Design Construction and Preliminary Results of a 250 Kw Organic Rankine Cycle System

sBen-Ran Fu , Yuh-Ren Lee, Jui-Ching Hsieh

h i g h l i g h t s

A 250-kW ORC system using turbine expande The experimentally maximal net power outpu The experimentally maximal system thermal The turbine isentropic efciency was 63.7% w

the hea

Turbine expander219.5 kW with a uctuation of 5.5 kW during prolonged operation. The maximal net power output andsystem thermal efciency were 225 kW and 7.94%, respectively. Under this condition, the isentropic

efcient methods for converting low-grade thermal energy, such asthat derived from waste heat recovery, geothermal and solar ther-mal sources, biomass combined heat and power (CHP), and oceanthermal energy into electricity [2,3]. Previous studies on ORCs have

ls, including con-[1,4], developinging application ofng waste heat re-[8], and at data

control strategy models [11], quasi-dynamic models [12]. In addi-tion, relevant studies have assessed the effect of the optimal pinch-point temperature range of evaporators on system performance[13], conducted prototype testing [14e16], and performed statisti-cal analysis of ORC-related patent data [17] and off-design perfor-mance analysis [18,19]. This section reviews several previousexperimental studies on the ORC systems in detail.* Corresponding author.

Contents lists available at ScienceDirect

Applied Therma

sev

Applied Thermal Engineering 80 (2015) 339e346E-mail address: [email protected] (B.-R. Fu).sources [1]. ORC is considered to be one of themost economical and centers [9], as well as proposing proof-of-concepts [10], optimal1. Introduction

An organic Rankine cycle (ORC) is identical to a steam Rankinecycle, except that it employs organic uids with a low boiling pointas working uids to generate power from low-temperature heat

applied various perspectives and research tooducting technical-economic-market surveysmethods for selecting working uids [5], reviewscroll expanders for ORC systems [6], evaluaticovery from a power plant [7], onboard shipsobvious. 2015 Elsevier Ltd. All rights reserved.Thermal efciencyWaste heat recovery efciency of the turbine was 63.7% with a rotational speed of 12 386 rpm, and the back-work ratio was

6.7%. In addition, the results of the dynamic testing demonstrated that the present system respondedvery rapidly as the heat source temperature changed. The experimental results also demonstrated thatthe system thermal efciency and net power output increased linearly with an increasing heat sourcetemperature. However, the effect of the heat source temperature on the turbine efciency was not The system responded very rapidly as

a r t i c l e i n f o

Article history:Received 3 December 2014Accepted 30 January 2015Available online 11 February 2015

Keywords:Organic Rankine cycle (ORC)http://dx.doi.org/10.1016/j.applthermaleng.2015.01.071359-4311/ 2015 Elsevier Ltd. All rights reserved.r was studied for waste heat recovery.t was 219.5 5.5 kW.efciency was 7.94%.ith a rotational speed of 12,386 rpm.t source temperature changed.

a b s t r a c t

This study involved designing and constructing a 250-kW organic Rankine cycle system, consisting of apump, preheater, evaporator, turbine, generator, condenser, as well as hot and cooling water circulationsystems. Refrigerant R245fa was used as a working uid. The design operating pressure levels of thepreheater/evaporator and condenser was 1.265 MPa and 0.242 MPa, respectively. Under design condi-tions, the net power output was 243 kW and the system thermal efciency was 9.5%. The preliminaryexperimental results under off-design conditions showed that the average net power output wasGreen Energy and Environment Research Laboratories, Industrial Technology Research Institute, Hsinchu 31040, TaiwanResearch paper

Design, construction, and preliminary reRankine cycle system

*

journal homepage: www.el7ults of a 250-kW organic

l Engineering

ier .com/locate/apthermeng

l EnManolakos et al. [20] experimentally evaluated the performanceof a low-temperature solar ORC system for reverse osmosis (RO)desalination using R134a as the working uid and scroll expander.In their experiments, the expander had amaximal efciency of 65%,power output of 2.05 kW, and ORC efciency of 4%. In addition,Manolakos et al. [21] presented on-site experimental results of asolar ORC system combined with an RO system. However, themaximal efciencies of the expander and the ORC system wereapproximately 45% and 1.75%, respectively.

Lemort et al. [22] and Quoilin et al. [23] have experimentallyinvestigated an ORC prototype with an open-drive oil-free scrollexpander and R123 as the working uid. In their experiments, themaximal isentropic effectiveness of the expander was 68%, and theshaft work of the expander was between 0.4 and 1.82 kW. Themaximal system efciency was 7.4%. In addition, they determinedthat the deviation between the experimental and predicted resultsof the proposed model was only approximately 5%.

Wang et al. [24] reported on the experimental results of an on-site micro-scale solar ORC system using R245fa and R245fa/R152mixtures as theworking uids. Themaximal power output and ORCefciency were 7.2 W and 5.59%, respectively. Their resultsdemonstrated that the ORC efciency was considerably improvedby using R245fa/R152 mixtures rather than using pure R245fa.

Wang et al. [25] designed and constructed a low-temperaturesolar ORC system using R245fa as the working uid and a rolling-piston expander. In their experiments, the average shaft poweroutput and isentropic efciency of the expander were 1.73 kW and45.2%, respectively. They achieved a maximal ORC efciency of12.9%.

Pei et al. [26] experimentally examined a kW-scale ORC systemusing a turbine and R123 as the working uid. The results showedthat the isentropic efciency of the turbine was 62.5%; the ORC

Nomenclature

Esys thermal efciency of the ORC system (%)Etur isentropic efciency of the turbine (%)mW mass ow rate of the hot water (kg/s)s mass specic entropy (kJ/kg K)T temperature (C)TC,in inlet temperature of the cooling water (C)TC,out outlet temperature of the cooling water (C)TW,in inlet temperature of the hot water (C)TW,out outlet temperature of the hot water (C)Wnet net power output of the system (kW)

B.-R. Fu et al. / Applied Therma340efciency was 6.8% at a temperature difference of 70 C betweenthe heat source and heat sink, and the power output was 1.36 kW.

Qui et al. [27] constructed and tested a micro-scale biomass-red CHP system with an ORC. Their experimental results showedthat the micro-CHP system with an ORC generated 0.861 kW and47.26 kW in electricity and heat, respectively, and the corre-sponding efciencies of the expander and ORC were 53.92% and3.78%, respectively. Consequently, the overall biomass CHP ef-ciency was 78.69%.

Zheng et al. [28] proposed a kW-scale rolling-piston expanderfor a low-temperature ORC system using R245fa as the workinguid and conducted a running test of the proposed ORC system. Theexperimental results showed that the expander operated at350e800 rpm with a maximal power output of 0.35 kW when theheat source temperature was below 90 C. The maximal isentropicefciency of the expander and cycle efciency were 43.3% and 5%,respectively.Declaye et al. [29] experimentally investigated an ORC systemusing R245fa as the working uid and a scroll expander. Theydemonstrated a maximal isentropic efciency of the expander andshaft power of 75.7% and 2.1 kW, respectively. A maximal cycleefciency of 8.5% was reached at evaporating and condensingtemperatures of 97.5 and 26.6 C, respectively.

Twomey et al. [30] reported on the dynamic performance of asmall-scale solar cogeneration system with an ORC using R134a asthe working uid and a scroll expander. The results demonstrated amaximal isentropic efciency of the expander of 59% with a cor-responding ORC efciency of 3.47%. In addition, the maximal poweroutput was 0.676 kW.

Hsu et al. [16] experimentally investigated the effect of inletpressure and the pressure ratio in the expander on the performanceof a 50-kW ORC system using a screw expander and R245fa as theworking uid. The results showed that for a given pressure ratio,higher inlet pressure resulted in both higher isentropic efciency ofthe expander and higher system efciency under an over-expansion condition, but resulted in lower isentropic efciency ofthe expander and system efciency under an under-expansioncondition. In their experiments, the maximal power output was50 kW and the system thermal efciency was 10.5%.

Jradi and Riffat [31] experimentally examined a small-scale tri-generation system, consisting of an ORC-based (using a scrollexpander and HFE7100 as the working uid) CHP unit and a com-bined dehumidication and cooling unit. They demonstrated thatthis combined system provided approximately 9.6, 6.5, and 0.5 kWfor heating, cooling, and electric power, respectively. Under amaximal electric power output condition, the isentropic efciencyof the expander was 74.2% and the corresponding cycle efciencywas 5.64%.

Avadhanula and Lin [32] proposed empirical models for a screwexpander based on the experimental data of a 50-kW ORC systemusing R245fa as theworking uid. The experimental results showedthat the pressure ratio, volume ratio, and power output of the ORCsystem were 2.70e6.54, 2.54e6.20, and 10e51.5 kW, respectively.In addition, their proposed models predicted the system poweroutput accurately, namely within 10% and 7.5% of experimentalvalues for the polytropic exponent model and isentropic workoutput model, respectively. The maximal isentropic efciency ofthe expander was nearly 70%, but no information of the ORC ef-ciency was provided.

Klonowicz et al. [33] designed and constructed a small-scaleturbine with a rotational speed of 3264 rpm for an ORC system us-ing R227ea as the working uid. In addition, they presented thenumerical and preliminary testing results of the system. Theexperimental power output was 9.9 0.2 kW, and the corre-sponding electrical efciency was 53 2%. However, the ORC ef-ciencywasnotprovided. In addition, the predicted systemefciencyexhibited a high consistency with the experimental result.

Chang et al. [34] experimentally investigated an ORC systemwith a scroll expander and R245fa as the working uid. Themaximal shaft power output and electricity power output were 1.74and 1.375 kW, respectively, at a temperature difference of 60.6 Cbetween the heat source and heat sink. The maximal system ef-ciency of the ORC system was 7.77%.

Zhang et al. [35] reported on the experimental results of an ORCsystemwith a single-screw expander and R123 as the working uidfor waste heat recovery from the exhaust of a diesel engine with amaximal horsepower of 336 kW. The results demonstrated that themaximal ORC power output was 10.38 kW. In addition, the ORCefciency and overall system efciency were 6.48% and 43.8%,respectively, resulting in 250 kW of diesel engine output. Using theORC improved the overall efciency of the diesel engine by

gineering 80 (2015) 339e346approximately 1.53%.

More detailed experimental results of ORC systems from pre-vious studies are summarized in Table 1. The cited studies experi-mentally investigated ORC systems with a power output of lessthan 50 kW. In addition, although numerous refrigerants wereexamined in numerous studies (e.g., on the selection of theworkinguid in an ORC system [5,36]), the typically used working uids inthe experimental ORC systems were R123, R134a, and R245fa.Table 1 indicates that scroll expanders were generally used in low-kW-level ORC systems, which is due to the compact size, low cost,high efciency, and low number of moving parts of scroll expanders[30]; screw-type expanders were used in tens-of-kW-level ORCsystems; the turbine expander type was generally used inhundreds-of-kW-level ORC systems. However, based on literaturereview, no studies have reported on experimental results of largeORC systems (>100 kW) using turbine expanders. As shown inTable 1, ORC thermal efciency was generally lower than 10%.

The present study constructed a 250-kW ORC system (usingR245fa as the working uid), consisting of a pump, preheater,evaporator, turbine, generator, condenser, as well as hot and cool-ing water circulation systems. In addition, this paper presents thepreliminary results of the present ORC system under off-design

was 39 C), respectively. The design points of the heat source (hotwater) temperature andmass ow rate were 133.9 C and 15.39 kg/s, respectively. The details of mathematical model for eachcomponent and system performance were described in the previ-ous studies [18,19], which were preliminary analyses of the off-design performance of the ORC system investigated in this study.Under design conditions, the net power output and system thermalefciency were 243 kW and 9.5%, respectively.

Table 1Some experimental results of the ORC system from available literature (sorted byexpander type).

Authors Expander type(max. efciency, %)

Workinguid

Max.electricityor shaftpower (kW)

Max. ORCthermalefciency(%)

Wang et al. [24] e R245fa/R152a

0.0072 5.59

Qiu et al. [27] Multi-vane (53.92) HFE7000 0.861 3.78Peris et al. [40] Volumetric (65.33) R245fa 15.93 10.88Wang et al. [25] Rolling-piston (45.2) R245fa 1.73 12.9Zheng et al. [28] Rolling-piston (43.3) R245fa 0.35 5Manolakos et al. [20] Scroll (65) R134a 2.05 4Manolakos et al. [21] Scroll (45) R134a 1 1.75Lemort et al. [22] Scroll (68) R123 1.82 7.4Declaye et al. [29] Scroll (75.7) R245fa 2.1 8.5Twomey et al. [30] Scroll (59) R134a 0.676 3.47Jradi and Riffat [31] Scroll (74.2) HFE7100 0.5 5.64Chang et al. [34] Scroll (76) R245fa 1.375 7.77

B.-R. Fu et al. / Applied Thermal Engineering 80 (2015) 339e346 341Peterson et al. [41] Scroll (49.9) R123 0.256 7.2Wang et al. [42] Scroll (77.5) R123 0.625 eMathias et al. [43] Scroll (83) R123 2.96 eLemort et al. [44] Scroll (71.03) R245fa 2.032 eBracco et al. [45] Scroll (74) R245fa 1.5 8Liu et al. [46] Scroll (41) R123 0.76 2.9Zhou et al. [47] Scroll (57) R123 0.645 8.5Saitoh et al. [48] Scroll (65) R113 0.35 11Tarique et al. [49] Scroll (64) R134a 0.92 8.5Li et al. [50] Scroll (83) R245fa/

R601a0.55 4.45

Miao et al. [51] Scroll (81) R123 3.25 6.39Gao et al. [52] Scroll (55.3) R245fa 0.151 3.2Lee et al. [14,15] Screw (65) R245fa 50 8.05Hsu et al. [16] Screw (72.5) R245fa 50 10.5Avadhanula

and Lin [33]Screw (70) R245fa 51.5 e

Zhang et al. [35] Screw (57.88) R123 10.38 6.48Pei et al. [26] Turbine (62.5) R123 1.36 6.8Klonowicz et al. [33] Turbine (59) R227ea 9.9 eNguyen et al. [53] Turbine (49.8) n-Pentane 1.5 4.3Liu et al. [54] Turbine () HFE7000,

HFE71000.284 e

Li et al. [55] Turbine (53) R123 6.57 eKang [56] Turbine (82.2) R245fa 32.7 5.22Li et al. [57] Turbine (68) R123 6 7.98Borsukiewicz-Gozdur Turbine () R227ea 9.87 4.88[58]conditions. The effect of the heat source temperature on systemperformance was investigated.

2. System design and construction

This 250-kW ORC prototype, as shown in Fig. 1, was built at theIndustrial Technology Research Institute, Taiwan. The ORC unitmeasured 450 cm (length) 270 cm (width) 310 cm (height),weighing approximately 11,000 kg. The working uid was refrig-erant R245fa, one of the most appropriate uids for low-gradewaste heat recovery of ORC systems [37]; the mass ow rate was11.58 kg/s. The thermodynamic properties of R245fa were evalu-ated using REFPROP 9.0 [38], which was developed by the NationalInstitute of Standards and Technology (USA). The working uidR245fa circulated on the shell side of the heat exchangers, and thehot and cooling water circulated in the tube side. Fig. 2 depicts thedetailed scheme of the experimental test rig of the system.

2.1. Design conditions

Fig. 3 shows the Tes diagram of the present ORC system. Thedesign operating pressure levels in the preheater/evaporator andcondenser were 1.265 MPa (the evaporation/saturation tempera-ture was 100 C) and 0.242 MPa (the condensation temperature

Fig. 1. Photograph of the 250-kW ORC prototype (pump is located at back side). (1)Preheater, (2) evaporator, (3) turbine, (4) generator, (5) condenser, (6) pump, (A)cooling water inlet, (B) cooling water outlet, (C) hot water inlet, (D) hot water outlet.Fig. 2. Detailed scheme of the experimental test rig.

Shell inside diameter 32.45 cmBundle hole diameter 1.61 cmBundle diameter 31.66 cmSealing strips number 0Nozzle inside diameter 10 cmBafe plate diameter 31.95 cmBafe thickness 0.4 cmBafe spacing 20 cmBafe cut 30%Bafe plate number 17

(b) Evaporator

Tube inside/outside diameter 1.639/1.765 cmTube thickness 0.063 cmTube number 300Tube bundle 4 passTube inside type RiedTube outside type Low-nnedFin height 0.06 cmFin thickness 0.02 cmTube arrangement StaggeredTube pitch transverse 2.375 cmTube pitch longitudinal 2.057 cmTube/Shell length 360 cmDistance upper row/center 1.9 cmTube in upper row 28Number of tube rows 12Free space above upper tube row 47%Shell inside diameter 69.59 cm

(c) Condenser

Tube outside diameter 1.905 cmTube number 480Tube bundle 2 passTube inside type RiedTube outside type Low-nnedTube arrangement StaggeredTube pitch transverse 2.375 cmTube pitch longitudinal 2.090 cmTube/Shell length 360 cmDistance upper row/center 2.23 cmTube in upper row 23Number of tube rows 18

l En2.2. Heat exchangers

The present ORC system featured three heat exchangers, namelya preheater, an evaporator, and a condenser. The preheater,comprising 200 heat transfer tubes, was of a one-pass shell-and-tube type, in which the heat transfer tube was of a ried type in theinside and low-nned type in the tube outside. The evaporator,consisting of 300 heat transfer tubes, was of a four-pass shell-and-tube ooded type. The condenser, composed of 480 heat transfertubes, was of a two-pass shell-and-tube ooded type. The detailedparameters of the used preheater, evaporator, and condenser aresummarized in Table 2.

2.3. Pump, turbine, and generator

The pump used in the present system was manufactured byGrundfos (model: CR 32-5, 60 Hz). The 250-kW turbine wasdesigned at the Industrial Technology Research Institute (Taiwan)and by a SoftInWay Inc. (USA) engineering team, and fabricated atthe Industrial Technology Research Institute. Detailed informationon the turbine performance was provided in a previous study [39].The manufactured blade row is shown in Fig. 4a, and the rotorwheel mounted with a stator ring and shaft is shown in Fig. 4b. Therotational speed of the turbine was 12,000 10% rpm with anisentropic efciency of approximately 80%. The shaft powergenerated by the turbinewas transferred to the generator through agearbox at a rotational speed of 3600 rpm; the generator efciencywas 90%. To prevent leakage of the working uid, the turbine,gearbox, and generator were arranged in a hermetic space. Thistype of hermetic turbogenerator was also employed by Klonowiczet al. [33].

2

cooling water

hot water

5s

43

2s1T

empe

ratu

re (o

C)

Mass specific entropy (kJ/kgK)

5

Fig. 3. Tes diagram of the studied ORC system.

B.-R. Fu et al. / Applied Therma3422.4. Hot and cooling water circulation systems

Hot water, considered the waste heat source, was supplied bythe pressurized hot water boiler with a maximal capacity of3788 kW, as shown in Fig. 5a. In the currently proposed system,natural gas was used as a fuel for the boiler. Heated water wascollected in the container, and owed into the ORC system throughthe piping. The cooling water circulation system consisted of two500-RT cooling towers and had a total maximal capacity of3860 kW, as shown in Fig. 5b.

2.5. Measurement and control

Pressure transducers with the full scale range of 2500 kPa,manufactured by WIKA (model A-10, Germany), and Pt100 (threewires) thermometers, designed by MorShine (Taiwan), werelocated at each point of the system (i.e., Points 1 to 5 in Fig. 3). TheTable 2Detailed parameters of the used heat exchangers.

(a) Preheater

Tube inside/outside diameter 1.471/1.587 cmTube thickness 0.058 cmTube number 200Tube in window 83Tube bundle 1 passTube inside type RiedTube outside type/Fin per inch (FPI) Low-nned/42Tube arrangement StaggeredTube pitch transverse 1.984 cmTube pitch longitudinal 1.718 cmTube/Shell length 360 cm

gineering 80 (2015) 339e346mass ow rate of the working uid was measured at Point 2 using avortex ow meter made by BNC (model 1100/LRT-B/A, Taiwan). Inaddition, level meters were employed in the evaporator andcondenser. The rotational speed of the turbine was measured usinga tachometer sensor. Themass ow rate and temperature of the hotand cooling water were measured using an electromagnetic owmeter made by BNC (model BMS1000-A100, Taiwan) in the hot andcooling water circulation systems. The electric power of the in-duction generator was measured using a power analyzer made byHIOKI (model 3169-20 with a 9661 clamp on sensor, Japan). Theoverall system control (including the processes of startup, shut-down, and emergency turn off) and data acquisition were per-formed using an AB MicroLogix 1500 programmable logiccontroller (USA). The measured data were then transmitted to a

Shell inside diameter 71.7 cm

personal computer for detailed system analysis. The measurementuncertainties in temperature, pressure, mass ow rate of theworking uid, mass ow rate of the water, heat transfer rate, tur-bine efciency, system thermal efciency, and net power output are0.25 C, 1.0%, 0.7%, 0.5%, 1.72%, 1.48%, 3.10%, and 2.36%,respectively.

3. Preliminary results and discussion

Fig. 6 shows the measured inlet and outlet temperatures of thehot and cooling water and the system net power output (Wnet)

experiments, we increased themass ow rate of the hot water from15.39 kg/s to 26.70 kg/s to enable the heat transfer rate to meet therequirement. In addition, because of a lower heat source temper-ature, a higher evaporation temperature (106 C) was applied tomaintain the power output level as design value.

Fig. 6 also shows that the uctuations of the heat source andsink temperatures were considerably low (1 C), indicating thatthe boiler and cooling towers were functioning adequately; theaverage net power output was 219.5 kW with a uctuation of5.5 kW (approximately 2.5% of the average net power output). Themaximal net power output and system thermal efciency were

Fig. 4. (a) Manufactured rotor wheel and (b) rotor wheel mounted with stator ring and shaft.

B.-R. Fu et al. / Applied Thermal Engineering 80 (2015) 339e346 343during prolonged operation. In this gure, TW,in and TW,out are theinlet and outlet temperatures of the hot water, respectively; andTC,in and TC,out are the inlet and outlet temperatures of the coolingwater, respectively. The ORC system required approximately 30min(excluding the period for the boiler to warm up before the exper-iments) to reach a steady-state condition and was longer than that(approximately 15 min) required for the 50-kW ORC system in ourprevious study [14]. When the system was under a steady-statecondition, the average inlet temperature of hot water was119.2 C; however, this was lower than it should have been ac-cording to the design (i.e., 133.9 C). This was due to our infra-structure limiting the supply of natural gas. Therefore, during theFig. 5. Hot and cooling water circulation systems. (a) H225 kW and 7.94%, respectively. Under the maximal net poweroutput condition, the isentropic efciency of the turbine was 63.7%with a rotational speed of 12,386 rpm. The pump and turbine ef-ciencies were lower than the designed value; the generator ef-ciency reached 91.5%. In addition, the back-work ratio, dened asthe ratio between the works consumed by the pump and producedby the expander, was 6.7%. The heat transfer rates in the preheaterand evaporator were 1161 and 1560 kW, respectively, close to thedesigned values. Heat loss in the preheater and evaporator was113 kW, which was approximately 4% of the heat input from the hotwater. The temperatures of two particular cycle points, namely thesuperheat at the evaporator outlet and the subcooling at theot water system, (b) two 500-RT cooling towers.

condenser outlet, which were not considered in the previousanalysis of the design condition, were 1.7 and 2.5 C, respectively. Aslight degree of superheat was applied to ensure that no liquid-phase working uid entered the turbine. The detailed experi-mental results at themaximal power output and design parameters

0 60 120 180 240 300 3600

50

100

150

200

250

Wnet

TC,outTC,in

TW,out

TW,in

T (

C)

Time (min)

0

50

100

150

200

250

Wne

t (kW

)

Fig. 6. Temperatures of hot and cooling water and system net power output.

0 60 120 180 24080

90

100

110

120

(d)(c)(b)(a)

Wnet

TW,out

TW,in

T (

C)

Time (min)

100

150

200

250

Wne

t (kW

)

Fig. 7. Dynamic behavior of the system.

B.-R. Fu et al. / Applied Thermal Engineering 80 (2015) 339e346344are summarized in Table 3.After prolonged operation, a dynamic testing was also con-

ducted. Fig. 7 shows the dynamic behavior of the system under thecontinued change in the inlet temperature of the heat source, i.e.,TW,in. During the dynamic testing, there were 4 periods, which theoperation conditions were described as follows: (a) TW,in decreasedrapidly from the steady-averaged value of 119.2 Ce102.7 C; (b)TW,in increased rapidly from 102.7 C to 115.0 C; (c) TW,in decreasedslowly from 115.0 C to 102.7 C; and (d) TW,in increased slowly from102.7 C to 109.4 C. The corresponding dynamic responses of thesystem net power output (Wnet) were described as follows: (a)Wnetdecreased rapidly from 223.4 kW to 132.0 kW; (b) Wnet increasedrapidly from 132.0 kW to 197.9 kW; (c)Wnet decreased slowly from197.9 kW to 135.8 kW; and (d)Wnet increased slowly from 135.8 kWto 169.7 kW. In addition, the outlet temperature of the heat source,i.e., TW,out, shows the same dynamic behavior with Wnet. Mostimportantly, those dynamic results demonstrated that the presentsystem responded very rapidly as the heat source temperaturechanged.

Table 3Detailed experimental results at the maximum net power output.Parameter Design Experiments

Mass ow rate of hot water (kg/s) 15.39 26.70Inlet temperature of hot water (C) 133.9 119.8Outlet temperature of hot water (C) 94.4 94.7Mass ow rate of R245fa (kg/s) 11.58 11.85Evaporation temperature (C) 100 106Evaporation pressure (MPa) 1.265 1.440Superheat at evaporator outlet (C) 0 1.7Total heat input from hot water (kW) 2573 2834Heat transfer rate in the preheater (kW) 1016 1161Heat transfer rate in the evaporator (kW) 1557 1560Heat loss in the preheater and evaporator (kW) 0 113Condensation temperature (C) 39 40Condensation pressure (MPa) 0.242 0.251Subcooling at condensation outlet (C) 0 2.5Rotational speed of turbine (rpm) 12,000 12386Back work ratio (%) 4.1 6.7Pump efciency (%) 90 68.1Turbine efciency (%) 80 63.7Generator efciency (%) 90 91.5ORC thermal efciency (%) 9.5 7.94Net power output (kW) 243 225Furthermore, we studied the effect of the heat source temper-ature (TW,in) on system performance, as shown in Fig. 8. Theexperimental results show that as TW,in increased from 102.4 to119.8 C, the isentropic efciency of the turbine slightly increasedfrom 58.8% to 63.7%; the net power output increased linearly andsubstantially from 135 to 225 kW. We observed such a linear in-crease of net power output in our previous off-design analysis [18].Furthermore, the system thermal efciency increased from 6.31% to7.94% for the studied range of TW,in, indicating that the overallthermal efciency increased by 0.94%, as the heat source temper-ature increased by 10 C. The increase in thermal efciency wasconsistent with previous analytical results [18], which demon-strated a 0.91% efciency increase for a 10 C of TW,in.

4. Conclusions

The present study involved designing and constructing a 250-kW ORC system, consisting of a pump, preheater, evaporator, tur-bine, generator, condenser, as well as hot and cooling water cir-culation systems. The ORC unit measured 450 cm (length) 270 cm(width) 310 cm (height), weighing approximately 11,000 kg.Refrigerant R245fa was used as the working uid. The designoperating pressure levels of the preheater/evaporator andcondenser were 1.265 MPa and 0.242 MPa, respectively. Underdesign conditions, the net power output was 243 kW and the sys-tem thermal efciency was 9.5%.

The preliminary experimental results of the ORC system underoff-design conditions showed that the average net power outputwas 219.5 kW with a uctuation of 5.5 kW during prolonged100 105 110 115 120 1250

50

100

150

200

250

Etur

Esys

Wnet

Wne

t (kW

) or E

tur (

%)

TW,in ( C)

5

6

7

8

9

E sys

(%)

Fig. 8. Net power output, turbine and system efciencies as a function of heat sourcetemperature.

al Enoperation. The maximal net power output and system thermal ef-ciency were 225 kW and 7.94%, respectively, at evaporation andcondensation temperatures of 106 and 40 C. Under this condition,the isentropic efciency of the turbine was 63.7% with a rotationalspeed of 12,386 rpm, and the back-work ratio was 6.7%. In addition,the dynamic behavior of the system under the continued change inthe inlet temperature of the heat source was also studied. The re-sults of the dynamic testing demonstrated that the present systemresponded very rapidly as the heat source temperature changed.The experimental results also demonstrated that the system ther-mal efciency and net power output increased linearly and sub-stantially with an increase of the heat source temperature.However, the effect of the heat source temperature on the turbineefciency was not obvious.

Future study will focus on obtaining more experimental resultsand constructing a map of system performance under off-designheat source conditions. Such research could provide operationalguidelines and indicate an optimal control strategy for off-designoperation. In addition, we will also develop an articial-intelligence feedback control system for the ORC unit.

Acknowledgements

The authors express their gratitude for the Energy R&D foun-dation funding from the Bureau of Energy of the Ministry of Eco-nomic Affairs, Taiwan, under the grant number of 104-E0207.

References

[1] S. Quoilin, M.V.D. Broek, S. Declaye, P. Dewallef, V. Lemort, Techno-economicsurvey of organic Rankine cycle (ORC) systems, Renew. Sustain. Energy Rev.22 (2013) 168e186.

[2] B.F. Tchanche, Gr Lambrinos, A. Frangoudakis, G. Papadakis, Low-grade heatconversion into power using organic Rankine cycles e a review of variousapplications, Renew. Sustain. Energy Rev. 15 (2011) 3963e3979.

[3] J. Wang, Z. Yan, M. Wang, S. Maa, Y. Dai, Thermodynamic analysis and opti-mization of an (organic Rankine cycle) ORC using low grade heat source,Energy 49 (2013) 356e365.

[4] F. Velez, J.J. Segovia, M.C. Martn, G. Antoln, F. Chejne, A. Quijano, A technical,economical and market review of organic Rankine cycles for the conversion oflow-grade heat for power generation, Renew. Sustain. Energy Rev. 16 (2012)4175e4189.

[5] J. Bao, L. Zhao, A review of working uid and expander selections for organicRankine cycle, Renew. Sustain. Energy Rev. 24 (2013) 325e342.

[6] P. Song, M. Wei, L. Shi, S.N. Danish, C. Ma, A review of scroll expanders fororganic Rankine cycle systems, Appl. Therm. Eng. 75 (2015) 54e64.

[7] D. Gewald, K. Siokos, S. Karellas, H. Spliethoff, Waste heat recovery from alandll gas-red power plant, Renew. Sustain. Energy Rev. 16 (2012)1779e1789.

[8] G. Shu, Y. Liang, H. Wei, H. Tian, J. Zhao, L. Liu, A review of waste heat recoveryon two-stroke IC engine aboard ships, Renew. Sustain. Energy Rev. 19 (2013)385e401.

[9] K. Ebrahimi, G.F. Jones, A.S. Fleischer, A review of data center cooling tech-nology, operating conditions and the corresponding low-grade waste heatrecovery opportunities, Renew. Sustain. Energy Rev. 31 (2014) 622e638.

[10] S. Aghahosseini, I. Dincer, Comparative performance analysis of low-temperature organic Rankine cycle (ORC) using pure and zeotropic workinguids, Appl. Therm. Eng. 54 (2013) 35e42.

[11] G. Manente, A. Toffolo, A. Lazzaretto, M. Paci, An organic Rankine cycle off-design model for the search of the optimal control strategy, Energy 58(2013) 97e106.

[12] M.O. Bamgbopa, E. Uzgoren, Quasi-dynamic model for an organic Rankinecycle, Energy Convers. Manag. 72 (2013) 117e124.

[13] Y.R. Li, J.N. Wang, M.T. Du, S.Y. Wu, C. Liu, J.L. Xu, Effect of pinch point tem-perature difference on cost-effective performance of organic Rankine cycle,Int. J. Energy Res. 39 (2013) 1952e1962.

[14] Y.R. Lee, C.R. Kuo, C.C. Wang, Transient response of a 50 kW organic Rankinecycle system, Energy 48 (2012) 532e538.

[15] Y.R. Lee, C.R. Kuo, C.H. Liu, B.R. Fu, J.C. Hsieh, C.C. Wang, Dynamic response of a50 kW organic Rankine cycle system in association with evaporators, Energies7 (2014) 2436e2448.

[16] S.W. Hsu, H.W.D. Chiang, C.W. Yen, Experimental investigation of the per-formance of a hermetic screw-expander organic Rankine cycle, Energies 7(2014) 6172e6185.

B.-R. Fu et al. / Applied Therm[17] B.R. Fu, S.W. Hsu, C.H. Liu, Y.C. Liu, Statistical analysis of patent data relating tothe organic Rankine cycle, Renew. Sustain. Energy Rev. 39 (2014) 986e994.[18] B.R. Fu, S.W. Hsu, Y.R. Lee, J.C. Hsieh, C.M. Chang, C.H. Liu, Effect of off-designheat source temperature on heat transfer characteristics and system perfor-mance of a 250-kW organic Rankine cycle system, Appl. Therm. Eng. 70(2014) 7e12.

[19] B.R. Fu, S.W. Hsu, Y.R. Lee, J.C. Hsieh, C.M. Chang, C.H. Liu, Performance of a250 kW organic Rankine cycle system for off-design heat source conditions,Energies 7 (2014) 3684e3694.

[20] D. Manolakos, G. Papadakis, S. Kyritsis, K. Bouzianas, Experimental evaluationof an autonomous low-temperature solar Rankine cycle system for reverseosmosis desalination, Desalination 203 (2007) 366e374.

[21] D. Manolakos, G. Kosmadakis, S. Kyritsis, G. Papadakis, On site experimentalevaluation of a low-temperature solar organic Rankine cycle system for ROdesalination, Sol. Energy 83 (2009) 646e656.

[22] V. Lemort, S. Quoilin, C. Cuevas, J. Lebrun, Testing and modeling a scrollexpander integrated into an organic Rankine cycle, Appl. Therm. Eng. 29(2009) 3094e3102.

[23] S. Quoilin, V. Lemort, J. Lebrun, Experimental study and modeling of anorganic Rankine cycle using scroll expander, Appl. Energy 87 (2010)1260e1268.

[24] J.L. Wang, L. Zhao, X.D. Wang, A comparative study of pure and zeotropicmixtures in low-temperature solar Rankine cycle, Appl. Energy 87 (2010)3366e3373.

[25] X.D. Wang, L. Zhao, J.L. Wang, W.Z. Zhang, X.Z. Zhao, W. Wu, Performanceevaluation of a low-temperature solar Rankine cycle system utilizing R245fa,Sol. Energy 84 (2010) 353e364.

[26] G. Pei, J. Li, Y. Li, J. Ji, Construction and dynamic test of a small scale organicRankine cycle, Energy 36 (2011) 3215e3223.

[27] G. Qiu, Y. Shao, J. Li, H. Liu, S.B. Riffat, Experimental investigation of a biomass-red ORC-based micro-CHP for domestic applications, Fuel 96 (2012)374e382.

[28] N. Zheng, L. Zhao, X.D. Wang, Y.T. Tan, Experimental verication of a rolling-piston expander that applied for low-temperature organic Rankine cycle,Appl. Energy 112 (2013) 1265e1274.

[29] S. Declaye, S. Quoilin, L. Guillaume, V. Lemort, Experimental study on an open-drive scroll expander integrated into an ORC (organic Rankine cycle) systemwith R245fa as working uid, Energy 55 (2013) 179e183.

[30] B. Twomey, P.A. Jacobs, H. Gurgenci, Dynamic performance estimation ofsmall-scale solar cogeneration with an organic Rankine cycle using a scrollexpander, Appl. Therm. Eng. 51 (2013) 1307e1316.

[31] M. Jradi, S. Riffat, Experimental investigation of a biomass-fuelled micro-scaletri-generation system with an organic Rankine cycle and liquid desiccantcooling unit, Energy 71 (2014) 80e93.

[32] V.K. Avadhanula, C.S. Lin, Empirical models for a screw expander based onexperimental data from organic Rankine cycle system testing, J. Eng. GasTurbines Power Trans. ASME 136 (2014) 062601.

[33] P. Klonowicz, A. Borsukiewicz-Gozdur, P. Hanausek, W. Kryowicz,D. Brggemann, Design and performance measurements of an organic vapourturbine, Appl. Therm. Eng. 63 (2014) 297e303.

[34] J.C. Chang, C.W. Chang, T.C. Hung, J.R. Lin, K.C. Huang, Experimental study andCFD approach for scroll type expander used in low-temperature organicRankine cycle, Appl. Therm. Eng. 73 (2014) 1444e1452.

[35] Y.Q. Zhang, Y.T. Wu, G.D. Xia, C.F. Ma, W.N. Ji, S.W. Liu, K. Yang, F.B. Yang,Development and experimental study on organic Rankine cycle system withsingle-screw expander for waste heat recovery from exhaust of diesel engine,Energy 77 (2014) 499e508.

[36] Z.Q. Wang, N.J. Zhou, J. Guo, X.Y. Wang, Fluid selection and parametric opti-mization of organic Rankine cycle using low temperature waste heat, Energy40 (2012) 107e115.

[37] E.H. Wang, H.G. Zhang, B.Y. Fan, M.G. Ouyang, Y. Zhao, Q.H. Mu, Study ofworking uid selection of organic Rankine cycle (ORC) for engine waste heatrecovery, Energy 36 (2011) 3406e3418.

[38] E.W. Lemmon, M.L. Huber, M.O. McLinden, NIST Standard Reference Database23: Reference Fluid Thermodynamic and Transport PropertieseREFPROP,Version 9.0, National Institute of Standards and Technology, Standard Refer-ence Data Program, Gaithersburg, USA, 2010.

[39] L. Moroz, C.R. Kuo, O. Guriev, Y.C. Li, B. Frolov, Axial turbine ow path designfor an organic Rankine cycle using R-245fa, in: Proceedings of ASME TurboExpo 2013: Turbine Technical Conference and Exposition, Paper No.GT2013e94078, pp. V05AT23A004, San Antonio, Texas, USA, June 3e7, 2013.

[40] B. Peris, J. Navarro-Esbr, F. Moles, R. Collado, A. Mota-Babiloni, Performanceevaluation of an organic Rankine cycle (ORC) for power applications from lowgrade heat sources, Appl. Therm. Eng. 75 (2015) 763e769.

[41] R.B. Peterson, H. Wang, T. Herron, Performance of a small-scale regenerativeRankine power cycle employing a scroll expander, Proc. Ins. Mech. Eng. Part AJ. Power Energy 222 (2008) 271e282.

[42] H. Wang, R.B. Peterson, T. Herron, Experimental performance of a compliantscroll expander for an organic Rankine cycle, Proc. Ins. Mech. Eng. Part A J.Power Energy 223 (2009) 863e872.

[43] J.A. Mathias, J.J.R. Johnston, J. Cao, D.K. Priedeman, R.N. Christensen, Experi-mental testing of gerotor and scroll expanders used in, and energetic andexergetic modeling of, an organic Rankine cycle, J. Energy Resour. Technol.Trans. ASME 131 (2009) 012201.

[44] V. Lemort, S. Declaye, S. Quoilin, Experimental characterization of a hermetic

gineering 80 (2015) 339e346 345scroll expander for use in a micro-scale Rankine cycle, Proc. Inst. Mech. Eng.Part A J. Power Energy 226 (2012) 126e136.

[45] R. Bracco, S. Clemente, D.Micheli,M. Reini, Experimental tests andmodelizationof a domestic-scale ORC (organic Rankine cycle), Energy 58 (2013) 107e116.

[46] G. Liu, Y. Zhao, Q. Yang, L. Wang, B. Tang, L. Li, Theoretical and experimentalresearch on scroll expander used in small scale organic Rankine cycle System,Proc. Inst. Mech. Eng. Part E J. Process Mech. Eng. 229 (2015) 25e35.

[47] N. Zhou, X. Wang, Z. Chen, Z. Wang, Experimental study on organic Rankinecycle for waste heat recovery from low-temperature ue gas, Energy 55(2013) 216e225.

[48] T. Saitoh, N. Yamada, S. Wakashima, Solar Rankine cycle system using scrollexpander, J. Environ. Eng. 2 (2007) 708e719.

[49] M.A. Tarique, I. Dincer, C. Zamrescu, Experimental investigation of a scrollexpander for an organic Rankine cycle, Int. J. Energy Res. 38 (2014) 1825e1834.

[50] T. Li, J. Zhu, W. Fu, K. Hu, Experimental comparison of R245fa and R245fa/R601a for organic Rankine cycle using scroll expander, Int. J. Energy Res. 39(2015) 202e214.

[51] Z. Miao, J. Xu, X. Yang, J. Zou, Operation and performance of a low temperatureorganic Rankine cycle, Appl. Therm. Eng. 75 (2015) 1065e1075.

[52] P. Gao, L. Jiang, L.W. Wang, R.Z. Wang, F.P. Song, Simulation and experimentson an ORC system with different scroll expanders based on energy and exergyanalysis, Appl. Therm. Eng. 75 (2015) 880e888.

[53] V.M. Nguyen, P.S. Doherty, S.B. Riffat, Development of a prototype low tem-perature Rankine cycle electricity generation system, Appl. Therm. Eng. 21(2001) 169e181.

[54] H. Liu, G. Qiu, Y. Shao, F. Daminabo, S.B. Riffat, Preliminary experimental in-vestigations of a biomass-red micro-scale CHP with organic Rankine cycle,Int. J. Low-Carbon Technol. 5 (2010) 81e87.

[55] M. Li, J. Wang, L. Gao, X. Niu, Y. Dai, Performance evaluation of a turbine usedin a regenerative organic Rankine cycle, in: Proceedings of ASME Turbo Expo2012: Turbine Technical Conference and Exposition, Paper No. GT2012-68441,June 11e15, 2012, pp. 425e432. Copenhagen, Denmark.

[56] S.H. Kang, Design and experimental study of ORC (organic Rankine cycle) andradial turbine using R245fa working uid, Energy 41 (2012) 514e524.

[57] M. Li, J. Wang, W. He, L. Gao, B. Wang, S. Ma, Y. Dai, Construction and pre-liminary test of a low-temperature regenerative organic Rankine cycle (ORC)using R123, Renew. Energy 57 (2013) 216e222.

[58] A. Borsukiewicz-Gozdur, Experimental investigation of R227ea applied asworking uid in the ORC power plant with hermetic turbogenerator, Appl.Therm. Eng. 56 (2013) 126e133.

B.-R. Fu et al. / Applied Thermal Engineering 80 (2015) 339e346346

Design, construction, and preliminary results of a 250-kW organic Rankine cycle system1. Introduction2. System design and construction2.1. Design conditions2.2. Heat exchangers2.3. Pump, turbine, and generator2.4. Hot and cooling water circulation systems2.5. Measurement and control

3. Preliminary results and discussion4. ConclusionsAcknowledgementsReferences