Characterization of Multiphase Distribution in Frequency Spectrum based on Electrical Capacitance Tomography Technique

    Marlin Ramadhan Baidillah1,2, and Masahiro Takei1

    9th International Symposium on Measurement Techniques for Multiphase Flow, Sapporo, Japan, 23rd-25th September 2015

    1Graduate School of Engineering, Div. Artificial System Science, Chiba University, 1-33, Inage-ku, Chiba-shi, Chiba, 263-0022, Japan 2CTECH Labs Edwar Technology, Spectra 23 BC No.10-12, Alam Sutera, Tangerang Selatan, Banten, 15325, Indonesia

    Corresponding author: M.R. Baidillah, E-mail:


    The conducted study is based on the fact that the relative permittivity value of a material depends on frequency capacitance measurement. Due to the complexity of capacitance sensor design of ECT in a multiphase system makes the sensor have several different behavior, i.e. positive and negative sensitivity matrix. Therefore, the measured capacitance in ECT technique is also highly depends on the capacitance sensor geometry. In ECT system, the measured capacitance of an object is normalized between the capacitance of two known relative permittivity medium to predict the object one in a single frequency capacitance measurement. An issue that still existed in a single-frequency measurement is always suffering a changing of capacitance measurement records in overtime, which makes the normalization of predicted value is not accurate and hardly to distinguish the multiphase system distribution which the relative permittivity is quite small different.

    In this study, we use the multi-frequency signal to predict multiphase distribution in frequency spectrum based on ECT technique. ECT sensor which has 12 electrodes in 1 layer was used to measure the multi-frequency sinusoidal signal (from 1 KHz to 5 MHz) of several variances of multiphase distribution (mixed between pellet PMMA distribution er ~ 2.8 to 3 and Air er ~ 1 in frequencies below 5 MHz).

    Although the relative permittivity of mixed phases inside the region of interest is quite small different and tends to be constant in frequency spectrum range 1 KHz to 5 MHz, the results of our study show that the integration between multi-frequency signal and ECT performed the variance of measured capacitance which is highly correlated with the mixed phases distribution. As the conclusion, the mixed phase distribution can be easily distinguished in frequency spectrum based on ECT technique.


    Electrical Capacitance Tomography, Multi-frequency signal measurement