Novel ERT sensor structures for 3D imaging of mixing process
Z. Ren and T.L. Rodgers
School of Chemical Engineering and Analytical Science, University of Manchester, Oxford Road,
Manchester, M13 9PL, United Kingdom email@example.com
Electrical Resistance Tomography (ERT) has been widely applied in various two-phase flow measurements to extract flow information, such as mixing time and concentration. Current research into ERT on flow measurement focuses on real-time 3D imaging, which provides more comprehensive flow information of entire vessel rather than 2D images providing cross-sectional information only. However, current 3D images either taking full measurement of multiple planes or measure each plane separately as the 2D imaging system does. The former method requires longer time for data acquisition and may result in unbalanced sensitivity between different electrode planes, while the latter suffers from the information lost between each electrode plane (along z-direction). This paper introduces novel ERT sensor structures to solve these problems. The first sensor is based on a circular tank with 4 planes of 16 electrodes on a vessel wall with an extra 8 electrodes at the base. The sensor allows utilising a zigzag electrodes configuration with the measurement strategy for each zigzag plane the same as 2D measurement strategy. The second circular sensor is a multi-linear array system for sensing inside a vessel. The performance of these circular sensor structures is compared to the conventional 2D sensor with single plane measurement strategy. The performance of a linear sensor is also provided in this paper. A linear sensor with 4 planes of 8 rectangular electrodes is used to find out the best electrode combination for 3D imaging. The discussion into the linear sensor includes the performance of different electrode combinations, i.e. 2 planes of 16 electrodes and one plane of 32 electrodes, and the comparison of their performance to the conventional 4 planes of 8 electrodes. Simulation models have been built in a MATLAB based ERT toolkit – EIDORS. The experimental data are taken by ITS P2000 and processed in EIDORS for image reconstruction, using an L-curve Tikhonov method for solving the ill-posed inverse problem. The results show that the novel sensors have good performance to provide 3D information compared to the conventional structures.
Keywords 3D ERT, zigzag, linear, L-curve Tikhonov, EIDORS
Copyright © International Society for Industrial Process Tomography, 2016. All rights reserved.