Analysis of Highly Swirled, Turbulent Flows in Dump Combustor With Exit Contraction

Author(s):  
Daniel A. Nickolaus ◽  
Clifford E. Smith

Highly swirled flows are commonly used in gas turbine combustors to stabilize the flame and enhance fuel-air mixing. Experiments by D. G. Lilley, 1985 have shown that swirling flow patterns (i.e. recirculation zones) are dramatically impacted by a downstream contraction. For unconstricted swirling flow, a large, central recirculation zone is formed, while for constricted swirling flows, the recirculation zone can be annular in shape and high (positive) axial velocity is seen on the centerline of the combustor. Over the past 20 years, steady-state Reynolds Averaged Navier Stokes (RANS) solutions with various turbulence models have not been able to mimic the flowfield patterns for swirling flow with a downstream contraction. In this study, Large Eddy Simulation (LES) calculations were performed that correctly predicted the recirculation flow patterns for swirling flow with a downstream contraction. In addition, LES predicted radial profiles of swirl velocity agreed well with measurements, including the solid body vortex core on the centerline of the combustor. RANS produced inferior predictions. Two cases with 45° swirlers and a dump combustor with and without a downstream contraction were modeled. The LES predictions were compared with RANS predictions and Lilley’s measurements. The computational domain included flow through the swirl vanes, the combustor, and the contraction area. The unstructured, parallel CFD-ACE+ code was used, with the Localized Dynamic kinetic Energy Model (LDKM) for subgrid turbulence.

2015 ◽  
Vol 789-790 ◽  
pp. 477-483
Author(s):  
A.R. Norwazan ◽  
M.N. Mohd Jaafar

This paper is presents numerical simulation of isothermal swirling turbulent flows in a combustion chamber of an unconfined burner. Isothermal flows of with three different swirl numbers, SN of axial swirler are considered to demonstrate the effect of flow axial velocity and tangential velocity to define the center recirculation zone. The swirler is used in the burner that significantly influences the flow pattern inside the combustion chamber. The inlet velocity, U0 is 30 m/s entering into the burner through the axial swirler that represents a high Reynolds number, Re to evaluate the differences of SN. The significance of center recirculation zone investigation affected by differences Re also has been carried out in order to define a good mixing of air and fuel. A numerical study of non-reacting flow into the burner region is performed using ANSYS Fluent. The Reynolds–Averaged Navier–Stokes (RANS) realizable k-ε turbulence approach method was applied with the eddy dissipation model. An attention is focused in the flow field behind the axial swirler downstream that determined by transverse flow field at different radial distance. The results of axial and tangential velocity were normalized with the U0. The velocity profiles’ behaviour are obviously changes after existing the swirler up to x/D = 0.3 plane. However, their flow patterns are similar for all SN after x/D = 0.3 plane towards the outlet of a burner.


2018 ◽  
Vol 140 (7) ◽  
Author(s):  
Paulius Vilkinis ◽  
Nerijus Pedišius ◽  
Mantas Valantinavičius

Flow over a transitional-type cavity in microchannels is studied using a microparticle image velocimetry system (μPIV) and commercially available computational fluid dynamics (CFD) software in laminar, transitional, and turbulent flow regimes. According to experimental results, in the transitional-type cavity (L/h1 = 10) and under laminar flow in the channel, the recirculation zone behind the backward-facing step stretches linearly with ReDh until the reattachment point reaches the middle of the cavity at xr/L = (0.5 to 0.6). With further increase in ReDh, the forward-facing step lifts the reattaching flow from the bottom of the cavity and stagnant recirculation flow fills the entire space of the cavity. Flow reattachment to the bottom of the cavity is again observed only after transition to the turbulent flow regime in the channel. Reynolds-averaged Navier–Stokes (RANS) equations and large eddy simulation (LES) results revealed changes in vortex topology, with the flow regime changing from laminar to turbulent. During the turbulent flow regime in the recirculation zone, periodically recurring vortex systems are formed. Experimental and computational results have a good qualitative agreement regarding the changes in the flow topology. However, the results of numerical simulations based on RANS equations and the Reynolds-stress-baseline turbulence model (RSM-BSL), show that computed reattachment length values overestimate the experimentally obtained values. The RSM-BSL model underestimates the turbulent kinetic energy intensity, generated by flow separation phenomena, on the stage of transitional flow regime.


2016 ◽  
pp. 92-97
Author(s):  
R. E. Volkov ◽  
A. G. Obukhov

The rectangular parallelepiped explicit difference schemes for the numerical solution of the complete built system of Navier-Stokes equations. These solutions describe the three-dimensional flow of a compressible viscous heat-conducting gas in a rising swirling flows, provided the forces of gravity and Coriolis. This assumes constancy of the coefficient of viscosity and thermal conductivity. The initial conditions are the features that are the exact analytical solution of the complete Navier-Stokes equations. Propose specific boundary conditions under which the upward flow of gas is modeled by blowing through the square hole in the upper surface of the computational domain. A variant of parallelization algorithm for calculating gas dynamic and energy characteristics. The results of calculations of gasdynamic parameters dependency on the speed of the vertical blowing by the time the flow of a steady state flow.


Author(s):  
Andrew P. S. Wheeler ◽  
Richard D. Sandberg

In this paper we use direct numerical simulation to investigate the unsteady flow over a model turbine blade-tip at engine scale Reynolds and Mach numbers. The DNS is performed with a new in-house multi-block structured compressible Navier-Stokes solver purposely developed for exploiting high-performance computing systems. The particular case of a transonic tip flow is studied since previous work has suggested compressibility has an important influence on the turbulent nature of the separation bubble at the inlet to the gap and subsequent flow reattachment. The effects of free-stream turbulence, cross-flow and pressure-side boundary-layer on the tip flow aerodynamics and heat transfer are investigated. For ‘clean’ in-flow cases we find that even at engine scale Reynolds numbers the tip flow is intermittent in nature (neither laminar nor fully turbulent). The breakdown to turbulence occurs through the development of spanwise modes with wavelengths around 25% of the gap height. Cross-flows of 25% of the streamwise gap exit velocity are found to increase the stability of the tip flow, and to significantly reduce the turbulence production in the separation bubble. This is predicted through in-house linear stability analysis, and confirmed by the DNS. For the case when the inlet flow has free-stream turbulence, viscous dissipation and the rapid acceleration of the flow at the inlet to the tip-gap causes significant distortion of the vorticity field and reductions of turbulence intensity as the flow enters the tip gap. This means that only very high turbulence levels at the inlet to the computational domain significantly affect the tip heat transfer. The DNS results are compared with RANS predictions using the Spalart-Allmaras and k–ω SST turbulence models. The RANS and DNS predictions give similar qualitative features for the tip flow, but the size and shape of the inlet separation bubble and shock positions differ noticeably. The RANS predictions are particularly insensitive to free-stream turbulence.


2005 ◽  
Author(s):  
Sowjanya Vijiapurapu ◽  
Jie Cui

The Reynolds averaged Navier-Stokes (RANS) equations were solved along with three turbulence models, namely κ-ε, κ-ω, and Reynolds stress models (RSM), to study the fully developed turbulent flows in circular pipes roughened by repeated square ribs. The spacing between the ribs was varied to form three representative types of surface roughness; d–type, intermediate, and k–type. Solutions of these flows at two Reynolds numbers were obtained using the commercial computational fluid dynamics (CFD) software Fluent. The numerical results were validated against experimental measurements and other numerical data published in literature. Extensive investigation of effects of rib spacing and Reynolds number on the pressure and friction resistance, flow and turbulence distribution was presented. The performance of three turbulence models was also compared and discussed.


2015 ◽  
Vol 72 (4) ◽  
Author(s):  
Norwazan A. R. ◽  
Mohammad Nazri Mohd. Jaafar

The numerical simulations of swirling turbulent flows in isothermal condition in combustion chamber of burner were investigated. The aim is to characterize the main flow structures and turbulence in a combustor that is relevant to gas turbines. Isothermal flows with different inlet flow velocities were considered to demonstrate the effect of radial velocity. The inlet velocity, Uo is varied from 30 m/s to 60 m/s represent a high Reynolds number up to 3.00 X 105. The swirler was located at the upstream of combustor with the swirl number of 0.895. A numerical study of non-reacting flow in the burner region was performed using ANSYS Fluent. The Reynolds–Averaged Navier–Stokes (RANS) approach method was applied with the standard k-ɛ turbulence equations. The various velocity profiles were different after undergoing the different inlet velocity up to the burner exit. The results of velocity profile showed that the high U0 give better swirling flow patterns.


1996 ◽  
Vol 118 (2) ◽  
pp. 308-315 ◽  
Author(s):  
M. D. Durbin ◽  
M. D. Vangsness ◽  
D. R. Ballal ◽  
V. R. Katta

A prime requirement in the design of a modern gas turbine combustor is good combustion stability, especially near lean blowout (LBO), to ensure an adequate stability margin. For an aeroengine, combustor blow-off limits are encountered during low engine speeds at high altitudes over a range of flight Mach numbers. For an industrial combustor, requirements of ultralow NOx emissions coupled with high combustion efficiency demand operation at or close to LBO. In this investigation, a step swirl combustor (SSC) was designed to reproduce the swirling flow pattern present in the vicinity of the fuel injector located in the primary zone of a gas turbine combustor. Different flame shapes, structure, and location were observed and detailed experimental measurements and numerical computations were performed. It was found that certain combinations of outer and inner swirling air flows produce multiple attached flames, aflame with a single attached structure just above the fuel injection tube, and finally for higher inner swirl velocity, the flame lifts from the fuel tube and is stabilized by the inner recirculation zone. The observed difference in LBO between co- and counterswirl configurations is primarily a function of how the flame stabilizes, i.e., attached versus lifted. A turbulent combustion model correctly predicts the attached flame location(s), development of inner recirculation zone, a dimple-shaped flame structure, the flame lift-off height, and radial profiles of mean temperature, axial velocity, and tangential velocity at different axial locations. Finally, the significance and applications of anchored and lifted flames to combustor stability and LBO in practical gas turbine combustors are discussed.


1997 ◽  
Vol 119 (4) ◽  
pp. 900-905 ◽  
Author(s):  
X. Zheng ◽  
C. Liao ◽  
C. Liu ◽  
C. H. Sung ◽  
T. T. Huang

In this paper, computational results are presented for three-dimensional high-Reynolds number turbulent flows over a simplified submarine model. The simulation is based on the solution of Reynolds-Averaged Navier-Stokes equations and two-equation turbulence models by using a preconditioned time-stepping approach. A multiblock method, in which the block loop is placed in the inner cycle of a multi-grid algorithm, is used to obtain versatility and efficiency. It was found that the calculated body drag, lift, side force coefficients and moments at various angles of attack or angles of drift are in excellent agreement with experimental data. Fast convergence has been achieved for all the cases with large angles of attack and with modest drift angles.


2005 ◽  
Vol 73 (3) ◽  
pp. 405-412 ◽  
Author(s):  
Hermann F. Fasel ◽  
Dominic A. von Terzi ◽  
Richard D. Sandberg

A flow simulation Methodology (FSM) is presented for computing the time-dependent behavior of complex compressible turbulent flows. The development of FSM was initiated in close collaboration with C. Speziale (then at Boston University). The objective of FSM is to provide the proper amount of turbulence modeling for the unresolved scales while directly computing the largest scales. The strategy is implemented by using state-of-the-art turbulence models (as developed for Reynolds averaged Navier-Stokes (RANS)) and scaling of the model terms with a “contribution function.” The contribution function is dependent on the local and instantaneous “physical” resolution in the computation. This physical resolution is determined during the actual simulation by comparing the size of the smallest relevant scales to the local grid size used in the computation. The contribution function is designed such that it provides no modeling if the computation is locally well resolved so that it approaches direct numerical simulations (DNS) in the fine-grid limit and such that it provides modeling of all scales in the coarse-grid limit and thus approaches a RANS calculation. In between these resolution limits, the contribution function adjusts the necessary modeling for the unresolved scales while the larger (resolved) scales are computed as in large eddy simulation (LES). However, FSM is distinctly different from LES in that it allows for a consistent transition between RANS, LES, and DNS within the same simulation depending on the local flow behavior and “physical” resolution. As a consequence, FSM should require considerably fewer grid points for a given calculation than would be necessary for a LES. This conjecture is substantiated by employing FSM to calculate the flow over a backward-facing step and a plane wake behind a bluff body, both at low Mach number, and supersonic axisymmetric wakes. These examples were chosen such that they expose, on the one hand, the inherent difficulties of simulating (physically) complex flows, and, on the other hand, demonstrate the potential of the FSM approach for simulations of turbulent compressible flows for complex geometries.


Author(s):  
G. K. Feldcamp ◽  
A. M. Birk

Cold flow experiments were conducted to study swirling flows in an annular diffuser with various strut configurations. Experimental results at 0°, 20°, and 40° of inlet swirl were obtained. Measured properties included detailed inlet and exit traverses using three and seven hole pressure probes, surface pressure taps on the diffuser wall, and surface flow visualization techniques. Evaluation of the diffuser and strut performance was based on pressure recovery, and detailed pressure distributions. The experimental results have been studied further using Reynolds Averaged Navier Stokes (RANS) based Computational Fluid Dynamics (CFD). These models are of modest size (less than four million volumes). Several turbulence models have been used to predict the performance of the annular diffuser with struts. Both high and low Reynolds number implementations of the turbulence models have been investigated. The results show that modest CFD models can be used with confidence to design these devices.


Sign in / Sign up

Export Citation Format

Share Document