6th World Congress on Industrial Process Tomography
Threedimensional Tomographic Reconstruction for Analysis of Droplet Behaviors from Electrohydrodynamic Jet
Chao Chen1, YongJae Kim2, SooHong Lee1, SeungBae Kim1 and Han Seo Ko1
1 School of Mechanical Engineering, Sungkyunkwan University, Suwon, Korea, hanseoko@skku.edu
2 School of Information and Communication Engineering, Sungkyunkwan University, Suwon, Korea
ABSTRACT
Droplet behaviours ejected from an electrohydrodynamic (EHD) jet have been analyzed by performed experiments initially. Since the droplet shape is not axisymmetric and the droplet does not go straight, threedimensional tomography techniques have been developed as an effective tool in order to investigate the size, location, and shape of the droplets in this study. Computersynthesized phantoms for liquid meniscus with one and two small droplets have been reconstructed by four threedimensional tomography methods such as an algebraic reconstruction technique (ART), an adaptive algebraic reconstruction technique (AART), simultaneous iterative reconstruction technique (SIRT) and a multiplicative algebraic reconstruction technique (MART) to obtain accurate results for the droplet movements using the multiple projection angles of view. Four different basis functions including cubic Bspline basis function, cubic cosine basis function, cubic oMoms basis function and keys basis function have been adopted to improve the performance of the developed tomographic reconstructions. From comparison of the four tomography results, the MART method with the cubic cosine basis function provides a significant improvement in the reconstruction accuracy for analysis of the droplets. In addition, the four different tomography methods with the aforementioned basis functions have been confirmed using experimental projection data obtained from two high speed cameras in two projection angles of view.
Keywords Droplet, Electrohydrodynamic, Tomography, Basis function
Sign-in to access the full text
Copyright © International Society for Industrial Process Tomography, 2010. All rights reserved.