Aero-elastic simulations of Counter-Rotating Open-Rotors (CROR)
10èmes Journées Des Doctorants MFE/ Nord de l’Onera
Benjamin FRANCOIS
2nd year PhD Student (Airbus/Cerfacs)
Funding : CIFRE Airbus
Ph.D Supervisors: Michel COSTES (ONERA-
DAAP/H2T)Guillaume DUFOUR (ISAE)
Airbus Advisor :Florian BLANC (EGAMT2)
2
IntroductionNew engine concepts for civil aircrafts are driven by
the fuel burn reduction motivated by : Prediction of rise of oil prices due to the growth of air traffic Strong environmental concern (reduction of CO2, NOx)
Counter Rotating Open Rotors (CROR) appears to be one suitable option ...
CROR concept
Example of aircraft powered by open rotorshttp://www.flightglobal.com
The unducted nature of the open-rotors lead to strong aerodynamic interactions with the aircraft and to coupled complex phenomena
3
IntroductionSome challenges adressed by this Ph.D :
1P-forces Transverse forces due to non-homogeneous inflow
(incidence, installation effect …) Essential for HTP and VTP sizing
Whirl flutter phenomenon Unsteady phenomenon coupling pitching and
yawing motion of the engine system (nacelle + rotors)
1P-forces contributes to whirl flutter Two major accidents referenced (1959,1960) on
Lockheed Electra L188
Whirl flutter on aircraft
1P-forces
α
Incidence effect on a aircraft powered by Open-Rotors
Vinflow
4
Scientific issue
An aircraft powered by CROR contains a large spectrum of frequencies : High frequency (HF) sources :
Front and rear interactions (wakes, potentials effects) driven by BPF ~ 250 Hz – 400 Hz (cruise – Take off) + harmonics Low Frequency (LF) sources :
Interactions between the wake of the pylon and the two rotors ~ 13 Hz – 20 Hz Incidence effect due to rotation speed of CROR ~ 13 – 20 Hz Aircraft modes ~ 10 Hz Whirl flutter of the open rotors ~ 1-10 Hz
Single timestep integration technique becomes too expensive as soon as the spectrum contains wide-separated frequencies
The goal of this Ph.D is to develop moderate-cost numerical methods and tools able to capture phenomena with wide-separated frequency
5
Consolidate experiences and best practices for unsteady simulations of isolated CROR with 360° configurations
=> Reference computations to be compared with moderate-cost techniques
Investigation and development of alternative techniques to compute at moderate-cost unsteady phenomena with wide-separated frequency
Contribution to improve the physical analysis of whirl flutter and its prediction tools
Extend these methods and tools to installed configurations on aircraft
Scientific objectives of the Ph.D
6
Outline
1. Introduction/Context
2. Unsteady simulations with 360° configurations of CROR1. Mesh strategy2. Aerodynamic interactions3. Incidence effect4. Whirl flutter
3. Moderate-cost techniques for CROR computations1. Bibliography2. Spectral Method for CROR
7
Isolated test-case : AI-PX7 configurationThe isolated configuration AI-PX7 was selected as a test-case to assess numerical methods and tools in the frame work of this Ph.D
AI-PX7 is the Airbus generic design used on European Research Platform Cleansky JTI-SFWA used to :
Develop and validate dedicated methods and tools for the aerodynamic simulations of CROR Enhance the understanding of the complex aerodynamic flows of CROR
Some features about the geometry: 11x9 bladed pushed configuration Front rotor diameter = 14 ft (4.2672m) Aft rotor blades are cropped by 10%
Several flight conditions were simulated on this isolated CROR configuration with an URANS approach (flight point ; unsteadiness to capture)
1.Cruise conditions (M=0.73, alt=35 kft), no incidence; aerodynamic interactions 2.Cruise conditions (M=0.73, alt=35 kft), incidence ; aerodynamic interactions + incidence effect3.Cruise conditions (M=0.73, alt=35 kft) , pitching motion; aerodynamic interactions + incidence effect + pitching motion effect
Airbus generic geometry (AI-PX7)
8
Mesh strategy360° configurations were chosen because
suitable to model : - incidence effect - installation effect - whirl flutter motion
Mesh Topology : Structured multi-blocks mesh Full 360° mesh (53M pts) Mesh is split into three domains :
Background fixed mesh : Far-field and nacelle area Cylindrical shaped rotating meshes : Front rotor & Rear
rotorThis decomposition offers two advantages : Mesh strategy suitable for the modelling of
installation effect No propagation of refined cell of rotor meshes into
the far-field area
Chimera/Sliding mesh interfaces btw domains
Mesh domain for isolated CROR
13 Rblade
8 LCROR
9
Aerodynamic interactions around an open-rotors
Flow-fields exhibits strong aerodynamic periodic phenomena called aerodynamic interactions :
Pressure waves
Propagation of acoustic waves upstream and downstream Wakes
Deficit of velocity propagating downstream
Ωcur/opp rotation speed of current/opposite stageNopp number of blades in the opposite stage
BPF = (Ωopp – Ωcur).Nopp Front Rotor Aft Rotor
Ω
ΩBPFFR = 231.5 Hz BPFAR = 291.5 Hz
These phenomena are well-known in the literature and their frequencies Blade Passing Frequency (BPF) are deterministic :
References : J.M. Tyler and T.G. Sofrin, "Axial Flow Compressor Noise Studies", Society of Automotive Engineers Transactions, vol. 70, p. 309-332, 1962
Aerodynamic interactions are high frequencies sources (BPFFR/AR , 2BPFFR/AR ,3BPFFR/AR … )
What are the impact on these interactions on aerodynamic loads?
10
Cruise conditions (M=0.73, alt=35 kft), incidence : Computation of aerodynamic loads
In case of non-homogeneous inflow (incidence for instance), 360° pressure field is necessary to compute 1P-forces
Pressure field on rotating blade shows unsteady variations with wide-separated frequencies Incidence effect (Low-frequency) Aerodynamic interactions (High frequency)
1P-forces
Unsteady pressure coefficient field on the rear
blade intrados
Vinflow
11
Cruise conditions (M=0.73, alt=35 kft), incidence : Do we need to capture aerodynamic interactions ?
8% 2%
3%
0.5%
Y
FRONT ROTOR
Cy+Cz
Z
Cy
Cz
Several integration timesteps (0.5°/it, 1°/it, 2°/it, 4°/it) were tested for these computations for the prediction of 1P-forces
Previous studies in this Ph.D (not shown here) have proved that maximum timestep of 0.5°/it is necessary to capture 1st harmonic of aerodynamic interactions
Analysis of 1P-forces shows that accurate prediction of lateral force requires at least 1°/iteration
4°/it2°/it1°/it0.5°/it
Vinflow
12
V0cos
V0sin
FRONT BLADE
Th
rust
(N
)
Cruise conditions (M=0.73, alt=35 kft), incidence : Impact on the blade thrust level
60%
r
V0sin sin
loc
V0cos
V
r
V0sin sin
loc
V0cos
V
Incidence effect leads to strong variation of thrust over one rotation
These results have been compared to NLR results (different mesh and solver) and show good agreement
13
Cruise conditions (M=0.73, alt=35 kft), pitching motion : Features
Computation features Integration timestep : 0.5° of rotation per timestep 4 pitching cycles ,10.6 rotations, 7632 iterations CPU time : 20 days using 128 cores on HPC3
Forced pitching motion Kinematics
Rotation around Y-axis Center of rotation : (-2.15,0,0) => this point has been chosen to model the effect of
a CROR cantilevered in the pylon (pusher configuration) Magnitude : Frequency : Equation
Modelling Rigid motion No structure model -> aerodynamic forces
Z
X+1°
-1°
(t) msin(2ft)
1mHzf 5
Vinf
)(tCenter of rotation
14
Cruise conditions (M=0.73, alt=35 kft), pitching motion : Impact on blade thrust
Unsteady phenomena involved : Rotor-rotor interactions (BPF = 231.5Hz, 291.5Hz, …) Incidence effect (frot = 13.25Hz ) Pitching effect (f = 5Hz)
There is no periodicity because pitching mode is not linked to the rotation speed !
15
Outline
1. Introduction/Context
2. Unsteady simulations with 360° configurations of CROR1. Mesh strategy2. Aerodynamic interactions3. Incidence effect4. Whirl flutter
3. Moderate-cost techniques for CROR computations1. Bibliography2. Spectral Method for CROR
16
References :1.C.Farhat, M. Chandesris, Time-decomposed parallel time-integrators : theory and feasibility studies for fluid, structure, and fluid-structure applications, International Journal for numerical methods in engineering, 20032. J-L LIONS and al. , Résolution d’EDP par un schéma en temps pararéel, Comptes rendus de l'Académie des sciences. Série 1, Mathématique, 20013. C.K.W. Tam and K.A. Kurbatskii, Multi-Size-Mesh Multi-Time-Step Dispersion-Relation-Preserving Scheme for Multiple-Scales Aeroacoustics Problems, International Journal of Computational Fluid Dynamics4. T. Le Garrec, X. Gloerfelt, C. Corre, Multi-Size-Mesh, Multi-Time-Step Algorithm for Noise Computation Around an Airfoil in Curvilinear Meshes, 13th AIAA/CEAS Aeroacoustics Conference (28th AIAA Aeroacoustics Conference)5. Zhi Yang and Dimitri Mavriplis, Time Spectral Method for Periodic and Quasi-Periodic Unsteady Computations on Unstructured Meshes2010 - AIAA- 40th Fluid Dynamics Conference and Exhibit6. K.Ekici and K.C.Hall : Non-linear Analysis of Unsteady Flows in Multtistage Turbomachines Using the Harmonique Blance Technique. AIAA Journal, 45(5) : 1047-1057, 2007. AIAA Paer 2006-422
YesNo
Multi-timescale approaches
Multi-timescale approaches
Time–parallel approach1,2
Multi-size meshMulti-timestep3,4
Coupling methods
Coupling of two calculations solving its own timescale
Hybrid spectral approach coupled
with URANS5
Spectral approach
for Low Frequency
Timescales are spatially separated ?
Spectral methods(HBT)6
17
Time-Spectral Method (TSM)
Spectral Approach
Principle :
Only one blade passage is meshed and computed
Periodic conditions are applied on the boundaries of the domain
CROR mesh suitable for TSM
TSM solve directly the periodic flow without solving useless unsteady transient part CPU cost expected to be lower than URANS
TSM (mono-frequency) can only applied to isolated CROR without incidence
References :A Time-Domain Harmonic Balance Method for Rotor/Stator Interactions Frédéric Sicot, Guillaume Dufour and Nicolas Gourdain J. Turbomach. -- January 2012 -- Volume 134, Issue 1, 011001 (13 pages) doi:10.1115/1.4003210
18
Extension to multi-frequencies techniques : Harmonic Balance Technique Harmonic Balance Technique is based on the same principle as TSM but enables to
solve periodic flows with multiples frequencies
Extension to multi-frequencies methods (HBT) for simulations with incidence and whirl flutter
Mesh for HBT One channel per rotor 360° Far-field mesh : incidence effect, installation
effect Development of duplicating sector sliding interfaces to
allow communication between 360° sliding interface
and sector sliding interface HBT simulations without incidence, with incidence …
sector sliding interface
360° sliding interface
Mesh suitable for HBT
References :"Non evenly spaced timelevels for multifrequential harmonic balance computations" T. Guédeney, A. Gomar, F. Sicot, G. Dufour , will be submitted in 2012 for AIAA Journal"Evaluation of two numerical approaches for the simulation of multi-frequential turbomachinery flows ”, T. Guédeney, N. Gourdain, L. Castillon , will be submitted in 2012 for AIAA Journal or Journal of Turbomachinery
19
ConclusionsAchievements 360° CROR simulations were performed and analysed for :
Aerodynamic performances 1P-loads prediction Whirl flutter forced motion
Deep unsteady analysis of wide-separated phenomena around CROR Development/Validation of moving grid techniques (chimera/sliding) for
CROR=> Proceeding at ISABE 2011 Implementation of whirl flutter motion in elsA code Development and validation of TSM capabilities for CROR applications
Future work TSM simulations currently running and to be compared with 360° config with
URANS (no incidence) Extension to multi-frequencies applications with Harmonic Balance Technique
20
Publications and training ProceedingsFrançois, Dufour and Costes, Comparison of Chimera and Sliding Mesh Techniques for
Unsteady Simulations of Counter Rotating Open-Rotors, ISABE proceeding, Göteborg (Sweden), sept 2011
JournalOn-going 2012 : B.Francois, M.Costes and G.Dufour,1P-loads prediction at cruise conditions
on Counter-Rotating Open Rotors, AIAA Journal
Training « Aérodynamique des hélices », Jean-Marc Bousquet, SUPAERO, 3ème
year course, 2010 « Physique Général de l’avion », Airbus Training, 2011
21
© AIRBUS Operations S.A.S. All rights reserved. Confidential and proprietary document. This document and all information contained herein is the sole property of AIRBUS Operations S.A.S. No intellectual property rights are granted by the delivery of this document or the disclosure of its content. This document shall not be reproduced or disclosed to a third party without the express written consent of AIRBUS Operations S.A.S. This document and its content shall not be used for any purpose other than that for which it is supplied. The statements made herein do not constitute an offer. They are based on the mentioned assumptions and are expressed in good faith. Where the supporting grounds for these statements are not shown, AIRBUS Operations S.A.S. will be pleased to explain the basis thereof.AIRBUS, its logo, A300, A310, A318, A319, A320, A321, A330, A340, A350, A380, A400M are registered trademarks.
Thank you for your attention