This paper reviews the physical and practical considerations that need to be addressed to develop an accurate and robust computational model of the transportation of multiphase fluids across large-scale pipelines. It then demonstrates the accuracy of the Real-Time Simulation System, which was designed according to the scientific principles that are laid out in the paper. To demonstrate the accuracy of the model, computed pressures, temperatures, and mass flow rates along a commercial pipeline are compared against experimentally measured values.
Pipelines in oil and gas production systems normally transport multiphase mixtures of oil, water, and gas. Simulation of complex multiphase flow processes is an important aspect of efficient oil field operation. Safety and optimization rely on a detailed understanding of multiphase flow behavior.1
The multiphase flow model that is built into the simulation software employs conservation laws and well-established empirical formulas to simulate the changes that occur to the mass flow rate and thermodynamic properties (pressure, temperature, etc.) of single or multiphase fluids that are transported across industrial scale pipelines.2,3 A diverse set of equations of state are accessed by the model to simulate a wide variety of materials and to ensure that the model can handle the difficulties that arise from fluid compression and phase transitions.4,5,6 When modeling the transportation of water/steam mixtures (that do not include contributions from oils or other gases), the multiphase flow model also determines the change that will occur to the quality of the steam (i.e. the composition of the water/steam mixture).
Multiphase flow processes involve a combination of materials having different properties and often exhibit relative motion among the phases.1,2 The multiphase flow model is designed to predict the fluid properties, mass flow rate ?, pressure P, and temperature T of a fluid at different distances along the pipeline x and at different moments in time t. In order to make those predictions, the initial mass flow rate and composition (relative amounts of oil, water, and gas) of the fluid must be specified. Values of the key physical properties (density, viscosity, etc.) of the fluid must also be identified either by using empirical formulas4-6 to calculate those values or by specifying them as input parameters.
Once the necessary input parameters have been specified several mathematical algorithms work in consort to determine the changes to ?, P, and T. The specific algorithms that are used at a given point of the pipeline are chosen to account for the dominant forces that effect the properties of the fluid at that point.