Image reconstruction on electrical resistance tomography measurements in a settling process

This thesis uses Electrical Resistance Tomography (ERT) to measure a settling process. The measurement hardware for this system was developed during previous work carried out by the author. This research is the continuation of this work by focusing on image reconstruction from the measured data. The reconstruction algorithm attempts to find the conductivity distribution of the interior of a rig from measurements made on its boundary. The method used in this thesis is the Newton-Raphson (NR) Algorithm, which employs iterative solutions of the Finite Element Method (FEM) in order to converge to a solution for conductivity. FEM is a numerical simulation method that discretises a region into elements and performs a global optimisation or minimisation in order to arrive at a solution. The NR-method is similar in principle to Newton's method, but is designed to operate on complex data and matrix equations in multiple dimensions. This thesis discusses the methods in reasonable detail and highlights several issues, including ill-conditioning, and regularisation as a method to improve the conditioning of the data. The results contain images of successful reconstruction of a settling process.

Grantor dc:publisher.institution

Department of Electrical Engineering

Year dc:date.issued

2004

Author dc:creator

• Herholdt, Siegfried

Advisor dc:contributor.advisor

• Tapson, Jonathan

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/11427/6768

OAI identifier oai:identifier

oai:open.uct.ac.za:11427/6768