Practical Capacities and Challenges of 3D CFD Modelling: Feedback Experience in Engineering Projects

In recent years, hazard classifications for many existing embankment dams have changed because hydrologic conditions have been altered. Consequently, many of these dams no longer provide adequate spillway capacity according to state and federal dam safety regulations. Stepped spillways are a popular choice for providing increased spillway capacities to existing embankment dams. Stepped spillways in these applications are typically placed over the existing embankment or auxiliary spillway; thereby, the chute slope is the same as the downstream embankment face or auxiliary spillway slope. Design guidelines and literature in general for these stepped spillways are very limited, so further research on these stepped spillways is warranted. A two-dimensional, physical model was constructed to evaluate the inception point, velocities, and energy dissipation in a 4(H):1(V) spillway chute having 38 mm (1.5 in.) high steps. Model unit discharges ranging from 0.11 m3 s-1 m-1 (1.2 cfs ft-1) to 0.82 m3 s-1 m-1 (8.8 cfs ft-1) were tested. Water surfaces, bed surfaces, and velocities were collected during the tests. An inception point relationship provided by H. Chanson may be used to determine the inception point for slopes as flat as 4(H):1(V) when F* ranges between 10 and 100. The velocity profiles transition from uniform at the crest to approaching a one-sixth power law distribution at the inception point for all tested flows. Energy losses increase in a linear fashion from near zero at the crest to approximately 30% near the inception point for all tested flows. This research will assist engineers with the design of stepped spillways applied on relatively flat embankment dams.

This study involves a computational analysis of the flow of iron ore (with a mean particle diameter of 55 microns) slurry flow through a pipeline of 2-inch diameter. The analysis covers a flow velocity range of 1 to 3.5 m/s at high concentrations ranging from 60 to 72 % w/w without and with additive (Sodium-hexametaphosphate) at different dosages from 0.1 to 1.5% w/w of solids. A three-dimensional Computational Fluid Dynamics (3D CFD) model is developed and validated using experimental data collected in our previous studies (Part I). The Eulerian multiphase model is used with K-epsilon turbulence settings to simulate the flow. Based on comparison with experimental data, it is observed that the 3D CFD model can predict the pressure drop with an error band of ±30%. However, the 3D CFD model predicts the pressure drop very accurately for the velocity range of 2 to 3m/s for the entire range of solids concentrations considered in the present study. Distributions of concentration, velocity, and slip velocity along with granular pressure and granular viscosity are also computed and presented using the 3D CFD model developed in the present study for the entire ranges of solids concentration and flow velocity covered in the present study.

In a stepped spillway, the spillway face is provided with a series of steps from near the crest to the toe. This type of spillway is attractive for water resource management because it dissipates energy that is within the flow. When energy is dissipated by the steps, it reduces the energy reduction that otherwise must occur at the toe of the spillway. In the present study, nine physical stepped spillway models with slopes of 15, 25, 45° and with a number of steps varying from 5 to 50 have been constructed. The crest of the spillways have standard Ogee profiles and a constant slope extends from the crest to the spillway toe. Two different-sized flumes were used in the experiments (25 and 50 cm wide). Several physical models have been constructed in order to measure the energy dissipation. The purpose of this research is to investigate energy dissipation rate in spillways with different slopes. It was found that the two most important dimensionless factors influencing energy dissipation are the Froude number and q2/(gHdam3). The spillway slope and step number have less impact on the rate of energy dissipation. However, for a constant discharge over a stepped spillway, increasing spillway slope, and increasing the number of steps will increase the energy dissipation. Using multiple regressions, some useful relations for determination energy dissipating are obtained.

Physical and numerical models can be employed together to better understand the SAGD process. When there are inadequate measured physical properties as input data, and insufficient built-in mathematical processes, it is beneficial to validate a numerical model by history matching physical model results. Based on Chung's experimental data from a physicalmodel, a two-dimensional, three-phase (bitumen, water and steam) and two-component (water & bitumen) black oil numerical model for the SAGD process was developed and tested, using the CMG "STARS" thermal simulator. Both the "Spreading Steam Chamber" and the "Rising Steam Chamber" processes were simulated. The numerical results agree reasonably well with the measured data for cumulative oil production, recovery percentage and temperature profiles in the model at different times. The results from the numerical simulation were found to be relatively insensitive to the form of the relative permeability functions employed. Linear relative permeability curves were employed to simulate the experimental production results. The model can be employed as a tool to aid in sensitivity studies and in the interpretation of the laboratory results. Introduction A reservoir can only be produced once, and it is often difficult and costly to correct mistakes in the approach, once the project has been initiated. However, a model study can be conducted several times and alternatives examined. Therefore, before launching any new technology, it is advisable to conduct extensive testing, using physical and numerical models, to evaluate various scenarios for the optimization of pattern type, size, well completions and rates. It is important, technically and financially, to have a prediction of the performance that is likely to be achieved from a given proposed recovery scheme for a given reservoir. Models, both physical and numerical, improve the understanding of reservoir processes. A physical model can provide a means for checking the accuracy and assumptions of the numerical model. Validating the numerical model is advisable by history matching with a well-defined physical model to overcome some of the limitations of numerical modelling. In some recovery processes, such as the Steam-assisted Gravity Drainage (SAGD) process(l), the physics of the situation is so complex that both physical and numerical models should be employed as complementary tools to obtain insight into different mechanisms and determine which strategies will optimize the process. Several field pilots using the SAGD process are underway. These include the three Esso Resources horizontal pilot projects at Cold Lake, Alberta, the Sceptre Resources pilot project in the Tangleflags North field near Lloydminster, Saskatchewan, and the Alberta Oil Sands Technology and Research Authority's (AOSTRA) Underground Test Facility (UTF) project in the Athabasca field near Fort McMurray, Alberta. Scope and Objective of Study The scope of this numerical simulation study is to match the oil recovery and the steam temperature interface position as observed in Chung's physical model(2,6) of the SAGD process. The objective is to investigate the feasibility, using a commercial CMG simulator ("STARS"), to history match the SAGD process, particularly the "spreading steam chamber" phase and the "rising steam chamber" phase.

A revision of the hydrology study for the Galaube dam catchment area has been made as part of the dam safety report. This study highlights an increase of the design flood. A 3D CFD free surface hydraulics model of the Galaube dam is built with ANSYS CFX in order to verify the discharge capacity for the new design flood. The 3D CFD model is validated on two elements: the results of the initial studies and the model’s sensitivity on the roughness coefficients and the mesh size. The different attempts on the model’s sensitivity do not lead to important modifications of the spillway rating curve. Moreover, it appears that, in a range of flow, the rating curve calculated with the 3D computational model is smaller than the rating curve obtained in the initial studies. This reduction is due to a drowning of the weir. This phenomenon is highlighted with the 3D CFD model. Furthermore, the model permits the identification of the limit flow rate before the saturation of the spillway and the representation of the transition in flow characteristics. Otherwise, the results are used to define some works to insure the dam safety: the dam freeboard and the wave wall safety are checked, and the height of the spillway training wall is increased.

Susu Dam is an RCC Dam under construction in Peninsula Malaysia. The dam design has been carried out by SMEC International in its offices in Kuala Lumpur. The dam is some 90 m tall. The overfall stepped spillway provides for discharges to enter a hydraulic jump stilling basin for energy dissipation, prior to passing through a culvert under a roadway. The spillway was subject to hydraulic model testing at the Utah Water Research Laboratory of Utah State University (USA), at a length scale of 1:30. The hydraulic performance of the spillway design was evaluated up to the 4,700 m3/s Probable Maximum Flood (PMF) discharge (almost 1,000 L/s model scale). The spillway was tested through a number of configurations, prior to the development of the final arrangement. The purpose of the paper is to describe the modelling detail, and then focus on the flow behaviour at the stair-stepped spillway chute with 2.4 m high steps, in particular the piezometric pressures and transient pressures on the tread and riser of the steps. The spillway chute converged from 100 m at the crest to 78 m at the bottom at entrance to the stilling basin; the unit discharge entering the stilling basin was 60 m2/s for the PMF. Of particular interest was the occurrence of negative pressures on the steps, and the paper will describe the transients for several discharges from the AEP 1 in 1,000 up to the PMF; the results indicating very low pressures into a cavitation region. The design provides for an aerator across the spillway in order to counter the effects of possible cavitation. Results will be presented with and without the aerator operating. The paper provides useful design information for the hydraulic design of stepped spillways.

Validation of a 3D multiphase-multicomponent CFD model for accidental liquid and gaseous hydrogen releases

Flip buckets are usually placed at the end of chute spillways or outlet conduits of high dams. The projected jets from these buckets jump into plunge pools to dissipate the destructive energy of plunging jets. Compound flip bucket is a special bucket with a non-zero degree of cross sectional slope, followed by a curvature in its longitudinal direction. These structures are very suitable for restricted geometries and conditional topographies. Although, studies of flow over flip buckets turn back to many years and there are valuable information for simple buckets, but still there are uncertainties regarding the flow over compound flip buckets. Therefore, in this study the hydraulic parameters such as pressure, velocity and depth of flow on these structures are presented. Two chute spillways with their flip buckets were used to check the pressure distribution and its variation with different hydraulic characteristic along the compound flip buckets. The scaled model of Gotvand dam (south of Iran) at Water Research Institute of Iran was used for this study. The Froude number varies in the range of 3.5 to 7.5. The results are presented in the form of dimensionless equation, which was calibrated based on the present results. The location of maximum dynamic pressure and pattern of flow obstruction on the bed of the compound flip bucket is also presented.

This paper covers a technical comparison of results from two physical scale models and a computational fluid dynamics (CFD) model that was undertaken for a masonry stepped spillway at Grassholme reservoir. The spillway is known to be deficient in terms of the depth of the collection channel and the size of the stepped spillway. Using the performance of the spillway during a flood in 1991 as a benchmark, the two modelling methods are compared. It is found that the physical scale models cannot match the full scale behaviour of spillway. The CFD model was undertaken in a way so as to account for air entrainment and flow bulking, and this formed a better estimate of the chute behaviour witnessed during 1991.

Some very impressive modeling work lies behind this paper, but leaves the reader with many questions which Dr. Mahkamov may care to answer.We are told that the Biomass Stirling Engine “has been experimentally tested” and that “experimental data, which is available (where?) of the mechanical brake power output and speed, indicates that”—these “are reasonably close to those predicted by using the 3D CFD model.” Could the author provide the figures so that we can judge for ourselves how close is “reasonably close?”One observes that, with fluid flow losses included, the γ engine achieves an inferior predicted performance to the modified α, whether using the second-order model, or the 3D CFD model. Mode2nd order3D CFDγ250W737W4Hz3.33Hzα2.970W3.870W7Hz5Hz So the α mode power output would seem to be 6.8 times better than the γ using the results of the second-order model, or 3.5 times better using the 3D CFD model. However, this is misleading; we are not comparing like with like.The γ engine had apparently a simple geometrical error which caused much loss of power owing to the crown of the power piston restricting flow as it approached TDC. Furthermore the γ engine model is run with a 31% porosity regenerator, while the α engine is run with an improved 40% regenerator. Did the author run his two programs for the γ engine with the crown restriction removed and with 40% porosity? Might he be able to provide the figures so that α and γ modes can be compared on a level playing field?On the basis that the pressures, P7, P5, and P1 in Fig. 13 correspond, respectively, to pressures P2 (of P2V2), P3 (of P3V3) and P1 of (P1V1) of Fig. 14, then one can make an estimate for how much the gas entrapment is costing the γ engine in power output. From Fig. 13 it can be seen that for some 25° either side of the 180° crank angle (where the power piston crown is near TDC) there is a significant difference in pressure between P7 and P5 and the gas is restricted in its passage from compression space 5 to compression space 6 (see Fig. 1).Transferring this pressure difference so as to modify the anticlockwise compression loop P2V2 one is able to deduce that the γ engine output will be raised at least from 0.737to1.77kW. Does Dr. Mahkamov feel this is a reasonable estimate?As a measure of goodness (proportional to the Beale number) one can compare the outputs in the modified Fig. 14 (γ mode) and Fig. 17 (α mode) on the basis of kW∕(Chargepressure×rpm×powersweptvolume). In changing from γ mode to α mode the power swept volume changes. In γ mode this volume is provided by the power piston (4 in Fig. 1) together with a small contribution from the unbalanced stanchion of the displacer piston. In α mode it is the larger expansion piston that becomes effectively the power piston. Using the outputs from the 3D CFD model and a nominal charge pressure of 15bar then one obtains kWrpmPower sweptvolume (liter)kWBar×liters×rpmγ1.772002.460.24×10−3α3.873003.420.25×10−3 It would seem that using the “measure of goodness factor” in the right-hand column, not very much has been gained in moving from γ to α mode. Indeed, much has been lost in terms of the mechanical soundness of the engine crank shaft gas loads, etc.The value of 0.25×10−3kWbar−1liter−1rpm−1 seems to be low. Might the 3D CFD model be significantly underestimating the performance?

Modelling and parametric study of an efficient Alpha type Stirling engine performance based on 3D CFD analysis

The present paper deals with the redesign of cyclic variation of a set of fan outlet guide vanes by means of high-fidelity full-annulus CFD. The necessity for the aerodynamic redesign originated from a change to the original project requirement, when the customer requested an increase in specific thrust above the original engine specification. The main objectives of this paper are: 1) make use of 3D CFD simulations to accurately model the flow field and identify high-loss regions; 2) elaborate an effective optimisation strategy using engineering judgement in order to define realistic objectives, constraints and design variables; 3) emphasise the importance of parametric geometry modelling and meshing for automatic design optimisation of complex turbomachinery configurations; 4) illustrate that the combination of advanced optimisation algorithms and aerodynamic expertise can lead to successful optimisations of complex turbomachinery components within practical time and costs constrains. The current design optimisation exercise was carried out using an in-house set of software tools to mesh, resolve, analyse and optimise turbomachinery components by means of Reynolds-averaged Navier-Stokes simulations. The original configuration was analysed using the 3D CFD model and thereafter assessed against experimental data and flow visualisations. The main objective of this phase was to acquire a deep insight of the aerodynamics and the loss mechanisms. This was important to appropriately limit the design scope and to drive the optimisation in the desirable direction with a limited number of design variables. A mesh sensitivity study was performed in order to minimise computational costs. Partially converged CFD solutions with restart and response surface models were used to speed up the optimisation loop. Finally, the single-point optimised circumferential stagger pattern was manually adjusted to increase the robustness of the design at other flight operating conditions. Overall, the optimisation resulted in a major loss reduction and increased operating range. Most important, it provided the project with an alternative and improved design within the time schedule requested and demonstrated that CFD tools can be used effectively not only for the analysis but also to provide new design solutions as a matter of routine even for very complex geometry configurations.

<div class="section abstract"><div class="htmlview paragraph">A promising strategy for increasing thermal efficiency and decreasing emissions of a spark ignited (SI) internal combustion engine is the application of lean mixtures. The flammability limit of lean mixtures can be increased by using an active pre-chamber containing an injector and a spark plug, resulting in a combustion mode commonly called Turbulent Jet Ignition (TJI). The optimization of the combustion chamber shape and operating parameters for TJI combustion can be a demanding task, since the number of design parameters is significantly increased and is today supported by numerical simulations. In this paper, the process of the development of a numerical framework based on 3D CFD and 1D/0D numerical models that will support the research of the pre-chamber design and optimization of operating parameters will be shown. For 3D CFD modelling the AVL Fire™ code is employed, where the full combustion chamber model with intake and exhaust ports of the experimental engine is prepared. In the pre-experimental phase the 3D CFD model, although not validated due to the absence of experimental results, is used to evaluate initial boundaries in terms of combustion flammability, positions of multiple flame jets and to create verification results of in-cylinder turbulence and combustion required for 0D model tuning. Since there is no commercially available pre-chamber model with multiple jets for the 0D modelling approach, the 0D combustion model of AVL Boost™ was extended to take into account multiple flame propagation in the main chamber. The evaluation of geometry of multiple flame jets was performed with CAD software. The presented numerical framework, once validated with experimental results, can be a useful tool for the definition of optimal pre-chamber geometry and operating parameters.</div></div>

Classically, dimensional analysis is used to establish the fundamental scaling laws to predict prototype behaviour from observed model results. However, the same dimensional analysis demonstrates that it is not possible to avoid scale effects if the prototype fluid - normally water - is used in the model studies. Flow through hydraulic structures is typically gravity-driven, and equality of Froude Numbers at homologous points in model and prototype leads to the determination of fundamental scaling laws. However, it is well known that, as a consequence, homologous values of other non-dimensional parameters, such as Reynolds Number and Weber Number, are not equivalent and this leads to scale effects due to the imperfect replication in the model of (for example) friction and surface tension effects. In such cases, it is necessary to use process functions to determine the magnitude of the scale effects and to inform the adjustment of measured flow properties in the model to reflect the true value of the flow properties in the prototype. The present paper extends the use of process functions to annular flows, such as those that occur in vortex drop structures and plunge pools. Within the vortex drop, it is shown that the process functions cannot be used to evaluate scale effects directly, but, instead, must be used in conjunction with the model data to develop an analytical model for the prediction of corresponding prototype behaviour. Within the plunge pool, it is hypothesised that the momentum flux at the exit of the vortex pipe must be correctly modelled to ensure the proper replication of flow conditions within the plunge pool. The techniques are developed, and their use illustrated through application to a large physical model study of a vortex drop and plunge pool in a municipal sewerage system. It is shown, further, that ignoring the insight gained from the developed process functions would lead to large errors in the prediction of prototype flow behaviour from measured model results.

The United States Department of Agriculture (USDA), Natural Resources Conservation Service (NRCS) estimates that nearly half of the small watershed dams constructed with their assistance will reach the end of their planned service life within the next 10 years. While many of these structures were originally built in rural environments, urban communities have encroached upon these structures over the years, making it difficult to address issues such as the maintenance of adequate spillway capacity to meet current dam safety standards. The urban infrastructure surrounding these structures limits the options available for dam rehabilitation. For instance, raising the dam or widening the existing auxiliary spillway to increase spillway capacity is often not a viable option due to land right constraints or changes in topography. Therefore, one of the more popular choices for increasing spillway capacity is the use of roller compacted concrete (RCC) stepped spillways. Two advantages of RCC stepped spillways are that they can be placed over the top of the dam and that they can provide considerable energy dissipation within the spillway chute. When placing an RCC spillway over an existing embankment dam, it is often desirable for the width of hydraulic control section to be greater than the width available to outlet the flow into the downstream channel. A generalized study utilizing a three-dimensional, 1:22 scale physical model was conducted to evaluate the influence converging training walls have on the flow characteristics on a stepped spillway. Vertical walls having convergence angles ranging from 0 degrees to 70 degrees were tested in conjunction with a spillway having a chute slope of 3(H):1(V) and a prototype step height of 0.3 m (1 ft). Water surface profiles in the central area of the chute and along the walls were collected and analyzed for each spillway configuration. Results of the study are presented in a format that will assist engineers in determining the required wall height for this type of a converging chute spillway.