PROCEEDINGS OF ECOS 2020 - THE 33RD INTERNATIONAL CONFERENCE ON

EFFICIENCY, COST, OPTIMIZATION, SIMULATION AND ENVIRONMENTAL IMPACT OF ENERGY SYSTEMS

JUNE 29-JULY 3, 2020, OSAKA, JAPAN

Optimizing the power hub of offshore multi-

platform for oil and gas production

Ronaldo L. A. Freirea, Daniel Flórez-Orregob, Julio A. M. da Silvac, Cyro

Albuquerque Netod and Silvio de Oliveira Juniore
a
University of São Paulo, São Paulo, Brazil, [email protected], CA
b
University of São Paulo, São Paulo, Brazil, [email protected]
c
Federal University of Bahia, Salvador, Brazil, [email protected]
d
FEI University Center, São Bernardo do Campo, Brazil, [email protected]
e
University of São Paulo, São Paulo, Brazil, [email protected]

Abstract:
Offshore oil and natural gas production is an energy-intensive activity and is responsible for the emission of
significant amounts of carbon dioxide into the atmosphere. The main emitting source is the simple-cycle gas
turbines (SCGT) of the utility system which supplies heat and power to the production platforms. Severe vessel
area and weight constraints are often cited as the main reason why production platforms are unable to allocate
high-efficiency combined-cycle gas turbines (CCGT) common in land-based power plants. Published work
suggests that in production development projects of giant offshore oil fields, the thermodynamic efficiency of
the utility system may be increased significantly, without prejudice to project economic viability, through an
additional vessel dedicated to generating power in CCGT. The best results are obtained when the power
demand is split between the power hub and local gas turbines which are used in cogeneration mode to
additionally produce heat for separation purposes. Therefore, this work proposes a methodology for optimizing
the power block of the power hub. The first step is the selection of combined cycle configurations from the
commercially available aero-derivative gas turbines. At sequence, evolutionary algorithms are used in the
multi-objective optimization (MOO) of the steam bottoming cycle, whose objective is to obtain the
configurations that produce the best results in terms of atmospheric CO2 emissions, occupied area, and capital
cost. A method is then proposed to select the best solution from the non-dominated solutions that compose
the Pareto front, taking into account the constraints imposed by the vessel of the central power plant and the
objectives to be optimized. The power hub solution presented average exergy efficiency 8.2p.p. above the
conventional, thus reducing the fuel gas consumption by 1.74 million ton and consequently avoiding the
emission of 4.75 million ton of CO2. Finally, in the context of growing environmental concern and taxation of
CO2 emissions, this work contributes to highlighting the advantages of the central power plant in future
maritime production development projects in large oil and gas fields.

Keywords:
Offshore, Power Hub, Combined-Cycle, Multi-Objective Optimization, CO2 Emission.

1. Introduction
Offshore oil production plays a strategic role in meeting the world’s energy demand. According to
the U.S. Energy Information Administration, in 2015, almost a third of the world’s oil production
occurred at sea, with Brazil ranking as one of the top five offshore producers [1]. According to the
Brazilian Agency of Petroleum, Natural Gas and Biofuels, offshore oil production represented 93%
of the country’s total production in 2015 [2]. Since 2006, giant offshore oil fields are being discovered
in Brazil, resulting in the largest offshore production development projects today, which are
characterized by clusters of production platforms.

Offshore oil production is energy intensive and significant amounts of carbon dioxide (CO2) are
released into the atmosphere by low-efficiency utility plants on production platforms. These systems
typically have simple-cycle gas turbines (SCGT) due to severe area and weight restrictions on vessels.

1
Although equipped with waste heat recovery units (WHRU) to convert part of the flue gas energy
into useful heat, the cogeneration demand is usually less than the system potential. Additionally,
power and heat demands undergo profound changes throughout the platform operating life, resulting
in partial load operation most of the time. The operation of rotating machines far from the best
efficiency point incorporates even more inefficiencies in the power plant.

Kloster [3] indicates that the increasing commercial value of natural gas and the taxation of
greenhouse gas emissions are the main factors driving the technological development of steam
bottoming cycles (SBC) for offshore platforms. According to Nguyen et al. [4], the integration of
SBC on offshore platforms is currently regarded as the most promising option for improving the
performance of these energy-intensive systems. However, weight and area requirements are the major
obstacles to SBC in real projects. A heat recovery steam generator (HRSG) may require 25-50% extra
area than a WHRU [3]. Nevertheless, as the heat exchanger is placed on the top of the gas turbine
main skid, the required extra area should not be a real problem.

Følgesvold et al. [5] assessed a GE LM2500+G4 gas turbine as topping cycle, and a once-through
heat recovery steam generator with extraction and back-pressure steam turbines (ST) as bottoming
cycles. According to the authors, the thermal efficiency could be increased by 12.3p.p. and 8.9p.p.
and CO2 emission reduced 26% and 21%, respectively, compared do the SCGT configuration. Nord,
Martelli and Bolland [6] optimized the weight-to-power ratio of a SBC for offshore platform. A
reduction of 4% compared to a knowledge-based design was obtained, and a Pareto front allow the
designer to select the solution which best matches the installation constraints. Riboldi and Nord [7]
investigated the optimum design approach for offshore SBC. Authors suggest that, rather than at peak
conditions, better overall performance is obtained when designing the plant at the end-life conditions.

A power hub can be attractive in production development projects with clusters of platforms, common
in giant oilfields. Some recent Brazilian pre-salt projects, such as those in the Lula, Búzios and Mero
oil fields, have up to four platforms distant up to 10 km from each other. The electrical interconnection
among platforms and to a power hub is technically possible and can be a solution to eliminate the
area and weight constraints that currently prevent the adoption of SBC on offshore platforms.

Vidoza et al. [8] assessed a power hub connected to three pre-salt platforms. The power hub with
power blocks containing three gas turbines coupled to a dual-pressure HRSG and a ST resulted in the
best cost-weight ratio and 53.2% thermal efficiency. Defining the best configuration for the utility
plant in a production cluster with power hub is a hard task. A hybrid solution with power generation
in both production units and power hub increased the Second Law performance by 9.1p.p. in the work
of Freire and Oliveira Jr [9]. An incremental financial analysis including the sensitivity to carbon
taxation shows that power hubs may not only be technically, but also financially viable in the future.

This work adds to the body of knowledge a comparative analysis which includes the optimization of
the SBC based on three objective functions: cost of capital, area occupied by the main equipment and
atmospheric CO2 emissions. The results obtained highlight the potential benefits of combined-cycle
power generation in the power hub associated with the combined heat and power (CHP) production
in the production platforms.

2. Production development project

Offshore platforms are normally equipped with SCGT-WHRU power blocks to supply efficiently
heat and power to the entire vessel considering its operating life. This section defines the case study
under investigation, which integrates multiple production units to a power hub equipped with CCGT
power blocks to supply part of the demanded electricity.

2
2.1. Energy demand

Silva and Oliveira Jr. [10] estimated the lifetime energy demand of a primary processing plant in a
pre-salt platform. The commercial oilfield life was set as 22-year. Authors proposed to estimate the
thermal demand as for the multiplication of predicted annual oil production and specific heat
consumption rates, varying according to the operating mode. Similarly, energy demand is obtained
from polytropic compression equations and predicted associated gas production. Gallo et al. [11] also
estimated the energy demand over a pre-salt platform life through the simulation of the primary
processing plant in a commercial software. Authors used available well fluids’ data and Weibull
statistic models with production peak near 7.5 years since production starting and the field life of 25
years to estimate power and heat demands. This work uses the electricity and heat demand curves
from Gallo et al. [11]. However, a change has been made to the original electricity to consider CO2
compressors driven by electric motors instead of SCGT. The combined heat and power demand
profile in shown in Figure 1.

Figure 1. CHP profile of the production platform (adapted from [11])

A production development project with four identical platforms is evaluated in this work, that is, with
identical heat and power as in Figure 1. In Cases 1 and 2 the start of production is delayed in 1 year
and the platforms are relatively close so that the transmission losses were neglected. Figure 2 illustrate
the full project taking into account the alternative design proposed in Case 2 (see item 3.1).

Figure 2. Production development project with four production platforms and a power hub

3
2.2. Fuel gas supply

The CO2 separation in the referenced pre-salt platform occurs between 3rd and 15th operating year
[11]. The high-pressure fuel gas is supplied by the primary processing plant within this period. In
contrast, from the production starting up to the end of the 2nd operating year and from the 16th
operating year onwards, the high-pressure fuel gas supply comes from the exporting/importing
pipeline. Table 1 shows the chemical composition of the fuel gas according to its source.

Table 1. Fuel gas composition by source (% molar)

Substance Exporting/importing pipeline (Mode A) Internal production (Mode B)
Methane 76.05 73.72
Ethane 10.63 11.54
Propane 6.30 7.36
I-butane 0.83 1.09
Butane 1.76 1.84
I-pentane 0.22 0.29
Pentane 0.49 0.32
Hexane 0 0.05
Nitrogen 0.64 0.77
Carbon dioxide 3.08 3.00

2.3. Environmental condition in Brazilian pre-salt area

Climatological standard normals are widely used as an indicator of the conditions likely to be
experienced in a given location. According to the World Meteorological Organization, climatological
standard normals can be obtained through averages of climatological data calculated for consecutive
30-year periods [12]. The Brazilian Institute of Meteorology [13] provides climatological standard
normals at various stations throughout the country. Cabo Frio station, localized in Rio de Janeiro
state, is representative of the site conditions in the Brazilian pre-salt region. In this context, Table 2
summarizes the annual average ambient condition set for this case study.

Table 2. Climatological standard normals in Cabo Frio station: 1961-1990 [13]

Atmospheric pressure Ambient temperature (°C) Relative humidity
(hPa) (%)
1,015 23.2 81

3. Utility plant analysis

This section is dedicated to the definition, modelling and simulation of the utility plants that support
the production system defined above.

3.1. Plant configurations

The utility plant in a pre-salt facility typically has four SCGT-WHRU units to supply heat and power
[10][11]. The system is configured at 4x33%, which means that the peak demand can be met by 3
power blocks, while one remains as backup for reliability reasons. This arrangement constitutes the
baseline of this comparative assessment, in which the performance of the proposed system based on
a power hub is evaluated. The two systems under comparison are detailed below:
4
 • Case 1 (baseline): this configuration represents the reference design case composed of a
typical pre-salt production platform. Each platform has its own utility plant with 4x33%
Siemens SGT-A35 aeroderivative gas turbine and waste heat recovery unit (see Figure 3);
• Case 2 (proposal): a central power plant with 5x25% GE LM6000PF+ combined-cycle. Each
production platform has one Siemens SGT-A35 aeroderivative gas turbine with waste heat
recovery unit and 4x25% hot water boilers (HWB) (see Figure 4).

Figure 3. Schematic diagram of Case 1

Figure 4. Schematic diagram of Case 2

3.2. Power block simulation

Utility plant processes were simulated with Thermoflow®, a comprehensive simulator suite for gas,
steam and renewable power plants [14]. According to Liu and Karimi [15], Thermoflow® is one of
the preferred computer programs for studying power plants. Nord, Martelli and Bolland [6] and
Vidoza et al. [8] exemplify the use of Thermoflow® in the design and optimization of energy systems
for offshore production. GT PRO® is used to generate cycle heat balance as well as physical design
of major equipment and balance-of-plant. The off-design performance of the modelled plant is then
simulated in GT MASTER® according to control set-points, loads, and ambient conditions. Modelling
a power block in GT PRO® involves selecting the SCGT built-in model calibrated from manufacturers
data, which is capable to simulate design and off-design performance for a given load, fuel, and
5
environmental condition. On the other hand, WHRU is modelled from real project data and HRSG is
modelled based on knowledge as no real project data is available (see Table 3). ST is multi-valve
type, with 14 stages, 2.5% inlet valve pressure drop, 0.05% miscellaneous auxiliary load, 0.25%
mechanical loss, 1% inlet leakage, and 90% leakage readmission.

Table 3. Design assumptions of the heat recuperator equipment

Equipment WHRU HRSG
Type Simple recovery Once-through
Arrangement Vertical Vertical
Pressure level Single Single
Pressure drop (Pa) 2400 1900
Tube arrangement Staggered Staggered
Tube material ASTM A335 Gr. T11 ASTM A213 Gr. T22
Number of flow passes 16 33
Number of tubes per pass 23 30
Longitudinal pitch (mm) 110 79
Transversal pitch (mm) 127 92
Fin type Serrated Serrated
Fin height (mm) 11 15
Fin thickness (mm) 1.5 1
Fin spacing (mm) 3.5 3.5
Fin material AISI 409 AISI 409

The simulation of the Siemens SGT-A35 gas turbine with waste heat recovery unit operating with
fuel compositions from exporting/importing pipeline (Mode A) and internal production (Mode B)
resulted in the following data (see Table 4):

Table 4. Simulated performance of Siemens SGT-A35 with waste heat recovery unit
Load (%) 25 50 75 100
Plant net power (MW) 7.36 14.83 22.22 29.51
Plant net LHV electric efficiency (%) 20.63 29.62 34.49 36.65
CHP LHV efficiency (%) 72.05 73.92 75.15 75.95
GT exhaust temperature (°C) 465 473 490 524
Stack temperature (°C) 169 180 191 204
Fuel flow – Mode A (kg/s) 0.788 1.106 1.423 1.779
Fuel flow – Mode B (kg/s) 0.789 1.107 1.424 1.780
Hot water production capacity (kg/s) 84 101 119 144
CO2 emission - Mode A (kg/s) 2.126 2.984 3.841 4.801
CO2 emission - Mode B (kg/s) 2.135 2.996 3.856 4.819

3.2.1. Multi-objective optimization of the GE LM6000PF+ combined-cycle

Designing SBC for offshore application is a challenging task, which involves minimizing at the same
time the occupied area, capital cost and CO2 emission. Multi-objective optimization is a multi-
criterion decision-making technique used to determine the vector of design variables within the

6
feasible region that represents the optimal solutions to a given problem, i.e., the vector of solutions
that minimize or maximize simultaneously multiple functions subject to a set of constraints. Multi-
objective optimization can be expressed mathematically as follows, where k is the number of
objective functions 𝑓(𝑥⃗), 𝑥⃗ is the vector of design variables, m is the number of inequality constraints
𝑔(𝑥⃗), and p is the number of equality constraints ℎ(𝑥⃗).

𝑀𝑖𝑛𝑖𝑚𝑖𝑧𝑒: 𝐹(𝑥⃗) = {𝑓1 (𝑥⃗), 𝑓2 (𝑥⃗), … , 𝑓𝑘 (𝑥⃗)}

𝑆𝑢𝑏𝑗𝑒𝑐𝑡 𝑡𝑜: 𝑔𝑖 (𝑥⃗) ≥ 0, 𝑖 = 1,2, … 𝑚 (1)
ℎ𝑖 (𝑥⃗) = 0, 𝑖 = 1,2, … 𝑝

Multi-objective optimization in engineering and industry is often very challenging to solve,

necessitating sophisticated techniques to tackle. Metaheuristic approaches have shown promise and
popularity in recent years [16]. The genetic algorithm has been applied extensively to solve various
practical industrial problems [17]. A multi-objective optimization algorithm based on NSGA-II is
applied to search for the design variables vector that minimizes the following objective functions:

• Objective 1: total area per net power, m²/MW

• Objective 2: capital cost per net power, US$/kW
• Objective 3: hourly CO2 emission per net power, kg/MWh

The optimization framework consists of interlinking GT PRO® and MATLAB® [18] to run the MOO
NSGA-II algorithm. The following parameters were set: generation size 400, population size 50,
crossover fraction 0.8, constraint tolerance 0.001 and function tolerance 0.0001. The design variables
are steam pressure and temperature before steam turbine stop valve. The following constraints were
considered in this work:

• 1000 kPa ≤ steam pressure before ST stop valve ≤ 9000 kPa abs
• 200 °C ≤ steam temperature before ST stop valve ≤ 520 °C

The MOO problem resulted in the following normalized non-dominated solutions (see Figure 5). Red,
green and blue dots are the projection of the solutions (black dots) in the respective planes.

Figure 5. Pareto front solutions

Thus, the MOO was performed based on the premises and assumptions mentioned above. The design
variable set that minimizes the distance between zero and normalized points is considered the solution
of compromise among the non-dominated solutions that represent the trade-off between various
conflicting objectives. The distance from each point to zero is calculated by the simple spatial
geometry formula given below, where 𝑁𝑂𝐹𝑖 means the normalized value of the 𝑖 − 𝑡ℎ objective.
7
 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 = √𝑁𝑂𝐹12 + 𝑁𝑂𝐹22 + 𝑁𝑂𝐹32 (2)

Table 5 presents the Pareto front solution that minimizes the normalized distance (which also
minimizes objective 2), as well as solutions that minimize occupied area and emissions.

Table 5. Optimized SBC inlet steam pressure and temperature

Norm. Norm. Norm. Norm.
Inlet steam Obj. 1 Obj. 2 Obj. 3
obj. 1 obj. 2 obj. 3 distance
kPa (abs) °C m²/MW US$/kW kg/MW.h - - - -
2510 458 2.857 704.4 409.5 0.60 0.00 0.24 0.64
2140 436 2.842 706.8 410.1 0.90 0.42 0.00 0.99
1740 476 2.899 710.0 408.3 0.00 1.00 0.92 1.36

The solution selected (called “pseudo-optimum” because it does not eliminate the possibility of better
unknown solutions) was simulated and the obtained performance is shown in Table 6. It is noteworthy
that the fuel gas from the power hub is supplied only by the exporting pipeline.

Table 6. Performance of the optimized CCGT power block with GE LM6000PF+ gas turbine
Load (%) 25 50 75 100
Plant net power (MW) 20.33 35.42 47.92 61.76
GT gross power (MW) 11.92 23.72 35.36 46.88
Plant net LHV electric efficiency (%) 37.01 44.57 49.50 52.43
Fuel flow (kg/s) 1.214 1.756 2.139 2.603
GT exhaust temperature (°C) 496 532 498 512
Stack temperature (°C) 136 140 149 155
ST inlet flow (kg/s) 9.16 12.49 13.44 15.64
CO2 emission (kg/s) 3.274 4.737 5.771 7.023

3.4. Power block operating strategy

The overall performance of SCGT plants are influenced also by its operation strategy. Riboldi and
Nord [7] discussed the relationship between load and thermal efficiency in parallel CCGT operation.
According to authors, the best choice for very high and low power outputs is a uniform load share
between CCGT. However, in intermediate power outputs the thermal efficiency is maximized by
keeping one CCGT load at high levels, while the other unit handles the remaining power output.
Despite the potential benefit of operating parallel CCGT at very different load, operators normally
maintain CCGT at the same load for stability reasons. This work assumes that parallel SCGT and
CCGT load is equally split and the smallest amount of power blocks are operated to accomplish a
certain service. Regarding to Case 2, the first level of the operating hierarchy is the SCGT in the
production platform, in which the required heat governs its load. HWB supplements the hot water
production when required, as well as the power hub supplies power to the production platforms.

3.5. Hot water boiler

Design and performance data of commercial HWB are hard to find in the literature. For this reason,
this work assumes a simplified approach to model this equipment. A LHV thermal efficiency equal

8
to 95% and First Law equations were used to obtain the overall off-design performance and emissions
(see Table 7). The obtained results are in line with commercial HWB of similar capacity [19].

Table 7 - Hot water boiler performance

Parameter/Load (%) 25 50 75 100
Heat supply (MW) 2.88 5.75 8.62 11.50
Thermal efficiency (%) 95.0 95.0 95.0 95.0
Fuel flow – Mode A (kg/s) 0.067 0.135 0.202 0.271
Fuel flow – Mode B (kg/s) 0.067 0.135 0.202 0.271
CO2 emission - Mode A (kg/s) 0.183 0.367 0.551 0.734
CO2 emission - Mode B (kg/s) 0.182 0.366 0.550 0.733

3.6. Exergy efficiency and carbon emission

The exergy of flows was calculated for the reference environment and standard chemical exergy
proposed by [20]. The exergy effects caused by nuclear, magnetic, electric, and surface tension forces
were neglected due to its irrelevance to the case study. The specific exergy 𝑏 is defined as the sum of
the following exergy components: potential 𝑏𝑝𝑜𝑡 , kinetic 𝑏𝑘𝑖𝑛 , thermomechanical or physical 𝑏𝑝ℎ ,
and chemical 𝑏𝑐ℎ , according to equation (1).

𝑏 = 𝑏𝑝𝑜𝑡 + 𝑏𝑘𝑖𝑛 + 𝑏𝑝ℎ + 𝑏𝑐ℎ (1)

There are some definitions in the literature to calculate the exergy efficiency of a given energy
conversion process. The exergy efficiency of the power and heat system on offshore platforms can
be determined by equation (2).

Useful exergy effect 𝐵̇𝑒𝑙 + 𝑚̇𝑤 (𝑏𝑤

𝑜𝑢𝑡 𝑖𝑛
− 𝑏𝑤 )
𝜂𝑒𝑥 = = (2)
Driving exergy 𝑚̇𝑓 𝑏𝑓

The lifespan exergy efficiency of the utility plant configurations is shown in Figure 6. A continuous
increase in the exergy efficiency of the conventional plant (Case 1) is observed following the
behaviour of the thermal demand curve (see Figure 1).

4. Results
The performance indicators of the conventional pre-salt platform (Case 1) and the proposed power
hub (Case 2) are summarized in Table 8. Case 2 presented Second Law efficiency throughout the
oilfield lifespan 8.2p.p. above Case 1, thus reducing the fuel gas consumption and increasing the
exportation of natural gas by 1.74 million ton and avoiding the CO2 emission by 4.75 million ton.

Based on Figure 6, one can note that exergy efficiency of Case 1 varies significantly compared to
Case 2 in the production beginning as well as when the oilfield becomes mature. Generically, Case 1
is characterized by decentralized plants prone to partial load operation, whereas Case 2 benefits either
by a relatively flat HWB efficiency and power blocks operation closer to the design point. For
instance, the largest exergy efficiency gap occurs in the 25th year (16.3p.p.), when Case 1 has six
SCGT-WHRU power blocks with average load of 60.3%, whereas Case 2 has two CCGT power
blocks at 86.4% average load to supply the entire power demand and HWB for heat purposes.

9
Table 8 – Utility plant performance indicators comparison throughout the oilfield lifespan
Key performance indicator Case 1 Case 2
Average exergy efficiency (%) 35.8 44.0
6
Overall fuel gas consumption (10 ton) 9.98 8.24
6
Overall CO2 emission (10 ton) 27.00 22.25

In fact, larger inefficiencies are expected in mature oilfields [21]. Even considering an optimized
CCGT power block in this work, previous work from Freire and Oliveira Jr [9] evidenced an average
exergy efficiency gain of 9.1p.p. for a configuration similar to Case 2. However, it is noteworthy that
in the referenced work the power blocks have different gas turbine models. Despite the peculiarities
of the performance curves, the results obtained in this work is in line with those from the previous
work and also suggest the need for an appropriate machinery selection process.

Figure 6 – Comparative exergy efficiency (left) and CO2 emissions (right)

Despite the promising results, the commercial viability of power hubs depends on numerous factors,
such as reliability not only in converting fuel energy into electricity, but also in supplying it to
production platforms. Freire and Oliveira Jr [9] presented a simplified incremental financial analysis
of energy hubs capable of producing part or all of the electric energy required by a set of four
platforms, including a sensitivity analysis of the net present value to the taxation of CO2 emissions.
According to the authors, power hubs can be financially viable in a scenario without CO2 taxation,
although the adoption of this component radically increases the attractiveness of this solution.

5. Conclusion

Exergy efficiency and total CO2 emission over the oilfield lifespan were used to compare the
performance of utility plants in a production cluster with four platforms in the Brazilian pre-salt
region. The proposed CCGT power hub increased the second-law efficiency and, thus, reduced the
environmental impact of the offshore oil and natural gas production. A hybrid concept composed of
a power hub equipped with optimized combined-cycle power blocks and production platforms with
simple-cycle gas turbines and supplementary hot water boilers increased the lifespan average exergy
efficiency by 8.2p.p. when compared to the configuration used in some commercial projects. The fuel
gas consumption over the oilfield lifespan was reduced by 1.74 million ton and this reduction would
avoid the emission of 4.75 million ton of carbon dioxide. This expressive environmental impact
reduction is in line with operator objectives in economies with growing interest in carbon taxation.

10
Acknowledgments
The second author acknowledges the Brazilian National Council for Scientific and Technological
Development (CNPq), grant 304935/2016-6.

Nomenclature
𝐴 boiler capacity, kW
b specific exergy, J/kg
𝐵̇ exergy rate, W
LHV lower heating value, MJ/kg
𝑚̇ mass flow rate, kg/s
NOF normalized objective function
OPEX operational expenditure, US$/MWh
𝑃 operating pressure, barg
R revenue, US$
ROCE return on capital employed, %
Abbreviations
CCGT combined-cycle gas turbine
CHP combined heat and power
FPSO Floating, production, storage and offloading
GE General Electric
HRSG heat recovery steam generator
HWB hot water boiler
MOO multi-objective optimization
SBC steam bottoming cycle
SCGT simple-cycle gas turbine
ST steam turbine
WHRU waste heat recovery unit
Greek symbols
η efficiency
Subscripts and superscripts
1,2,3 objective functions
ch chemical
el electric
ex exergy
f fuel
in inlet
kin kinetic
out outlet
ph physical
pot potential
w water

11
References
[1] U.S. Energy Information Administration, Offshore production nearly 30% of global crude oil
output in 2015 – Available at: <https://www.eia.gov/todayinenergy/detail.php?id=28492>
[accessed 27.02.2019].
[2] Anuário Estatístico Brasileiro do Petróleo, Gás Natural e Biocombustíveis: 2016. Rio de Janeiro,
Brazil: National Agency of Petroleum, Natural Gas and Biofuels; 2016 Technical Report No.:
ISSN 1983-5884. (in portuguese).
[3] Kloster P., Energy Optimization on Offshore Installations with Emphasis on Offshore Combined-
cycle Plants. In: Proceedings of the 1999 Offshore Europe Oil and Gas Exposition and
Conference; 1999 Sep 7-10; Aberdeen, Scotland. Society of Petroleum Engineers Inc.
[4] Nguyen T.-V., Tock L., Breuhaus P., Maréchal F., Elmegaard B. Oil and gas platforms with
steam bottoming cycles: System integration and thermoenvironomic evaluation. Applied
Energy 2014;131:222–237.
[5] Følgesvold E., Skjefstad H. S., Riboldi L., Nord L. O. Combined heat and power plant on offshore
oil and gas installations. Journal of Power Technologies 2017; 97(2):117–126.
[6] Nord L. O., Martelli E., Bolland O. Weight and power optimization of steam bottoming cycle for
offshore oil and gas installations. Energy 2014; 76:891-898.
[7] Riboldi L. Nord L. O. Lifetime Assessment of Combined-cycles for Cogeneration of Power and
Heat in Offshore Oil and Gas Installations. Energies 2017; 10(6), 744.
[8] Vidoza J. A., Andreasen J. G., Haglind F., Reis M. M. L, Gallo W., Design and optimization of
a power hub for Brazilian off-shore oil production units. Energy 2019; 176:656-666.
[9] Freire R. L. A., Oliveira Jr S., Technical and economic assessment of power hubs for offshore oil
and gas application. In: Proceedings of the ECOS 2019 – The 32nd International Conference on
Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems; 2019
Jun 23-28; Wroclaw, Poland.
[10] Silva J. A. M, de Oliveira Jr S. Unit exergy cost and CO2 emissions of offshore petroleum
production. Energy 2018;147:757-766.
[11] Gallo W. L. R., Gallego A. G., Acevedo V. L., Dias R., Ortiz H. Y., Valente B. A., Exergy
analysis of the compression systems and its prime movers for a FPSO unit. Journal of Natural
Gas Science and Engineering 2017;44:287–298.
[12] http://www.wmo.int/pages/prog/wcp/wcdmp/GCDS_1.php [accessed 29.09.2019].
[13] http://www.inmet.gov.br/portal/index.php?r=clima/graficosClimaticos [accessed 29.09.2019].
[14] Thermoflow Inc. GT Pro and GT Master Version 28.2. 2019.
[15] Liu Z., Karimi I. A. Simulating combined cycle gas turbine power plants in Aspen Hysys.
Energy Conversion and Management 2018; 171:1213-1225.
[16] Yang X.-S., Nature-inspired optimization algorithms. Amsterdam: Elsevier; 2014.
[17] Pintarič Z. N., Kravanja Z. Towards Outcomes-Based Education of Computer-Aided Chemical
Engineering. In: Proceedings of the ESCAPE 26 – The 26th European Symposium on Computer
Aided Process Engineering; 2016 Jun 12-15; Portorož, Slovenia.
[18] Mathworks, Matlab®, Version R2019a; 2019.
[19] Bosch Industriekessel GmbH. Efficient technology
for major tasks. Hot water boilers. – Available at: <https://www.bosch-thermotechnology.com/>
[accessed 03.01.2020].
[20] Szargut J., Morris D.R., Steward F.R., Exergy analysis of thermal, chemical and metallurgical
processes. New York: Hemisphere Publ. Corp; 1988.
[21] Nguyen T.-V., Jacyno T., Breuhaus P., Voldsund M., Elmegaard B. Thermodynamic analysis
of an upstream petroleum plant operated on a mature field. Energy 2014;68:454-469.
12