Abstr:AC6-14: Difference between revisions
m (moved Lib:AC6-14 to Abstr:AC6-14) |
|||
| (One intermediate revision by one other user not shown) | |||
| Line 10: | Line 10: | ||
=Abstract= | =Abstract= | ||
<!--{{Demo_AC_Guidance}}--> | <!--{{Demo_AC_Guidance}}--> | ||
This test case concerns flow unsteadiness generated in a swirl apparatus which was investigated experimentally | |||
and numerically. | and numerically. | ||
The swirl apparatus, shown in figure below, | The swirl apparatus, shown in the figure below, was designed in | ||
Timişoara, Romania. | Timişoara, Romania, where also the measurements were carried out. LDA measurements were performed along three survey axes in the test section to | ||
provide the characteristics of the swirling flow in the conical diffuser. | provide the characteristics of the swirling flow in the conical diffuser. | ||
The swirling flow configuration corresponds to part load operation of a Francis turbine. | The swirling flow configuration corresponds to part-load operation of a Francis turbine. | ||
A series of numerical simulations | A series of numerical simulations undertaken to study the highly swirling turbulent | ||
flow generated by rotor-stator interaction in | flow generated by rotor-stator interaction in the test-case swirl generator are reported here. | ||
The purpose is to assess the applicability of different turbulence models in swirling | The purpose is to assess the applicability of different turbulence models in swirling | ||
flow with a high level of unsteadiness and a significant production and dissipation of | flow with a high level of unsteadiness and a significant production and dissipation of | ||
| Line 24: | Line 23: | ||
Nine turbulence models are compared: four high-Reynolds number URANS, | Nine turbulence models are compared: four high-Reynolds number URANS, | ||
two low-Reynolds number URANS and three hybrid URANS-LES models. | two low-Reynolds number URANS and three hybrid URANS-LES models. | ||
The URANS models are capable of capturing the main unsteady | The URANS models are capable of capturing the main unsteady features of this flow, | ||
the so-called helical vortex rope, which is formed by the strong centrifugal force and | the so-called helical vortex rope, which is formed by the strong centrifugal force and | ||
an on-axis recirculation region. | an on-axis recirculation region. | ||
| Line 35: | Line 34: | ||
The hybrid methods also capture the blade wakes better than the other models, | The hybrid methods also capture the blade wakes better than the other models, | ||
elucidating the wake interaction with the vortex rope. | elucidating the wake interaction with the vortex rope. | ||
The frequency of the vortex rope is predicted well and the total turbulence | The frequency of the vortex rope is predicted well, and the total turbulence | ||
(resolved and modeled) | (resolved and modeled) as delivered by the hybrid method corresponds reasonably well to | ||
the experimental results. | the experimental results. | ||
Latest revision as of 09:54, 30 August 2016
Swirling flow in a conical diffuser generated with rotor-stator interaction
Application Area 6: Turbomachinery Internal Flow
Application Challenge AC6-14
Abstract
This test case concerns flow unsteadiness generated in a swirl apparatus which was investigated experimentally and numerically. The swirl apparatus, shown in the figure below, was designed in Timişoara, Romania, where also the measurements were carried out. LDA measurements were performed along three survey axes in the test section to provide the characteristics of the swirling flow in the conical diffuser. The swirling flow configuration corresponds to part-load operation of a Francis turbine. A series of numerical simulations undertaken to study the highly swirling turbulent flow generated by rotor-stator interaction in the test-case swirl generator are reported here. The purpose is to assess the applicability of different turbulence models in swirling flow with a high level of unsteadiness and a significant production and dissipation of turbulence in the flow away from the wall. Nine turbulence models are compared: four high-Reynolds number URANS, two low-Reynolds number URANS and three hybrid URANS-LES models. The URANS models are capable of capturing the main unsteady features of this flow, the so-called helical vortex rope, which is formed by the strong centrifugal force and an on-axis recirculation region. However, the size of the on-axis recirculation region is overestimated by the URANS models. It is shown that a more detailed resolution, which is provided by the hybrid URANS-LES methods, is necessary to capture the turbulence and the coherent structures. The flow contains a strong disintegration of the vortex rope which is predicted well by the hybrid URANS-LES models. The hybrid methods also capture the blade wakes better than the other models, elucidating the wake interaction with the vortex rope. The frequency of the vortex rope is predicted well, and the total turbulence (resolved and modeled) as delivered by the hybrid method corresponds reasonably well to the experimental results.
|
| Schematic view of Timişoara swirl generator |
Contributed by: A. Javadia, A. Bosiocb, H Nilssona, S. Munteanc, R. Susan-Resigab — aChalmers University of Technology, Göteborg, Sweden; bUniversity Polytehnica Timişoara, Timişoara, Romania; cCenter for Advanced Research in Engineering Sciences, Romanian Academy, Timişoara Branch, Timişoara, Romania
© copyright ERCOFTAC 2011