Large Eddy Simulation of Labyrinth Seals and Rib Shapes for Internal Cooling Passges

The turbine is one of the key components in gas turbine engines. To prevent the turbine blades from being badly damaged by their harsh working environment, it is necessary to keep them cool. This can be achieved by enhancement of the heat transfer performance through internal cooling passages. However, the large quantity of flow within this internal cycle inevitably results in mass flow loss, which is a major source of loss in turbomachinery. Therefore labyrinth seals are also investigated in this study, attempting to reduce the flow leakage and further increase the turbine efficiency. Large Eddy Simulation ( LES ) is used for its capability to capture the complex unsteady flow features in this study. Different rib shapes in a fully developed ribbed channel are investigated, aiming to improve the heat transfer performance. An immersed boundary method ( IBM ) is used with LES to generate complex geometries. With the use of IBM , the range of geometries can be represented on a background Cartesian grid. To obtain the best sealing performance, an investigation is undertaken into the possibility of optimising labyrinth seal planforms using a genetic algorithm ( GA ). By making use of the large number of populations, a much faster calculation can be achieved toward the objective function. Three hundred LES calculations are carried out, and an optimised design is generated that maximises the sealing effectiveness. The optimised design shows a leakage reduction of about 27.6% compared to the baseline geometry. The optimisation process employing a GA will be continued. It is expected that automated optimisation as presented will become increasingly important in the design process of future turbomachines, particularly for flows with strong parameter interactions, with an aim to further improve the overall efficiency of gas turbines.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2018

Author dc:creator

• Dai, Yushuang

Advisor dc:contributor.advisor

• Tucker, Paul

• LES
• GA
• turbomachinary

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/9b66b61f-36a7-4dbf-9680-90c18d3f4d35/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

en

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.18745

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/271753