Design optimization model for the integration of Renewable and
1
Nuclear energy in the United Arab Emirates’ power system
2 3 4
Ali Almansoori*, Alberto Betancourt-Torcat
5
Department of Chemical Engineering, The Petroleum Institute, Abu Dhabi, P.O. Box 2533,
6
United Arab Emirates
7 8
Abstract
9
A mixed integer linear programming (MILP) formulation is presented for the optimal design of the
10
United Arab Emirates’ (UAE) power system. The model was formulated in the General Algebraic
11
Modeling System (GAMS), which is a mathematical modeling language for programming and
12
optimization. Previous studies have either focused on the estimation of the UAE’s energy demands or the
13
simulation of the operation of power technologies to plan future electricity supply. However, these studies
14
have used international simulation tools such as “MARKAL” and “MESSAGE”; whereas the present
15
work presents an optimization model. The proposed design optimization model can be used to estimate
16
the most suitable combination of power plants under CO2 emission and alternative energy targets, carbon
17
tax, and social benefits of air emissions avoidance. Although the proposed model was used to estimate the
18
future power infrastructure in the UAE, the model includes several standard power technologies; thus, it
19
can be extended to other countries. The proposed optimization model was verified using historical data of
20
the UAE power sector operation in the year 2011. Likewise, the proposed model was used to study the
21
2020 UAE power sector operations under three scenarios: domestic vs. international natural gas prices
22
(considering different carbon tax levels), social benefits of using low emission power technologies (e.g.,
23
renewable and nuclear), and CO2 emission constraints. The results show that the optimization model is a
24
practical tool for designing the UAE power infrastructure, evaluating future production technologies and
25
scenarios, and identifying key parameters affecting the UAE power sector.
26
Keywords: Optimization, UAE, Renewable energy, Nuclear energy, Power system.
27
*
Corresponding author: e-mail:
[email protected].
1
28
1. Introduction
29
The current increasing social pressures on global warming issues and high oil prices have
30
attracted the international attention. Social pressure aims to prevent serious impacts on both the
31
environment and economic growth. According to the International Energy Agency (IEA), in
32
2009 approximately 68% of the electricity generated originated from fossil fuels such as: coal
33
(40.6%), natural gas (21.4%) and oil (5.1%). The remaining share of electricity was produced
34
from hydro (16.2%), nuclear (13.4%) and renewable sources (3.3%) [1, 2]. The production of
35
electricity from fossil fuels is higher in the developing world. Thus, the use of renewable and
36
cleaner energy sources is needed to secure electricity supply in developing regions, including the
37
United Arab Emirates (UAE).
38
The UAE’s power sector completely depends on conventional fossil fuels. For example, in 2009
39
approximately 98% of the electricity was generated using natural gas-based power plants [3]. On
40
the other hand, electricity demand growth has accelerated in recent years to 9% [4]. Although the
41
country holds one of the largest energy endowments in the world [5]; it became a net importer of
42
natural gas in 2007 [6]. The increasing gas requirements result, in part, from domestic gas
43
resource constraints (i.e., high sulfur content). The country’s gas shortage will continue growing;
44
unless, new domestic gas resources are exploited, or alternative energy sources are introduced to
45
supply the national power grid.
46
Between the years 1990 and 2008, the UAE’s CO2 emissions grew from 60.8 million tonnes
47
(MT) to 146.9 MT. The extended use of fossil fuels in the UAE’s power sector is expected to
48
increase the share of CO2 emissions produced by the sector. Currently the power sector alone
49
contributes approximately to 50% of the total UAE’s CO2 emissions [3]. Despite UAE’s fossil
50
energy abundance, the country has acknowledged the importance of environmental conservation.
2
51
Accordingly, the country has taken necessary measures such as: environmental conservation
52
programs [7], zero flaring targets [8], and steps towards sustainable energy transition (e.g.,
53
Masdar Initiative [9-11] and the Estidama building code [12-14]).
54
Over the years important efforts and contributions have been made to optimize the operation of
55
electric power systems; both in academic research and industrial applications. Davidson et al.
56
[15] developed a mathematical model to optimize the operation of a power system. The
57
optimization was based on the selection of loading modes in power generating units by price
58
bids. Varympopiotis et al. [16] investigated the potential advantages of fuel switching in power
59
plants according to operational and financial criteria and conditions. The optimal switch timing is
60
derived to ensure increasing yields of an average capacity power plant. Moreover, stochastic
61
programming models have been developed to deal with uncertainties related to power system
62
planning. For example, Li and Huang [17] formulated a multistage stochastic model for planning
63
electric power systems and manage greenhouse gas (GHG) emissions. That model can be used to
64
determine electricity generation schemes and capacity expansion plans under GHG mitigation
65
strategies. Pereira and Pinto [18] presented a methodology for the solution of multistage
66
stochastic problems. The methodology was based on the approximation of the expected-cost-to-
67
go functions by piecewise linear functions. Piao et al. [19] developed a stochastic simulation
68
optimization model for energy system planning. The model can predict electricity demand using
69
support-vector-regression and Monte Carlo simulation, and optimize energy allocation.
70
Multiperiod optimization models have also been developed to determine the optimal pathway in
71
long term energy planning. For example, Zhang et al. [20-22] have used a multiperiod
72
optimization superstructure for optimal planning in the power sector. The model considers the
73
construction and decommissioning of power plants as well as carbon capture and storage (CCS)
3
74
methods on coal-fuelled plants. Furthermore, multi-region optimization models have been
75
developed to capture the differences in the power sector among regions. Cheng at al. [23]
76
proposed a multi-region model for optimal planning in the power sector. The model considers
77
spatial distribution of resource, generation and demand. Hoster [24] and Vorspools and
78
D’haeseleer [25] developed models based on the inter-connections of European countries’ power
79
grids. Those models were used to analyze CO2 emission control policies. Watcharejyothin and
80
Shrestha [26] built a MARKAL-based model for electricity trading between Laos and Thailand.
81
The modelling focused on the environmental impacts of CO2 emissions. Also, several multi-
82
regional power models have been developed to study China’s power sector. Those multi-regional
83
models usually consider electricity market integration [27] and emission mitigation strategies
84
[28-30]. Similar modeling studies for the UAE energy sector have been reported in the literature.
85
AlFarra et al. [31] using the “MESSAGE” model estimated the CO2 emissions that could be
86
avoided using nuclear energy, renewable energy, and CCS in the UAE by 2050. Also, Sgouridis
87
et al. [14] developed an energy-financial model for the UAE. That model considers the coupled
88
nature of energy production and water desalination, and the associated trade-offs. Mondal et al.
89
[32] evaluated future energy-supply strategies for the UAE using the “MARKAL” energy model.
90
Although previous studies have been made on the power sector, the present work differs from
91
previous studies because it includes key economic and social measures. Such measures include
92
energy pricing and social benefits of emissions reductions. Both of these measures aim to
93
mitigate air emissions in the optimal design of the power system. The proposed optimization
94
model aims to assess the process economics of the UAE’s power system using a steady-state
95
model. Conventionally, most of the models used in Chemical Engineering to examine the
96
process economics of a system are first developed at steady-state, especially those systems that
4
97
involve large-scale plants [33-35]. This approach was followed in the present study to assess the
98
economics of the UAE’s power system since it includes large power utility-scale production. The
99
approach represents the traditional method used to evaluate the process economics in energy
100
systems. Although the process economics can be assessed using models that involve time [36-
101
38], the development of such time-dependent model usually requires a prior steady-state
102
mathematical formulation. Based on the above, the present model aims to provide a stationary
103
analysis of the UAE power system.
104
The proposed design optimization model can account for carbon tax and reduced emission social
105
benefits. Both strategies involve the use of CCS systems and alternative energy sources.
106
Additionally, the model takes into account CO2 emission targets as well as renewable and
107
nuclear energy targets. The problem’s dilemma stems from using the design optimization model
108
to minimize UAE’s power costs vs. cost increases due to the adoption of emission mitigation
109
strategies. This dilemma arises because environmental protection is a public good, and public
110
good are underprovided by markets [39].
111
The optimization model proposed in this work was verified using data from the Abu Dhabi
112
Water & Electricity Company [40] and the Statistics Centre of Abu Dhabi [41]. Moreover, three
113
case studies considering different techno-economics (alternative power targets and gas prices),
114
environmental (CO2 emission targets), and fiscal policy (carbon tax) parameters in the year 2020
115
are presented in this work.
116
This article is organized as follows: Section 2 presents the main features of the power
117
optimization model proposed in this work. Section 3 presents a case study for the year 2011;
118
which was used for the verification of the design optimization model. Section 4 considers three
119
case studies to design the optimal UAE’s power system according to: i) local and international
5
120
natural gas prices at different carbon tax levels, ii) considering the social benefits of air
121
emissions avoidance in economic terms, and iii) including a CO2 emission target for the year
122
2020. Concluding remarks and future work are presented in Section 5.
123
2. Model formulation
124
This section presents the main features of the proposed model to design the optimal
125
infrastructure of the UAE’s power sector. The superstructure of the optimization model is shown
126
in Figure 1. From left to right across the figure can be found the main components of the model’s
127
superstructure: 1) the energy input sources (energy-supply side) shown as colored ovals (i.e.,
128
natural gas, wind, solar, and nuclear). 2) The power plants denoted by colored boxes (i.e., natural
129
gas-based [42-47], wind turbines [48-50], solar-based [51-53], and power reactors [54-57]). 3)
130
The outputs represented by the product (power) and by-products (air emissions and nuclear
131
waste) displayed at the center. 4) The end-users (energy-demand side) shown on the right hand
132
side end (country’s economy sectors); whose demands are met by the power plants.
133
2.1. Problem Statement
134
Given are a set of power production technologies with their corresponding capacities and air
135
emission factors. Also, given are the capital, fuel and operating costs for each technology. In this
136
work, the integration of renewable and nuclear energy in the UAE’s power sector is formulated
137
as a Mixed Integer Linear Programming (MILP) model. The aim is to determine the optimal
138
power infrastructure, while meeting the electricity demand under environmental constraints. The
139
problem is associated to a sustainable energy transition path for the UAE power sector, as part of
140
a larger national energy transition strategy adopted by the country [58].
141
2.2. Inputs
6
142
In the proposed power optimization model the inputs are represented by: the total electricity
143
demand expected for a given year in the UAE, the carbon dioxide emission target, the expected
144
shares of renewable and nuclear power, and the minimum/maximum number of power plants
145
available per type of technology. Figure 2 shows the key model inputs and outputs considered in
146
the present energy model. The electricity demand value, CO2 tax and emission target [58, 59],
147
fuel price, and alternative power share values are obtained from energy forecasts. On the other
148
hand, the minimum/maximum number of power plants available for deployment can be defined
149
by the user. Accordingly, the total electricity demand is formulated using the following
150
constraint:
151
TEP TCP ED
152
where TEP represents the total electricity produced by the power infrastructure (see (A.7) in
153
Appendix A for details), TCP is the total compression power used in carbon capture and storage,
154
and ED is a model input representing the total electricity demand. Additionally, the CO2
155
emission constraint is given as:
156
TCE CET
157
where TCE represents the carbon dioxide equivalent generated by the power plants’ fleet (see
158
(A.9) in Appendix A for details), and CET is an input that represents the maximum allowable
159
CO2 emission for the county’s power system (i.e., CO2 emission target) for a specific year. The
160
present work only takes into account the CO2 emissions generated during the power production
161
process.
162
The set of power technologies is denoted in the model by the index p, p g , w, s, n , where g
163
identifies the natural gas-based power plants, w the wind turbine farms, s the solar-based plants,
(1)
(2)
7
164
and n the nuclear plants. The share of alternative (i.e., wind, solar and nuclear) power required in
165
the fleet is constrained as follows:
166
AEp IEp ICp UCFp ,
167
where AEp is a model input that represents the minimum installed generation capacity expected
168
from power plant type p in a given year, IEp is an integer variable that denotes the number of
169
plants p selected for the power infrastructure, ICp represents the plant’s p installed capacity, and
170
UCFp indicates the unit capacity factor for plant p.
171
The number of power plants available per type of technology can be constrained as follows:
172
EpU IEp EpL ,
173
where E pU and E pL are upper and lower bounds for the variable IE p .
174
2.3. Electricity Production
175
The present optimization model considers conventional (i.e., fired-gas plants) and alternative
176
(i.e., renewable and nuclear) power production technologies. Both technologies are used for
177
planning the optimal power infrastructure of a country (e.g., UAE) under environmental and
178
supply constraints. Accordingly, the power production balance by technology is given as
179
follows:
180
EPp IEp ICp CFp ,
181
where EPp represents the amount of electricity produced using the power plant p, whereas ICp
182
and CFp are model’s parameters that represent the installed capacity and capacity factor of the pth
183
power plant, respectively. Furthermore, the total balance of natural gas consumed (TNG) by the
184
gas-based power plants can be estimated as follows:
185
TNG IEp NGp , p
p w, s, n
p
(3)
(4)
p
p g
(5)
(6)
8
186
where NGp is the amount of natural gas consumed by the power plants g.
187
2.4. Greenhouse Gas (GHG) emissions and Criteria Air Contaminants (CAC)
188
The quantification of the GHG and CAC emissions from the power infrastructure is a key
189
environmental factor; which has been considered in the present optimization model. Only natural
190
gas-based plants are considered to generate air emissions. The emissions considered in the power
191
model are defined by set e, which is included as an index in the formulation of the problem.
192
Accordingly, the set of air emissions is given as follows: e CO2 , CH 4 , N 2O, NOx, SO2 , PM 10 .
193
Where the term CO2 correspond to the carbon dioxide emissions, CH4 methane, N2O nitrous
194
oxide, NOx nitrogen oxides, SO2 sulfur dioxide, and PM10 coarse particles (2.5 μm to 10 μm in
195
size). Accordingly, the GHG emissions (e.g., CO2, CH4 and N2O) and CAC (e.g., NOx, SO2 and
196
PM10) balance considered in the model can be calculated as follows:
197
EGe ,p EPp AEFe,p ,
198
where EGe,p is the amount of emission e generated by plant p, and AEFe,p is a parameter that
199
represents the air emission e produced by the pth power plant. The amount of CO2 eq. emission
200
can be constrained according to governmental regulations or global climate change agreements.
201
An important feature included in the present power optimization model is the carbon tax. The
202
carbon tax is an incentive-based policy that levies taxes on burning fossil fuels or emitting
203
carbon dioxide [60]. The motivation is to reflect the negative externalities caused by fossil-based
204
electricity generation, but not directly accounted for in energy prices. As a result, the present
205
model includes a financial penalty per unit of CO2 eq. generated in the power sector as follows:
206
ECc ,p CE p CTAX t ,
c tax , dis
e, p
(7)
c tax , p g
(8)
9
207
where the index c represents the type of externality associated to the air emissions. The
208
externality can stand for: 1) added costs in terms of a carbon tax “tax” paid due to the generation
209
of air emissions (i.e., extra cost on power production). 2) Cost discounts “dis” as a result of
210
emission abatements (i.e., deduction in the power production cost). In this particular case, Eq. 8
211
represents the carbon tax associated costs. Moreover, ECc,p represents the externality c associated
212
to the power plant p, CEp is the CO2 eq. produced by the pth plant (see (A.8) in Appendix A for
213
details), CTAX is a parameter that indicates the amount of money paid per unit of CO2 eq.
214
($/CO2 eq.) emitted, and t is the annual operating hours of the plants.
215
2.5. Carbon dioxide capture systems (CCS)
216
The CO2 capture systems enable the reduction of the carbon dioxide emissions to the
217
atmosphere. The CCS considered in the present work include pre-combustion (i.e., for NGCC
218
plants) and post-combustion (i.e., for Oxyfuel plants) as carbon capture methods. Additionally,
219
CO2 injection in oil fields for enhanced oil recovery (EOR) and permanent sequestration in
220
suitable geological sites (e.g., depleted oil fields) are considered as CO2 storage methods.
221
Accordingly, the balance of CO2 captured in the gas-based plants can be estimated as follows:
222
CC p EPp CCFp ,
223
where CCp represents the amount of CO2 captured in plants p using CCS methods, and CCFp is
224
the CO2 capture factor associated to the pth power plant. Accordingly, the annual cost associated
225
to the transportation of CO2 from the capture unit to the storage sites can be estimated as follows:
226
TCT CC p CTF PLp t ,
227
where TCT is the total CO2 transport cost, CTF is a parameter that denotes the transport cost per
228
unit of CO2 and length [61], PLp represents the pipeline length traveled by the CO2 captured at
229
plat p. Similarly, the annual CO2 sequestration cost (TCS) can be estimated as follows:
p
p g
(9)
p g
(10)
10
230
TCS CC p CSF t ,
231
where CSF is a parameter that denotes the unit CO2 underground sequestration cost [61]. The
232
total compression power balance required to transport via pipelines the CO2 captured is
233
calculated as:
234
TCP CC p CPF PLp ,
235
where TCP is the total compression power used to transport the captured CO2 to the
236
sequestration sites, and CPF is the compression power factor that denotes the amount of power
237
consumed per unit of CO2 and distance traveled.
238
2.6. Air emissions avoidance and associated social benefits
239
The avoidance of GHG and CAC emissions results in significant benefits to the society in terms
240
of both reduced public health issues and environmental damages. The emissions avoided using
241
alternative energy sources (renewables and nuclear), instead of fossil-based plants, are included
242
as a key model’s feature. Accordingly, the balances of the air emissions avoided are given as
243
follows:
244
AEAe ,p EPp ENGe ,
245
where AEAe,p is the amount of air emission e avoided using the pth plant ( p g ), and ENGe is a
246
parameter that indicates the average emission e that would be generated using conventional
247
NGCC plants (as those currently used in the UAE power sector). These latter emissions can be
248
avoided using alternative power plants. Accordingly, the annual social benefits or damage costs
249
avoided using alternative energy sources is given as follows:
250
ECc ,p AEAe,p ESC e t , e
p
p
p g
(11)
p g
p g
(12)
(13)
c dis , p g
(14)
11
251
where ECc,p represents in this case the discount cost c associated to the generation of clean
252
electricity from alternative power plants p, ESCe is a model parameter that defines the social cost
253
avoided by evading emission type e through the use of alternative energy plants. Similarly, the
254
annual social benefit obtained through fossil-based power plants with CCS methods can be
255
defined as follows:
256
ECc ,p CC p SCC t ,
257
where SCC is a parameter that specifies the avoided social cost by capturing CO2 emissions in
258
fossil-based power plants.
259
2.7. Power production Costs
260
The power production costs associated with the energy infrastructure are primarily composed of:
261
capital costs, operating and maintenance costs, fuel costs, and external production costs (e.g.,
262
carbon tax or social benefits of emissions avoidance). Additionally, cost penalty factors are
263
included to account for public concern issues such as nuclear energy risks. Accordingly, the
264
annual direct power cost (PCp) can be calculated as follows:
265
PCp IEp Cap p OMp FCp PPCp ECc ,p , c
266
where Capp represents the annual amortized fraction of the capital cost associated to the power
267
plant p, OMp denotes the operating and maintenance cost of the plants, FCp represents the fuel
268
costs of the plants, PPCp denotes the cost associated with the public perception on the
269
deployment of the pth power plant (e.g., nuclear reactor), and ECc,p represents the external cost c
270
associated to plant p. As previously discussed in Sections 2.4 and 2.6, the external cost ECc,p
271
may either represent an additional or discount cost; this will depend on the scenario under
272
analysis. For example, if the scenario considers the application of a carbon tax this will be
c dis , p g
(15)
p
(16)
12
273
represented as an additional cost to the power generation, whereas the considerations of air
274
emission avoidance as a social benefit will result in a discount over the overall power cost.
275
Accordingly, the general form of the capital cost can be given as:
276
Cap p PCFp ICp AFp ,
277
where PCFp represents the capital factor of the pth power plant, and AFp is the capital
278
amortization factor associated to plant p (see (A.14) in Appendix A for details). More details
279
about the capital cost per type of power plant are given in (A.11)-(A.13) of Appendix A.
280
Furthermore, the operating and maintenance costs of the power plants can be expressed as
281
follows:
282
OMp ICp CFp OMFp RRFp t,
283
where OMFp represents the operating and maintenance cost factor, and RRFp is the repair and
284
replacement cost factor. More specifications of the operating and maintenance costs for
285
individual power plants are provided in (A.15)-(A.17) of Appendix A. On the other hand, the
286
power plants’ fuel costs can be calculated as follows:
287
FCp ICp CFp HR p FCFp t ,
288
where HRp represents the heat rate of plant p, and FCFp indicates the fuel cost factor.
289
Additionally, the public perception cost associated to the pth power plant (e.g., nuclear) is given
290
as follows:
291
PPCp ICp CFp WCFp DCFp ECFp ,
292
where WCFp is a factor that denotes the waste repository costs (e.g., storage and disposal costs of
293
wasted nuclear fuel), DCFp represents the decommissioning costs, and ECFp denotes the external
294
costs generally paid by the community in relation to health, safety, and environment related to
p
(17)
p
(18)
p g, n
(19)
pn
(20)
13
295
the deployment of power plants p in its proximity. The latter type of cost was only considered for
296
nuclear power given its conflicting public perception.
297
2.8. Optimization Model
298
Based on the inputs, environmental factors, types of power plants and production costs discussed
299
in the previous sub-sections, the conceptual formulation of the design optimization model
300
considered in this study can be expressed as follows:
min CF p PCp TCT TCS
( 21)
η
301
subject to Power Demand Type of Power Plants Power Plants Installed Capacities Available Number of Power Plants Alternative Power Share target Carbon Dioxide (CO 2 ) emission target i ) Typesof Power Plants η ii ) Number of Power Plants iii ) Power Plants operating capacities
302
where CF is the model’s objective cost function that represents the annual power generation cost.
303
As shown in problem 21, CF is defined in terms of the direct power costs (PCp) and emission
304
mitigation costs (TCT and TCS) associated with power generation. The objective function is
305
given as an annualized cost ($/year). Also, as shown in problem 21, the formulation is subject to
306
the following constraints: power demand, types of power plants, plants installed capacities,
307
available number of plants, alternative power shares target, and CO2 emission target. Moreover,
308
the variable η denotes the set of decision variables in the design optimization model. The
309
variables’ set includes: the types of power plants, the number of power plants, and the plants’
310
operating capacities. The proposed optimization model searches for the most suitable 14
311
combination of power plants. Also, the number of power plants and corresponding production
312
capacities are determined for the optimal design of the UAE’s power system under
313
environmental constraints.
314
The resulting power optimization model (1)-(20) and (A.1)-(A.17) (see details for the latter
315
equations in Appendix A) is a Mixed Integer Linear Programming (MILP) model. The model
316
was developed in the General Algebraic Modeling System (GAMS) [62], and solved using the
317
CPLEX solver [63]. In the present work, all discussed solutions represent local optimal.
318
Although the CPLEX MIP algorithm is based on a branch-and-bound search that aims to find
319
global optimum by tree search; the algorithm in GAMS includes a default termination criterion
320
(as shown in Figure 2) based on the relative gap between the objective value of the “best
321
estimate solution” and “best integer solution”. The “best integer solution” is the best result found
322
that satisfies all integer requirements; whereas the “best estimate solution” provides a bound for
323
the optimal integer solution. Accordingly, when the relative gap between both objectives values
324
drops below GAMS default criterion, the algorithm terminates. The proposed mathematical
325
model can be used as a practical tool to: 1) design the optimal infrastructure of the UAE’s power
326
sector, 2) study the introduction of alternative energy sources in the UAE power grid, 3)
327
forecasts future power production scenarios, and 4) plan the expansion of the UAE’s power
328
production infrastructure.
329
3. Case Study 2011: Model Verification
330
The first step considered in the present work was the verification of the optimization model.
331
Accordingly, the mathematical model presented in the previous section was initially used to
332
simulate Abu Dhabi’s power sector operations in the year 2011. The year 2011 was selected for
333
verification purposes due to the availability of information for Abu Dhabi’s power sector. Such 15
334
information includes: 1) types of power plants with corresponding generation outputs [64], 2)
335
fuel consumption [65], 3) air emissions [65], and 4) average electricity unit cost [66]. The
336
Emirate of Abu Dhabi was selected in this study because it currently leads the country and
337
Middle East & North Africa (MENA) Region in alternative energy initiatives [9, 67]. This aspect
338
is key for the analysis of future power production scenarios in the coming years (see following
339
section). The proposed design optimization model was verified for a specific electricity
340
generation scenario. The scenario considers: 1) fixed type of power plants (see Table 1 and 2,
341
ENG1-ENG2 and ENG5-ENG6 [68-72]), and 2) gross electricity output per plant (EPp) [64]. Both
342
measures were specified according to the power sector’s operations in 2011. Consequently, only
343
the aforementioned gas-based power plants were considered for verification purposes. The inputs
344
for the 2011 case study are listed in Tables 2 and 3.
345
The present model verification did not account for carbon tax costs (8), CO2 capture costs (10)-
346
(11), and social benefits of emissions avoidance (14)-(15). These variables were not accounted
347
given that they were not in placed in 2011. Moreover, the CO2 emission target (2) was also
348
neglected because such a measure was not implemented in the year 2011. Accordingly, the
349
systems of equations composed by (1), (3)-(7), (9), (12)-(13), and (16)-(20) were considered for
350
the verification of the model. In the present verification, the number and type of power plants
351
were not considered as decision variables and they remained fixed during the calculations.
352
However, for the following case studies (Section 4) the power plants are considered to be
353
decision variables in the optimization process. Consequently, the present case study reproduces
354
the 2011 Abu Dhabi’s power sector operations only for verification purposes. This approach
355
enabled a comparison between the model’s key outputs and historical data for the power system
356
operations in 2011. 16
357
3.1. Verification Approach
358
In the proposed model the natural gas price is a key economic parameter. According to the
359
reported information, the prices of the gas consumed by Abu Dhabi’s power sector during 2011
360
are not clearly defined. The average gas price in Abu Dhabi can be estimated from the costs of
361
the domestic gas production and imports with their corresponding total volumes. However, only
362
data for the domestic gas supply price is publically available [73, 74]. On the other hand, the
363
average price for imported Qatari gas (i.e., contracted and uncontract [75]) reported in the
364
literature differs from one source to another. Accordingly, given the reservations on the natural
365
gas prices, it is worth performing a sensitivity analysis for average gas price deviations. The
366
point is to cover the whole range of probable natural gas prices. This method allows evaluating
367
the impact that these deviations may have on the levelized electricity cost. As a result, the
368
sensitivity analysis was performed assuming different average gas prices for 2011. The selected
369
average gas prices were calculated assuming the followings: i) a price of 1.50 $/MMBtu for the
370
contracted gas volumes, ii) the share of imported gas for the power sector was varied between
371
20-60%, and iii) the average price of domestic gas supply was assumed to be 4.20 $/MMBtu
372
[73]. Furthermore, for this sensitivity analysis, the effect of the uncontract gas volume (sold to
373
Abu Dhabi) was neglected. This latter volume is considered to be small compared with the
374
overall annual consumption rate in the Emirate.
375
3.2. Verification Results
376
Figure 3 shows a comparison between the levelized electricity costs obtained from the natural
377
gas price sensitivity analysis simulations and the reported historical electricity cost in Abu Dhabi
378
for 2011. As shown in the figure, the range of values obtained in the simulations are close and
17
379
even intersect with the historical value [66]. Accordingly, the compared values can be considered
380
to be in reasonable agreement. It should be noticed that only the average annual delivered cost of
381
electricity is reported in the literature. Accordingly, in order to make a fair comparison with the
382
average annual electricity cost obtained from the simulations; the average annual transmission
383
and distribution costs (0.011 $/kWh and 0.031 $/kWh, respectively [76]) were deducted from the
384
delivered cost (0.076 $/kWh [66]). Thus, following this procedure the net electricity generation
385
cost (0.034 $/kWh) was obtained. The net electricity generation cost is represented by a dashed
386
line in Figure 3. All costs are given in US$ (2010).
387
As shown in Figure 3, the historical electricity cost for 2011 agrees well with the simulations
388
results. For example, the maximum electricity generation cost obtained from the natural gas
389
sensitivity analysis was 0.040 $/kWh (i.e., maximum gas price); whereas the minimum cost was
390
0.029 $/kWh (i.e., minimum gas price). These values indicate that the historical electricity cost is
391
enclosed between the maximum and minimum gas price scenarios considered for the present
392
case study. Furthermore, the electricity cost (0.035 $/kWh) obtained using the mean natural gas
393
price (i.e., ~ 0.0027 $/MJ) matches the historical average electricity cost (0.034 $/kWh) for 2011.
394
Additionally, the verification results show that the fuel consumption (natural gas (6)) in Abu
395
Dhabi’s power sector was 319,746 Billion BTU/yr in 2011. There is not available data for Abu
396
Dhabi’s power sector fuel consumption alone. The data reported by the Statistics Centre of Abu
397
Dhabi reflects the combined fuel consumption of the water and power sector (i.e., 543,643
398
Billion BTU/yr) for 2011 [41]. However, the same source estimated that the power to water
399
production ratio was approximately 46.40 kWh/m3 in that year. This value corresponds
400
approximately to a fuel utilization ratio of 58.93% for power production in co-generation cycles
401
similar to those used in the UAE [77]. Accordingly, by making use of the fuel utilization ratio, 18
402
UAE’s historical gas consumption for the power sector alone can be estimated. As a result, the
403
fuel requirement for Abu Dhabi’s power sector alone (6) was estimated to be 320,369 Billion
404
BTU/yr. This value only differs in 0.2% with that of the verification simulation (see Table 4).
405
Following a similar approach, the CO2 emissions generated by the power sector alone (7) in 2011
406
was estimated from data reported in the literature (i.e., they depend on fuel consumption). Then,
407
this value was compared with the verification result. As shown in Table 4, the CO2 emissions
408
(16.97 MT/yr) from reported public data [65] agrees well with the verification result (17.14
409
MT/yr), i.e., they only differ in 1%. The results of the verification, for the mean natural gas price,
410
are reported in Table 4 along with the reported historical data for 2011 [65].
411
The 2011 power infrastructure and their corresponding generation outputs (5), average levelized
412
electricity cost, fuel consumption (6), and CO2 emissions (7) match reasonably well with those
413
reported in the literature [64-66]. Therefore, the optimization model presented in this work can
414
be used as a complementary tool to estimate: future power production scenarios, average power
415
costs, and future power production infrastructures. Additionally, the introduction of new power
416
plant technologies (e.g., nuclear and renewable) along with stringent environmental policies
417
(e.g., carbon tax and CO2 emission targets) can be analyzed using this mathematical tool.
418
4. Case Study 2020
419
The proposed design optimization model was also used to determine the most suitable power
420
production infrastructure, potential electricity costs, and emission mitigation strategies for Abu
421
Dhabi in the year 2020. This was done considering different operating scenarios. The year 2020
422
was selected because estimates of primary energy prices [32, 78], and projected power
423
configurations [79] are readily available in the literature. Also, the year 2020 represents a
19
424
landmark period for the introduction of alternative energies in Abu Dhabi’s power sector. Table
425
5 lists the key model inputs for this case study. The Abu Dhabi Water and Electricity Company
426
(ADWEC) baseline electricity gross demand forecast was used as key model input [80]. To
427
propose a realistic picture, all the power plants shown in Table 1, except for ESOL4 and ESOL6, are
428
considered for this case study. The disregarded technologies require sea depth of approximately
429
1,000 m to operate, whereas the average depth of UAE’s Arab sea is about 100 m. Also, the
430
share of renewable power is considered to be 7% of the total installed power capacity for 2020
431
(i.e., approximately 1,500 MW) [79, 81]. The share of renewable power is constrained as shown
432
in (A.1) of Appendix A. The minimum generation capacity for each power technology is listed in
433
Table 5.
434
The minimum gas-based power generation was assumed to be at least equal to that of the year
435
2012 [64]. This given that it represents the readily available gas installed capacity in the Emirate.
436
On the other hand, for the renewable power plants, the generation capacities were assumed
437
according to projections from the Masdar Initiative [82] and information available in the
438
literature [79]. Whereas the minimum and maximum number of nuclear power plants available
439
for the year 2020 were set to be 2 and 4, respectively [31, 83]. Additionally, no technological
440
advancements in the power technologies inventory while moving towards 2020 were considered
441
for this study. This is, the present analysis comprises a short-term forecast of the power
442
generation scenario in Abu Dhabi. Therefore, no major technological breakthroughs in such a
443
short period of time are considered to occur. The natural gas price was assumed to be at local
444
subsidized levels for the different analyzed case studies unless otherwise stated. The studied
445
scenarios are compared with a reference scenario (described in section 4.1.1). This was done to
446
evaluate the benefits of reduced emissions or improved economics. The studied scenarios
20
447
considered: fuel cost levels, tax penalties, social benefits of emission reductions, and
448
environmental policies. Furthermore, the optimization problem converged for the analyzed case
449
studies after 10 to 15 CPU seconds. The optimization was computed in an Intel(R) Core(TM) i5
450
M-560 with 2.67 GHz CPU processor and 4.00 GB of RAM memory machine. The solutions
451
were found after 4 to 8 iterations of the branch-and-bound algorithm.
452
4.1. Case study 1: Local subsidized natural gas price vs. international market price
453
The present case study showcases the main differences between adopting subsidized gas prices
454
or international market prices in Abu Dhabi’s power sector. The differences are evaluated in
455
terms of: the power infrastructure configuration, electricity costs, and emission mitigation levels.
456
Furthermore, a sensitivity analysis on the carbon tax value is conducted to evaluate its influence
457
on the final power plant’s configuration. Additionally, the carbon tax capacity to equate both gas
458
price scenarios is analyzed. The assumed carbon tax values are shown in Tables 6 and 7 along
459
with the scenarios’ main optimization results. Additionally, base cases, where the carbon tax is
460
assumed to be zero, are analyzed for both gas price scenarios.
461
4.1.1. Local subsidized natural gas price
462
The reference scenario in this work assumes local subsidized gas prices and no carbon tax, such
463
as the current business-as-usual (BAU) operation mode in Abu Dhabi’s power sector. However,
464
alternative energy targets are also included for this scenario. As shown in Table 6, for the
465
reference scenario approximately 80% of the power is generated using gas-based power plants.
466
On the other hand, nuclear and renewable power account for 14% and 6% of the total power
467
supply, respectively. The prevalence of gas generated power is due to the highly subsidized gas.
468
The fuel represents over 67% of the cost in gas-based generation, whereas the capital represents
469
approximately 27% of the cost. The remaining share of the cost corresponds to operating and 21
470
maintenance costs. Overall, this scenario’s context is similar to the current characteristics of Abu
471
Dhabi’s power sector. Consequently, the low fuel price favors the generation of gas-based
472
power. The share of electricity generated through nuclear and renewable plants allow meeting
473
the corresponding minimum expected installed capacity goals (see Figure 4). By comparing the
474
annual electricity costs and power outputs of the different technologies (see Table 6) the
475
following results were obtained: 1) Gas-based power represents approximately 58% of the total
476
annual cost while it produces 80% of the electricity. 2) Nuclear energy represents around 19% of
477
the total costs while it produces 14% of the power supply. 3) Renewable sources account for
478
23% of the cost, but they only contribute with 6% of the supply (see Table 6). Therefore, power
479
plants with lower levelized electricity costs (i.e., natural gas-based) are favored over those with
480
higher costs (e.g., nuclear and renewable).
481
Additionally, as the carbon tax increases from zero to 150 $/tonne CO2, the share of the carbon
482
tax on the total costs increases from zero to 34%. This shows the significant cost burden that a
483
carbon tax may add to the annual power cost. As a result, at the highest carbon tax rate the share
484
of gas generated power decreases to a minimum of 75%; whereas nuclear increases to a
485
maximum of 19%. Although, the number of nuclear facilities remained fixed at two units (see
486
Table 8), the nuclear plants switched from lower (ENUC2) to higher (ENUC1) generation capacity
487
plants. Which allows an increase in the share of nuclear power (e.g., carbon-free) without
488
incurring in significantly larger overhead costs. For example, significant overhead costs would
489
result if a third nuclear facility is built. On the other hand, the amount of power generated from
490
renewable plants remained fixed. Renewable power is not affected by the carbon tax (i.e., they
491
are carbon-free). Additionally, the generation costs of renewable power are already significantly
492
elevated compared to conventional technologies. The wind turbines are found to be the most 22
493
suitable renewable technologies. Thus, they reach their maximum allowable installed capacity.
494
Wind turbines represent a fairly mature renewable technology option that can produce electricity
495
at relative low costs in suitable locations. Nonetheless, the average renewable generation cost is
496
considerably higher than that of gas-based plants. Therefore, its generation share is mainly used
497
to meet the target.
498
As shown in Table 8, at low carbon tax values, the conventional gas-based power plants without
499
CO2 capture (i.e., ENG1-ENG2, ENG5-ENG6) dominate the energy infrastructure. This is, their
500
levelized electricity costs are the lowest among the power plants (see Figure 5). Additionally,
501
ENUC2 produce the minimum share of nuclear power expected to be supplied in the year 2020.
502
Furthermore, as the carbon tax reaches approximately the value of 50 $/tonne CO2, the power
503
production infrastructure is modified (see Table 8). For example, ENG2 seems reduced its
504
electricity output more than half (i.e., to 27%). Most of the generation capacity loss by plants
505
ENG2 migrated to ENG3 plants (NGCC plants with CO2 capture). This is, the carbon tax rises the
506
levelized electricity costs in ENG2 plants such as they converge with ENG3’s unit costs at
507
approximately 50 $/tonne CO2. Moreover, at higher tax values ENG2’s costs exceed those of ENG3
508
since the carbon tax burden surpasses the energy cost penalties of CCS methods. Furthermore,
509
the remaining part of ENG2’s loss capacity migrates to nuclear. Comparing the total costs between
510
the reference and maximum carbon tax scenarios, for gas generated power the expenses increase
511
over 100%, for nuclear by 23% (due to capacity increase), whereas for renewables the cost
512
remained fixed (see Table 8).
513
4.1.2. International natural gas market price
23
514
Regarding the international natural gas price scenario, gas generated power also dominates the
515
power infrastructure. However, the predominance of gas power significantly decreases to an
516
average of 55% compared with the local gas price scenario (e.g., over 75%). This is due to
517
higher fuel costs (i.e., 2.5 times higher). For example, in the base case scenario (i.e., no carbon
518
tax), the fuel (natural gas) represents approximately 82% of the cost in gas-based generation;
519
whereas the capital represents around 14% of the cost. On the other hand, for carbon taxes higher
520
than 75 $/tonne CO2, ENG3 plants replace most of the generation capacity loss by ENG2 plants
521
(similar to the local gas price scenario). The generation capacity of the remaining conventional
522
power plants (i.e., ENG1, ENG5-ENG6) remained fixed near their minimum expected production
523
levels. Thus, at international gas price levels, they are no longer cost competitive compared with
524
nuclear power (see Figure 6).
525
As shown in Figure 6, the levelized electricity cost of nuclear plant ENUC1 is 0.062 $/kWh. Since
526
nuclear plants are considered to be carbon-free technologies; their costs do not vary as a function
527
of carbon tax values. On the other hand, the least costly gas-based power technology corresponds
528
to plants type ENG2 (0.070 $/kWh) for the base case scenario. Accordingly, nuclear power
529
averages 38% of the total supply, which represents an increment of over 100% with respect to
530
the local gas price scenario (see Figure 4). For all carbon tax values, the number of nuclear plants
531
remained fixed to its maximum allowable number (i.e., 4). Nevertheless, some changes take
532
place in the final configuration of the nuclear plants in terms of capacities (see Table 8). As for
533
the renewable power sources, their electricity output remained fixed at 6% of the total supply
534
(similar to the local gas price scenario). This is the result of their comparatively high generation
535
costs. As previously shown, under competitive natural gas market prices the nuclear option
24
536
becomes the most suitable power production technology. This is, despite the associated external
537
costs to society from the operation of the nuclear reactors.
538
As shown in Table 7, for the international gas price’s base scenario (i.e., no carbon tax), gas
539
generated power accounts for 55% of the annual cost. This number is very similar to that
540
obtained previously for the reference scenario (58%). Nevertheless, the power output from the
541
gas fleet is significantly smaller compared with the reference scenario (55% vs. 80%). This
542
illustrates the effect of considering international gas prices in the operation of the Emirate’s
543
power sector (see Table 8). An international gas price level causes a significant increase in the
544
levelized electricity cost of gas generated power; thus, decreasing its power share.
545
4.1.3. Carbon Tax and Air emissions
546
As shown in Figure 7, for the local gas price, the CO2 offset by applying different carbon tax
547
levels can be divided among 3 regions. The first region comprises carbon tax values up to 48
548
$/tonne CO2. In this region the annual CO2 offset remained fixed at 9 MT CO2 eq./yr. Similar
549
CO2 offset levels are observed for the reference case (i.e., no carbon tax) as a result of the
550
minimum projected installed capacities of alternative power sources. This is, alternative power
551
allows reducing the CO2 emission levels compared with a BAU operation mode in the year 2020.
552
Therefore, the first region denotes a zone where the carbon tax is not high enough to produce
553
changes in the power infrastructure. On the other hand, in the second region (that includes
554
carbon tax values between 49-54 $/tonne CO2) the amount of carbon offset increases to nearly 14
555
MT CO2 eq./yr. This results from changes in the power infrastructure toward lower carbon
556
emitter plants. For instance, both the deployment of gas plants with CCS and the increase in the
557
nuclear generation capacity take place. In contrast, the third region starts at carbon tax greater
25
558
than 54 $/tonne CO2. In this region the number of gas plants with CCS proliferates (i.e., ENG3);
559
while the number of conventional gas plants decrease (i.e., ENG2).
560
Regarding the international gas price scenarios, the price level by itself produces a large impact
561
in the amount of CO2 offset. For example, for the base case (no carbon tax) the CO2 offset is
562
comparable to that attained at local gas price and tax values up to 54 $/tonne CO2. Furthermore,
563
the maximum CO2 abatement level reached for the local gas price only exceeds by 12% those of
564
the base international gas price scenario. This is an indicative of the enormous impact that gas
565
prices have on the configuration of the power infrastructure. This is directly reflected on the
566
amount of CO2 offset. Moreover, as shown in Figure 7, for domestic gas prices (low priced) the
567
carbon tax is a suitable measure to enforce reductions in the CO2 emission levels. On the other
568
hand, international gas price by itself is a valuable instrument to mitigate CO2 emissions. This
569
given that higher feedstock fuel prices incentivize the deployment of alternative low emission
570
energy sources such as nuclear.
571
As shown in Figure 8, for the local gas price scenario, the annual carbon costs increase at the
572
steepest slope between 0-48 $/tonne CO2. Subsequently, there is a slope change triggered by a
573
power infrastructure shift toward less CO2 emitting power technologies. Similarly, beyond 54
574
$/tonne CO2 tax values another slope change (lowest slope) takes place. This change is due to
575
modifications in the power infrastructure toward additional CO2 capture plants (i.e., ENG3 plants).
576
On the other hand, for international gas price only one slope changed takes place (i.e., beyond 75
577
$/tonne CO2), which also reflects a shift toward a less carbon emitting power infrastructure.
26
578
As shown in Figure 9, the levelized electricity cost for the base international gas price scenario is
579
comparable to that of local gas price with a carbon tax near to 125 $/tonne CO2. This reflects the
580
significant impact of gas prices over the levelized electricity cost in Abu Dhabi.
581
4.2. Case study 2: Considering the social benefits of using alternative energy sources
582
The present case study illustrates the case when the social benefits of using alternative energy
583
sources are to be accounted as part of the problem’s objective function (i.e., ECc,p). This can lead
584
to changes in the power production infrastructure, air emissions, and electricity costs. These
585
social benefits are incorporated as a discount in the power generation costs. This discount results
586
from air emissions reductions (i.e., GHG and CAC) with respect to a BAU operation. These
587
benefits are a reflection of the welfares to society attributed to the avoidance of air emissions.
588
These benefits correspond to save healthcare and environmental remediation costs. Moreover,
589
this case study considers no carbon tax. Accordingly, two scenarios are studied: the capacity
590
constrained and unconstrained social benefit scenarios. In the capacity constrained social benefit
591
scenario, maximum capacities have been set for renewable and nuclear power generation as
592
shown in (4) and (A.1)-(A.2) (see Appendix A for the latter equations). On the other hand, in the
593
capacity unconstrained social benefit scenario, the maximum capacities of renewable and
594
nuclear power have not been restricted; whereas the renewable target was not set. The latter
595
scenario allows studying the level of impact that the unconstrained deployment of alternative
596
energy sources can produce in the Emirate’s power sector. Furthermore, in both scenarios the
597
influence of deducted costs over the power system is analyzed in terms of the final power
598
infrastructure design (see Figures 10-12).
599
The power plant infrastructure results for the capacity constrained scenario are similar to those of
600
the reference scenario (see previous section: Case Study 1) to a certain extent (see Figure 10). 27
601
This can be seen in the electricity output from the power technologies. For example, the
602
electricity outputs from renewable technologies are comparable (e.g., solar and wind) with those
603
of the reference scenario. However, the overall gas power generation is reduced in 5.6% mainly
604
due to a major fall in ENG2’s generation capacity (approximately 50%). This loss of generation
605
capacity is not completely balanced by the migration of capacity to ENG3 plants (see Figure 11).
606
Furthermore, ENG3 plants take part in the power infrastructure for this scenario because they
607
include CO2 capture. This allows discounting costs in terms of social damages avoided by
608
reducing the carbon emissions. On the other hand, as shown in Figure 11, the nuclear power
609
output increases by 32% compared to the reference scenario. This is the result of deploying 2
610
nuclear reactors with higher generation capacities (ENUC1) instead of smaller size reactors
611
(ENUC2). This allows an increment in the production of low-emission electricity, which results in
612
further cost discounts.
613
As a result, Figure 12 shows that the overall nuclear costs are reduced by 30%, the renewable
614
power costs decreased by 13%, whereas the gas-based costs are reduced by 10% with respect to
615
the reference scenario. Although nuclear power generation largely increases (32%), its
616
generation cost decreased notoriously (30%). This uncommon development is caused by the
617
increasingly high financial discounts per additional unit of clean power produced. This
618
overweighs the costs of the extra generated nuclear power. Similarly, even though the generation
619
of renewable power remained constant, their corresponding costs also decreased; nonetheless at a
620
minor rate. This is, although the renewable power output remained unchanged, there are visible
621
economic benefits related to the use of clean energy sources (i.e., in terms of costs discounts)
622
compared with conventional gas-fired turbines. Moreover, natural gas-based power generation
623
presents the smallest cost reduction among the power production technologies for this scenario. 28
624
Partly, due to reductions in its overall output compared with the reference scenario. Also, the
625
introduction of ENG3 plants (with CO2 captured) plays an important role. These two factors allow
626
costs discounts by the avoidance of carbon emissions. Consequently, the average levelized
627
electricity cost for the capacity constrained scenario was approximately 0.044 $/kWh (12%
628
below the reference scenario’s cost). Additionally, the amount of CO2 eq. offset significantly
629
increases to 20.7 MT/yr (130% over the reference scenario’s value). This shows the effect of
630
considering emissions mitigation in the power sector (by using alternative energies or CCS
631
methods) as a social benefit.
632
Regarding the capacity unconstrained social benefit scenario, there are a few differences
633
compared with the capacity constrained social benefit scenario described above. For example,
634
the overall renewable electricity generation declined by 50%, since the total renewable power
635
target was neglected. Additionally, nuclear power remained constant whereas gas-based power
636
increases by 4%. Renewable power decreases given its associated higher production costs, which
637
exceeds the economic benefits of emission avoidance. As a result, the total installed capacities of
638
PV (ESOL2) and CSP (ESOL3) plants remained at their minimum expected levels. This is, only
639
minimum renewable capacity constraints were kept. On the other hand, wind power (EWIN2) is
640
the only renewable technology that increases its electricity output (525%). This is due to its
641
lowest generation costs among renewable energy sources. Whereas nuclear power output
642
remained unchanged with 2 ENUC1 facilities (minimum expected number). However, the nuclear
643
plants selected are medium capacity facilities (e.g., ENUC2 plants are smaller size reactors). This
644
shows that under social benefits considerations nuclear power is a suitable option, but until a
645
limited extent. On the other hand, total gas-based power generation grows due to an increase in
646
ENG3’s output. This to cover part of the generation capacity loss by renewable sources and ENG2 29
647
plants (see Table 9). ENG3’s power output offset significant amounts of CO2 that otherwise would
648
be generated using conventional power plants ENG2. Also, ENG3 replaces part of the renewable
649
power capacity loss, but at much lower financial costs.
650
4.3. Case study 3: CO2 constrained scenarios
651
This scenario considers an annual CO2 emission reduction target for the Abu Dhabi power
652
sector. This target is assumed analogous to the emission goal anticipated for the US and Canada
653
under the United Nations Framework Convention on Climate Change (UNFCCC) [59].
654
Accordingly, a CO2 emission target of 17% reduction compared with the 2005 levels was set for
655
the present scenario [84]. Consequently, the total CO2 emissions for the Abu Dhabi power sector
656
(see (A.10) in Appendix A for details) were constrained to be lower than 8.31 MT of CO2
657
eq./year in 2020 [65]. Although this reduction target of 17% may appear modest compared with
658
other nation’s goals; the fact that the UAE is under no obligation to reduce its emissions
659
(according to the UNFCCC) makes the considered environmental goal ambitious. In order to
660
obtain a feasible solution, no minimum installed generation capacities or power production
661
shares were specified for the plant’s infrastructure a priori. Furthermore, no carbon tax or social
662
benefits associated to the avoidance of air emissions were considered in the present case study.
663
As shown in Table 10, the power production infrastructure for this case study is entirely based on
664
natural gas power plants; especially power plants ENG3 that include CO2 capture. Accordingly,
665
approximately 96.3% of the overall power demand is met using ENG3 plants. Furthermore, only a
666
small fraction (3.3%) of the total power is generated through plants ENG2, which is the least
667
expensive electricity production option for local natural gas price ranges. This small generation
668
fraction represents the largest potential power output from plants ENG2 to comply with the carbon
669
emission constraint. On the other hand, plants ENG5 are only deployed to fill a miniscule 30
670
generation gap (0.3%) needed to meet the overall electricity demand. This type of plants features
671
the second lowest generation capacity among the natural gas-based power plants considered in
672
the model. As previously mentioned, for this case study the minimum generation capacity
673
constraints were relaxed in order to find a feasible solution. Under none of the minimum
674
generation capacity constraints used in the previous case studies, the problem converges to a
675
feasible solution. Thus, attaining such low carbon emission levels requires a very specific power
676
infrastructure. Nonetheless, similar CO2 abatement levels also can be reached with the
677
deployment of alternative energy sources, such as nuclear and renewable power, but a higher
678
generation costs compared with those attained using a gas-based infrastructure fed with highly
679
subsidize fuel.
680
The CCS costs represent approximately just over 9% of the total annual power generation cost.
681
This shows the cost weight of the carbon mitigation methods included in the power
682
infrastructure. Moreover, the annual power cost (see Table 10) is equivalent to that of the
683
reference scenario (see Table 6) since they only differ in 0.6% from each other. This cost
684
correspondence shows that the implementation of alternative energy policy in the Emirate
685
(reference scenario) is comparable to the present case study in terms of annual costs. However,
686
when the amounts of CO2 eq. offset from both scenarios are compared, it is clearly shown that
687
significantly higher CO2 emissions can be offset (see Tables 6 and 10) under the CO2 targeted
688
scenario. For example, the emissions offset are 4 times higher compared with the reference
689
scenario. Furthermore, none of the case studies discussed in the previous sections match the
690
amount of carbon offset in the present case study. This is, under none of the gas prices, carbon
691
tax values or social benefit scenarios previously discussed, the amount of CO2 eq. offset reaches
692
the levels of the present case study. This shows that gas plants with CO2 capture are the most 31
693
economical and suitable options to accomplish significant carbon emission reduction levels. This
694
when highly subsidized natural gas prices are assumed for the operation of the power sector.
695
The amount of carbon emissions reduced in the present case study is considerably high if we
696
consider that the Emirate’s power infrastructure is already gas-based. This characteristic only
697
leaves room to the deployment of CCS systems or alternative energies to mitigate CO2
698
emissions. Comparatively, most of the US (from which we have mirrored the emission target)
699
mitigation strategy relies on substituting coal-based electricity generation by gas power.
700
Nevertheless, the present case study would require extensive amounts of natural gas amounting
701
2.56 Bcf/d. This amount of gas is greater than the current UAE’s gas imports from Qatar (1.73
702
Bcf/d). Consequently, Abu Dhabi would have to find ways to boost its gas supply either through
703
new gas import options (e.g., via pipeline or LNG cargos) or new domestic sour gas
704
developments. Both gas sources seem unlikely to be obtained at low costs, which imply that the
705
government would have to increase the subsidy levels to the power sector to maintain low tariffs.
706
Additionally, the projected capacity of the carbon transport network would have to be largely
707
expanded to handle the high CO2 captured volumes. This will require high economic investments
708
and time to be deployed.
709
The enactment of more ambitious environmental mitigation strategies by the government of Abu
710
Dhabi, such as that discussed in the present case study, would be a significant challenge for the
711
Emirates in techno-economical terms. Furthermore, this case study represents the optimistic
712
scenario in term of CO2 abatement. However, it was intended to illustrate what it would take to
713
dramatically reduce the carbon emissions in Abu Dhabi. This does not imply a suggestion from
714
the authors toward an entirely gas-based power fleet with high proportion of plants with CCS
715
methods. Which given the restricted availability of cheap natural gas (e.g., at current price levels) 32
716
resources in the region seems unlikely in the future. This would represent a significant increase
717
in the governmental subsidies to the power sector and impact the energy security of the Emirate.
718
5. Conclusions
719
In the current work, a comprehensive design optimization model has been proposed. The
720
proposed mathematical model can be used to determine the optimal power infrastructure for the
721
UAE under environmental constraints. The results presented in this work show that at local gas
722
price levels (subsidize), the gas-based plants are the most economically attractive options;
723
namely plants ENG2. Conversely, for international (Asia-Pacific market) gas price levels, the gas-
724
based plants also dominate the power production infrastructure, but at lower levels. Moreover,
725
for both gas price scenarios as the carbon tax level increases, the power infrastructure migrates
726
towards CO2 captured plants (e.g., ENG2) and higher nuclear output shares. On the other hand, the
727
renewable power output remains fixed given its relatively high costs. The increase in gas prices
728
and the adoption of a carbon tax are both suitable options to promote emissions reductions in the
729
Abu Dhabi’s power sector.
730
Accounting for the social benefits of emissions avoidance in economic terms, the results show
731
that nuclear output significantly increases compared with the reference scenario. This is the
732
result of nuclear power’s relatively low generation costs and the economic benefits of producing
733
low emission electricity. Additionally, a CO2 reduction target similar to that of the US and
734
Canada for the year 2020 can be achieved in the Abu Dhabi’s power sector. This at cost levels
735
similar to those of the reference scenario. Yet, the latter scenario would require for the domestic
736
gas prices to remain low in the foreseeable future and a complete carbon capture-based gas
737
power infrastructure. Also, the planned CO2 transport network would have to be considerably
738
expanded. Overall, the use of emission mitigation strategies increases the levelized electricity 33
739
costs compared with a BAU operation; unless emissions reductions are treated as public benefits
740
and valued in monetary terms.
741
The results presented in this work show that the proposed mathematical model can be used as a
742
valuable tool to: 1) design the expansion of the UAE’s power plant’s infrastructure, 2) analyze
743
the impact of introducing nuclear and renewable power technologies into the UAE power
744
system, 3) examine the introduction of carbon tax fees and social benefits to mitigate air
745
emissions in the UAE power sector, 4) forecasts future power generation scenarios, and 5)
746
evaluate annual costs in the UAE power sector. Future work on this research includes the co-
747
generation of electricity and desalinated water along with stand-alone power plants and
748
desalination facilities. This will enable the minimization of water losses during the summer
749
season, which is caused by the mismatch between the desalinated water requirements and the on-
750
peak electricity demands. Also, the development of a multiperiod model that takes into account
751
the evolution of the power network over time as certain technologies reached maturity. The
752
current model estimates the potential power plant’s infrastructure for the projected total power
753
demand at steady state. Another relevant realistic approach may consider uncertainty in the
754
model’s key parameters subject to fluctuations, e.g., natural gas price, renewable sources costs,
755
gas supply, CO2 emission target, renewable energy targets, and social damages avoided values.
756
As a result, the proposed deterministic approach will be expanded into a stochastic model, which
757
will be useful to determine the most likely distributions amongst the potential power
758
infrastructures, unit power generation costs, and air emissions.
34
759
Appendix A. Supplementary Equations of the Design Optimization Model
760
The supplementary equations describing in more detailed the proposed design optimization
761
model are included in the present section. For instance, the overall installed capacity of
762
renewable power plants is constrained as follows:
763
AEp RET,
764
where RET indicates the minimum overall installed capacity expected from renewable sources in
765
a specific year. This parameter can be obtained from issued government energy policies or
766
environmental agency recommendations.
767
The share of nuclear power generated by the reactors can be constrained as follows:
768
EPp SE p ED,
769
where EPp represents the amount of electricity produced using the power production technology
770
p, SEp is a model input that indicates the share of the total electricity demand produced by
771
technology p.
772
The power production balance by wind turbines is given by the following expression:
773
EPp IEp ICp AN FLH,
774
where ICp is a parameter of the model that represents the power plant p installed capacity, AN
775
denotes the total array number of wind turbines in the farm, and FLH denotes the average full
776
load hours for wind turbines in the country.
777
The total balance of electricity produced by power source (e.g., fossil, renewable and nuclear)
778
can be estimated as follows:
779
EF EPp ,
p g
(A.4)
780
ER EPp ,
p w, s
(A.5)
p
p
p
p w, s
(A.1)
pn
(A.2)
pw
(A.3)
35
781
EN EPp ,
782
where EF, ER and EN represent the amount of electricity produced by fossil, renewable and
783
nuclear energy sources, respectively.
784
Similar to Eqs. A.4-A.6, the total balance of electricity produced by the country’s power
785
infrastructure is estimated as follows:
786
TEP EPp
pn
p
(A.6)
(A.7)
p
787
where TEP represents the total amount of electricity produced by the power infrastructure in the
788
country.
789
The balance of carbon dioxide equivalent (CO2 eq.) generated by the power plants’ fleet can be
790
estimated as follows:
791
CE p EGe ,p ,
792
where CEp represents the amount of emission e produced by the natural gas-based power plants p
793
(i.e., p = g). Accordingly, the total balance of CO2 equivalent produced by the power fleet can be
794
calculated as follows:
795
TCE CE p ,
796
where TCE represents the total amount of carbon dioxide equivalent produced by the natural gas-
797
based power plants in the fleet.
798
On the other hand, the total balance of carbon dioxide captured (TCC) in the power infrastructure
799
is given as follows:
800
TCC CC p ,
801
Where CCp denotes the amount of CO2 captured by power plant p.
p g , e CO2 , CH 4 , N2O
e
p g
p
p
p g
(A.8)
(A.9)
(A.10)
36
802
The power technologies included in the present design optimization model can be divided in sub-
803
sets of power plants. Accordingly, the type of natural gas-based power plants is given by the
804
subset g, g cc, ox, st, gt , where cc represents the natural gas combined cycle (NGCC) plants,
805
ox denotes the oxyfuel plants, st represents the steam turbines, and gt indicates the gas turbines.
806
Similarly, the type of solar plants is denoted by the subset s, s pv, cs , ot, sp , where pv
807
represents the photovoltaic (PV) power plants, cs denotes the concentrating solar power (CSP)
808
plants, ot indicates the ocean thermal energy conversion (OTEC) plants, and sp symbolizes the
809
solar land ponds plants.
810
The capital cost per type of power plant is given as follows:
811
Cap p PCFp ICp AFp ,
812
Cap p PCFp ICp WT AFp ,
813
Cap p PCFp BOSp Insp A AFp ,
814
where PCFp represents the capital factor of power plant p, AFp is the capital amortization factor
815
associated to the plant p, WT is the wind turbine’s array number in the farm, BOSp is the balance
816
of the system cost for photovoltaic plants, Insp is the installation cost for photovoltaic plants, and
817
A the surface covered by the photovoltaic cells.
818
The capital amortization factor for the power plants (AFp) is calculated as follows:
819
AI 1 AI p AFp 1 AI DTp 1
820
where AI denotes the annual interest rate and DTp is the depreciation time associated to plant p.
821
The operating and maintenance costs of the individual power plants are calculated as follows:
822
OMp
p g, n, s , s cs , ot , s pw
(A.11) (A.12)
p s pv
(A.13)
DT
Cap p AFp
(A.14)
OMF RRF , p
p
p g, s
s pv
(A.15)
37
823
OMp ICp CFp OMFp t,
824
OMp ICp AN FLH OMFp ,
825
where OMFp represents the operating and maintenance cost factor and RRFp is the repair and
826
replacement cost factor. Furthermore, the operating and maintenance costs of the solar plants,
827
except for the PV plants ( s pv ), are given as fixed yearly amounts.
pn
(A.16)
pw
(A.17)
828
38
829
Acronyms
830
ADWEC = Abu Dhabi Water and Electricity Company
831
BAU = business-as-usual
832
CAC = criteria air contaminants (e.g., NOx, SO2 and PM10)
833
CCS = carbon capture and storage
834
CH4 = methane emissions
835
CO2 = carbon dioxide emissions
836
CO2 eq. = carbon dioxide equivalent
837
CSP = concentrating solar power
838
GHG = Greenhouse Gases (e.g., CO2, CH4 and N2O)
839
MILP = mixed integer linear program
840
N2O = nitrous oxide emissions
841
NGCC = natural Gas Combined Cycle
842
NOx = nitrogen oxides emissions
843
OTEC = ocean thermal energy conversion
844
RSB = Regulation and Supervisory Bureau of the Emirate of Abu Dhabi
845
SO2 = sulfur dioxide emissions
846
UAE = United Arab Emirates
847
yr = year
39
848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898
Nomenclature Model Variables AEAe,p = emission e avoided using plant p [tonne/h] Capp = capital cost of the power plant p [$/yr] CCp = CO2 captured by plant p [tonne/h] CEp = CO2 eq. produced by plant p [tonne CO2/h] CF = objective cost function [$/yr] ECc,p = external cost c associated to the air emissions of the power plant p [$/yr] EF = electricity produced by fossil sources [kW] EGe,p = emission e generated by the plant p [tonne/h] ER = electricity produced by renewable sources [kW] EN = electricity produced by nuclear sources [kW] EPp = electricity produced by the power plant p [kW] FCp = fuel cost of the plant p [$/yr] NGp = natural gas consumed by the plant p [Nm3/h] OMp = operating and maintenance cost of p [$/yr] PCp = total annual power production cost [$/yr] PPCp = cost associated to the public perception on the deployment of the pth power plant [$/yr] TCC = total CO2 captured in the power fleet [tonne CO2/h] TCE = CO2 eq. produced in the fleet [tonne CO2/h] TCP = total compression power used to transport the captured CO2 to the sequestration sites [kW] TCS = total CO2 sequestration cost [$/yr] TCT = total CO2 transport cost [$/yr] TEP = total electricity produced by the fleet [kW] TNG = natural gas consumed by the plants [Nm3/h] Integer Variables IEp = number of power plants p Sets c = set of external cost associated to the air emissions e = set of gaseous air emissions p = set of power plants η = set of decision variables in the design optimization model Sub-sets g = subset of natural gas-based power plants n = subset of nuclear power plants s = subset of solar-based power plants w = subset of wind turbine farms Sets and subsets elements cc = natural gas combined cycle (NGCC) plants cs = concentrating solar power (CSP) plants dis = cost discount due to emission abatements gt = power gas turbines ot = ocean thermal energy conversion (OTEC) plants ox = oxyfuel power plants pv = photovoltaic power plants sp = solar land pond power plants st = power steam turbines
899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929
tax = toll paid due to the generation of air emissions Model Parameters A = surface covered by the photovoltaic cells [m2] AEFe,p = emission e from the plant p [tonne/kWh] AFp = capital amortization factor of plant p [% / yr] AI = annual capital interest rate [%] AN = total array number of wind turbines in the farm [units] BOSp = balance of the system cost for PV [$/m2] CCFp = CO2 capture factor of plant p [tonne CO2/kWh] CET = maximum allowable CO2 emission from the country’s power infrastructure [tonne CO2/h] CFp = capacity factor of power plant p [%] CPF = power consumed per unit of CO2 and traveled distance [kWh/(tonne) (km)] CSF = CO2 sequestration cost [$/tonne CO2] CTAX = CO2 tax cost [$/tonne CO2] CTF = CO2 transport cost [$/tonne km] DTp = depreciation time of plant p [yr] ED = total electricity demand input [kW] ENGe = average emission e generated by the conventional NGCC fleet in the UAE [tonne/kWh] ESCe = avoided social cost associated to the emission e by using alternative energy plants [$/tonne] FCFp = fuel cost factor [$/MJ] FLH = full load hours for wind turbines in a given geographic location [%] HRp = heat rate of the power plant p [MJ/kWh] ICp = installed capacity of power plant p [kW] Insp = installation cost for photovoltaic plants [$/m2]
930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947
E pU = maximum available number of plants p that can take
E pL = minimum allowable number of plants p [units]
part in the power infrastructure [units] OMFp = operating and maintenance cost factor [%], [$/kWh] or [$/yr] PCFp = power plant p capital factor [$/kW] or [$/m2] PLp = distance traveled by the CO2 captured at p [km] AEp = minimum installed generation capacity expected of power plants type p [kW] RET = minimum overall installed capacity expected from renewable sources [kW] RRFp = repair and replacement cost factor [%] SCC = avoided social cost of CO2 emitted to the atmosphere [$/tonne] SEp = share of electricity produced by p [%] t = total operating time of the infrastructure [h/yr] UCFp = unit capacity factor of plant p [units] WT = number of wind turbines in the farm’s arrays [units]
40
948
Table Captions
949
Table 1. List of power plants included in the design optimization model.
950
Table 2. List of key techno-economic parameters used for modeling the power plants.
951
Table 3. Case Study 2011: Economic and operational input parameters and assumptions used for
952
the verification of the model.
953
Table 4. Case Study 2011: Comparison of the verification results and literature values for Abu
954
Dhabi’s power sector operation in 2011.
955
Table 5. Case Study 2020: Key economic and operational modeling parameters used for the
956
optimal design of Abu Dhabi’s power system for 2020.
957
Table 6. Case Study 2020: Optimization results for local gas price at different CO2 carbon tax
958
levels.
959
Table 7. Case Study 2020: Optimization results for international gas price at different CO2
960
carbon tax levels.
961
Table 8. Case Study 2020: Power Plant Infrastructure at local and international gas prices.
962
Table 9. Case Study 2020: Optimization results considering the social benefits of CO2 emission
963
avoidance in the power sector.
964
Table 10. Case Study 2020: Optimization results under CO2 constrained power production.
41
965
Figure Captions
966
Figure 1. Superstructure of the power optimization model.
967
Figure 2. General layout of the power optimization model.
968
Figure 3. Comparison of the historical levelized electricity cost in Abu Dhabi and the verification
969
electricity costs at typical gas prices in 2011.
970
Figure 4. Comparison of the power infrastructure design at local and international gas price
971
without carbon tax in 2020 (4)-(5).
972
Figure 5. Comparison of the levelized electricity costs for the power plants at local gas price
973
under carbon tax in 2020 (5),(16).
974
Figure 6. Comparison of the levelized electricity costs for the power plants at international gas
975
price under carbon tax in 2020 (5),(16).
976
Figure 7. Comparison of CO2 offset at local and international gas price under carbon tax in 2020
977
(9),(13).
978
Figure 8. Comparison of the annual carbon cost at local and international gas price under carbon
979
tax in 2020 (8).
980
Figure 9. Comparison of the average levelized electricity cost at local and international gas price
981
under carbon tax in 2020 (5),(16).
982
Figure 10. Comparison of the power infrastructure for: capacity constrained and unconstrained
983
social benefit scenarios and the reference scenario for the year 2020.
984
Figure 11. Comparison of the annual power production for: capacity constrained and
985
unconstrained social benefit scenarios and the reference scenario for the year 2020.
986
Figure 12. Comparison of the annual power cost for: capacity constrained and unconstrained
987
social benefit scenarios and the reference scenario for the year 2020.
42
988
Table 1. List of power plants included in the design optimization model Type of power plant
Literature
Natural gas Natural Gas Combined Cycle (NGCC)- class 7FA (ENG1) Black & Veatch Holding Company [42] Natural Gas Combined Cycle (NGCC)- class 7FB (ENG2) Rubin, et al. [43, 44] Natural Gas Combined Cycle (NGCC)- class 7FB- with Rubin, et al. [43, 44] 90% CO2 capture using MEA (ENG3) Natural Gas Oxyfuel with CO2 capture (ENG4) Davison [45] Steam Turbine (ST) (ENG5) ICF International Company [46] Gas Turbine (GT) (ENG6) US Energy Information Administration [47] Wind turbine Nordex N43/600 (EWIN1) Harijan, et al. [48] Nordtank 500/41 (EWIN2) Janajreh, et al. [49] Sonkyo 3.5 kW (EWIN3) Janajreh, et al. [49] Gaia–Wind 133-11 kW (EWIN4) Shawon, et al. [50] Solar Sanyo single crystalline silicon solar cells (ESOL1) Radhi [51] Mono-silicon BP solar 90 W modules (ESOL2) Harder, et al. [52] Concentrating solar power (CSP) (ESOL3) Straatman, et al. [53] Ocean thermal energy conversion (OTEC) (ESOL4) Straatman, et al. [53] Solar land pond (SLP) (ESOL5) Straatman, et al. [53] Hybrid of ocean thermal energy conversion with an Straatman, et al. [53] offshore solar pond (OTEC–OSP) (ESOL6) Nuclear APR-1400 (ENUC1) World Nuclear Association [54] Nuclear Energy Institute [55] AP-1000 (ENUC2) International Atomic Energy Agency [56] EPR-1650 (ENUC3) World Nuclear Association [54] Areva [57]
989
43
990
Table 2. List of key techno-economic parameters used for modeling the power plants
Nuclear
Solar
Wind
Natural gas
Type of power plant
991 992 993 994 995
ENG1 ENG2 ENG3 ENG4 ENG5 ENG6 EWIN1 EWIN2 EWIN3 EWIN4 ESOL1 ESOL2 ESOL3 ESOL4 ESOL5 ESOL6 ENUC1 ENUC2 ENUC3
Installed Capacity (kW) (kW/m2)a 580,000 507,000 432,000 440,000 60,000 85,000 600 (25)d 500 (30)d 3.5 (100)d 11 (45)d 0.15a 0.143a 16,700 50,000 50,000 50,000 1,400,000 1,100,000 1,650,000
Capital Cost ($/kW) ($/m2)b 1,250 595 978 1,308 681 973 1,620 547 6,084 12,955 2,595b 1,315b 14,228 13,500 7,938 2,970 3,643 3,582 4,100
Operating & Maintenance Cost Factor 0.0045 $/kWh 1.8 % 3.7 % 8.6 % 0.005 $/kWh 0.0163 $/kWh 0.011 $/kWh 0.039 $/kWh 0.047 $/kWh 0.238 $/kWh 6% 0.8 % 5,400,000 $/yr 9,450,000 $/yr 6,750,000 $/yr 4,050,000 $/yr 0.002 $/kWh 0.0054 $/kWh 0.002 $/kWh
Heat Rate
7.07 MJ/kWh 7.17 MJ/kWh 8.41 MJ/kWh 7.70 MJ/kWh 8.37 MJ/kWh 11.45 MJ/kWh N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 2.77E-6 kg/kWh 7.48E-6 kg/kWh 2.77E-6 kg/kWh
a
Installed Capacity of the Photovoltaic Solar plants given in (kW/m2). Capital Cost of the Photovoltaic Solar plants includes the Balance of the System (BOS) and Installation costs given in ($/m2). c Operating and Maintenance cost (including repair and replacement cost for PV) given as a percentage (%) of the plant’s total capital cost. d Number enclosed by brackets indicates the total array number of turbines in the wind farm. N/A Power plants that do not consider Heat Rate related to fuel consumption. b
44
996
Table 3. Case Study 2011: Economic and operational input parameters and assumptions
997
used for the verification of the model. Input type Mean natural gas price (FCFp) Natural gas CO2 emission factor (AEFe,p) Annual capital interest rate (AI) Annual electricity generation (EPp) Annual operating hours (t) Power plants’ depreciation time (DTp) Gas-based plants’ capacity factor (CFp) Transmission cost Distribution cost Power to Water ratio (P/W) Power’s fuel utilization ratio
Value 0.0027 $/MJ 367 g CO2/kWh 10% 46,367 GWh/yr 8,760 h/yr 30 yr 0.90 0.010 $/kWh 0.040 $/kWh 46.40 kWh/m3 58.93%
Literature Source [73, 75, 85, 86] [47] [47] [65] [47] [47] [42, 45-47] [76] [76] [41] [77]
998
45
999 1000
Table 4. Case Study 2011: Comparison of the verification results and literature values for Abu Dhabi’s power sector operation in 2011. Simulated values to verify the design optimization model
Reported values in the literature for Abu Dhabi’s power system in 2011
Literature Source
ENG1 (GWh/yr) ENG2 (GWh/yr) ENG5 (GWh/yr) ENG6 (GWh/yr) Total Electricity produced (TEP) (GWh/yr) Techno-economic Evaluation
18,290 23,983 3,784 670 46,727
18,230 22,797 4,500 840 46,367
[64] [68-72] [64] [68-72] [64] [68-72] [64] [68-72] [65]
Average levelized electricity cost ($/kWh) Annual fuel consumption (TNG (6)) (Billion BTU/yr) Annual CO2 eq. emissions (TCE (7)) (MT CO2 eq./yr)
0.035
0.034
319,746 17.14
320,369 16.97
[66, 76] [65] [65]
Variables Power generation share (5)
1001
46
1002
Table 5. Case Study 2020: Key economic and operational modeling parameters used for the
1003
optimal design of Abu Dhabi’s power system for 2020. Input Parameter Local natural gas price (FCFp) International market natural gas price (FCFp) Renewable power generation capacity (REp) CO2 transport cost (TCT) CO2 sequestration cost (TCS) Annual capital interest rate (AI) Annual electricity generation (EPp) Annual operating hours (t)
NG plants minimum annual electricity generation NGCC (ENG1-ENG2) ST (ENG5) GT (ENG6) Renewable plants minimum installed capacities Wind turbine (EWIN1-EWIN4) PV (ESOL1-ESOL2) CSP (ESOL3) Nuclear plants minimum number of facilities Nuclear power reactor (ENUC1-ENUC3)
Input Value 0.0033 $/MJ 0.0085 $/MJ 7% 1.40 $/tonne CO2/100 km 8 $/tonne CO2 10% 120,023 GWh/yr 8,760 h/yr
Literature Source [32] [78]
45,289 GWh/yr 4,205 GWh/yr 683 GWh/yr
[64] [64] [64]
130 MW 510 MW 100 MW
[79] [79] [79]
2 units
[31, 83]
[79, 81] [61] [61] [47] [80] [47]
1004
47
1005 1006
Table 6. Case Study 2020: Optimization results for local gas price at different CO2 carbon tax levels. Variables Power generation (GWh/yr) (5) ENG1 ENG2 ENG3 ENG5 ENG6 EWIN2 EWIN3 ESOL2 ESOL3 ESOL5 ENUC1 ENUC2 Power cost (MM$/yr) (16) ENG1 ENG2 ENG3 ENG5 ENG6 EWIN2 EWIN3 ESOL2 ESOL3 ESOL5 ENUC1 ENUC2 CCS cost (10)-(11) Total cost (21) Carbon Tax cost (8) Average levelized electricity cost ($/kWh)
0 18,290 71,950 0 4,205 1,340 270 5 1,250 750 4,870 0 17,345 830 2,380 0 180 90 16 3.5 535 185 635 0 1,160 0 6,015 0 0.050
15
30
18,290 18,290 71,950 71,950 0 0 4,205 4,205 1,340 1,340 270 270 5 0 1,250 1,250 750 750 4,870 4,870 0 0 17,345 17,345 930 2,780 0 200 98 16 3.5 535 185 635 0 1,160 0 6,540 530 0.054
1,030 3,180 0 225 105 16 0 535 185 635 0 1,160 0 7,070 1,060 0.059
Carbon tax ($/tonne CO2) 45 60 75 100 18,290 71,950 0 4,205 1,340 255 15 1,250 750 4,870 0 17,345
18,290 31,980 34,060 4,625 1,340 270 0 1,250 750 4,870 22,910 0
18,290 31,980 34,060 4,625 1,340 270 0 1,250 750 4,870 22,910 0
18,290 31,980 34,060 4,625 1,340 270 0 1,250 750 4,870 22,910 0
1,130 3,580 0 250 112 15 10 535 185 635 0 1,160 0 7,610 1,590 0.063
1,230 1,770 1,660 300 120 16 0 535 185 635 1,420 0 160 8,030 1,350 0.067
1,340 1,940 1,690 325 127 16 0 535 185 635 1,420 0 160 8,370 1,690 0.070
1,500 2,240 1,740 365 140 16 0 535 185 635 1,420 0 160 8,940 2,260 0.074
125
150
18,290 31,980 34,060 4,625 1,340 270 0 1,250 750 4,870 22,910 0
18,290 31,980 34,060 4,625 1,340 270 0 1,250 750 4,870 22,910 0
1,670 1,840 2,530 2,830 1,780 1,830 410 450 152 164 16 16 0 0 535 535 185 185 635 635 1,420 1,420 0 0 160 160 9,500 10,065 2,820 3,390 0.079 0.084
1007
48
1008 1009
Table 7. Case Study 2020: Optimization results for international gas price at different CO2 carbon tax levels. Variable Power generation (GWh/yr) (5) ENG1 ENG2 ENG3 ENG5 ENG6 EWIN2 ESOL2 ESOL3 ESOL5 ENUC1 ENUC3 Power cost (MM$/yr) (16) ENG1 ENG2 ENG3 ENG5 ENG6 EWIN2 ESOL2 ESOL3 ESOL5 ENUC1 ENUC3 CCS cost (10)-(11) Total Cost (21) Carbon Tax cost (8) Average levelized electricity cost ($/kWh)
0
15
30
Carbon tax ($/tonne CO2) 45 60 75
100
125
150 18,290 31,975 10,215 4,205 1,340 270 1,250 750 4,870 34,365 13,300
18,290 43,970 0 4,205 1,340 270 1,250 750 4,870 45,820 0
18,290 18,290 43,970 43,970 0 0 4,205 4,205 1,340 1,340 270 270 1,250 1,250 750 750 4,870 4,870 45,820 45,820 0 0
18,290 43,970 0 4,205 1,340 270 1,250 750 4,870 45,820 0
18,290 43,970 0 4,205 1,340 270 1,250 750 4,870 45,820 0
18,290 43,970 0 4,205 1,340 270 1,250 750 4,870 45,820 0
18,290 31,975 10,215 4,205 1,340 270 1,250 750 4,870 34,365 13,300
18,290 31,975 10,215 4,205 1,340 270 1,250 750 4,870 34,365 13,300
1,500 3,090 0 360 170 16 535 185 635 2,840 0 0 9,330 0 0.077
1,600 1,700 3,330 3,580 0 0 385 410 177 185 16 16 535 535 185 185 635 635 2,840 2,840 0 0 0 0 9,700 10,100 375 750 0.080 0.084
1,800 3,820 0 430 192 16 535 185 635 2,840 0 0 10,450 1,130 0.087
1,910 4,060 0 455 200 16 535 185 635 2,840 0 0 10,840 1,500 0.090
2,010 4,310 0 480 207 16 535 185 635 2,840 0 0 11,220 1,880 0.093
2,180 3,430 967 515 219 16 535 185 635 2,130 910 48 11,770 2,110 0.097
2,340 2,510 3,720 4,020 980 995 555 595 232 245 16 16 535 535 185 185 635 635 2,130 2,130 910 910 48 48 12,300 12,800 2,640 3,170 0.102 0.106
1010
49
1011
Table 8. Case Study 2020: Power Plant Infrastructure at local and international gas prices. Variable Power plant (units) (4) ENG1 ENG2 ENG3 ENG5 ENG6 EWIN2a EWIN3 ESOL2b ESOL3c ESOL5 ENUC1 ENUC2 ENUC3
1012 1013 1014
0
Local NG price results / International NG price results Carbon tax ($/tonne CO2) 15 30 45-48 49-54 55-60 75 100 125
4/4 4/4 4/4 4/4 4/4 18 /11 18 /11 18 /11 18 /11 14 /11 0/0 0/0 0/0 0/0 3/0 10 / 10 10 / 10 10 / 10 10 / 10 10 / 10 2/2 2/2 2/2 2/2 2/2 5/5 5/5 5/5 5/5 5/5 14 / 0 14 / 0 0/0 40 / 0 40 / 0 5/5 5/5 5/5 5/5 5/5 1/1 1/1 1/1 1/1 1/1 13 / 13 13 / 13 13 / 13 13 / 13 13 / 13 0/4 0/4 0/4 0/4 2/4 2/0 2/0 2/0 2/0 0/0 0/0 0/0 0/0 0/0 0/0
4/4 8 /11 10 / 0 11/10 2/2 5/5 0/0 5/5 1/1 13/13 2/4 0/0 0/0
4/4 8 /11 10 / 0 11/10 2/2 5/5 0/0 5/5 1/1 13/13 2/4 0/0 0/0
4/4 8 /8 10 /3 11/10 2/2 5/5 0/0 5/5 1/1 13/13 2/3 0/0 0/1
4 /4 8/8 10 /3 11/10 2/2 5/5 0/0 5/5 1/1 13/13 2/3 0/0 0/1
150 4/4 8/8 10 /3 11/10 2/2 5/5 0/0 5/5 1/1 13/13 2/3 0/0 0/1
a
Number of units were converted to represent the equivalent quantity of wind farms with 100 turbines. Number of units were converted to represent the equivalent number of 100 MW PV plants (analogous to Nour I PV). c Number of units were converted to represent the equivalent numbers of 100 MW CSP plants (analogous to Shams I). b
50
1015 1016
Table 9. Case Study 2020: Optimization results considering the social benefits of CO2 emission avoidance in the power sector. Type of Power Plant Constrained Alternative Power Share Case ENG1
ENG2
ENG3
ENG5
ENG6
EWIN2
EWIN3
ESOL2
ESOL3
ESOL5
ENUC1
18,290
35,980
30,650
4,200
1,340
255
15
1,250
750
4,870
22,910
830
1,190
850
180
90
8
10
500
165
500
820
Number of units (4)
4
9
9
10
2
5
40
5
1
13
2
Power generation (GWh/yr) (5) Power cost (MM$/yr) (16)
18,290
Unconstrained Alternative Power Share Case 31,980 37,470 4,625 1,340 1,590 0 1,250
750
0
22,910
830
1,060
1,040
200
90
55
0
500
165
0
820
Number of units (4)
4
8
11
11
2
30
0
5
1
0
2
Variable Power generation (GWh/yr) (5) Power cost (MM$/yr) (16)
1017
51
1018
Table 10. Case Study 2020: Optimization results under CO2 constrained power production. Variable Values Average power unit cost ($/kWh) 0.050 CO2 offset (MT/yr) (9)-(13) 36.5 CO2 emissions (MT/yr) (7) 7.86 CO2 transport costs (MM$/yr) (10) 188 CO2 storage costs (MM$/yr) (11) 359 Total annual power cost (MM$/yr) (21) 5,980 Natural gas consumption (Bcf/d) (6) 2.56 Power Infrastructure Variables Type of Power Plant ENG2 Power generation (GWh/yr) (5) Power cost (MM$/yr) (16)
1019
Number of production units (4)
ENG3
4,000 115,800 132 5,280 1 34
ENG5
420 18 1
52
1020
Figures INPUTS / ENERGY SOURCES
POWER PLANTS
OUTPUTS / ENERGY PRODUCTS
END-USERS
EMISSIONS: GHG, CAC NGCC 7FA NGCC 7FB
NATURAL GAS
CO2 STORAGE
NGCC 7FB MEA OXY FUEL CCS
INDUSTRIAL
ST GT
NORDEX NORDTANK WIND
POWER
COMMERCIAL
SONKYO GAIA
SANYO BP CSP SOLAR
RESIDENTIAL OTEC SLP OTEC-OSP
APR-1400 URANIUM
AP-1000 EPR-1650
1021 1022
NUCLEAR WASTE
Figure 1. Superstructure of the power optimization model. 53
1023 MODEL KEY INPUTS
MODEL KEY OUTPUTS
• EXPECTED TOTAL
• ANNUAL
ELECTRICITY
PRODUCTION COST
DEMAND (ED).
(PCp) • EXPECTED
PROBLEM
SHARES OF
DILEMMA
ALTERNATIVE
MINIMIZE
POWER (REp).
COSTS
AVAILABLE (
).
SELECTION NATURAL
PLANT (PCp/EPp)
INCREASE COSTS DUE
GHG WIND
Is the termination criterion satisfy?
SELECTED (IEp) YES
CAC WASTE
PLANTS WITH THEIR
CAPACITIES (IEp ,ICp)
NUCLEAR
MITIGATION
• FUEL UNIT
• NUMBER OF POWER
CORRESPONDING
SOLAR
TO EMISSION
• TYPES OF PLANTS
ELECTRICITY
GAS VS
NUMBER OF
POWER PLANTS
• LEVILIZED COST BY
PRODUCTS
POWER
• MINIMUM / MAXIMUM
POWER PLANTS
• AIR EMISSIONS:
STRATEGIES
GHG, CAC (EGe,p)
PRICE (FCFp) NO • CO2 TAX AND
• FUEL CONSUMPTION (NGg ,TNG)
EMISSION TARGET (CTAX ,
1024 1025
CET).
Figure 2. General layout of the power optimization model.
54
Levelized electricity cost ($/kWh)
Simulated electricity costs at different gas prices for 2011 (Verification results) Electricity costs found in the literature for 2011 (Historical)
0.040 0.038
Simulation Results
0.036
0.034 0.032
Historical Value
0.030 0.028 0.0018
0.002
0.0022 0.0024 0.0026 Natural gas price ($/MJ)
0.0028
0.003
0.0032
1026 1027
Figure 3. Comparison of the historical levelized electricity cost in Abu Dhabi and the verification
1028
electricity costs at typical gas prices in 2011.
55
1029 Local gas price scenario w/o carbon tax
International gas price scenario w/o carbon tax
20
Number of power plants (unit)
18 16 14 12 10 8 6 4 2 0 ENG1
1030 1031 1032
ENG2
ENG5
ENG6 EWIN2 EWIN3 ESOL2 Types of power plants
ESOL3
ESOL5
ENUC1 ENUC2
Figure 4. Comparison of the power infrastructure design at local and international gas price without carbon tax in 2020 (4)-(5).
56
ENG1
ENG2
ENG3
ENG5
ENG6
ENUC1
ENUC2
Levelized electricity cost at local gas price ($/kWh)
0.130 0.120 0.110 0.100 0.090 0.080 0.070
0.060 0.050 0.040 0.030 0
1033 1034 1035
20
40
60
80
100
120
140
160
Carbon tax ($/tonne CO2)
Figure 5. Comparison of the levelized electricity costs for the power plants at local gas price under carbon tax in 2020 (5),(16).
57
Levelized electricity cost at international gas price ($/kWh)
ENG1
1037 1038
ENG3
ENG5
ENG6
ENUC1
0.180
0.160
0.140
0.120
0.100
0.080
0.060 0
1036
ENG2
20
40
60
80
100
120
140
160
Carbon tax ($/tonne CO2)
Figure 6. Comparison of the levelized electricity costs for the power plants at international gas price under carbon tax in 2020 (5),(16).
58
Local gas price
International gas price
25
CO2 offset (MT/yr)
23
Region 3
21 19 17 Region 2
15 13
11 Region 1
9 7 0
1039 1040 1041
20
40
60
80
100
120
140
160
Carbon tax ($/tonne CO2)
Figure 7. Comparison of CO2 offset at local and international gas price under carbon tax in 2020 (9),(13).
59
Local gas price
International gas price
Annual carbon cost (Billion $/yr)
4 3.5 3 2.5 2 1.5 1 0.5 0
0
1042 1043 1044
20
40
60
80
100
120
140
160
Carbon tax ($/tonne CO2)
Figure 8. Comparison of the annual carbon cost at local and international gas price under carbon tax in 2020 (8).
60
Local gas price
International gas price
Average levelized electricity cost ($/kWh)
0.110 0.100 0.090 0.080 0.070 0.060
0.050 0
1045 1046 1047
20
40
60
80
100
120
140
160
Carbon tax ($/tonne CO2)
Figure 9. Comparison of the average levelized electricity cost at local and international gas price under carbon tax in 2020 (5),(16)
61
Capacity constrained social benefit scenario
Capacity unconstrained social benefit scenario
Reference scenario
Number of power plants (unit)
40 35 30 25 20 15 10 5 0 ENG1
1048 1049 1050
ENG2
ENG3
ENG5
ENG6 EWIN2 EWIN3 ESOL2 ESOL3 ESOL5 ENUC1 ENUC2 Types of power plants
Figure 10. Comparison of the power infrastructure for: capacity constrained and unconstrained social benefit scenarios and the reference scenario for the year 2020.
62
Annual power production (TWh/yr)
Capacity constrained social benefit scenario
Capacity unconstrained social benefit scenario
Reference scenario
70 60 50 40 30
20 10
0 ENG1
1051 1052 1053
ENG2
ENG3
ENG5
ENG6 EWIN2 EWIN3 ESOL2 ESOL3 ESOL5 ENUC1 ENUC2 Types of power plants
Figure 11. Comparison of the annual power production for: capacity constrained and unconstrained social benefit scenarios and the reference scenario for the year 2020.
63
Annual power cost (Million $/yr)
Capacity constrained social benefit scenario
Capacity unconstrained social benefit scenario
Reference scenario
2000
1500
1000
500
0 ENG1 ENG2 ENG3 ENG5 ENG6 EWIN2 EWIN3 ESOL2 ESOL3 ESOL5 ENUC1 ENUC2
1054 1055 1056
Types of power plants
Figure 12. Comparison of the annual power cost for: capacity constrained and unconstrained social benefit scenarios and the reference scenario for the year 2020.
64
1057
REFERENCES
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100
[1] International Energy Agency. Key World Energy Statistics 2011. 2011. [2] Chong H-Y, Lam W-H. Ocean renewable energy in Malaysia: The potential of the Straits of Malacca. Renewable and Sustainable Energy Reviews. 2013;23:169-78. [3] The World Bank. World Development Indicators 2011. Washington 2011. [4] Dubai Statistics Center. Statistical yearbook – Emirate of Dubai. Dubai2010. [5] U.S. Energy Information Administration. Energy Outlook 2011. Washington, DC: U.S. Department of Energy; 2011. [6] Bachellerie IJ. Renewable Energy in the GCC Countries: Resources, Potential, and Prospects. Gulf Research Center; 2012. p. 230. [7] United Nations Environment Programme. Memorandum of Understanding on the Conservation and Management of Dugongs and Their Habitats Throughout Their Range. 2010. [8] Wasfi AK. The First Real Zero Gas Flaring Project in the Middle East and Gulf Region. Abu Dhabi International Conference and Exhibition. Abu Dhabi, United Arab Emirates: Society of Petroleum Engineers; 2004. [9] Nader S. Paths to a low-carbon economy—The Masdar example. Energy Procedia. 2009;1:3951-8. [10] Reiche D. Renewable Energy Policies in the Gulf countries: A case study of the carbonneutral “Masdar City” in Abu Dhabi. Energy Policy. 2010;38:378-82. [11] Sgouridis S, Kennedy S. Tangible and fungible energy: Hybrid energy market and currency system for total energy management. A Masdar City case study. Energy Policy. 2010;38:174958. [12] Al-Sallal KA, Al-Rais L, Dalmouk MB. Designing a sustainable house in the desert of Abu Dhabi. Renewable Energy. 2013;49:80-4. [13] Benkari N. The “Sustainability” Paradigm in Architectural Education in UAE. Procedia Social and Behavioral Sciences. 2013;102:601-10. [14] Sgouridis S, Griffiths S, Kennedy S, Khalid A, Zurita N. A sustainable energy transition strategy for the United Arab Emirates: Evaluation of options using an Integrated Energy Model. Energy Strategy Reviews. 2013;2:8-18. [15] Davidson MR, Dogadushkina YV, Kreines EM, Novikova NM, Seleznev AV, Udal’tsov YA, et al. Mathematical model of power system management in conditions of a competitive wholesale electric power (capacity) market in Russia. J Comput Syst Sci Int. 2009;48:243-53. [16] Varympopiotis G, Tolis A, Rentizelas A. Fuel switching in power-plants: Modelling and impact on the analysis of energy projects. Energy Conversion and Management. 2014;77:650-67. [17] Li YP, Huang GH. Electric-power systems planning and greenhouse-gas emission management under uncertainty. Energy Conversion and Management. 2012;57:173-82. [18] Pereira MVF, Pinto LMVG. Multi-stage stochastic optimization applied to energy planning. Mathematical Programming, Series B. 1991;52:359-75. [19] Piao MJ, Li YP, Huang GH. Development of a stochastic simulation–optimization model for planning electric power systems – A case study of Shanghai, China. Energy Conversion and Management. 2014;86:111-24. [20] Zhang D, Ma L, Liu P, Zhang L, Li Z. A multi-period superstructure optimisation model for the optimal planning of China's power sector considering carbon dioxide mitigation: Discussion on China's carbon mitigation policy based on the model. Energy Policy. 2012;41:173-83.
65
1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145
[21] Zhang D, Liu P, Ma L, Li Z. A multi-period optimization model for planning of China's power sector with consideration of carbon dioxide mitigation—The importance of continuous and stable carbon mitigation policy. Energy Policy. 2013;58:319-28. [22] Zhang D, Liu P, Ma L, Li Z, Ni W. A multi-period modelling and optimization approach to the planning of China's power sector with consideration of carbon dioxide mitigation. Computers & Chemical Engineering. 2012;37:227-47. [23] Cheng R, Xu Z, Liu P, Wang Z, Li Z, Jones I. A multi-region optimization planning model for China’s power sector. Applied Energy. 2015;137:413-26. [24] Hoster F. Effects of a European electricity market on the German electricity industry: results from a simulation model of the European power systems. Applied Economics. 1999;31:107-22. [25] Voorspools KR, D’haeseleer WD. Modelling of electricity generation of large interconnected power systems: How can a CO2 tax influence the European generation mix. Energy Conversion and Management. 2006;47:1338-58. [26] Watcharejyothin M, Shrestha RM. Effects of cross-border power trade between Laos and Thailand: Energy security and environmental implications. Energy Policy. 2009;37:1782-92. [27] Gnansounou E, Dong J. Opportunity for inter-regional integration of electricity markets: the case of Shandong and Shanghai in East China. Energy Policy. 2004;32:1737-51. [28] Ye M, Cai W, Wang C. A Multi-Period Multi-Region Optimization Model of China's Power Sector Considering Synergetic CO2 and Air Pollutants Control. Procedia Environmental Sciences. 2013;18:397-403. [29] Chen W, Li H, Wu Z. Western China energy development and west to east energy transfer: Application of the Western China Sustainable Energy Development Model. Energy Policy. 2010;38:7106-20. [30] Ziwei M, Can W. Assessing the Climate Impact of Renewable Energy Targets by BottomUp Modeling. Intelligent System Design and Engineering Application (ISDEA), 2010 International Conference on2010. p. 394-9. [31] AlFarra HJ, Abu-Hijleh B. The potential role of nuclear energy in mitigating CO2 emissions in the United Arab Emirates. Energy Policy. 2012;42:272-85. [32] Mondal MAH, Kennedy S, Mezher T. Long-term optimization of United Arab Emirates energy future: Policy implications. Applied Energy. 2014;114:466-74. [33] Turton R, Bailie R, Whiting W, Shaeiwitz J. Analysis, Synthesis and Design of Chemical Processes 3rd ed: Prentice Hall; 2009. [34] Seider W, Seader J, Lewin D, Widagdo S. Product and Process Design Principles. 3rd ed: Wiley; 2009. [35] Biegler L, Grossmann I, Westerberg A. Systematic Methods of Chemical Process Design. 1st ed1997. [36] Georgis D, Jogwar SS, Almansoori AS, Daoutidis P. Design and control of energy integrated SOFC systems for in situ hydrogen production and power generation. Computers & Chemical Engineering. 2011;35:1691-704. [37] Liu H, Almansoori A, Fowler M, Elkamel A. Analysis of Ontario's hydrogen economy demands from hydrogen fuel cell vehicles. International Journal of Hydrogen Energy. 2012;37:8905-16. [38] Almansoori A, Shah N. Design and operation of a future hydrogen supply chain: Multiperiod model. International Journal of Hydrogen Energy. 2009;34:7883-97.
66
1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
[39] T. Tietenberg. Environmental and Natural Resource Economics. MA: Addison-Wesley Reading; 1996. [40] Abu Dhabi Water & Electricity Company. Statistical Report 1998 - 2012. 2013:73-4. [41] Statistics Centre - Abu Dhabi. Energy & Environment Statistics 2011. Abu Dhabi2012. [42] Black & Veatch Holding Company. Cost and Performance data for power generation technologies. National Renewable Energy Laboratory; 2012. [43] Rubin ES, Chen C, Rao AB. Cost and performance of fossil fuel power plants with CO2 capture and storage. Energy Policy. 2007;35:4444-54. [44] Rubin ES, Rao AB, Chen C. Comparative assessments of fossil fuel power plants with CO2 capture and storage. In: Rubin ES, Keith DW, Gilboy CF, Wilson M, Morris T, Gale J, et al., editors. Greenhouse Gas Control Technologies 7. Oxford: Elsevier Science Ltd; 2005. p. 285-93. [45] Davison J. Performance and costs of power plants with capture and storage of CO2. Energy. 2007;32:1163-76. [46] ICF International Company. Technology Characterization: Steam Turbines. Washington, DC: Environmental Protection Agency; 2008. [47] U.S. Energy Information Administration. Updated Capital Cost Estimates for Utility Scale Electricity Generating Plants. Washington, DC: U.S. Department of Energy; 2013. [48] Harijan K, Uqaili MA, Memon M, Mirza UK. Assessment of centralized grid connected wind power cost in coastal area of Pakistan. Renewable Energy. 2009;34:369-73. [49] Janajreh I, Su L, Alan F. Wind energy assessment: Masdar City case study. Renewable Energy. 2013;52:8-15. [50] Shawon MJ, El Chaar L, Lamont LA. Overview of wind energy and its cost in the Middle East. Sustainable Energy Technologies and Assessments. 2013;2:1-11. [51] Radhi H. Trade-off between environmental and economic implications of PV systems integrated into the UAE residential sector. Renewable and Sustainable Energy Reviews. 2012;16:2468-74. [52] Harder E, Gibson JM. The costs and benefits of large-scale solar photovoltaic power production in Abu Dhabi, United Arab Emirates. Renewable Energy. 2011;36:789-96. [53] Straatman PJT, van Sark WGJHM. A new hybrid ocean thermal energy conversion– Offshore solar pond (OTEC–OSP) design: A cost optimization approach. Solar Energy. 2008;82:520-7. [54] World Nuclear Association. The Economics of Nuclear Power. London: World Nuclear Association; 2014. [55] Nuclear Energy Institute. Costs: Fuel, Operation, Waste Disposal & Life Cycle. Washington, DC Nuclear Energy Institute; 2013. [56] Alonso G, Palacios JC, Ramirez JR, Longoria LC, del Valle E. Alternatives of Financing for New Nuclear Reactors in Mexico. In: Agency IAE, editor. Estado de Mexico, Mexico: Instituto Nacional de Investigaciones Nucleares; 2013. [57] Areva. Key assets of the EPR reactor: Competitiveness. Areva; 2013. [58] Abu Dhabi Council for Economic Development. The Abu Dhabi Economic Vision 2030. 2008. [59] United Nations. United Nations Framework Convention on Climate Change 1992. [60] Hoeller P, Wallin M. Energy prices, taxes and carbon dioxide emissions. OECD Economic Studies1991. p. 91-105.
67
1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235
[61] Dahowski RT, Li X, Davidson CL, Wei N, Dooley JJ, Gentile RH. A preliminary cost curve assessment of carbon dioxide capture and storage potential in China. Energy Procedia. 2009;1:2849-56. [62] GAMS Development Corporation. General Algebraic Modeling System (GAMS). 2013. [63] GAMS Development Corporation. Cplex Solver Manual. GAMS Development Corporation 2013. [64] Abu Dhabi Water and Electricity Company. ADWEC Statistical Report 1998 - 2012. Abu Dhabi, UAE: Abu Dhabi Water & Electricity Company; 2012. p. 73-4. [65] Abu Dhabi Statistic Centre. Energy & Environment Statistics 2011. Abu Dhabi, UAE: Statistic Centre - Abu Dhabi; 2012. [66] The Abu Dhabi Regulation & Supervision Bureau. Annual Report 2011. Abu Dhabi, UAE: The Abu Dhabi Regulation & Supervision Bureau; 2012. p. 6. [67] Hopwood D. Abu Dhabi's Masdar plan takes shape Renewable Energy Focus. 2010;11:1823. [68] Abu Dhabi Water and Electricity Company. Existing Projects. Abu Dhabi, UAE: Abu Dhabi Water & Electricity Company; 2012. [69] Taweelah Asia Power Company. Project Details. Abu Dhabi, UAE: Taweelah Asia Power Company 2011. [70] Al Mirfa Power Company. Production. Abu Dhabi, UAE: Al Mirfa Power Company; 2014. [71] Arabian Power Company. All Projects. Abu Dhabi, UAE: Arabian Power Company; 2014. [72] Canty D. Shuweihat S2 in Abu Dhabi reaches full operation. Middle East Utilities; 2012. [73] PR Newswire. Gasco Energy announces 2012 Financial and Operating results and estimated proved reserves. Denver, USA: Market Watch - The Wall Street Journal; 2013. [74] PR Newswire. Gasco Energy Announces First Quarter 2011 Results. Denver, USA: Alaska Dispatch; 2011. [75] Krane J. Energy Policy in the Gulf Arab States: Shortage and Reform in the World's Storehouse of Energy. In: USAEE, editor. Cleveland, US2013. [76] Energy Information Administration. Annual Energy Outlook 2014. Washington, DC: Energy Information Administration 2014. [77] Hamed OA, Al-Sofi MA, Imam M, Mustafa GM, Ba-Mardouf K, Al-Washmi H, et al. Thermodynamic Analysis of Al-Jubail Power/Water Co-Generation Cycles. Al-Jubail, Saudi Arabia: Saline Water Conversion Corporation, Al-Jubail Desalination & Power Plants; 2000. p. 20. [78] Medlock III KB. Natural gas price in Asia: What to expect and what it means Houston: Rice University; 2014. [79] Krane J. An expensive diversion: Abu Dhabi’s renewables investments in the context of its natural gas shortage. Cambridge Working Papers in Economics. Cambridge, UK: University of Cambridge; 2012. [80] Miller K. Electricity and Water Demand Forecasts Abu Dhabi Abu Dhabi Water and Electricity Company 2012. p. 17. [81] Mezher T, Dawelbait G, Abbas Z. Renewable energy policy options for Abu Dhabi: Drivers and barriers. Energy Policy. 2012;42:315-28. [82] Masdar - Abu Dhabi Government. Delivering Sustainability: Sustainability Report 2012. Masdar; 2012. [83] Emirates Nuclear Energy Corporation. ENEC's Program: Timeline. Abu Dhabi: Emirates Nuclear Energy Corporation 2014. 68
1236 1237 1238 1239 1240 1241 1242 1243
[84] United Nations. Compilation of economy-wide emission reduction targets to be implemented by Parties included in Annex I to the Convention. 2011. [85] Energy Information Administration. Country Analysis: United Arab Emirates. Washington, DC U.S. Department of Energy; 2013. [86] Al Kindi H. Abu Dhabi's gas dilemma. The Economic Review Autumn 2010. Abu Dhabi, UAE: Abu Dhabi Council for Economic Development 2010.
69
