A fast pseudo-composition model is put forth to accurately predict the gas mass fraction (GMF) of production fluid. This model accurately predicts retrograde behaviour characterized by a region in PVT space where decreasing pressure increases liquid fraction. While tuned to normal operating conditions, GMF predictions are within 2% of PVTSim when the system deviates from these conditions.
It was shown by Dhoorjaty et al. [1] that a good approximation to the GMF can be made by assuming that a production fluid with many components can be modelled as consisting of only two components which are "heavy" and "light". The heavy component has a low vapour pressure, high boiling point, and the light component has a high vapour pressure and low boiling temperature. The model proposed in [1] was applicable to black oil. Here the model is extended to wet gas pipelines and to production fluid that exhibits retrograde behaviour. This behaviour is characterized by a non-monotonic change in quality with respect to pressure so that in the retrograde region the amount of liquid increases with decreasing pressure. A simulation is required that can track compositional changes downstream of a separator for both the vapor and liquid streams as well as mixing of different production lines as illustrated in Figure 1. The operational methodology involves a separation between the wells and the production pipelines in order to operate a gas pipeline network as well as a liquid pipeline network. The conditions of the separation are controlled by the dynamically changing pressure and temperature caused by transient operations. As a result, the compositions into the vapour and liquid pipelines will vary constantly throughout any transient operation. The requirement from the client was to create a full network simulator that includes the wells, separators and production pipelines.
