Fluid Handling Equipment

Background
• The regime by which the fluids moves generally classified as follows: laminar flow, referring to a flow pattern
without significant mixing of the particles, and turbulent flow, which is a random flow pattern and particle
randomly mixed
• To differentiate it is required to introduce the definition of Reynolds number:

• Here V and L correspond to the speed and length characteristics of the fluid, and θ
to the kinematic viscosity of the fluid
• For Reynolds numbers less than 2000 laminar flow regime is present; for higher Reynolds 4,000 turbulent
flow, and between these two values is known as transition regime.
• Applying a fluid energy balance moving along a flow line leading to obtain an expression known as Bernoulli’s
Equation:
• where V1 and V2 are the velocities of the fluid at two specified points, pi fluid pressure, fluid
density ρi and fluid height hi.
• This equation of Newton’s second law should not be confused with the energy equation.
Several simplifications are also included:
• Inviscid flow, i.e., without shear
• Continuous flow
• Constant density
• Balance along a streamline
• Bernoulli’s equation is widely used in studies of fluids. However, care must be taken because
this equation only complies with the restrictions mentioned above. Also not to be confused
with the energy equation, since the Bernoulli results comes from applying Newton’s second
law on a differential of fluid.
• Aside from making a force balance on a particle of fluid, it is possible to perform an energy
balance on the same particle. Omitting the mathematical development of this, the energy
equation is:
• where HP and HT correspond to the added energy and removal system and hL represents the
energy losses in the system. The losses are given by accessories and by friction with the tube;
thus simply brings flowing pressure losses
• if some accessories are added, these losses increase. Loss calculation due to flow is made
based on the following equation:

• This equation is valid for laminar and turbulent flow in both circular and noncircular pipe
• where L is the path length through the fluid, D is the pipe diameter, and f is the friction factor,
which is obtained by Moody diagram or other experimental correlations. Moreover, in the
case of pressure losses due to accessories the equation has the following form:
• where K corresponds to an empirical factor depending on the type of accessory used. All
these correlations can be easily found in any text on fluid mechanics.
• Meanwhile, the simulators are based on the differential expression of the energy balance to
calculate losses through a pipe.

• The first term on the right side of Eq corresponds to the pressure gradient which is as a result
of gravity, where ρm is the fluid density and the inclination angle θ of the pipe.
• The second term corresponds to the losses caused by fluid friction.The last term represents
the angular acceleration component on the pressure drop, and is proportional to the change
in fluid velocity v
• Depending on the situation, either of these terms can be more representative when
calculating the pressure drops through piping (Potter and Wiggert)
• To perform this calculation, a very accurate model should be used for predicting the fluid
density, as this is involved in the calculation of all terms of the equation
• If there is more than one phase, the simulator calculates the density from the fluid holdup in
the pipe, using the same correlation which depends on the operating conditions.
• Similarly to the holdup calculation, the simulator uses different correlations to determine the
quantity of the friction term.
• In this case there is no correlation that successfully meets all conditions, since many of these
have been determined for very specific situations. The following section illustrates some of
the most important correlations available in process simulators.
Aziz, Govier and Fogarasi
• This model consider that the flow rate is independent of phase viscosity and pipe diameter
but it is proportional to the cube root of the gas density
• From this statement, the superficial gas and liquid velocities are modified according to a map
of flow regimes. Once the system is determined, the proper correlation is selected to calculate
the pressure gradient and other parameters as the flow rate
Baxendell and Thomas
• The Baxendell and Thomas model is the result of an extension of the Poettman and Carpenter
model, which is intended to include much larger flows. It is based on a homogeneous pattern
in which the friction factor for the two phases is calculated using a pilot correlation.
• It is appropriate to use this model in situations of horizontal flow; this is in addition to the
vertical flows considered by Poettman and Carpenter model
• Likewise, this correlation does not take into account the different flow regimes which can
occur in a given time, and it is assumed that the pressure gradient is independent of viscosity.
Beggs and Bril
• It is the method used by default in modules in Aspen HYSYS® pipes, due to the versatility and
the calculation restrictions, unlike several of the other models which do not correspond to
systems with very specific conditions.
• The method is based on the work developed for air–water mixture under different conditions;
this module can be applied to inclined flow conditions.
• When using Beggs and Brill correlation, the flow rate is determined using the Froude number
and content of incoming liquid. The flow map is based on horizontal flow systems; regimens
depend on inlet liquid fraction, i.e., the stream quality, and accordingly define four regimens,
namely segregated, intermittently distributed and transition.
• Each of these systems has its own characteristics, as shown below:
 Segregated flow: corresponds to a stratified and annular flow.
 Intermittent flow: rich in viscous liquid.
 Flow transition presents an intermediate behavior between the segregated flow
and intermittent flow.
 Distributed flow: bubbles and mist
Beggs and Bril
• It is the method used by default in modules in Aspen HYSYS® pipes, due to the versatility and
the calculation restrictions, unlike several of the other models which do not correspond to
systems with very specific conditions.
• The method is based on the work developed for air–water mixture under different conditions;
this module can be applied to inclined flow conditions.
• When using Beggs and Brill correlation, the flow rate is determined using the Froude number
and content of incoming liquid. The flow map is based on horizontal flow systems; regimens
depend on inlet liquid fraction, i.e., the stream quality, and accordingly define four regimens,
namely segregated, intermittently distributed and transition.
• Each of these systems has its own characteristics, as shown below:
 Segregated flow: corresponds to a stratified and annular flow.
 Intermittent flow: rich in viscous liquid.
 Flow transition presents an intermediate behavior between the segregated flow
and intermittent flow.
 Distributed flow: bubbles and mist
• Once the flow rate is determined, it proceeds to calculate the fluid holdup to a horizontal pipe
through appropriate correlation. A correction factor is applied if there is any inclination. Then,
with the predicted holdup friction factor for the two phases and finally, the pressure gradient
is calculated
Duns and Ros
• Duns and Ros model is based on a laboratory research about the vertical flow of air,
oil, and water systems. The model uses three flow sections, namely:
 Region I: liquid phase is continuous.
 Region II: liquid and gas phases alternate.
 Region III: gas phase is continuous
• According to the region in which the flow is found, the proper correlation is used. The regions
are distinguished according to two dimensionless groups, one corresponding to the gas
velocity and other to the liquid content. All correlations for this model are defined in terms of
these dimensionless groups.
Gregory Aziz Mandhane Pressure Gradient
• An elaborated model to predict the pressure drop for two-phase flow is used. It makes use of
a flow map to set the flow rate of a stream. After knowing the flow rate, this method uses the
adequate correlations to estimate parameters of the pipe.
Hagedorn and Brown
• The systems data on which the construction of this model was made for upward flow of air or
air–water mixtures or crude. Pressure drop is calculated using a derivative factor Moody chart,
using a Reynolds number for two phases
• Which in case of single phase flow should be reduced to the Reynolds number for this phase.
This model developed a curve in which the void fraction is related to the same dimensional
parameters for Duns and Ros model.

HTFS Models (Aspen Technology Inc. 2009)

• The two HTFS models share calculation method for calculating the pressure gradient. However
they differ in the method for calculating the gradient of static pressure. For the homogeneous
model, the void fraction is assumed to be a homogeneous fraction, while for the other model
for the fraction is calculated. These modules have been validated for horizontal and vertical
flow in both directions.
OLGAS 2000
• This model allows flow calculation in two or three phases. It is one of the most powerful
models at the time of simulating transport operations in the oil and gas industry models. For
calculations, OLGAS 2000 has models for every four schemes, namely stratified annular and
dispersed flow like mud.
• The multiphase flow is a complex process between the different phases involved considering
fluid properties, geometry, and reservoir conditions, the well, pipeline and processing
plant.
• To model this, OLGAS model has the ability to handle up to three-phase flow, and different
substances such as oil, gas, and sand, among others. OLGAS predicts the pressure gradient,
fluid holdup, and flow rate. It has been designed from trials involving horizontal flow to
vertical and inclined flows
Orkisewski
• Orkisewski correlation is a relationship for vertical upward flow based on various methods. In
these four flow regions, different correlations are used according to the rules defined. Among
the correlations used by Orkisewski, the Duns and Ros correlation is included
Poettman and Carpenter
• This model assumes that the contribution of the acceleration term in the overall pressure
drop is negligible and the pressure drop can be calculated using a homogeneous model. Also
assumes that the pressure drop can be estimated using a homogeneous two-phase density.
• In the model, a correction factor for calculating friction factor based on experimental data for
gas ascending wells under varying conditions is proposed. This model assumes that the
pressure gradient is independent of viscosity
Tacite Hydrodynamic Mode
• This module is mainly used to simulate multiphase flow, and for the design and control of oil
and gas pipelines.
• The module provides two options for the calculation: one for gas–liquid and another based on
the correlation of Zuber-Findlay. In this module three regions of flow are identified:
• Stratified: The model assumes a time balance between the phases present in the pipe section.
• Flashing this type of flow is solved as a problem of two regions. The gas phase is considered as stratified
flow, and the flow of liquid is dispersed.
• Sparse: This is a special case of the previous region.
Tulsa (Aspen Technology Inc 2005b, Operation Guide)
• This model proposes a mechanical approach to predict the flow, pressure drop, and fluid
holdup in an upward vertical flow for two phases
• There five regions are included, each with its specific correlations, mostly based on existing
models such as Aziz and Hagedorn method. Tulsa model has been tested in more than two
thousand wells and, in general, it is recognized to get very accurate results
Pumps
• From the energy balance equation, there are two terms corresponding to the added energy
and system withdraws. This energy, expressed in the equation as height, belonging to a given
by a pump or a turbine work taken. This makes it possible to know the work used in both
cases by the following expressions:

• There WP corresponds to the work and η to the efficiency. Similarly, there is an

additional parameter for pump calculation known as the net positive suction head (NPSH).
• The NPSH is related to the difference between the pressure at a given point and the vapor
pressure of the liquid, and is of vital importance because if the fluid pressure at a point is
found to be less than the vapor pressure, a cavitation phenomenon occurs.
• NPSH is defined in two different ways. NPSH required is related to the minimum value to avoid
cavitation given by the manufacturer and the NPSH available, which depends on the liquid and
the characteristics of the system in which it will operate.
• These parameters can be calculated with the following equations:
For NPSH required:

• Where Hz is the minimum at the rotor inlet pressure and v is the velocity
For NPSH Available:
• Where Pa is the suction pressure, the suction height Ha, Pds pressure drop in the suction line
and Pv the vapor pressure at pumping temperature. To avoid cavitation, the NPSH available
must be greater than required.
• In process design, NPSH available is calculated and with a factor of 2 ft (according to design
practice), the NPSH required is specified.