Neural network based multi-criterion optimization image reconstruction technique for imaging two- and three-phase flow systems using electrical capacitance tomography

• Published in: Measurement science & technology Vol. 12; no. 12; pp. 2198 - 2210
• Main Authors: Warsito, W, Fan, L-S
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.12.2001; Institute of Physics
• Subjects: Biological and medical sciences; Computers in experimental physics; Electrical impedance tomography (EIT); Exact sciences and technology; Image processing; Instruments, apparatus, components and techniques common to several branches of physics and astronomy; Investigative techniques, diagnostic techniques (general aspects); Medical sciences; Miscellaneous. Technology; Neural networks, fuzzy logic, artificial intelligence; Pathology. Cytology. Biochemistry. Spectrometry. Miscellaneous investigative techniques; Physics; Multiphase flow; Multicriteria analysis; Neural networks; Optimization method; Tomography; Capacitance; Modelling; Experimental study; Flow diagram; Image reconstruction
• ISSN: 0957-0233; 1361-6501
• DOI: 10.1088/0957-0233/12/12/323

In this work a new image reconstruction technique for imaging 2- and 3-phase flows using electrical capacitance tomography (ECT) has been developed. The technique is based on multicriterion optimization using an analogue neural network combined with the so-called linear back projection technique commonly used for ECT. The multicriterion optimization image reconstruction problem is solved using Hopfield model dynamic neural network computing. The technique has been tested on a capacitance data set obtained from simulated measurement as well as experiment using a 12-electrode sensor. The performance of the technique has been compared with iterative linear back projection and the simultaneous image reconstruction techniques for 2-phase system imaging and has shown great improvements in accuracy and consistency. The technique has also shown the capability of 3-phase image reconstruction with high accuracy. (Original abstract - amended)