Energy 94 (2016) 422e430

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

2050 LCOE (Levelized Cost of Energy) projection for a hybrid PV

(photovoltaic)-CSP (concentrated solar power) plant in the
Atacama Desert, Chile
C. Parrado a, *, A. Girard b, F. Simon c, E. Fuentealba a, d
a
Centro de Desarrollo Energ
etico Antofagasta (CDEA), University of Antofagasta, Av. Universidad, de Antofagasta 02800, Antofagasta, Chile
b
n
Universidad Adolfo Iba ~ ez, Av. Padre Hurtado 750, Vin
~ a del Mar, Chile
c
Department of Civil Engineering, University of Granada, Av. Severo Ochoa s/n, 18071, Granada, Spain
d
Solar Energy Research Center (SERC-Chile), Av. Tupper 2007, Piso 4, Santiago, Chile

a r t i c l e i n f o a b s t r a c t

Article history: This study calculates the LCOE (Levelized Cost of Energy) on the PSDA (Atacama Solar Platform) for a solar
Received 29 May 2015 esolar energy mix with the objective of evaluate new options for continuous energy delivery. LCOE was
Received in revised form calculated for three 50 MW (megawatt) power plants: A PV (photovoltaic), a CSP (concentrated solar
14 October 2015
power) plant with 15 h TES (thermal energy storage) and a hybrid PV-CSP plant constituted with 20 MWp
Accepted 5 November 2015
Available online 28 November 2015
of PV and 30 MW of CSP with 15 h TES. Calculations present two scenario projections (Blue Map and
Roadmap) until 2050 for each type of plant. Due to the huge solar resource available in northern Chile, the
PV-CSP hybrid plant results to be a feasible option for electricity generation, as well as being effectively
Keywords:
LCOE
able to meet electricity demand profile of the mining industry present in the area. This type of energy
PV could mitigate long-term energy costs for the heavy mining activity, as well as the country CO2 emissions.
CSP Findings point out that PV-CSP plants are a feasible option able to contribute to the continuous delivery of
Mining industry sustainable electricity in northern Chile. Moreover, this option can also contribute towards electricity
Solar resource price stabilization, thus benefiting the mining industry, as well as reducing Chile's carbon footprint.
Hybrid © 2015 Elsevier Ltd. All rights reserved.

1. Introduction displeasure in the society that is increasingly aware of this issue.

The north of Chile has been bolstering its electricity consump-
tion in the last 20 years due to new mining investments in the first,
second and third region [1]. Economical pressures and population
growth have pushed government to look at cost effective options,
mainly those exploiting fossil fuel sources, to cope with the
increased demand for electricity in the northern Chile inter-
connected power system (called SING e Sistema Interconectado del
Norte Grande). New thermal power plants running on fossil fuel
sources were built because the generated electricity has lower cost
than other conventional technologies [1]. The SING's electricity
comes mainly from Coal and LNG (Liquefied Natural Gas) sources
(approximately 90% of total electricity generation) [1]. This
dependence on fossil fuels has a great impact on the environment,
allowing the emission of many tons of CO2 each year into the at-
mosphere, causing the rise of environmental concerns and
* Corresponding author.
E-mail address: cristobal.parrado@uantof.cl (C. Parrado).
Moreover, Chile is not a fossil energy producer; the country satisfies
its internal consumption based mainly on imported fuels [2]. It is
then dependent of international energy markets in order to secure
its needs, which makes Chile vulnerable against supply disruptions
and price volatility. The SING provides 90% of its electricity pro-
duction to mining industry located in the northern regions, and
electricity price inflation has direct influence on their operation
costs affecting product selling prices and then consumers. It is
essential to build a safe and reliable energy matrix and to diversify
energy generation with the inclusion of clean and safe technologies
that take full advantage of the great potential of the natural re-
sources in northern Chile. In addition, the Ministry of Energy and
Environment aims to regulate the CO2 emissions from the mining
industry since this sector is the main source of CO2 emissions in the
country [3]. In response, companies are developing various clean
energy generation technologies to meet their electricity demands.
The great amounts of CO2 emitted from the mining industry are
translated into great pollution taxes, and thus, it is important for
this industry to explore the current and future options for clean and
safe electricity [4,5].

http://dx.doi.org/10.1016/j.energy.2015.11.015
0360-5442/© 2015 Elsevier Ltd. All rights reserved.
 C. Parrado et al. / Energy 94 (2016) 422e430
In order to meet future electricity demands with clean and
reliable energy, it is necessary to exploit the natural resources of the
country. Northern Chile, specifically the Atacama Desert, is known
as the most arid desert in the world and has the highest solar ra-
diation ranging between 7 and 7.5 kWh/m2 daily [6e8]. DNI (Direct
Normal Irradiation) reaches approximately 3000 kWh/m2 per year,
meaning that it is an optimal location for the placement of solar
plants. Despite the high levels of radiation available in northern
Chile, solar power generation is variable and discontinuous, and
this means that it is not a dependable choice for end consumers.
Presently, the market offers different options to complement solar
plants and increase their productivity. TES (Thermal Energy Stor-
age) allows CSP (Concentrating Solar Power) plants to increase
electricity production in a range between 3 and 17 h, making
generation more continuous and reliable. Another way to cope with
reduced or curtailed energy production when the sun sets or the
sky is cloudy is fuel back-up or hybridisation. Several sources show
that PV (photovoltaic) or CSP plants can produce electricity during
approximately 6e10 h per day without TES, and that they can be
coupled with conventional fossil fuel plants as back-up systems
when solar resource is unavailable, in order maintain a high ca-
pacity factor [6,8,9]. Even more recently studies have proven that is
possible make a continuous production of electricity implementing
SolareSolar plants [10,11]. PV and CSP can be integrated for reliable
production of electricity.
The paper presents an economic study on the feasibility of a
combination of PV-CSP solar plants in northern Chile with the
purpose of making a clean, secure and reliable Chilean energy
matrix. The first approach calculates the LCOE (Levelized Cost of
Energy) for current PV and CSP technologies. The second approach
evaluates economically the LCOE for a hybrid Solar- Solar plant, PV
and CSP. Also, a projection of the LCOE for these plants was made in
order to show the current and future scenario of the solar energy in
the north of Chile.
2. Energy in Chile
The SING has an installed capacity of 4308 MW (megawatt), and
meets the demand of the first, second and fifteenth regions of Chile.
A growth of 6.5% is expected between 2012 and 2018, due to sky-
rocketing mining activity in the coming years [1]. The electricity
demand in the SING system is expected to reach about 5000 MW in
2020 [1]. The actual demand of SING is held by 90% of unregulated
customers, most of them mining companies, and the remaining 10%
Fig. 1. SING installed capacity per technology in 2014 [12].
423
by regulated customers. The SING installed capacity is distributed
according to Fig. 1.
In order to secure energy supply, the SING electricity sector rely
heavily (97.2%) on coal, gas or oil-fired power plants as shown in
Fig. 1. The Centre of Energy Discharge (CDEC-SING), an entity that
regulates the generation of energy in terms of the lowest marginal
cost has not allowed until now the exploitation of REs because their
energy costs are high in comparison with conventional energies.
The decision to meet power demands only with fossil energy has
kept electricity prices low, but has significantly incremented CO2
emissions.
It is important to mention that northern Chile does not have
optimal conditions to install hydropower and cogeneration plants
due to the low amount of rain and snowmelt, suggesting that coal,
gas and diesel plants will remain in this region in the future.
However, this is not a sustainably viable solution for the future of
Chile, considering the environmental issues and CO2 emissions
related with the exploitation of fossil fuel sources. Moreover, it has
become crucial for the country to be able to provide adequate en-
ergy supplies in order to continue its economic growth. Hanel and
Escobar [2] have noted that it is of critical importance to ensure the
development of indigenous energy sources at a sufficient rate such
as needed for the substitution of imported energy resources in
order to rapidly achieve energy security and a degree of energy
independence. Indeed, national energy debates and investments
into the country's energy security have been strengthened and
relying more on RE sources has become a target with the intro-
duction of the law 20/25 (20% of RE by 2025), aiming also at taking
more control in the future over the increasing electricity cost.
Although RE systems are generally more expensive than traditional
fossil fuels, they are recognized for entailing lower environmental
and social impacts [13,14].
As seen in Table 1, solar energy is first in the ranking for energy
projects in the country, reaching 1618 MW in qualification and
2881 MW of permit approved in 2014. This shows also that in-
vestors are already interested in to exploiting the country's radia-
tion, and solar is likely to provide large quantities of energy to the
system in the future.
3. Potential for solar plants in northern Chile
With some of the highest solar DNI (direct normal irradiance)
rates and clear sky indexes in the world, Chile presents a great
potential for solar power [9,16,17]. Therefore, political actors seri-
ously started to take into consideration solar power a genuine
sustainable solution. Actually, solar energy development in Chile is
small, mostly focusing on water heating applications for the resi-
dential sector. The total contribution of solar energy to the primary
energy consumption of Chile is negligible [18]. However, the rise of
electricity cost [19] is driving industrial companies to consider self-
generation or collaboration on solar projects [20].
CNE (Chile's National Energy Commission) says it is feasible to
connect up to 2.2 GW of solar PV plants to the national grid over the
next 15 years, in a newly published plan for the expansion of the
national transmission system, claiming that the investment price of
solar PV (photovoltaic) power plant is equivalent or lower to the
price of building coal generation plants [21e23]. As announced at
the PV Insider Latin America conference, as well as by the Deutsche
Bank, Chile is currently one of the most exciting markets for PV [24].
The photovoltaic industry in Chile has seen a considerable change
from 2013 to 2014 with a growth of 103% of PV plants in operation,
while the plants under construction has growth 184%, according to
the PV Insider report “PV industry roadmap in Chile 2014” [25].
The solar thermal power technology (i.e. CSP) presents also a
significant potential for covering a large part of the electricity
424 C. Parrado et al. / Energy 94 (2016) 422e430

Table 1
Present and future generation projects in SING in 2014 [15].

Type of source In certification Approved

Capacity (MW) Investment (US$ million) Capacity (MW) Investment (US$ million)

Solar 1.618 6.019 2.881 9.944

Liquefied Natural Gas (LNG) 1.300 1.158 0 0
Wind 140 345 1.592 3.373
Coal 0 0 1.770 3.500
Diesel 0 0 207 340
Fuel oil n
6 0 0 216 302
Geothermic 0 0 50 180
Total 3.058 7.522 6.717 17.639

demand [26]. This well developed technology is mostly established during summertime at values close to 8 kWh/m2/day, with winter
in the United States and Spain, with market shares of 57% and 42% minima nearing 3 kWh/m2/day.
respectively in 2013 [9], while in Chile it is still scarcely known. The As electricity production is directly dependent on solar irradi-
potential of CSP is greater than PV due to higher capacity factors ation available, similar trends of annual electricity production
(with the use of TES systems). In 2012, Spain's based Ibereolica was would observed as those of radiation in Fig. 3. The solar electricity
one of the first companies to move to Chile and has received
environmental approval for its first project, Termosolar Pedro de
Valdivia of 360 MW [27]. The company has also filed the Environ-
mental Impact Assessment for a second project also in Antofagasta
Region, Maria Elena, a 400 MW CSP complex to use tower tech-
nology with molten salt as heat transfer fluid and thermal storage
medium [28]. It is only very recently that funding programs are
delivered to develop CSP plants in Chile [29,30]. For example in
January 2014, Chile's Ministry of Energy and public sector devel-
opment agency Coporacion de Fomento de la Produccion (Corfo)
awarded Abengoa the contract to build a 110 MW (megawatt) CSP
plant in Antofagasta region. It is estimated that the plant would
avoid about 643,000 tons of CO2 emissions per year [31e33].
Fig. 2 presents the distribution of annual GHI (global horizontal
irradiation) in Chile [34]. The GHI (kWh/m2/year) is represented as
the yearly sum of DNI and DIF (direct normal and diffuse horizontal
irradiations) incident per unit area on a surface horizontal to the
ground.
Herna
ndez-Moro and Martínez-Duart [35] observed that PV
seems always more appropriate for areas located in middle to high
latitudes of the Earth, while CSP systems, preferably with thermal
storage incorporated, yield their best performance in arid areas
located at relatively low latitudes. This is explained because PV and
CSP technologies do not make the same use of the solar resource.
CSP technology uses only the direct radiation of the sun to
concentrate it, while PV uses both direct and diffuse irradiation.
Therefore, while projects of PV power plants will look at the global
irradiation value, CSP developers tend to consider the value of DNI
(Direct Normal Irradiance), since sun tracking systems allow con-
centrators to always be normally orientated towards the sun. The
low limit value of annual DNI for CSP plants to achieve economic
performance is generally set between 1900 kWh/m2 and
2000 kWh/m2 [36,37], although Purohit et al. [38] expects that in
near future locations with lower DNI values (over 1600 kWh/m2)
could also become financially feasible with the development of new
technologies, advancement of materials, economy of scale,
manufacturing capability along with the enhanced policy measures
etc. In 2012, most of the CSP power plants are installed in the
Southwest of the United States and in the South of Spain, where
DNI values range from 2000 to 2850 kWh/m2/year [6,35,39,40]. In
the case of fixed optimally-inclined south-oriented PV systems,
investors have set the boundary conditions for global irradiation
values from 1300 to 2300 kWh/m2/year and that include locations
with low direct radiation such as Germany [41e43].
The NREL model estimations of monthly average global hori-
zontal solar radiation for four locations, representative of northern
Chile are shown in Fig. 3 [44]. As seen, the solar radiation peaks Fig. 2. Global Horizontal Irradiation (GHI) in Chile [34].
 C. Parrado et al. / Energy 94 (2016) 422e430
Fig. 3. Daily average GHI on monthly mean at selected locations, representative of northern Chile.
generation would be greater between the months of October and
March and lower between the months of April and September,
unless the solar plant is equipped with a fossil fuel back-up system
able to compensate during winter months. For PV generation, a
high inclination angle of the PV modules can be used to optimized
generation in winter, and thus minimize seasonal differences at one
particular location [10]. However, at equivalent plant specifications,
the electricity generation would be greater in locations where the
solar irradiation is higher, such as in Location 1 and 2 compared to
Location 3 and 4.
CSP plants have the possibility to be hybridized with the fossil
fuel to make the plant produce electricity for 24 h or/and can be
integrated with a storage system to store a part of the concentrated
heat during the day to generate electricity in the cloudy days or
when the sun sets in order to improve its thermal efficiency and
capacity factor and hence reduce the cost of the technology.
4. The different technologies to the mix
4.1. Configurations of the studied plants
Solar energy is unattractive to some investors because of its
lower capacity factor and higher electricity cost compared to con-
ventional power plants. To make solar energy economically viable
and supply secure, it is necessary to increase the efficiency of the
plants. To do so, solar power requires storage (for load-balancing) to
cope with its output intermittency resulting from weather/seasonal
fluctuations.
EES (Electrical energy storage) such as batteries represents an
effective solution to mitigate the imbalance associated to PV
plants [45]. A background on the development of various energy
storage options could be found in the Sandia National Laboratories
handbook [46]. They are available at various sizes from a few
kilowatts to hundreds of megawatts. Today, there are numerous
commercial batteries including lead-acid, lithium-ion, sodium-
sulphur, nickel-cadmium, or vanadium redox flow, each with
different characteristics including energy capacity cost, round trip
efficiency, depth of discharge, life, discharge duration, cycle fre-
quency, energy/power density, environmental impact, etc.
[46e48]. However, in spite of the recent developments on EES
systems efficiency by the use of pulse width modulator inverters
425
[49] or by ultracapacitors [50], batteries still require further work
to improve their cycle life and thus reduce the overall costs [51,52].
According to Cervone et al. [53], only the lead-acid batteries can
reach a suitable level of cost effectiveness in utility scale appli-
cations, while lithium-ion, sodium-sulphur or flow batteries still
remain unsuitable until their costs decrease. However, lead-acid
batteries have the disadvantages to be slow to charge, to have a
limited depth of discharge and a limited amount of charge/
discharge cycles [54]. EES are mainly used for off grid applications.
For large-scale grid-connected PV power plants, EES is not an
economically viable option. For these reasons, the integration of
EES to the PV plant system is not considered in the cost projection
analysis.
Regarding CSP technology, TES systems are an effective solution
to increase the plant capacity factor allowing electricity generation
for an extra 3e17 h a day. TES systems use molten salt, typically a
mixture of sodium and potassium nitrates (KNO3eNaNO3) which
melts at about 220
C [55], as a heat carrier and storage medium.
The capacity factor of a CSP plant is approximately 40% without
storage and 60% with TES [6,9]. The other option to increase the
capacity factor of CSP plants is to incorporate a back-up system that
can generate either electricity or the remaining thermal energy
required in order to reach the nominal power output of the steam
turbine in a continuous operation mode. The back-up system can
make the use of other energy sources, such as the conventional
fossil fuels, but also the wind, biomass, or the sun [8].
An accurate assessment of the energy resources available at the
plant location is important to select the most appropriate energy
mix that can provide continuous delivery. On the PSDA, the DNI
and GHI values are some of the highest of the world, thus making
hybrid PV-CSP plants an interesting option to investigate for this
location.
Three types of solar plants of 50 MW capacity each are included
in the analysis. Authors assumed a fixed-angle PV power plant
without storage systems. The second plant is a parabolic trough
CSP system with 15 h of TES and the third one is a hybrid PV-CSP
plant enabling 24-h electricity generation composed with
20 MWp fixed-angle PV and 30 MW parabolic trough CSP with
15 h of TES. In order to optimize the hybrid PV-CSP system output,
the plant is set to prioritize the generation of electricity from the
PV side when solar resource is available. During this time the CSP
426 C. Parrado et al. / Energy 94 (2016) 422e430
side is used to collect molten salt in order to produce electricity
later at night. For both the PV and the hybrid PV-CSP plants, the PV
collector area is constituted with south orientated mono-
crystalline silicon PV modules, inclined at the yearly optimum
inclination angle.
4.2. LCOE model
According to Herna
ndez-Moro and Martínez-Duart [56], an
analytical model can be used for the projection of the LCOE of solar
technology. The LCOE is equal to the sum of all the cost incurring
during the lifetime of the project divided by the units of energy
produced during its lifetime [16,56]. For a precise LCOE projection
for CSP and PV, the most convenient approach is to use the life-
cycle cost method as given by Herna
ndez-Moro and Martínez-
Duart [35,56]. The LCOE of the solar plant installed in a year t can be
calculated using Equation (1) as follows:
0  1
P
C ðt Þ þ L þ Ti¼1 C ðt ÞðO&MþI Þ
B ð1þdÞ
i
C in solar energy industry in different global locations [5]. Unlike CSP,
LCOEðt Þ ¼ B
@   C
A (1)
PT STFhð1DRÞi
i¼1 ð1þdÞ
i
where C(t) is the total installation cost of the system ($/W), L is the
land cost ($/W), O&M represent the operation and maintenance
costs (%), I the insurance costs (%), S stands for the solar Resource
(kWh/m2/year), TF is the tacking factor (%), h is the performance
factor (m2/W), DR the degradation rate (%), d the discount rate (%),
and T the estimated lifetime of the system (years). This expression
represents the LCOE for a one plant type per year of the projection.
The computational simulation was made in Matlab. The most
important variables used are C(t), S and h because they depend on
the plant location. The cost of the system at a certain time t, CðtÞ,
depends of the cumulative installed capacity at that time, qðtÞ, and
is defined as following:
 b
qðt2 Þ
Cðt2 Þ ¼ Cðt1 Þ
qðt1 Þ
The learning coefficient, b, is related to the learning rate, LR, and
the progress ratio, PR, that reflects the advance of technology [57].
The learning rate and progress ratio are given by:
PR ¼ 2b
Fig. 4. Global CSP cumulative installed capacity and annual CSP electricity production [9,36,58].
LR ¼ ð1  PRÞ (4)
Therefore, the learning coefficient can be calculated as follows:
logðLR  1Þ
b¼ (5)
logð2Þ
5. Simulation
5.1. Calculating global parameters for PSDA
Several sources in the literature were consulted in order to
obtain precise initial values for annual CSP electricity production
and the cumulative installed capacity [9,36,58,59]. Fig. 4 shows the
global cumulative installed capacity between 1990 and 2014. It can
be observed that during the period between 2008 and 2014 the
global cumulative installed capacity increased rapidly, due to the
large number of technological improvements. This fact has helped
many international companies to increase their capital investments
Fig. 5 shows that the cumulative installed capacity of PV has been
increasing from the year 2000 to the present. This trend is pro-
duced because of the early adoption of PV technology by the energy
industry and researchers.
A projection for global cumulative installed capacity and annual
production of electricity for the two technologies was made. At the
beginning of 2014 the cumulative installed capacity, qð0Þ, was
191 GW for PV systems and 3.93 GW for CSP systems, and it was
taken as a reference year for the projection to 2050. In order to
obtain the correct projection to the qðtÞ, it was necessary to use the
two different studies made by the IEA (International Energy
Agency), Blue Map and Roadmap scenarios [36,59]. Each scenario
gives a projection of PV and CSP technologies up to 2050 with
different market penetrations. According to the Blue Map and
Roadmap scenario, PV and CSP systems would provide respectively
6% and 5% of the annual global electricity production in 2050
[9,36,59]. The Roadmap scenario has a high expectation for massive
(2)
installation of future solar technologies worldwide; unlike the Blue
Map scenario, which is more conservative with the evolution of the
solar market.
For PV technology and for both projection scenarios, a logistic
function called the S-shaped curve can be used because it best fits
the data. For CSP technology in the Blue Map scenario the same
(3) logistic function is utilized, while for the Roadmap Scenario a
 C. Parrado et al. / Energy 94 (2016) 422e430 427

Fig. 5. Global PV cumulative installed capacity and annual PV electricity production [9,58,59].

second grade polynomial function is used [56]. The S-shaped curve have published new methods to boost the efficiency of solar
and the second grade polynomial function used to calculate the panels, reaching higher values compared to current panels. Unlike
projections of cumulative installed capacity in the Blue Map and PV, CSP does not reach the highest value of cumulative installed
Roadmap scenarios are expressed by the following Equations (6) capacity, but in contrast with the present situation of the tech-
and (7). nology, the maximum at 2050 (1000 GW) is an excellent projec-
tion. Several researchers have investigated different options of
erðt2014Þ CSP designs such as the parabolic trough, dish stirling, linear
qðt Þ ¼    1 erðt2014Þ  (6)
1  M þ fresnel and solar tower [7,36]. These technologies are helping to
qð0Þ M
expand the solar energy across the world because each one has its
own characteristics, making them feasible for different global
qðtÞ ¼ 0:459ðt  2014Þ2 þ 9:0725ðt  2014Þ þ 3:9 (7) scenarios.

where q(0) and M are values of the initial and maximum cumulative
installed capacity, respectively, and r is the growth parameter. Ac-
cording to Hern
andez-Moro and Martínez-Duart [56], the specific
values for PV according to the Blue Map and Roadmap scenario are
respectively: q(0) is 191 GW and 191 GW in 2014, M is 1150 GW and
3155 GW in 2050 and r is 0.102 and 0.185. For CSP technology ac-
cording to the Blue Map q(0) is 3.93 GW, M is 630 and r is 0.32. For
the Roadmap Scenario to CSP only a q(0) of 3.93 GW is necessary,
given that the expression is a second grade polynomial [56].
Fig. 6 demonstrates that the photovoltaic technology will lead
solar generation in 2050 with approximate 3000 GW of cumula-
tive installed capacity, while the CSP will reach almost 1000 GW.
This situation could be produced because PV technology was
investigated previous to CSP [9]. Presently, several researchers
5.2. Calculating specific parameters for PSDA
According to the model from Herna
ndez-Moro and Martínez-
Duart [35,56], there are three factors that are specific to the location
of the plant: the cost of the system in each year, CðtÞ, the solar
resource, S, and the performance factor, h.
The SAM software was used to obtain the costs of the system
and the LCOE from the different studied technologies [60]. The cost
obtained for the base year C(0) is 1.24 US$/W for PV and 6.12 US$/W
For CSP [61e63].
For the studied location in the PSDA region (Lat: 24.087
,
Long: 69.913
), the values of DNI and GHI are 3421.9 kWh/m2/
year and 2576 kWh/m2/year respectively [44]. The performance

Fig. 6. Projection of cumulative installed capacity for Blue Map and Roadmap scenario to PV and CSP technologies.
428 C. Parrado et al. / Energy 94 (2016) 422e430
factors for the CSP technology, hCSP, and for PV, hPV can then be
calculated using the following Equations (8) and (9).
 
Electricity produced per installed watt kWh
W
 
hcsp ¼ (8)
Utilized solar resource ðDNIÞ kWh
m2
 
Electricity produced per installed watt peak kWh
Wp
hpv ¼   (9)
Utilized solar resource ðGHIÞ kWh
m2
The studied parabolic trough solar plant in the PSDA location
has an electricity production per installed watt of 4715.2 kWh/W.
This value was calculated dividing the electricity produced by the
plant in one year, 235.75 GWh, by the plant capacity, 50 MW. The
utilized solar resource corresponding to DNI was 3421.9 kWh/m2/
year. Therefore, the CSP performance factor for the chosen location
is 1.377 m2/kWh. For the PV plant, the electricity production per
unit of installed capacity is 1374.61 kWh/Wp. This value was ob-
tained dividing the electricity produced by the PV plant in one year,
68.73 GWh, by the plant capacity, 50 MWp. With a GHI value of
2576 kWh/m2/year at the studied location, the PV performance
factor is then 0.533 m2/kW. In summary, two performance factors
were obtained, 1.377 m2/kW and 0.533 m2/kW for CSP and m2/kW
PV technology respectively. These values are very different, due to
the fact that CSP technology uses only the direct radiation of the
sun, while PV uses both direct and diffuse irradiation. The CSP
electricity production value was obtained from the SAM software
for two types of CSP plants, a parabolic trough and a Flat Plate of
50 MW each.
Values for hybrid PV-CSP plant was prorated according to the
percentage of each technology in the mix, using the values in
Table 2.
6. Results and discussion
The main goal of the simulation is to compare the LCOE of the
two most used solar technology plants (PV, CSP), as well as with the
LCOE of a hybrid PV-CSP plant with 24 h of electricity generation.
The LCOE was calculated for three types of solar plants: a fixed
optimally-inclined south-oriented 50 MWp PV system; a 50 MW
parabolic trough CSP plant with 15 h of TES; and a PV-CSP hybrid
plant able to operate 24 h a day constituted with a fixed optimally-
orientated and inclined 20 MWp PV system and a 30MW parabolic
trough CSP system with 15 h of TES. The parameters for the LCOE
projection were calculated according to the model by Herna
ndez-
Moro and Martínez-Duart [56]. Table 2 gives the specific parame-
ters for the location in the Atacama Desert. The cost of PV tech-
nology is lower than CSP because the PV technology is more mature
and its market penetration is stronger than CSP [59]. However, CSP
technology cost tend to decrease and will become more affordable
in the future, due to recent developments in TES systems intro-
ducing new materials for thermal storage [65].
Figs. 7 and 8 shows the LCOE of PV, CSP and PV-CSP plants for
two scenarios: Blue Map (1) and Roadmap (2) scenarios. For the
scenario (1), LCOE values obtained varied between 12.88 and
8.43 cUS$/kWh for PV technology and between 15.29 and
9.02 cUS$/kWh for CSP technology, between 2014 and 2050
respectively. For the scenario (2), LCOE values obtained varied be-
tween 10.74 and 7.79 cUS$/kWh for PV technology and between
14.93 and 7.57 cUS$/kWh for CSP technology. LCOE values obtained
for both PV and CSP technology and both scenarios in 2014 are in
the same range than those obtained by Kost et al. [63] in the PSDA
location, showing that such LCOE values are from the lowest in the
present market. According to Fraunhofer ISE study [63], the LCOE
for solar technologies reaches values between 9.21 and 17.68 cUS$/
kWh for PV and between 17.27 and 24.36 cUS$/kWh for CSP. The
competitive costs obtained for the two technologies are explained
by the high solar radiation available in the Atacama Desert and the
low capital cost ($/W) of the technology calculated for the current
location. From 2014 to 2050, further reductions in capital costs are
expected, which will allow solar technology to reach grid parity
with than conventional fossil fuel technologies [9].
In addition, Figs. 7 and 8 show the LCOE for the studied 50 MW
hybrid PV-CSP plant according to the two mentioned scenarios. For
the scenario (1), LCOE values obtained varied between 14.69 and
8.57 cUS$/kWh between 2014 and 2050 respectively. For the sce-
nario (2) LCOE values obtained varied between 13.88 and 7.74 cUS$/
kWh. In Fig. 7, according to the Blue Map scenario, the three plants
LCOE have tendency to decrease over time. The CSP plant will reach
the lowest LCOE value compared to the other two, followed closely
by the hybrid PV-CSP plant. According to the Roadmap scenario in
Fig. 8, the three plants have the same tendency, their LCOE de-
creases over time, but it is the PV plant which will reach the lowest
LCOE value in 2050 followed by the hybrid PV-CSP plant. According
to these calculations, in the Blue Map scenario, it is predicted that
CSP market will increase faster and stronger than PV market, due to
a greater growth of CSP global installed capacity. The Roadmap
scenario, which is based on different projection criteria, is pre-
dicting a greater growth of global PV installed capacity in the
future, given that PV is a more mature technology than CSP, that it
has lower cost and thus it is more accessible [59].
These two different situations are explained by the value of
learning rate LR included in the model. The LR studies on learning
rate factors are very useful for modeling technical change and
informing policy decisions related to energy technology. Learning
rate models have a variety of forms to describe the relationship
between cumulative capacity and cost [57]. The learning rate of a
technology is derived from the accumulation of experiences in
production. Its value gives an indication on how fast the LCOE of the
technology can be expected to decrease over the years [57]. In the
Blue Map scenario, according to the LR learning rate of CSP tech-
nology, it is expected that CSP will be further developed in the
coming years, the global installed capacity is expected to signifi-
cantly increase, thus reaching a lower LCOE than the other studied
plants. In the Roadmap scenario, however, it is expected a higher
growth in PV installed capacity in the future, thus leading to a lower
LCOE for the PV technology compared to the other two types.
Although the PV and CSP plants separately reach a lower LCOE
than the hybrid combination, either in the Blue Map and the
Roadmap scenarios, the hybrid PV-CSP plant enable 24-h electricity
generation and reach higher capacity factors than the other two.
Therefore, this type of hybrid PV-CSP plants can effectively provide
power supply and match the 24-h 7-day power demand of the
mining industry in Chile. The mining industry is also the first sector
of CO2 emissions in Chile, so it is important to search clean and
reliable options to supply its electricity demand. Hybrid PV-CSP
plants are an option to ensure 24 h continuous supply of elec-
tricity and reduce the country's dependence on fossil fuels. The
similar LCOE for future years between Hybrid, PV and CSP can be an
important factor to determinate new options to satisfy the mining
sector power demand, allowing hybrid PV-CSP plants to be the
most economically and environmentally viable option even if its
LCOE is not the lowest. Also, the mitigation of CO2 emissions from
mining industries resulting from the use of solar resource instead of
fossil fuel electricity would lead to non-negligible reductions in
pollution taxes, and directly affecting business cash flows.
 C. Parrado et al. / Energy 94 (2016) 422e430 429

Table 2
Values for the projection of LCOE for PV and CSP.

Factor Symbol Units PV CSP

Cost of the system in 2014 C(0) US$/W 1.24 6.12

Cumulative installed capacity in 2014 q(0) GW 191 3.935
Cumulative installed capacity in year t q(t) GW Fig. 6 Fig. 6
Learning rate LR % 18 [35] 10 [35]
Land cost L US$/kW 20 [64] 20 [64]
Discount rate d % 10 [35] 10 [35]
O&M costs O&M % 1.5 [35] 2 [35]
Annual insurance rate I % 0.25 [35] 0.5 [35]
Solar resource S kWh/m2/year 2576 3421.9
Tracking factor TF Dimensionless 1 [35] 0.9711 [35]
Performance factor h m2/kW 0.533 1.377
Lifetime of the system T Years 25 [35] 30 [35]
Annual output degradation rate DR % 0.6 [35] 0.2 [35]

promising option for sustainable power generation. It represents a

reliable and clean solution to the concern of power supply in Chile,
particularly because it could satisfy the continuous power demand
of the large local industries that mostly rely on fossil fuel resources.
Additionally, it is important to note that the solar resource in the
Atacama Desert is greater than that of other sunny areas where
huge solar energy projects have been developed, such as the two
leading countries in solar investment, the US and Spain. This means
that there is an untapped natural resource that can also be con-
verted into a profitable endeavor for public and private companies.
Foreign energy investors should pay special attention to this
geographic area, given that the unused solar resource could be the
future source of power for the mining industry in Chile, which is in
the same time seeking solutions for reducing its environmental
impact.

Fig. 7. 2050 LCOE projection to PSDA between 2014 and 2050 for PV, CSP and PV-CSP
Acknowledgments
in Blue Map scenario (1).

The authors would like to acknowledge the financial support

provided by CONICYT/FONDAP 15110019 “Solar Energy Research
Center” SERC-Chile, Fondecyt Postdoctoral grant n
3140014, FIC-R
30137092 funded by Atacama Government and the Education
Ministry of Chile Grant PMI ANT 1201.

References

Fig. 8. 2050 LCOE projection to PSDA between 2014 and 2050 for PV, CSP and PV-CSP
in Roadmap scenario (2).
7. Conclusion
This paper has important ramifications for the energy industry
in Chile, especially for mining companies located in the studied
area. Chile has a rich solar resource, which exploitation is still in its
infancy. The main concern regarding solar energy is the disconti-
nuity, intermittency of electricity production. However, the hybrid
PV-CSP mix modeled in the paper is a sound solution to this
problem. The hybrid PV-CSP mix in the Atacama Desert is a
[1] Tokman RM, Huepe C. Política Energe
tica, Nuevos Lineamientos: Trans-
formando la Crisis Energe
tica en una oportunidad. Santiago: Comisio
n
Nacional de Energía (CNE); 2008.
[2] Hanel M, Escobar R. Influence of solar energy resource assessment uncertainty
in the levelized electricity cost of concentrated solar power plants in Chile.
Renew Energy 2013;49:96e100.
[3] Ministry of Energy, 17 10 2014. [Online]. Available: http://www.minenergia.
cl/.
[4] National Energy Commission (CNE). Proyeccio
n de la evolucio

n de las emi-
siones de gases de efecto invernadero en el sector energía an ~ os 2000e2025.
Santiago: CNE; 2008.
[5] Ministry of Energy. Energy 2050. Antofagasta: Ministry of Energy; 2014.
[6] International Renewable Energy Agency (IRENA). Renewable energy tech-
nologies: costs analysis series. Conc Sol Power June 2012;1(2).
[7] World CSP. CSP 2012 a review. January 2013 [Online]. Available: www.csp-
world.com.
[8] REN21. Renewables 2014: global status report. Paris: REN21; 2014.
[9] International Energy Agency (IEA). World energy outlook 2013: renewable
energy outlook. Paris: OECD/IEA; 2013.
[10] Green A, Diep C, Dunn R, Dent J. High capacity factor CSP-PV hybrid systems.
Energy Procedia 2014;69:2049e59.
[11] Larchet Kevin. Solar PV-CSP hybridisation for baseload generation, a techno-
economic analysis for the Chilean market. Master of Science Thesis. Stock-
holm: KTH School of Industrial Engineering and Management; 2015.
[12] National Energy Commission (CNE). Capacidad instalada SING. CNE; 2014.
[13] European Commission. External costs: research results on socio-
environmental damages due to electricity and transport. Brussels: Direc-
torate-General for Research, Directorate J-Energy; 2003.
430 C. Parrado et al. / Energy 94 (2016) 422e430
[14] European Commission. ExternEdexternalities of energy: methodology 2005
update. Brussels: Directorate-General for Research, Sustainable Energy Sys-
tems; 2005.
[15] Systep. Catastro de Proyectos ERNC. 2014.
[16] Corral N, Anrique N, Fernandes D, Parrado C, C
aceres G. Power, placement and
LEC evaluation to install CSP plants in northern Chile. Renew Sustain Energy
Rev 2012;16:6678e85.
[17] Escobar RA, Corte
s C, Pino A, Pereira EB, Martins FR, Cardemil JM. Solar energy
resource assessment in Chile: satellite estimation and ground station mea-
surements. Renew Energy 2014;71:324e32.
[18] Renewable Energy Center (CER). CER report November 2013, NCRE project
status in Chile. CER; 2013.
[19] Economic Center of Charge Service (CDEC). Statistics of CDEC. CDEC-SIC Ltda.
Available at: http://www.cdec-sic.cl/.
[20] Gallego B. CSP and mining in Chile. Renewable EnergyWorld.com; 16 04 2013
[Online]. Available: http://www.renewableenergyworld.com/rea/companies/
first-conferences/news/article/2013/04/csp-and-mining-in-chile [accessed
25.02.14].
[21] Pekic V. Chile prepares to add 2.2 GW of solar power generation. PV Maga-
zine; 21 01 2013 [Online]. Available: http://www.pv-magazine.com/news/
details/beitrag/chile-prepares-to-add-22-gw-of-solar-power-generation_
100009903/ [accessed 11.02.14].
[22] Miller A. Chile proving to be a hot market for solar. Renew-
ableEnergyWorld.com; 06 09 2013 [Online]. Available: http://www.
renewableenergyworld.com/rea/blog/post/2013/09/chile-proving-to-be-a-
hot-market-for-solar [accessed 18.02.14].
[23] Montgomery J. Latin America report: Chile's road to solar grid parity.
RenewableEnergyWorld.com; 11 09 2013 [Online]. Available: http://www.
renewableenergyworld.com/rea/news/article/2013/09/latin-america-report-
chiles-road-to-solar-grid-parity [accessed 18.02.14].
[24] Gonzalez B. Profiting from PV in Chile. PV Insider; 19 03 2013 [Online].
Available: http://news.pv-insider.com/photovoltaics/profiting-pv-chile
[accessed 12.02.14].
[25] Gonzalez B. PV plants under construction in Chile grow 184% from last year.
PV Insider; 3 02 2014 [Online]. Available: http://news.pv-insider.com/
photovoltaics/pv-plants-under-construction-chile-grow-184-last-year
[accessed 20.02.14].
[26] Scanlon B. Thermal storage gets more solar on the grid. Renew-
ableEnergyWorld.com; 24 02 2012 [Online]. Available: http://www.
renewableenergyworld.com/rea/news/article/2012/02/thermal-storage-gets-
more-solar-on-the-grid [accessed 25.02.14].
[27] World CSP. Termosolar Pedro de Valdivia receives environmental approval
from Chilean authorities. CSP-World.com; 24 08 2012 [Online]. Available:
http://www.csp-world.com/news/20120824/00427/termosolar-pedro-de-
valdivia-receives-environmental-approval-chilean-authorities [accessed
25.02.14].
[28] World CSP. Chile's attractiveness for CSP companies keeps growing. CSP-
World.com; 18 04 2013 [Online]. Available: http://www.csp-world.com/
news/20130418/00830/chile-s-attractiveness-csp-companies-keeps-growing
[accessed 25.02.14].
[29] World CSP. Chile, an emerging market for concentrated solar power, receives
$200 million support from World Bank. CSP-World.com; 17 05 2012 [Online].
Available: http://www.csp-world.com/news/20120517/00225/chile-
emerging-market-concentrated-solar-power-receives-200-million-support
[accessed 25.02.14].
[30] World CSP. Chile receives $67 M from climate investments funds to develop
CSP projects. CSP-World.com; 25 09 2012 [Online]. Available: http://www.
csp-world.com/news/20120925/00491/chile-receives-67-m-climate-in-
vestments-funds-develop-csp-projects [accessed 25.02.14].
[31] Burger A. South America's largest solar thermal power plant to be built in
Chile. TriplePundit.com; 13 01 2014 [Online]. Available: http://www.
triplepundit.com/2014/01/latin-americas-largest-solar-thermal-power-plant-
built-chile/ [accessed 25.02.14].
[32] World CSP. Abengoa wins first CSP tender in Chile, will build a 110 MW tower
plant to generate around the clock. CSP-World.com; 09 01 2014 [Online].
Available: http://www.csp-world.com/news/20140109/001295/abengoa-
wins-first-csp-tender-chile-will-build-110-mw-tower-plant-generate
[accessed 25.02.14].
[33] Parkinson G. Abengoa to build 110MW solar tower storage plant in Chile.
Renew Economy; 13 01 2014 [Online]. Available: http://reneweconomy.com.
au/2014/abengoa-to-build-110mw-solar-tower-storage-plant-in-chile-24839
[accessed 25.02.14].
[34] SolarGIS. GHI Solar map e Chile. © 2010e2015 GeoModel Solar.
[35] Herna
ndez-Moro J, Martínez-Duart JM. Analytical model for solar PV and CSP
electricity costs: present LCOE values and their future evolution. Renew
Sustain Energy Rev 2013;20:119e32.
[36] International Energy Agency (IEA). Concentrating solar power technology
roadmap. Paris, France: IEA/OECD; 2010. p. 1e52.
[37] Viebahn P, Kronshage S, Trieb F, Lechon Y. D 12.2 Final report on technical
data, costs, and life cycle inventories of solar thermal power plants. NEEDS
(New Energy Externalities Developments for Sustainability) Project no:
502687. 2008.
[38] Purohit I, Purohit P, Shekhar S. Evaluating the potential of concentrating solar
power generation in Northwestern India. Energy Policy 2013;20:157e75.
[39] Tsikalakis A, Tomtsi T, Hatziargyriou ND, Poullikkas A, Malamatenios Ch,
Giakoumelos E, et al. Review of best practices of solar electricity resources
applications in selected Middle East and North Africa (MENA) countries.
Renew Sustain Energy Rev 2011;15(6):2838e49.

[40] Súri M, Remund J, Cebecauer T, Hoyer-Klick C, Dumortier D, Huld T, et al.
Comparison of direct normal irradiation maps for Europe. In: SolarPACES
Conference “Concentrated Solar Power and Chemical Energy Systems; 2009.
Berlin, Germany.

[41] Súri M, Huld TA, Dunlop ED, Ossenbrink HA. Potential of solar electricity
generation in the European Union member states and candidate countries. Sol
Energy 2007;81:1295e305.
[42] International Energy Agency (IEA). Technology roadmap - solar photovoltaic
energy. Paris, France: IEA/OECD; 2010. p. 1e48.
[43] European Commission: Joint Research Center. Photovoltaic Geographical In-
formation System (PVGIS). Geographical Assessment of Solar Resource and
Performance of Photovoltaic Technology; 2012 [Online]. Available: http://re.
jrc.ec.europa.eu/pvgis/.
[44] National Renewable Energy Laboratory (NREL). Solar and Wind Energy
Resource Assessment (SWERA). NREL; 2015 [Online]. Available: https://maps.
nrel.gov/swera/#/ [accessed 22.09.15].
[45] Koohi-Kamali S, Tyagi VV, Rahim NA, Panwar NL, Mokhlis H. Emergence of
energy storage technologies as the solution for reliable operation of smart
power systems: a review. Renew Sustain Energy Rev 2015;25:135e65.
[46] Akhil AA, Huff G, Currier AB, Kaun BC, Rastler DM, Chen SB, et al. DOE/
EPRI2013 Electricity storage handbook in collaboration with NRECA. US
Department of Energy and Electric Power Research Institute (EPRI), Albu-
querque (New Mexico) and Livermore (California). 2013.
[47] Chen HS, Cong TN, Yang W, Tan CQ, Li YL, Ding YL. Progress in electrical en-
ergy storage system: a critical review. Prog Nat Sci 2009;19:291e312.
[48] Battke B, Schmidt TS, Grosspietsch D, Hoffmann VH. A review and probabi-
listic model of lifecycle costs of stationary batteries in multiple applications.
Renew Sustain Energy Rev 2013;25:240e50.
[49] Fathabadi H. Two novel techniques for increasing energy efficiency of
photovoltaic-battery systems. Energy Convers Manag 2015;105:149e66.
[50] Glavin ME, Hurley WG. Optimisation of a photovoltaic battery ultracapacitor
hybrid energy storage system. Sol Energy 2012;86:3009e20.
[51] Barnhart CJ, Benson SM. On the importance of reducing the energetic and
material demands of electrical energy storage. Energy Environ Sci 2013;6:
1083e92.
[52] Mulder G, Six D, Claessens B, Broes T, Omar N, Van Mierlo J. The dimensioning
of PV-battery systems depending on the incentive and selling price condi-
tions. Appl Energy 2013;111:1126e35.
[53] Cervone A, Carbone G, Santini E, Teodori S. Optimization of the battery size for
PV systems under regulatory rules using a Markov-chain's approach. Renew
Energy 2016;85:657e65.
[54] Poullikkas A. A comparative overview of large-scale battery systems for
electricity storage. Renew Sustain Energy Rev 2013;27:778e88.
[55] Morisson V, Rady M, Palomo E, Arquis E. Thermal energy storage systems for
electricity production using solar energy direct steam generation technology.
Chem Eng Process 2008;47:499e507.
[56] Herna
ndez-Moro J, Martínez-Duart JM. CSP electricity cost evolution and grid
parities based on the IEA roadmaps. Energy Policy 2012;41:184e92.
[57] Nemet GF. Beyond the learning curve: factors influencing cost reductions in
photovoltaics. Energy Policy 2006;34:3218e32.
[58] International Energy Agency (IEA). Energy technology perspectives 2010:
scenarios and strategies to 2050. Paris, France: IEA/OECD; 2010. p. 1e650.
[59] International Energy Agency (IEA). Technology roadmap e solar photovoltaic
energy. Paris: IEA/OECD; 2014.
[60] National Renewable Energy Laboratory (NREL). System advisor model (SAM).
NREL; 2013.
[61] Kolb GJ, Ho CK, Mancini TR, Gary JA. Power Tower technology roadmap and
cost reduction plan. SAND2011-2419. Albuquerque, USA: Sandia National
Laboratories; 2011.
[62] Turchi CS. Parabolic trough reference plant for cost modeling with the solar
advisor model (SAM). NREL/TP-550e47605. 2010.
[63] Kost C, Mayer JN, Thomsen J, Hartmann N, Senkpiel C, Philipps S, et al. Lev-
elized cost of electricity renewable energy technologies. Freiburg, Germany:
Fraunhofer Institute for Solar Energy Systems ISE; 2013.
[64] Ministry of National Goods (MBN). Cuentas Públicas Nacionales. MBN; 2014.
[65] Dinter F, Gonzalez DM. Operability, reliability and economic benefits of CSP
with thermal energy storage: first year of operation of ANDASOL 3. Energy
Procedia 2014;49:2472e81.