Improved magnetic induction tomography

Tools

Turton, Logan James (2023) Improved magnetic induction tomography. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only until 12 December 2025. Subsequently available to Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (28MB)

Abstract

Imaging through magnetic induction has inherent advantages over rival technologies, including being safe, contactless, and capable of both the location and characterisation of underlying electrical properties. Applications are wide-ranging, varying from geophysical investigations of the sub-surface, through security screening of baggage, to medical imaging. This research aims to improve the quality of image reconstructions from inversion.

A derived set of magnetic fields use orthogonality to maximise the available information. Necessary generalisations of existing forward models then include accounting for magnetometer measurements and inhomogeneous magnetic fields. Adapted sensitivity formulations move to a rectangular matrix with scaling computational advantages as image resolution increases. Well suited to augmentation, this also allows the combination of multiple magnetic profiles into a single inversion. An examination of relevant inverse theory covers appropriate analysis tools and a library of algorithms. Individually algorithms have varying convergence properties, computational requirements, and regularising effects, with the specific choice subsequently problem dependent.

An approach to produce a set of the aforementioned magnetic fields with a single modular system by varying the driving currents is evaluated, alongside a framework to quantitatively assess whether a particular profile is a unique component of that set. Various designs demonstrate the robust nature of the design procedure and the influence of parameters such as component composition, driving current precision, and the regularisation term.

Finally, an evaluation considers the culmination of the preceding work applied to three dimensional inversion in MIT. Increasing the number of spatially unique magnetic fields leads to a demonstrable improvement in image quality, with significant enhancement possible using improved orthogonality. Orthogonality also appears to reduce the computational effort required to reach semi-convergence for multiple algorithms, reducing inversion time. Expanding the number of fields explicitly enables overdetermined problems to be constructed, which show reduced noise sensitivity as the number of orthogonal fields increases. This behaviour holds for various, less ideal conductivity distributions, as well as for low conductivities and higher resolutions. A demonstration also covers the applicability of numerous novel inversion algorithms and the potential associated with optimisation.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Fromhold, Mark
Wilkinson, Paul
Keywords: Magnetic induction; Image reconstructions from inversion; Magnetic fields; Three-dimensional inversion
Subjects: Q Science > QC Physics > QC501 Electricity and magnetism
Faculties/Schools: UK Campuses > Faculty of Science > School of Physics and Astronomy
Item ID: 74388
Depositing User: Turton, Logan
Date Deposited: 12 Dec 2023 04:40
Last Modified: 12 Dec 2023 04:40
URI: https://eprints.nottingham.ac.uk/id/eprint/74388

Actions (Archive Staff Only)

Edit View Edit View