Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Turbulence modeling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,135 records · Page 63
All sourcesNASA NTRSDOE OSTI
NASA NTRS1990

Flux-difference split scheme for turbulent transport equations

This paper describes a high accuracy upwind method to solve the mean compressible flow equations closed with two-equation turbulence models. A flux-difference splitting scheme is developed to compute the flowfield without introducing oscillations into the solution. A high-order accuracy scheme is used to predict both the mean flow variables and the turbulence variables. Results comparing two different two-equation turbulence models are presented for supersonic flow over a flat plate and a compressible, free shear-layer. The non-oscillatory behavior of the upwind scheme is demonstrated on the free shear-layer.

Morrison, J. H.↗
NASA NTRS1999

Prediction of Heat and Mass Transfer in a Rotating Ribbed Coolant Passage With a 180 Degree Turn

Numerical results are presented for flow in a rotating internal passage with a 180 degree turn and ribbed walls. Reynolds numbers ranging from 5200 to 7900, and Rotation numbers of 0.0 and 0.24 were considered. The straight sections of the channel have a square cross section, with square ribs spaced one hydraulic diameter (D) apart on two opposite sides. The ribs have a height of 0.1D and are not staggered from one side to the other. The full three dimensional Reynolds Averaged Navier-Stokes equations are solved combined with the Wilcox k-omega turbulence model. By solving an additional equation for mass transfer, it is possible to isolate the effect of buoyancy in the presence of rotation. That is, heat transfer induced buoyancy effects can be eliminated as in naphthalene sublimation experiments. Heat transfer, mass transfer and flow field results are presented with favorable agreement with available experimental data. It is shown that numerically predicting the reattachment between ribs is essential to achieving an accurate prediction of heat/mass transfer. For the low Reynolds numbers considered, the standard turbulence model did not produce reattachment between ribs. By modifying the wall boundary condition on omega, the turbulent specific dissipation rate, much better agreement with the flow structure and heat/ mass transfer was achieved. It is beyond the scope of the present work to make a general recommendation on the omega wall boundary condition. However, the present results suggest that the omega boundary condition should take into account the proximity to abrupt changes in geometry.

Rigby, David L.↗
NASA NTRS1993

A numerical analysis of supersonic flow over an axisymmetric afterbody

The performance of a pressure-based algorithm and the k-epsilon type turbulence models, including the original formulation proposed by Jones and Launder (1972), model extensions for refined treatment of time scales, and model corrections developed for treating compressible flows is assessed by analyzing the near-wake region of a Mach 2.5 flow over an axisymmetric cylindrical afterbody. The mechanisms responsible for the different levels of performance of the various turbulence models for the flow are investigated. It is shown that the k-epsilon models extended to allow for better response to the mean strain rate and corrected for compressibility result in better flowfield predictions than the original k-epsilon model.

Tucker, P. K.↗
NASA NTRS2004

Study of CFD Variation on Transport Configurations from the Second Drag-Prediction Workshop

This paper describes and analyzes a series of nearly 90 CFD test cases performed as a contribution to the second Drag Prediction Workshop, held in association with the AIAA in June 2003. Two configurations are included: DLR-F6 wing-body and wing-body-nacelle-pylon. The ability of CFD to predict the drag, lift, and pitching moment from experiment-including the "delta" arising from the addition of the nacelle and pylon-is assessed. In general, at a fixed angle of attack CFD overpredicts lift, but predicts the delta C (sub L) reasonably well. At low lift levels (C (sub L) less than 0.3)), delta C (sub D) is 20-30 drag counts (30-45%) high. At the target lift coefficient of C(sub L) = 0.5, delta C (sub D) is overpredicted by between 11-16 counts. However, the primary contribution of this paper is mot so much the assessment of CFD against experiment, but rather a detailed assessment and analysis of CFD variation. The series of test cases are designed to determine the sensitivity/variability of CFD to a variety of factors, including grid, turbulence model, transition code, and viscous model. Using medium-level grids (6-11 million points) at the target lift coefficient, the maximum variation in drag due to different grids is 5-11 drag counts, due to code is 5-10 counts, due to turbulence model is 7-15 counts, due to transition is 10-11 counts, and due to viscous model is 4-5 counts. Other specific variations are described in the paper.

Christopher L Rumsey↗
NASA NTRS1989

Numerical investigation of chemically reacting flows in ramjet dump combustors

The time-dependent Navier-Stokes equations, including second-order turbulence model, are numerically integrated by using four-stage Runge-Kutta scheme to predict the steady-state supersonic flow structures in ramjet dump combustors. The formulation is derived for reacting flows with finite-rate chemistry. In the present study, it is firstly attempted to assess the accuracy of existing high-order turbulence model in supersonic flows. The comparison shows reasonable agreement between calculated and measured data in terms of velocity distributions. It is indicated that a modified constant C-mu for calculating turbulent eddy viscosity is needed in the supersonic flow regime and the adaptive meshing is preferred to capture the recirculation zone. In the reacting flow calculation, the results from a test case of hydrogen and air combustion at premixed conditon show that the rearward facing step is able to increase flow residence time and stabilize the flame in supersonic flows.

Hsieh, Kwang-Chung↗
NASA NTRS1981

Numerical Prediction Methods (Reynolds-Averaged Navier-Stokes Simulations of Transonic Separated Flows)

During the past five years, numerous pioneering archival publications have appeared that have presented computer solutions of the mass-weighted, time-averaged Navier-Stokes equations for transonic problems pertinent to the aircraft industry. These solutions have been pathfinders of developments that could evolve into a major new technological capability, namely the computational Navier-Stokes technology, for the aircraft industry. So far these simulations have demonstrated that computational techniques, and computer capabilities have advanced to the point where it is possible to solve forms of the Navier-Stokes equations for transonic research problems. At present there are two major shortcomings of the technology: limited computer speed and memory, and difficulties in turbulence modelling and in computation of complex three-dimensional geometries. These limitations and difficulties are the pacing items of the continuing developments, although the one item that will most likely turn out to be the most crucial to the progress of this technology is turbulence modelling. The objective of this presentation is to discuss the state of the art of this technology and suggest possible future areas of research. We now discuss some of the flow conditions for which the Navier-Stokes equations appear to be required. On an airfoil there are four different types of interaction of a shock wave with a boundary layer: (1) shock-boundary-layer interaction with no separation, (2) shock-induced turbulent separation with immediate reattachment (we refer to this as a shock-induced separation bubble), (3) shock-induced turbulent separation without reattachment, and (4) shock-induced separation bubble with trailing edge separation.

Mehta, Unmeel↗
NASA NTRS1994

Numerical Simulation of Supersonic Compression Corners and Hypersonic Inlet Flows Using the RPLUS2D Code

A two-dimensional computational code, PRLUS2D, which was developed for the reactive propulsive flows of ramjets and scramjets, was validated for two-dimensional shock-wave/turbulent-boundary-layer interactions. The problem of compression corners at supersonic speeds was solved using the RPLUS2D code. To validate the RPLUS2D code for hypersonic speeds, it was applied to a realistic hypersonic inlet geometry. Both the Baldwin-Lomax and the Chien two-equation turbulence models were used. Computational results showed that the RPLUS2D code compared very well with experimentally obtained data for supersonic compression corner flows, except in the case of large separated flows resulting from the interactions between the shock wave and turbulent boundary layer. The computational results compared well with the experiment results in a hypersonic NASA P8 inlet case, with the Chien two-equation turbulence model performing better than the Baldwin-Lomax model.

Kapoor, Kamlesh↗
NASA NTRS2013

Wind-US Code Contributions to the First AIAA Shock Boundary Layer Interaction Prediction Workshop

This report discusses the computations of a set of shock wave/turbulent boundary layer interaction (SWTBLI) test cases using the Wind-US code, as part of the 2010 American Institute of Aeronautics and Astronautics (AIAA) shock/boundary layer interaction workshop. The experiments involve supersonic flows in wind tunnels with a shock generator that directs an oblique shock wave toward the boundary layer along one of the walls of the wind tunnel. The Wind-US calculations utilized structured grid computations performed in Reynolds-averaged Navier-Stokes mode. Four turbulence models were investigated: the Spalart-Allmaras one-equation model, the Menter Baseline and Shear Stress Transport k-omega two-equation models, and an explicit algebraic stress k-omega formulation. Effects of grid resolution and upwinding scheme were also considered. The results from the CFD calculations are compared to particle image velocimetry (PIV) data from the experiments. As expected, turbulence model effects dominated the accuracy of the solutions with upwinding scheme selection indicating minimal effects.

Georgiadis, Nicholas J.↗
DOE OSTI2024

ECAR-7932 Rev 0 Large Eddy Simulation of MARVEL Reactor Core Subchannel to Evaluate Model Uncertainty of Reynolds-Averaged Navier-Stokes Equation Based Computational Fluid Dynamics Analysis

In the previous work (ECAR-7210), the peak cladding temperature of the MARVEL microreactor has been evaluated by steady-state Reynolds-Averaged Navier-Stokes (RANS) based computational fluid dynamics (CFD) simulations. Although numerical uncertainties of RANS-based CFD simulations has been assessed in ECAR-7210, the model uncertainty of RANS turbulence models must be investigated to resolve the issues related to inaccurate prediction of turbulent heat flux and flow pulsation in a tight lattice rod bundle using the steady-state RANS simulations. Consequently, this ECAR conducted a high-fidelity CFD analysis utilizing Large Eddy Simulation (LES) to generate reference data and investigated the model uncertainty of RANS-based CFD simulations.

21 - SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLAN↗
NASA NTRS2018

PIV and Rotational Raman-Based Temperature Measurements for CFD Validation in a Single Injector Cooling Flow

NASA's Turbulent Heat Flux (THX) task under the Transformational Tools & Technologies project is focused on acquiring benchmark temperature and velocity data in turbulent shear layers for validation of computational fluid dynamics (CFD) codes. Standard CFD turbulence models lack the ability to accurately calculate a number of fundamental flow phenomena, including the turbulent transport of heat. In nearly all production class Reynolds-averaged Navier-Stokes (RANS) CFD codes, a gradient-diffusion approximation is used whereby a constant turbulent Prantdl number, Prt, is used to relate an eddy viscosity calculated for the momentum terms, to the turbulent heat flux terms. The default in most CFD codes is Prt = 0.9, however this is not representative of all flows. As discussed in Yoder et al. (2015) and Reynolds (1974), Prt = 0.7 is felt to be more appropriate for jets. Recent work, as discussed in Yoder (2016a), has explored more complex formulations including variable turbulent Prantdl number models. PARTICLE IMAGE VELOCIMETRY; TEMPERATURE MEASURING INSTRUMENTS; FILM COOLING; HIGH TEMPERATURE GASES; COMPUTATIONAL FLUID DYNAMICS; PREDICTION ANALYSIS TECHNIQUES; TURBULENCE MODELS; RAMAN SPECTRA; VELOCITY; MACH NUMBER; GAS TURBINE ENGINES; NOZZLE FLOW; THERMOCOUPLES; LASER INDUCED FLUORESCENCE; COOLING FLOWS (ASTROPHYSICS); RAMAN SPECTROSCOPY; HEAT FLUX

Wernet, Mark P.↗
NASA NTRS1978

Critical issues in viscous flow computations

In developing computer programs to numerically solve the Navier-Stokes equations, the purpose of the computation must be clearly kept in mind. In the Air Force, the purpose is to provide design information on non-linear aerodynamic phenomenon for aircraft that perform throughout the flight corridor. This translates into the requirement for a computer program which can solve the time averaged compressible Navier-Stokes equations (with a turbulence model) in three dimensions for generalized geometries. The intended application of the results then controls the priorities in addressing critical issues. Recurrent problem areas encountered in the study of viscous flow include: (1) grid generation for arbitrary geometry; (2) numerical difficulties; (3) turbulence models; (4) accuracy and efficiency; and (5) smearing of discontinuities.

Hankey, W. L., Jr.↗
NASA NTRS1985

Prediction of turbine blade heat transfer

It is planned to incorporate a number of low Reynolds number turbulence models in a general two-dimensional boundary layer calculation procedure. This will be applied to different flow conditions over turbine blades and the predictions will be compared with experimental data. The prediction activity will lead to a recommendation about a satisfactory turbulence model for turbine blade heat transfer.

Patankar, Suhas V.↗
NASA NTRS1984

A computational study of complex three-dimensional compressible turbulent flow fields

Solutions of the time dependent, Reynolds-averaged, Navier-Stokes equations are presented and are compared with a family of experimental results for the three-dimensional interaction of a shock wave with a turbulent boundary layer. The solutions correctly predict the major features of the flow field independent of the shock strength and the extent of separation when using a two-equation turbulence model with wall functions. The experimentally observed boundary between cylindrical and conical flow regimes is also predicted. However, for the flow fields with large separated zones, the details of the measured pressure distributions are not accurately computed, which indicates a need for improved turbulence modeling and/or grid resolution.

Horstman, C. C.↗
NASA NTRS1989

Boundary layer simulator improvement

Boundary Layer Integral Matrix Procedure (BLIMPJ) has been identified by the propulsion community as the rigorous boundary layer program in connection with the existing JANNAF reference programs. The improvements made to BLIMPJ and described herein have potential applications in the design of the future Orbit Transfer Vehicle engines. The turbulence model is validated to include the effects of wall roughness and a way is devised to treat multiple smooth-rough surfaces. A prediction of relaminarization regions is examined as is the combined effects of wall cooling and surface roughness on relaminarization. A turbulence model to represent the effects of constant condensed phase loading is given. A procedure is described for thrust decrement calculation in thick boundary layers by coupling the T-D Kinetics Program and BLIMPJ and a way is provided for thrust loss optimization. Potential experimental studies in rocket nozzles are identified along with the required instrumentation to provide accurate measurements in support of the presented new analytical models.

Praharaj, Sarat C.↗
NASA NTRS2002

Convective Heat Transfer in the Reusable Solid Rocket Motor of the Space Transportation System

This simulation involved a two-dimensional axisymmetric model of a full motor initial grain of the Reusable Solid Rocket Motor (RSRM) of the Space Transportation System (STS). It was conducted with CFD (computational fluid dynamics) commercial code FLUENT. This analysis was performed to: a) maintain continuity with most related previous analyses, b) serve as a non-vectored baseline for any three-dimensional vectored nozzles, c) provide a relatively simple application and checkout for various CFD solution schemes, grid sensitivity studies, turbulence modeling and heat transfer, and d) calculate nozzle convective heat transfer coefficients. The accuracy of the present results and the selection of the numerical schemes and turbulence models were based on matching the rocket ballistic predictions of mass flow rate, head end pressure, vacuum thrust and specific impulse, and measured chamber pressure drop. Matching these ballistic predictions was found to be good. This study was limited to convective heat transfer and the results compared favorably with existing theory. On the other hand, qualitative comparison with backed-out data of the ratio of the convective heat transfer coefficient to the specific heat at constant pressure was made in a relative manner. This backed-out data was devised to match nozzle erosion that was a result of heat transfer (convective, radiative and conductive), chemical (transpirating), and mechanical (shear and particle impingement forces) effects combined.

Ahmad, Rashid A.↗
NASA NTRS1990

Physical considerations governing computation of turbulent flows over bodies at large incidence

The physical aspects governing accurate numerical simulation of turbulent flows having large regions of crossflow separation are re-examined. Time-accurate, three-dimensional fine-grid Navier-Stokes solutions were obtained for turbulent subsonic and supersonic flows over a slender ogive-cylinder body of revolution at large angles of attack. These flowfields are complex and contain regions of crossflow separation and an organized leeward-side vortex structure. An algebraic eddy-viscosity turbulence model has been modified to correctly account for the effects of the vortices on the underlying viscous layers. The numerical results show that the vortical flow structure of both subsonic and supersonic high incidence flows are similar, and can be accurately treated by the turbulence model. As the angle of attack is increased, the effectiveness of the model increases since the influence of the vortices on the underlying boundary layer decreases.

Degani, David↗
DOE OSTI2022

Automated and efficient local adaptive regression for principal component-based reduced-order modeling of turbulent reacting flows

Principal Component Analysis can be used to reduce the cost of Computational Fluid Dynamics simulations of turbulent reacting flows by reducing the dimensionality of the transported variables through projection of the thermochemical state onto a lower-dimensional manifold. However, because of the nonlinearity of the principal component source terms, nonlinear regression techniques must be utilized for the source terms in terms of the principal components. Unfortunately, widely available and utilized nonlinear regression techniques can have prohibitive computational requirements and/or accuracy that is highly dependent on user experience in ad hoc tuning of model architecture and hyperparameters. Here, in this work, a new nonlinear regression approach is proposed that is both computationally efficient and automated so does not require any user input. The approach is evaluated through a priori prediction of principal component source terms using data from a Direct Numerical Simulation of a turbulent nonpremixed n-heptane/air jet flame. In particular, the proposed framework consists of local regressions whose complexity is adapted according to the local nonlinearity of the data: local linear regression when accurate enough and local Artificial Neural Networks when nonlinear regression is required. The number of local clusters for local regression is determined automatically using the Davies-Bouldin index. In addition, Bayesian optimization is utilized for model training (i.e., to select the best architectures and hyperparameters of the nonlinear regressions in an unsupervised fashion), eliminating ad hoc hand-tuning and/or expensive grid searches. Overall, compared to a single, global neural network, the new local adaptive regression approach is shown to have comparable accuracy but 69% less training time due to the utilization of local linear regression and faster training of local neural networks.

42 ENGINEERING↗
DOE OSTI2021

Machine Learning Assisted Safety Modeling and Analysis of Advanced Reactors

With the advances in computational power and numerical methods, analysts can now rely on first-principle simulations to predict ultra-fine details in a variety of applications. Advances in machine learning (ML) have produced algorithms that can now learn high-level abstractions via hierarchical models. This project aims to leverage advances in ML techniques and the available high-resolution simulation data to develop a novel modeling and simulation (M\&S) methodology for reactor safety analysis. While application-agnostic ML techniques are available, complex physics constraints need to be incorporated into ML techniques to build ML-based closures for computationally efficient predictive simulations. This project intends to develop a physics-guided data-driven multi-scale methodology for M\&S of advanced reactors. The project focuses on thermal fluid (T/F) phenomena, which play major roles in advanced reactor safety. Specifically, we propose a data-driven coarse-mesh turbulence model based on local flow features for the transient analysis of thermal mixing and stratification in a sodium-cooled fast reactor (SFR). The model has a coarse-mesh setup to ensure computational efficiency, while it is trained by fine-mesh computational fluid dynamics (CFD) data with Reynolds-averaged Navier-Stokes (RANS) turbulence model to ensure accuracy. Three different neural networks are developed and tested for loss-of-flow transients in the hot pool of SFR, i.e. the densely connected convolutional neural network (DCNN), long-short-term-memory network based on proper orthogonal decomposition (POD-LSTM), and the DCNN informed by LSTM (DCNN-LSTM). The performances of these three neural networks are evaluated based on baseline models. The DCNN-LSTM model has been chosen for further hyperparameter optimization. Furthermore, based on a simplified two-dimensional case, uncertainty quantification (UQ) of the developed ML-based closure are investigated with three methods, i.e. Monte Carlo dropout, deep ensemble, and Bayesian neural network. The developed ML-based turbulent viscosity closure relation based on deep ensemble is then integrated into the system analysis module SAM and serves as a term in the conservation equations. Such a SAM-ML based procedure guarantees that the obtained results are consistent with the physical constraints of the thermal-fluid system. The SAM-ML simulation on the same loss-of-flow transient showed comparable accuracy with the CFD simulation but with a much coarser mesh setup. Last but not least, the ML-based closure improvement with the support of higher-fidelity data from large eddy simulation (LES) is discussed. As a first step towards this direction, a baseline LES simulation is performed to obtain comparable data with RANS results. Based on the early results, future investigation on further improving the ML-based closure is discussed. We believe the developed approach that combines scientific machine learning with nuclear system analysis code can benefit the advanced reactor community as more accurate safety analyses will better characterize reactor safety margins and reduce licensing efforts.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗