What Do We Really Know About Fluid Dynamics?

By Pamela Boschee 26 Jun 2014

Multiphase flow, flow dynamics and modeling, and flow assurance are familiar terms in the oil and gas industry, but they’re not always well understood. While data-based flow modeling invites confidence, the inherent uncertainty in flow dynamics remains to be quantified.

At a luncheon hosted by the SPE Flow Assurance Technical Section, “Multiphase Flow, Flow Assurance, Dynamic Modeling…Why Bother?”, Sam Kashou, flow assurance consultant at Chevron Energy Technology, emphasized the importance of understanding the uncertainties. Quoting P.J. Plauger, he said, “My definition of an expert in any field is a person who knows enough about what’s really going on to be scared.” Kashou said, “Contrary to popular belief, not all measured data is based on good data. When you’re working with ‘real’ data, you need to ask if it’s reliable and where did it come from.”

Flow assurance covers all methods to ensure the safe and efficient delivery of hydrocarbons from the well to the collection facilities. It is a multidisciplinary activity involving a number of engineering disciplines, including mechanical, chemical, process, control, instrumentation, and software engineering. Kashou said, “The bottom line is to optimize production in an economic and technologically feasible way to make money.”

The variables involved in flow dynamics, such as slugging, gas hydrates, scale, waxes, corrosion, erosion, and asphaltenes, are complex and may interact in unexpected ways. Kashou said, “Production may have surprises that can be costly to overcome. Do we trust our models? In fluid flow modeling, steady state and dynamic simulation is a must in understanding the system performance and operability. Simulations allow for debugging the system dynamics and what to do during upsets. If you can’t simulate, you already have a problem.”

He described the difficulty in basing flow models on the design parameters of facilities. “You have a design basis, build it,” he said. “If something changes, let the flow assurance engineers know, so they can change the model. Often, the flow assurance engineer isn’t told that the facility was not build per specs.” The modeling results will be inaccurate, misleading, and may lead to unnecessary and costly interventions, such as chemical injections.

Kashou identified several areas requiring further study to improve the quantification of uncertainties, which would allow for added confidence in the extrapolation of observed and modeled data to the flow patterns of a production system from input through exit. For example, erosion due to the fines in gas-dominated flows, improved characterization of fluids entering a separator, flow regimes, scaling, and sand transport and accumulation.

In new deepwater applications, “subsea pipelines are pushing the envelope. Multiphase simulations should not be treated as a ‘mature’ technology,” Kashou said. Traditional approaches are not suited to deepwater production because of the extreme distances, depths, temperatures, or economic constraints. The introduction of subsea process equipment between the wells and the flowlines to shore affects the pressure and temperature conditions, the system capacity, and hydrate profiles. 

Flow assurance requires a system approach. In deepwater systems, flow and process models must be combined throughout the production and injection system to diminish the pressure drop between the wellhead and the receiving facilities. The pressure drop is influenced by many parameters in multiphase flow, including, the amount of liquid, length of the flowline, velocity (higher velocities increase the pressure drop), temperature, density (in multiphase flow, the density is a function of the oil/gas/water phases, temperature, and pressure), friction between the flow and fluid in the pipe wall, gravity forces, and valves, bends, and process modules.

Kashou called for wider industry efforts to collaborate and share information on the modeling approaches that worked and did not work. He acknowledged that the knowledge about what did not work is unlikely to be generally shared, even though it would be especially useful in acquiring sufficient data about uncertainties to extrapolate to other systems.

Pam Boschee is the Senior Editor for Oil and Gas Facilities.