Chapter 1 - Introduction

1.1. Background
Pipeline networks are systems with hundreds or thousands of kilometers of
pipes and production facilities, storage and distribution centers, compression
stations, and many other devices like valves and regulators.
These types of systems work at high pressures and use compression stations and
pumps to supply to the oil enough energy to be moved along long distances.
For a mature oil field gas and liquid compositions tends to change due to the aging
of the field because water cut will increase thus may increase the artificial lift
requirement. Also the smaller accumulations nearby tend to be developed as
satellite to the main facilities.
Analyzing the entire upstream supply chain in an integrated manner, compared to
looking at the individual elements, is becoming increasingly important in the
current business, because the system bottlenecks could shift from year to year
(Thijssen and Mittendorff 2007).
Network analysis will look at the impact of changes on the entire network over the
entire time period with the aim to determine the optimal timing of developments
of new fields, to identify bottlenecks in the network and evaluate options for
removal of these bottlenecks. This analysis can also be used to underpin
investment decisions, to help optimize the product slate and analyze trade-offs
between, for example, energy efficiency, production and overall recovery. Overall
it supports the activity known as strategic or investment planning.
Modeling approach is often used in energy master planning studies, where the
focus is on the commercial value of oil & gas and related value streams in the
market while balancing the costs of production, transport, processing and storage.
It provides a structured framework for analysis across the whole energy value
chain in order to focus on the key business opportunities.
The same approach can be used for a smaller section of the energy value chain,
for example when the focus is only on the upstream sector. In this sector all
individual elements in the oil & gas upstream supply chain are typically modeled
in isolation using simulation programs. Analyzing the upstream supply chain in an

1
integrated manner, compared to looking at the individual elements, is becoming
increasingly important in the current business, because the system bottlenecks
could shift from year to year (Lasschuit and Thijssen 2004).
For these integrated studies, one typically considers the oil & gas supply chain
starting from wellhead platforms via multi-phase pipeline to a production
platform, then through a network of pipelines to the export facilities.

1.2. Problem Statement

Hamra field is facing challenges to sustain its production, which is decreasing
annually.
Flow lines network plays important role in delivering the production from
wellhead to Field Processing Facility (FPF).Bottlenecks in pipeline can cause rise
in wellhead pressure, which can have a very strong impact on production
sustainability.

1.3. Research objectives

1. To build physical model and compare simulation result with field data
2. To identify production bottlenecks and constraints
3. To optimize production from the networks

1.4 HAMRA FIELD BACKGROUND

Hamra field is located in block 4 & operated by Greater Nile operating company
(GNPOC) which has concession in Western Upper Nile area includes the field
from sedimentary basin of Muglad in interior Sudan. This Basin is characterized
by thick clastic sequence of cretaceous and tertiary age. The depositional
sequence includes thick lacustrine shale and clay stones, flood plain clay stone,
and lacustrine, fluvial, and alluvial sandstone and conglomerates.
These lacustrine clay stones deposited in a suboxic environment provide good oil-
prone source rocks. Reservoir sandstone has been found in wide variety of no
marine sandstone faces. Tectonic activity created formation of several deep fault
bounded troughs, major interbasinal highs and complex basin flanks. Thus variety
of structure has been created and many of them have hydrocarbon traps.
Hamra field consist of 38 wells connected to four oil gathering manifolds.
Most of the wells are completed in multiple formations and being produced

2
commingled. These formations have wide variation in reservoir properties, oil
type and pressure regime.

3
Chapter 2 - Literature Review

2.1. Modeling approach

Recent studies confirm that network modeling approach is often used in energy master
planning studies, where the focus is on the commercial value of oil & gas and related
value streams in the market while balancing the costs of production, transport, processing
and storage (Lee & Thijssen, 2007), It provides a structured framework for analysis
across the whole energy value chain in order to focus on the key business opportunities.
For these studies, one typically considers the oil & gas supply chain starting from
wellhead platforms via multi-phase pipeline to a production platform, then through a
network of pipelines to the export facilities.
The aim of the network analysis is to help determine the optimal timing of developments
of new fields, to identify bottlenecks in the network and evaluate options for removal of
these bottlenecks. This analysis can also be used to help underpin investment decisions,
optimize the product slate and to analyze trade-offs between, for example, energy
efficiency, production and overall recovery (Thijssen and Mittendorff 2007).
In 1995 a published literature showed that Bottlenecks that arise in the upstream supply
chain can be a result of:
1 Field composition (gas-liquid) changing over time and increased water content in the
oil-water liquid stream due to ageing of fields.
2 Increasing use of existing infrastructure for the implementation of new projects to
compensate the production of ageing fields or to increase overall production.
3 New projects that come on stream typically will use part of the existing facilities,
which may have sufficient capacity to handle increase of production from one
individual project, but will not be adequate for other projects (Litvak and
Darlow1995).
The model helps to optimize flow and production of oil and gas, between wellhead
platforms and demand locations over the defined time period given the infrastructure
Production, Pipelines, and Compressor constraints. The economic analysis converts
output from the program into analysis of individual assets and scenarios based on costs,
capabilities and prices.
The model can be split into four key modules. These are the Supply Module, which
include production profiles for each existing and future wellhead platform (the so-called
technical potential); the Processing Module describing production/processing facilities &
constraints; the Demand Module, which covers any demand point (oil & gas demand,
fuel- and re-injection gas requirement, etc); and the Interconnection Module, which ties
together supply, processing and demand to account for pipeline or distribution constraints
and capacities. General data input is required of supply sources (wellhead platforms),
production and processing facilities infrastructure and capacities (processing platforms,
compressors, pumps etc.), transportation capacity (pipelines), costs of processing
(operating cost and fuel gas consumption), and other business constraints in production,
processing and distribution (e.g. minimum and maximum demands, no venting after a
specific year).
The model can help to optimize the entire network over the entire time horizon. The
objective function is based on the variables in the system and is typically the Net Present
Value of the profit margin over the time-horizon, based on the GOR & Water cut of the
respective wellhead platforms and the Unit Production Cost.

The profit margin is in general the revenues from oil and gas sales minus fixed and
variable operating costs (including cost of venting/carbon dioxide) minus investment
costs (as illustrated above).
This means that the model will only drive towards producing oil and gas if the value of
the oil and gas exported is greater than all associated costs (operating, fixed and capital
expenditures). Depending on the oil, gas and water content of the production from a
wellhead platform and the constraints in the entire network, the model will maximize
production from certain wellhead platforms. For example, when the bottleneck occurs
only in the oil pipelines and the driver is the oil price the model will try to maximize oil
production from those wellhead platforms, which have relative low water content. If there
is also gas export restrictions in the network the model will try to maximize the wellhead
platforms that have relative low water content and a low gas/oil ratio (GOR), because it

5
does not want to produce the gas with the oil as the gas evacuation system might
constrain the oil production (Lee & Thijssen,2007).

2.2 Surface pipeline network Model

In most simulations, well production rates are constrained by limited oil, gas, and water
handling capacities of separator banks and gas plants as well as pressure limits in
separation facilities. Therefore, integrated reservoir, well tubing string, and surface-
pipeline modeling is required for accurate prediction of well rates from reservoir and
facility constraints.
The surface pipeline-network model simulates steady-state multiphase flow in well
tubing strings from wellbores to wellheads, well chokes, and pipeline systems from
wellheads to separator banks. It also determines well artificial lift rates and well
connections to different pressure systems. Different elements of the surface-pipeline-
network model are described below.

2.2.1 Well Inflow Performance.

Molar rates of the hydrocarbon components and water for production wells are
determined by the reservoir model as functions of well bottomhole pressure and grid
block properties saturations, compositions, and pressure for grid blocks with well
perforations(Litvac 1993). a Newton-Raphson iteration procedure is applied to obtain the
solution of the nonlinear reservoir flow equations at each time step. If the values of the
grid-block variables from a previous Newton-Raphson iteration are used, the well inflow
relationships can be represented as functions of the bottom hole pressure only
………………………………………………………………………… (2.1)

……………………………………………………………………….. (2.2)

2.2.2 Tubing Strings.

Different methods available in the petroleum industry ~e.g., Hagedorn and Brown are
applied for pres-sure gradient calculations in well tubing strings. In these methods, the
pressure distribution along the tubing is calculated from the solution of the following
ordinary differential equation (Hagedorn, and Brown1965)
6
 …………………………………………………………………. (2.3)

To evaluate the pressure gradient on the left-hand side of Eq. (2.3), the density, viscosity,
and in-situ velocities of oil, gas, and water phases need to be determined. The equation of
state is used to obtain the oil and gas compositions and their densities as a function of the
pressure, temperature, and molar rates of the hydrocarbon components. Oil and gas
viscosities are then calculated using viscosity correlation ~e.g., the Lohrenz-Bray-Clark
correlation (Lohrenz and Clark 1964).

2.2.3 Well and Flowlines.

Production wells are tied to separator facilities by means of a surface-pipeline-network

system. Different correlations available in the petroleum industry ~e.g., Beggs and Brill
are applied for pressure gradient calculations in the flow-lines. Therefore, the pressure
distribution along each flow line is calculated from the solution of the ordinary equation
similar to Eq. (2.3).

2.2.4 Chokes.

The Perkins method is applied for modeling of critical and subcritical multiphase flow in
well chokes. Look-up tables are used to correlate choke setting to internal diameter.

2.3 Coupling of reservoir simulators with surface

networks
The design of a multiphase flow gathering network requires an estimate of flow potential
from each well. Rarely is the aerial extent of the reservoir so well defined at the outset
that all potential well locations can be identified. The flow potential of the fluid phases
produced up the wellbore into the gathering network vary between wells depending upon
the characteristics of the reservoir, and pressure depletion or “drive” mechanism of the
reservoir. Accordingly, the efficient transport of reservoir fluids through the gathering
network is usually difficult to predict over the producing life of the wells. (Ghorayeb,
2003).
The forecast of production rates from wells using a numerical reservoir simulation model
is likewise dependent on backpressures caused by pressure losses in the wellbore and

7
surface gathering network. The historical approach taken by the reservoir engineer has
been to prepare flowing bottom hole pressure versus flow rate (VFP) “look-up” tables to
approximate the backpressure caused by the wellbore and surface network
(Kosmala,2003). Each table is unique with respect to tubular dimensions, lengths, etc.
When dealing with multiphase flow many permutations (e.g. different water-gas ratios,
condensate-gas ratios) are often required to construct a VFP table over potential range of
flowing conditions. Such an approach is suitable for single well, single flow line
gathering networks however; such configurations are not typical of most pipeline
networks. It is not uncommon for networks to include numerous wells, tubing
descriptions, pipeline branches and loops, as well as a variety of surface equipment such
as pumps, compressors, line heaters, separators, etc.
Attempts by the reservoir engineer to forecast the flow rate potential of the reservoir, and
the pipeline design engineer to design a network to efficiently transport reservoir fluids is
particularly difficult in high rate, low pressure systems. Simplification of the hydraulic
component of the gathering network or the reservoir model component usually introduces
an unacceptable error to engineering calculations. The use of conventional surface
network models, that use one-dimensional, tank-type reservoirs, may address the
backpressure problem, but they cannot accurately forecast transient production profiles.
This is especially true in low permeability, hydraulically or naturally fractured reservoirs,
or those reservoirs exhibiting multiphase flow effects.
Reservoir management is normally achieved using numerical simulation to model the
performance of the reservoir under different scenarios of well placement, number of
wells, and production and/or injection profiles. However, reservoir simulators do not
generally model the production downstream of the wellhead, and so the production
network effects on the behavior of the overall system are not fully acknowledged. Flow
simulation of the reservoir system also does not account for all the boundary conditions
set at the surface, such as the suction pressure of the separator. This may have a direct
impact on the evaluation of the production targets that will actually be achieved. On the
other hand, production management typically uses surface network nodal analysis tools
that fully account for those effects but can only model the reservoir as a homogeneous
‘tank’ of uniform properties.

8
Moreover, reservoir management aims at optimizing the reservoir performance over the
field life by maximizing the recovery factor at the minimum cost, while production
management is concerned with optimizing the production system capabilities on a day-to-
day basis. Thus, reservoir and production management have complementary goals in field
development, but on different time scales, and by using separate tools there is no
guarantee that one will achieve a solution that satisfies both aims. Therefore, the
integration of the capabilities of both reservoir and production system simulators appears
to be a critical technology for field development and optimization.
The problems described have long been understood as an impediment to improving the
accuracy of reservoir simulation forecasts and the design of pipeline networks to
transport the produced fluids. The solution to the problem demands integration of various
engineering disciplines, and their software technologies.
Recently, some vendors of reservoir simulation software have created interfaces allowing
third party pipeline. (Beliakova, 2000)
The system that was developed therefore consisted of a reservoir simulator coupled to a
network modeling and optimization tool through a controller (Ghorayeb,2003) The
controller acts as a data link between the two simulators to facilitate optimal reservoir and
production management and achieve realistic targets while respecting defined field and/or
well constraints. The reservoir simulator models the flow from the reservoir to the bottom
of the wells (either to the sand face or upstream of the downhole valves) and the
production network from this point up to the separator, so there is no overlap in the
models.
Two functionalities appear as essential in the development of a coupling tool between the
reservoir and production network simulators: 1) the creation of a “physical” link between
the two systems so that the results of the simulations are dynamically passed from one
simulator to the other and 2) the design of a component that can optimize the overall
system.

2.4 Modeling software

The PIPESIM steady-state multiphase flow simulator offers complex production and
injection networks analysis. The well, pipeline, and flow assurance capabilities are all

9
within a shared common environment, powered by the most rigorous field wide solver.
The solver is suitable for networks of any size and topology, including complex loop
structures and crossovers. By modeling the entire production or injection system as a
network, the interdependency of wells and surface equipment can be accounted for, and
the deliverability of the system can be determined.
PIPESIM network simulation and optimization capabilities enable users to
 Engineer the best well, pipeline, and facilities design
 Identify production bottlenecks and constraints
 Optimize production from complex networks
 Handle multiple system constraints
 Quickly identify locations in the system most prone to flow assurance issues such
as erosion, corrosion, and hydrate formation
 Quantify the benefits of adding new wells, compression, pipelines, etc.
 Determine optimal locations for pumps and compressors
 Design and operate water or gas injection networks
 Analyze hundreds of variables such as pressure, temperature, and flow assurance
parameters through complex flow paths
 Evaluate benefits of loops and crossovers to reduce backpressure
 Calculate full field deliverability to ensure contractual delivery rates can be met
 Optimize the allocation of lift gas amongst wells
Modern production systems require designs that ensure safe and cost-effective
transportation of fluids from the reservoir to the processing facilities. Once these systems
are brought into production, the ability to ensure optimal flow is critical to maximizing
economic potential. From complex individual wells to vast production networks, the
PIPESIM steady-state multiphase flow simulator enables production optimization over
the complete lifecycle (Brill, 2012).

2.4.1Continuous innovation

For over 30 years, the PIPESIM simulator has been continuously improved by
incorporating not only the latest science in the three core areas of flow modeling—
multiphase flow, heat transfer, and fluid behavior—but also the latest innovations in

10
computing, and oil and gas industry technologies. The simulator includes advanced three-
phase mechanistic models, enhancements to heat transfer modeling, and comprehensive
PVT modeling options. The ESRI-supported GIS map canvas helps deliver true spatial
representation of wells, equipment, and networks. Networks can be built on the GIS
canvas or generated automatically using a GIS shape file. The interactive graphical
wellbore enables rapid well model building and analysis. Faster simulation runtime has
also been achieved for all modeling though the implementation of a new parallel network
solver to spread the computational load across all processors.

2.4.2 Steady-state flow assurance, from concept to operations

The PIPESIM simulator offers the industry’s most comprehensive steady-state flow
assurance workflows for front-end system design and production operations. The flow
assurance capabilities of the simulator enable engineers to ensure safe and effective fluid
transport—from sizing of facilities, pipelines, and lift systems, to ensuring effective
liquids and solids management, to well and pipeline integrity. In addition, where dynamic
analysis is needed to add further insight, the PIPESIM-to-OLGA converter tool enables
rapid conversion of models. Shared heat transfer, multiphase flow, and fluid behavior
methodologies ensure data quality and consistency between the steady-state and transient
analyses.

2.5 Modeling History

According to recent literature many surface network modeling has been conducted in
variety of networks configuration and in combination of deferent simulator in this section
we will discuss three case studies that were conducted in the last decade and how they are
related to this study in term of network configuration, type of field,wells type, procedure
and optimization result in addition to type of soft ware used in conducting the study.

2.5.1 ADCO Abudhabi 2012

A giant onshore field producing from multilayered under saturated carbonate reservoir
w a s presenting many challenges. The field is producing since early seventies
supported with peripheral water injection leading to wide variation in reservoir pressure
and water cuts. The field has been produced with the help of 6 gathering manifolds. The

11
northern manifold presents additional challenges as this area is affected with
asphaltenes deposition problems in production tubing and flow lines. Tubing obstruction
due to asphaltenes adversely affects flow besides cause difficulty in lowering pressure
gauges resulting into scarcity of pressure survey data. Additionally, some wells are
operating on gas lift and a gas injection pilot is located on the western side leading to
Gas Oil Ratio (GOR) variations. The horizontal wells completed in the low
permeability layer tend to cease production as the water cut reaches >35%. Production
allocation, optimization and de-bottlenecking become difficult in such a scenario. It
was, therefore, decided to build a production network model as a tool to overcome such
problems.
It was decided to construct an integrated network model of the field to address various
issues as enumerated above. Well modeling and conducting Nodal Analysis to
ascertain well behavior is an industry practice. Connecting the well models with
surface facilities through flow lines and transfer lines has been attempted by many
workers (Nader et al. 2006 and Kumar et. al. 2012). The calculation of flow rates and
pressure drop in such a network is attempted by linear programming method. Network
modeling by utilizing this method has reportedly provided excellent match with the
field data. Jha et. al. (2009) has attempted to elaborate a plan for implementing a field
wide integrated network optimization model incorporating real time data measurement
for Bombay High Field. Many operators are implementing smart field solutions with
real-time network modeling through collaborative working environment (Bin Amro et.
al. 2010). The cost of implementing smart field solutions to this maturing giant
carbonate field would be cost intensive as well as time consuming. It was, therefore,
decided to build a network model for the field to address various issues.
A network model was built using PIPESIM software using workflow
Data gathering All the data regarding each well was collected in an Excel sheet. The data
included depth reference, well diagram, deviation survey data, and pressure survey data of
the well and nearby wells, production test data and well history. These Excel sheets would
serve as reference documents during the model construction. An area was allocated in the
sheet to keep a log of modeling activity like calibration and any changes made in the
model. All the information was validated to avoid any uncertainty in the future.

12
PVT Matching the giant carbonate field, which is producing from three distinct pays, is
presumed to be behaving like a single under saturated reservoir unit. Review of the PVT
studies conducted in the field does not show wide variation. Therefore, the results of only
available PVT study for this area were adopted. Slight tuning of the PVT parameters was
done to match separator conditions as well as during flow correlation matching. Standing
(gas solubility), Vasquez and Beggs (OFVF) and Chew Connelly (viscosity) were found
satisfactory.
Build well Models Individual well models were constructed based on well diagram and
deviation data. In case of vertical wells PI model for completion was adopted but
horizontal completions were formed on the basis of Joshi model with zero anisotropy
along the open hole. Each string of the dual completion well was modeled separately.
GLV sizes were included as per latest design.
Pressure Traverse Matching Several pressure traverse matching’s conducted to select
the most appropriate correlation for the project. This exercise was conducted on many
bottom hole pressure data having good control on production test data. Many correlations
like Beggs and Brill, Hagedorn and Brown, Duns and Ros, and Orkiszewski were tried.
Hagedorn and Brown correlation was selected with holdup factor as 1.
Network model of one of the manifolds of a giant carbonate field was built.
Satisfactory PVT tuning and well calibration was achieved for all the wells.
Calibration of wells in this area proved challenging due to lack of pressure data. The
model predicts oil production in excess of 6-10% with the allocation data. Unstable
flow in many of the horizontal wells, outflow constraint in asphaltenes affected
wells and pressure fluctuation in RDS-CDS transfer line may be the reason behind
over estimation besides allocation methodology. Further tuning of model for such
factors and conciliation with production allocation would greatly reduce this
uncertainty. Horizontal wells producing from low permeability reservoirs tend to flow
in unstable flow regime because of larger tubing size. Such unstable flow also causes
uncertainty in flow prediction by the model. Premature shut down of these wells may
be avoided by reduction in tubing size (Al Sayari, 2013). It can be seen that network
modeling is an effective tool for production estimation, optimization, allocation and
debottlenecking.

13
2.5.2 Shell 2007

To support integrated network analysis Shell Global Solutions International has

developed the strategic planning tool: GMOS/NetSim (Global Manufacturing & logistic
Optimization System/Network Analysis and Supply Chain Optimization System.
The model helps to optimize flow and production of oil and gas, between wellhead
platforms and demand locations over the defined time period given the infrastructure
(e.g. production, pipelines, and compressor) constraints. The economic analysis converts
output from the program into analysis of individual assets and scenarios based on costs,
capabilities and prices.
The GMOS/NetSim model can be split into four key modules. These are the Supply
Module, which include production profiles for each existing and future wellhead
platform (the so-called technical potential); the Processing Module describing
production/processing facilities & constraints; the Demand Module, which covers any
demand point (oil & gas demand, fuel- and re-injection gas requirement, etc); and the
Interconnection Module, which ties together supply, processing and demand to account
for pipeline or distribution constraints and capacities. General data input is required of
supply sources (wellhead platforms), production and processing facilities infrastructure
and capacities (processing platforms, compressors, pumps etc.), transportation capacity
(pipelines), costs of processing (operating cost and fuel gas consumption), and other
business constraints in production, processing and distribution (e.g. minimum and
maximum demands, no venting after a specific year).
GMOS/NetSim helped to optimize the entire network over the entire time horizon. The
objective function is based on the variables in the system and is typically the Net Present
Value of the profit margin over the time-horizon, based on the GOR & Water cut of the
respective wellhead platforms and the Unit Production Cost.
The potential or the results of the study are typically expressed in oil and/or gas that can
be accelerated through the debottlenecking exercise.

14
 Figure (2.1) Shell optimized oil production (Lee & Thijssen, 2007)

Approximately ten percent of the fifteen years figure (2.1) oil production can be
accelerated if the debottlenecking options will be implemented substantiation required for
this still These options vary from extra processing facilities, via small piping changes to
bypassing compressors and swapping lines in direction or by type of flow going thru the
pipeline. The thick line indicates the maximum of oil that can be produced if all wellhead
platforms are producing at their technical potential.

2.5.3 Aramco Khafji field 2005

Khafji field commenced oil production in 1961, and currently 300MBOPD of average
daily crude production, among which 260MBOPD is sweet crude as Khafji crude and
40MBOPD is sour crude as Hout crude, has been maintained from seven (7) oil
reservoirs (Ghoniem, 2005).
The offshore production facilities consist of approximately 180 production wells on the
scattered 80 unmanned well jackets, a lot of different size of flowlines connecting from
wells to one of four (4) gas oil separation platforms for Khafji crude, gathering station
complex (gas lift compressor platform and pump platform, offshore control center, living
quarter platform, utility platform, and gas oil separation platform for Hot crude). The gas
lift operation by means of artificial lift in Khafji field has successfully contributed to
sustain field target since it was introduced in 1988with the capacity of 25MMSCF/D lift
gas.

15
Field study was conducted to establish the most optimum Al-Khafji Joint Operations
(KJO)long term field development plan (FDP), which includes finding the optimum
target rate over 30 years and optimum artificial lift among different artificial lift methods
with facility scenarios.
FDP study concluded the necessity of a lot of capital investment in near future to sustain
the optimum target rate for Khafji crude, such as additional infill wells, expansion of
handling capacity of GOSPs, and expansion of gas lift gas compression facility,
introduction of ESPs, and etc. due to the increase of field water cut and depletion of
reservoirs. However, most of the above facilities require a longer lead time until their
installations and commissioning.
In order to make up production decline and prolong to sustain the field target rate until
the commissioning the new facility expansions and possibly to defer such a large
investment, production optimization, which comprise lift gas optimization and de-
bottlenecking of the existing flowlines network to reduce backpressure, was found one of
the cost effective solutions to be able to be executed shortly.
The production optimization study, for Khafji field, was successfully conducted in two
steps as the followings:
Lift Gas Optimization. The lift gas optimization study, by using the models, involves two
activities, the first is screening sensitivity run between the existing gas lift wells (65
wells), and then the available lift gas capacity (21 MMSCF/D allocated for Khafji sweet
crude facilities) was optimized.
As a result, the most effective gas lift wells to increasing oil rate were prioritized for lift
gas re-allocation. Also, it was found that a total of forty two (42) wells can be lifted
successfully by the available lift gas capacity. The saved amount of gas was re-allocated
to open more wells on gas lift and for some high productivity wells.
Result shows the comparison of gas injection rate and oil rate for the forty two (42) gas
lift wells between before and after lift gas optimization. it shows also 2,800 BPD of oil
gain from well No.38, can be obtained by applying lift gas optimization, although it was
closed before conducting the optimization study due to lack of gas lift volume. Also,
three (3) well Nos.5, 40 and 42, were producing by natural flow
With lower oil rates of 1600, 2200, 1200 BPD respectively before lift gas optimization.

16
However, after applying the screening by the models, approximately 1.3 MMSCF/D of
lift gas was allocated hence producing with higher oil rates of 3850, 4600, and 3700 BPD
respectively, which can obtain additional oil gain of 7150 BPD (Ghoniem, 2006). The
rest of saved amount of lift gas was allocated for some high productivity wells like wells
Nos. 7, 9, 15, 25, 35 and 37, which also resulted in higher oil rate. Figure (2.2) shows the
oil rate for the forty two (42) wells before and after lift gas optimization.

Figure (2.2) Gas Lift Wells Optimization ( Ghoniem,2005)

Also, as the results of lift gas optimization, a performance curve at the optimum operating
conditions was obtained for each gas lift well. As an example, gas lift performance curves
for wells Nos. 4, 9 & 10 at optimum gas injection rate.
Consequently, a marginal gain of approximately 12,000 BOPD was obtained by
optimizing lift gas within the existing lift gas capacity with the proper utilization of the
existing flow line network.

17
Chapter 3 Methodology

In this study it was decided to follow following steps while and building network
model and various sensitivity study:
 Data Collection and Validation (QC)
 Physical Model Building and Validation
 PVT modeling
 Multiphase Flow Correlation Matching
 Network Balancing and fine tuning

3.1 Data Collection

Data Collection is the first and foremost requirement of a model building effort. Since
field is structurally and otherwise a dynamic environment it was essential that model
building and validation should be done by matching model result to certain cut-off date
instead of trying to match a moving target.
In order to ensure speed and efficiency on data Collection process, a detailed list of data
requirement was prepared upfront. The data included depth reference, well diagram,
deviation survey data, and pressure survey data of the well and nearby wells, production
test data and well history. Meetings among various discipline and groups were organized
to ensure clear understanding of data requirement and objective of the study. The data
were manually collected from various groups and locations of GNPOC.

3.2 Physical Model Building

A hydraulic network in PIPESIMTM is made up of single branches or segments
connected at points called nodes. The segment may be just a connector or it may contain
pressure loss devices such as pipes and piping equipment connected in series. Nodes can
be boundary nodes (Sources and Sinks) or internal nodes (junctions). The net flow in a
junction node is zero. A boundary node can be a:
1. Source node: where fluids flow into the network; node flow rate is positive.
2. Sinks node: where fluids flow out of the network; node flow rate is negative.

18
3.2.1 Layer 1: FPF, OGMs and Trunk lines

In PIPESIMTM Graphical user interface (GUI), the network layout has been logically
organized using PIPESIM’s folder option to enable easy navigation to various parts of the
model Figure (3.1)

Figure (3.1) Network Layout

Three trunk lines connecting OGM’s together and one trunk line connects to HE FPF
these trunk lines data such as:
1. Horizontal distance.
2. Inner diameter.
3. Wall thickness.
4. Roughness
5. Ambient temperature was input for each line. Figure(3.2)

19
 Figure (3.2) Flow lines Data

3.2.2 Layer 2 Sources and flow lines connected to OGMs

Sources and production wells are connected to each OGM by a flow line data were input
to this flow line as same as the data used to build the trunk lines illustrated earlier
Figure (3.3) show how these wells and flow lines are distributed.

20
 Figure (3.3) Wells and Flow lines layout

3.2.3 Layer 3: Wells Model

1. Wells operated by PCP and sucker road pump were treated as sources due to
PIPESIME 2011 version limitation and for the fact that they are operated at surface
input data required are:
1) Daily production rate
2) Temperature,
3) Fluid properties
As illustrated in the figure (3.4)

21
 Figure (3.4) Sources Well Data

2. Wells with Electrical submersible pumps

ESP wells were treated with different approach since PIPESIM allow production
optimization for this kind of wells with downhole data in term of determining optimum
operating condition to obtain pump performance curve but required additional data
include
a) Pump manufacturing design such as:
i. Frequency.
ii. Number Of Stages.
iii. Pump Efficiency.
b) Completion Data
c) Fluid Properties
Minimum data entry was uploaded as shown in figure (3.5).

22
 Figure (3.5) ESP Wells Data

3.2.4 PVT Data and fluids properties

A black oil model was selected since it is typically applicable for GOR less than 2,000
STB/SCF and compositional data were not available, data required include
1) GOR
2) API
3) WOR
This data were input as illustrated in figure (3.6).

23
 Figure (3.6) Black Oil Model Data

Viscosity data were collected from the laboratory and it was input as specified viscosity
in a specified temperature as illustrated in figure (3.7).

Figure (3.7) Black Oil Viscosity Data

24
3.2.5 Flow Correlations

Flow correlation was selected from variety of correlation provided by software based on
best match for vertical flow correlation Hagedom & Brown correlation was selected
And For horizontal correlation Beggs & Brill revised correlation was selected as
illustrated in figure (3.8).

Figure 3.8 Flow Correlation Selections

Flowlines inlet pressure matching was carried out to ensure that the measured flowlines
inlet pressure and that calculated by the models are consistent. Matching the surface
flowlines pressure was done to confirm the applicability of selected flow correlations for
surface network.

3.3 Running the model

After applying previous steps to construct the model and balancing the data a model was
checked and verified for errors, no error was found and the model became ready for
running and execution as illustrated in figure (3.9).

25
 Figure (3.9) Model Data Check

The model then was run successfully as illustrated in figure (3.10) and detailed result will
be discussed in next chapter.

Figure (3.10) Successful Model Running

26
CHAPTER 4- RESULT

In this chapter we will discuss the result after running and validating the data
A flow correlation matching result concluded based on best match correlation of the
actual data for vertical flow correlation Hagedom & Brown correlation was selected and
For horizontal correlation Beggs & Brill revised correlation was selected as illustrated in
figure (4.1)

Figure (4.1) Vertical Flow Correlation Matching

The production optimization study for Hamra field was successfully conducted and the
result and main findings are illustrated in two steps as the followings:

4.1 OGM’s Trunk Lines

The trunk lines at all OGM’s (OGM3~OGM6) which have excessive pressure in
comparison with the model pressure were identified and assessed table (4.1) illustrate the
comparison between the actual data and the model data in term of pressure

27
OGM Actual Model Difference Trunk Equivalent Percentage Remark
outlet outlet (psi) line size trunk line of area
pressure pressure (in) size opened
(Psi) (psi) (in) (%)
OGM-3 170 155 15 10 10 100 Ok
OGM-4 185 162 23 6 4.7 61 partially
blocked
OGM-5 185 156 29 10 10 100 Ok
OGM-6 190 174 16 6 5.5 84 Started to
block
Key :
(0%=ok), (less than 25%= Mostly blocked), (25-75%= partially blocked), (above 75% started to block)

Table (4.1) Actual Data And Model Data Comparison For OGM’s

The trunk lines which have an excessive backpressure were successfully identified by
using the prediction mode of the models. As a result, the excessive backpressure was
observed in many 10” size flowlines and some commingled flowlines which handle high
rate and in 6’’ lines due to wax or asphaltenes deposition.
The trunk line that connects between OGM 4 and OGM 3 there is one bottleneck due to
long distance (6562 ft) and the high rate handled (3,200 BPD). A pressure drop of 23 psi
is created through the line, which shall cause the 119 BPD of production reduction for the
connected wells Figure (4.2). It was found that Cleaning the line using hot water and
pigging technique will enable us to reduce the back pressure and gain 119 BPD this is in
short term for long term solution laying an additional new line to the nearby jacket to
another OGM is the most cost effective de-bottlenecking solution, which results in an oil
gain of at least 207 BPD as listed in Table (4.3) And figure (4.3)

28
Figure (4.2) Pressure Drop simulation For OGM-4 Trunk line

Figure (4.3) Oil Production Rate Comparison for OGM -04

29
OGM No of Actual Model Difference Actual Model Difference
wells pressure pressure (psi) flow rate flow rate (BBL)
(Psi) (psi) (BPD) (BPD)
OGM-03 10 170 155 15 9067 9049 -18
OGM-04 7 185 174 11 3112 3231 +119
OGM-05 13 185 156 29 4505 4502 -3
OGM-06 8 190 174 16 1296 1383 +87
Total 207
OIL gain

Table (4.2) Actual Flow Rate and Model Flow Rate Comparison

The other observation is in trunk line that connect OGM -06 to OGM-05 the model show
a 16 psi difference and the prediction mode showing a reduction in equivalent pipe
diameter of 0.5 in indicating that the line is starting to face a bottlenecking issue that
cause a loss of 87 bbl in compare with the model , this line has to be considered for clean
up and pigging before the severity of bottlenecking increases and cause further reduction
in production Figure (4.4) showing the model detailed tabular report of the 4 OGM’s
being examined .

30
 Figure (4.4) OGM’s Trunk lines Tabular report

4.2 Wells Flow Lines

Extending the analyses to the OGM-06 wells flow lines predicted another issues where
by six bottlenecks were identified as shown in table (4.2) and figure (4.5).

31
HA-OGM- Actual Model Difference Trunk Equivalent Percentage Remark
06(wells) outlet outlet (psi) line trunk line of area
pressu pressure size size opened
re (psi) (in) (in) (%)
(Psi)
HAE-20 250 213 37 6 2.3 15 Mostly blocked

HAE-25 235 203 32 6 3.8 40 partially

blocked
HAE-27 210 184 26 6 4 45 partially
blocked
HAE-29 215 190 25 6 5.5 84 Started to
block
HASE-10 220 181 39 6 4 45 partially
blocked
HASE-11 235 195 40 6 2.8 21 Mostly blocked

HASE-12 176 173 3 6 6 0 ok

HASE-14 210 176 34 6 3.5 34 partially
blocked
Key :
(0%=ok), (less than 25%= Mostly blocked), (25-75%= partially blocked), (above 75% started to block)

Table (4.3) Actual Data and Model Data Comparison for OGM-06 Wells

It crucial to remove this bottlenecks by identifying it as early as possible and clean it

before it comes to point of costly process such as pigging or replacing the line in order to
restore all wells productivity and minimize back pressure exerted by strong wells on
weaker ones.
This practice if implemented regularly will help on maintain optimum production from
all the wells in addition to minimize the cost of pigging and new lines installation.

32
 Figure (4.5) OGM-06 Wells Flow Lines Tabular report

4.3 Findings:
Hamra OGM-04 to OGM-03 trunk line excessive pressure, the recommended approach to
tackle this issue in short term solution is to flush with hot water.
Study the pigging possibility because the effectiveness of 6" trunk line
On the long term solutions it is recommended to install a Surface Heater that will
increase the temperature in the line and thus reduce wax and asphaltenes accumulation
resulting in decreasing of flowing pressure.
33
Hamra- OGM-06 to OGM-05 trunk line:-
Since the line is just started to bottleneck the recommended solution is to flush the line
with hot water and study the pigging possibility to reduce the pressure to the model
pressure.

34
Chapter 5 Conclusion and recommendation

5.1 Conclusion
Hamra field models which comprise surface flowlines network and wells have been
successfully constructed.
Hamra Field surface flowlines network deliverability was investigated in the current
operating condition.
The more economic and effective network de-bottlenecking was successfully assessed.
A net oil gain of approximately 207 BPD is expected.
A net production gain by the above production optimization can assist to sustain Hamra
target production rate for coming years.
The models are ready for field optimization under different operating conditions and
should be updated regularly.
This study confirms that modeling network analysis can help to bring production closer
to the technical potential of the Field production. It can help to identify the impact that all
changes together have on the performance of the network.

. 5.2 Recommendation
1. As this studies focus on identifying actual bottlenecks and future bottlenecks,
accurate representation of the network is crucial. This means that accurate data is
key to success and that design data of equipment and pipelines alone is not
sufficient. If actual performance data is not available on site, performance testing
prior/during these types of studies will be required. It is crucial that the client is a
member of the study team.
2. This study would present more accurate data if it is been merged with OLGA
flow assurance software in order to upload dynamic data .

35
References

 Beliakova, N. et al.: "Hydrocarbon Field Planning Tool for Medium to Long Term
Production Forecasting from Oil and Gas Fields Using Integrated Surface-Sub-Surface
Models," paper SPE 65160 presented at the SPE European Petroleum Conference,
Paris, France (October 24-25, 2000).
 Ghoniem, E.O., et al.: “Dynamic Production Optimization in Khafji Offshore Field”,
SPE 101388 paper presented at the 2006 Abu Dhabi International Petroleum
Exhibition and Conference held in Abu Dhabi, U.A.E., 5–8 November 2006.
 Ghoniem, E.O., et al.: “Successful Application of Production Optimization Models to
Sustain Field Target and Defer Capital Investment in Offshore Khafji Field”, SPE
93555 paper presented at the 14th SPE Middle East Oil and Gas Show, Bahrain,
March 12-15, 2005.
 Hagedorn, A.R. and Brown, K.E.: ``Experimental Study of Pressure Gradients
Occurring During Continuous Two-Phase Flow in Small Diameter Vertical Conduits,''
JPT ~October 1965! 475; Trans., AIME, 234.
 Jim Brill "Multiphase Flow Technology- Past, Present and Future "PIPESIM User
Group Meeting, Houston, September 5, 2012.
 Kassem Ghorayeb A General Purpose Controller for Coupling Multiple Reservoir
Simulations and Surface Facility Networks, SPE,
 KJ Li, Nort Thijssen, and Ilse Mittendorff(2007), “Enhancement of Oil and Gas
Production using Integrated Network Analysis” paper SPE IPTC 11638 presented at
the International Petroleum Technology Conference held in Dubai, U.A.E., 4–6
December 2007.
 Lasschuit, Winston, Thijssen, Nort (2004), “Supporting supply chain planning and
scheduling decisions in the oil & chemical industry”, Computers and Chemical
Engineering 28, 863-870, Elsevier.
 Litvak, M.L. and Darlow, B.L.: ``Surface Network and Well Tubing-head Pressure
Constraints in Compositional Simulation,'' paper SPE 29125 presented at the 1995

36
 SPE Symposium on Reservoir Simula-tion, San Antonio, Texas, 12-15 February.
 Litvak, M.L.: ``New Procedure for the Phase-Equilibrium Computa-tions in the
Compositional Reservoir Simulator,'' paper SPE 25252 presented at the 1993 SPE
Symposium on Reservoir Simulation, New Orleans, 28 February±3 March.
 Lohrenz, J., Bray, B.G., and Clark, C.B.: ``Calculating Viscosities of Reservoir Fluids
From Their Compositions,'' JPT ~October 1964! 1171; Trans., AIME, 231.
 Thijssen, Nort, Li, K.J., Mittendorff, Ilse (2007), “Enhancement of Oil & Gas
Production via Network optimization”, Process Systems Engineering Vol. 3; Supply
Chain Optimisation, Wiley.

37
Appendix
Flow line
WELL VISCOSITY (CP) W.C (%) API PUMP TYPE RATE (BBLS/D)
THP( psi) DISTANCE(m)
0.0
HAE-20 230 142 36 BPU 300 100
0.4
HAE-25 220 354 35.29 BPU 500 900
0.0
HAE-26 135 870 34 BPU 700

HAE-27 210 227 2.0 34.32 PCP 500 450

HAE-28 Water Injection Well

HAE-29 215 880 2.0 32.59 PCP 505 500

HASE-10 220 225 4.8 34.95 BPU 600 1450

HASE-11 235 249 0.3 33 PCP 500 1600

HASE-12 230 179 70.0 27 PCP 700 1700

HASE-14 210 150 50.0 29.62 PCP 1860 1350

Table 1 OGM-06 wells Actual data

WELL THP VISCO W.C API PUMP RATE [Link] [Link] DSITANCE
(PSI) SITY (%) TYPE (BPD) ESS(PSI) SS(PSI) (M)

HAE- 200 251 10.0 32.07 ESP 500 343 2249 650
03

HAE- 190 285 96.3 34 ESP 2500 1285 1610 50

07

HAE- 200 114 0.0 33 ESP 300 1530 2083 300

12

HAE- 200 60 16.0 32 ESP 600 1717 2056 600

16

Table 2 OGM-05 ESP wells Actual data

38
Figure 1 OGM-05 & OGM-05 Laboratory data

39
 Figure 2 HAE-02 Data Matching

Figure 3 HAE-02 Performance curve

40
Figure 4 OGM -03 Tabular Reports

Figure 5 HAE-02 Tabular report

41
Figure 6 OGM-06 Model Data

42