Electrical capacitance tomography with variable diameter and potential application for pole inspection

Abstract

Electrical capacitance tomography (ECT) has been applied in various industrial processes as its merits over other tomography modalities. The reconstructed images can show the permittivity distribution of the cross-section of an object by measuring the capacitance. Because this is first time that ECT is proposed for pole inspection, the conventional ECT needs to be optimised to overcome the related shortages for this purpose. In the PhD study, same key issues are solved including the design of a variable diameter sensor, the adaptive selection of relaxation factor in Landweber iterative algorithm and metallic samples imaging in ECT. The diameter of a conventional ECT sensor is fixed. This study presents a new structure of ECT sensor, which can achieve adjustable inner diameter of the sensor within a certain range, by re- engineering the traditional structure so that the electrode-to-gap ratio (EGR) of the sensor can be changed. The relationship between EGR and image quality was studied by simulation. A prototype ECT sensor with variable diameter was fabricated and evaluated with an impedance-analyser based ECT system. The experimental results show that the ECT sensor with variable diameter can provide similar reconstructed images to a conventional ECT sensor, when the sensor diameter varies in a certain range, i.e. EGR above 6:1. Therefore, this new structure can enable the ECT sensor to adapt itself to the change in diameter of the measured objects. It is crucial to select a suitable relaxation factor in Landweber iterative algorithm for ECT because it affects the convergence and convergence rate. Simulation study shows that the relaxation factor should be selected adaptively according to the sensor structure (i.e. the number of electrodes), permittivity distribution and noise level in capacitance data. With different number of electrodes and four typical permittivity distributions, the relaxation factor and the related convergence are investigated in consideration of the change in relative image error and relative capacitance residual. By adding noises with different levels to noise-free data, their influences on the selection of relaxation factor and convergence are characterized. For a typical permittivity distribution, the corresponding relaxation factor is selected based on the upper bound of all relaxation factors, which are determined by a sensor design. The performance of Landweber algorithm with an adaptively selected relaxation factor is compared with constant relaxation factors and updated relaxation factor, showing that the proposed method can ensure convergence with less computation time than other relaxation factors. ECT has been mainly used to image non-conductive medium. Concrete pole has unique cross- sectional structure, which consists of multiple rebars, concrete and internal air cavity. Therefore, a concrete pole presents both dielectric permittivity and electrical conductivity. An accurate image 15 Abstract cannot be reconstructed using the conventional methods. In this study, some forward models proposed for imaging the metallic materials by other researchers are introduced, and the sensitivity maps based on these forward models are compared. The experiments are conducted to test the single rebars and multiple rebars inside a simulated internal structure of the concrete pole with a 12 electrodes sensor. Then, a simulated defect, i.e. air cylinder, is also tested with the influence of a single rebar. It is found that ECT cannot image the grounded rebars inside the concrete poles, while it can recover the shape the floating rebar and distinguish the multiple floating rebars clearly.

Date of Award	1 Aug 2018
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Hujun Yin (Supervisor) & Wuqiang Yang (Supervisor)