Energy for Sustainable Development 42 (2018) 121–128

Contents lists available at ScienceDirect

Energy for Sustainable Development

Cost reduction potential of parabolic trough based concentrating solar

power plants in India
Chandan Sharma a,⁎, Ashish K. Sharma b, Subhash C. Mullick b, Tara C. Kandpal b
a
Government Engineering College Ajmer, N.H. 8, Barliya Circle, Ajmer 305025, India
b
Centre for Energy Studies, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India

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

Article history: In this paper, we estimate the cost reduction potential of parabolic trough based concentrated solar Power
Received 21 October 2016 systems in India and consequently their implications for levelized cost of electricity. Using the past as well as
Revised 11 July 2017 envisaged global cumulative diffusion of CSP systems and the learning rates reported in the literature, the
Accepted 10 October 2017
expected capital cost of parabolic trough based CSP systems in India has been estimated. Present capital cost of
Available online xxxx
parabolic trough based CSP plant in India has been taken as US $2540/kW. Local manufacturing of CSP compo-
Keywords:
nents and consequently reduced logistics is expected to reduce the capital cost by 14% and 8% respectively. A
Concentrating solar power learning rate of 10% for the global cumulative diffusion in the base case scenario is expected to reduce capital
Cost reduction in CSP technology cost to 49% of the present cost by the end of 2050. Present LCOE of US $151/MWh of CSP plant in India is expected
Levelized cost of electricity to reduce to US $76/MWh by the end of 2050. Provision of 6 hour thermal storage is expected to reduce LCOE of
Learning rates CSP plant in India by 18% as compared to LCOE of CSP plant without storage. Analysis for different learning rates
(5%, 10% and 15%) for CSP plant and different discount rates (6%, 8% and 10%) has also been undertaken and
results obtained are presented.
© 2017 International Energy Initiative. Published by Elsevier Inc. All rights reserved.

Introduction
Concentrated solar power (CSP) is an option for generation of
electricity using solar energy. Worldwide, around 4815 MW capacity
plants are operational up to December 2016 (SolarPACES, 2017) signify-
ing the importance of CSP technology for providing electricity in a sus-
tainable manner without adversely affecting the environment. Four
different solar concentrator technologies, namely the parabolic trough,
central tower receiver, linear Fresnel reflector and parabolic dish have
been developed for power generation. A schematic of parabolic trough
based CSP plant is presented in Fig. 1. A solar collector field redirects
the beam component of solar radiation incident on its aperture on to a
receiver that absorbs the concentrated flux falling on its surface. The in-
cident photons thus converted into heat that is extracted and trans-
ferred to a heat transfer fluid. The hot fluid then exchanges heat in a
steam generator where water is converted into steam that is expanded
in a turbine and power is generated.
The levelized cost of electricity (LCOE) of CSP plants is relatively
higher than several of the competing renewable energy based electricity
generation options such as solar photovoltaic power. This is primarily
due to the fact that there has been considerable advancements in
photovoltaic technology leading to a drastic reduction in cost per peak
⁎ Corresponding author.
E-mail address: sharmac1975@ecajmer.ac.in (C. Sharma).
Watt of solar cells. As a consequence, the LCOE of PV plants has reduced
considerably in last 3–4 years, thus making the solar photovoltaics as
the preferred option for solar power generation.
Despite the fact that presently capital cost of CSP systems is
considerably higher than the capital cost of solar photovoltaic systems,
the interest in CSP technology is growing for several reasons that
include (i) possibility of including relatively inexpensive thermal
storage component in CSP systems so as to facilitate better dispatch
ability, (ii) relatively higher conversion efficiency (solar to electric) as
compared to that normally achievable with existing commercial
photovoltaic plants and (iii) relatively higher capacity utilization factor
(23% to 28%) of CSP systems (Purohit et al., 2013) as compared to
photovoltaic systems (12% to 18%). CSP systems are generally tracked
whereas PV systems are rarely tracked, resulting in higher capacity
utilization factor.
Several studies (ESMAP, 2012; IRENA, 2012; Sundaray and Kandpal,
2013; IEA, 2014; Sharma et al., 2017) have reported that the cost of
CSP technology can be reduced in the future. The measures that can
contribute towards such a cost reduction of CSP technology include,
technological advancements in the components (concentrators,
receiver tubes, heat transfer fluid, etc.) leading to efficiency improve-
ments, local manufacturing of CSP components and by experience
and/or enhanced learning about the CSP plants in operation.
In view of the above and the fact that the present capital cost of a CSP
system is significantly higher than the capital cost of photovoltaic based
systems, it would be useful to study different cost reduction possibilities

https://doi.org/10.1016/j.esd.2017.10.003
0973-0826/© 2017 International Energy Initiative. Published by Elsevier Inc. All rights reserved.
122
Fig. 1. Schematic of parabolic trough based CSP plant.
of CSP systems so as to make their LCOE competitive with that of photo-
voltaic systems.
A preliminary attempt has been made in this paper to review the
cost reduction possibilities of parabolic trough based CSP systems in
India and their implications for levelized cost of electricity. Parabolic
trough based CSP plants based on simple Rankine cycle and without
the use of any supplementary fossil fuel (such as natural gas) have
been considered for the analysis. Since the modelling of cost reduction
possibilities is often related to the cumulative diffusion of the technology
(Neij, 2008; Moro and Duart, 2012; Huenteler et al., 2014), a brief review
of the status of CSP technology and the time trend of the installed
capacity has been presented. The current status of CSP technology in
India has also been presented. A summary of different possibilities of
cost reduction in parabolic trough based CSP plants (with primary
focus on India) is then presented. Finally, the expected cost of parabolic
trough based CSP plants in the future has been estimated and the same
has been used to estimate the likely values of the levelized cost of
electricity delivered by parabolic trough based CSP plants in India.
Global status of CSP plants
The majority of the plants installed worldwide are in the United
States of America (USA) and Spain with a small capacity installed in
China, Australia, Algeria, South Africa and India. Despite the fact that
plants installed in mid 80s in the USA were successful as CSP technology,
no new CSP plants were commissioned until 2006. Significant capacity
addition (Fig. 2) has however been made after 2006 (SolarPACES,
2017). By the end of December 2016, CSP plants of cumulative capacity
of 4815 MW were operational across the world, 1260 MW capacity
5.5
Cumulative installed capacity (GW)
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Fig. 2. Cumulative installed capacity of CSP plants worldwide (SolarPACES, 2017).
C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128
plants were under construction and 2709 MW capacity plants were in
the development stage (SolarPACES, 2017).
Details of few operational CSP plants in United States and Spain are
presented in Table 1 (SolarPACES, 2017). There is significant variation
in the reported capital cost of the CSP plants. The capital cost is varying
from US $3540/kW to US $7060/kW.
The average LCOE of CSP based systems by region varies from a low
of US $0.20/kWh in Asia to a high of US $0.25/kWh in Europe with the
LCOE of individual projects varying significantly depending on location
and solar resource (IRENA, 2012). However, as costs are falling, recent
projects are being built with LCOE of US $0.17/kWh and power purchase
agreements are being signed at even lower values where low cost
financing is available.
Average capital cost projections for all types of CSP plants without
storage and with 6 h of storage are presented in Fig. 3 (IEA, 2014). As
can be seen from Fig. 3, projected capital cost of CSP plants is expected
to be US $2300/kW (without storage) to US $3100/kW (with provision
for 6 h of thermal storage) by the year 2050. As a consequence of
reduction in capital cost and also of improved performance, a significant
lowering in the LCOE (Fig. 4) is expected.
Status of CSP plants in India
With the launch of Jawaharlal Nehru National Solar Mission
(JNNSM) in 2010, activities towards large scale dissemination of CSP
plants in India has gathered considerable momentum. Several research
groups have undertaken research and development activities on CSP
plants in recent years (CSTEP, 2012; Reddy et al., 2012; Desai et al.,
2014).
A brief summary of operational CSP plants and plants under
construction is presented in Table 2 (SolarPACES, 2017). As shown in
Table 2, CSP plants of cumulative capacity 225 MW are operational
and 292 MW capacity plants are under construction. Comparing
reported capital cost of operational CSP plants in United States and
Spain (Table 1) to the reported capital cost of few operational CSP plants
in India (US $2520/kW to US $2540/kW) in Table 2, It is found that there
is significant variation in the reported capital cost. To a large extent, the
difference in the reported values of the capital cost of CSP plants can be
attributed to the following:
• Relatively longer project preparation cycle time and higher costs in
obtaining permits and clearances
• Higher cost of labor
• Relatively more stringent environmental regulations and
requirements
Though the cost of a CSP plant in India is lower than the cost of a CSP
plant in United States and Spain, it is still higher than several renewable
energy based electricity supply options (Table 3) such as wind power
and solar PV power in India (WEC, 2013).
Results of a preliminary attempt to review the potential avenues for
cost reduction and to assess the effect of any such cost reduction on
LCOE are presented in the following sections.
Drivers of cost reduction of CSP plants in India
A CSP power plant consists of several important hardware
components that critically affect the performance and cost of the
plant. For ensuring market competitiveness of a CSP plant, each one of
the components must satisfy certain basic requirements. Table 4
summarizes some of these requirements. Cost of CSP technology is
expected to reduce through technological breakthroughs/innovations
and local manufacturing of components used in CSP plants (ESMAP,
2012; IRENA, 2012; IEA, 2014).
Significant cost reductions are expected to be achieved through the
use of alternative mirror materials, using mirrors of high reflectivity,
 C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128 123

Table 1
Details of few operational CSP plants worldwide (SolarPACES, 2017).

Project Location Technology Capacity (MW) Storage (hours) Electricity output (GWh/year) Approx. cost (US $/kW)

Mojave solar project California, USA Parabolic trough 280 None 600 5710
Ivanpah SEGS California, USA Central tower receiver 392 None 1079 5610
Solana GS Arizona, USA Parabolic trough 280 6 944 7142
Martin NGSEC Florida, USA Parabolic trough 75 None 155 6340
Andasol - 3 Granada, Spain Parabolic trough 50 7.5 175 7060
Borges Thermosolar Spain Parabolic trough 25 None 98 6840
Termesol 50 Spain Parabolic trough 49.9 7.5 175 6050

new support structure design, improved receiver characteristics, alter-
native heat transfer fluids, and increased operating temperature of
cycle for improved thermal efficiency and storage system developments
(ESTELA, 2010). A summary of some such possibilities is presented in
Table 5.
Currently, the majority of the components used in CSP plants in India
are being imported. These include – mirrors, receivers and absorbers,
heat transfer fluid (HTF), power block etc. This is one of the important
reasons of high capital cost of solar thermal power plants in India as
compared to other renewable energy based options. Looking at the
large scale potential of solar thermal power in India, several companies
actively involved globally in CSP industry have ventured with Indian
companies to set up their manufacturing base in India. This will
significantly reduce the cost of CSP components due to reduced cost in
logistics and relatively lower labor costs in India. Table 6 presents the
possibility of manufacturing of different CSP components in India with
strengths, limitations and possible solutions. Table 7 presents expected
cost reduction in CSP components owing to local manufacturing and
reduced logistics (ESMAP, 2012).
For the purpose of analysis, the cost of a 50 MW CSP plant based on
parabolic trough technology (without storage) is assumed to be
consisting of three components: (a) Solar energy collection system,
(b) thermal energy transfer system and (c) power generation system.
The solar energy collection system is assumed to be comprising of
mirrors, support structures, drive mechanism and site preparation.
The cost of solar energy collection system is assumed as 37% of the
total cost. The thermal energy transfer system includes heat transfer
fluid, receiver tube, ball joints, piping, valves etc. and its cost is assumed
as 25% of the total cost. Power generation system takes into account
power block, balance of plant (BOP) and related work. The cost of
power generation system is assumed as 38% of the total cost (ESMAP,
2012). For CSP systems with thermal storage, an additional storage
system has been considered that comprises of storage fluid and storage
tanks. For a parabolic trough based CSP system with 6 h of storage, the
relative contribution of various subsystems in the overall cost is solar
energy collection system 40%, thermal energy transfer system 20%,
power generation system 25% and storage system 15% (ESMAP, 2012).
Table 7 presents expected cost reduction in CSP components owing
to local manufacturing and reduced logistics (ESMAP, 2012). Consider-
ing various sub-systems of parabolic trough based CSP plant (without
storage) discussed above and data presented in Table 7, it is expected
that due to local manufacturing, within the next two decades, there
shall be a reduction of 13% each in the cost of solar energy collection
system and thermal energy transfer system and a reduction of 15% in
the cost of power generation system. In view of the cost share of various
sub-systems in the overall cost, the overall reduction due to local
manufacturing in the cost of parabolic trough based CSP plant without
storage is expected to be 14% in next two decades.
Similarly, reduced logistics shall also reduce the cost of parabolic
trough based CSP plant. As presented in Table 7, reduced logistics is
expected to decrease cost of solar energy collection system by 13%,
thermal energy transfer system by 5% and power generation system
by 4%. Overall reduction in the cost of parabolic trough based CSP
plant due to reduced logistics is expected to be 8%.
Cost assessment using learning rate approach
The approach of using learning rate to mathematically model the
likely impact of cumulative diffusion of renewable energy technology
on the unit capital cost is widely used (Price and Carpenter, 1999;
McDonald and Schrattenholzer, 2001; Trieb, 2005; Broek et al., 2009;
Ferioli et al., 2009). The expected future cost C (t) of a particular compo-
nent in achieving a cumulative diffusion Q (t) can be estimated from the
base cost C (0), corresponding to a base level diffusion Q (0), using the
following expression (Neij, 2008):
 
CðtÞ Q ðtÞ b
¼ ð1Þ
Cð0Þ Q ð0Þ
where b is the learning parameter for each component. The ratio of
C (t) to C (0) is defined as the Progress Ratio (PR) – a measure of cost
(as a fraction of base cost) on each doubling of cumulative diffusion.
i:e: Progress Ratio ðPRÞ ¼ 2b ð2Þ

Without storage With 6 hour storage

6500 180
Capital cost (US$/kW)

160
LCOE (US$ / MWh)

5500
140
4500
120
3500
100
2500
80
1500 60
2015 2020 2030 2040 2050 2015 2020 2025 2030 2035 2040 2045 2050
Year (t) Year

Fig. 3. Projected capital cost of CSP plants with and without storage (IEA, 2014). Fig. 4. LCOE projections for CSP plants worldwide (IEA, 2014).
124 C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128

Table 2
Status of solar thermal power in India (SolarPACES, 2017).

Project Location Technology Capacity (MW) Status Approx. cost (USD/kW) Expected electricity output (GWh/year)

Godawari Green Energy Nokh, Rajasthan Parabolic trough 50 Operational – 118

Megha Engineering Anantpur, AP Parabolic trough 50 Operational 2540 110
Dhursar Pokhran, Rajasthan Compact LFR 125 Operational 2520 280
Gujarat Solar One Kachch, Gujarat Parabolic trough 28 Under construction – 130
KVK Solar Energy Askandra, Rajasthan Parabolic trough 100 Under construction – –
Diwakar Solar Askandra, Rajasthan Parabolic trough 100 Under construction – –
Abhijeet Solar Phalodi, Rajasthan Parabolic trough 50 Under construction – –
Dadri ISCC Plant Dadri, Uttar Pradesh Compact LFR 14 Under construction – 14

Learning rate (LR) is a measure of the decrease in the cost as a frac- Neij (2008). As presented in Table 10, a reduction of approximately
tion of the base cost with each doubling of cumulative diffusion. 49% in the cost of CSP plants in India is expected by the end of 2050.

Cð0Þ−CðtÞ CðtÞ Envisaged capacities of CSP plants in India

Learning Rate ¼ ¼ 1− ¼ 1−PR ¼ 1−2b ð3Þ
Cð0Þ Cð0Þ
Although CSP plants of total installed capacity of 517 MW were
To assess the future capital cost, expected cumulative diffusion given financial closure in the batch I of phase I of JNNSM, plants of
and learning rates are required. With the use of Eq. (3), for the given cumulative capacity of 225 MW have become operational until the
learning rate, learning parameter b is obtained and using Eq. (1), future end of 2015 (Table 2). Anticipating that expected cost reduction will
capital cost is estimated for the given cumulative diffusion and calculated facilitate more and more CSP plants to come up in the near future and
value of learning parameter. considering the JNNSM targets, future cumulative installed capacity
considered in the study is presented in Table 11. The numbers presented
Learning effects of different CSP components on future capital cost due to for cumulative diffusion in Table 11 are as obtained from widely used
expected cumulative diffusion S-diffusion curve.

To assess the future capital cost of CSP plants worldwide, Viebahn Estimation of the levelized cost of electricity delivered by CSP plants in India
et al. (2011) proposed three possible scenarios of cumulative diffusion:
Very optimistic, Optimistic-realistic and Pessimistic (Table 8). In the The levelized cost of electricity (LCOE) has been estimated from the
present study, optimistic-realistic scenario has been considered for the following expression (Kandpal and Garg, 2003):
analysis and referred to as base case scenario. Further, a learning rate
of 12% each was proposed for solar energy collection system and ther- ðCapital cost  capital recovery factorÞ þ annual O & M cost
LCOE ¼
mal energy transfer system respectively and 5% for power generation Annual electricity output
system (Viebahn et al., 2011). An illustration of effect of learning rate ð4Þ
of CSP sub-systems on their future cost for the base case scenario is
presented in Table 9. The capital cost for the parabolic trough based Capital recovery factor (CRF) is given by:
CSP plant (without storage) in India has been taken as US $2540/kW
n
(Table 2) i.e. Indian National Rupee (INR) 169,240/kW (1 US $ = INR dð1 þ dÞ
CRF ¼ n ð5Þ
66.63 as on February 1, 2016). ð1 þ dÞ −1
A significant reduction in the capital cost of various sub-systems of
parabolic trough based CSP plants (without storage) is expected up to where d is discount rate and n is useful life of the plant.
2050 (solar energy collection system 56%; thermal energy transfer sys- Annual electricity output (AEO) is estimated by:
tem 56%; power generation system 28%) for the cumulative diffusion
and learning rates proposed by Viebahn et al. (2011). AEO ðMWhÞ ¼ Installed capacity ðMWÞ  8760 hours
While Viebahn et al. (2011) proposed different learning rates for  Capacity utilization factor ð6Þ
various CSP sub-systems, Neij (2008) proposed a moderate learning
rate of 10% for the CSP plant as a whole with uncertainty of ±5%. The
reason for an overall learning rate of 10% for CSP plants might be limited
experience of CSP plants worldwide. For the rest of the calculations in Table 4
the present study, learning rate of 10% for CSP plants has been assumed. Requirements on the components of CSP plants (ESMAP, 2012; IRENA, 2012).

Table 10 presents estimates of future capital cost of CSP plants in Component Specific requirements
India using Eqs. (1)–(3) due to expected cumulative diffusion in the Mirrors Different from traditional mirrors in reflectivity, durability
base case scenario and an overall learning rate of 10% as proposed by and strength and have a low iron content
Support structures Since exposed to open air, specific grades of steel or aluminium
are required
Receiver tubes Selective surfaces of receiver tubes must be highly absorbent
Table 3 and must have a low thermal emissivity at operating
Comparison of capital cost and LCOE for various renewable energy supply based options in temperature
India (WEC, 2013). Tracking and drive Should ensure optimal positioning of reflectors during the
mechanism day to track Sun's position with a high degree of accuracy
Renewable energy technology Capital cost (US $/kW) LCOE (US $/MWh)
HTF (synthetic oil) Solar field outlet temperature is restricted by HTF properties
Onshore wind 1080–1250 47–113 that means fluid that can perform functions are also limited
Small hydro 1400–3680 19–314 Turbine Must endure frequent load variations and numerous startups
Large hydro 1590–4150 24–302 and shutdown procedures. Besides, these usually have the
Biomass (incineration) 830–1200 65–86 relatively small capacity (50–100 MW)
Solar PV (C-Si, without tracking) 1530–1810 87–137 HTF pumps Design should permit wide fluctuations in operating
Solar thermal (parabolic trough) 3080–4550 123–248 conditions expected in CSP plants
 C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128 125

Table 5
Brief description of technological improvements in CSP components and their implications (ESMAP, 2012; IRENA, 2012).

Technology Parameter/component Improvement/innovation Implication

PTC Support structure • New designs with materials like stamped steel or aluminium • Lighter construction
• Larger trough dimensions • Total number of rows decreases for same solar field leading to
• Improved resistance to external forces saving in capital cost
• Precise Sun concentration, plant efficiency increases
Mirrors • Alternative materials such as thin glass mirrors, aluminized • Higher reflectivity resulting in increased plant efficiency
reflector, polymer reflector
Receiver tube • Special selective coatings having high absorptance and low infra- • Increased operating temperature of heat transfer fluid leading
red emittance to improvements in power cycle efficiency
Heat transfer fluid • Alternative fluids such as molten salts, direct steam generation, • Higher operating temperature leading to increased plant
inorganic fluids etc. efficiency
CTR Solar field • Different heliostat designs according to their location in the solar field
• Reduction in cost of solar field
Heliostats • Large size heliostats • Reduction in total number of tracking devices, reducing tracking
cost per m2
Heat transfer fluid • Alternative fluids such as molten salts, superheated steam, • Improvement in power cycle efficiency leading to reduced cost
pressurized air that can operate at high temperature of electricity generation
LFR Primary reflectors • Automation of primary reflectors • Capital cost reduction
Heat transfer fluid • Use of superheated steam in place of saturated steam • Improvement in power cycle efficiency due to high operating
temperature of superheated steam
Receiver • Enhanced Absorptance, low emittance and more resistant receiver pipe • Can withstand high pressure, high temperature superheated steam
All Mirrors • Anti-soil coatings, hydrophobic coatings on mirrors • Decrease mirror cleaning requirements, reduced mirror washing
cycles, reduced O&M cost
Storage • Thermocline storage that permits storage of cold and hot fluid in a • Reduced cost of storage as compared to two-tank storage
single tank
Turbine • Enhanced internal seals to decrease leakage, improved blade • Improvement in power cycle efficiency
designs and manufacturing methods

For estimating LCOE, capital cost of parabolic trough based CSP
plants (without storage) in India in 2015 is taken as US $2540/kW
(i.e. INR 169240/kW) and capacity utilization factor as 25% (Table 2).
Other assumptions are: (i) a discount rate (d) of 10%, (ii) useful life
(n) of plant as 25 years and (iii) annual operation and maintenance
(O&M) cost as 2% of the capital cost.
Using Eqs. (4) to (6), and future capital cost estimated in Table 10,
LCOE of parabolic trough based CSP plants (without storage) in India
has been estimated and presented in Table 12.
As presented in Table 12, a reduction of about 40% in the LCOE of
parabolic trough based CSP plants (without storage) is expected by
2030 and about 50% by 2050.
As already discussed, due to possibility to relatively inexpensive
thermal storage associated with parabolic trough based CSP plants,
capacity utilization factor can be increased and that may reduce LCOE
depending on the additional cost incurred in solar collection field and
storage devices. Some preliminary calculations have been made to
demonstrate the effect of thermal storage on LCOE. Since in Indian
context, cost data for CSP plants with provision of thermal storage was
not available, data pertaining to cost and capacity utilization factor has
been taken for two operational plant (one without storage and other
with 6 hour storage) in the United States.
The data for Mojave solar project in California, USA (280 MW
capacity, without storage, 25% CUF and approx. cost US $5710/kW)

Table 6
Possibility of manufacturing of different CSP components in India with strengths, limitations and possible solutions (ESMAP, 2012).

Component/industry Strength Limitations Possible solutions

Mirrors • Existing setup of float glass manufacturing industries
• Availability of technology with presence of International
players in India
Support structures • Fairly mature fabrication industry
• Automotive industry capable of manufacturing support
structures
• Huge potential for cost reduction
Receiver tubes • Few Indian industries actively involved in indigenously
developing selective coatings for receiver tubes
• Huge market potential as receiver tubes can also be
used for industrial process heating applications
Tracking and drive • Some players are supplying tracking solutions for mil-
mechanism itary applications and can do so for CSP industry also
HTF (synthetic oil) • Presence of international players Dow and Solutia in
India
• Indian oil and reliance petrochemicals may
manufacture
Turbine • Highly mature turbine industry
manufacturing • Technological knowhow available with Indian
industries
HTF pumps • Mature industry for refineries and thermal power
plants
• Unavailability of low iron sand
• Testing facility and industry
standards not available
• Testing facility not available in
India
• Structures should be tested in
Indian conditions
• Technical expertise for metal--
glass seal not available
• Proprietary know how for sen-
sors and controllers
• Technology transfer from global
companies is required
• Unavailability of highly pure
propylene crude
• Technical expertise for blending
and sulphonation not available
• Knowhow for meeting com-
plete specifications still lacking
• Relatively smaller capacity
(50–100 MW) required
• Opportunities for year on year
cost reduction is small
• Low iron sand can be imported
• Low iron flat float glass can be imported and bending
and subsequent operations can be performed locally
• From JVs/licensing route testing facility in India can be
developed
• Licensing/JVs with prominent International companies
• Indigenous development of technical know how
• Indigenous R&D by Indian companies
• Low manpower cost can provide a competitive edge
• R&D collaboration with prominent international
companies
• Collaboration with CSP technology providers
• Indian industries can meet the demand from other
regions to increase profitability
126 C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128

Table 7 Table 10
Expected percentage reduction in cost of CSP components in India due to local Future capital cost of CSP plants in India (learning rate of 10%).
manufacturing and reduced logistics (ESMAP, 2012).
t Global cumulative Future capital cost C(t)
Component Percentage reduction due to diffusion Q(t), GW
US $/kW INR/kW
Local manufacturing Reduced logistics
2015 4.6 2540 169,240
Mirrors 3 5 2020 29 1920 127,926
Tracking devices and mechanisms 3 2–5 2025 63 1706 113,695
Receiver tubes 3 2 2030 138 1515 100,921
HTF pumps 8–10 3 2040 267 1370 91,288
Turbines 5–10 2 2050 405 1286 85,686
PTC structures 4–7 3
Solar steam generators 3–5 2

Table 11
Envisaged cumulative installed capacity of CSP plants in India.
Table 8 Year Expected cumulative installation (MW)
Three scenarios for global cumulative diffusion of CSP plants (Viebahn et al., 2011).
2015 225
Year Cumulative diffusion (GW) 2020 500
2025 1000
Very optimistic Optimistic-realistic Pessimistic
2030 5000
scenario scenario scenario
2040 10,000
2015 4.6 4.6 4.6 2050 20,000
2020 40 29 14
2025 89 63 26
2030 200 138 47
2040 630 267 83 All the estimates of LCOE have been obtained assuming a discount
2050 1000 405 120
rate of 10%. In view of the present cumulative installation and a learning
rate of 10%, it is expected that future capital cost shall be significantly
lower than the present capital cost and project developers might
show increased interest in installation of CSP plant. That scenario may
and Solana generating system in Arizona, USA (280 MW capacity,
facilitate borrowing at lower interest rates due to lower risk involved
6 hour storage, 38.4% CUF and approx. cost US $7142/kW) has been
and consequently lower discount rates. Lower discount rates shall
taken from Table 1 (SolarPACES, 2017) for the analysis and it is assumed
further lead to lower values of LCOE. Fig. 7 presents expected variation
that same cost differentials for the CSP plants with and without 6 hour
in LCOE for three discount rates (6%, 8% and 10%). Lowering of discount
storage shall prevail in India. Table 13 presents future capital cost of
rate from 10% to 8% is expected to lower LCOE by 13%.
parabolic trough based CSP plants (with 6 hour storage) in India and
As reported (Neij, 2008), we have used a learning rate of 10% for the
its implication on LCOE. Comparing Tables 12 and 13, it is seen that
present study. There is always an uncertainty associated with assump-
with the present cost assumptions, by the end of 2050, LCOE of parabolic
tion of learning rate and the same shall have considerable impact on
trough based CSP plant in India with 6 hour storage is expected to
future capital cost and consequently LCOE. Neij (2008) had suggested
be 18.6% less than the LCOE of parabolic trough based CSP plant
an uncertainty of ± 5% for the learning rate of CSP plants. An analysis
without storage. A comparison of future capital cost and LCOE for
for effect on future capital cost for three different learning rates (5%,
CSP plants without storage and with 6 hour storage is shown in Figs. 5
10% and 15%) for CSP plants has also been undertaken in the present
and 6.
study (Fig. 8). As shown in Fig. 8, by the end of 2050, future capital
cost for parabolic trough based CSP plant in India with learning rate of
5% is expected to be 41% higher than the future capital cost with learn-
ing rate of 10%. This signifies the fact that the assumption of learning
Table 9
Effect on future cost of sub-systems due to cumulative diffusion and learning rates. rate has highest impact on the future capital cost of parabolic trough
based CSP plant in India.
CSP Learning Learning Year Cumulative Future capital
sub-system rate parameter diffusion cost C(t)
(b) (GW) Concluding remarks
US $/kW INR/kW

Solar energy 12% (−) 0.1844 2015 4.6 940 62,632 A preliminary attempt has been made to analyze cost reduction
collection 2020 29 669 44,575 possibilities of parabolic trough based CSP plants in India. The present
system 2025 63 580 38,645
2030 138 502 33,448
capital cost of plant (with and without storage) and capacity utilization
2040 267 445 29,650
2050 405 412 27,452
Thermal energy 12% (−) 0.1844 2015 4.6 635 42,310 Table 12
transfer 2020 29 452 30,117 Effect on LCOE of parabolic trough based CSP plants (without storage) in India due to cost
system 2025 63 392 26,119 reduction.
2030 138 339 22,588
2040 267 300 19,989 t Expected cumulative Future capital cost LCOE
2050 405 278 18,523 installation (MW)
US $/kW INR/kW US $/MWh INR/MWh
Power 5% (−) 0.074 2015 4.6 965 64,298
generation 2020 29 842 56,102 2015 225 2540 169,240 151 10,046
system 2025 63 795 52,971 2020 500 1920 127,926 114 7594
2030 138 750 49,973 2025 1000 1706 113,695 101 6749
2040 267 715 47,640 2030 5000 1515 100,921 90 5991
2050 405 693 46,175 2040 10,000 1370 91,288 81 5419
2050 20,000 1286 85,686 76 5086
(1 US $ = INR 66.63 as on February 1, 2016).
 C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128 127

Table 13 160
Effect on LCOE of parabolic trough based CSP plants (with 6 hour storage) in India due to d = 10% d = 6% d = 8%
cost reduction. 140

LCOE (US$/MWh)
t Expected cumulative Future capital cost LCOE 120
installation (MW)
US $/kW INR/kW US $/MWh INR/MWh 100
2015 225 3175 211,550 123 8176
2020 500 2400 159,912 93 6180 80
2025 1000 2133 142,088 82 5491
60
2030 5000 1894 126,181 73 4876
2040 10,000 1713 114,104 66 4410
40
2050 20,000 1608 107,108 62 4139
2015 2020 2025 2030 2035 2040 2045 2050
Year (t)

Fig. 7. Expected variation in LCOE for CSP plant in India for different discount rates.
3500
without storage with 6 hour storage
Capital cost (US$/kW)

3000
2500
2000
1500
1000
2015 2020 2025 2030 2035 2040
Year (t)
Fig. 5. Future capital cost of CSP plant in India without storage and with 6 hour storage.
factor has been taken from the operational plants in India and other
countries that was available in public domain (SolarPACES, 2017). It
is expected that learning effects due to global cumulative diffusion,
innovation and breakthroughs in various CSP components and indigeni-
zation of CSP components shall have significant impact on the future
cost of parabolic trough based CSP plant and will subsequently impact
its cost of generation.
A brief description of technological improvements in CSP compo-
nents and their implications have been discussed. Further, possibility
of manufacturing different CSP components in India with strengths,
limitations and solutions have also being presented. Several measures
(such as local manufacturing and consequently reduced logistics) that
could lower the cost of CSP plants have been summarized. For example,
local manufacturing of CSP components is expected to reduce capital
cost by 14% in the next two decades. Similarly, reduced logistics is
expected to reduce capital cost by 8%.
Possibility of including 6 h of storage and its effect on LCOE has also
been considered. A provision of 6 h of storage is expected to increase
capital cost by about 25% of the cost without storage but the same
shall also result in increase of capacity utilization factor from 25%
(without storage) to 38.4% (with 6 hour storage) leading to lower values
of LCOE. For example, LCOE of a CSP plant with 6 hour storage is
expected to be 18% lower than LCOE of CSP plant without storage by
the end of 2050.
Though there is a possibility of reduction in the capital cost of CSP
plants with increased cumulative diffusion of the technology, the
expected learning rate is rather moderate (10%). Using a learning rate
2045 2050 of 10%, the capital cost and LCOE of parabolic trough based CSP plants
in India in 2050 is expected to reduce to 50% of the present capital
cost and LCOE. Further, as reported, uncertainty in learning rates shall
have significant effect on future capital cost and consequently LCOE. A
sensitivity analysis for three different learning rates (5%, 10% and 15%)
has been undertaken. The cost of parabolic trough based CSP plant in
India (without storage) is expected to vary from US $1824/kW (for a
learning rate of 5%) to US $891/kW (for a learning rate of 15%) by the
end of 2050.
With increased cumulative diffusion, there is a possibility of
borrowing at lower interest rate from international market leading to
reduced discount rate. A sensitivity analysis for three discount rates
(6%, 8% and 10%) has also been undertaken. The LCOE for parabolic
trough based CSP plant in India (without storage) is expected to vary
from US $58/MWh (for a discount rate of 6%) to US $76/MWh (for a
discount rate of 10%).
The uncertainty in assumptions such as present capital cost with and
without storage, learning rates for CSP plant, capacity utilization factor,
annual operation & maintenance cost of CSP plant shall significantly
affect the cost reduction estimates presented in the study. It is expected
that with increased cumulative diffusion, there shall be a possibility of
having more realistic capital cost assumptions and operational data of
parabolic trough based CSP plants in India leading to more accurate
future cost estimates.

170
without storage with 6 hour storage
3000
150
LR = 5% LR = 10% LR = 15%
LCOE (US$/MWh)

Capital cost (US$/kW)

130 2500

110 2000

90 1500

70 1000

50 500
2015 2020 2025 2030 2035 2040 2045 2050 2015 2020 2025 2030 2035 2040 2045 2050
Year (t) Year (t)

Fig. 6. Expected LCOE of CSP plant in India without storage and with 6 hour storage Fig. 8. Effect of different learning rates on future capital cost of CSP plant in India.
128 C. Sharma et al. / Energy for Sustainable Development 42 (2018) 121–128

Acknowledgement Kandpal TC, Garg HP. Financial evaluation of renewable energy technologies. New Delhi:
Macmillan India; 2003.
McDonald A, Schrattenholzer L. Learning rates for energy technologies. Energy Policy
Chandan Sharma acknowledges the encouragement provided by the 2001;29:255–61.
Department of Technical Education, Government of State of Rajasthan Moro JH, Duart JMM. CSP electricity cost evolution and grid parities based on the IEA
roadmaps. Energy Policy 2012;41:184–92.
(India), Government Engineering College, Ajmer and the Quality Neij L. Cost development of future technologies for power generation – a study based on
Improvement Program at Indian Institute of Technology Delhi. experience curves and complementary bottom-up assessments. Energy Policy 2008;
36:2200–11.
Price HW, Carpenter S. The potential for low-cost concentrating solar power systems.
References NREL/CP-550-26649; 1999. [Golden Colorado].
Purohit I, Purohit P, Shekhar S. Evaluating the potential of concentrating solar power
Broek MVD, Hoefnagels R, Rubin E, Turkenburg W, Faaij A. Effects of technological
generation in North-western India. Energy Policy 2013;62:157–75.
learning on future cost and performance of power plants with CO2 capture. Prog
Reddy KS, Kumar KR, Devraj VA. Feasibility analysis of megawatt scale solar thermal
Energy Combust Sci 2009;35:457–80.
power plants. J Renewable Sustainable Energy 2012;4, 063111.
CSTEP. Engineering economic policy assessment of concentrated solar thermal power
Sharma C, Sharma AK, Mullick SC, Kandpal TC. Solar thermal power generation in India:
technologies in India. Bangalore, India: Centre for Study of Science Technology and
effect of potential incentives on unit cost of electricity. Int J Sustainable Energy
Policy (CSTEP); 2012http://www.cstep.in/publications/reports/index/1/30.
2017;36(8):722–37.
Desai NB, Bandyopadhyay S, Nayak JK, Banerjee R, Kedare SB. Simulation of 1MWe solar
SolarPACES. CSP projects around the world. Almeria, Spain: Solar Power and Chemical
thermal power plant. Energy Procedia 2014;57:507–16.
Energy Systems (SolarPACES); 2017http://www.solarpaces.org/csp-technology/
ESMAP. Development of local supply chain: the missing link for concentrated solar power
csp-projects-around-the-world.
projects in India. Washington D.C.: The Energy Sector Management Assistance
Sundaray S, Kandpal TC. Preliminary feasibility evaluation of solar thermal power
Program (ESMAP), World Bankhttps://www.esmap.org/Publications, 2012.
generation in India. Int J Sustainable Energy 2013;33(2):461–9.
ESTELA. Solar thermal electricity 2025, clean electricity on demand: attractive STE cost
Trieb F. Concentrating solar power for the Mediterranean region. BMU, Germany: German
stabilize energy production. Brussels, Belgium: European Solar Thermal Electricity
Aerospace Center (DLR); 2005. http://www.dlr.de/Portaldata/1/Resources/portal_
Association (ESTELA)http://www.estelasolar.org/wp-content/uploads/2015/11/
news/newsarchiv2008_1/algerien_med_csp.pdf.
2010-Solar-Thermal-Electricity-2025-ENG.pdf, 2010.
Viebahn P, Yolanda L, Trieb F. The potential role of concentrated solar power (CSP) in
Ferioli F, Schoots K, Van der Zwaan BCC. Use and limitations of learning curves for energy
Africa and Europe - a dynamic assessment of technology development, cost develop-
technology policy: a component-learning hypothesis. Energy Policy 2009;37:
ment and life cycle inventories until 2050. Energy Policy 2011;39:4420–30.
2525–35.
WEC. World energy perspective: cost of energy technology. London, United Kingdom:
Huenteler J, Niebuhr C, Schmidt TS. The effect of local and global learning on the cost of
World Energy Council (WEC); 2013https://www.worldenergy.org/publications/
renewable energy in developing countries. J Clean Prod 2014;128:6–21.
2013/world-energy-perspective-cost-of-energy-technologies/.
IEA. Technology roadmap - solar thermal electricity. 75739 Paris Cedex 15, France: Interna-
tional Energy Agency (IEA); 2014https://www.iea.org/publications/2014edition.pdf.
IRENA. Renewable energy technologies: Cost analysis series, concentrating solar power.
Masdar city, United Arab Emirates: International Renewable Energy Agency (IRENA);
2012www.irena.org/documentdownloads/…/re_technologies_cost_analysis-csp.pdf.