Schlieren 3D-CT Reconstruction of Instantaneous Density Distributions of Spark-Ignited Flame Kernels of Fuel-Rich Propane-Air Premixture

Autor: Yuta Ishiko, Yojiro Ishino, Kazuma Kakimoto, Yu Saiki, Kimihiro Nagase, Naoki Hayashi, Ahmad Zaid Nazari
Rok vydání: 2016
Předmět:
Zdroj: Volume 1: Heat Transfer in Energy Systems; Thermophysical Properties; Theory and Fundamentals in Heat Transfer; Nanoscale Thermal Transport; Heat Transfer in Equipment; Heat Transfer in Fire and Combustion; Transport Processes in Fuel Cells and Heat Pipes; Boiling and Condensation in Macro, Micro and Nanosystems.
DOI: 10.1115/ht2016-7423
Popis: For 3D observation of high speed flames, non-scanning 3D-CT technique using a multi-directional quantitative schlieren system with flash light source, is proposed for instantaneous density distribution of unsteady premixed flames. This “Schlieren 3D-CT” is based on (i) simultaneous acquisition of flash-light schlieren images taken from numerous directions, and (ii) 3D-CT reconstruction of the images by an appropriate CT algorithm. In this technique, for simultaneous schlieren photography, the custom-made 20-directional schlieren camera has been constructed and used. This camera consists of 20 optical systems of single-directional quantitative schlieren system. Each system is composed of two convex achromatic lenses of 50 mm in diameter and 300 mm in focal length, a light source unit, a schlieren stop of a vertical knife edge and a digital camera. The light unit has a flash (9 micro-sec duration) light source of a uniform luminance rectangular area of 1 mm × 1 mm. Both of the uniformity of the luminosity and the definite shape are essential for a quantitative schlieren observation. Sensitivity of the digital cameras are calibrated with a stepped neutral density filter. Target flames are located at the center of the camera. The image set of 20 directional schlieren images are processed as follows. First the schlieren picture brightness is shifted by no-flame-schlieren picture brightness in order to obtain the real schlieren brightness images. Second, brightness of these images is scaled by Gladstone-Dale constant of air. Finally, the scaled brightness is horizontally integrated to form “density thickness images”, which can be used for CT reconstruction of density distribution. The density thickness images are used for CT reconstruction by MLEM (maximum likelihood-expectation maximization) CT-algorithm to obtain the 3D reconstruction of instantaneous density distribution. In this investigation, the “density thickness” projection images of 400(H) × 500(V) pixel (32.0 mm × 40.0 mm) are used for 3D-CT reconstruction to produce 3D data of 400(x) × 400(y) × 500(z) pixel (32.0 mm × 32.0 mm × 40.0 mm). The voxel size is 0.08 mm each direction. In this investigation, the target flame is spark-ignited flame kernels. The flame kernels are made by spark ignition for a fuel-rich propane-air premixed gas. First, laminar flow is selected as the premixed gas flow to establish the spherically expanding laminar flame. The CT reconstruction result show the spherical shape of flame kernel with a pair of deep wrinkles. The wrinkle is considered to be caused by spark electrodes. Next turbulent flows behind turbulence promoting grid is selected. The corrugated shape flame kernel is obtained. The schlieren 3D-CT measurements are made for the complicated kernels. CT results expresses the instantaneous 3D turbulent flame kernel shapes.
Databáze: OpenAIRE