• Katrine Arnoldsen Juhl
4. semester, Energiteknik, Kandidat (Kandidatuddannelse)
The purpose of this project was; to investigate the influence of aerated flow in a Coriolis flowmeter based on experiments of two different Coriolis flowmeters, to perform Computational Fluid Dynamics, CFD simulations of the fluid behaviour in a SITRANS FC400 sensor to the Coriolis flowmeter and to achieve an understanding of the theories applied in two-phase flow. The theory of how the Coriolis flowmeter works was explained through the means of Coriolis forces, beam theory and phase-shift graph. For describing the fundamental theory of two-phase flow, flow regimes for horizontal and vertical pipes were examined. The Weisman model and the Mandhane flow regime map were both used. A patented method by which the gas void fraction, GVF is estimated, was examined. Two Coriolis flowmeters, of two different manufacturers, were tested on an Aerated flow test rig provided by Siemens A/S, Flow Instruments (Nordborg, Denmark). During the tests the back-pressure, the air and liquid fluid parameters were specified and the mass flow rate, density and drive level were measured with the Coriolis flowmeter. The measurement error for the fluid mass flow rate and the fluid density increased as the GVF increased. Meter 2 could measure the drive level which increased as both the GVF and the liquid mass flow rate were increased. A CFD simulation was designed in order to see the aerated fluid behaviour in the sensor to the Coriolis flowmeter. The CFD simulation was performed using ANSYS workbench and FLUENT. The mixture model and the k-ω turbulence model were used. Based on the FC400 Coriolis flowmeter the mesh for the flow behaviour simulations was made. A grid convergence study was performed using the Grid Convergence Index, GCI, which is based on Richardson Extrapolation. Three cases were investigated; a steady single-phase case, a transient single-phase case and a transient two-phase case. Recirculation zones in the flowsplitters were first noticed when running the single-phase cases. Three transient two-phase simulations were performed. The inlet parameters changed between a liquid flow rate of 350 kg/h and 3500 kg/h and a GVF of 5% and 10%. There are areas of a higher gas volume fraction present in all the simulations, than are present in the inlet boundary conditions. These areas increased as either the liquid flow rate increased or the GVF increased, which fits well with the theory of decoupling and the quantified errors of measurements. An assessment of the validity of the simulation results without oscillations were made with a crude estimation of the forces impact on the flow. The Coriolis force was at least eight times larger that any other force affecting the fluid. It was concluded that the error seen in the experiment, when increasing the GVF, was not caused by the larger Coriolis force but by the decoupling effect. There was no significant change to the size of the Coriolis force when the fluid was pure water or a mixture with a GVF of 5%. This report contains the first step in improving the measurement accuracy of the Coriolis flowmeter with aerated flow. Several recommendations regarding experiments and simulations are given, these should be seen as the next step in improving the measurement accuracy of the FC430 Coriolis flowmeter.
SpecialiseringsretningTermisk energi og procesteknik
Udgivelsesdato2 jun. 2014
Antal sider92
Ekstern samarbejdspartnerSiemens A/S, Flow Instruments
Nils Basse nils.basse@siemens.com
ID: 198440524