Identification of flow patterns for coarse particles transported in a non-Newtonian carrier using electrical resistance tomography

Abstract

Flow features provide considerable guidance for the rational selection of techniques to predict hydraulic behaviour and for suitable operating conditions for pipelines. Traditionally, water was used to transport coarse particles, and it was necessary to operate at velocities at which the flow was turbulent in order to avoid blockage. Consequently the friction losses were too high for economic operation. In addition, wear on pipes, fittings and pumps presented serious problems. Nowadays, it is well established that it is possible to operate at very high solids concentration in a heavy vehicle (carrier fluid). Similar solids throughputs may be achieved at very much lower velocities by operating in the laminar flow regime. This results not only in lower power requirement, but it also reduces wear and water consumption. In spite of these potential benefits, only a few studies dealing with the transport of coarse particles in heavy media have been reported. Since the distinction between different flow patterns is of paramount importance for modelling purposes, as equations are flow pattern dependent, and given the importance of avoiding excessive wear of pipes at low and high velocities, the present work was carried out in the context of dense or non-Newtonian carrier fluid. This project comprised analysis of existing data acquired at the Flow Process and Rheology Centre of the Cape Peninsula University of Technology. Kaolin in the range of 6% to 15% volumetric concentration was used as a carrier fluid and coarse material in the range of 10% to 30% volumetric concentration was simulated by silica sand ranging in size from 1 mm to 3 mm. For the purpose of this study flow patterns derived from resistance curves for various mixtures, particle concentrations, particle grading and flow conditions were compared with “concentration profiles” and images obtained from electrical resistance tomography (ERT). It appeared from this work that the sand concentration does not change the flow pattern but increases or reduces the pressure gradients depending on the case. The concentration of kaolin carrier can change the flow patterns from layered to homogeneous flow, inducing an increase in total pressure gradients as it increases. Flow patterns obtained from ERT compared reasonably well with those derived from pressure gradients profiles. The transition velocities from layered to heterogeneous flow obtained from both methods were similar, especially for low and moderate carrier concentrations. As the kaolin carrier concentration or as the sand concentration increased it became more difficult to distinguish the transition velocity between heterogeneous and layered flow. More work is still needed to improve the ERT instrument and its image reconstruction software.