Study on Dual-Frequency Imaging of Capacitively Coupled Electrical Impedance Tomography: Frequency Optimization

Abstract

In this work, a new dual-frequency imaging framework of capacitively coupled electrical impedance tomography (CCEIT) is presented. Unlike conventional singlefrequency imaging, the dual-frequency imaging adopts two different working frequencies to obtain the real part and the imaginary part of the impedance respectively. With the real part image and the imaginary part image at the two frequencies, the framework further introduces image fusion to obtain the fused image. This work focuses on optimization of the two working frequencies. To achieve the optimal selection of the two frequencies, data collection in a wide range frequency is carried out. The multifrequency data is then analyzed in depth and investigated from several aspects, including the measurement data, the sensitivity distribution and the imaging quality. Experiment was carried out with a 12-electrode CCEIT sensor, an impedance analyzer, and a computer to obtain the real part and imaginary part measurements of the impedance. Research results show that a low working frequency is recommended for the real part, while a relatively high working frequency is recommended for the imaginary part. Within the investigated frequency range of 200 kHz -20 MHz, the optimized frequencies for the real part and the imaginary part are 1 MHz and 15 MHz respectively. Results of verification experiment shows the superior performance of the two selected frequencies. Additionally, this paper demonstrates the advantages of the dual-frequency imaging framework. Compared with CCEIT in individual frequencies, CCEIT based on dualfrequency imaging with the two optimized frequencies has much better imaging performance.

A new image reconstruction strategy for capacitively coupled electrical impedance tomography

“…CCEIT takes the 1-electrode excitation and 1-electrode detection strategy, i.e., for each impedance measurement, one electrode is selected as the excitation electrode and one electrode is selected as the detection electrode, and other electrodes are kept floating [21]. In a measurement cycle, a frame of data has 66 independent impedance measurements, including 66 real part measurements and 66 imaginary part measurements.…”

“…When an AC voltage source is applied to e4, an output current can be obtained at e10 for the impedance measurement. The equivalent impedance measured by the electrode pair can be described as [20][21][22]:…”

“…The sensitivity distributions in figure 3(a) are obtained under the conductive liquid background and those in figure 3(b) are obtained under the non-conductive gas background. To quantitatively evaluate the sensitivity distributions, the average sensitivity sa and the uniformity coefficient su are introduced, and the calculations are as follows [21]: where l is the total number of positive sensitivities in the six distributions. vn and s n a are the standard deviation and the average sensitivity of the sensitivities in the nth sensitivity distribution, respectively.…”

“…In the past several years, significant progress has been achieved in CCEIT [15,[20][21][22]. It has been verified that utilizing both the real and imaginary parts of the fluid impedance makes sense in improving the imaging performance [23][24][25][26].…”

“…That means the problems resulting from contact measurement in the traditional ERT no longer exist in CCEIT. Meanwhile, inspired by the research results that both the real part and imaginary part of the fluid impedance contain useful information [17][18][19][20][21][22], CCEIT regards the gas-liquid two-phase flow as a complete impedance and uses both the two parts for imaging. Therefore, CCEIT has attracted much attention from researchers since it was proposed and has shown great potential in industrial applications [8][9][10][20][21][22].…”