Page 231 - Introduction to Petroleum Engineering

218 UPSTREAM FACILITIES

Dispersed bubble

Intermittent
Superficial
liquid Annular
velocity
Stratified
Stratified wavy
smooth

Superficial gas
velocity
FIgURE 11.8 Illustration of a flow pattern map. (Source: Brill and Arirachakaran (1992).)

for kinetic energy, and

v
p f mm 2 (11.30)
L fric 2 D
for friction. The subscript m attached to variables on the right‐hand side of
Equations 11.28 through 11.30 denotes that the associated variable is calculated
for a mixture.
Models based on mixture variables are called homogeneous models. In addition
to homogeneous models, two other approaches are often used: empirical correlations
and mechanistic models. Empirical correlations depend on fitting experimental data
and field data to models that contain groups of physical parameters. The empirical
correlations approach can yield useful and accurate results quickly, but does not
provide a scientific basis for extrapolation to significantly different systems. By
contrast, mechanistic models are based on physical mechanisms that describe all
significant flow mechanisms. Modern mechanistic modeling still requires some
empiricism to determine poorly known or difficult to measure parameters.
Homogeneous models do not account for slip between fluid phases. Drift‐flux
models are designed to model slip between fluid phases flowing in a pipe, as well as
model countercurrent flow. Countercurrent flow is the movement of heavy and
light phases in opposite directions when fluid flow is slow or there is no net fluid
flow in the pipe.

11.6 WELL PATTERNS

The effectiveness of a hydrocarbon recovery process depends on many factors,
including factors that are beyond our control, such as depth, structure, and fluid type.
Other factors that influence efficiency can be controlled, however. They include the