Planar Velocity Measurements at 100 kHz in Gas Turbine Combustors With a Continuous Laser Source

2017 
This work aims at presenting a novel approach to measure planar velocity in gas turbine combustors at very high sampling frequencies. For this purpose, a continuous wave laser is used in order to illuminate particles that are seeded into the flow. The Mie scattering images are acquired with a high-speed camera at 100 kHz with a constant time between each frame. The velocity fields are then obtained by applying classical PIV algorithms on successive particle scattering images. While this approach has been recently used in other research fields, such as aerodynamics or hydrodynamics, it is relatively new in combustion studies, where pulsed laser systems with higher power levels are usually preferred. The proposed technique is an economical and ergonomic solution to determine velocity fields at very high sampling frequencies. It is highly portable and safe and convenient to use and align. The main drawback is the long image exposure duration due to the low laser energy. This leads to a smearing effect of the captured particles and acts as a low-pass filter. It has the consequence that the PIV algorithm does not determine the displacement of “dots”, but of “traces”. The measurement technique is tested experimentally on a model gas turbine combustor at a laboratory scale. The test is performed in three steps: (1) The instantaneous velocity fields are analysed in order to verify, whether the flame topology is represented correctly. (2) The mean and RMS velocity fields that are obtained with the present technique are compared with those obtained by classic low speed PIV. (3) Instantaneous synthetic Mie scattering fields are generated from a large eddy simulation (LES) on a similar combustor to test the algorithms. The planar velocity fields are calculated from these images and compared for the two techniques. Finally, possible error sources of the new technique are discussed.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []