Volume 2: Computational Fluid Dynamics最新文献

{"title":"Curvature Estimation Modeling Using Machine Learning for CLSVOF Method: Comparison With Conventional Methods","authors":"Majid Haghshenas, Ranganathan Kumar","doi":"10.1115/ajkfluids2019-5415","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5415","url":null,"abstract":"\u0000 Despite extensive progress in recent decades, curvature estimation in two-phase models remains a challenge. Well-established curvature computing techniques such as distance function, smoothed volume fraction and height-function directly estimate the interface curvature from the implicit representation of the interface. Most recently, machine learning approach has been incorporated in computational physics simulation. Machine learning is a set of algorithms that can be utilized for training a system which allows predicting the output in the future. In this work, we train a curvature estimation model using machine learning approach for Coupled Level Set Volume of Fluid (CLSVOF) method in which both distance function and volume fraction implicitly represent the interface. Three datasets for the curvature are generated: a) curvature as a function of volume fraction (nine inputs), b) curvature as a function of distance function (nine inputs), and c) curvature as a function of both volume fraction and distance function (eighteen inputs). For each interfacial cell, curvature and input parameters (nine volume fraction and nine distance function values) at nine grid points across the interface are stored. Datasets are utilized to train different curvature computing models using neural network (NN) learning algorithm. Comparison of different datasets reveals that the distance function dataset is the best input for curvature function training. Different available learning algorithms on built-in NN toolbox in Matlab are examined. The curvature estimation function is examined for a dimensional 2D well-defined droplet on different grid resolution. In addition, the curvature estimation model by machine learning approach is compared with conventional methods such as the level set method and height function method for couple of cases. First, the case of elliptical droplet is used to evaluate curvature estimation of different methods in comparison with the analytical solution. Then, the standard case of equilibrium droplet is simulated by CLSVOF solver using different curvature estimation methods to evaluate parasitic currents generation and droplet pressure prediction. The results show that the machine learning curvature function outperforms conventional methods even on coarse grids. Finally, the curvature estimation methods are is utilized to solve a practical case of rising bubble. We observed that the terminal velocity of capped bubble reported by curvature function simulation has the lowest error.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114060253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and Validation of a Hybrid RANS-LES Approach Based on Temporal Filtering","authors":"Vladimir Duffal, Benoît de Laage de Meux, R. Manceau","doi":"10.1115/ajkfluids2019-4937","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-4937","url":null,"abstract":"\u0000 To address the challenge of controlling the energy partition in hybrid RANS-LES methods, the use of a consistent operator based on temporal filtering is desirable. This formalism leads to the development of a consistent continuous hybrid RANS-LES approach called Hybrid Temporal LES (HTLES). In this paper, an upgraded version of HTLES is presented, focusing on improving the model for wall-bounded flows. Notably, a shielding function is integrated in the model to impose the RANS behavior in the near-wall regions. The calibration and validation of the hybrid method applied to the standard k-ω-SST model is then carried out on several test cases: decaying isotropic turbulence, channel flow and periodic-hill flow. The new version of the model fulfills the specifications: the correct subfilter dissipation; the correct migration from RANS to LES in the boundary layer; the robustness of the results to grid coarsening; the accuracy of the predictions at a reasonable computational cost.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121358246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Numerical Simulations of Turbulent Flow Through a 90° Elbow","authors":"R. Venters, B. Helenbrook, G. Ahmadi","doi":"10.1115/ajkfluids2019-5498","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5498","url":null,"abstract":"\u0000 Turbulent flow in an elbow has been numerically investigated. The flow was modeled using two approaches; Reynolds Averaged Navier-Stokes (RANS) and Direct Numerical Simulation (DNS) methods. The DNS allows for all the scales of turbulence to be evaluated, providing a detailed depiction of the flow. The RANS simulation, which is typically used in industry, evaluates time-averaged components of the flow. The numerical results are accompanied by experimental data, which was used to validate the two methods. Profiles of the mean and root-mean-square (RMS) fluctuating components were compared at various points along the midplane of the elbow. Upstream of the elbow, the predicted mean and RMS velocities from the RANS and DNS simulations compared well with the experiment, differing slightly near the walls. However, downstream of the elbow, the RANS deviated from the experiment and DNS, showing a longer region of flow re-circulation. This caused the mean and RMS velocities to significantly differ. Examining the cross-section flow field, secondary motion was clearly present. Upstream secondary motion of the first kind was observed which is caused by anisotropy of the reynolds stresses in the turbulent flow. Downstream of the bend, the flow transitions to secondary motion of the second kind which is caused by streamline curvature. Qualitatively, the RANS and DNS showed similar results upstream of the bend, however downstream, the magnitude of the secondary motion differed significantly.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114359819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reconstruction of RANS Model and Cross-Validation of Flow Field Based on Tensor Basis Neural Network","authors":"Xu-dong Song, Zhen Zhang, Yiwei Wang, Shu-ran Ye, Chenguang Huang","doi":"10.1115/ajkfluids2019-5572","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5572","url":null,"abstract":"\u0000 The solution of the Reynolds-averaged Navier-Stokes (RANS) equation has been widely used in engineering problems. However, this model does not provide satisfactory prediction accuracy. Because the widely used eddy viscosity model assumes a linear relationship between the Reynolds stress and the average strain rate tensor and these linear models cannot capture the anisotropic characteristics of the actual flow. In this paper, two kinds of flow field structures of two-dimensional cylindrical flow and circular tube jet are calculated by using the RANS model. Secondly, in order to improve the prediction accuracy of the RANS model, the Reynolds stress of the RANS model is reconstructed by the tensor basis neural network algorithm based on nonlinear eddy viscosity model. Finally, the model trained by neural network is cross-validated, and compare the cross-test results with the traditional RANS k-eps model. The results show that the multi-layer neural network method has achieved good results in turbulence model reconstruction.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114880542","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effects of Stenosis Geometry on Flow in Arteriovenous Fistula Patients","authors":"Jeffrey Krampf, R. Agarwal, S. Shenoy","doi":"10.1115/ajkfluids2019-4689","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-4689","url":null,"abstract":"\u0000 This focus of this study is to understand the relationship between the fluid properties present in and the geometric parameters of stenoses developed in end-stage renal disease (ESRD) patients, after creation of an arteriovenous fistula (AVF). Stenosis is the leading cause of failure in AVF creation and maturation. A fistula is meant to provide an access point for hemodialysis treatment necessary for ESRD patients, but large failure rates in fistula creation and maturation cause reoccurring problems for patients and a disproportionately high amount of spending on ESRD patient care. In the United States alone, ESRD patients account for 1% of the Medicare patient population, but the Centers for Medicare & Medicaid Services spent $35.4 billion, 7.2% of the 2016 Medicare budget on their treatment (United States Renal Data System, 2018 Annual Report). This study uses CFD to simulate blood flowing through venous stenoses of varying lengths and initial flow conditions. Computational modeling allows for specific control of geometric conditions as well as simple generation of resulting properties, such as wall shear stress, that are difficult to acquire in vivo.\u0000 For this study, five different geometric models were constructed to represent straight vascular segments with varying lengths of stenosis. Each vessel was 4-millimeters in diameter with a 2-millimeter diameter stenosis. The lengths of the inlet and outlet vessel segments adjacent to the stenosis were each four times the vessel diameter. Stenosis lengths of 5, 15, 30, 45 and 60-millimeters were used. Vessels were treated as rigid tubes, and the geometries were created using PTC Creo Parametric (PTC Inc., Needham, MA), a commercially available CAD software. CFD analysis of the flow through the vessel segments was performed using ANSYS Fluent (ANSYS, Inc., Canonsburg, PA) for each geometric model with a range of boundary conditions. The working fluid was blood, treated as a Newtonian fluid for the shear rates present, with dynamic viscosity of 2.55 × 10−3 kg/m-s and density of 1060 kg/m3. To model the range of pressure experienced by vessels during the cardiac cycle, simulations were performed using a range of pressure values at the vessel inlet. The boundary condition used at the inlet was a static pressure ranging from 50 to 160 mmHg in increments of 10 mmHg for each geometric model. Outflow pressure values of 10, 15, and 20mmHg were used on the outlet boundary. As expected, flow rate through the system was found to increase linearly with inlet pressure for each geometry and outlet pressure. Flow rate decreased logarithmically as stenosis length increased for each inlet and outlet pressure. Flow rate through the system also decreased as outflow pressure increased, as it would in the presence of further downstream blockages in patients. The data collected here shows under which flow conditions different stenosis geometries can result in a failed fistula, as well as under which conditions the sten","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"123 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116380500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Numerical Investigation of Flow and Heat Transfer Characteristics for Attached and Separated Low-Pr Flows","authors":"S. Bhushan, M. Elmellouki, W. Jock, D. K. Walters, J. Lai, Y. Hassan, A. Obabko, E. Merzari","doi":"10.1115/ajkfluids2019-5273","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5273","url":null,"abstract":"\u0000 This study performs a comprehensive analysis of the effect of flow separation and reattachment, convective conditions and Pr to understand their effect on heat transfer characteristics and the predictive capability of low- and hig-fidelity turbulence models are assessed. To achieve the objective DNS is performed for plane channel flow at Reτ = 640, Pr = 0.71 and 0.025 involving mixed forced and natural convection condition, and RANS, hybrid RANS/LES, and LES calculations are performed for backward backing step with expansion ratio 1.5, Pr = 0.71 and 0.0088 and Ri = 0 and 0.338. Channel flow simulations reveal that the convective conditions affect the near-wall turbulent structures and thermal diffusion more significantly in high-Re flows that in low-Re flows. Thus, the generated DNS database provides a challenging test case for turbulence model validation. For backward facing step case, all the turbulence models predict the overall flow characteristics, and Ri = 0 case is a more challenging validation test case than Ri = 0.338, as the former involves complex turbulent diffusion, whereas the latter is dominated by large scale buoyancy driven convection. Results show that well resolved PANS and LES predictions can help in improve understanding of turbulent diffusion under complex convection and flow separation/ reattachment regimes. RANS results are also quite encouraging and indicates that they may represent a reasonable compromise between computational expense and accuracy for cases in which high resolution simulations are not feasible.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122355426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computational Evaluation of Stability in Slosh Dynamics of Tank Vehicles Using Zero Moment Point","authors":"M. Usman, M. Sajid","doi":"10.1115/ajkfluids2019-5611","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5611","url":null,"abstract":"\u0000 Sloshing characterized by inertial waves has an adverse effect on the directional dynamics and safety of partially filled tank vehicles, limiting their stability and controllability during steering, accelerating or braking maneuvers. A mathematical description of the transient fluid slosh in a horizontal cylindrical tank should consider the simultaneous lateral, vertical and roll excitations assuming potential flows and a linearized free-surface boundary condition. While the determination of vehicle stability would require coupling this model to a dynamic roll plane model of a tank vehicle resulting in a computationally expensive analysis. Considering the need for a simpler method to predict roll stability for partially filled tank vehicles, we explore the Zero Moment Point of a liquid domain as a novel solution to this challenge. Numerical investigations are carried out in a three-dimensional partially filled tanks while tracking the movement of the liquid-air interface by employing the volume of fluid method in OpenFOAM. The center of Mass and Zero Moment Point were calculated from the computational results using analytical expressions. The movement of free surface is found to be in good agreement with available literature. The center of mass of the liquid domain was traced as a practical means to quantify the slosh in the tanker. The analyses are performed for different fluid fill heights at varying speeds. The results suggest that the roll stability of tank vehicles can be efficiently analyzed using the zero moment point with significantly lower computational effort.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126026357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reynolds Averaged Navier-Stokes Simulation of Highly Turbulent, Under-Expanded Jets and Effects of Impingement at Elevated Injection Pressure","authors":"Jacob Riglin, A. Wachtor, R. Morgan, Ryan Holguin, J. Bernardin","doi":"10.1115/ajkfluids2019-5564","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5564","url":null,"abstract":"\u0000 Under-expanded jets have wide range of application from fuel injection to rocket propulsion. In the present work, a numerical model was generated to investigate the fluid mechanics behavior of under-expanded jet formation and wall interaction of a jet produced by exhausting a high pressure cylinder through a narrow tube into a low pressure cylinder. Axisymmectic, Reynolds Averaged Navier Stokes simulations were conducted employing the ANSYS FLUENT explicit, Coupled Pressure-Velocity solver to determine the stagnation pressure at the wall downstream of the orifice. Transient cases were conducted using timestep sizes of 1.0 × 10−8 s and 5.0 × 10−9 s. Various gases were investigated with Hydrogen being the primary working fluid with pressure ratios ranging from 10 to 100. This paper will focus primarily on the Hydrogen jets for pressure ratios of 10, 20, and 70. Numerical results were validated from both experimental results and higher fidelity Large Eddy Simulation results specifically analyzing the jet formation. Error between Mach disk height, Mach disk width, and Prandtl-Meyer expansion fan angles of the jet for pressure ratios of 10 and 70 were kept below 5%. The peak stagnation pressures at the center of the far wall for pressure ratios of 10, 20, and 70 were observed to be 86,843 Pa, 127,786 Pa, and 315,843 Pa, respectively. The predicted peak pressures show a linear relationship with respect to the initial pressure ratio existing between the high pressure and low pressure regions when the ratios are bounded between 10 and 70.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125531931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Direct Numerical Simulation of Radial Convection in a Cylindrical Annulus With and Without Rotation","authors":"D. Pitz, W. Wolf","doi":"10.1115/ajkfluids2019-5025","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5025","url":null,"abstract":"\u0000 In rotating systems with temperature gradients, convection may occur due to gravitational or centrifugal effects. In cases where rotation is strong enough so that the centrifugal acceleration is higher than gravity, the flow is induced by centrifugal buoyancy and gravitational effects can be neglected. The problem of flow induced by centrifugal buoyancy in a cylindrical annulus has been used as a canonical setup to investigate industrial configurations, such as buoyancy-driven flows occurring in gas turbine secondary air systems, as well as geophysical flows, such as convection in the core of planets and the global circulation of the atmosphere. Due to the constraints imposed by the Taylor-Proudman theorem, such flows are quasi-homogeneous along the axial direction, and heat transfer as well as turbulent fluctuations tend to be suppressed by the action of the Coriolis force. Previous work has demonstrated that when the annulus is bounded by parallel disks, boundary layers scaling consistently with laminar Ekman layers are formed near each of the disks, even though the flow is purely buoyancy-induced. Also, the Nusselt number measured on the outer cylindrical surface has been shown to scale with the Rayleigh number as in natural convection between horizontal plates. In the present work we use direct numerical simulation (DNS) to investigate buoyancy-induced flow in an air-filled cylindrical annulus bounded by two adiabatic parallel disks, with and without rotation around the axis. In both cases the outer cylindrical surface is at a higher temperature than the inner one, so that a radial acceleration directed outwards induces an unstable stratification. In the case with rotation, the flow is induced by the centrifugal acceleration in the radial direction, and Coriolis forces are considered. For the case without rotation, the Coriolis terms are suppressed in the calculations, whereas the radial acceleration is the same as in the rotating case. Statistics are obtained and compared in the two cases, including the time-averaged Nusselt number, mean temperature profiles, velocity and temperature fluctuations, as well as terms of the turbulent kinetic energy equation. By analysing such statistics, the extent to which rotation suppresses heat transfer and turbulent fluctuations, as well as the contribution of each term to the turbulent kinetic energy budget, can be assessed.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128523936","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Numerical Simulation of Proppant Transportation in Hydraulic Fracture Based on DDPM-KTGF Model","authors":"Yan Zhang, Xiaobing Lu, Xuhui Zhang, Peng Li","doi":"10.1115/ajkfluids2019-5613","DOIUrl":"https://doi.org/10.1115/ajkfluids2019-5613","url":null,"abstract":"\u0000 Hydraulic fracturing is an efficient way to improve the conductivity of the tight oil or gas reservoirs. Proppant transportation in hydraulic fractures need to be investigated because the proppant distribution directly affects the oil or gas production. In this paper, the dense discrete particle model (DDPM) combined with the kinetic theory of granular flow (KTGF) are used to investigate the proppant transportation in a single fracture. In this model, the effects of proppant volume fraction, proppant-water interaction, proppant-proppant collision, and proppant size distribution are considered. The proppant-proppant collision is derived from the proppant stress tensor. This model is applicable from dilute to dense particulate flows. The simulated results are similar to the experimental data from other researchers. In further study, the two-phase flow in the cross fractures will be considered for engineering application.","PeriodicalId":346736,"journal":{"name":"Volume 2: Computational Fluid Dynamics","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122364030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}