore was: I 1 0.0
Figure 4.3.1 Grid of the cylinder, seen from above. The water flow direc-
tion is from left to right.
Figure 4.3.2 Exam-
ple of time series of
bed elevation in
front of the cylinder
(p.s. not from this
case)
24. CFD Modelling for Hydraulic Structures 24
Figure 4.3.3 Contour map of computed bed elevations (meters)
Defining the outblocked region
Modelling local scour around an obstacle, it is often useful to block out
a part of the grid where the obstacle is located. An example is given in
Fig. 4.3.3.:
(i=18,j=19)
(i=11,j=12)
Level 2
Figure 4.3.3 Detail of a grid around a cylinder, with cell numbers for the two cells defining
the outblocked region.
The definition of the cells that are outblocked is given on the G 13 data
set in the control file. Modelling the cylinder in Fig. 4.3.3, and assuming
there are 20 cells in the vertical direction, with index from 2 to 21, the
data set would be:
G 13 1 11 18 12 19 2 21
25. 25 4. Local scour
The first integer indicates which sides of the outblocked region has wall
laws. The index 1 defines wall laws on the vertical sides. It is also pos-
sible to model a submerged outblocked region, in which case the index
should be 2. Such cases are described in Chapter 3.
When making the G 13 data set, it is advisable to test the numbers by
looking at the grid in SSIIM. Red lines will be shown around the out-
blocked region when velocity vectors are shown. Then it usually can be
seen if the correct numbers are given.
Specifying curved surfaces
The Grid Editor is often used to make the grid for SSIIM cases. The ed-
itor does not have any tools for specifying curved surfaces. Modelling a
circular cylinder, this is necessary.
In the current case, the cylinder geometry was generated using a
spreadsheet. Coordinates for the lines at the boundary of the cylinder
can be given in the koordina.mod file. An example of such a file is given
in Text Box 4.3.1. This file can be read by the Grid Editor. Then these
points can be fixed in the Grid Editor by defining them as NoMovePoints.
An elliptic generator can then be used to get a smooth grid outside the
cylinder.
The koordina.mod file does not have to contain all the points in the grid,
as opposed to the koordina file.
line i=10 line i=18
line j=19
line j=12
Level 2
Figure 4.3.4 Detail of a grid around a cylinder, with line numbers for the bound-
ary of the outblocked region. The numbers correspond with the cell numbers in
Fig. 4.3.3.
26. CFD Modelling for Hydraulic Structures 26
Text Box 4.3.1 The 10 12 3.317643 3.481477 0
file koordina.mod for 10 13 3.31355 3.626353 0
the cylinder case in 10 14 3.312283 3.75 0
Fig. 4.3.3 and Fig. 10 15 3.31355 3.873647 0
4.3.4. Note that only 10 16 3.317643 4.018523 0
10 17 3.32434 4.185914 0
the lines on Fig. 4.3.4
10 18 3.333322 4.373602 0
are given, and only on 10 19 3.374111 4.559002 0
the cylinder walls. The 18 12 4.546907 3.486203 0
file can be generated 18 13 4.580889 3.620445 0
using a spreadsheet, 18 14 4.589555 3.75 0
where the formula for 18 15 4.580889 3.879555 0
a circle can be used. 18 16 4.546907 4.013797 0
The file is then placed 18 17 4.496345 4.15109 0
on the same directory 18 18 4.446408 4.307081 0
as SSIIM. When the 18 19 4.433154 4.521925 0
Grid Editor is started, 11 12 3.486356 3.480756 0
the file is read by in- 12 12 3.61656 3.479054 0
13 12 3.744036 3.477904 0
voking a menu com-
14 12 3.894862 3.477595 0
mend. 15 12 4.067451 3.477815 0
16 12 4.254552 3.475646 0
17 12 4.44291 3.462987 0
11 19 3.502721 4.596612 0
12 19 3.622959 4.624805 0
13 19 3.747236 4.6369 0
14 19 3.883652 4.628207 0
15 19 4.022004 4.599478 0
16 19 4.157544 4.561292 0
17 19 4.289437 4.531614 0
27. 27 5. Intakes
5. Intakes
SSIIM is an acro-
Design of run-of-river water intakes is a particular problem in sediment- nym for Sediment
carrying rivers. Often, a physical model study is done to assist in hydrau- Simulation In In-
lic design. However, it is very difficult to model fine sediments in a phys- takes with Multi-
ical model. This was the main motivation for making the SSIIM model. block option
The SSIIM model has been used to compute trap efficiency of a sand
trap (Olsen and Skoglund, 1994), also shown in Chapter 5.3. Later it has
been used to model bed changes in a sand trap (Olsen and Kjellesvig,
1998), further described in Chapter 5.4. Flow in a very complex intake
structure, the Himalayan Intake is presented in Chapter 5.5.
Complex intakes has also been modelled by Demny et. al. (1998), using
an in-house finite element model of the Aachen University of Technolo-
gy, Germany. Sediment flow in intakes has been computed by Atkinson
(1995), for two prototype cases. The CFD model PHOENICS was used.
5.1 Intake design and sediment problems
Most hydraulic structures will be surrounded by a fairly complex three-
dimensional flow field. For water flow with sediments, the suspended
particles will move with the flow. A good hydraulic design of an run-of-
the-river intake is often essential for its performance.
There are some hydraulic principles involved in the intake design. One
guideline is to avoid recirculation zones. These create turbulence, keep-
ing suspended sediments in the water for a longer time, instead of set-
tling out. When it comes to bed load, the creation of secondary currents
is often used to prevent the sediments from entering the intake. An ex-
ample is to have the intake located at the outside of a river bend, using
the natural secondary current to sweep the bed load away from the in-
take. It is also possible to use the secondary currents emerging from
flow around obstructions to move the bed load away from the intake.
A CFD model including all these processes has to be fully three-dimen-
sional, capable of modelling both secondary currents, recirculation
zones and turbulence. It also should take into account bed changes as
a function of sediment deposition/scour.
5.2 Modelling complex structures with SSIIM
Often, a hydraulic structure is best modelled by SSIIM 1. This is because
of its outblocking options and the possibility to create walls along grid
lines. A hydraulic structure often has vertical walls with channel open-
ings, and this is problematic to model for SSIIM 2. The outblocked areas
are modelled with the G 13 data sets in the control file. The walls be-
tween cells are modelled with the W 4 data set.
An intake may also have a complex inflow and outflow. The default in-
flow for SSIIM 1 is over the whole of the upstream cross-section. And
the default outflow is over the whole downstream cross-section. The
sides are modelled with walls as default. It is possible to change this, by
28. CFD Modelling for Hydraulic Structures 28
adding walls to parts of the upstream or downstream cross-section. Or
specifying inflow/outflow on the side walls or in the bed. Removing or
adding walls are done with the W 4 data set. Specifying inflow/outflow is
done with the G 7 data set. Note that if the G 7 data set is used, one
should specify inflow/outflow over the whole geometry, and not use the
default inflow/outflow. Also, the W 4 data set must be used to remove
the walls in the areas where the G 7 data set is used.
The water surface is modelled with zero gradients as default, but it is
also possible to change this. There are two approaches:
1. Specify a closed conduit simulation on the K 2 data set
2. Add walls on the surface by using the W 4 data set
If option 1 is used, the option K 2 0 0 should be used. The coordinates
for the top of the geometry then has to be given in the koordina file. An
extra double is then given on each line, specifying the vertical elevation
of the roof of the conduit. This can also be used to model a curved roof.
If option 2 is used, the wall is specified with the W 4 data set and 3 -1 as
the first and second integer.
Note that multiple G 7, G 13 and W 4 data sets can be given, enabling
multiple walls, outblocked cells and inflow/outflows.
The default sediment inflow is at the upstream cross-section, and it is
specified with the I data set of the control file. There is one I data set for
each sediment fraction. This will distribute the sediment over the up-
stream cross-section, so if this is partially blocked out by walls or out-
blocked cells, the specification will not be correct. Then the G 20 data
set should be used instead. This can also be used if there is sediment
inflow from other locations than the upstream cross-section.
Modelling an intake, it is sometimes important to know how much sedi-
ments enter the intake, and how much enters the spillway area. This can
be determined using multiple G 21 data set. A region is then defined,
where the water and sediment flows through. For each G 21 data set,
the sediment flux will be written to the boogie file.
5.3 Example 1. Sediment deposition in a sand trap
The flow of water and sediments in a three-dimensional sand trap was
calculated for a steady-state situation. A physical model study carried
out to verify the results. It showed the recirculation zone to be shorter
than the numerical model result, but modifications in the k-ε turbulence
model gave better agreement. The sediment concentration calculations
compare well with the experimental procedures. The concentration cal-
culated by using the flow field from the original k-ε model gave 87.1 %
trap efficiency, whereas calculation with the more correct flow field gave
88.3 %.
29. 29 5. Intakes
Figure 5.3.1 Entrance region of the sand trap. The water is flowing from left to right. The
units are in meters.
Figure 5.3.2. Location of measure-
ment points in three profiles (above),
together with measured (crosses)
and computed (lines) concentrations
(right). The line A denotes computa-
tions with the modified k-ε model, with
the more correct velocity field. The
line B denotes computations with the
original k-ε model.
In the current case, the SSIIM model was not used for computing the
water flow, only the sediments. In SSIIM it is not possible to modify the
k-ε turbulence model.
Fig. 5.3.2 was generated using the VerifyProfile graphics in SSIIM. The
measured sediment concentrations are then given in the verify file. The
file also contains the geometrical locations of the measurements. Pro-
files similar to Fig. 5.3.2 can then be generated.
5.4 Example 2. Bed changes in a sand trap
The computation of the sediment concentration in Example 1 only con-
sidered a steady state. When sediment deposit, the geometry changes,
30. CFD Modelling for Hydraulic Structures 30
together with the flow field. From an engineering/design point of view, it
is important to assess how the bed changes affect the hydraulic struc-
ture. The purpose of the current case was therefore to see how well the
CFD model could predict bed changes in a sand trap. A physical model
study was carried out to verify the numerical model. An existing physical
model of the sand trap was used. The model had previously been used
in a study of the sand trap at Svartisen Hydropower Project in Norway.
The model was 5.18 meters long, 0.3 m wide and 0.3 m high. The sand
trap did not have a free surface, and was modelled as a closed conduit.
This meant the K 2 data set in the control file was set to K 2 0 0, and the
roof of the sand trap was specified in the koordina file.
The water discharge was kept constant at 15 litres/second during the
study. Sediments were added upstream, at a rate of 0.91 kg/minute. The
sediment feeding and water inflow were interrupted every 22 minutes,
when the bed levels were measured. This was done four times, giving a
total sediment inflow of 80 kg after 88 minutes.
The numerical model used a grid with 96 x 13 x 15 cells in the stream
wise, cross-streamwise and vertical direction respectively. The geomet-
rical data given to the numerical model was in scale 1:1 in relation to the
physical model.
Five sediment sizes were used to simulate the grain size distribution,
given on five S data sets. The inflow sediment load was given on five
corresponding I data sets. The roughness of the walls and the bed were
specified according to the physical model. A roughness of 0.01 mm was
given at the plexiglass wall, 3 mm was given where the roughness ele-
ments were placed, and 1 mm was given on the bed where the sand de-
posited.
Fig. 5.4.1 shows a contour map of the bed changes in the sand trap at
different times. There entrance region of the sand trap was formed so
that the water entered like a jet following the bed. A recirculation zone
was formed at the roof. The recirculation zone was observed in both the
physical model and on the CFD results. Fig. 5.4.1 shows the movement
of the deposits in the flow direction. It also shows the vertical growth of
the deposits.
Figure 5.4.1 Plan view of the
sand trap, including the comput-
ed bed elevation changes after
a) 22 min, b) 44 min, c) 66 min
and d) 88 min. The values are
given in cm.
Fig. 4.5.2 shows the calculated sediment concentrations in a longitudi-
nal plane at various times. There is a reduction in the sediment concen-
tration along the flow path, corresponding to the deposition of
sediments. Also, the sediment concentration is higher close to the bed
31. 31 5. Intakes
than near the roof. This is in accordance with the theory. The figure
shows the time evolution of the deposit. At the upstream end the mag-
nitude of the deposit decrease as sediments are eroded at this location.
This was also observed in the physical model.
Figure 5.4.2 Longitudinal pro-
files of the sand trap with pro-
files of suspended sediment
concentrations. The bed
changes at the centerline of
the profile is also shown, after
a) 22 min, b) 44 min, c) 66 min
and d) 88 min.
Fig. 5.4.3 shows a comparison between the measured and calculated
bed elevation changes after 88 minutes. There is good correspondence
between the calculated and measured values. Some deviation at the
start and at the front of the deposition will be discussed later.
Figure 5.4.3 Computed (A) and meas-
ured (B) bed changes in the sand
trap.The numbers on the axis are in
meters.
A parameter sensitivity test was carried out to assess the effects of
some important input parameters on the result. The chosen roughness
(F 16 data set) affected the shear stress at the bed, which affected the
sediment transport capacity. The parameter test showed that it is fairly
important to use a correct roughness in the simulation.
The angle of repose for the sediments was changed from 40 degrees to
35 degrees (F 56 data set), only resulted in marginal changes. The for-
mula for sediment concentration close to the bed was changed (F 6 data
set), reducing the concentration by 33 %. The sediment deposits then
became greater in magnitude and did not reach as far out in the sand
trap as for the original simulation. This result is a logical consequence of
the decrease in transport capacity. However, the changes are not very
great, considering that the transport capacity was decreased by 33 %.
32. CFD Modelling for Hydraulic Structures 32
The numerical model was therefore not very sensitive to the accuracy of
the formula for concentration at the bed for this case.
Other parameter tests show very little effect of changing the number of
inner iterations from 5 to 10 (F 33 data set) for each time step for the Na-
vier-Stokes equations, and changing the time step 10 to 5 seconds (F
33 data set).
The deviation between the calculated and measured bed level changes
can be caused by several phenomena. Fig. 5.4.3 shows that the meas-
ured bed levels are lower than the calculated levels at the downstream
part of the deposit. This may be caused by false diffusion as there is a
recirculation zone in this region. Another explanation may be the shape
of the deposit in this area. Looking at Fig. 5.4.1, after 88 minutes, the
deposit is much greater at the centerline of the sand trap than at the
sides. This effect is also observed in the physical model, but not to the
same extent as in the calculated results.
5.5 Example 3. Complex geometries - Himalaya intake
An intake construction sometimes has a very complex geometry. One of
the most complex flow cases modelled with SSIIM was the Himalayan
Intake. The intake was designed by Prof. H. Støle as a mean of decreas-
ing sediment problems for run-of-the-river hydropower plants taking
water from steep rivers. The geometry of the CFD model was made by
H. Kjellesvig (1995). (Kjellesvig and Støle, 1996).
The intake has gates both close to the surface and close to the bed. This
allows both floating debris and bedload to pass the intake dam. The dam
itself has a large intake in form of a tube, parallel to the dam axis. The
longitudinal profiles shown in Fig. 5.5.1 are cross-sections of the dam
and the tube. At the exit of this tube, the water goes to the hydropower
plant. Most of the water enters at the upper part of the tube, where the
sediment concentration is lowest. There is also a opening at the bottom
of the tube, allowing deposited sediments to fall down into the river and
be flushed out.
The tube and other parts of the intake are modelled partly as outblocked
regions and partly with walls between cells. The outblocked regions are
modelled with G 13 data sets, and the walls are modelled with W 4 data
sets. These data sets and the corresponding location is given in Text
Box 5.5.1 and Fig. 5.5.1.
Each wall that has water on both sides must be modelled with two W4
data sets. One data set will only provide wall functions for one side of
the wall. In Text Box 5.5.1, the two corresponding W 4 data sets are
placed after each other. As an example the data sets for wall g is given
below:
W 4 1 -1 7 2 16 8 11 wall g
W 4 1 1 8 2 16 8 11 wall g
33. 33 5. Intakes
Figure 5.5.1 Longitudinal b
profile of the Himalayan In- f
g
take, close to the side facing
away from the power plant d
(j=2). The top figure shows a h
the walls, inlet and outlet ar-
eas, where letters are also i
e
given to be described in c
more detail in the text. The
middle figure shows the
grid, with indexes for some
of the cells. The lower figure
shows a velocity vector plot.
Cells i=2 Cells i=6 Cells i=21
G 7 0 1 2 16 2 14 0 0 126.75 1 0 0 inlet a
Text box 5.5.1 Data sets in
the control file, defining in- G 7 1 -1 2 4 2 3 0 0 25.85 1 0 0 outlet b
flow, outflow, outblocked re-
gion, internal and external G 7 1 -1 2 16 12 14 0 0 37.20 1 0 0 outlet c
walls. The letters corre-
spond to Fig. 5.5.1. Note G 13 3 20 21 2 16 8 11 outblocked region d
only the data sets relevant
for Fig. 5.5.1 are shown. W 4 1 -1 21 2 16 4 7 wall e
W 4 1 -1 7 2 16 8 11 wall g
W 4 1 1 8 2 16 8 11 wall g
W 4 3 -1 7 8 16 2 16 wall h
W 4 3 1 8 8 16 2 16 wall h
W 4 3 1 12 13 19 2 16 wall f
W 4 3 -1 11 13 19 2 16 wall f
W 4 3 1 8 18 19 2 16 wall i
W 4 3 -1 7 18 19 2 16 wall i
34. CFD Modelling for Hydraulic Structures 34
The first integer (1) says that the wall is in a cross-section. On the first
line, the wall is in cell i=7 (third integer). The second integer, -1, says that
the wall is in the direction of an increasing cell number. That is, in the
direction of cell i=8.
On the second line, the wall function is in cell i=8, but the wall is now in
the direction of decreasing i numbers (cell i=7).
The last four integers indicate the location of the wall in the cross-sec-
tion. The fourth and fifth integer indicates the cell numbers in the trans-
verse direction, from j=2 to j=16. Note that the cross-section shown in
Fig. 5.5.1 is in layer j=2. The last two integers says the wall is in the ver-
tical layer from k=8 to k=11. This can be verified by counting the cells in
Fig. 5.5.1, noting the bed cell has cell number k=2.
Vertical distribution of grid cells
Looking at the grid in Fig. 5.5.1, the vertical distribution of grid cells var-
ies greatly. The variation is given on the G 16 data sets. The distribution
is made so that the location of the intake geometry should be as close
to the prototype as possible. Also, the grid qualities, expansion ratio, or-
thogonality etc. can be controlled by the G 16 data sets.
Text Box 5.5.2 gives all the G 16 data sets for the grid in Fig. 5.5.1.
G16 1 2 1 16 0.0 7.69 15.38 23.08 30.77 38.46 46.15 53.85 61.54 69.23 76.92 84.62 92.31 100.0
G16 3 3 1 16 0.0 10.01 20.01 30.02 40.02 50.03 60.03 64.97 69.90 74.83 79.77 86.51 93.26 100.0
G16 4 7 1 16 0.0 12.32 24.64 36.96 49.28 61.59 73.91 76.09 78.26 80.43 82.61 88.41 94.20 100.0
G16 8 8 1 16 0.0 9.42 18.84 28.26 37.68 47.10 58.66 65.15 71.64 78.12 84.61 89.74 94.87 100.0
G16 9 9 1 16 0.0 8.33 16.67 25.00 33.33 41.67 50.00 58.06 67.13 76.50 86.26 90.85 95.44 100.0
G16 10 10 1 16 0.0 7.25 14.49 21.74 28.99 36.23 43.48 50.52 63.56 76.00 87.65 91.76 95.86 100.0
G16 11 11 1 16 0.0 6.38 12.75 19.13 25.51 31.88 38.26 43.48 60.87 75.00 89.13 92.75 96.38 100.0
G16 12 12 1 16 0.0 6.09 12.17 18.26 24.35 30.43 37.33 43.48 60.87 75.17 89.48 92.98 96.48 100.0
G16 13 13 1 16 0.0 5.65 11.30 16.96 22.61 28.26 33.91 43.48 60.87 76.09 91.30 94.20 97.10 100.0
G16 14 14 1 16 0.0 5.51 11.01 16.52 22.03 27.54 33.04 43.48 60.87 76.09 91.30 94.20 97.10 100.0
G16 15 18 1 16 0.0 5.36 10.72 16.09 21.45 26.81 32.17 43.48 60.87 76.09 91.30 94.20 97.10 100.0
G16 19 19 1 16 0.0 5.01 10.01 15.02 20.03 25.04 30.04 41.70 59.64 75.34 91.03 94.02 97.01 100.0
G16 20 20 1 16 0.0 4.55 9.09 13.64 18.18 22.73 28.18 40.91 59.09 75.00 90.91 93.94 96.97 100.0
G16 21 21 1 16 0.0 4.55 9.09 13.64 18.18 22.73 27.27 40.91 59.09 75.00 90.91 93.94 96.97 100.0
Text box 5.5.2 G 16 data sets in the control file, defining the grid.
35. 35 Literature
Literature
Ackers, P. and White, R. W. (1973) "Sediment Transport: New
Approach and Analysis", ASCE Journal of Hydraulic Engineering, Vol.
99, No. HY11.
Atkinson, E. (1995) “A Numerical Model for Predicting Sediment Exclu-
sion at Intakes”, HR Wallingford, UK, Report OD130.
Blench, T. (1970) “Regime theory design of canals with sand beds”,
ASCE Journal of Irrigation and Drainage Engineering, Vol. 96, No. IR2,
Proc. Paper 7381, June, pp. 205-213.
Brooks, H. N. (1963), discussion of "Boundary Shear Stresses in
Curved Trapezoidal Channels", by A. T. Ippen and P. A. Drinker, ASCE
Journal of Hydraulic Engineering, Vol. 89, No. HY3.
Demny, G., Rettemeier, K., Forkel, C. and Köngeter, J. (1998) “3D-
numerical investigation for the intake of a river run power plant”,
Hydroinformatics - 98, Copenhagen, Denmark.
Einstein, H. A., Anderson, A. G. and Johnson, J. W. (1940) “A distinc-
tion between bed-load and suspended load in natural streams”, Trans-
actions of the American Geophysical Union’s annual meeting, pp. 628-
633.
Einstein, H. A. and Ning Chien (1955) "Effects of heavy sediment con-
centration near the bed on velocity and sediment distribution", UCLA -
Berkeley, Institute of Engineering Research.
Engelund, F. (1953) "On the Laminar and Turbulent Flows of Ground
Water through Homogeneous Sand", Transactions of the Danish Acad-
emy of Technical Sciences, No. 3.
Engelund, F. and Hansen, E. (1967) "A monograph on sediment trans-
port in alluvial streams", Teknisk Forlag, Copenhagen, Denmark.
Fisher-Antze, T., Stoesser, T., Bates, P. and Olsen, N. R. B. (2001) “3D
Numerical Modelling of Open-Channel Flow with Submerged Vegeta-
tion”, IAHR Journal of Hydraulic Research, Vol. 39, No. 3.
Kjellesvig, H. M. and Stoele, H. (1996) "Physical and numerical model-
ling of the Himalayan Intake", 2nd. International Conference on Model-
ling, Testing and Monitoring for Hydro Powerplants", Lausanne,
Switzerland.
Kjellesvig, H. M. (1996) “Numerical modelling of flow over a spillway”,
Hydroinformatics 96, Zürich, Switzerland.
Krüger, S., Bürgisser, M., Rutschmann, P. (1998). Advances in calcu-
lating supercritical flows in spillway contractions. Hydroinformatics' 98,
163-170.
Krüger, S. and Olsen, N. R. B. (2001) "Shock-wave computations in
channel contractions", XXIX IAHR Congress, Beijing, China.
Lysne, D. K. (1969) “Movement of sand in tunnels”, ASCE Journal of
Hydraulic Engineering, Vol. 95, No. 6, November.
Mayer-Peter, E. and Mueller, R. (1948) "Formulas for bed load trans-
port", Report on Second Meeting of International Association for
Hydraulic Research, Stockholm, Sweden.
Melaaen, M. C. (1992) "Calculation of fluid flows with staggered and
36. CFD Modelling for Hydraulic Structures 36
nonstaggered curvilinear nonorthogonal grids - the theory", Numerical
Heat Transfer, Part B, vol. 21, pp 1- 19.
Olsen, N. R. B. and Melaaen, M. C. (1993) "Three-dimensional numeri-
cal modelling of scour around cylinders", ASCE Journal of Hydraulic
Engineering, Vol. 119, No. 9, September.
Olsen, N. R. B., Jimenez, O., Lovoll, A. and Abrahamsen, L. (1994)
"Calculation of water and sediment flow in hydropower reservoirs", 1st.
International Conference on Modelling, Testing and Monitoring of
Hydropower Plants, Hungary.
Olsen, N. R. B. and Skoglund, M. (1994) "Three-dimensional numerical
modelling of water and sediment flow in a sand trap", IAHR Journal of
Hydraulic Research, No. 6.
Olsen, N. R. B. and Stokseth, S. (1995) "Three-dimensional numerical
modelling of water flow in a river with large bed roughness", IAHR Jour-
nal of Hydraulic Research, Vol. 33, No. 4.
Olsen, N. R. B. (1996) “Three-dimensional numerical modelling of local
scour”, Hydroinformatics 96, Zürich, Switzerland.
Olsen, N. R. B. (1999) "Computational Fluid Dynamics in Hydraulic and
Sedimentation Engineering", Class notes, Department of Hydraulic and
Environmental Engineering, The Norwegian University of Science and
Technology.
Olsen, N. R. B. and Kjellesvig, H. M. (1998a) "Three-dimensional
numerical flow modelling for estimation of maximum local scour depth",
IAHR Journal of Hydraulic Research, No. 4.
Olsen, N. R. B. and Kjellesvig, H. M. (1998b) "Three-dimensional
numerical flow modelling for estimation of spillway capacity", IAHR
Journal of Hydraulic Research, No. 5.
Olsen, N. R. B. (1999) "Two-dimensional numerical modelling of flush-
ing processes in water reservoirs", IAHR Journal of Hydraulic
Research, Vol. 1.
Olsen, N. R. B. and Kjellesvig, H. M. (1999) "Three-dimensional numer-
ical modelling of bed changes in a sand trap", IAHR Journal of Hydraulic
Research, Vol. 37, No. 2. abstract
Pasche, E. (1984) “Turbulence Mechanism in Natural Streams and the
Possibility of its Mathematical Representation” (in German) Mitteilungen
Institut für Wasserbau und Wasserwirtschaft No. 52. RWTH Aachen,
Germany.
Patankar, S. V. (1980) "Numerical Heat Transfer and Fluid Flow", Mc-
Graw-Hill Book Company, New York.
Reinauer, R. (1995). Kanalkontraktionen bei schiessendem Abfluss und
Stosswellenreduktion mit Diffraktoren. PhD thesis, ETH Zurich, Diss No.
11320.
Rhie, C.-M, and Chow, W. L. (1983) "Numerical study of the turbulent
flow past an airfoil with trailing edge separation", AIAA Journal, Vol. 21,
No. 11.
Richardson, J. E. (1997) "Control of Hydraulic Jump by Abrupt Drop,"
XXVII IAHR Congress, San Francisco, USA.
37. 37 Literature
Richardson, J. E. and Panchang, V. G. (1998) “Three-dimensional sim-
ulation of scour-inducing flow at bridge piers“, ASCE Journal of Hydrau-
lic Engineering, Vol. 124, No. 5, May, page 530-540.
van Rijn, L. C. (1982) “Equivalent roughness of alluvial bed”, ASCE
Journal of Hydraulic Engineering, Vol. 108, No. 10.
van Rijn, L. C. (1987) "Mathematical modelling of morphological proc-
esses in the case of suspended sediment transport", Ph.D Thesis, Delft
University of Technology.
Roulund, A. (2000) “Three-dimensional numerical modelling of flow
around a bottom mounted pile and its application to scour”, PhD Thesis,
Department of Hydrodynamics and Water Resources, Technical Univer-
sity of Denmark.
Schlichting, H. (1979) "Boundary layer theory", McGraw-Hill.
Shields, A. (1936) “Use of dimensional analysis and turbulence
research for sediment transport”, Preussen Research Laboratory for
Water and Marine Constructions, publication no. 26, Berlin (in Ger-
man).
Seed, D. (1997) "River training and channel protection - Validation of a
3D numerical model", Report SR 480, HR Wallingford, UK.
Spaliviero, F. and May, R. (1998) “Numerical modelling of 3D flow in hy-
draulic structures”, IAHR/SHSG seminar on Hydraulic Engineering,
Glasgow, UK.
Sæterbø, E., Syvertsen, L. and Tesaker, E. (1998) “Handbook of rivers”
(Vassdragshåndboka), Tapir forlag, Norway, ISBN 82-519-1290-3 (in
Norwegian).
US Bureau of Reclamation (1973) “Design of small dams”, Washington,
USA.
Vanoni, V., et al (1975) "Sedimentation Engineering", ASCE Manuals
and reports on engineering practice - No54.
Yang, T. C. (1973) "Incipient Motion and Sediment Transport", ASCE
Journal of Hydraulic Engineering, Vol. 99, No HY10.
Yang, J. and Johansson, N. (1998) “Determination of Spillway Dis-
charge Capacity - CFD Modelling and Experiment Verification”, 3rd Int.
Conf. on Advances in Hydroscience and -Engineering, Cottbus, Ger-
many.
Wright, N. (2001) “Conveyance implications for 2-D and 3-D model-
ling”, EPSRC Network on Conveyance in River/Floodplain Systems,
UK, http://ncrfs.civil.gla.ac.uk/workplan.htm.