The development of kilohertz planar laser diagnostics for applications in high power turbulent flames
Abstract
In modern gas-turbine combustors, flame stabilization is achieved by inducing exhaust gas circulation within the flame zone through swirl-induced vortex breakdown. Swirling flows exhibit strong shear regions resulting in high turbulence and effective mixing. In combustion, these flows are characterized by complex unsteady interactions between turbulent flow structures and chemical reactions. Developments in high-resolution, quantitative, experimental measurement techniques must continue to improve fundamental understanding and support modeling efforts. This work describes the development of a gas turbine combustion experiment to support the application of advanced optical measurement techniques in flames operating at realistic engine conditions. Facility requirements are addressed, including instrumentation and control needs for remote operation when working with high energy flows. The methodology employed in the design of the optically-accessible combustion chamber is elucidated, including window considerations and thermal management of the experimental hardware under extremely high heat loads. Experimental uncertainties are also quantified. The stable operation of the experiment is validated using multiple techniques and the boundary conditions are verified. The successful prediction of operating conditions by the design analysis is documented and preliminary data is shown to demonstrate the capability of the experiment to produce high-fidelity datasets for advanced combustion research. Building on this experimental infrastructure, simultaneous measurements of velocity and scalar fields were performed in turbulent nonpremixed flames at gas turbine engine operating conditions using 5 kHz Particle-Image Velocimetry (PIV) and OH Planar Laser Induced Fluorescence (OH-PLIF). The experimental systems and the challenges associated with acquiring useful data at high pressures and high thermal powers are discussed. The quality of the particle scattering images used in the two-dimensional, two-component velocity field measurements is discussed. The effects of high flame luminosity and particle defocusing on the signal-to-noise ratio are discussed. Laser sheet absorption effects, which have been reported to be severe in many previous high pressure OH-PLIF attempts, were not observed to be significant in this work. The time-averaged peak and (spatial) mean signal to noise ratios were 12.7 and 6.3, respectively, at the flame B operating condition; 550 kW total thermal power and 1.0 MPa combustion chamber pressure. Simultaneous 5 kHz PIV and OH-PLIF measurements showed good agreement between single-shot flow-flame interactions, but unresolved, out-of-plane velocity components restricted the interpretation of the temporal context. At a 5 kHz interrogation frequency, the temporal resolution of the measurements was found to be sufficient for only the largest scales within the turbulent flame. The development of an analysis library for the extraction of physical data from highly-resolved planar measurements is also described. The resolution of the measurements, in space and time, is described with respect to the integral scales of the flow. The mean flow structure and its resultant effect on flame behavior is discussed. A method to perform mass-weighted averaging of flow variables was developed for direct comparison of turbulent flow properties between experimental measurements and computations. Conditional statistical sampling and length-scale filtering were used to elucidate details of flow-flame interactions as they pertain to sub-grid modeling in large-eddy simulations.
Degree
Ph.D.
Advisors
Lucht, Purdue University.
Subject Area
Aerospace engineering|Mechanical engineering|Optics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.