Evaluation of Multiphase Flow Rate Models for Chokes Under Subcritical Oil/Gas/Water Flow Conditions

Abstract

Summary Knowing the mass flow rate is important in relation to production control in the oil and gas industry. If the change in pressure and temperature across a choke can be correlated with the mass flow rate, the actuator position, and the properties of the well stream, this may constitute a mass flow rate meter that is both simple and inexpensive compared to other designs. However, knowledge of the predictive ability and accuracy of available mass flow rate models is required to qualify such a solution. Predictions from several mass flow rate models are compared with data from a crude oil/natural gas/water system at pressures varying from 8 to 16 bara. The fluids used were recombined oil from the Njord field in the North Sea, natural gas from the Kaarstoe terminal in Norway, and water with added salts to give typical produced-water properties. Two different choke geometries (orifice and cage type) were tested for three different opening areas. The experimental results are compared with eight mass flow rate models for multiphase flow through chokes. These are the two Hydro models originally developed by Selmer-Olsen, the Sachdeva et al. model, the Perkins' model, and four two-phase multiplier models—the Morris, the Chisholm, the Simpson, and the homogeneous equilibrium model (HEM), respectively. For the orifice-type geometry, the Hydro short model predicted the results most accurately. For the cage-type geometry, the Hydro long model, which includes losses in the choke geometry, predicted the results most accurately. A modification to the slip model improves the results of the Hydro models, predicting all the 367 test points with a standard deviation of 7.8%. The average error of absolute values was 5.8%.