Simulation of the Relationship between the Current Pattern and Particle Distribution in Electrical Resistance Tomography

抄録

In the present study, cross-sectional images of the assumed particle distribution of a particle-liquid two-phase flow are simulated using a newly proposed algorithm and electrical resistance tomography (ERT) with four current patterns. The first step of the algorithm is a forward problem, which involves calculating the electrical potential on electrodes from the assumed particle distribution by solving the Maxwell equation. The second step of the algorithm is an inverse problem, which involves calculating the particle distribution from the electrical voltage calculated by the forward problem. In the simulation conditions of the four current patterns, namely, opposite, adjacent, one-reference, and trigonometric methods are used to discuss the most accurate current patterns in three types of particle-liquid two-phase flow, namely, annular, stratified, and dispersed flows. The simulation results reveal that the current pattern depends on the flow conditions. In particular, among the four current patterns, the trigonometric method is found to be preferable.