0
Research Papers: Gas Turbines: Turbomachinery

Development and Applications of a Coupled Particle Deposition—Dynamic Mesh Morphing Approach for the Numerical Simulation of Gas Turbine Flows

[+] Author and Article Information
Peter R. Forsyth

Southwell Laboratory,
Department of Engineering Science,
University of Oxford,
Oxford OX2 0ES, UK
e-mail: peter.forsyth@eng.ox.ac.uk

David R.H. Gillespie

Southwell Laboratory,
Department of Engineering Science,
University of Oxford,
Oxford OX2 0ES, UK
e-mail: david.gillespie@eng.ox.ac.uk

Matthew McGilvray

Southwell Laboratory,
Department of Engineering Science,
University of Oxford,
Oxford OX2 0ES, UK
e-mail: matthew.mcgilvray@eng.ox.ac.uk

1Corresponding author.

Contributed by the Turbomachinery Committee of ASME for publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received July 9, 2017; final manuscript received July 20, 2017; published online October 3, 2017. Editor: David Wisler.

J. Eng. Gas Turbines Power 140(2), 022603 (Oct 03, 2017) (11 pages) Paper No: GTP-17-1325; doi: 10.1115/1.4037825 History: Received July 09, 2017; Revised July 20, 2017

The presence and accretion of airborne particulates, including ash, sand, dust, and other compounds, in gas turbine engines can adversely affect performance and life of components. Engine experience and experimental work have shown that the thickness of accreted layers of these particulates can become large relative to the engine components on which they form. Numerical simulation to date has largely ignored the effects of resultant changes in the passage geometry due to the build-up of deposited particles. This paper will focus on updating the boundaries of the flow volume geometry by integrating the deposited volume of particulates on the solid surface. The technique is implemented using a novel, coupled deposition-dynamic mesh morphing (DMM) approach to the simulation of particulate-laden flows using Reynolds-averaged Navier–Stokes modeling of the bulk fluid, and Lagrangian-based particulate tracking. On an iterative basis, the particle deposition distributions are used to modify the surface topology by altering the locations of surface nodes, which modifies the mesh. The continuous phase solution and particle tracking are then recalculated. The sensitivity to the modeling time steps employed is explored. An impingement geometry case is used to assess the validity of the technique, and a passage with film cooling holes is interrogated. Differences are seen for all sticking and solid phase motion models employed. At small solid particle sizes, considerable disparity is observed between the particle motion modeling approaches, while the position and level of accretion is altered through the use of a nonisotropic stick and bounce model.

Copyright © 2018 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Comparison of simulations of Liu and Agarwal [22] pipe flow experiments using the CRW and DRW models. Figure taken from Ref. [24].

Grahic Jump Location
Fig. 2

Diagram demonstrating dynamic mesh morphing approach

Grahic Jump Location
Fig. 3

“Boundary layer smoothing” method demonstrating retention of boundary layer mesh despite significant deformation: (a) undeformed mesh, (b) after one mesh morphing iteration, and (c) after four mesh morphing iterations

Grahic Jump Location
Fig. 4

Computational domain for Clum et al. [10]. Top: schematic. Flow top to bottom from inlet, through impingement holes, onto impingement plate, and exhausted front and back from the lower volume. Bottom: Impingement hole geometry. See original paper for further details.

Grahic Jump Location
Fig. 5

Particle size distribution for impingement simulations, as sampled from Ref. [10]

Grahic Jump Location
Fig. 6

Build up of impingement plate surface due to particle deposition. Top: 15 mg injected. Bottom: 30 mg injected. Contours show δ/Dh, height normalized by impingement hole diameter (0.71 mm).

Grahic Jump Location
Fig. 7

Averaged deposition profiles plotted on Clum et al. [10] data. Contours of injected mass; simulation: 25 mg; experiment: 100 mg.

Grahic Jump Location
Fig. 8

Effect of changing solver precision on solution. Left: single-precision solver. Right: double precision solver. Contours indicate equal mass injection; profiles averaged radially.

Grahic Jump Location
Fig. 9

Effect of boundary conditions (BS: bounce-stick model; AS: all-stick model) on the transient nature of build up of deposition. PRO (projected) indicates one deposition iteration (1 mg actual injection) projected out to 25 mg linearly.

Grahic Jump Location
Fig. 10

Effect of previous deposition on current particle rebound characteristics. Deposition statistics in Table 1: (a) particle impacts on (left) initial, undeformed mesh, and (right) deformed mesh. Color scale particle velocity magnitude (m/s). dp = 4 μm, peak displacement δ/Dh  = 0.32, and (b) normal coefficient of restitution for particle sizes considered (solid lines, 10, 7, 4 μm, outside in), and ensemble-averaged particle velocity at impact (dashed lines, 10, 7, 4 μm, mean flow, left to right).

Grahic Jump Location
Fig. 11

Computational domain of Wylie et al. [9] test piece. Flow right to left, exiting through film cooling holes.

Grahic Jump Location
Fig. 12

Particle size distribution as sampled from Wylie et al. [9]

Grahic Jump Location
Fig. 13

Comparison of numerical and experimental deposition in the film cooling holes (top row; flow in the main channel is left to right) and main channel (bottom row). Both looking into their respective flows. Photographs from preliminary work in Ref. [9]: (a) numerical deposition in film cooling hole, (b) experimental deposition in film cooling hole, (c) numerical deposition in main channel, and (d) experimental deposition in main channel.

Grahic Jump Location
Fig. 14

Mesh in film cooling hole undeformed (left) and substantially deformed (right). Flow moved axially right to left.

Grahic Jump Location
Fig. 15

Comparison of deposited mass (points, left) and consequent RFP (lines, right) for CRW and DRW models

Grahic Jump Location
Fig. 16

Comparison of different ash “compositions” and temperatures. Experimental RFP values from Wylie et al. for 1 g injected ash: Chaiten 14–19%, Eyja (Tw = 1193 K): < 0.5%, Eyja (Tw = 1300 K): 6%.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In