Multiphase Flow Through Chokes - Evaluation of five models for prediction of mass flow rates
MetadataVis full innførsel
This thesis has evaluated five choke models described in the literature by comparing their predictions to measured flow rates for two different data sets. The models are the Bernoulli Equation with two-phase multiplier, Asheim's model, the Sachdeva et al. model, Al-Safran and Kelkar's model and the Hydro Model. Abstract For single-phase flow, the prediction of mass flow rate across a choke for a given pressure drop is a rather straight-forward process. But for two-phase flow, it is more difficult. Many authors have developed models for two-phase mass flow rate predictions, but they are not as accurate as for single-phase flow. Abstract Most emphasis has been given to the Hydro Model. Earlier work shown that there is room for improvement. Therefore, it was divided into parts and each part was evaluated separately. Some improvements were found, but all in all the resulting revised Hydro Model presented here is very similar to the original model. The largest difference is the removal of pressure recovery, which greatly simplifies the model and reduces run time drastically. Abstract It was the Hydro Model that was best for one of the data sets, but worst for the other. The Bernoulli Equation with Simpson et al.'s two-phase multiplier was seen to be one of the best models for both data sets, in spite of its relative simplicity and the fact that it does not separate between critical and sub-critical flow. Abstract It was found that the most important feature seems to be the inclusion of slip between the gaseous and liquid phases. Pressure recovery after the choke seems to be negligible in most cases. For calculation of density for a two-phase liquid, the mometum density appears to be the most suited for use in choke models.