Experimental design to reduce inductive coupling in spectral electrical impedance tomography (sEIT) measurements

Abstract

Summary Spectral electrical impedance tomography (sEIT) is a promising method to image the subsurface complex resistivity distribution in a broad frequency range (i.e. mHz to kHz). However, inductive coupling between cables is known to affect measurement accuracy for frequencies above 50 Hz. Previous studies have proposed correction methods, but these have not been widely adopted yet. In this study, we evaluated the influence of inductive coupling on the measured complex impedance for different electrode and cable configurations. We propose a novel index to evaluate the inductive coupling strength and use it to develop a filter that selects data with limited inductive coupling. In a first step, the inductive coupling strength of a fan-shaped and parallel cable layout were evaluated. It was found that the fan-shaped layout provided more measurements with low inductive coupling strength. Using a synthetic modeling study with a fan-shaped cable layout, we then showed that it is possible to achieve good inversion results without data correction if measurements with high inductive coupling strength are filtered out before inversion. In a final step, we use the novel filtering approach based on inductive coupling strength with actual surface sEIT measurements. The EIT inversion results based on the filtered data corresponded well with inversion results using data corrected for inductive coupling and also showed good spectral consistency. It was concluded that it is possible to achieve reliable inversion results without data correction for inductive coupling when a fan-shaped layout and configurations with sufficiently low inductive coupling strength are used.