Solar thermal powered Organic [601107]

Licență

Solar thermal powered Organic [601107]

Byadmin ianuarie 1, 2024

Solar thermal powered Organic
Rankine Cycles 16
M. Orosz1, R. Dickes2
1Massachusetts Institute of Technology, Cambridge;2University of Lige, Lige, Belgium
16.1 Introduction to solar Organic Rankine Cycle
systems
Sunlight is the primordial energy source for most of the work that has occurred on
Earth in the past 4.54 billion years. The most basic transformations are physical,e.g., sunlight evaporates water from oceans, raising it into the atmosphere and trans-porting it hundreds of miles to fuel global precipitation patterns. The original mecha-nism for converting sunlight into electro-mechanical work, however, was developedby single celled organisms, probably the progenitors of today ’s chloroplasts. Photo-
synthesis, the reduction of atmospheric carbon dioxide mediated by water and light,
produced oxygen (a by-product) and energy to power cellular functions, such as
DNA transcription, organelle formation, metabolism, and cell division. Over timethis sunlight-powered work ﬁlled the earth ’s atmosphere with oxygen, creating an
atmosphere similar to today ’s, by around 2.3 billion years ago.
In the human era, the earliest references that lay a foundation for the concept of con-
verting sunlight into work involve a compilation of two ancient Sicilian/Greek inno-vations. Though likely apocryphal, the AD 2nd Century author, Lucian, reportedthat during the Siege of Syracuse (214 e212 BC), Archimedes defended the city by
setting ﬁre to Roman ships using mirrors in a parabolic arrangement to concentrate
sunlight. Nearly three centuries later, Hero of Alexandria described a device for con-verting heat into work with an “aeolipile ”in his Pneumatica. His device converted
thermal energy into rotary motion by means of a radial expander using two opposingsteam jets emanating from a central, spinning boiler, though it is not known whether itwas ever put to productive use.
In the ﬁrst story, sunlight is concentrated to create heat (the autoignition tempera-
ture of wood is around 250
/C14C), while in the second, heat creates steam in a boiler,
which then expands out of ori ﬁces at opposite tangents to create rotation. When com-
bined in this way, these two innovations embody the essence of concentrating solarpower (CSP). In the midst of the Industrial Revolution, this synthesis was ﬁnally
put to practice, ﬁrst by French inventor Augustin Mouchot, and later by Swedish
American John Ericsson, although dates of international exhibition are Ericsson ﬁrst,
whose solar air engine was featured at the Centennial Exhibition in Philadelphia in1876, followed by Mouchot at the Paris Universal Exhibition in 1878.
When the ﬁrst solar power plant was commissioned in 1913 in Maedi, Egypt, it was
still several decades before Chapin, Fuller, and Pearson would demonstrate a working
Organic Rankine Cycle (ORC) Power Systems. http://dx.doi.org/10.1016/B978-0-08-100510-1.00016-8
Copyright ©2017 Elsevier Ltd. All rights reserved.

silicon photovoltaic (PV) cell (4% ef ﬁciency, at Bell Labs in 1954). The Maedi plant,
designed by American engineer Frank Shuman, used ﬁve parabolic trough collec-
tors (PTCs), each 62 m long and 4 m wide, to concentrate the sun ’sr a y so nab o i l e r
tube supplying a 75 kW atmospheric (lo w-pressure) steam engine coupled to a
water pump capable of lifting 22 cubic meters of Nile water per minute ( American
Inventor Uses Egypt ’s Sun for Power, 1916 ). Though technologi cally successful,
Shuman ’s demonstration plant never gain ed traction due to the onset of the ﬁrst
World War and the widespread availability of cheap fossil fuels, however, experi-
mentation with solar powered motors prolif erated in the ensuing decades, including:
the SOMOR solar pump invented by Italians Daniele Gasperini and FerruccioGrassi in the 1940s; a 1 kW
eSolar ORC pump installed in Mali in 1966; and
a3 k W eSolar ORC developed by Harry Zvi Tabor and Lucien Bronicki and
demonstrated to the United Nations in 1961, leading to the formation of Ormat
(Bronicki, 2013 ). The 354 MW ecombined Solar Energy Generating System plants,
operating in California, starting in 1984, es tablished the commercial feasibility of
utility-scale CSP using primarily steam Ra nkine power blocks, creating an industry
that has over 4.5 GW eof installed capacity in 2015. Subsequently, the commercial
viability of solar-driven ORCs has been explored at many scales: in 2006, agrid-connected 1 MW
eSolar ORC plant ( “Saguaro ”)w a sc o m m i s s i o n e di nR e d
Rock, Arizona (Southwest US A) by Arizona Public Service ( Canada et al.,
2005b ), while more recently, the focus on solar ORC systems has tended towards
smaller scale, experimental systems optim ized for niche applications, as discussed
inSection 16.1.1 .
16.1.1 Applications dgrid connected power generation,
desalinization, distributed power generation,
cogeneration, and hybrid systems
Power from a solar ORC (SORC) can be useful in a variety of applications, from the
ordinary supply of electrons via a traditional distribution grid, to islanded microgrids,to cogeneration for community or industrial use. Hybrid systems, involving other gen-eration sources or one or more additional sources of, or uses for, thermal energy, areincreasingly considered for solar ORC applications. This section highlights a range oftypical uses for solar ORCs.
16.1.1.1 Grid connected power generation
There seems to be little scope for grid connected standalone solar ORC projects underthe current technical and economic scenario. The 1 MW
egrid connected solar ORC,
referenced in Table 16.2 , is an unlikely application for deployment in the future due
to the recent decline in the cost of the main competing technologies, i.e., fossil fuels
and solar PV. Grid scale CSP using steam Rankine, higher temperature sCO 2, or air
Brayton cycles are still being commercially explored, particularly in the case wherethe value of the energy produced lies in its dispatchability via thermal energy storage(TES) (for more detail see Section 16.2.3 ). At grid scale, in 2015, PV installs for under570 Organic Rankine Cycle (ORC) Power Systems

USD 2/W peak whereas CSP typically has a speci ﬁc cost of over USD 4/W of name-
plate capacity ( Orosz, 2015 ). The cost of CSP with TES, however, remains lower than
the cost of PV with electrochemical batteries at large scales; whereas large-scale bat-teries for shifting PV production into the evening are commercially unavailable, TESoperational experience exists at the GWh scale in several CSP plants (see Table 16.1 ).
This indicates that a solar ORC with TES may have a role to play in meeting energyneeds as a dispatchable power source in a future where the grid is comprised of a high
fraction of irregular, renewable sources.
16.1.1.2 Distributed power generation
Whereas the value of dispatchability in grid connected solar power plants is only
beginning to gain recognition in the market, the demand variance and the need for stor-
age of intermittent sun power is already inherent in any load-following application
such as an isolated industrial or community mini-grid. Operational experience withislanded solar ORC systems is limited and commercially immature, however, recentanalyses suggest that the application of solar ORC to support energy needs in remoteareas for commercial or community loads may, if properly sized and designed, beeconomically attractive in comparison with alternative systems capable of meetingthe demand dynamic, such as diesel generators ( Mitterhofer and Orosz, 2015 )o rP V
arrays with large battery storage systems. The attractiveness of the solar ORC for
distributed generation can also be increased when hybrid solutions are relevant, as
discussed in Section 16.1.1.5 .
16.1.1.3 Desalination
Direct solar distillation of brackish water or seawater has been practiced from an early
time through the coupling of a solar collector with a traditional still. Indirect desalina-
tion, through multiple stage ﬂash distillation or reverse osmosis (RO), utilizes solar
energy to derive either solar-derived electricity or mechanical work (and potentiallyheat) to purify water. Investigation of solar ORC desalination has included bothdual use (electricity and water) and cogeneration systems, in which the water desa-lination mode directly couples the ORC shaft output to a high-pressure pump feedingbrackish water to an RO membrane. The feedwater in these systems can be used forheat rejection from the ORC, and the increased temperature of the water improves the
efﬁciency of the RO ( Tchanche et al., 2010; Manolakos et al., 2007; Delgado-Torres
and García-Rodríguez, 2007 ).
16.1.1.4 Irrigation
Solar-powered ORCs can also be used for irrigation duty for agricultural production. In
such systems, solar energy supplied to an ORC is converted into mechanical power
which is directly exploited to drive an irrigation pump. The advantages of solar water
pumping include the coincidence between seasons of high solar irradiation and the de-mand for irrigation, and the fact that, as a load, irrigation is relatively insensitive to thediurnal intermittency of solar power ( Pytilinski, 1978 ).Solar thermal powered Organic Rankine Cycles 571

Table 16.1 Comparison of recently installed large-scale solar power plants using photovoltaic (PV) or
concentrating solar power (CSP) technology, including speci ﬁc costs and hours of storage ( Orosz, 2015 )
Plant name TypeNameplate
(MW)Pricetag
(USD 000,000)Reported
production(GWh/year) Storage (h)Installed cost
(USD/W)Installed cost
(USD/kWh/year)
Lalackama PV 60 110 160 0 1.83 0.69
Cha~nares PV 40 70 94 0 1.75 0.74
Desert Sunlight PV (CdTel) 550 1900 1024 0 3.45 1.86
Crescent Dunes CSP tower 110 975 485 10 9.09 2.01Antelope Valley PV (CdTel) 230 1400 623 0 6.09 2.25Topaz PV (CdTel) 550 2400 1053 0 4.36 2.28Agua Caliente PV (CdTel) 290 1800 741 0 6.21 2.43Andasol 1 CSP trough 150 380 150 7.5 2.53 2.53
Solana CSP trough 280 2000 600 6 7.14 3.33
Ivanpah CSP Tower 392 2200 519 0 5.61 4.24572 Organic Rankine Cycle (ORC) Power Systems

16.1.1.5 Hybrid solar organic Rankine cycle
Solar ORCs with or without TES may be used in conjunction with a variety of other
generation technologies, some of which are simply used in parallel or potentiallyscheduled sequentially via a macro level control strategy. Other sources of energyinclude PV arrays, concentrating PV/thermal (CPVT) collectors, fuel-based generatorsets, fuel cells, wind turbines, biomass boilers ( Fig. 16.1 ), waste heat recovery,
geothermal, etc. A solar ORC may also be con ﬁgured to bypass heat around the
ORC to directly meet thermal loads, such as industrial process heat (e.g., beverage
bottle cleaning, aiding ﬂash distillation processes, enhanced oil recovery, etc.), an
absorption chiller, or building heating or hot water supplies. These con ﬁgurations,
while involving increasing degrees of complexity, may justify the capital expensebased on fuel or battery cost savings (in the case of islanded operation), increasedcapacity factor, and the leveraging of locally available resources.
The bene ﬁts of hybridizing solar ORC with combustion based generators include,
reduced fuel consumption (replacing fuel-derived power with solar-derived power)
and the potential for improved overall fuel ef ﬁciency via recuperating waste heat
from the exhaust stream of the genset. Hybridizing with PV arrays can be attractiveif the daytime loads are supplied via the relatively inexpensive PV, while solar thermalproduction is dispatched from the TES, to supply nighttime loads, displacing the needfor costly batteries. CPVT with ORC represents an opportunity to further reduce theinfrastructure involved in a PV eCSP hybrid.
The capital cost of CSP collectors is especially justi ﬁed when local demand for heat
exists, or when available waste heat is at low temperatures that could bene ﬁt from solar
boosting (e.g., from server datacenters). Some industrial processes may be supplied by
CSP alone without the use of an ORC, but dynamics in load and solar resource (seeSection 16.3.3 ) will tend to create a mismatch in the availability and requirement
for heat. Excess heat from the collectors can, in this case, be exploited by an ORC,offsetting plant power consumption. Similarly, in cases where the thermal loadsare seasonal (e.g., wintertime heating of buildings) the addition of an ORC can
Figure 16.1 1M W eShive village electri ﬁcation plant (Thermax India), a hybrid plant
consisting of a concentrating solar power (CSP) collector ﬁeld, steam Rankine cycle, Organic
Rankine Cycle, and a backup biomass- ﬁred boiler. Left: Thermax PTC collector. Right:
GE/Calnetix 100 kW eORC ( Thakur, 2013 ).Solar thermal powered Organic Rankine Cycles 573

productively utilize otherwise wasted summertime CSP production. Another option
making use of seasonal surplus solar resource, with or without ORC, would be to inte-grate a multieffect absorption chiller, although the cost of this approach compared toPV driving an electric heat pump would be worth exploring. Other hybridizationscenarios are possible such as combining ORC with vapor compression cooling(Dumont et al., 2015 ); the key considerations in selecting and specifying the solar
ORC components is the value of dispatchability and thermal products, and the costs
otherwise incurred via alternate solutions. This highlights the fact that hybrid solutions
tend to be unique and site speci ﬁc, and one must recognize that the cost of investi-
gating each situation and engineering the integration of multiple systems, whichmay require novel control strategy implementations, poses an obstacle to widespreadadoption.
16.1.2 Meteorological and solar resource dynamics
As noted in the introduction to this chapter, a solar ORC obtains its heat input from thesun at the center of our solar system. While the astrophysics and celestial mechanicsof this star are beyond the scope of this book, a basic overview of the functionality
of the solar resource and its in ﬂuence on the design and optimization of thermal power
systems will be summarized.
16.1.2.1 Sunlight energy
The sun is a complex thermonuclear reactor whose mass holds the earth and other solarsystem objects in its gravitational ﬁeld, while releasing 3.86 /C210
26J/s via the fusion
of hydrogen into helium. At its outermost layers, the sun surface is maintaining atemperature of approximately 5778 K. According to Planck ’s well-known equation
for the spectral radiance of a body at thermal equilibrium at temperature T:
B
l¼2hc2
l51
ehc
lkBT/C01(16.1)
(where Bis the power in Watts per m2per nm, lis the wavelength, kBis the Boltzmann
constant, his the Planck constant, and cis the speed of light in the medium) the sun
emanates a constant spherical beam of electromagnetic radiation with a distinctspectral signature approximating that of a blackbody at 5778 K.
16.1.2.2 The solar spectrum
Due to the size of the earth (DIA ¼12,742 km) and its distance from the sun
[1.496 /C2108km or 1 astronomic unit (AU)] the amount of solar energy intercepted
by the earth is 1.74 /C21017J/s. The shape of the solar spectrum is comparable to a
blackbody at 5778 K, however, the actual solar spectrum is modulated by the sun ’s
composition and partially attenuated by absorption bands in the earth ’s atmosphere
in proportion to the depth of atmosphere traversed by a ray emanating from the sun.574 Organic Rankine Cycle (ORC) Power Systems

Due to the rotation of the earth on an w23.5 degrees tilted axis, this depth varies by
latitude, seasonality, and time of day. The sunlight impinging on a ﬂat plane at the
surface of the earth is nearly collimated (the sun shape has an acceptance angle of0.55 degrees); the intersection of a spherical beam of sunlight projected out from
the sun across 1 AU distance striking the earth results in an extraterrestrial solar con-
stant of approximately 1367 W/m
2. An approximation of the direct normal irradiance
(DNIdthe beam component as opposed to diffuse light) solar spectrum at 1.5 times
the thickness of the atmosphere is a frequently used standard for CSP known asAM1.5D (where D stands for direct, shown in Fig. 16.2 ). At prospective CSP sites,
historical DNI measurements, obtained with a pyrheliometer, can be used to estimatelocal solar resources.
While the detailed characteristics of the solar spectrum have implications for the
optical performance of solar collectors (specular re ﬂectivity of primary mirrors,
absorption, etc.), it is not uncommon to simply deploy the integral of the spectrum(in W/m
2), as a ready means of ascertaining power ﬂux impinging on a collector aper-
ture. The solar constant at the top of the atmosphere ( w1367 W/m2) is reduced by 25%
or more by atmospheric conditions; at high DNI locations, at the earth ’s surface, it can
reach above 1000 W/m2and 7.5 kWh/m2/day. A more typical range for areas still
considered appropriate for use of CSP is above 800 W/m2and 5 kWh/m2/day, and
standards or speci ﬁcations for collector output will typically specify the DNI condi-
tions at which the measurements were made.
16.1.2.3 Tracking
Most CSP collectors employ imaging optics that require the aperture of the collector to
be normal to the sun. From dawn until dusk the sun is constantly moving in azimuth
and zenith (zenith is 90-elevation, where elevation is the angle between the horizon
and the sun) with respect to a ﬁxed location on a rotating earth. Point focus systems,
unlike line-focus systems (both described further in Section 16.2.1 ) must move to track
the sun position in both azimuth and zenith coordinates (known as A-Z tracking).
00.511.52
500 1500 1000 2000 2500
Wavelength (nm)I (W m–2 nm–1)Extraterrestrial solar spectrum
AM 1.5 direct normal irradiance
Figure 16.2 Extraterrestrial and Direct Normal Irradiance (DNI) from ASTM G173 DNI is at an
air mass of 1.5 ( ASTM, 2012 ).Solar thermal powered Organic Rankine Cycles 575

In line-focus systems 2-D tracking is possible, however, the more common approach is
the minimization of the cosine of the angle f(between the collector aperture and the
sun) by means of tracking around a single north-south or east-west axis. Compared to ahorizontal surface, the increased solar irradiance on a single-axis tracking surface isapproximately 10 e20% or 20 e30% for east-west and north-south parallel axes
respectively, depending on latitude, with an additional 5 e10% possible for two-axis
tracking ( Helwa et al., 2000; Duf ﬁe et al., 1994 ). Algorithms for calculating the sun
position (azimuth and elevation) as a function of latitude, longitude, and time are
published by National Renewable Energy Laboratory (NREL) in the USA and thePlataforma Solar de Almería (PSA) in Spain Reda and Andreas, 2008 ). Calculating
the tilt angle for a single-axis platform in an east-west or north-south orientationCSP system involves minimizing the angle of incidence between the collector andimpinging sunlight; equations for obtaining this result are widely available in the liter-ature ( Dufﬁe et al., 1994 ).
16.1.2.4 Intermittency
ORCs driven by geothermal or waste heat sources can usually be designed with near
constant heat input rates, or rates that vary in a predictable and controllable manner.ORCs driven exclusively or partially via solar input must contend with thermal inputsthat can vary with unknown frequency and magnitude, superimposed on other ﬂuctu-
ations with a wide range but having a greater degree of predictability. Solar concentra-tors (discussed in more detail in Section 16.2 ), experience varying intensity of sunlight
on their aperture on the timescale of seconds to minutes during cloud passing, on the
timescale of hours due to the diurnal cycle, and on seasonal timescales on the basis of
latitude driven changes in the time varying pathlength of rays through the atmosphere(as well as any attending angle of incidence effects). Landscape shadowing effects canhave both diurnal and seasonal components. The DNI on a single-axis tracking surfacefor a typical cloudless day is shown in Fig. 16.3 and it is contrasted to cloudy weather
at the same location; the effect of cloud passing can be seen to produce rapid changesin energy ﬂux that require further design considerations when coupling CSP and ORC
technologies (see Section 16.2.3 ).
16.1.2.5 Solar resource maps
The spatial distribution of the availability of beam (relatively collimated) sunlight
(i.e. DNI), versus global irradiance (DNI țdiffuse light) on an average daily basis
(integrating the ﬂuctuations due to local weather) per square meter for locations on
earth ’s surface has been characterized by a combination of ground based and satellite
measurements to produce useful maps for planning solar projects, an example of whichis shown in Fig. 16.4 :
For a more granular representation of local conditions, Typical Meteorological Year
(TMY) datasets that include pryheliometer measurements (for measuring DNI) areparticularly useful for simulating the dynamic behavior of solar systems (discussedin more detail in Section 16.3.3 ).576 Organic Rankine Cycle (ORC) Power Systems

16.1.3 Installed capacity dsurvey of existing sites
Since the ﬁrst development of heat engines, solar energy has been considered as a
viable heat source. While none of the earliest experimental investigations of solar-
based ORCs are in existence today, records and examples of experimental power
plants are available, including unique pilot systems for R &D and commercial power
generation units built and tested in the 20th century to the present. These SORC facil-ities range from “micro-power ”generation (typically de ﬁned as lower than 1 MW
e)t o
large power plants of several megawatts, representing many variants of the technology.
0
02004006008001000
51 0 1 5 2 0 2 5
Hour of the dayW m–2
Figure 16.3 DNI on a single-axis tracking surface (Lesotho, 2009) for a cloudless day (red) and
various cloudy days.
Adapted from Ireland et al. (2014).
50°S30°S10°S10°N30°N50°N
50°S30°S10°S10°N30°N50°N140°W120°W100°W 80°W 60°W 40°W 20°W 0° 20°E 40°E 60°E 80°E 100°E 120°E 140°E 160°E
140°W120°W100°W 80°W 60°W 40°W 20°W 0° 20°E 40°E 60°E 80°E 100°E 120°E 140°E 160°EDirect normal irradiation
averaged annual sum
<800 kWh/m /y
>2800 kWh/m /y801 – 1000
1001 – 1200
1201 – 1400
1401 – 1600
1601 – 1800
1801 – 2000
2001 – 2200
2201 – 2400
2401 – 2600
2601 – 2800
Figure 16.4 Map of direct normal irradiance (DNI) at the earth ’s surface in kWh per square
meter annually.
Adapted from NASA SSE 6.0 (1993 e2006) and DLR (2008).Solar thermal powered Organic Rankine Cycles 577

The records for these systems are incomplete, but an effort to compile a list of known
plants was undertaken by Mueller et al. (2016) and from this dataset of diverse solar-
based ORC systems, a nonexhaustive list of experimental and commercial powerplants with known characteristics is provided in Table 16.2 .
16.2 Solar Organic Rankine Cycle components and
architecture
16.2.1 Solar thermal collectors
Solar-based ORCs use solar energy as the heat source from which to generate elec-
tricity or mechanical power. The heat is obtained by means of solar thermal collectors
which intercept incoming sunlight and collect or re ﬂect it onto a heat collection
element (HCE). At the HCE, light is converted to sensible or latent heat and typicallytransferred to a ﬂowing transport medium, called the heat transfer ﬂuid (HTF). In most
situations, the HTF acts an intermediate ﬂuid that transports heat to the ORC evapo-
rator, however, direct generation can also be performed if the ORC working ﬂuid is
circulated through the HCE in place of a separate HTF, as discussed in Section
16.2.2 . The thermal ef ﬁciency of solar collectors depends signi ﬁcantly on the oper-
ating conditions and the insulation quality of the HCE, with values for the thermal ef-
ﬁciency typically ranging from 30% to 70%. For an HCE, the higher the temperature
difference between the ambient air and the HCE, the higher the thermal losses.
Many types of thermal collectors have been developed in past decades and a major
distinction can be made between concentrating and nonconcentrating technologies. Anonconcentrating collector has the same surface area for intercepting (aperture) and forabsorbing (HCE surface) sunlight (the radiation ﬂux onto the HCE is not increased),
and a tracking system is not usually required. Both direct and diffuse portions of
sunlight are exploited, and non-concentrating collectors can be well-suited for low-
temperature applications. In the case of solar-based ORCs, three main technologiesare used, namely solar ponds, ﬂat plate collectors, and evacuated tube collectors.
16.2.1.1 Salt-gradient solar pond
A salt-gradient solar pond (SGSP), or simply solar pond , is a pool of saltwater used for
collecting solar energy. Within the reservoir, a vertical salt concentration pro ﬁle
creates a three-layer strati ﬁcation dividing the brine into an upper convective zone
(UCZ) at the surface, a nonconvective zone (NZC) in the middle, and a lower convec-tive zone (LCZ) at the bottom (see Fig. 16.5 ). While the top and bottom zones have a
quasi-uniform density (low and high respectively) the nonconvective zone features adensity pro ﬁle that increases continuously from the upper to the lower boundary. As
such, any natural convection is eliminated in the middle zone and a stable insulation
layer is formed. During sunny periods, solar radiation passes through the layers andis absorbed by the lower zone. Thanks to the stable intermediate layer (NCZ inFig. 16.5 ), the temperature of the bottom zone rises (up to 95
/C14C) whereas the top of578 Organic Rankine Cycle (ORC) Power Systems

Table 16.2 Nonexhaustive review of existing experimental solar Organic Rankine Cycle (SORC) facilities
Location (date)Collector
technologyaCollectorareaHeat transfer
ﬂuidPower
outputWorking
ﬂuid Thermal storage References
Mali (1966) FPC 43 m
2e 600 W e None Einav (2004)
Ein Bokek, Israel (1977) SGSP 7000 m2Brine 150 kW e Solar pond Tabor (1981)
Pasadena, CA – USA
(1978)PDC 116.9 m2Toluene 30 kW Toluene None Kiceniuk (1985)
Willard, NM – USA (1979) PTC 1276 m2Mineral oil 19 kW R113 Thermocline
directFenton et al. (1984)
Kuwai city, Kuwait (1981) PDC 1100 m2Synthetic oil 100 kW/
700 kW thToluene Thermocline
directMoustafa et al. (1984)
Vignola, France (1982) PTC 1176 m2Thermal oil 100 kW Fluoroinert
FC75Thermocline
directSimonnot et al. (1987)
Perth, Australia (1984) PTC e Thermal oil 35 kW C 8F16 e Barutti et al. (1984) and
Bado et al. (1979)
Beith Ha ’Arava, Israel
(1982)SGSP 250,000 m2Brine 5 MW e Solar pond Tabor and Doron (1990)
Lausanne, Switzer (2001) LFC 100 m3Water 15 kW R123/R134a None Kane et al. (2003)
Sendai, Japan (2002) CPC 5.75 m2Water <1 kW R113 Buffer Saitoh et al. (2007)
Red Rock, AZ eUSA
(2006)PTC 10340 m2Mineral oil 1 MW n-pentane None Canada et al. (2005a)
Newcastle, Australia
(2006)PTC 132 m2Mineral oil 6 kW HFE 7100 n.a. Kohlenbach et al. (2006)
Almeria, Spain (2007) PTC n.a. Thermal oil 5 kW SES36 n.a. Galvez (2010)
Continued

Table 16.2 Continued
Location (date)Collector
technologyaCollectorareaHeat transfer
ﬂuidPower
outputWorking
ﬂuid Thermal storage References
Tianjin, China (2008) FPC 0.6 m
2R245fa 13 W R245fa None Wang et al. (2011)
Tianjin, China (2008) FPC/ETC 44 m2R245fa 1.7 kW R245fa None Wang et al. (2010)
Morphou, Cyprus (2010) PTC 216 m2Water 18 kW R245fa n.a. Electratherm (2010)
Burkina Faso (2012) SCT 180 m2Mineral oil 10 kW R245fa Two-tank direct N’Tsoukpoe et al. (2014)
Berea, Lesotho (2012) PTC 75 m2Glycol 3 kW/
25kWthR245fa n.a. Orosz (2012)
Crowley, LA eUSA
(2013)PTC 1051 m2Water 50 kW R245fa Buffer Chambers et al. (2014)
Tampa, FL eUSA (2013) PTC n.a. Water/
ethylene50 kW R245fa PCM Goswami et al. (2013)
Cadarache, France (2013) PTC 550 m2Water 10 kW R245fa Thermocline
directRieu (2012)
Ait-Baha, Morocco (2014) PTC 6159 m2Air 3 MW n.a. Sensible packed-
bedAit-Baha
Rende, Italy (2014) LFC 9780 m2Mineral oil 1 MW n.a. None NREL
Fallon, NV eUSA (2015) PTC 656 m2Water 33 MW Isobutane None NREL
Li/C18ege, Belgium (2015) PTC 78 m2Synthetic oil 3 kW R245fa None Georges et al. (2013) and
Dickes et al. (2014)
Busan, Korea (2015) ETC n.a. Water 1.5 kW R245fa n.a. Baral et al. (2015)
aCPC , compound parabolic collector; ETC, evacuated tube collector; FPC,ﬂat plate collector; LFC, linear Fresnel collector; PDC , parabolic dish collector; PTC, parabolic trough collector; SCT, solar central tower;
SGSP , salt-gradient solar pond.

the pond remains close to the ambient temperature. Depths of solar ponds range from
0.5 to 5 m and large surface areas are required ( >200,000 m2) for powering large-scale
systems. Thermal energy is transferred to the ORC by pumping some of the hot brineto the evaporator and the upper cold water can be used as a heat sink for the ORCcondenser. A key advantage of solar ponds is their ability to both collect and storethermal energy. Thanks to the high thermal inertia of the reservoir, heat power canbe delivered to the ORC at a stable temperature.
16.2.1.2 Flat plate collector
Aﬂat plate collector (FPC) is made of ﬂuid tubes ( ﬁlled with ﬂowing HTF) connected
to a darkened (high absorptivity) ﬂat plate absorber that collects the sunlight and trans-
fers the heat energy to the tubes. To reduce both convection and radiation losses, atransparent cover protects the absorber, while conduction losses are limited with an
efﬁcient thermal insulation of the collector casing. An example FPC architecture is
shown in Fig. 16.6 . Initially designed for domestic hot water generation, this technol-
ogy is well-suited for low-temperature applications (below 100
/C14C) and achieves good
thermal ef ﬁciencies thanks to the limited heat losses.
NCZ (gradient zone) LCZ (salt saturated)UCZ (nearly fresh)
Figure 16.5 Salt-gradient solar pond. LCZ, lower convective zone; NCZ, nonconvective zone;
UCZ , upper convective zone.
TubesAbsorberInsulationGlass cover
Figure 16.6 Flat plate collector.Solar thermal powered Organic Rankine Cycles 581

16.2.1.3 Evacuated tube collector
To decrease heat losses at high temperature or in unfavorable weather conditions, an
evacuated tube can be placed around the HCE of a solar collector. Evacuated tubecollectors (ETC) can be found in two different con ﬁgurations: in the ﬁrst, the HTF
ﬂows directly through the tube placed in the middle of the evacuated envelope while
in the second con ﬁguration, as depicted in Fig. 16.7 , a heat pipe is placed in the
vacuum-sealed tube and an auxiliary phase-change medium is used to transfer solar
energy to the main HTF by means of an evaporation econdensation cycle. More
precisely, solar radiation evaporates the auxiliary medium, after which the vapor phasenaturally migrates to a condenser situated at the top of the heat pipe. The heat energy isthen transferred to the main HTF, and the cooled auxiliary medium ﬂows down the
heat pipe in liquid phase. Although more complex, this mechanism permits collectionof solar energy at constant temperature and avoids issues with overheating or freezing.Like ﬂat plate collectors, ETCs exploit both direct and diffuse solar radiation, but
ETCs have better thermal performance than FPCs for similar operating conditions
and are suitable for a somewhat higher temperature range (up to 150
/C14C).
In many cases, however, it is desirable to operate solar collectors at temperatures
above even 150/C14C, and this requires use of a concentrating collector (using mirrors
or lenses to focus sunlight) to increase the ﬂux density at the HCE. Concentrating col-
lectors therefore have a larger aperture area than HCE area, and the ratio between thesetwo surfaces is known as the concentration ratio ( C
f). Depending on the technology,
the concentration ratio can be slightly larger than one or reach over one thousand.At a given operating temperature, the higher the concentration ratio, the higher the
thermal ef ﬁciency. The heat loss from an HCE of a given surface area and optical prop-
erties is proportional to the average temperature difference between the HCE and theambient environment. Therefore, for a given operating condition, the ef ﬁciency in-
creases with the concentration ratio C
fsince the losses are reduced in proportion to
the heat input. Solar collectors can be designed and operated to achieve high temper-atures for the heat source and thus a better thermodynamic ef ﬁciency of the ORC.
Note, however, that a tradeoff must be made for this ef ﬁciency gain, as higher
Main HTFCondenser
Evacuated tubeHeat pipe + auxiliary HTF
Figure 16.7 Evacuated tube collector.582 Organic Rankine Cycle (ORC) Power Systems

temperatures at the HCE also increase thermal losses of the collector. Unlike non-
concentrating technologies, concentrating collectors usually require continuoustracking to ensure an ef ﬁcient focus of incoming radiation onto the HCE, and their op-
tical imaging nature means they exploit only DNI and demonstrate poor performancein diffuse solar conditions. The most common concentrating technologies employed inSORC are presented in the following Sections 16.2.1.4 e8.
16.2.1.4 Parabolic trough collector
A parabolic trough collector (PTC) is a linear concentrating system made of long,
parabolic-shaped mirrors and a receiver tube placed along the focal axis of theparabola. DNI is concentrated onto the receiver tube (as illustrated in Fig. 16.8 ), where
solar energy is absorbed by the HTF. A glass envelope is often placed around the HCEto limit convection losses and further improve the collector ef ﬁciency; the annulus
space between the glass envelope and the receiver tube can be under vacuum. Com-mon PTCs achieve concentration ratios of 50, and the HTF temperature can reachup to 400
/C14C(Lovegrove and Stein, 2012 ). Parabolic troughs are highly modular
and can be arranged in solar ﬁelds of various sizes and architecture, however, to mini-
mize losses, the collector axis must be oriented either in an east-west or in a north-south direction, both of which require single-axis tracking. In the case of a smallersolar ﬁeld, dual-axis tracking can be used to reduce optical losses, however, this is rela-
tively uncommon for linear concentrators.
16.2.1.5 Linear Fresnel collector
A linear Fresnel collector (LFC) is a line-focused concentrating system that re ﬂects
DNI onto an elevated stationary receiver tube (see Fig. 16.9 ). Unlike PTCs which
consist of a continuous parabola-shaped re ﬂector, LFCs are composed of many long
strips of mirror that can be moved independently (single-axis tracking) to achievereﬂection of solar radiation onto the HCE. Due to its architecture, LFCs are highly
modular and can be constructed and assembled inexpensively in various-scale solar
Figure 16.8 Parabolic dish/trough collectors.Solar thermal powered Organic Rankine Cycles 583

ﬁelds. Concentration ratios of LFCs without secondary optics are slightly lower than
for PTCs, and LFCs are generally characterized by lower optical performance but po-tential cost savings.
16.2.1.6 Compound parabolic collector
A compound parabolic collector (CPC) (see Fig. 16.10 ) is another trough-type technol-
ogy that concentrates solar energy onto a tube receiver. The re ﬂector geometry is built
by the combination of two symmetric parabolic segments with different focal lengths.This geometrical arrangement enables collection of any solar radiation entering thecollector aperture within an acceptance angle (depending on geometry but rangingfrom 10 to 80 degrees) onto the tube receiver by means of multiple internal re ﬂections.
This important feature allows CPCs to operate without continuous tracking and to
exploit both DNI and some portions of diffuse sunlight. CPCs are characterized by
low concentration ratios ( <5) and are well-suited for medium-temperature applications
(up to 200
/C14C). CPCs also ﬁnd use as secondary down-facing re ﬂectors in LFC systems
to increase the effective concentration ratio.
Figure 16.9 Linear Fresnel collector.
Figure 16.10 Compound parabolic collector.584 Organic Rankine Cycle (ORC) Power Systems

16.2.1.7 Parabolic dish re ﬂector
A parabolic dish re ﬂector (PDR) is a point-focus system with a paraboloid geometry
given by the revolution of one half of a parabola around its normal axis. Sunlightentering the collector aperture with a nor mal incidence is concentrated onto a heat
receiver located at the focal point of th e dish. Parabolic dishes exploit only DNI
and require a two-axis tracking mechanism to ensure a proper focus throughoutthe day. Typical concentration ratios of PDRs range from 500 to 3000, making
this technology suitable for high t emperature applications (up to 450
/C14C for SORCs),
but unlike other collector types, PDRs are rarely connected together in a solar ﬁeld.
Instead, they often operate as distributed power systems with independent engineunits directly located at the focal point of e ach collector. Due to their high tempera-
ture capabilities, ORC is not often the cycle of choice for a PDR collector, withStirling and thermoacoustic examples being more common ( Kiceniuk, 1985; D ﬁd,
2010; Zhang et al., 2014 ).
16.2.1.8 Solar central tower
A solar central tower (SCT) is another t ype of point-focus system that uses a ﬁeld
of independently actuated mirrors (called h eliostats) to concentrate DNI onto a cen-
tral receiver at the top of a tower ( Fig. 16.11 ). Each heliostat is controlled with a 2D
tracking mechanism calibrated to its loca tion and physical placement relative to the
tower. Solar towers achieve high concentr ation ratios (up to 2000) and are most often
used in high-temperature large-scale st eam power plants, however, some pilot pro-
jects (see Table 16.2 ) also use this collector technology in ORC-based power
systems.
The selection of an appropriate solar collector technology for an ORC application
will involve speci ﬁc cost considerations, or the comparison of alternatives, along with
project speci ﬁcations including any limitations in mounting space, operating temper-
atures, and the availability of any reservoir for heat rejection (discussed further inSection 16.3.4 ). The main characteristics of the different collector technologies
presented in this section are summarized in Table 16.3 .
Figure 16.11 Solar tower system with heliostat array and central receiver.Solar thermal powered Organic Rankine Cycles 585

Table 16.3 Solar collectors used in solar Organic Rankine Cycles (SORCs) and their characteristics
(Lovegrove and Stein, 2012; Kalogirou, 2004; Weinrebe, 2007 )
Collector type SGSP FPC ETC CPC LFC PTC PDR SCT
Tracking None None None None/1D 1D 1D/2D 2D 2D
Concentration ratio 1 1 1 <52 0 e40 30 e50 500 e3000 400 e2000???
Solar exploitation Total Total Total Total Direct only Direct only Direct only Direct onlyReceiver temperature <100
/C14C<100/C14C 100 e150/C14C 100 e200/C14C 100 e300/C14C 100 e400/C14C>300/C14C >300/C14C
CPC , compound parabolic collector; ETC, evacuated tube collector; FPC,ﬂat plate collector; LFC, linear Fresnel collector; PDR , Parabolic dish re ﬂector; PTC, parabolic trough collector; SCT,
solar central tower; SGSP , salt-gradient solar pond.586 Organic Rankine Cycle (ORC) Power Systems

16.2.2 Heat transfer ﬂuid
In general, a ﬂowing heat carrier ﬂuid (HTF) is used to facilitate transfer of energy
collected in the solar ﬁeld to the ORC unit.
A simpler option is to directly use the working ﬂuid of the organic Rankine cycle as
the heat transport medium ( Fig. 16.12 left side), referred to as direct steam generation
(DSG), vaporizing the organic ﬂuid within the solar collectors. This con ﬁguration
avoids the cost of a heat delivery heat exchanger and the parasitic losses associated
with a pump circulating a second ﬂuid, and the simple architecture makes it convenient
for use in micro-scale systems ( Lovegrove and Stein, 2012 ). DSG is rarely utilized in
medium- and large-scale SORC systems due to the large (costly) volume of organicﬂuid required to ﬁll the solar ﬁeld; DSG implies high-pressure operations in the solar
collectors which requires a more expensive collector design and may actually increasethe overall cost of the project.
The second, more common, SORC architecture employs an intermediate HTF to
transport collected heat energy from the solar collectors to the ORC system. In this
case, a heat exchanger between the solar loop and the ORC is used to evaporate the
working ﬂuid (heat delivery similar to that used in more traditional ORC applications),
and a secondary pump circulates the HTF in the solar collectors.
The selection of an ef ﬁcient and cost-effective HTF is one of the important design
parameters affecting the overall power plant performance. The main solar plant HTFrequirements are summarized here:
1.Low melting temperature, to avoid solidi ﬁcation and an adequate pumpability of the HTF
during cold weather;
2.High temperature stability, to ensure the HTF integrity at high-temperature operating
conditions;
3.High thermal conductivity, to maximize the heat transfer rate in the solar collectors;
4.High thermal capacity, to limit the mass ﬂow rate required for transporting a given amount of
heat power;
5.Low coef ﬁcient of expansion, to limit the volume variations of the HTF between different
operating conditions;
6.Low viscosity, to reduce power consumption of circulating pumps;
Figure 16.12 Direct steam generation (DSG) con ﬁguration (left). Heat transfer ﬂuid (HTF)
conﬁguration (right).Solar thermal powered Organic Rankine Cycles 587

7.Low corrosion, to ensure the long-term integrity of the power plant;
8.Limited hazard and environmental issues, i.e., low vapor pressure, no toxicity, limited
ﬂammability, etc.;
9.Low cost and availability within the project target market.
A common, effective, and inexpensive heat carrier used in low-medium tempera-
ture SORC is pressurized water. Aside from low cost, wide availability, and theabsence of environmental issues, water has good heat transfer properties with a lowviscosity and a high thermal capacity. A major drawback of water, however, is the
sharp increase of pressure required to ensure a liquid-phase ﬂow at high temperature,
as shown in Fig. 16.13 . Pressures in water-based solar ﬁelds easily reach 20 bar, mak-
ing the installation more hazardous (from a mechanical safety perspective) and moreexpensive. Working HTF pressure can become a limiting factor in solar ﬁeld operating
temperature, which limits in turn the maximum temperature of the working ﬂuid inside
the ORC.
Low-pressure operations in the solar ﬁeld can be preserved by admixture of water
with soluble ﬂuids that lower the boiling point, such as the glycol diols, or using alter-
native ﬂuids such as thermal oils or other organic or inorganic materials. These HTFs
can be vegetal (e.g., palm oil), synthetic (i.e., arti ﬁcially-made), mineral (i.e.,
petroleum-derived), or lique ﬁed salts; HTFs of these types have been developed for
use in a wide range of temperature ranges and conditions (see Table 16.4 ). It is impor-
tant to recall that HTF properties are generally temperature dependent; Table 16.4 pro-
vides one point of comparison among ﬂuids (at 150
/C14C) and the coef ﬁcients required to
calculate the density and the speci ﬁc heat capacity of the ﬂuids in function of the
temperature, i.e.,
rT¼r0țbrTc p T¼cp0țbcpT (16.2)
0 50 150 250 100 200 3000102030405060708090Saturated pressure (bar)
Temperature (șC)
Figure 16.13 Saturated pressure of water as a function of the temperature.588 Organic Rankine Cycle (ORC) Power Systems

Table 16.4 Nonexhaustive list of range of commercial heat transfer ﬂuids (HTFs)
Fluid Tmin Tmax Tauto,ig r150 8C cp150 8C k150 8C m150 8Cr0 br cp0 bcp
(8C) ( 8C) ( 8C) (kg/m3) (J/kg K) (W/m K) (cP) (kg/m3) (kg/m38C) (J/kg K) (J/kg K 8C)
Water (20 bar) ee e 918 4302 0.682 0.18 1024 L0.725 4099 1.49
Xceltherm 600 /C029 316 349 773 2440 0.126 1.07 864 /C00.607 1927 3.41
Xceltherm MK1 12 400 621 957 1916 0.121 0.59 1093 /C00.942 1495 2.77
Xceltherm LV1 7 371 604 955 1944 0.120 0.70 1080 /C00.835 1519 2.79
Dynalene MT e 350 410 862 2083 0.118 1.10 972 /C00.732 1772 1.72
Dynalene HT /C034 350 450 951 2033 0.113 1.51 1058 /C00.715 1474 3.73
Dynalene SF /C060 315 330 789 2441 0.124 1.66 890 /C00.672 1906 3.60
Dynalene SGXT /C027 176 n.a. 931 7863 0.433 7.51 1054 /C00.640 3445 5.83
Dynalene 600 /C065 288 n.a. 833 1517 0.130 17.00 983 /C01.000 1234 1.88
Therminol 66 /C03 345 399 921 2014 0.110 1.52 1028 /C00.735 1475 3.65
Therminol XP /C029 315 324 795 2394 0.117 1.48 893 /C00.668 1747 4.13
Therminol 59 /C061 315 404 878 2110 0.110 0.74 990 /C00.771 1614 3.32
Therminol 55 /C054 315 343 783 2364 0.113 1.29 887 /C00.705 1836 3.53
Therminol SP /C040 315 366 783 2368 0.113 1.29 888 /C00.716 1836 3.54
Therminol 62 /C042 325 407 859 2252 0.111 1.06 973 /C00.794 1907 2.15
Therminol VP1 12 400 621 957 1913 0.121 0.59 1098 /C00.972 1486 2.82
Dowtherm A 12 400 599 952 1940 0.118 0.58 1096 /C00.995 1501 2.93
Dowtherm G 4 360 432 946 2000 0.111 0.96 1062 /C00.775 1476 3.50
Dowtherm Q /C035 330 412 867 2058 0.104 0.45 981 /C00.758 1596 3.03
ContinuedSolar thermal powered Organic Rankine Cycles 589

Table 16.4 Continued
Fluid Tmin Tmax Tauto,ig r150 8C cp150 8C k150 8C m150 8Cr0 br cp0 bcp
(8C) ( 8C) ( 8C) (kg/m3) (J/kg K) (W/m K) (cP) (kg/m3) (kg/m38C) (J/kg K) (J/kg K 8C)
Water (20 bar) ee e 918 4302 0.682 0.18 1024 L0.725 4099 1.49
Dowtherm RP /C020 350 385 937 2007 0.115 1.32 1046 /C00.738 1561 2.98
Dowtherm MX /C023 330 420 868 2032 0.109 0.96 981 /C00.775 1545 3.25
Dowtherm 4000
(ethylene
glycold90% vol)/C030 177 e 1031 3138 0.269 0.62 1149 /C00.753 2243 5.96
Syltherm 800 /C060 400 385 820 1830 0.111 1.70 960 /C00.978 1574 1.71
Syltherm XLT /C0111 260 350 722 2045 0.080 0.34 876 /C01.027 1730 2.10
Syltherm HF /C082 260 355 740 2002 0.075 0.43 892 /C01.011 1633 2.46
Paratherm HR /C011 343 416 860 2200 0.107 0.93 965 /C00.746 1884 2.30
Paratherm HE 3 310 371 781 2600 0.116 2.10 872 /C00.611 1791 5.06
Paratherm GLT 8 288 n.a. 781 2500 0.110 1.30 887 /C00.707 1929 3.60
Paratherm NF /C043 332 366 797 2500 0.098 1.50 899 /C00.683 1469 7.97
Pirobloc mineral /C010 305 n.a. 782 2374 0.131 n.a. 879 /C00.649 1811 3.72
Duratherm 450 /C045 232 329 774 2473 0.133 0.77 876 /C00.681 2021 3.02
Duratherm 600 /C010 315 360 768 2336 0.136 2.01 869 /C00.672 1844 3.27
Duratherm HTO /C015 315 360 758 2288 0.135 2.07 862 /C00.665 1815 3.24
Duratherm S /C066 204 436 928 1926 0.117 0.98 968 /C00.263 1641 1.91
Solar salt 220 600 ee e e e 2090 /C00.636 1443 0.172
Hitec salt 142 535 ee e e 18 2079 /C00.732 ee590 Organic Rankine Cycle (ORC) Power Systems

where brandbcpare the linear coef ﬁcients of the density and the speci ﬁc heat capacity
correlations respectively. From Table 16.4 , it can be seen that organic heat carriers
have lower thermal capacities, higher viscosities, and lower thermal conductivitiesthan pressurized water. They are limited at high temperature because of stability issuesdue to their molecular structures, however, even at high temperature, they can oftenbe operated at or near atmospheric pressure, creating cost savings and avoiding themechanical hazards of high-pressure systems. Molten salts present a good thermal
stability at high temperature and properties similar to water (high thermal capacity and
conductivity, and low viscosity) at low vapor pressure. However, they are character-ized by high-temperature melting-points ( >150
/C14C) which make them, in general, not
suitable for ORC-based CSP systems.
Finally, besides thermophysical properties, the cost of the heat transfer carrier must
be accounted in the design process of a solar power system. Price is function of theproduct quality, the seller, but also of the volume container. In the case of 200 Ldrums, cheap and low-grade HTFs are about 1 e3V/L, standard HTFs cost around
4e6V/L, whereas high-grade ﬂuids can cost up to 10 e15V/L.
16.2.3 Thermal energy storage
The intermittent nature of sunlight is an inherent drawback of solar power that can lead
to imbalances between consumer demand and heat source availability. By adding
energy storage, however, it is possible to shift excess energy from high-insolation
periods to nighttime or periods of unfavorable meteorological conditions. The wholepower system is thus more ef ﬁcient (avoiding waste during periods where insolation
outstrips demand), reliable (buffering system output during cloud passage), and ﬂex-
ible (higher capacity factor) despite transient external conditions. A key advantage ofsolar ORCs over photovoltaic technologies is the opportunity to use cost-effectivethermal energy storage (TES), which stores excess heat energy in thermal ratherthan electrical form, in place of electrochemical batteries. TES, which is currently
cheaper than commercial batteries and has a longer usable lifetime, comes in many va-
rieties, and design/selection of the TES component of a SORC system must be under-taken with the following criteria in mind:
1.High energy density minimizes storage size;
2.High heat transfer rate permits fast storage and release of thermal energy;
3.Chemical compatibility is required between the storage medium, the container, and the HTF
(if in direct contact);
4.High storage ef ﬁciency is needed over the complete charge estandby edischarge cycle;
energy and exergy losses should be minimized;
5.Easy control and good ﬂexibility;
6.Fluid selections should avoid environmental issues and limit hazards;
7.Lowest cost systems amplify advantage over electrochemical batteries.
Various TES technologies can be distinguished by the mechanism employed for
storing heat energy, i.e. sensible, latent and chemical TES systems. These TES tech-nologies are discussed in more detail in the following sections.Solar thermal powered Organic Rankine Cycles 591

16.2.3.1 Sensible thermal energy storage
In sensible TES, heat storage is achieved by raising the temperature (without changing
phase) of a single-phase medium. The amount of energy stored is proportional to themass of the medium employed, its speci ﬁc heat capacity, and the temperature differ-
ence between initial and ﬁnal states. The storage material can either be the HTF ﬂow-
ing in the solar ﬁeld (i.e., direct storage) or another medium (i.e., indirect storage). In
the latter case, an additional heat exchanger is required to interface the SORC plant to
the TES unit. Cost reductions for either direct or indirect storage can further be
achieved by using inexpensive solid materials, preferably with high heat capacity, topartially ﬁll the storage container (i.e., packed-bed TES). A list of common ﬁller ma-
terials is given in Table 16.5 .
Sensible TES can be integrated into SORC systems in three distinct architectures:
Single buffer
In this con ﬁguration a small volume capacity of HTF is placed in series with the solar
loop (see Fig. 16.14 ). The buffer mitigates short ﬂuctuations of temperatures at the
outlet of the solar ﬁeld (e.g., due to passing cloud cover). The storage capacity of
the buffer tank is limited and is generally not used to extend operation of the SORC
into nonsun periods.
Two-tank storage
Two-tank storage involves two separate reservoirs, which are used to store, respec-tively, hot and cold liquids ( Fig. 16.15 ) in various system con ﬁgurations. Two-tank
Table 16.5 Properties of potential solid ﬁller materials for packed-bed
storage ( Kuravi et al., 2013; Singh et al., 2010; Grirate et al., 2013 )
Filler materialDensity Heat capacity Conductivity Thermal expansion
(kg/m3) (J/kg K) (W/m K) (1e-5/K)
Concrete 2000 e2400 750 e900 0.4 e1.5 2.5 e4.5
Brick 1600 e2000 850 e1000 0.6 e1.5 0.9 e1.7
Basalt 2200 e2800 800 e1150 1.0 e2.5 2.4 e2.8
Granite 2500 e2700 700 e850 1.7 e4.0 1.2 e2.4
Limestone 2500 e2800 830 e1000 1.3 e2.5 2.4 e3.6
Marble 2600 e2800 800 e1150 2.0 e3.0 1.7 e3.6
Quartzite 2510 e2860 700 e1100 3.3 e7.0 3.3 e3.9
Sandstone 2100 e2700 710 e930 1.7 e2.9 3.0 e3.3
Aluminium 2700 e2800 870 e890 205 e215 6.5 e6.9
Cast iron 7200 e7900 460 e600 37 e55 3.3 e3.5
Steel 7750 e7830 465 e490 36 e54 3.1 e4.2592 Organic Rankine Cycle (ORC) Power Systems

storage can have high thermal charge edischarge ef ﬁciencies (up to 95% if well insu-
lated) and good operational ﬂexibility. They are easily scalable and can be designed to
widely extend the daily operational time of SORC by up to several hours.
Thermocline storage
As shown in Fig. 16.16 , thermocline storage is a single tank used to store both hot and
coldﬂuid where a natural separation is created by the density difference between the
two zones of ﬂuid. By using a single reservoir, cost reductions (up to 33% ( Kolb,
2011 )) can be achieved compared to two-tank storage, however, thermocline systems
Figure 16.14 Buffer reservoir.
Figure 16.15 Direct (left) and indirect (right) two-tank thermal energy storage.
Figure 16.16 Direct (left) and indirect (right) thermocline storage.Solar thermal powered Organic Rankine Cycles 593

are inherently less ef ﬁcient than two-tank TES because of the imperfect separation
between the hot and cold ﬂuid. Thermal diffusion and mixing phenomena (e.g. turbu-
lence due to ﬂow) induce a transition zone of increasing thickness which degrades the
high-temperature energy initially stored. In spite of this, thermocline TES is easilyscalable and can be designed to extend the operational window of SORCs for manyhours past sunset.
Overall, sensible heat storage is the most common TES technology used in SORC
systems as it is simple, cost-effective, and has good heat transfer performance. How-
ever, the lower energy density ( wkJ/m
3) implies large storage size compared to TES
varieties discussed in Section 16.2.3.2 .
16.2.3.2 Latent thermal energy storage
In this design, thermal energy is stored in the latent heat of a phase-change material
(PCM). Since latent energy of phase-change transitions is much higher than speci ﬁc
heat capacity, PCM storage has the potential for much higher energy storage density
(wMJ/m3) than sensible TES. Among the different phase-change transitions possible,
solideliquid transitions are generally preferred because of the limited volumetric
expansion of the storage medium during the process. Generally encapsulation of thestorage media is required to avoid mixing between the liquid-phase PCMs andthe HTF.
Another key advantage of latent storage is that the phase transition is a quasi-
isothermal process, a feature that facilitates TES integration and control within theSORC plant. The tradeoff, however, is the low thermal conductivity of the storage me-
dia which results in an excessive charging and discharging time for thermal storage.
Faster response times can be achieved by adding high-conductivity materials (e.g.,metals or graphite) within the PCM.
Media for latent storage include organic materials (e.g., paraf ﬁns and fatty acids),
inorganic compounds (e.g., salt hydrates), and various eutectic mixtures (seeTable 16.6 ). Despite its technical advantages compared to sensible technologies, latent
heat storage is more expensive and still at the prototype phase.
16.2.3.3 Thermochemical thermal energy storage
Thermochemical TES relies on reversible chemical reactions to store heat energy.In the charging process, injected heat is used to drive an endothermic chemical reac-tion; the chemical products are later used to restore thermal energy by performing thereverse (exothermic) reaction. Examples of potential materials for thermochemicalstorage in CSP are provided in Table 16.7 . Among the different storage technologies,
thermochemical TES has the highest energy density potential ( wGJ/m
3) thanks to the
large enthalpy change in chemical reactions. Additional advantages to thermochemical
TES are quasi-isothermal storage and release, as well as storage of chemical products
at ambient temperatures which largely reduces long-term thermal losses. Thermo-chemical storage is not yet mature, however, with high cost and technical complexityremaining barriers to commercialization.594 Organic Rankine Cycle (ORC) Power Systems

Table 16.6 Examples of phase-change materials with their properties
Chemical
formula Medium nameMelting
point DensityLatent
energyEnergy
density
(8C) (kg/m3) (kJ/kg) (MJ/m3)
C8H8Cl2 p-Xylene
dichloride100 1432 139 199
C8H8O4 Methyl
fumarate102 1370 242 332
C8H4(OH) 2 Catechol 104.3 1344 207 278
C6H4O2 Quinone 115 1318 171 225
C8H9NO Acetanilide 118.9 1219 222 271
C4H4O3 Succinic
anhydride119 1230 204 251
C6H5COOH Benzoic acid 121.7 1266 143 181
C14H12 Stibene 124 971 167 162
C6H5CONH 2 Benzamide 127.2 1341 169 227
KNO 3țNaNO 3- 222 2257 108 244
NaNO 2 Sodium nitrite 282 2168 212 460
NaNO 3 Sodium nitrate 310 2257 174 393
NaOH Sodium
hydroxide318 2130 158 337
KNO 3 Potassium
nitrate337 2109 116 245
KOH Potassium
hydroxide360 2100 167 351
Table 16.7 Examples of reactions for thermochemical storage
Chemical reactionTemperature
range ( 8C)Energy density
(GJ/m3)
FeCO 34FeOțCO 2 180e200 2.6
CH 3OH4COț2H2 200e250 e
Ca(OH) 24CaOțH2O 400 e600 3
CaCO 34CaOțCO 2 800e900 4.4
6Mn 2O344Mn 3O4țO2 900e1000 1Solar thermal powered Organic Rankine Cycles 595

16.3 Solar Organic Rankine Cycle systems
16.3.1 Design and speci ﬁcation
While there may be differences in the operating temperatures, working ﬂuids, sizes
and types of pumps, expanders, and heat exchangers for an SORC system, thefundamental methods for predicting the perf ormance of a solar-integrated ORC sys-
tem are similar to the thermodynamic mode ling techniques described in Chapters 1,
6, and 7. The selection of cycle paramete rs optimized for a solar ORC, however, is
driven mainly by the cost and temperature of the available heat source and coldreservoir (air in dry or wet cooling modes, or a local water body). In particular,the optimum operating temperature of a solar-driven ORC can vary with the type
of collector (described in Section 16.2.1 ) and its relative ef ﬁciency as a function
of temperature (ef ﬁciency decreases with increasing HCE temperature). This is a
fundamental aspect of solar-integrated ORC systems; there is an opposing trend
between ef ﬁciency and operating temperature for any solar collector (negative
derivative) as compared to a thermodynamic cycle [positive derivative (ef ﬁciency
increases with increasing expander inlet te mperature)]. This implies that for any
collector architecture, and taking into cons ideration the cooling strategy at the loca-
tion of the plant, there is an optimum ope rating temperature where the tradeoff
between collector and ORC ef ﬁciency is balanced to achieve the best overall plant
efﬁciency.
16.3.2 Steady state performance prediction of solar ORC
systems
A steady state model of a solar ORC describes an instantaneous snapshot of the state of
the system. On a clear day sunlight impinges on a solar collector, photons are con-verted to thermal energy at an ef ﬁciency calculated from optical and material proper-
ties, thermal energy is then conveyed by an HTF at a prescribed ﬂow rate to the
evaporator of an ORC, where a working ﬂuid reaches a two-phase, saturated, or super-
heated state at high pressure, and mechanical work is performed when the working
ﬂuid transits through the expander. Finally the high-volume, low-pressure working
ﬂuid is condensed, heat is rejected (to ambient air, water, or a secondary thermal
loop in, for example, cogeneration), and power output from the expander/generatoris divided by the amount of energy captured from the sun to calculate the ef ﬁciency.
System performance is analyzed assuming that all environmental variables (solar inso-lation, ambient temperature, etc.) remain “constant enough ”over a “long enough ”
period of time for their effects to propagate through the system (note that the deviation
of actual conditions from these assumptions can be quanti ﬁed to determine the extent
of validity of steady state results). This conceptual framework provides the basis forsimple models, discussed in this section, and even for complex models (e.g., dynamicmodels, as described in Section 16.3.3 ).596 Organic Rankine Cycle (ORC) Power Systems

16.3.2.1 Solar collector modeling
To determine the performance of a solar collector (and therefore quantify the HTF
temperature and energy content as supplied to the ORC) it is necessary to quantifythe energy input:
Energy in ¼DNI/C3aperture of the collector /C3cos/C28 (16.3)
(where /C28is the angle of incidence) and a few other relevant parameters which affect the
heat loss terms of the model, such as the temperature and mass ﬂow rate of the HTF
entering the collector, the ambient temperature and the wind speed. A typical solarcollector model will calculate the fraction of sunlight energy lost via the followingoptical mechanisms:
Cosine losses (as the sunlight impinges on the aperture at an angle of incidence theta, the
actual power intercepted by the collectors is cos f, and this can lead to further losses
through “walk off ”of the beam past the receiver end at the collector side furthest from
the sun)
absorption or scattering at the primary or secondary mirrors (re ﬂective coef ﬁcient)
missing the absorber target (intercept factor)
shadowing of aperture area (by collector structures or nearby obstructions)
absorption or scattering at any glass envelope
reﬂection off of the absorber surface (inverse of absorbtivity)
After these losses are accounted for, the remaining energy is assumed to be
absorbed as heat by the system, however, this heat does not entirely result in enthalpygain of the HTF due to the inevitability of heat losses from the absorber. In typical so-lar collector models, such as the one developed by Forristall (2003) , an energy balance
is calculated on the HCE, taking into account its optical parameters (emissivity of
absorber tube and glass glazing, if there is glazing), convection transfer at the HCE
outer surface to air (as a function of ambient temperature and wind speed) and withinany annulus (unless it is evacuated), and radiative losses propagating out from theabsorber through any glazing. This involves a solution of the radial temperature distri-bution across the HCE ( Fig. 16.17 ), taking into consideration conduction through any
HCE materials and the heat transfer from the inner surface of the HCE to the HTF (as afunction of the thermophysical properties of the HTF, e.g., thermal conductivity, heatcapacity, viscosity, and velocity). Alternatively, empirical correlations can be used to
derive the effective heat power absorbed by the ﬂuid in function of the operating
conditions ( Dickes et al., 2015 ).
Such 1D energy balances may be performed on subsections of the HCE line and
solved sequentially, where the output HTF temperature of the ﬁrst node is used as
the input HTF temperature for the second node, and so forth ( Fig. 16.18 ). This method
reproduces a temperature gradient along the absorber line, aggregates convection andradiation losses along the line, and estimates the outlet temperature and heat gain of theHTF, which is the relevant input parameter for the ORC model.Solar thermal powered Organic Rankine Cycles 597

Note that the heat gain of the HTF (in Watts) divided by the original input term at
the collector aperture, is the de ﬁnition of solar collector ef ﬁciency:
hcol¼_mDh
DNI$A$cosf(16.4)
where fis the angle of incidence of sunlight on the collector aperture.
Once the outlet temperature of the collector is known, it can be input into a detailed
ORC model (iteratively, if the HTF inlet temperature of the solar loop is to be deter-mined by the outlet temperature of the ORC vaporizer or preheater) or substitutedinto a simpli ﬁed representation of a heat engine such as, e.g., a Chambadal-Novikov
(De Parga, 2009 ) alternative to the Carnot ef ﬁciency approximating the operating
experience of actual heat engines:
h
ORC¼1/C0ﬃﬃﬃﬃﬃ
Tc
Thr
(16.5)
where T cand T hare the hot source and cold sink temperatures, respectively.
Glass envelopeTube receiverHTFq5SolAbs.q3SolAbs.
q56,conv.
q34,conv.
q12,conv.q57,rad.
q45,cond.q23,cond.q34,rad.
Figure 16.17 Radial temperature distribution across heat collection element boundaries
beginning with the interface between the heat transfer ﬂuid (HTF) and the inner diameter of the
absorber (1-2), conduction across the absorber (2-3), radiation and convection from the outer
diameter of the absorber to the inner diameter of the glass envelope (3-4), conduction across
the glass envelope (4-5), and ﬁnally convection and radiation from the outer diameter of the
glass envelope to the ambient air (5-6) and the sky (5-7) respectively.
1 2 3 i–1 N–1 N i+1i
Figure 16.18 A 1D solution for energy balance is applied to sequential nodes with outlet
temperatures of the nth node becoming the inlet temperature of the nț1 node.598 Organic Rankine Cycle (ORC) Power Systems

The overall system ef ﬁciency is simply a product of these two component ef ﬁ-
ciencies (sunlight-to-heat and heat-to-power).
The drawback to this steady state approach lies in the assumption of steady or
average DNI and average air temperature, whereas in reality these two driving param-eters are constantly varying during normal operation of the ORC and “off-design ”ef-
ﬁciencies may differ substantially from the average ef ﬁciency. At a minimum, the
ﬂuctuation in heat gain due to these changes frequently necessitates the use of thermal
buffering, if not outright TES, to stabilize the inlet conditions at the ORC expander (as
discussed previously). In order to consider the effects of thermal inertia (of the ORC,the collectors, or any buffering or storage components), it is necessary to make use of adynamic modeling approach.
16.3.3 Dynamic performance prediction of solar organic
Rankine cycles with thermal energy storage using typical
meteorological year data
A thermal power plant operating from a constant heat source (such as a metered feed-
stock of combusting coal) is a good candidate for steady state calculations to determine
typical performance parameters and speci ﬁcations of major components, such as ex-
panders and heat exchangers; in general, the system will always function at or veryclose to the steady state design point. A solar power plant, however, especially onethat is expected to handle any load-following (such as an islanded solar microgrid),represents the opposite extreme; both energy delivery and demand will vary potentiallyquite far from the average values, and this points directly to the inadequacies of asteady state model.
For a solar ORC, the availability and inte nsity of sunlight is intermittent, and
usually not synchronized with the variab ility in demand. If the operating tempera-
ture differential between source and sink i s relatively small, as it tends to be with a
solar ORC, ﬂuctuations in ambient temperature s (and their dynamic with respect to
heat availability) play a co rrespondingly larger role in the performance of the
system.
Typical meteorological year or TMY datasets have been developed to stan-
dardize the characterization of such local ﬂuctuations in ambient temperature and
irradiance. These consist of historical weather observations for a set of 12 ‘typical ’
months, usually derived from a multiyear da taset, with hourly values. This type of
data provides the baseline design cond itions for engineerin g a system that can
handle dynamics.
A parametric steady state approach to characterizing thermal power generation
(solving steady state models for differen t ambient conditions) can lead to an under-
standing of off-design behavior, but only a dynamic model can lead to an under-
standing of which nominal design , given the certitude of continual off-design
conditions, optimizes for the ﬁgure of merit (typically minimized total cost per
unit of energy delivered). The following sections explain the motivation and ap-
proaches used to develop dynamic models to solve optimization problems for solar
ORC systems.Solar thermal powered Organic Rankine Cycles 599

16.3.3.1 Conservation of energy and thermal capacitance
Energy balance, or conservation of energy, is the principal framework for modeling
power systems. At steady state, the sum of inputs and outputs is equal to zero for apower block. When energy inputs and outputs of a power system do not match forevery point in time, the designer can smooth these discrepancies by introducing energystorage. Thus, a solar collector array can operate at its full output, without defocusingcollectors, even when this heat ﬂux exceeds the heat ﬂux consumed by the ORC, or an
ORC can operate at full output during a brief period of cloudy weather. The difference
is simply enthalpy gain (or loss) in the TES component. Even if the available (sensible)TES temperature is declining (e.g., during operation after dusk), varying the HTF ﬂow
rate can maintain constant heat ﬂux and evaporating temperature in the ORC as long as
the temperature remains above an operating threshold.
In its most simple form, a dynamic solar ORC model extends the energy balance
framework by adding mathematical terms for thermal capacitance. TES is generallythe largest capacitance in the system, and its relation to other system components is
inherently dynamic, as its role in the power system involves either charging or dis-
charging energy while varying its internal energy (in practice, any TES is constantlydischarging through heat leakage unless it reaches equilibrium with the surroundings).These transfers of energy happen in time, and consideration of the time discretizationof a dynamic model is critical to avoid numerical diffusion type errors.
16.3.3.2 Selection of model timestep
Intuitively, if the residence time of a TES tank is 30 min, a 1 h timestep will fail toconserve and balance energy ﬂows; energy storage calculations based only on initial
andﬁnal temperatures will fail to account for the increase in temperature of the
HTF volume that ﬂowed into, and then out of, the TES tank during the timestep. If
a solar collector ﬁeld with a network of HCEs is used with an ORC, the change in solar
collector outlet temperature for a step change in HTF inlet temperature will only man-ifest after the travel time in the collector (and will reach steady state, if nothing elsechanges, only after the thermal capacity of the HCEs have equilibrated). If the timestep
is shorter than the travel time in a steady state solar collector model, energy conserva-
tion residuals will ensue, because within the timestep the gradient of temperatures frominlet to outlet is resolved, when in reality the collectors did not have enough time topropagate the change in inlet temperature all the way through to the outlet. Thistype of numerical diffusion error is framed by the Courant-Friedrichs-Lewy (CFL)condition, which holds that, in numerical simulations using ﬁnite difference methods,
the time discretization must contain the analytical phenomenological domain in ﬂu-
enced by the initial conditions ( Courant et al., 1967 ). Selection of an appropriate time-
step as a function of the spatial discretization of components in Solar ORC dynamic
simulation meeting the CFL condition can thus include:
thermal capacity ðJȚ
thermal power ðJ=sȚ>timestep <LHCEðmȚ
vHTFðm=sȚ(16.6)600 Organic Rankine Cycle (ORC) Power Systems

(the second boundary above the timestep could be eliminated by the use of a dynamic
solar collector model, similar to the approach taken with a TES tank). Other condi-tions, for example, volumes and ﬂow rates in tanks, may be considered as well, but in
general the consideration of dynamic simulation timestep involves the tradeoff be-tween impractical computational effort at in ﬁnitesimal timesteps and numerical failure
to converge at timesteps too large to meet the CFL condition.
16.3.3.3 Control strategy
The dynamic simulation of a solar ORC requires de ﬁnition of control parameters and
strategies (de ﬁning initial and operational modes), and presents an opportunity to opti-
mize these in software before implementing them in practice. In a dynamic solar ORCsimulation, the various components, including TES, are connected to each other viamass and energy ﬂows, which may in turn depend on external conditions. Once the
operational modes have been de ﬁned (e.g., by system temperatures or pressures, or
by external conditions), conditional statements can be used with parameter thresholdvalues to trigger selection between modes based on conditions for the given timestep.
The major controllable variables in a solar ORC are the ﬂow rates of the HTF
and the working ﬂuid, the speed of the cooling fan/pump, the tracking of the collectors
(on/off), and the setpoint of the expander rotational speed. Various scenarios withdifferent operating setpoints for these components can be articulated, with real timeacquisition of sensor data triggering the shift from one operational mode to anotherin the operation of a real system (and conditional statements embodying this behaviorin the dynamic simulation).
The number of unique modes and parameter combinations triggering mode shifts
are too extensive and application speci ﬁc to describe comprehensively, but a few
examples are given here for illustration.
Typical startup sequence
When DNI reaches a certain threshold (e.g., 200 W/m2), solar collectors are set to tracking
mode
When temperature in the absorber of the solar collector reaches a certain threshold (depend-
ing on the application) HTF ﬂow is initiated
When temperature of the HTF entering the ORC vaporizer (potentially supplied from TES)
reaches a certain threshold, the ORC is engaged, i.e., working ﬂuid pump (and potentially
expander variable frequency drive) and condenser cooling fan/pump are activated
Example operational considerations
To maintain the design pressure ratio across the expander, the HTF ﬂow rate and condenser
heat rejection rate may be modulated on the basis of, for example, degrees of working ﬂuid
superheat at the expander inlet or temperature pinch in the condenser
In the event of a cloud passing or dusk (DNI lower threshold being reached), the HTF ﬂow in
the collector ﬁeld may be stopped to avoid rejecting heat at the CSP HCE, while HTF ﬂow in
the TES eORC loop may be maintained for continuous power output (provided suf ﬁcient
HTF temperature, above a set threshold, is available from the TES).Solar thermal powered Organic Rankine Cycles 601

Operational systems may be shut down normally (according to a reverse of the
above sequence) or due to some perturbation such as a cloud passing, low temperaturein the HTF loop, or error states such as locked rotors, cavitation in the working ﬂuid
pump, excessively high temperature, etc. One specialized case of exceptional shut-down that can be easily diagnosed is frequent defocusing of a solar ﬁeld due to
overtemperature conditions; generally this points to an undersizing of the TES and/or the ORC components relative to the solar ﬁeld in the system design.
The various thresholds deployed in the control system and the modes associated
with various combinations of sensor inputs are themselves also subject to dynamicoptimization. For example, Ireland et al. (2014) performed a parametric sweep of
HTF temperature thresholds for engaging an ORC and found that system performance(in dynamic simulation of a solar ORC with TES) is highest when the ORC is onlyturned on when approaching the maximum allowable HTF temperature, as opposedto the minimum viable HTF temperature or some intermediate value ( Ireland et al.,
2014 ).Mitterhofer and Orosz (2015) conducted a parametric sweep of condenser
size, temperature defect, and pressure ratios in a dynamic simulation of a pilot
3k W
esolar ORC (installed by STG International at Eckerd College in St. Petersburg,
FL) to identify optimum expander volume ratio, condenser size, temperature defect,and coolant fan effort setpoints for variable ambient temperatures, using TMY dataand levelized cost as an objective function ( Mitterhofer and Orosz, 2015 ). Observa-
tions from this study included the cost advantages of using smaller air cooledcondensers with a higher temperature defect in the ORC cycle; the ef ﬁciency gains
from a large condenser are outweighed by the relative expense of this component.
Because of the speci ﬁcity of ambient climate conditions and project purpose and
sizes, results from such studies, while informative, are typically not generalizable,and detailed analysis are required to identify optimized control strategies, operationalmodes, and threshold variables for each application. Furthermore, even for a giventechnology design, control strategies and threshold variables must be validated inde-pendently for the particular load pro ﬁle and meteorological conditions experienced
at each new deployment location.
16.3.4 Solar Organic Rankine Cycle optimization and economics
A physics-based representation of a solar ORC with thermal capacitance is essentialfor accurate prediction of performance and sizing of components. While perfor-mance and ef ﬁciency are in ﬂuential metrics, several other factors may bear on
the design of the system, as conceived in the context of a real-world application.The decision to deploy solar collector s with an ORC (and potentially TES) and
determining appropriate sizing amo ngst the components, will usually be made
only after evaluating speci ﬁc project criteria; cost effectiveness compared to alter-
native options usually being foremost am ong them. The analysis of performance
and cost of various con ﬁgurations necessarily entails design engineering of the
system for the particular application which, to the extent that applications varywidely and are sited across a range of locat ions with unique irra diance and climate
characteristics, poses both a technical and economic challenge (cost of engineering602 Organic Rankine Cycle (ORC) Power Systems

hours, time required to complete designs and dynamic simulations, etc.). These
challenges have undoubtedly limited the d eployment of solar ORCs historically.
16.3.5 Application engineering and system analysis
Many variables impact solar ORC project design, and it can be dif ﬁcult to know where
to start because the most important factors may be different in each individual case.The following list of considerations is neither comprehensive nor necessarily in orderof priority but can provide a starting point for conducting a prefeasibility analysis.
Before engaging in a solar ORC project, relevant information to gather can include:
What is the location of the application? With this known, the closest station with TMY data
can be used to characterize the available irradiance and ambient temperature patterns.
What is the shape and variability of the load pro ﬁle? What is the peak load (kW), total daily
load (kWh), and variability in load throughout the day? What is the form of the load to be
supplied (i.e., are there both thermal and electrical loads)? Is a pro ﬁle of the demand well
constrained in the time domain across major operating modes? Is thermal demand in the
form of process or building heating, or does it require cooling? What are the target heating
or cooling temperatures? Is there seasonality to the load? These data are critical for accuratelyparameterizing a dynamic simulation of the system and assessing the need for TES.
Is there integration with any other thermal source (e.g., biomass, fossil fuel, geothermal,
waste heat, etc), and if so, what is the capacity, availability, and cost of the alternative
source? These data will in ﬂuence the relative sizing of both the solar component and TES
and will drive overall project economics.
Given the temperatures involved, what type of solar collector, HTF, and ORC working ﬂuid
are most suitable?
Given the size of the application and availability of expander egenerator packages, which
expander type (positive displacement or turbomachine) is suitable?
Is the type of heat rejection (dry, wet, or water) constrained (e.g., lack of available water in
desert regions), or will the decision be made on a least cost basis?
What is the source of project ﬁnance and what are the terms, e.g., for debt ﬁnancing what
would be the loan tenure and interest rate?
What is the market value of the system outputs?
What standard alternative systems (gas burners, heat pumps, PV and battery systems, etc.)
are available for the application, and what would be the total cost of deploying a competing
technology?
These data and others speci ﬁc to the project will inform the decision process for
whether a solar ORC is an economically viable option and will guide the modelingprocess for deciding how to implement a solar ORC or hybrid system.
16.3.5.1 Design of Organic Rankine Cycle systems for use with
solar collectors
Once the temperature regime of the system is determined, the next important speci ﬁ-
cation is the selection of an ORC working ﬂuid, which in ﬂuences the expander con ﬁg-
uration via the pressure evolume ratio of the ORC and the size of heat exchangers for
thermal input and heat rejection. In a parametric study of solar ORC performance andSolar thermal powered Organic Rankine Cycles 603

economics, Garg et al. (2015) investigated the speci ﬁc costs of power for 16 zero-
Ozone Depletion Potential (ODP) and positive condenser pressure (saturation pressurecorresponding to 45
/C14C) working ﬂuids for HTF supply temperatures between 125 and
275/C14C (the top ﬁve most cost ef ﬁcient ﬂuids are shown in Fig. 16.19 ). This analysis
showed that speci ﬁc costs of electricity between 1.25 and 2 USD/W are achievable
for optimized solar ORC systems using a parabolic trough solar collector, with costadvantages at increasing operating temperatures ( Garg et al., 2015 ). Note that a signif-
icant result of this study is that cost ef ﬁciency and thermodynamic ef ﬁciency are not
necessarily correlated when decisions, for example, to recuperate or not and variablecondenser size/fan duty are optimized for cost.
16.3.6 Speci ﬁc and levelized costs of power generation
Solar ORCs have historically been developed more often for research and demonstra-
tion purposes than for commercial applications. Where the decision to deploy dependson market forces, various ﬁnancial metrics such as internal rate of return (IRR),
payback period, total cost of ownership, discounted cash ﬂow, etc., may be selected
as aﬁgure of merit for a project developer. To varying degrees, poor economic perfor-
mance coupled with the necessity for complex dynamic simulation to arrive at calcu-lated performance and optimized cost con ﬁgurations may account for the scarcity of
commercial solar ORC installations to date.
Speci ﬁc capital costs for a nominal power output are relatively straightforward to
calculate and compare among alternatives when the output is electrons (i.e., $/W)but may be more dif ﬁcult to assess or present in a simple metric, when the outputs
are combined heat and power. Even for electricity, $/W for solar plants is typically
indicated in peak Watts ($/W
peak), whereas the daily energy generated by an equivalent
nameplate capacity system could vary signi ﬁcantly between, for example, ﬁxed tilt PV
systems and tracking CSP systems due to the latter ’s higher production levels in the100 150 200 250 3004.08.012.016.0
THTF,in (°C)Cycle efficiency (%)
$/We
1.271.461.651.842.022.21
Figure 16.19 Speci ﬁc cost ($/W e) and cycle ef ﬁciency for various working ﬂuids evaluated at
heat transfer ﬂuid (HTF) inlet temperatures from 125 to 275/C14C for a 5 kW eOrganic Rankine
Cycle.6, butane; 8, isobutane; 9, R134a; B, R161;7, R152a ( Garg et al., 2015 ).604 Organic Rankine Cycle (ORC) Power Systems

morning and evening. Levelized energy costs can be calculated by dividing capital
plus operating and maintenance costs by the total amount of energy produced(estimated via simulation methods), though this too is generally more useful for elec-tricity than for thermal outputs and is dif ﬁcult to assess in scenarios where the value of
outputs changes (such as with time-of-use pricing for electricity).
If the objective of the project is to maximize IRR for a given amount of investment
through the sale of electricity to a single off-taker, the evaluation of the project would
utilize different metrics than in the case of, say, an industrial microgrid with cogene-
ration. In the former project, the use of storage and dynamics of time-of-use pricingwould be important considerations, while in the latter the evaluation framework couldbe minimizing the total cost of ownership or comparison against the status quoalternative, which is likely to have a well-known cost. Projects that are designed tomaximize pro ﬁt through sale of energy as a commodity are therefore conceptually
different from projects where the objective is to meet a particular energy demandproﬁle at minimized cost.
Where cost minimization for a particular demand curve is sought, a general ﬁgure of
merit is the net present cost (NPC) ( Orosz et al., 2013 ):
NPC¼I
oțXn
t¼0OtțMt
ð1țrȚt (16.7)
where Iois the initial cost of the equipment, nis the service life in years, O tand M t
represent Operating and Maintenance costs in year t, and ris the discount rate ( Guth,
2009 ). Operating cost ( Ot) is a unique function of the technology (e.g., fuel con-
sumption) while maintenance cost ( Mt) can be generalized as a function of the initial
costs in two terms: a constant annual service component and a linearly increasingcomponent for repairs due to wear with age, as shown in Eq. (16.8) .
M
t¼f
nțf$ð1/C0aȚ
n2$ð2t/C01Ț (16.8)
where fis the fraction of total maintenance costs with respect to initial costs, and ais a
coefﬁcient expressing the fraction of regular service costs with respect to total main-
tenance costs (including incremental repairs). Various values for these coef ﬁcients can
be found in the literature or in manufacturers manuals and maintenance procedures(NREL, 2013 ).
For demand driven projects, NPC is a valuable tool for comparison between alter-
native energy system con ﬁgurations serving the same purpose, but other ﬁgures of
merit may be preferable depending on the planner ’s objectives and type of available
ﬁnance and the target applications of the system.
16.3.7 Optimization tools
Once a ﬁgure of merit (typically economic) is selected, the process for optimizing
according to this metric involves varying system level approaches (solar ORC orSolar thermal powered Organic Rankine Cycles 605

some competing solution), components, con ﬁgurations, control strategies, etc., while
evaluating the systems against the application conditions and known dynamics (DNI,ambient temperature, demand, and possibly time-of-use pricing). The calculation forthe economic ﬁgure of merit is added to the already complex dynamic simulation of
energy ﬂows in a solar ORC or hybrid system, and in most cases an exhaustive search
of the inherently irregular parameter space will be prohibitively expensive/slow tocompute even using multicore cloud computing resources. This motivates the use
of optimization algorithms, a diverse and specialized set of tools that can achieve
maximization or minimization of an objective function with drastically reducedcomputational effort.
The various approaches, described in more detail in Chapter 7, include evolutionary
and gradient methods, among others. The genetic algorithm (GA), inspired by biolog-ical evolution, “breeds ”a population of ever more optimized solutions through succes-
sive generations, escaping the local maxima or minima that can trap gradient-ascent(“hill-climbing ”) methods via “mutations ”that sample more widely in the search
space. Particle swarm optimization (PSO) similarly employs a population of solutions,
with the candidates initially dispersed and the vector of the “swarm ”in the search
space is parametrically updated to move towards the “leader ”(optimum metric for a
given search timestep) without reference to any evolutionary operators such as“breeding ”,“crossover ”,o r “mutation. ”Once the leader is established in the mono-
tonic region of the global optimum, the optimization can drop the swarm and switchto a faster gradient-ascent/descent from the leader ’s position, however, this requires
either some understanding of the search space or a metric for identi ﬁcation of the
globally optimum ‘hill. ’GA and PSO, as well as more traditional gradient methods,
can be applied to solar ORC optimization to greatly reduce computation time; fordeeper analysis into their relative bene ﬁts the reader is referred to the literature ( Hassan
et al., 2004 ).
16.3.8 Case study: hybrid solar Organic Rankine Cycle
microgrid system
While the recent advent of inexpensive crystalline silicon PV panels may have under-
mined the economic rationale for standalone solar ORC systems, viable applications
for the technology exist in niches created by the need for storage, the utility of cogen-
eration, and the availability of inexpensive alternative supplies of thermal energy. Onescenario that can meet these conditions is the deployment of a solar ORC within anisolated, hybrid microgrid consisting of PV and fossil fueled backup generation, andserving community combined heat and power loads ( Fig. 16.20 ).
While renewable energy technologies are preferable for many reasons (long-term
sustainability, emissions reductions, and even price considerations in extremely rural lo-cations), to achieve 100% availability (capacity factor near to 1) in such a system with
only PV panels and without ever resorting to a backup generator would be prohibitively
expensive under the current price structure of deep cycle batteries ( >250 USD/kWh).
The addition of a solar ORC with TES can reduce (though not fully eliminate) theneed for battery storage, but may only be economically justi ﬁed in cases where a second606 Organic Rankine Cycle (ORC) Power Systems

use of the solar heat is speci ﬁed, such as for heating institutions or dormitories. The need
for a reciprocating internal combustion engine generating set to ensure high availability,however, also provides an opportunity for waste heat recovery from the approximately60% of the fuel exergy that leaves with the exhaust ( Teng and Regner, 2006 ). Note that
while an optimum operating temperature for the ORC can be found for a given location
and solar collector type (discussed in Section 16.3.1 ), the introduction of a second heat
stream may in ﬂuence the optimum operating temperature of the hybrid system due to the
tradeoffs involved with a second heat exchanger and the temperature of the waste heatstream, driving the need to simulate the entire system simultaneously.
Deﬁnition of loads for a new hybrid microgrid could be derived from a survey and
inventory method to predict energy consumption, but a more reliable approach wouldbe to produce a probabilistic distribution function from a dataset of measured loads atcomparable communities (i.e., ones which were only recently connected to the grid
and have similar size and economic activity levels) ( Orosz and Mueller, 2015 ). For
institutional thermal loads, a building envelope model such as developed under AnnexCo f ISO 13790 (2008) could be parameterized for the structure of interest and exer-
cised in dynamic simulation with local TMY data to produce a time distribution ofthermal demand. Once the load pro ﬁles are established with reasonable granularity
Parabolic trough collectorsCollector bypass
Recuperator
WF pumpAir cooled
condenserScroll-expander
with asynchronous
generatorEvaporatorORC bypassPV panelsHTF pumpExhaustBackup generator
Battery
Thermal
storagePower
load
Heat
load
Air
HTF
WF
HeatSolar energy
Propane
AC
DCExhaust hea
tPropane
tank
Figure 16.20 Hybrid Microgrid including photovoltaic (PV), concentrating solar power (CSP),
Organic Rankine Cycle (ORC), thermal energy storage (TES), and a backup generator usingliqueﬁed petroleum gas (LPG).
Adapted from Mitterhofer, M., Orosz, M., 2015. Dynamic simulation and optimization of an
experimental micro-CSP power plant. PowerEnergy. doi:10.1115/ES2015-49333.Solar thermal powered Organic Rankine Cycles 607

(hourly or better), dynamic simulation of the hybrid microgrid can be performed where
the allocation of PV panels, batteries, CSP collectors, and TES can be varied paramet-rically through an optimization method (such as PSO) within a control framework forprioritizing energy ﬂows between components and to the load. A simple example
framework for this application would be:
1.While the sun is shining, the load is supplied (via inverter) from the PV directly to the extent
that PV power is suf ﬁcient to cover the load
2.If the PV power is insuf ﬁcient to meet the load, the load is supplied from a battery bank (via
inverter), as long as the battery storage is not depleted
3.If the PV power exceeds the load, the excess energy charges the battery bank
4.While the sun is shining the CSP collectors charge the TES, until the TES capacity is
reached
5.Thermal loads are supplied by discharging the TES
6.If the TES is insuf ﬁcient to supply the thermal load, a backup (fueled) source is engaged for
heat production
7.If TES reaches overtemperature limits, the ORC is engaged to convert heat to electricity
(except in case 9 below)
8.If the ORC is engaged while the PV is supplying power, its effect is to augment the PV
power and conforms to 2 and 3 (above)
9.If the battery bank reaches a full state of charge (SOC) while PV power and/or ORC power
is exceeding the load, the excess PV power is directed to a load dump or sections of the array
are open circuited, and the ORC is disengaged
10.If the TES reaches overtemperature limits while case 9 holds, the CSP collectors are defocused
11.If the battery reaches a low SOC threshold while the TES is charged, the ORC is engaged
12.If the battery reaches a lowest SOC threshold while the TES is discharged and there is no PV
power, the backup generator is engaged
13.If the generator is engaged and the battery state of charge or the TES SOC reaches a (design
point) recharge threshold, the generator is disengaged.
14.If the generator is engaged and the sun resumes shining, the generator is disengaged
(to preserve battery capacity for the solar systems).
Other special cases can be envisioned but the above is an outline of a simple hier-
archy of power ﬂows that can be considered when sizing and optimizing a hybrid
microgrid and implementing its control strategy.
A one year simulation is likely to be suf ﬁcient for extrapolation across the project life-
time, as a multiyear simulation would not produce results signi ﬁcantly different enough
to justify the additional computational effort. The annual simulation would proceed withload and TMY input at each timestep, track the SOC of the storage components, andtrack the cumulative fuel usage in the backup generator (which constitutes the operatingcost), all while ensuring that the demand (at each timestep) is met.
If the objective function of the microgrid simulation is determination of the mini-
mized cost-recovery tariff for electricity and heat, then for a given combination of
load dynamic and location there will be some con ﬁguration of the PV, battery, CSP,
TES, and ORC components that is optimum (note that under some circumstancesthis could be a size of zero for a particular component). When performing the optimi-zation, it is useful to set boundaries for the parameter space to be explored, i.e.,whereas the generator is sized to meet the peak load with a factor of safety, the PV608 Organic Rankine Cycle (ORC) Power Systems

or ORC components might be varied parametrically from zero to some multiple of the
peak load (not more than 5). The battery and TES components could likewise be variedfrom zero to some multiple of an average day ’s energy requirements, calculated from
integrating the load pro ﬁle (not more than 3).
The cost functions of all components and the ﬁnancial terms of the project ﬁnance
(loan tenure and interest) being known, the objective function of minimized cost-recovery tariff can be solved for with an optimization algorithm searching the tariff
space in a project lifetime cash ﬂow model accounting for initial and recurring costs:
operations, maintenance, loan repayment, and battery replacement (a function of bat-tery type and the number of charge cycles which is tracked in the dynamic simulation).Identi ﬁcation of the cost minimizing solution enables one to proceed with speci ﬁc
component selection and provides a foundation for sizing and engineering design ofthe microgrid. To date, the case study just described has not been realized, andtherefore validation of the simulation and hybrid design approach awaits futureresearch and piloting programs.
References
Ait-Baha CSP Pilot Plant – Technical Datasheet.
American Inventor Uses Egypt ’s Sun for Power, July 2, 1916. Appliance Concentrates the Heat
Rays and Produces Steam. New York Times.
ASTM, 2012. Standard tables for reference solar spectral irradiances: direct normal and hemi-
spherical on 37/C14tilted surface. ASTM 1 e21.
Bado, G., Tomei, G., Angelino, T., Gaia, M., Macchi, E., 1979. The Ansaldo 35 kW solar power
system. In: Proceedings of the Silver Jubilee Congress, pp. 1090 e1094.
Baral, S., Kim, D., Yun, E., Kim, K., 2015. Experimental and thermoeconomic analysis of
small-scale solar organic rankine cycle (SORC) system. Entropy 17 (4), 2039 e2061.
Barutti, A., Pedrick, W.G., Angelino, G., Gaia, M., Macchi, E., 1984. Ansaldo solar thermal and
photovoltaic plants located at Ballajura, western Australia. In: Proceedings of the 8th Solar
World Congress 1983, Biennial Congress of the International Solar Energy Society, vol. 3,pp. 1572 e1576.
Bronicki, L., 2013. Short review of the long history of ORC power systems. In: ASME ORC13.
Canada, S., Brosseau, D., Kolb, G., Moore, L., Cable, R., Price, H., 2005a. Status of APS 1-Mwe
parabolic trough project. In: 2005 DOE Solar Energy Technologies.
Canada, S., Cohen, G., Cable, R., Brosseau, D., Price, H., 2005b. Parabolic Trough Organic
Rankine Cycle Solar Power Plant NREL CP/-550-37077. Denver, CO.
Chambers, T., Raush, J., Russo, B., 2014. Installation and operation of parabolic trough organic
rankine cycle solar thermal power plant in south Louisiana. Energy Procedia 49,
1107e1116.
Courant, R., Friedrichs, K., Lewy, H., March, 1967. On the partial difference equations of
mathematical physics. IBM Journal 11.
De Parga, G.A., 2009. The Curzon-Ahlborn Ef ﬁciency for Three Different Energy Converters,
vol. 1, pp. 1 e4.
Delgado-Torres, A.M., García-Rodríguez, L., October 2007. Comparison of solar technologies
for driving a desalination system by means of an organic Rankine cycle. Desalination 216
(1e3), 276e291.Solar thermal powered Organic Rankine Cycles 609

Dﬁd, 2010. A potential role for an AMC in supporting dish/Stirling concentrating solar power,
case study annex. Advance Market Commitments for Low-carbon Development. http://
www.vivideconomics.com/wp-content/uploads/2010/03/Micro_CSP_case_study.pdf .
Dickes, R., Dumont, O., Declaye, S., Quoilin, S., Bell, I., 2014. Experimental investigation of an
ORC system for a micro-solar power plant. In: Proceedings of the Purdue Conferences 2014.
Dickes, R., Lemort, V., Quoilin, S., 2015. Semi-empirical correlation to model heat losses along
solar parabolic trough collectors. In: Proceedings of ECOS 2015.
Dufﬁe, J.A., Beckman, W.A., Worek, W.M., 1994. Solar Engineering of Thermal Processes,
second ed. Wiley.
Dumont, O., Quoilin, S., Lemort, V., June 2015. Experimental investigation of a reversible heat
pump/organic Rankine cycle unit designed to be coupled with a passive house to get a Net
Zero Energy Building. International Journal of Refrigeration 54, 190 e203.
Einav, A., 2004. Solar energy research and development achievements in Israel and their
practical signi ﬁcance. Journal of Solar Energy Engineering 126 (3), 921.
Electratherm METU NCC Solar ORC Report. Electratherm Website, 2010.Fenton, D.L., Abernathy, G.H., Krivokapich, G.A., Otts, J.V., 1984. Operation and evaluation of
the Willard solar thermal power irrigation system. Solar Energy 32 (6), 735 e751.
Forristall, R., 2003. Heat Transfer Analysis and Modeling of a Parabolic Trough Solar Receiver
Implemented in Engineering Equation Solver. Golden, CO.
Galvez, J.B., 2010. Powersol: Mechanical Power Generation Based on Solar Thermodynamic
Engines – Final Report.
Garg, P., Orosz, M.S., Kumar, P., Dutta, P., 2015. Thermo-economic Evaluation of ORCs for
Various Working Fluids, pp. 1 e16.
Georges, E., Declaye, S., Dumont, O., Quoilin, S., Lemort, V., 2013. Design of a small-scale
organic rankine cycle engine used in a solar power plant. International Journal of Low-Carbon Technologies 8 (Suppl. 1), 1 e8.
Goswami, D.Y., Stefaakos, E., Rahman, M.M., Aydin, S., Reedy, R., 2013. Design,
Construction and Operation of CSP Solar Thermal Power Plants in Florida.
Grirate, H., Zari, N., Elamrani, I., Couturier, R., Elmchaouri, A., Belcadi, S., Tochon, P., 2013.
Characterization of several Moroccan rocks used as ﬁller material for thermal energy
storage in CSP power plants. Energy Procedia 49, 810 e819.
Guth, J.H., 2009. Resolving the Paradoxes of Discounting in Environmental Decisions, vol.
2008, pp. 95 e114.
Hassan, R., Cohanim, B., de Weck, O., Venter, G., 2004. A comparison of particle swarm
optimization and the genetic algorithm. American Institute of Aeronautics and Astronautics1e13.
Helwa, N.H., Bahgat, A.B.G., El Shafee, A.M.R., El Shenawy, E.T., Bahgat, A.B.G., El
Shafee, A.M.R., El Shenawy, E.T., November 2000. Maximum collectable solar energy bydifferent solar tracking systems maximum collectable solar energy by different solar
tracking systems. Energy Sources 22.
ISO 13790:2008 Energy performance of buildings eCalculation of energy use for space heating
and cooling.
Ireland, M., Orosz, M., Brisson, J., Desideri, A., Quoilin, S., 2014. Dynamic modeling and
control system de ﬁnition for a micro-CSP plant coupled with thermal storage unit. In:
ASME Turbo Expo 2014: Turbine Technical Conference and Exposition GT2014-27132.
Kalogirou, S.A., 2004. Solar thermal collectors and applications. Progress in Energy and
Combustion Science 30 (3), 231 e295.
Kane, M., Larrain, D., Favrat, D., Allani, Y., 2003. Small hybrid solar power system. Energy 28
(14), 1427 e1443.610 Organic Rankine Cycle (ORC) Power Systems

Kiceniuk, T., 1985. Development of an Organic Rankine-Cycle Power Module for a Small
Community.
Kohlenbach, P., Mcevoy, S., Stein, W., Burton, A., Wong, K., Lovegrove, K., Burgess, G.,
Joe, W., Coventry, J., 2006. A new parabolic trough solar collector. In: Australian and New
Zealand Solar Energy Society, pp. 1 e8.
Kolb, G.J., 2011. Evaluation of annual performance of 2-Tank and thermocline
thermal storage systems for trough plants. Journal of Solar Energy Engineering
133 (3).
Kuravi, S., Trahan, J., Goswami, D.Y., Rahman, M.M., Stefanakos, E.K., 2013. Thermal energy
storage technologies and systems for concentrating solar power plants. Progress in Energy
and Combustion Science 39 (4), 285 e319.
Lovegrove, K., Stein, W., 2012. Concentrating Solar Power Technology – Principle,
Developments and Application. Woodhead Publishing.
Manolakos, D., Papadakis, G., Kyritsis, S., Bouzianas, K., February 2007. Experimental eval-
uation of an autonomous low-temperature solar Rankine cycle system for reverse osmosisdesalination. Desalination 203 (1 e3), 366e374.
Mitterhofer, M., Orosz, M., 2015. Dynamic simulation and optimization of an experimental
micro-CSP power plant. PowerEnergy. http://dx.doi.org/10.1115/ES2015-49333 .
Moustafa, S., Hoe ﬂer, W., El-Mansy, H., Kamal, A., Jarrar, D., Hoppman, H., Zewen, H., 1984.
Design speci ﬁcations and application of a 100 kWe (700 kWth) cogeneration solar power
plant. Solar Energy 32 (2), 263 e269.
Mueller, A., Orosz, M., Kumar Narasimhan, A., Kamal, R., Hemond, H., Goswami, Y., 2016.
Evolution and feasibility of decentralized concentrating solar thermal power systems for
modern energy access in rural areas. MRS Energy Sustainability 3.
N’Tsoukpoe, K.E., Azoumah, K.Y., Clerc, E., Zmuda, J., Gaye, M., Seshie, Y.M., Kenda
Nitedem, S.E., 2014. CSP4Africa : développement d ’un pilote économiquement viable
d’une mini-CSP pour la production d ’électricité pour mini-réseau en Afrique.
NREL, 2013. “Distributed Generation Renewable Energy Estimate of Costs (Online). Available:
http://www.nrel.gov/analysis/tech_lcoe_re_cost_est.html .
NREL – CSP Projects Review.
Orosz, M., Mueller, A., 2015. Dynamic simulation of performance and cost of hybrid PV-CSP-
LPG generator micro grids with applications to remote communities. In: ASME Power &
Energy Conference 2015.
Orosz, M.S., Quoilin, S., Hemond, H., 2013. Technologies for heating, cooling and powering
rural health facilities in sub-Saharan Africa. Proceedings of The Institution of MechanicalEngineers Part A-journal of Powerand Energy 227 (7), 717 e726.
Orosz, M.S., 2012. Thermosolar and Photovoltaic Hybridization for Small Scale Distributed
Generation: Application for Powering Rural Health. Massachusetts Institute of Technology.
Orosz, M., 2015. Photovoltaics and concentrating solar power: why hybridization makes sense.
SPIE Newsroom 1 e4.
Pytilinski, J.T., 1978. Solar energy installations for pumping irrigation water. Solar Energy 21
(4), 255 e262.
Reda, I., Andreas, A., 2008. Solar Position Algorithm for Solar Radiation Applications.
Rieu, V., September 2012. A 10 kW solar power plant for rural electri ﬁcation. In: Proceedings of
SolarPACES 2012.
Saitoh, T., Yamada, N., Wakashima, S., 2007. Solar Rankine cycle system using scroll
expander. Journal of Environmental Engineering 2 (4), 708 e719.
Simonnot, G., Louche, A., Decanini, Y., 1987. Three years exploitation of the 100 kWe Ajaccio
solar power plant. In: Proceedings of the ISES Solar World Congress, pp. 1588 e1592.Solar thermal powered Organic Rankine Cycles 611

Singh, H., Saini, R.P., Saini, J.S., 2010. A review on packed bed solar energy storage systems.
Renewable &Sustainable Energy 14 (3), 1059 e1069.
Tabor, H.Z., Doron, B., 1990. The Beith Ha ’Arava 5 MW(e) Solar Pond Power Plant
(SPPP)dProgress report. Solar Energy 45 (4), 247 e253.
Tabor, H., 1981. Solar ponds. Solar Energy 27 (3), 181 e194.
Tchanche, B.F., Lambrinos, G., Frangoudakis, a., Papadakis, G., April 2010. Exergy analysis of
micro-organic Rankine power cycles for a small scale solar driven reverse osmosis desa-
lination system. Applied Energy 87 (4), 1295 e1306.
Teng, H., Regner, G., 2006. Achieving high engine ef ﬁciency for heavy-duty diesel engines by
waste heat recovery using supercritical organic- ﬂuid Rankine cycle. In: SAE Pap., No. 724.
Thakur, D., 2013. A case study on the solar biomass hybrid distributed power generation project
at Shive village. In: Odisha Solar Conference.
Wang, X.D., Zhao, L., Wang, J.L., Zhang, W.Z., Zhao, X.Z., Wu, W., March 2010. Performance
evaluation of a low-temperature solar Rankine cycle system utilizing R245fa. Solar Energy
84 (3), 353 e364.
Wang, X.D., Zhao, L., Wang, J.L., 2011. Experimental investigation on the low-temperature solar
Rankine cycle system using R245fa. Energy Conversation and Management 52 (2), 946 e952.
Weinrebe, 2007. Renewable energy – technology, economics and environment. In: Solar
Thermal Power Plants (Chapter 5).
Zhang, S., Wu, Z.H., Zhao, R.D., Yu, G.Y., Dai, W., Luo, E.C., 2014. Study on a basic unit of a
double-acting thermoacoustic heat engine used for dish solar power. Energy Conversation
and Management 85, 718 e726.612 Organic Rankine Cycle (ORC) Power Systems

Copyright Notice

© Licențiada.org respectă drepturile de proprietate intelectuală și așteaptă ca toți utilizatorii să facă același lucru. Dacă consideri că un conținut de pe site încalcă drepturile tale de autor, te rugăm să trimiți o notificare DMCA.

Acest articol: Solar thermal powered Organic [601107] (ID: 601107)

Dacă considerați că acest conținut vă încalcă drepturile de autor, vă rugăm să depuneți o cerere pe pagina noastră Copyright Takedown.

Solar thermal powered Organic [601107]

Copyright Notice

Fiabilitate si Durabilitate – Fiability Durability Supplement No 1 2015 [601898]

TECHNICAL UNIVERSITY OF CLUJ-NAPOCA ACTA TECHNICA NAPOCENSIS Series: Applied Mathematics, Mechanics, and Engineering Vol. XX, Issue xx, Month, 2019 A… [608878]

Specializarea : PSIHOLOGIE -SUGESTIOLOGIE [619905]

Acoperirea solurilor în cartierele Copou, Păcurari și zona periurbană [303780]

POLITICI DE SECURITATE DUPĂ RĂZBOIUL RECE [609690]

Muzica nu este ceva întamplător pe acest pământ..ea este creată de om și pentru om, este [601433]

Copyright Notice

Similar Posts