A non-linear integrated aeroelasticity method for the prediction of turbine forced response with friction dampers

Citation
C. Breard et al., A non-linear integrated aeroelasticity method for the prediction of turbine forced response with friction dampers, INT J MECH, 43(12), 2001, pp. 2715-2736
Citations number
13
Categorie Soggetti
Mechanical Engineering
Journal title
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
ISSN journal
00207403 → ACNP
Volume
43
Issue
12
Year of publication
2001
Pages
2715 - 2736
Database
ISI
SICI code
0020-7403(200112)43:12<2715:ANIAMF>2.0.ZU;2-6
Abstract
The aim of this paper is to describe an integrated aeroelasticity model for turbine blade forced response predictions. Stich an approach requires a su ccessful integration of the unsteady aerodynamics with non-linear structura l dynamics, the latter arising from the use of root friction dampers to dis sipate energy so that the response levels can be kept as low as possible. T he inclusion of friction dampers is known to raise the resonant frequencies by up to 20% from the standard assembly frequencies, a shift that is not k nown prior to the aeroelasticity calculations because of its possible depen dence on the unsteady excitation. An iterative procedure was therefore deve loped in order to determine the resonance shift under the effects of both u nsteady dynamic loading and non-linear friction dampers. The iterative proc edure uses a viscous, non-linear time-accurate flow representation for eval uating the aerodynamic forcing, a look-up table for determining the aerodyn amic boundary conditions at any speed, and a time-domain friction damping m odule for resonance tracking. The methodology was applied to a high-pressur e turbine rotor test case where the resonances of interest were due to firs t torsion and second flap blade modes under 40 engine-order excitation. The forced response computations were conducted using a multi-bladerow approac h in order to avoid errors associated with "linking" single bladerow comput ations since the spacing between the bladerows was relatively small. Three friction damper elements, representing one actual friction damper, were use d for each rotor blade. The number of rotor blades was decreased by 2-90 to obtain a cyclic sector of 4 stator and 9 stator blades. Such a route allow ed the analysis to be conducted on a much smaller domain, hence reducing th e computational effort significantly. However, the stator blade geometry wa s skewed in order to adjust the mass flow rate. Frequency shifts of 3.2% an d 20.0% were predicted for the 40 engine-order resonances in torsion and be nding modes, respectively. The predicted frequency shifts and the dynamic b ehaviour of the friction dampers were found to be within the measured range . Furthermore, the measured and predicted blade vibration amplitudes showed a good agreement with available experimental data, indicating that the met hodology can be applied to typical industrial problems. (C) 2001 Elsevier S cience Ltd. All rights reserved.