Performance of Unsteady Reynolds-Averaged Navier-Stokes and Hybrid Scale-Resolving Simulation Approaches in Simulating a Low-Speed Axial Compressor Rotor

2021 ◽  
Author(s):  
Xiangyi Chen ◽  
Björn Koppe ◽  
Martin Lange ◽  
Wuli Chu ◽  
Ronald Mailach
Keyword(s):  
Experimental Data ◽  
Turbulence Modeling ◽  
Navier Stokes ◽  
Tip Leakage Flow ◽  
Low Speed ◽  
Turbulent Characteristics ◽  
Sst Model ◽  
Reynolds Averaged Navier Stokes ◽  
Sas Model ◽  
Les Model

Abstract Aerodynamics phenomena in compressors are highly unsteady and turbulent. Selecting a proper turbulence-modeling method is significant to reveal the complex flow mechanism in turbomachines. In the current paper, the shear stress transport (SST) model as an unsteady Reynolds-averaged Navier-Stokes (RANS) method, the scale-adaptive simulation (SAS) model, and the zonal wall-modeled large eddy simulation (zonal-LES) as two hybrid scale-resolving simulation approaches have been compared. These turbulence-modeling methods were employed to simulate a single rotor of a low-speed research compressor featuring a tip clearance of 1.3% of chord length. Comparisons were made between the simulation results and the experimental data at three operating points, and the flow fields at the design point have been specifically discussed in detail. The results show that the advantage of the zonal-LES model becomes obvious as the compressor throttles. The zonal-LES model brings a significant improvement over both the SST model and the SAS model in capturing the experimental data, especially the velocity distribution in the low-span region, as well as the loss near the endwalls. The SAS model as a scale-resolving method presents no benefits in predicting the relevant flow compared with the SST model, as the activation of the SAS source term is limited for this test case. For the loss prediction, the variation in the upper half-span region is mainly due to the different results in modeling turbulent characteristics of the tip leakage flow, whereas the mechanism behind the higher loss at the lower half-span predicted by the zonal-LES model is a consequence of the complex topology of the corner separation and the intensive mixing.

A hybrid viscous body force model for low-speed centrifugal compressors

2020 ◽  
pp. 095765092094426 ◽  
Author(s):  
Zhiyuan Liu ◽  
Qingjun Zhao ◽  
Xiaorong Xiang ◽  
Wei Zhao ◽  
Xiaoyong Zhou
Keyword(s):  
Body Force ◽  
Navier Stokes ◽  
Force Model ◽  
Tip Leakage Flow ◽  
Centrifugal Compressors ◽  
Low Speed ◽  
Force Method ◽  
Body Force Method ◽  
Body Forces ◽  
Reynolds Averaged Navier Stokes

The flow in centrifugal compressors is viscous and unsteady. Flow separation off the blades challenges the accuracy of simulations. A viscous body force model is expected to speed up numerical convergence and reduce the computational costs of unsteady simulations. In this paper, both stability and accuracy of the viscous body force model are investigated based on the case of a low-speed centrifugal compressor. First, two formulations of the viscous body forces are obtained from the expression of the viscous flux. Then, the numerical stability of two body force models is found to be related to drag coefficient and flow angle. For large negative drag coefficients, the viscous body forces would lead to divergences. Since unsteady Reynolds-averaged Navier–Stokes simulations show that two formulas have considerable accuracy, stability is considered as the main factor for modeling. With the findings, a hybrid viscous body force method is proposed. To assess the applicability of the hybrid model, two test cases are compared against the results obtained by unsteady Reynolds-averaged Navier–Stokes simulations. The first case is the capability evaluation of unsteady characteristics capture for low-speed centrifugal compressors. The simulation results show that the hybrid viscous body force model can capture main unsteady viscous characters, including wake vortexes and tip leakage flow. The other is the case in which the inlet total pressure is disturbed. It is found that fluctuations of pressure, temperature, and velocity predicted by the viscous body force method are close to unsteady Reynolds-averaged Navier–Stokes results. In addition, the time-accurate overall performance of the compressor with disturbance is also predicted satisfactorily. With the advantage in lowering computer resource requirement, the viscous body force model is a promising method for long length scale unsteady cases.

Thermofluid Characteristics of Czochralski Melt Convection Using 3D URANS Computations

2019 ◽  
Vol 11 (6) ◽  
Author(s):  
Sudeep Verma ◽  
Anupam Dewan
Keyword(s):  
Turbulence Modeling ◽  
Three Dimensional ◽  
Comparative Assessment ◽  
Navier Stokes ◽  
Thermal Gradients ◽  
Melt Flow ◽  
Turbulent Characteristics ◽  
Eddy Structures ◽  
Rans Models ◽  
Reynolds Averaged Navier Stokes

Turbulent characteristics of Czochralski melt flow are presented using the unsteady Reynolds-averaged Navier–Stokes (URANS) turbulence modeling approach. Three-dimensional, transient computations were performed using the Launder and Sharma low-Re k-ε model and Menter shear stress transport (SST) k-ω model on an idealized Czochralski setup with counterrotating crystal and crucible. A comparative assessment is performed between these two Reynolds-averaged Navier–Stokes (RANS) models in capturing turbulent thermal and flow behaviors. It was observed that the SST k-ω model predicted a better resolution of the Czochralski melt flow capturing the near wall thermal gradients, resolving stronger convective flow at the melt free surface, deciphering more number of characteristics Czochralski recirculating cells along with predicting large number of coherent eddy structures and vortex cores distributed in the melt and hence a larger level of turbulent intensity in the Czochralski melt compared with that by Launder and Sharma low-Re k-ε model.

Use of Experimental Data for an Efficient Description of Turbulent Flows

1990 ◽  
Vol 43 (5S) ◽  
pp. S240-S245 ◽  
Author(s):  
N. Aubry
Keyword(s):  
Experimental Data ◽  
Turbulent Flows ◽  
Turbulence Modeling ◽  
Stokes Equations ◽  
Original System ◽  
Navier Stokes ◽  
Lorenz Attractor ◽  
Wall Region ◽  
Lorenz Equations ◽  
Chaotic Solution

The proper orthogonal decomposition (POD), also called Karhunen-Loe`ve expansion, which extracts ‘coherent structures’ from experimental data, is a very efficient tool for analyzing and modeling turbulent flows. It has been shown that it converges faster than any other expansion in terms of kinetic energy (Lumley 1970). First, the POD is applied to the chaotic solution of the Lorenz equations. The dynamics of the Lorenz attractor is reconstructed by only the first three POD modes. In the second part of this paper, we show how the POD can be used in turbulence modeling. The particular case studied is the wall region of a turbulent boundary layer. In this flow, the velocity field is expanded into POD modes in the normal direction and Fourier modes in the streamwise and spanwise directions. Dynamical systems are obtained by Galerkin projections of the Navier Stokes equations onto the different modes. Aubry et al. (1988) applied the technique to derive and study a ten dimensional representation which reproduced qualitatively the bursting event experimentally observed. It is shown that streamwise modes, absent in Aubry et al.’s model, participate to the bursting events. This agrees remarkably well with experimental observations. In both examples, the dynamics of the original system is very well recovered from the contribution of only a few modes.

A 3-D Thin Layer Navier-Stokes Code for Supersonic Laminar and Turbulent Flows

2002 ◽  
Author(s):  
Omid Abouali ◽  
Mohammad M. Alishahi ◽  
Homayoon Emdad ◽  
Goodarz Ahmadi
Keyword(s):  
Experimental Data ◽  
Shock Wave ◽  
Mach Number ◽  
Turbulent Flows ◽  
Turbulence Modeling ◽  
Cross Flow ◽  
Navier Stokes ◽  
Cross Sectional ◽  
Laminar And Turbulent Flows ◽  
Circumferential Pressure

A 3-D Thin Layer Navier-Stokes (TLNS) code for solving viscous supersonic flows is developed. The new code uses several numerical algorithms for space and time discretization together with appropriate turbulence modeling. Roe’s method is used for discretizing the convective terms and the central differencing scheme is employed for the viscous terms. An explicit time marching technique and a finite volume space discretization are used. The developed computational model can handle both laminar and turbulent flows. The Baldwin-Lomax model and Degani-Schiff modifications are used for turbulence modeling. The computational model is applied to a hypersonic laminar flow at Mach 7.95 around a cone at different incidence angles. The circumferential pressure distribution is compared with the experimental data. The cross-sectional Mach number contours are also presented. It is shown that in addition to the outer shock, a cross-flow shock wave is also present in the flow field. The cases of supersonic turbulent flows with Mach number 3 around a tangent-ogive with incidence angles of 6° and a secant-ogive with incidence angles of 10° are also studied. The circumferential pressure distributions are compared with the experimental data and the Euler code results and good agreement is obtained. The cross-sectional Mach number contours are also presented. It is shown that in this case also in addition to the outer shock, a cross-flow shock wave is also present at the incidence angle of 10°.

On the Investigation of Flow around the Square Cylinder Based on Different LES Models

2012 ◽  
Vol 594-597 ◽  
pp. 2676-2679
Author(s):  
Zhe Liu
Keyword(s):  
Flow Characteristics ◽  
Turbulence Models ◽  
Navier Stokes ◽  
Detached Eddy Simulation ◽  
Square Cylinder ◽  
Rans Model ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Reynolds Averaged Navier Stokes ◽  
Les Model

Although the conventional Reynolds-averaged Navier–Stokes (RANS) model has been widely applied in the industrial and engineering field, it is worthwhile to study whether these models are suitable to investigate the flow filed varying with the time. With the development of turbulence models, the unsteady Reynolds-averaged Navier–Stokes (URANS) model, detached eddy simulation (DES) and large eddy simulation (LES) compensate the disadvantage of RANS model. This paper mainly presents the theory of standard LES model, LES dynamic model and wall-adapting local eddy-viscosity (WALE) LES model. And the square cylinder is selected as the research target to study the flow characteristics around it at Reynolds number 13,000. The influence of different LES models on the flow field around the square cylinder is compared.

Calculation of Debris Trajectories During High-Speed Snowplowing

2002 ◽  
Author(s):  
H. K. Nakhla ◽  
B. E. Thompson
Keyword(s):  
Particle Tracking ◽  
High Speed ◽  
Turbulence Modeling ◽  
Stokes Equations ◽  
Cutting Edge ◽  
Navier Stokes ◽  
Engineering Model ◽  
Navier Stokes Equations ◽  
Snow Plowing ◽  
Reynolds Averaged Navier Stokes

An engineering model is presented to calculate the trajectory of airborne debris that adversely affects visibility during high-speed snow plowing. Reynolds-averaged Navier-Stokes equations are solved numerically with turbulence-modeling, particle-tracking, and cutting-edge approximations. Results suggest snow can be divided into splash and snow-cloud when designing treatments to improve visibility for snowplow drivers and following traffic. Calculated results confirm the findings of windtunnel and road tests, specifically that the trap angle of overplow deflectors should be less than 50 degrees to eliminate snow debris blowing over top of the plow onto the windscreen.

scholarly journals Jet Noise in Airframe Integration and Shielding

2020 ◽  
Vol 10 (2) ◽  
pp. 511
Author(s):  
Saman Salehian ◽  
Reda Mankbadi
Keyword(s):  
Experimental Data ◽  
Large Eddy Simulation ◽  
Jet Noise ◽  
Numerical Approach ◽  
Navier Stokes ◽  
Complex Geometries ◽  
Eddy Simulation ◽  
Large Eddy ◽  
Reynolds Averaged Navier Stokes

This paper reviews and presents new results on the effect of airframe integration and shielding on jet noise. Available experimental data on integration effects are analyzed. The available options for the computation of jet noise are discussed, and a practical numerical approach for the present topic is recommended. Here, it is demonstrated how a hybrid large eddy simulation—unsteady Reynolds-averaged Navier-Stokes approach can be implemented to simulate the effect of shielding on radiated jet noise. This approach provides results consistent with the experiment and suggests a framework for studying more complex geometries involving airframe integration effects.

An Aerodynamic Investigation of an Isolated Stationary Formula 1 Wheel Assembly

2012 ◽  
Vol 134 (2) ◽  
Author(s):  
John Axerio-Cilies ◽  
Emin Issakhanian ◽  
Juan Jimenez ◽  
Gianluca Iaccarino
Keyword(s):  
Experimental Data ◽  
Large Scale ◽  
Turbulence Modeling ◽  
Cross Flow ◽  
Vortex Pair ◽  
Navier Stokes ◽  
Eddy Simulation ◽  
Flow Features ◽  
Large Scale Flow ◽  
Flow Blockage

The flowfield around a 60% scale stationary Formula 1 tire in contact with the ground in a closed wind tunnel at a Reynolds number of 500,000 was computationally examined in order to assess the accuracy of different turbulence modeling techniques and confirm the existence of large scale flow features. A simplified and replica tire model that includes all brake components was tested to determine the sensitivity of the wake to cross flow within the tire hub along with the flow blockage caused by the brake assembly. The results of steady and unsteady Reynolds averaged Navier-Stokes (URANS) equations and a large eddy simulation (LES) were compared with the experimental data. The LES closure and the RANS closure that accounted for unsteadiness with low eddy viscosity (unsteady kω-SST) matched closest to the experimental data both in point wise velocity comparisons along with location and intensity of the strong counter-rotating vortex pair dominating the far wake of the tire.

Sign in / Sign up

Export Citation Format

Share Document