NUMERICAL SIMULATION OF MECHANICAL AND THERMAL FLUID–STRUCTURE INTERACTION IN LABYRINTH SEALS

• Main Author: Du, Yu
• Format: Others
• Language: English; en
• Published: 2010
• Online Access: https://tuprints.ulb.tu-darmstadt.de/2253/2/mythesis_lebenslauf.pdf; Du, Yu <http://tuprints.ulb.tu-darmstadt.de/view/person/Du=3AYu=3A=3A.html> (2010): NUMERICAL SIMULATION OF MECHANICAL AND THERMAL FLUID–STRUCTURE INTERACTION IN LABYRINTH SEALS.Darmstadt, Technische Universität, [Ph.D. Thesis]

Labyrinth seals are widely used in jet engines to control the air leakage and thus reduce the specific fuel consumption. The mechanical and thermal interaction of the leakage flow and the rotor/stator has strong influences on the performance of labyrinth seals. In previous numerical studies, such fluid–structure interaction (FSI) effects are usually treated by decoupling the fluid and solid fields. However, fully coupled FSI modeling is indeed required due to the tightly coupled physics. This thesis aims at investigating various mechanical and thermal FSI effects in labyrinth seals by means of numerical simulations. In particular, an implicit partitioned approach is applied to study the rotor vibration induced by fluid forces, the fluid–solid heat transfer, and the impact of centrifugal growth. The results show that, periodic oscillations of the rotor can be induced by fluid forces, and the amplitude is linearly dependent on the pressure ratio and mass flow. In the study of thermal and centrifugal effects, the difference between FSI and CFD simulations is discussed in depth, which provides guidelines on the choice of models for future research. Moreover, the influences of various operating conditions on the seal performance are thoroughly investigated in terms of primary dimensionless numbers. Finally, the heat transfer across the fluid–rotor/stator interfaces is discussed in detail. By introducing fully coupled FSI simulations, the current research of labyrinth seals is enriched in respect of: 1) studying FSI effects that are beyond the scope of individual CFD/CSM simulations; 2) reducing potential errors introduced by single field approximations; 3) obtaining additional information and more accurate predictions of fluid and solid fields.