. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Thermodynamic and economic evaluation ofa small-scale organic Rankine cycle integratedwith a concentrating solar collector

Milad Ashouri1, Fatemeh Razi Astaraei1*, Roghaye Ghasempour1, M.H. Ahmadi1

and Michel Feidt21Renewable Energies and Environmental Department, Faculty of New Science andTechnologies, University of Tehran, Tehran, Iran; 2Laboratoire d’Energetique et deMecanique Theorique et Appliquee, ENSEM, 2, Avenue de laForetde Haye, 60604,Vandoeuvre 54518, France

*Corresponding author:

razias_m@ut.ac.ir

AbstractRecently, distributed power systems especially with renewable sources have shown an increasing demand

all over the world and have been a technical viable solution to demand growth for electricity. Among

these, solar-thermal power plants show a trustworthy source for electricity generation especially for rural

areas where small-scale plants are needed. Organic Rankine cycle (ORC) is a suitable power cycle for

electricity generation from low-grade heat and has shown a good compatibility with parabolic trough

solar collectors (PTCs). In this study, a PTC integrated with an ORC is being studied thermodynamically

and economically for small-scale electricity generation up to 100 kW electricity. Four schematics of the

cycle including the recuperation and superheating are examined. Effect of superheating and recuperating

was investigated on the thermal efficiency and costs of the system. A parametric study shows the effect of

key parameters such as turbine inlet temperature and pressure on the characteristics of the system such as

net work, thermal efficiency, oil temperature, overall heat transfer coefficient and heat transfer area of

shell-and-tube heat exchangers and also on costs of the system. Results show the dependence of the system

efficiency and system costs on the operating pressure of heat exchangers. Existence of the Recuperator

seems quite effective on increasing the cycle efficiency and, in some cases, lowering the total costs due to

lowering the condenser load. A comparison of different working fluids including benzene, butane,

pentane, isopentane, R123 and R245fa have been done to cover a wide range of operating pressures and

temperatures. Results show that benzene has the best thermodynamic performance among other fluids

followed by pentane, isopentane, R123, R245fa and butane. Also, benzene has the highest total cost

among other fluids followed by pentane, isopentane, butane, R123 and R245fa. This paper helps to

evaluate a solar ORC power plant both thermodynamically and economically.

Keywords: organic Rankine cycle; efficiency; solar collector; economic evaluation

Received 3 January 2015; revised 23 July 2015; accepted 24 July 2015

1 INTRODUCTION

Thermodynamics cycles have been put under investigation by

various authors. Ahmadi et al. [1] studied an Atkinson engine

and optimized the performance of the system using genetic algo-

rithm. Among these thermodynamic cycles, solar powered cycles

have been received much attention. Ahmadi et al. [2] optimized

a solar multi-step irreversible Brayton cycle based on normalized

power and thermal efficiency. They also designed a solar dish

Stirling for maximized thermal efficiency and power and also

performed a thermo-economic maximization to consider eco-

nomic parameters [3, 4]. Among these, organic Rankine cycles

(ORCs) have received much attention during last decade. This

cycle obeys the fundamental rules of conventional Rankine

cycles working with water in common plants but has some

advantages over water Rankine cycle which made it popular.

First, this cycle can work under low temperatures and pressures

in comparison to conventional Rankine cycle and shows a better

International Journal of Low-Carbon Technologies 2015, 0, 1–12# The Author 2015. Published by Oxford University Press.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), whichpermits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.doi:10.1093/ijlct/ctv025 1 of 12

International Journal of Low-Carbon Technologies Advance Access published August 30, 2015 by guest on Septem

ber 3, 2015http://ijlct.oxfordjournals.org/

Dow

nloaded from

performance than water especially from low-grade heat sources

because its working fluids include a variety of hydrocarbons and

refrigerants and according to the range of accessible heat source

temperatures and pressures, different outputs can be derived by

using suitable working fluids, second, it can work without feed-

water heaters and multi-stage turbines which makes it simple

and low cost. Among these, solar parabolic trough collectors are

a huge source of thermal energy but with a low-grade heat

which makes it only suitable for some kilowatts to few megawatts

electricity generation, also it shows a reliable means for electri-

city generation especially in rural areas or near factories to meet

their demand without the need for connection to grid which

may be expensive. As mentioned before, the organic fluids used

in ORCs are divided into hydrocarbons and refrigerants, some

of them are dry fluids which mean they have a positive slope

T–S diagram in the saturation vapor region. This makes it pos-

sible for some organic fluids to work properly without super-

heating to a great extent and cause no damage to turbine. As

shown in this study, a comparison of different dry organic fluids

with and without superheating and recuperation has been done

to show the variations in cycle efficiency and performance of

the system which help us make a decision to choose the system

conditions according to our needs. The ORC has been put

under investigation by many authors. Le et al. [5] performed a

thermodynamic analysis and optimization of a supercritical

ORC driven by low-grade heat. Two cases of basic and regenera-

tive ORC were investigated. Results showed that R1234ze has the

best performance. Rayegan and Tao [6] developed a procedure

to select the working fluids in a solar ORC plant and found that

11 working fluids are suitable in solar ORCs that under low- or

medium-temperature solar collectors. Wang et al. [7] presented

a detailed analysis of ORC coupled with solar collector with a

thermal storage system during a whole day. McMahan [8]

designed and optimized a solar-thermal ORC with various

working fluids. Quoilin et al. [9] presented an optimization and

sizing procedure of heat exchangers in a small-scale solar-driven

ORC by pinch and pressure drop and optimized it by turbine

input pressure and evaporator temperature. Ferrara et al. [10]

compared different organic fluids in a 20 kWe solar plant and

chose acetone as the best organic fluid choice with supercritical

pressure. Le et al. [11] performed a thermodynamic and eco-

nomic analysis of a subcritical ORC using zeotropic working

fluids and water temperature at 1508C as a hot source. Pentane

showed the best exergetic performance and the lowest levelized

cost of electricity. The best maximized exergy efficiency cor-

responds to the working fluid mixture with the smallest

temperature glide. Under exergy efficiency maximization, pure

fluid-based ORCs present higher thermodynamic performances.

Wang et al. [12] performed a multi-objective optimization of

ORC with waste heat using the exergy efficiency and total capital

costs of the system as the objectives. The turbine inlet tempera-

ture, turbine inlet pressure, pinch and approach points of the

heat exchangers were introduced to be the key parameters of the

system for performing the optimization. Al-Sulaiman [13] pre-

sented an exergy analysis of a parabolic trough solar collector on

a steam Rankine cycle and an ORC bottoming cycle as the con-

denser. Result showed that the main sources of destruction are

trough collectors and the vapor generator. Also, the bottoming

cycle increased the exergy efficiency compared with a steam

Rankine cycle without the bottoming cycle. Each of the afore-

mentioned works in the literature focused on one type of ORCs.

In this study, four schematics of the cycle have been chosen to

study with respect to both thermodynamic and economics to

give a better evaluation of the solar ORC power plants. A

small-scale ORC driven by a parabolic trough solar collector was

simulated using a commercial software [14] interacting with

MATLAB for doing the energy balance of the system and sizing

procedure of heat exchangers for the system described.

2 THERMODYNAMIC MODEL

Schematics of solar ORC is presented in Figure 1. Also, the

process in T–S diagram is shown in Figure 2.

ORC fluid is pumped through economizer and is heated to

its saturation temperature. Then it enters evaporator in which

the two-phase mixture is transformed to pure vapor in constant

temperature, the third heat exchanger superheats the fluid and

raise the temperature in a constant pressure. Vapor in state 6

enters the turbine and delivers the power, because of the high

energy content of vapor in state 7, we can reuse its energy by

entering the fluid to a Recuperator to preheat the incoming sub-

cooled fluid before entering the economizer. The working fluid

enters the condenser at state 7 and is cooled to its saturated

liquid. Then, the pump raises its pressure to economizer pres-

sure and the cycle completes. The basic relations used in the

model are as follows [15].

Mass balance equation:

X

_mi ¼X

_mo ð1Þ

where _mi is the inlet mass flow rate (in kg/s) and _mo is the outlet

mass flow rate (in kg/s).

Energy balance equation:

Q�W ¼X

_moho �X

_mihi ð2Þ

where Q indicates heat (in kW), W work (in kW) and h is en-

thalpy of hot fluid (in kJ/kg).

Vapor generator (including economizer (Eco), evaporator

(Eva) and superheater (SH)):

QVG ¼ UAEcoDTLMTD;Eco þ UAEvaDTLMTD;Eva

þ UASHDTLMTD;SH ð3Þ

where UA is product of heat transfer coefficient and heat transfer

area (in kW/8C), where U refers to overall heat transfer coeffi-

cient (in W/m K) and TLMTD is logarithmic mean temperature

difference.

M. Ashouri et al.

2 of 12 International Journal of Low-Carbon Technologies 2015, 0, 1–12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

Turbine (Tur) isentropic efficiency:

hTur ¼WTur;a

WTur;s

¼hi � ho;a

hi � ho;s

; ð4Þ

WTur;a ¼ m6h6 �m7h7 ð5Þ

where hTur is efficiency of the turpine, ho;s is isentropic enthalpy

of hot fluid and ho;a is actual enthalpy of hot fluid.

Condenser (Cd):

_QCd ¼ _m8h8 � _m1h1 ¼ UACdDTLMTD;Cd ð6Þ

Pump isentropic efficiency:

hpump ¼Wpump;s

Wpump;a

ð7Þ

Net output work of the system:

Wnet ¼ WTur;a �Wpump;a ð8Þ

Thermal efficiency ðhThÞ of the system:

hTh ¼Wnet

Qin

ð9Þ

In this study, turbine isentropic efficiency and mechanical effi-

ciency was supposed to be 80 and 99%, respectively, as a logical

supposition. A single-stage refrigerant turbine was used because

delivering small amounts of work up to 100 kWe was considered

[16]. The Recuperator model effectiveness was fixed to 80%

corresponding to the medium technology design [17] with a

minimum pinch of 58C and a maximum pressure drop of 3%.

A centrifugal ORC pump with an 85% isentropic efficiency was

selected. Also, a reciprocating pump was selected for oil circula-

tion through the collectors. The vapor generator consists of

three phases, first of them makes the quality of vapor equal to

zero (x ¼ 0), second one evaporates the fluid to its saturated

vapor in constant pressure (x ¼ 1) and the third part heats the

Figure 1. Schematic diagram of a solar ORC with superheater and recuperator.

Figure 2. T–S diagram for the ORC process (benzene).

Thermodynamic and economic evaluation of a small-scale organic Rankine cycle

International Journal of Low-Carbon Technologies 2015, 0, 1–12 3 of 12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

saturated vapor to some degree and makes it superheated. For

the sake of simplicity, the same system conditions described

above were taken into account for all the working fluids. The

three phases described are shown in Figure 3. The vapor gener-

ator and condenser use multi-pipe shell-and-tube heat exchan-

gers for relative low cost and commercial availability. The

working fluids are in the tube side and the shell side conveys the

oil and water in the vapor generator and condenser, respectively.

The schematic of a shell-and-tube heat exchanger is shown in

Figure 3.

Water-cooled condenser is used in the cycle with inlet water

temperature of 158C, minimum pinch of 58C was selected. It

is assumed that the outlet state of working fluid from the con-

denser is saturated. The condenser consists of two phases,

de-superheating region and condensation region which are

shown in Figure 4.

2.1 Single-phase regionIt is assumed that the flow is fully developed in the heat exchan-

gers and the condition is steady state. Also, heat losses to

ambient are neglected. The governing equations in the single-

phase region of heat exchangers in the tube side are as follows:

Re ¼rul

m; ð10Þ

where Re is the Reynolds number, u is the velocity (m/s) of the

organic fluid in the tube and is defined as

u ¼_m

rS; ð11Þ

l is defined as the characteristic length and, for a tube, is defined

as its inner diameter Di. Density (r, in kg/m3) dynamic viscosity

(m, in Pa s) and Prandtl number are found from fluid mean tem-

perature. S is the cross-sectional area of tubes (in m2)

The Stanton number is given by

St ¼ ERe�0:205Pr�0:505; ð12Þ

where

E ¼ 0:0225 expð�0:0225 ln ðPrÞ2Þ; ð13Þ

The convective heat transfer coefficient (a, in W/m2 C) for the

inner tube is defined as [18]

ai ¼ rucSt; ð14Þ

Figure 3. Temperature profile in vapor generator and schematic for a multi-pipe shell-and-tube heat exchanger.

Figure 4. Temperature profile in condenser.

M. Ashouri et al.

4 of 12 International Journal of Low-Carbon Technologies 2015, 0, 1–12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

where c is the heat capacity of the fluid, St is Stanton number.

For calculating the heat transfer coefficient across bundle of

tubes (shell side), the Nusselt number is given as

Nu ¼ aRemc Pr0:34FN; ð15Þ

where

Rec ¼rumDo

m; ð16Þ

where a, m and FN are constants related to Reynolds number.

The velocity um is calculated at the minimum cross-flow area in

the bundle as

um ¼_mORC

LwNtr; ð17Þ

where Nt is the number of tubes in a row, L is the length of tube

experiencing the cross-flow and w is dependent to the arrange-

ment of tubes in the bundle [18].

The convective heat transfer coefficient in the shell side is

therefore defined as:

ao ¼kNu

Do

ð18Þ

2.2 Two-phase regionFor calculating the boiling heat transfer coefficient in the vapor

generator and condenser, it is necessary to notice the significant

changes of fluid properties during phase change. The heat trans-

fer in the tubes is divided into small parts of constant properties

to gain more realistic results. The coefficient in the saturated

boiling region is the sum of convective and nucleate components

[18, 19] as follows:

ai ¼ anuc þ aconv; ð19Þ

where

aconv ¼ 0:023FcKL

Di

Re0:8L Pr0:4

L ð20Þ

and

anuc ¼ 0:00122K0:79L C0:45

L r0:49L

s0:5m0:29L ðlrGÞ

0:24

( )

ScðTw�TsÞ0:24ðpw� psÞ

0:75;

ð21Þ

where subscripts L and G accounts for the liquid and gas phase.

Fc and Sc are convective–correction factor and suppression

factor, respectively. The Reynolds number is calculated on the

assumption of liquid flowing in the tube.

ReL ¼ð1� xÞGDi

mL

; ð22Þ

PrL ¼CLmL

kL; ð23Þ

where x is the vapor quality (in %) of flow and G is the mass

velocity (in kg/m2 s) of the fluid defined as

G ¼_mORC

S; ð24Þ

where _mORC is the mass flow rate of organic fluid and S is the

tube flow area.

2.3 Condensation regionThe condensate in the condenser is assumed to form annular

flow and the coefficient is calculated based on the method of

Boyko [20] as described below.

The fluid is assumed to be liquid flow in the tube. The ReL(Eq. 22, when x ¼ 0) and PrL are used in the calculation of

Nusselt number. The equation becomes

aL ¼ 0:023KL

Di

Re0:8L Pr0:4

L ; ð25Þ

where K is the thermal conductivity (in W/ m C).

The convective heat transfer in the condenser is given as:

ai ¼aL

21þ

ffiffiffiffiffiffi

rLrG

r� �

: ð26Þ

Finally, the overall heat transfer coefficient is [18]:

1

U¼

1

ao

þ Ro þ1

ai

þ Ri

� �

Do

Di

þyw

kw

Do

Di

; ð27Þ

where the subscripts i and o refers to inside and outsider of

tubes, respectively, and yw refers to wall thickness of tubes. Also,

Rf is the fouling factor resistance.

2.3 Working fluidsSix fluids were chosen including two refrigerants R123 and

R245fa and four hydrocarbons benzene, isopentane, pentane

and butane to cover a wide range of working temperatures and

pressures. Properties of these fluids are shown in Table 1. One of

the major concerns of organic fluids is their safety, global

warming potential (GWP) relative to CO2, ozone depletion po-

tential (ODP) relative to R11 and atmospheric lifetime [22].

2.4 Collector modelIn this study, a parabolic trough solar collector was chosen as the

heat source. THERMINOL VP-1 was selected as the working

fluid [24] and its characteristics are shown in Table 2. Design

point collector nominal optical efficiency of the collector was

selected 0.82 according to Eq. (30).

hopt ¼ rmirror � hshadowing � hgeometry � hunaccounted ð28Þ

Site latitude, altitude and hour of the day were considered 358,

1420 m and 12 (solar time), respectively, according to Tehran.

Thermodynamic and economic evaluation of a small-scale organic Rankine cycle

International Journal of Low-Carbon Technologies 2015, 0, 1–12 5 of 12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

Other input characteristics of solar collector and ORC system

are presented in Table 3.

3 ECONOMIC MODEL

The economic model developed here is based on capital cost

equations of various components in the solar ORC model. For

all equipment, there is a strong dependence on material and op-

erating pressure. The method of accounting for changes in oper-

ating pressures is covered through the use of the pressure factor.

Also, the change in the material of construction is covered

through the use of the material factor. The total cost of the

system is the summation all components.

Ctot ¼ CVG þ Ctur þ CRec þ CCd þ Cpump þ CSolarpump

þ CSolarCollector; ð29Þ

The purchased cost for vapor generator is given by [21]:

CVG ¼ CoVG bVG

1 þ bVG2 FVGm FVGp

� �

; ð30Þ

where bVG1 and bVG

2 are constants of heat exchanger type. FVGm is

the material factor for stainless steel and FVGp is the pressure

factor for operating pressure of heat exchanger. The coefficients

used in the model are represented in Table 4.

The base cost of vapor generator CoVG is the cost of equipment

made from carbon steel at ambient working pressure and is

defined as

logCoVG ¼ kVG1 þ kVG2 log AVG þ kVG3 ðlogAVGÞ

2; ð31Þ

where kVG1 , kVG2 and kVG3 are constants of heat exchanger type.

Also, AVG is the calculated heat transfer area for vapor generator.

Also, FVGp is given by

log FVGp ¼ cVG1 þ cVG2 log pVG þ cVG3 ðlog pVGÞ2; ð32Þ

where cVG1 , cVG2 and cVG3 are constants of heat exchanger type and

pVG is the operating pressure of organic fluid in bar.

The cost of turbine is expressed as:

Ctur ¼ FbmC0tur; ð33Þ

where Fbm is the bare module factor and is related to turbine

type and material (stainless steel steam turbine). C0tur is the base

cost for turbine expressed as:

log CoTur ¼ kTur

1þ kTur2 log Wtur þ kTur3 ðlogWturÞ

2; ð34Þ

where Wtur is the turbine shaft power in kW. The cost functions

for recuperator and condenser are similar to vapor generator

with different heat transfer area and operating pressure as

Table 2. Thermal oil properties [24].

Substance Min/max

temperature

(8C)

Density

(kg/m3)

Specific

heat

(kJ/kg C)

Thermal

conductivity

(w/m C)

THERMINOL

VP-1

12.78/398.9 1067.6 1.532 0.1368

Table 3. Collector model parameters.

Parameter Value

rmirror 0.94

�shadowing 0.98

�geometry 0.93

�unaccounted 0.96

Receiver tube outer diameter 70 mm

Reflector aperture width 1 m

Reflector row pitch/aperture width 2.5

Reflector cleanliness factor 0.95

Receiver tube emissivity 0.14

Receiver glass envelope emissivity 0.86

Convective heat transfer coefficient outside glass envelope 56.78 w/m28C

Mass flow rate of thermal oil 4.5 kg/s

Minimum allowable pinch point temperature difference

(for vapor generator, condenser and recuperator)

58C

Mass flow rate of ORC fluid 0.5 kg/s

Table 1.Working fluid properties [22, 23].

Substance Physical data Tbpa (8C)

Tcrb (8C) Pcr

c (bar)

Molar mass

(kg/kmol)

Safety

data

ASHRAE

34 safety

group

Environmental

data

atmospheric

lifetimed

/GWP/ODP

Benzene 80.08 288.9 48.94 78.108 – –

pentane 36.06 196.55 33.70 72.149 A3 –

R123 27.823 183.68 36.618 152.93 B1 1.3/77/0.02

Isopentane 27.83 187.2 33.78 72.149 A3 –/20/0

R245fa 15.14 154.01 36.51 134.05 B1 7.6/1030/0

Butane 20.49 151.98 37.96 58.122 A3 –

aNormal boiling point.bCritical temperature.cCritical pressure.dTime in which the organic fluids remain in the atmosphere (year).

Table 4. Coefficients of cost estimation equations.

Component K1 K2 K3 B1 B2 Fm

Vapor generator 2.7652 0.7282 0.0783 1.74 1.55 2.8

Recuperator 2.7652 0.7282 0.0783 1.74 1.55 2.8

Condenser 2.7652 0.7282 0.0783 1.74 1.55 2.8

Turbine 3.4092 20.5104 0.0030 0 1 3.6

M. Ashouri et al.

6 of 12 International Journal of Low-Carbon Technologies 2015, 0, 1–12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

follows:

CCd ¼ CoCd bCd

1 þ bCd2 FCdm FCdp

� �

; ð35Þ

log CoCd ¼ kCd1 þ kCd2 log ACd þ kCd3 ðlog ACdÞ

2; ð36Þ

log FCdp ¼ cCd1 þ cCd2 log pCd þ cCd3 ðlog pCdÞ2; ð37Þ

CRec ¼ CoRec bRec

1 þ bRe2 FRecm FRecp

� �

; ð38Þ

log CoRec ¼ kRec1 þ kRec2 log ARec þ kRec3 ðlog ARecÞ

2; ð39Þ

log FRecp ¼ cRec1 þ cRec2 log pRec þ cRec3 ðlog pRecÞ

2: ð40Þ

The ORC pump uses a centrifugal pump and its cost is defined

as:

CORCpump ¼ CoORCpump

�

bORCpump2

þ bORCpump2 FORCpump

m FORCpumpp

�

; ð41Þ

where

log CoORCpump ¼ k

ORCpump1 þ k

ORCpump2 log WORCpump

þ kORCpump3 ðlog WORCpumpÞ

2: ð42Þ

bORCpump1 , b

ORCpump2 are the constants for centrifugal pump.

kORCpump2 , k

ORCpump2 and k

ORCpump3 are constants of pump type

and WORCpump is the pump shaft work. The FORCpumpm is the ma-

terial factor for stainless steel pump and FORCpumpp is defined as:

log FORCpumpp ¼ c

ORCpump1 þ c

ORCpump2 log PORCpump

þ cORCpump3 ðlog PORCpumpÞ

2; ð43Þ

where pORCpump is the operating pressure of pump in bar. Also,

cORCpump2 , c

ORCpump2 and c

ORCpump3 are constants of pump type.

The cost of solar pump is defined similar to ORC pump with

different type (reciprocal pump) given as:

CSolarpump ¼ CoSolarpump

�

bSolarpump1

þ bSolarpump2 FSolarpump

m FSolarpumpp

�

; ð44Þ

log CoSolarpump ¼ k

Solarpump1 þ k

Solarpump2 log WSolarpump

þ kSolarpump3 ðlog WSolarpumpÞ

2; ð45Þ

log FSolarpumpp ¼ cSolarpump

1þ c

Solarpump2 log PSolarpump

þ cSolarpump3 ðlog PSolarpumpÞ

2: ð46Þ

The cost of solar field is the summation of its components in-

cluding reflector, receiver, heat transfer fluid (thermal oil),

structure and miscellaneous (material and equipment) expressed

as:

CSolarfield ¼ CReflector þ CReceiver þ CHTF þ CStr þ CMisc: ð47Þ

The cost of solar collectors is a strong function of aperture area

of collectors. The cost data for parabolic trough solar collector

were extracted from the Thermoflow software [14] and were

used in the model.

4 RESULTS AND DISCUSSION

4.1 Thermodynamic analysis resultsTable 5 shows the ORC fluids and min/max pressures and tem-

peratures which were used in the modeling as the base case with

the existence of recuperator and superheater.

Table 6 summarizes the results for six working fluids with

four schematics, simple Rankine cycle without regenerator and

superheater, with superheating, with regenerating and with both

of them. Table 6 shows that the existence of recuperator has a

more effective role on increasing the system efficiency compared

with superheating. According to Figure 5, superheating can raise

the system efficiency to some extent for all working fluids

because the energy content of working fluid is raised, but it also

increases the temperature difference in the collectors which

requires more reflective area for collectors. This behavior is

based on the reason that dry behavior of organic fluids causes

the average temperature of heat rejection to increase together

with the average temperature of heat addition. Table 6 shows a

comparison of different schematics for all working fluids. It is

revealed that existence of recuperator plays an important role in

Table 5. ORC fluids and their temperatures and pressures at turbine input

and condenser output as the base case.

ORC fluid PT (bar) TT (8C) PC (bar) TC (8C)

Benzene 26.82 287 0.20 35.28

Pentane 25 190 1 35.67

R123 30 180 1.3 35

Isopentane 30 185 1.28 34.76

R245fa 23 140 2.11 35

Butane 25.02 137 3.30 35.17

Table 6. Net electrical efficiency for different working fluids (percent).

Working fluids Simple

Rankine

cycle

Addition of

superheater

Addition of

recuperator

Addition of

superheater and

recuperator

Benzene 22.28 22.61 24.19 26.55

Pentane 15.49 15.5 17.86 18.73

R123 15.21 15.47 15.95 16.86

Isopentane 15.12 15.17 17.37 18

R245fa 12.07 12.18 12.74 13.53

Butane 12 12.1 12.62 13.37

Thermodynamic and economic evaluation of a small-scale organic Rankine cycle

International Journal of Low-Carbon Technologies 2015, 0, 1–12 7 of 12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

increasing efficiency, this is because of dry behavior of organic

fluids which their positive slope of T–S diagram causes the

outlet state of turbine to be in superheated region. A lot of

energy could be recuperated from vapor at turbine outlet and

transported to incoming fluid before entering vapor generator

by means of a recuperator. Not only existence of recuperator

decreases the amount of heat addition to working fluids in heat

exchangers, but it also decreases the condenser load to a great

extent. This is effective on lowering the heat transfer area of con-

denser and solar collector which contributes to lower total cost

for entire system. Results of modeling different organic fluids are

shown in Table 6.

As can be seen in Table 6 among these six working fluids,

benzene shows the best performance for these conditions

described. Figures 6 and 7 show the effect of turbine inlet pres-

sure on the efficiency and work output.

According to Figure 6 there is an optimum pressure in each

temperature for fluids, the reason for this behavior lies in the

h– s diagram of working fluids, when outlet pressure of turbine

is fixed and input pressure rise, the enthalpy difference between

two pressures goes through a maximum which is due to the con-

stant pressure curves in h– s diagram of fluids [8].

Figure 7 shows the effect of turbine inlet pressure on work

output, the more pressure increases the more work produces.

But, on the other hand, the work consumed by pump to gener-

ate the required pressure also increases. As mentioned above, en-

thalpy difference across the turbine goes through a maximum

which plays a more important role in determining the network

output. Overall results are shown in Figure 7. Also a noticeable

interpretation from Figures 7 and 8 is that R123 produces less

power than butane in all working pressures but has a more

thermal efficiency. This is because butane needs more heat to

evaporate and reach the desired condition with the same mass

flow rate.

Figure 8 shows the multiplication of overall heat transfer

coefficient and heat transfer area (UA) in three regions of vapor

generator, including economizer, evaporator and superheater.

Also, Figure 9 shows the heat transfer rate in these regions,

respectively. When turbine input pressure increases with fixed

turbine back pressure, the amount of heat addition to econo-

mizer increases so as to make the sub-cooled fluid to be satu-

rated liquid (as shown in Figure 9) which leads to an increase in

UA. With rising the pressure, the enthalpy difference (latent heat

of evaporation) in two-phase area on the T–S diagram decreases

(see Figure 2) leading to a decrease in the amount of heat add-

ition in evaporator (as shown in Figure 9). On the other hand,

Figure 6. Effect of varying generator pressure on efficiency of the system.

Figure 7. Effect of generator pressure on net output work.Figure 5. Effect of superheating in different working fluids with regenerator.

Figure 8. Effect of generator pressure on characteristics of vapor generator

(benzene).

M. Ashouri et al.

8 of 12 International Journal of Low-Carbon Technologies 2015, 0, 1–12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

because of increasing the evaporation temperature with increas-

ing the pressure in evaporator, the UA of evaporator increases

due to decreased DTLMTD in evaporator. With the raised pres-

sures, the same amount of superheating is obtained with less

heat addition to organic fluid as shown on Figure 9 leading to a

decreased UA in superheater. Results are shown in Figures 8 and

9 for benzene. Also, the summation of heat transfers in sub-

sections of vapor generator is shown in Figure 9. Because the

outlet temperature of vapor generator is constant, the variation in

generator pressure does not affect the outlet temperature of heat

source as shown in Figure 8, but the inlet temperature of collector

rises to a little amount with increasing the generator pressure.

This is because of the fact that increasing the generator pressure

will reduce the overall required heat of vapor generator. Also,

Figure 10 shows the variation in generator pressure on pinch

values. As can be seen, increasing the working pressure, at first

stages, do not affect the pinch value because of constant generator

temperature, but increasing the pressure eventually causes the

pinch value to decrease to a minimum allowable value.

From Figures 11 and 12, it appears that rising the turbine

inlet pressure will decrease the required reflective area in solar

field. This is due to the fact that rising the pressure causes the

overall heat duty in vapor generator to fall (according to

Figure 9) and this will reduce the required reflective area in solar

field.

Figure 13 shows the variation in UA of collector against gen-

erator pressure. UA of collector is an indicator for heat loss in

the absorber tube. Therefore, with increasing the pressure,

according to a slight increase in collector inlet temperature, the

heat losses decrease and therefore the UA is subject to a small

decrement as shown in the figure.

4.2 Economic analysis resultFigure 14 shows the column charts for each working fluid with

different schematic and their corresponding efficiencies and

costs. As can be seen, for benzene, isopentane and pentane, the

total capital cost of the plant with recuperator is the lowest al-

though it has an extra equipment (recuperator) compared with

the simple plant and the plant with superheater. The reason is

that the existence of pecuperator lowers the load of condenser by

lowering the enthalpy of inlet fluid to condenser. Also, by

Figure 9. Effect of turbine inlet pressure on characteristics of vapor generator

(benzene).

Figure 10. Effect of generator pressure on pinch temperature difference of vapor

generator for different fluids.

Figure 11. Effect of turbine inlet pressure on solar collectors’ reflective area

(benzene, R123 and R245fa).

Figure 12. Effect of generator pressure on solar collectors’ reflective area

(Pentane, Isopentane and Butane).

Thermodynamic and economic evaluation of a small-scale organic Rankine cycle

International Journal of Low-Carbon Technologies 2015, 0, 1–12 9 of 12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

increasing the enthalpy of inlet fluid to vapor generator, it

decreases the reflective area required by solar collectors and con-

tribute to the lower cost of the overall plant. The plant with both

the superheater and recuperator has the highest efficiency but

with higher capital cost for all working fluids. Detailed results

are shown for each fluid according to Table 5 input parameters.

Figures 15 and 16 shows the detailed estimated costs of

various components including heat exchangers, pumps, turbine

and solar field against turbine inlet pressure for benzene. As can

be seen in the figures, turbine pressure is an effective parameter

on the total cost of the system. According to Figure 15, vapor

generator, recuperator and refrigerant pump show an increasing

trend with pressure because of increasing heat transfer surface

area in heat exchangers and pressure rise in the pump. On the

other hand, condenser cost decreases as the pressure increases.

This is due to the single-phase region in the condenser (see

Figure 4). With increasing pressure, the output temperature of

recuperator falls, due to improved heat transfer, and surface area

of condenser falls leading to a decreased costs. According to

Figure 16, solar pump cost remains approximately constant

because of its weak relation to turbine pressure. Other compo-

nents show a decreasing trend. This is due to the fact that asFigure 13. Effect of generator pressure on UA of solar collector (for benzene in

vapor generator).

Figure 14. Column charts indicating output work, thermal efficiency and total costs for each working fluid with different schematics.

M. Ashouri et al.

10 of 12 International Journal of Low-Carbon Technologies 2015, 0, 1–12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

operating pressure of heat exchangers goes up, overall tempera-

ture rise in the collector falls resulting in a decrease in reflector

surface. As it appears from Figures 16 and 17, the receiver and

reflector accounts for the main portion of solar field total price.

The relatively slight decrease in the receiver and reflector cost is

due to the small reduction in aperture surface as mentioned

above.

Figure 18 shows that solar field constitutes the main portion

of total cost of the system (≏40–45% of the total cost). Also,

with increasing pressure, the total cost of ORC equipment rises

due to the increased heat transfer areas in the heat exchangers

but the solar field shows a decreasing trend due to decreasing

aperture area of collectors which has the dominant effect among

other costs in the solar field.

Figure 19 shows the variation in total cost of the system and

thermal efficiency against turbine inlet pressure on the condition

that other parameters remain constant. According to Figure 19,

increasing the turbine inlet pressure causes the total cost of the

plant to increase. But the efficiency goes up at first and then

decreases. Therefore, choosing the suitable operating pressure is

a tradeoff between thermal efficiency and cost of the system.

However, as the efficiency does not change dramatically (,1%)

the lower pressures are more desirable due to low costs.

5 CONCLUSION

In this study, an ORC engine coupled with a parabolic trough

solar collector was investigated thermodynamically and

Figure 15. Cost variation of ORC components versus turbine pressure

(benzene).

Figure 16. Cost variation of solar field components versus turbine pressure.

Figure 17. Cost variation of reflector and receiver versus turbine pressure.

Figure 18. Total cost of ORC and solar field versus turbine pressure.

Figure 19. Total cost of the solar ORC system and thermal efficiency against

turbine inlet pressure (benzene).

Thermodynamic and economic evaluation of a small-scale organic Rankine cycle

International Journal of Low-Carbon Technologies 2015, 0, 1–12 11 of 12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from

economically for small-scale power generation. Six working

fluids were examined in cycle with different working tempera-

tures and pressures. Also, four schematics of ORC engine includ-

ing simple cycle, regenerative cycle, superheating cycle and both

superheating and regenerating cycle were compared. A

shell-and-tube heat exchanger consisting of three regions

(economizer, evaporator and superheater) was used as a vapor

generator. Sizing of heat exchangers was performed in order to

obtain capital costs of heat exchangers. The effect of key para-

meters such as turbine inlet pressure and temperature on cycle

efficiency, power output, oil temperature, current UA of heat

exchangers and costs of the equipment were investigated. Results

show that, among these fluids, benzene has the best performance

in terms of efficiency (26.55%) but with the highest total system

cost (≏76 0000 USD) followed by pentane, R123, isopentane,

butane and R245fa. Butane has the lowest efficiency (13.37%)

and R245fa has the lowest total cost of 34 000 USD. Th perform-

ance evaluation reveals that existence of recuperator is vital

because it increases the thermal efficiency by recovering the heat

from turbine outlet and lowering the loads of condenser and

solar collectors. Also, in some cases, recuperator causes the low-

ering of total cost of the system compared with simple cycle as a

result of lowering the condenser and solar collector loads. The

best schematic was found to be the cycle with recuperator and

superheater but with the cost of higher total cost. The paramet-

ric study reveals that turbine inlet pressure has an important

effect on thermal efficiency and total cost of the system.

REFERENCES

[1] Ahmadi MH, Ahmadi MA, Shafaei A, et al. Thermodynamic analysis and

optimization of the Atkinson engine by using NSGA-II. Int J Low Carbon

Tech 2015; doi:10.1093/ijlct/ctv001.

[2] Ahmadi MH, Ahmadi MA, Feidt M. Performance optimization of a solar-

driven multi-step irreversible brayton cycle based on a multi-objective

genetic algorithm. Oil Gas Sci Technol – Rev. IFP Energies Nouvelles 2014;

doi:10.2516/ogst/2014028.

[3] Ahmadi MH, Sayyaadi H, Dehghani S, et al. Designing a solar powered

Stirling heat engine based on multiple criteria: maximized thermal effi-

ciency and power. Energ Convers Manage 2013;75:282–91.

[4] Ahmadi MH, Sayyaadi H, Mohammadi AH, et al. Thermo-economic multi-

objective optimization of solar dish-Stirling engine by implementing evolu-

tionary algorithm. Energ Convers Manage 2013;73:370–80.

[5] Le VL, Feidt M, Kheiri A, et al. Performance optimization of low-

temperature power generation by supercritical ORCs (organic Rankine

cycles) using low GWP (global warming potential) working fluids. Energy

2014;67:513–26.

[6] Rayegan R, Tao YX. A procedure to select working fluids for Solar Organic

Rankine Cycles (ORCs). Renew Energ 2011;36:659–70.

[7] Wang J, Yan Z, Zhao P, et al. Off-design performance analysis of a solar-

powered organic Rankine cycle. Energ Convers Manage 2014;80:150–7.

[8] Mcmahan AC. Design & optimization of organic rankine cycle solar-

thermal powerplants. MSc Thesis, 2006.

[9] Quoilin S, Orosz M, Hemond H, et al. Performance and design optimiza-

tion of a low-cost solar organic Rankine cycle for remote power generation.

Sol Energ 2011;85:955–66.

[10] Ferrara F, Gimelli A, Luongo A. Small-scaled concentrated solar power

(CSP) plant: ORC’s comparison for different organic fluids. Energy Procedia

2014;45:217–26.

[11] Le VL, Kheiri A, Feidt M, et al. Thermodynamic and economic optimizations

of a waste heat to power plant driven by a subcritical ORC (Organic Rankine

Cycle) using pure or zeotropic working fluid. Energy 2014;76:622–38.

[12] Wang J, Yan Z, Wang M, et al. Multi-objective optimization of an organic

Rankine cycle (ORC) for low grade waste heat recovery using evolutionary

algorithm. Energ Convers Manage 2013;71:146–58.

[13] Al-Sulaiman FA. Exergy analysis of parabolic trough solar collectors inte-

grated with combined steam and organic Rankine cycles. Energy Convers

Manage 2014;77:441–9.

[14] Therrmoflow Inc. THERMOFLOW 19.0. Sudbury, MA, 2009.

[15] El-Wakil MM. Powerplant Technology. McGraw-Hill, 2002.

[16] Barbieri E, De Pascale A, Ferrari C, et al. Performance evaluation of the inte-

gration between a thermo-photo-voltaic generator and an organic Rankine

cycle. J Eng Gas Turb Power 2012;134:102301–1.

[17] Tchanche BF, Papadakis G, Lambrinos G, et al. Fluid selection for a low-

temperature solar organic Rankine cycle. Appl Therm Eng 2009;29:2468–76.

[18] Butterworth D. Introduction to Heat Transfer. Oxford University Press for

the Design Council, the British Standards Institution and the Council of

Engineering Institutions, 1977.

[19] Chen JC. A correlation for boiling heat transfer to saturated fluids in con-

vective flow. Ind Eng Chem Proc DD 1966;5:322.

[20] Boyko LD, Kruzhilin GN. Heat transfer and hydraulic resistance during con-

densation of steam in a horizontal tube and in a bundle of tubes. Int J Heat

Mass Tran 1967;10:361–73.

[21] Turton R, Bailie RC, Whiting WB, et al. Analysis, Synthesis, and Design of

Chemical Processes. 3rd edn. Prentice Hall, 2009.

[22] ASHRAE. Handbook Fundamentals, Chapter 29, Refrigerants, 2009.

[23] Lemmon EW, McLinden MO, Huber ML. NIST Standard Reference

Database 23: Reference Fluid Thermodynamic and Transport Properties-

REFPROP, Version 8.0. National Institute of Standard and Technology, 2007.

[24] Therminol Heat Transfer Fluids by Solutia Inc. Therminol VP-1. 2012. http://

www.therminol.com/pages/products/vp-1.asp.

M. Ashouri et al.

12 of 12 International Journal of Low-Carbon Technologies 2015, 0, 1–12

by guest on September 3, 2015

http://ijlct.oxfordjournals.org/D

ownloaded from