An integrated nonlinear approach for turbomachinery forced response prediction. Part II: Case studies

Citation
M. Vahdati et al., An integrated nonlinear approach for turbomachinery forced response prediction. Part II: Case studies, J FLUID STR, 14(1), 2000, pp. 103-125
Citations number
6
Categorie Soggetti
Mechanical Engineering
Journal title
JOURNAL OF FLUIDS AND STRUCTURES
ISSN journal
08899746 → ACNP
Volume
14
Issue
1
Year of publication
2000
Pages
103 - 125
Database
ISI
SICI code
0889-9746(200001)14:1<103:AINAFT>2.0.ZU;2-E
Abstract
This paper discusses the application of an advanced turbomachinery forced p rediction model to two representative turbomachinery cases: an HP turbine a nd a rig fan. The approach is based on an integrated nonlinear multi-passag e analysis that includes both the stator and the rotor blade-rows. The nume rical model has advanced features such as nonlinear friction damping for tu rbine blades, tip gap flows and blade vibratory motion. A series of invisci d and viscous computations were performed for an HP turbine with 36 stator and 92 rotor blades. The peak-to-peak maximum displacements were predicted with and without root friction dampers and the findings were compared with available experimental data. Good agreement was observed in most cases. It was found that most of the unsteady forcing was due to the potential effect s. In a second phase of the turbine work, the response of low engine-order excitation was predicted using a multi-row whole-annulus model. The stator assembly was assumed to have blades with varying throat widths and the magn itude of the unsteady aerodynamic forcing was found to increase with increa sing scatter in throat width variation. A second forced response study was conducted for a rig fan with 15 variable inlet guide vanes (VIGVs) and 20 r otor blades. For a 30 degrees VIGV opening, a good match was observed betwe en the predicted and measured wake profiles. Similarly, the measured and pr edicted rotor blade vibration levels were also found to be in good agreemen t. It is concluded that the proposed methodology can be applied, with reaso nable confidence, to the study of industrial cases. (C) 2000 Academic Press .