The use of gas turbines for power generation has increased significantly over the years, with the global market predicted to grow by 45% from 2013 to 2020. The energy market is also growing more diverse as renewables from different sources become more viable financially and technologically. Gas turbines are therefore likely to become the mainstay for providing the base load, thus driving the need for increased flexibility in terms of operational load as well as flexibility of fuels such as hydrogen and biofuels.
One main challenge towards this future of flexible operations and fuels is the combustor design and its effects on the turbine. Combustion chambers are typically designed and tested in isolation, with little reference to the downstream turbine except through targeted bulk flow properties. The design of the NGV then relies on simple models for the inlet profile such as isotropic turbulence intensity, a turbulence intensity profile, or a swirl profile that is static in space and time. The underlying assumption is that the flow field around the NGV is essentially steady.
However, most modern industrial gas turbines use lean premixed, swirl stabilised combustion which lowers the flame temperature, thus reducing the formation of NOx. The swirl is necessary to mix and stabilise the flame and prevent flashback or backfiring. A recirculation zone is created downstream of the swirler which results in a lower total pressure swirl core that persists far downstream of the swirler and even convects downstream of the NGV.
This central low pressure core is not simply the `wake' of the swirler centre body, but rather is created by so-called vortex breakdown of the swirl in the expanding transition duct. As a result, the flow downstream of a combustor is inherently unsteady and is characterised by large scale coherent structures. These structures have been shown to significantly affect the aerodynamics and heat transfer on an NGV.
This study investigates the steady and unsteady effects of different combustor swirl geometries on these flow structures and consequently, the aerodynamics and heat transfer on a film-cooled NGV. Time-varying, full surface maps of HTC across the vane are presented here which clearly resolve these structures.
A single large bypass swirler has been included in this study as it is representative of conventional swirlers in use today. It is shown in this study that this geometry results in the largest structures and consequently the highest fluctuations in HTC. These effects can be reduced through combustor design, hence showing that greater optimisation could be achieved through the integrated design of the combustor and NGV.
This study will be the first of its kind in investigating the unsteady effects of combustor flows using various high-speed experimental techniques such has high-speed infrared thermography, thin-film gauges, and high-speed surface mount pressure transducers. High-fidelity Large Eddy Simulations will also be carried out to full resolve the unsteady flow field.
This project falls within the EPSRC Energy and Engineering research area.