Large Eddy Simulation of a scale-model turbofan for fan noise source diagnostic

Leonard, Thomas M.; Sanjosé, Marlène; Moreau, Stéphane; Duchaine, Florent

doi:10.2514/6.2016-3000

Cited by 25 publications

(22 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The present study is the continuation of previous works (Leonard et al, 2016;Pérez Arroyo et al, 2019, and gives a computational analysis with Reynolds Averaged Navier-Stokes (RANS) and Large Eddy Simulations (LES) for the SDT case at approach conditions for which the rotor speed is reduced and the aerodynamic characteristics are far from the design point for which the blade geometries have been optimized. This numerical study aims at comprehending and highlighting the sensitivity of the flow around the rotor blades to different numerical parameters.…”

Section: Introductionmentioning

confidence: 75%

“…In the experimental configuration, there are 22 fan blades and 54 outlet guide vanes (OGV). The first numerical configuration noted as RS (Leonard et al, 2016;Pérez Arroyo et al, 2019), includes the nacelle, the external flow and an angular periodicity of 2π/11 with 2 fan blades. The modeling of the periodic domain shown in Fig.…”

Section: Domain Configurationsmentioning

confidence: 99%

“…All the RANS simulations are performed using ANSYS CFX 17.2. The RS configuration used in (Leonard et al, 2016) with ANSYS CFX 15.0 has been recomputed for the current version to ensure the consistency between results.…”

Section: Ransmentioning

confidence: 99%

See 2 more Smart Citations

CFD Modeling of a Realistic Turbofan Blade for Noise Prediction. Part 1: Aerodynamics

Arroyo¹,

Kholodov²,

Sanjosé³

et al. 2019

Proceedings of Global Power &Amp; Propulsion Society

View full text Add to dashboard Cite

Various aspects of the numerical modeling of a scale-model turbofan blade are investigated for the purpose of future noise assessment. The fan geometry considered is the baseline configuration of the "Fan Noise Source Diagnostic Test" (SDT) experimental setup. Both Reynolds Averaged Navier-Stokes (RANS) simulations and Large Eddy Simulations (LES) are performed for a single rotor blade without the stator vanes and compared against the full configuration from previous works. With RANS simulations the mesh convergence is systematically studied. In the LES, two wall models and two numerical schemes are considered to evaluate the sensitivity of the boundary-layer transition to turbulence on the blade suction side. The different LES results are compared with the RANS solutions obtained using fully turbulent and transitional turbulence models. All RANS results show similar performance, whereas in the LES the no-slip wall condition gives better performance than the log-law slip wall condition. The prism layers on the hub and the shroud change the boundary layer profile upstream of the blade but do not affect the performance. On the blade, the RANS simulations show a laminar recirculation bubble whereas the LES exhibit a leading-edge vortex spiralling radially. The transitional RANS turbulence model gives the best agreement with the LES on the pressure side. The LES results are more affected by the mesh resolution than by the wall model or the numerical scheme, especially in the tip. In the wake all results show a good agreement with the experiment. Nomenclature Variables y + Dimensionless wall distance τ w Wall shear stress C p = P−P ∞ 0.5ρ ∞ U 2 ∞ Mean pressure coefficient C f = τ w 0.5ρU ∞ Mean skin friction coefficient U Mean velocity P Mean pressure Indices w Wall quantity ∞ Free-flow quantity t Total (stagnation) quantity This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License CC-BY-NC-ND 4.0

show abstract

Section: Introductionmentioning

confidence: 75%

Section: Domain Configurationsmentioning

confidence: 99%

See 1 more Smart Citation

CFD Modeling of a Realistic Turbofan Blade for Noise Prediction. Part 1: Aerodynamics

Arroyo¹,

Kholodov²,

Sanjosé³

et al. 2019

Proceedings of Global Power &Amp; Propulsion Society

View full text Add to dashboard Cite

show abstract

“…64 or 117 in its simplified form in Hanson (2001)) provides the upstream and downstream acoustic spectra as a sum over azimuthal modes, turbulent wavenumbers and a scattering index of the upwash turbulent velocity three-dimensional wavenumber spectra Φ ww and an acoustic cascade power spectra. This model has been successfully applied to turbofan configurations (Hanson, 2001;Leonard et al, 2016).…”

Section: Investigated Noise Sourcesmentioning

confidence: 99%

CFD Modeling of a Realistic Turbofan Blade for Noise Prediction. Part 2: Analytical Acoustic Predictions

Sanjosé¹,

Kholodov²,

Arroyo³

et al. 2019

Proceedings of Global Power &Amp; Propulsion Society

View full text Add to dashboard Cite

Noise sources on the rotor of a scale-model turbofan are investigated at approach conditions. The fan geometry considered is the baseline configuration of the "Fan Noise Source Diagnostic Test" (SDT) experimental setup. Based on Reynolds Averaged Navier-Stokes simulations (RANS) competitive noise source mechanisms, the turbulent interaction noise and the trailingedge noise on the rotor are investigated. The noise estimations are compared to acoustic predictions from Large Eddy Simulations (LES) which consider only the rotor row and include a single blade, and to noise measurement performed by NASA. Focus is made on the modeling of the noise excitations in the analytical predictions. Nomenclature δ * Displacement thickness δ 99 Boundary layer thickness ∂p ∂s Streamwise wall pressure gradient Λ f Integral length scale of the turbulence ω Turbulent eddy frequency τ w Wall shear stress Θ Momentum thickness D Turbulent length scale coefficient I Intensity of the turbulence k Turbulent kinetic energy l y Spanwise coherence length Ma Mach number U c Turbulent eddy convective speeds U e Exterior stream velocity W Relative velocity FWH Ffowcs Williams & Hawking's analogy LE Leading-edge LES Large Eddy Simulation RANS Reynolds Averaged Navier-Stokes SDT Fan Noise Source Diagnostic Test TE Trailing-edge TEN Trailing Edge Noise TIN Turbulence Interaction Noise This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License CC-BY-NC-ND 4.0

show abstract

“…Many simulation methods have been proposed for predicting the broadband noise. Some that focus on the dominant interaction noise use a two-step approach in which the wake turbulence is modelled or obtained via experiment as a first step and then the interaction of the turbulence with the FEGV is considered as a second step [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20]. For the two-step approaches to be useful in the design process, simulated wake turbulence quantities upstream of the FEGV are required.…”

Section: Introductionmentioning

confidence: 99%

Analysis of fan-stage gap-flow data to inform simulation of fan broadband noise

Grace

Gonzalez-Martino

Casalino

2019

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

Time-resolved simulations present a new opportunity for studying the disturbances responsible for the broadband interaction noise created by the fan stage. In this paper, two vane configurations from the source diagnostic test at the approach rotor speed were computed with PowerFLOW's very large-eddy simulation (VLES) method using two solution strategies: a coarser mesh near the rotor and a trip to trigger turbulent transition on the rotor; and a much finer mesh near the rotor with no trip. The simulated data allow for an investigation of the potential effect from the vane configuration and an in-depth study of the mean and turbulent flow in the interstage gap. A challenge related to post-processing of high-resolution simulations is discussed. Comparison of the flow quantities with previously obtained Reynolds Averaged Navier–Stokes simulation results indicates that little advantage is gained by running a lattice Boltmann method (LBM)/VLES to simply recover the gap flow parameters for use with a lower-order fan broadband interaction noise calculation method. The true benefit of the LBM/VLES is that the noise calculation can be directly and simultaneously completed with the flow simulation. This article is part of the theme issue ‘Frontiers of aeroacoustics research: theory, computation and experiment’.

show abstract

Large Eddy Simulation of a scale-model turbofan for fan noise source diagnostic

Cited by 25 publications

References 26 publications

CFD Modeling of a Realistic Turbofan Blade for Noise Prediction. Part 1: Aerodynamics

CFD Modeling of a Realistic Turbofan Blade for Noise Prediction. Part 1: Aerodynamics

CFD Modeling of a Realistic Turbofan Blade for Noise Prediction. Part 2: Analytical Acoustic Predictions

Analysis of fan-stage gap-flow data to inform simulation of fan broadband noise

Contact Info

Product

Resources

About