Spatial considerations on electrical resistance tomography measurements

Abstract

This work reports experimental evidence of the effect of certain geometrical parameters on the accuracy of electrical resistance tomography (ERT) measurements in two-phase systems. Emphasis is given to millimetre-size electrodes that suffer from significant electric field distortion due to (a) extension of the measuring volume beyond the electrode plane (fringe effect) and (b) the non-uniform distribution of field strength at the electrode plane. Water constitutes the continuous (conductive) phase, whereas Teflon rods constitute the dispersed (non-conductive) phase. The examined parameters include the diameter of the cylindrical test vessel, the size of electrodes and the number and size (radius and length) of the submerged Teflon rods. The variable in these tests is the axial and radial position of the Teflon rods inside the test cell. It is found that for homogeneously (axially and radially) dispersed rods, the void fraction measured by a set of electrodes at a plane agrees pretty well with 2D theoretical predictions. However, in cases of non-homogeneously dispersed rods, void fraction measurements deviate considerably from theoretical values. This is a manifestation of severe electric field distortion associated with the employed small electrodes. Moreover, evidence is provided that in the examined system, the fringe effect is more significant than the topography of the field strength in distorting measurements. To allow a quantitative analysis of the present data, nonlinear regression combined with dimensional analysis is conducted to derive an expression that describes the void fraction measurements by a set of electrodes at a plane for different axial and radial positions of the submerged rods. A parametric analysis of this expression illustrates the significance of different parameters on ERT measurements.