Register or Log In

International Society for Industrial Process Tomography

8th World Congress on Industrial Process Tomography

Investigation of multiphase flow blind-T mixing effectiveness using high speed gamma-ray tomometry

R. Maad1, S-A. Tjugum2 and A. Hallanger3, B. T. Hjertaker1*

1Dept. of Physics and Technology, University of Bergen, P.O. Box 7803, NO-5020 Bergen, Norway.

2Roxar Emerson Process Management AS, Kokstadveien 23, NO-5257 Kokstad, Norway

3Christian Michelsen Research AS, P.O. Box 6031, NO-5892, Bergen, Norway.

*bjorn.hjertaker@uib.no

ABSTRACT

Most multiphase hydrocarbon flow instrumentation is flow regime dependent, i.e. the accuracy of the instrument depends on the cross sectional spatial distribution of the flow components. A common way to limit the influence of the flow regime dependency is to emulate homogeneous flow by installing a horizontal blind-T (blind-tee) flow obstruction just upstream of the vertically installed flow instrumentation. The objective of this research has been to investigate the effectiveness of multiphase hydrocarbon flow mixing by blind-T using a high speed gamma-ray tomograph developed at the Department of Physics and Technology, University of Bergen, Norway. The effectiveness of the blind- T flow obstruction is crucial for securing the measurement accuracy of the flow instrumentation. It is therefore of great interest to validate the effectiveness of the flow mixing caused by the blind-T as a function of vertical position. Possible deviation from homogeneous flow can to some degree be compensated for by more advanced flow regime interpretations which require higher computing power in the instrument flow computer. The experimental results obtained from the multiphase flow experiments conducted in the flow rig facility at Christian Michelsen Research AS, Bergen, Norway, demonstrates deviation from axisymmetric gas/ liquid flow distribution just after the blind-T, in addition to an increasing degree of non-homogeneous cross-sectional flow distribution as a function of vertical position from the blind-T. The experimental data obtained using the high speed gamma-ray tomograph has also been verified using CFD (Computational Fluid Dynamics) simulations of the blind-T mixing scenario.

Keywords Blind-T mixing, computational fluid dynamics, flow regime dependency, high speed gamma-ray tomography, multiphase flow mixing.

Sign-in to access the full text