Fill-level measurement using capacitance sensors with adaptively configurable electrodes for resolution enhancement

• Published in: Measurement science & technology 2019-04, Vol.30 (4), p.45101
• Main Authors: Kandlbinder-Paret, Christoph, Fischerauer, Alice, Fischerauer, Gerhard
• Format: Article
• Language: English
• Subjects: adaptivity; capacitance; configurable; ECT; electrodes; fill-level; resolution
• ISSN: 0957-0233; 1361-6501
• DOI: 10.1088/1361-6501/ab0486

It is of crucial interest to observe fluids and measure their properties in the field of process engineering. Capacitive measurement techniques have the advantage of being low-cost and having the ability to distinguish between higher-permittivity liquids such as water and gases such as air. A possible application for this technique is the non-invasive and non-intrusive determination of the water fill level in a pipe. In this paper we report on the design, implementation, and test of such a capacitive measurement system and on the finite element (FE) modeling of the system and typical application cases. The system consists of electrodes that are mounted around the circumference of a polycarbonate pipe. These elementary electrodes can be combined electronically via software to form larger 'synthetic' electrodes. The capacitances between the electrodes are measured to obtain information about the phase distribution inside the pipe (liquid or gaseous). While the spatial resolution depends on the size of the configurable electrodes, it is possible to first obtain coarse fill level information with larger electrodes and then obtain refined spatial resolution with a small-electrode configuration in a smaller region of interest. The disadvantage of using smaller electrodes is that the measurement time increases. The system was tested on a test bench with a pipe in which the liquid could be made to slosh by axial movements. The results of capacitance measurements in a 16-electrode system were compared to FE simulation results in Ansys. The results of the simulation were used to create a characteristic curve for the fill level as a function of the capacitance of every electrode combination. The fill level change was also observed optically to identify the most appropriate electrode combinations to fit the fill-level change best. An algorithm was developed to choose suitable 'essential' electrode combinations for adaptive spatial resolution refinement. In the end, 28 capacitance measurements were needed to observe the sloshing in a coarse way, followed by 56 measurements for resolution enhancement. The straightforward approach based on the 16 elementary electrodes would have required 120 measurements. It is also shown that the resolution enhancement using the 56 electrode combinations can determine the fill level more accurately than the standard approach using the information from all 120 electrode combinations.