Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “flow simulation”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Active- and transfer-learning applied to microscale-macroscale coupling to simulate viscoelastic flows

Active- and transfer-learning are applied to microscale dynamics of polymer flows for the multiscale discovery of effective constitutive approximations required in viscoelastic flow simulation. The result is macroscopic rheology directly connected to a microstructural model. Micro and macroscale simulations are adaptively coupled by means of Gaussian process regression (GPR) to run the expensive microscale computations only as necessary. This multiscale method is demonstrated with flows of a polymer solution as a model system. At the microscale level dissipative particle dynamics (DPD) is employed to model the fluid as a suspension of bead-spring micro-structures subjected to steady shear flow. The results yield the non-Newtonian viscosity and the first normal stress difference at strain rates as training data used in a GPR model. DPD parameters are calibrated with respect to experimental data for a real polymer solution. Compliance with these data requires adjustment of the DPD model's cutoff radius, which then becomes a function of the second invariant of the strain rate tensor. The FENE-P model is chosen for the macroscale description using the spectral element method (SEM) to simulate channel flow and flow past a circular cylinder. The DPD results at the lowest possible shear strain rate yield an estimate of the zero-shear rate viscosity, which allows the initiation of the macroscale flow by SEM as a Newtonian fluid. The resulting strain-rate field is surveyed to determine additional shear strain rate sampling points for the DPD system. This new information allows an initial fitting of parameters of the constitutive equation followed by new SEM simulations at the macroscale. Additionally, guided by active-learning GPR to select new sampling points, this process continues until convergence is achieved. The effectiveness of this new simulation paradigm for viscoelastic flows is tested with different macroscale operating conditions. The effective closure learned in the channel simulation is then transferred directly to the flow past a circular cylinder at low Reynolds number, where the results show that only two additional DPD simulations are required to achieve a satisfactory constitutive model. With an increase of the Reynolds number, the active-learning scheme automatically detects the inaccuracy of the learned constitutive model, and initiates additional DPD simulations for the extra data needed to once again close the microscale-macroscale coupled system. This new paradigm of active- and transfer-learning for multiscale modeling is readily applicable to other microscale-macroscale coupled simulations of complex fluids and other materials. Furthermore, the coupling between microscale and macroscale solvers can be seamlessly implemented with our open source multiscale universal interface (MUI) library.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Deep learning closure models for large-eddy simulation of flows around bluff bodies

Near-wall flow simulation remains a central challenge in aerodynamics modelling: Reynolds-averaged Navier–Stokes predictions of separated flows are often inaccurate, and large-eddy simulation (LES) can require prohibitively small near-wall mesh sizes. A deep learning (DL) closure model for LES is developed by introducing untrained neural networks into the governing equations and training in situ for incompressible flows around rectangular prisms at moderate Reynolds numbers. The DL-LES models are trained using adjoint partial differential equation (PDE) optimization methods to match, as closely as possible, direct numerical simulation (DNS) data. They are then evaluated out-of-sample – for aspect ratios, Reynolds numbers and bluff-body geometries not included in the training data – and compared with standard LES models. The DL-LES models outperform these models and are able to achieve accurate LES predictions on a relatively coarse mesh (downsampled from the DNS mesh by factors of four or eight in each Cartesian direction). We study the accuracy of the DL-LES model for predicting the drag coefficient, near-wall and far-field mean flow, and resolved Reynolds stress. A crucial challenge is that the LES quantities of interest are the steady-state flow statistics; for example, a time-averaged velocity component $\langle {u}_i\rangle (x) = \lim _{t \rightarrow \infty } ({1}/{t}) \int _0^t u_i(s,x)\, {\rm d}s$ . Calculating the steady-state flow statistics therefore requires simulating the DL-LES equations over a large number of flow times through the domain. It is a non-trivial question whether an unsteady PDE model with a functional form defined by a deep neural network can remain stable and accurate on $t \in [0, \infty )$ , especially when trained over comparatively short time intervals. Our results demonstrate that the DL-LES models are accurate and stable over long time horizons, which enables the estimation of the steady-state mean velocity, fluctuations and drag coefficient of turbulent flows around bluff bodies relevant to aerodynamics applications.

Mechanics↗
NASA NTRS1977

Evaluation of subgrid-scale turbulence models using a fully simulated turbulent flow

An exact turbulent flow field was calculated on a three-dimensional grid with 64 points on a side. The flow simulates grid-generated turbulence from wind tunnel experiments. In this simulation, the grid spacing is small enough to include essentially all of the viscous energy dissipation, and the box is large enough to contain the largest eddy in the flow. The method is limited to low-turbulence Reynolds numbers, in our case R sub lambda = 36.6. To complete the calculation using a reasonable amount of computer time with reasonable accuracy, a third-order time-integration scheme was developed which runs at about the same speed as a simple first-order scheme. It obtains this accuracy by saving the velocity field and its first-time derivative at each time step. Fourth-order accurate space-differencing is used.

Clark, R. A.↗
NASA NTRS1969

Flow field simulation Patent

Wind tunnel method for simulating flow fields around blunt vehicles entering planetary atmospheres without involving high temperatures

Jones, R. A.↗
NASA NTRS2013

Generating a Simulated Fluid Flow Over an Aircraft Surface Using Anisotropic Diffusion

A fluid-flow simulation over a computer-generated aircraft surface is generated using a diffusion technique. The surface is comprised of a surface mesh of polygons. A boundary-layer fluid property is obtained for a subset of the polygons of the surface mesh. A pressure-gradient vector is determined for a selected polygon, the selected polygon belonging to the surface mesh but not one of the subset of polygons. A maximum and minimum diffusion rate is determined along directions determined using a pressure gradient vector corresponding to the selected polygon. A diffusion-path vector is defined between a point in the selected polygon and a neighboring point in a neighboring polygon. An updated fluid property is determined for the selected polygon using a variable diffusion rate, the variable diffusion rate based on the minimum diffusion rate, maximum diffusion rate, and angular difference between the diffusion-path vector and the pressure-gradient vector.

Rodriguez, David L.↗
NASA NTRS2016

Generating a Simulated Fluid Flow over a Surface Using Anisotropic Diffusion

A fluid-flow simulation over a computer-generated surface is generated using a diffusion technique. The surface is comprised of a surface mesh of polygons. A boundary-layer fluid property is obtained for a subset of the polygons of the surface mesh. A gradient vector is determined for a selected polygon, the selected polygon belonging to the surface mesh but not one of the subset of polygons. A maximum and minimum diffusion rate is determined along directions determined using the gradient vector corresponding to the selected polygon. A diffusion-path vector is defined between a point in the selected polygon and a neighboring point in a neighboring polygon. An updated fluid property is determined for the selected polygon using a variable diffusion rate, the variable diffusion rate based on the minimum diffusion rate, maximum diffusion rate, and the gradient vector.

Rodriguez, David L.↗
DOE OSTI2022

Finding Hidden Patterns in High Resolution Wind Flow Model Simulations

Wind flow data is critical in terms of investment decisions and policy making. High resolution data from wind flow model simulations serve as a supplement to the limited resource of original wind flow data collection. Given the large size of data, finding hidden patterns in wind flow model simulations are critical for reducing the dimensionality of the analysis. In this work, we first perform dimension reduction with two autoencoder models: the CNN-based autoencoder (CNN-AE) [1], and hierarchical autoencoder (HIER-AE) [2], and compare their performance with the Principal Component Analysis (PCA). We then investigate the super-resolution of the wind flow data. By training a Generative Adversarial Network (GAN) with 300 epochs, we obtained a trained model with 2× resolution enhancement. We compare the results of GAN with Convolutional Neural Network (CNN), and GAN results show finer structure as expected in the data field images. Also, the kinetic energy spectra comparisons show that GAN outperforms CNN in terms of reproducing the physical properties for high wavenumbers and is critical for analysis where high-wavenumber kinetics play an important role.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Verification assessment of piston boundary conditions for Lagrangian simulation of compressible flow similarity solutions

This work is concerned with the use of similarity solutions of the compressible flow equations as benchmarks or verification test problems for finite-volume compressible flow simulation software. In practice, this effort can be complicated by the infinite spatial/temporal extent of many candidate solutions or “test problems.” Methods can be devised with the intention of ameliorating this inconsistency with the finite nature of computational simulation; the exact strategy will depend on the code and problem archetypes under investigation. For example, self-similar shock wave propagation can be represented in Lagrangian compressible flow simulations as rigid boundary-driven flow, even if no such “piston” is present in the counterpart mathematical similarity solution. The purpose of this work is to investigate in detail the methodology of representing self-similar shock wave propagation as a piston-driven flow in the context of various test problems featuring simple closed-form solutions of infinite spatial/temporal extent. The closed-form solutions allow for the derivation of similarly closed-form piston boundary conditions (BCs) for use in Lagrangian compressible flow solvers. Finally, the consequences of utilizing these BCs (as opposed to directly initializing the self-similar solution in a computational spatial grid) are investigated in terms of common code verification analysis metrics (e.g., shock strength/position errors and global convergence rates).

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2018

Burial-driven methane recycling in marine gas hydrate systems

Natural gas hydrate may be buried with sediments until it is no longer stable at a given pressure and temperature, resulting in conversion of hydrate into free gas. This gas may migrate upward and recycle back into the hydrate stability zone to form hydrate. As of yet, however, no quantitative description of the methane recycling process has been developed using multiphase flow simulations to model burial-driven gas hydrate recycling. In this study, we present a series of 1D multiphase transport simulations to investigate the methane recycling process in detail. By invoking the effects of capillary phenomena on hydrate and gas formation in pores of varying size, we find that a free gas phase can migrate a significant distance above the bulk base of hydrate stability. Since the top of the free gas occurrence is often identified as the base of the hydrate stability zone from seismic data, our results demonstrate that not only could this assumption mischaracterize a hydrate system, but that under recycling conditions the highest hydrate saturations can occur beneath the top of the free gas occurrence. We show that the presence of pore size distributions requires a replacement zone through which hydrate saturations progressively decrease with depth and are replaced with free gas. This replacement zone works to buffer against significant gas buildup that could lead to fracturing of overlying sediments. In conclusion, this work provides a framework for simulating flow and transport of methane within the 3-phase stability zone from a mass conservation perspective.

58 GEOSCIENCES↗
DOE OSTI2016

Designing a graphics processing unit accelerated petaflop capable lattice Boltzmann solver: Read aligned data layouts and asynchronous communication

The lattice Boltzmann method is a well-established numerical approach for complex fluid flow simulations. Recently, general-purpose graphics processing units (GPUs) have become available as high-performance computing resources at large scale. We report on designing and implementing a lattice Boltzmann solver for multi-GPU systems that achieves 1.79 PFLOPS performance on 16,384 GPUs. To achieve this performance, we introduce a GPU compatible version of the so-called bundle data layout and eliminate the halo sites in order to improve data access alignment. Furthermore, we make use of the possibility to overlap data transfer between the host central processing unit and the device GPU with computing on the GPU. As a benchmark case, we simulate flow in porous media and measure both strong and weak scaling performance with the emphasis being on large-scale simulations using realistic input data.

Computer Science↗
DOE OSTI2019

Interface-Resolved Simulations of Reactor Flows

The presented review paper outlines the recent progress of the high-resolution single- and two-phase flow simulations of reactor-relevant flows. Rapid development of high-performance computing capabilities is creating exciting opportunities to study complex reactor thermal hydraulic phenomena. Today’s advances in thermal hydraulic analysis, machine learning techniques and interface resolved simulations will help pave the way to the next level of understanding of two-phase flow behavior in complex geometries. This paper consists of two major parts: (i) history and review of interface tracking simulations for nuclear thermal hydraulics in recent years and (ii) several opportunities to apply those advanced tools in the future. First part will discuss typical computational methods used for those simulations, provide some examples of the past work, as well as computational cost estimates and affordability of such simulations for research and industrial applications. In the second part some specific examples are discussed which could be analyzed using exascale supercomputers being designed and projected to be online in the early 2020s. New generation methodologies are required in order to take full advantage of those capabilities to greatly enhance the scientific understanding of complex two-phase flow phenomena in nuclear reactors under normal operation and postulated accident conditions.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2015

Flow induced migration in polymer melts – Theory and simulation

Flow induced migration, whereby polymer melts are fractionated by molecular weight across a flow field, represents a significant complication in the processing of polymer melts. Despite its long history, such phenomena remain relatively poorly understood. Here a simple analytical theory is presented which predicts the phenomena based on well-established principles of non-equilibrium thermodynamics. It is unambiguously shown that for purely viscous materials, a gradient in shear rate is needed to drive migration; for purely viscometric flows no migration is expected. Molecular scale simulations of flow migration effects in dense polymer melts are also presented. In shear flow the melts exhibit similar behavior as the quiescent case; a constant shear rate across the gap does not induce chain length based migration. In comparison, parabolic flow causes profound migration for both unentangled and entangled melts. These findings are consistent with the analytical theory. The picture that emerges is consistent with flow induced migration mechanisms predominating over competing chain degradation mechanisms.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
NASA NTRS1991

Flow visualization studies of the internal flow characteristics in a simulated mixed flow vectored thrust ASTOVL engine configuration

The characteristics of internal flowfields in a Mixed Flow Vectored Thrust (MFVT) propulsion system for Short Take Off and Vertical Landing (STOVL) fighter aircraft were investigated using flow visualization techniques. A highly parametric 14 percent scale transparent model was used to simulate the MFVT propulsion system. The working medium was water. The mixing between the fan and core flow was studied using flow visualization for a variety of geometrical and flow parameters including three bypass ratio variations. The flow visualization technique involves the use of laser light sheet for illumination and fluorescent dye or air bubbles as tracers. Results indicate the existence of a strong vortex flowfield in the hover mode of operation that is responsible for good mixing between the fan and core flows.

Saripalli, K. R.↗