lable at ScienceDirect

Applied Thermal Engineering 67 (2014) 35e42

Contents lists avai

Applied Thermal Engineering

journal homepage: www.elsevier .com/locate/apthermeng

Analysis of plate multi-effect distillation system coupled with thermalpower generating unit

Zhihua Ge, Chunjie Yang, Yun Liu, Xiaoze Du*, Lijun Yang, Yongping YangKey Laboratory of Condition Monitoring and Control for Power Plant Equipment, North China Electric Power University, Ministry of Education,Beijing 102206, China

h i g h l i g h t s

� The practical potential of plate exchanger as evaporator in MED seawater desalination system.� Physico-mathematical model is developed for plate MED system coupled with thermal power generating unit.� The design of MED system with different plate evaporators is conducted and compared with tubular system.� Off-design performance of plate MED system is obtained under various conditions of feed brine and extracting vapor from turbine.

a r t i c l e i n f o

Article history:Received 25 December 2013Accepted 28 February 2014Available online 12 March 2014

Keywords:Seawater desalinationLow-temperature multi-effect distillationPlate heat exchangerFlow and heat transferOff-design operation

* Corresponding author. Tel.: þ86 (10)61773918; faE-mail address: duxz@ncepu.edu.cn (X. Du).

http://dx.doi.org/10.1016/j.applthermaleng.2014.02.071359-4311/� 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

In order to explore the practical potential of plate exchanger as evaporator in multi-effect distillation(MED) system, the physico-mathematical model is established for a plate MED system coupled withthermal power generating unit. A MED seawater desalination system with shell and tube evaporator inoperation producing freshwater 10,000 ton/day is selected as comparison. The system design withdifferent types of plate evaporators and off-design operating performance are conducted. The resultsindicate that the overall area can be significantly reduced with plate evaporator compared with that ofshell and tube evaporator. The plate MED system can achieve a greater gain of ratio (GOR) than that oforiginal tubular system by selecting suitable entrainment ratio of the last effect. The performance of theplate MED system can be improved by reducing the length of single plate of evaporators, for morechannels with shorter plate can enhance heat transfer in evaporator and restrain the boiling pointevaluation of evaporating brine. Influences of inlet temperature and flow rate of feed brine, flow rate ofextracting source vapor from steam turbine under off-design conditions on the performance of plateMED system are investigated either. The investigations may provide a reference for the application ofplate MED technology coupled with waste heat utilization of thermal power generating unit.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The energy consumption is the core portion of the cost of waterproduction by large-scale seawater desalination. On the other hand,thermal power generation is one of the most efficient technologiesof fossil fuel utilization at present. Nevertheless, most of the energyis exhausted into the environment as the low-grade waste heatduring thermal power generation, leading to tremendous energywaste. It has been found that the energy demand for seawaterdesalination and the residual heat of thermal power plant can be

x: þ86 (10)61773877.

2

complementary, and the energy efficiency can be increased from40% to 60% by combined water and electric power generation [1].

Dominating thermal technologies of seawater desalinationinclude that of multi-stage flash (MSF) and multi-effect distillation(MED). The research of MSF technology at present is mainly focuson the further improvement of system performance and connec-tion with reverse osmosis (RO) technology, etc. El-Dessouky et al.[2] proposed a new structure of multi-stage flash (MSF) system forreducing the residual heat taken away by seawater, increasing theutilization of the energy. El Din et al. [3] discussed some principlesof MSF systems design to avoid or reduce the effects of corrosion,improve efficiency and performance. Maheshwari et al. [4] indi-cated that the combine of multi-stage flash (MSF) and reverseosmosis (RO) is an effective way to improve power plant operatingefficiency and reduce energy consumption. Recently, multi-effect

Nomenclature

A heat transfer area, m2

cp heat capacity at constant pressure, J/(kg K)D0 flow rate of extracting vapor from turbine, kg/s (or t/h)h heat transfer coefficient, W/(m2 K)K overall heat transfer coefficient, W/(m2 K)L the width of plate, mM mass flow rate, kg/s (or t/h)_m mass flow rate, kg/s (or t/h)Rf fouling resistance, m2 K/WT temperature, KDTm logarithmic mean temperature differencex flash ratez length, mg latent heat of vaporization, J/kg

d thickness, ml thermal conductivity, W/(m K)

SubscriptB brinecon condensatione effecteva evaporationi ith effectin inletp platepre preheats saturationV vaporW water

Fig. 1. Schematic diagram of the plate heat exchanger.

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e4236

distillation (MED) has become a priority because of its high gain ofratio (GOR) compared with MSF [5]. The techno-economic analysishas revealed that the multi-effect distillation (MED) with reverseosmosis (RO) (MED-RO) technology has superiority on improvingenergy efficiency and reducing freshwater cost while integratingseawater desalination with gas-steam combined cycle [1,6]. Andthemulti-effect distillation coupled with the low-grade heat sourceof thermal power plant has higher economical efficiency than thatby reverse osmosis [7].

Falling film evaporation has a lot of advantages, including that ofshort contact time, high heat flux, low static pressure loss and lowtemperature difference of heat transfer [8,9]. These features endowMED seawater desalination with excellent water production per-formance. The inherent defect of MED is that the heating andevaporation of seawater are at the same process, which leads tofouling on seawater heat transfer surface. Low temperature MED(LT-MED) with medium top brine temperature (TBT) lower than70 �C can suppress fouling to some extent. However, whilecombining MED desalination and thermal power generating unit,significant exergy loss by throttling can be resulted in order toreduce the high temperature and pressure of extracting sourcesteam from turbine. At the same time, the increase of effects islimited by the top brine temperature, and hence the gain of ratio(GOR) of seawater desalination is reduced. These contradictionsbecome the main technical bottlenecks for the improvement ofseawater desalination performance by MED approach.

Plate heat exchanger uses the corrugated plate as the main heattransfer element. As shown in Fig. 1, the heat exchanger is made upby series of plates side by side. Cold and hot fluids flow in differentsides of the plate to complete heat exchange [10]. Plate heatexchanger has been widely used in the corrosive operating envi-ronment with high temperature and high pressure. High specificsurface area endows the plate heat exchanger with high heattransfer performance, because of which, the temperature differenceof plate heat exchanger used as seawater evaporator can be smallenough and hence inhibits brine fouling to some extent [10,11].Combing plate falling film evaporator with technique of suitableseawater pretreatment and metal material surface process, high-temperature multi-effect distillation technology coupled withcoal-fired power generating unit may be of great potential [12e15].

In the present study, base on the design parameters of anoperating low temperature MED (LT-MED) seawater desalinationsystem coupled with 2 � 300 MW coal-fired power generatingunits, physico-mathematical model with plate falling film evapo-rator is established for the optimization of the seawater

desalination system and operating performance analysis under off-design conditions. The results are compared with the performanceof the original seawater desalination system with shell and tubeevaporators. The investigations may provide a reference for thedesign of the multi-effect distillation systemwith plate evaporatorscoupled with waste heat utilization of thermal power generatingunit.

2. Physico-mathematical model

Fig. 2 shows the device flowchart of LT-MED seawater desali-nation system original designed by SIDEM Co. The system with nopreheater uses the process of spray falling brine film distillationoutside the horizontal tube bundles. It includes four effects, inwhich the temperature and pressure decrease successively fromleft to right, as shown in Fig. 2. The heat source of the first effectcontains the vapor extraction from steam turbine of power gener-ating unit and the vapor ejector from the last effect. The pressure is0.55MPa and the temperature is 320 �C of the vapor extracted fromthe turbine. The brine film is heated to the top temperature in eacheffect and releases secondary vapor, which condenses inside thehorizontal tube bundles for heating the brine in next effect. Sec-ondary vapor condenses to produce freshwater. Producing water ineach effect flows to flash tank with lower pressure, which flashes

Fig. 2. Schematic diagram of LT-MED seawater desalination system.

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e42 37

and also generates secondary vapor for the next effect. Concen-trated brine in the former effect flow to the flash tank in the bottomof next effect with lower pressure, which flashes and generatessecondary vapor either. The concentrated brine is discharged in lasteffect. The excess vapor in last effect flow to the condenser and becooled into producing water.

Thermal vapor compressor (TVC) is employed in the system, bywhich, the pressure of the secondary vapor entrained from the lasteffect can be promoted by the source vapor extracted from steamturbine for recycling utilization. It can also reduce the temperaturedifference in the first effect. This seawater desalination systemproducing freshwater 10,000 ton/day has been applied in a thermalpower plant of China, of which GOR is 8.33.

Three types of chevron plate heat exchangers are chosen as thebrine evaporator for comparing with the above-mentionedseawater desalination system, which are indicated by BRS1.2,BRS1.8 and BRS2.2, respectively. B stands for plate exchanger. Rstands for chevron bellows. S stands for vertical bellows. Verticalchevron plate which has sine shape and chevron angle is a kind ofcorrugated plate. Table 1 shows the main technical parameters ofthe three different plate heat exchangers.

Vertical chevron plates are assembled to form the plate evapo-rator for seawater desalination with down-flow arrangement bothfor source vapor and brine, as illustrated in Fig. 1. Feed brine ispumped and passes through the spray distributor at the inlet ofplate corrugated heat transfer element. Then it forms a continuousuniform thin film on the surface of vertical chevron plate. Sourcevapor imported from the other side of the plate condenses to heatthe brine, which evaporates and produces secondary vapor as theheat source of the next effect. Except for the plate evaporators, theother designs of the system flowcharts are the same as that shownin Fig. 2.

Mathematical model for performance analysis of plate fallingfilm evaporation is established as follows.

Table 1Main parameters of the plate heat exchanger.

BRS1.2 BRS1.8 BRS2.2

Plate length, H, m 1.64 2.16 2.48Plate width, L, m 1 1 1.25Plate thickness, m 0.0007 0.0007 0.0007Chevron angel 60� 60� 60�

Area of single plate, m2 1.2 1.8 2.2Chevron height, mm 0.0033 0.0033 0.0033

In present paper, the following assumptions have been adopted:

(1) The environmental heat loss is negligible in each effect of thesystem.

(2) The influences of incondensable gas both on evaporation andcondensation of the plate heat exchanger are negligible.

(3) The function of shearing force caused by flowing of vapor onthe phase interface is negligible.

(4) The decrease of pressure along flow directions both for feedbrine and source vapor in plate evaporator is negligible.However, the variations of physical properties caused bytemperature and concentration of brine are taken intoaccount.

Preset the inlet parameters both of brine and vapor, includingthat of initial temperature and flow rate. Assuming that the numberof channels of each plate evaporator is 2ni, the area of the effect canbe obtained by (2ni þ 1)Ae. The inlet flow of vapor and brine in eachchannel can be given respectively as follow, _mBi;0 ¼ _MBi;0=ni,_mVi;0 ¼ _MVi;0=ni. The heat and heat transfer area needed forheating brine to the saturated temperature in each effect areexpressed respectively by,

QBi;pre ¼ _mBi;0�hBi;s � hBi;in

�(1)

Ai;pre ¼ QBi;pre

Ki;preDtmi;pre(2)

by which, the total preheat area of each effect is obtained by niAi,pre.Dtmi,pre is the logarithmicmean temperature difference of brine andvapor in preheating section,

Dtmi;pre ¼ tBi;s � tBi;0log tVi�tBi;0

tVi�tBi;s

(3)

The overall heat transfer coefficient in preheating section, Ki,pre,is expressed by,

Ki;pre ¼ 1dplpþ Rf þ 1

hV;conþ 1

hB;pre

(4)

where Rf is the fouling resistance, which equals 0.00002 (m2 K)/W[16].

Table 2Correlations of surface heat transfer coefficient and resistance coefficient involved inmathematical model.

Correlations References

Condensation surfaceheat transfercoefficientof vapor

hV;con ¼ 4:118�

llDh

�Re0:4eq Pr1=3l [18]

Single-phase surfaceheat transfercoefficientof brine

hB;pre ¼ 0:2121Re0:78Pe13

�mmmwall

�0:14[19]

Evaporationsurface heattransfercoefficient of brine

hr;l ¼ 0:2092�ll;zDh

�Re0:78Pr

13

�mmmwall

�0:14hB,eva ¼ 88Bo0.5hr,l

[20,21]

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e4238

The quality of source vapor at the exit of preheating section inevaporator can be then obtained by,

xVi;eva ¼ 1� QBi;pre

gVi _mVi;0(5)

The variation of brine flow rate along flow direction during filmevaporation satisfies the following equation,

�gBid _mBidz

¼ gVid _mVidz

¼ Ki;evaL�tVi � tBi;s

�(6)

The change of brine concentration due to evaporation alongflow direction leads to the variations of physical properties, espe-cially the so-called boiling point evaluation (BPE) [17]. BPE meansthe variation of the boiling temperature by the dissolved salts inwater at a given pressure, which is improved with salt concentra-tion. As a result, the saturated feed brine turns subcooled whileevaporating, leading to the temperature difference decreases.Hence, both latent heat and sensible heat are consumed at thephase change interface. Considering the influence of BPE duringbrine evaporation, the above mentioned energy equation under thein-situ feed brine concentration becomes,

��gBi þ cp

�TBi;z � TBi;s

��d _mBidz

¼ gVid _mVidz

¼ Ki;evaL�tVi � tBi;z

�(6b)

of which, tBi,z is defined as the local saturated temperature of feedbrine.

The overall heat transfer coefficient in evaporation section, Ki,eva,is,

Ki;eva ¼ 1dplpþ Rf þ 1

hV;conþ 1

hB;eva

(7)

The evaporating area of each channel, Ai,eva, can be obtainedunder the condition that all the source vapor condenses,

gVixVi;eva _mVi;0 ¼ZAi;eva

0

Ki;eva�tVi � tBi;s

�dABi;eva (8)

from which, the overall evaporating area in each effect can be ob-tained as niAi,eva.

The heating vapor contains the vapor extraction from turbineand the vapor ejector from the last effect for the first effect, or isconsisted of vapor produced by brine of previous effect and flashvapor from both brine and freshwater for the other effects. Theflash rate from each source can be obtained homogeneously by thefollowing energy balance,

x ¼ h01 � h02h002 � h02

(9)

where the subscripts 1 and 2 stands for vapor thermodynamic statebefore and after flashing, respectively. The flash vapor from brineand freshwatermerges into the heating vapor of the next effect, andthe flash vapor from merges into the producing vapor of presenteffect.

Considering the function of thermal vapor compressor (TVC) inthe system, the flowand temperature of the first effect source vaporcan be acquired by the following equations, respectively,

_MV1;0 ¼ D0 þMV4;1 (10)

tV1;0 ¼ m0cp0t0 þmV4;1cpV4;1tV4;1�m0 þmV4;1

�cpV1;0

(11)

inwhich, D0 and t0 stands for extraction flow and temperature fromsteam turbine of power generating unit, respectively.

The gain of ratio (GOR) of seawater desalination systems is givenby,

GOR ¼P4

i¼1 MWi;1

D0(12)

The local GOR of seawater desalination systems with totalsource vapor flow of the first effect, GOR’, is given by,

GOR0 ¼P4

i¼1 MWi;1

MV1;0(13)

Based on the chosen plate structures of the present study thecorrelations of surface heat transfer coefficient Eqs. (4) and (7) arelisted in Table 2. Variations of physical properties of feed brine,which are caused by the salinity varying during brine evaporating,can be found in the literature [21].

In order to solve the mathematical model, the inlet flow andtemperature of feed brine at each effect, as well as the source vaporflow and temperature of the first effect are given in previous.Assuming that total number of channels in every plate evaporator is2ni, respectively, the overall heat transfer area of each effect can beobtained through the mathematical model. Check the result ac-cording to the following condition for each effect,��ð2ni þ 1ÞAe � ni

�Ai;pre þ Ai;eva

��� � ε (14)

The procedure will continue iteratively until the error satisfiesthe pre-set value, ε, under which, the transport characteristics ofseawater desalination can be obtained accordingly. Detailedcalculation processes also can refer to Fig. 3.

3. Results with discussions

Referring to the design parameters of the multi-effect distilla-tion system of the selected thermal power plant in operatingshown in Fig. 2, the inlet flow and temperature of the vaporextraction from steam turbine were given as design conditions ofthe present plate seawater desalination system. The results ofsystem design with plate evaporator as well as its operating char-acteristics were illustrated as follows and discussed compared withthe original MED system.

Fig. 3. The flowchart of calculation process.

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e42 39

Fig. 4 shows the variations of system GOR with entrainmentratio of secondary vapor from the last effect under different flow ofvapor extraction from steam turbine. Entrainment ratiomeans ratioof the flow of extracting vapor from turbine and the flow of lastevaporator vapor ejector. It can be found from the results that thereexist the maximum values of GOR while varying the entrainmentratio for various plate types of evaporator under the same vaporextraction flow of turbine. The increase of vapor ejector in last ef-fect leads to the increase of flow rate of heating vapor in the firsteffect, which can improve the second vapor production and henceGOR. However, with the increase of vapor ejector of last effect theinlet temperature of the heating vapor reduces, which reduces thetemperature difference in evaporator. The combined influencesresult in the optimal amount of vapor ejector corresponding to themaximum GOR, as shown in Fig. 4.

The plate type selected for evaporator also can affect the GOR ofthe system. Under the same vapor extraction from steam turbineand vapor ejector of the last effect, it can be found that GOR in-creases with the decrease of the heat transfer area of single chevron

plate. The heating transfer area of single plate is reduced, whichmeans that the brine side of the evaporator process in every effectis shortened, leading to the increase of channel number in eacheffect and the reduction of feed brine flow and falling film thicknessin each channel. At the same time the shorter flow channel at brineside restrains the influence of BPE and thus reduces the degree ofsubcooling. All the influences can benefit the feed brine evapora-tion and secondary vapor production.

The comparison of heat transfer area needed in each effect withplate evaporators and original tube and shell (tubular) evaporatorare shown in Fig. 5. It can be shown that using the chevronplate heatexchanger reduces the needed heat transfer area ofMED evaporatorsignificantly due to the structure of the plate arrangement, whichmakes the structure more compact and reduces the costs.

High heat transfer coefficient achieved by the plate heatexchanger is one of the main reasons reducing the total areaneeded by evaporator. Taking the first effect as an example, thevariations of overall heat transfer coefficient using different platetypes along the flow length are shown in Fig. 6.

Fig. 4. Changes of GOR with the changes of entrainment ratio in final effect.

0 2 41000

1500

2000

2500

3000

3500

4000

Hea

t tra

nsfe

r are

am

2

The number of effect

BRS1.2BRS1.8BRS2.2Tubular heat exchangers

Fig. 5. The overall heat transfer area for different types of plate.

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e4240

Theoverall heat transfer coefficient of thefirst effect evaporator inMED desalination system in the present selected thermal powerplant, which is 2506 W/(m2 K), is given as a contrast. The flows ofbrine andvaporbetween theplates canbedisturbedobviously by thechevrons. It is believed that the fluid can achieve a turbulent statewhen the Reynolds number reaches 100, which can enhance heat

transfer significantly. Besides, there aremore than three rows of tubebundles generally in traditional tubular heat exchanger. The first rowof tubes contacts thevaporfirst, and then theother rows successively.Uneven distributions of heat load and vapor flowmay exist betweenthe bundles, which may lead to drift and backflow. In the presentstudy, the design of one-through flow channel of between plates isemployed, which can avoid the uneven flows effectively.

The local heat transfer coefficients along flow direction in firsteffect are also illustrated in Fig. 6 for different plate types. Asanalyzed in previous, two aspects can lead to the reduction of heattransfer coefficient, one is the increase of area of single chevronplate, and the other is boiling point elevation due to the concen-tration increase of feed brine along flow direction.

In the present design, higher GOR in all three plate structures canbe obtained than the original MED systemwith tube and shell heatexchangerbyoptimizationof the entrainment ratio. The comparisonof GOR under different evaporator structure is listed in Table 3.

The operating characteristics of the present plate MED systemare also discussed under off-design conditions, of which the in-fluences of variations of inlet temperature and flow of feed brine, aswell as variation of source vapor extraction from steam turbine areconsidered.

The inlet temperatureof feedbrine intoeacheffect is set tobe25 �Cin the present design. Because of season or day and night alternation,inlet temperature discrepancy may usually occur and hence impactthe water producing capacity of seawater desalination system.

Fig. 7 shows the variations of system GOR with inlet tempera-ture of feed brine for different types of plate evaporator underdesigned flow of extracting vapor from steam turbine. If the inlettemperature of feed brine is lower than the nominal design con-ditions, more source vapor will be consumed to heat the subcooledfeed brine in order to make the temperature of brine reaching thepredetermined saturation temperature. And hence, the source heatused to evaporate brine is reduced and less secondary vapor can begenerated for the next effects correspondingly. As shown in Fig. 7, itcan be found that system GOR decreases with decreasing the inletfeed brine temperature for all types of plate evaporators.

There are two aspects of influence that the flow of feed brine hason the production of freshwater. The higher flow of brine leads tohigher proportion of the heating vapor for feed brine preheating,leading to the reduction of source vapor for brine evaporation. Onthe other hand, the increase of the flow rate of feed brine leads tothe increase of the flow in each channel of the chevron plateexchanger, and the Reynolds number increases, which cause theincrease of surface heat transfer coefficient. Therefore, more heatcan be transferred to generate vapor by brine, which can improvesystem GOR. It is logical to deduce that there exist the maximum

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.44000

4200

4400

4600

4800

5000

3.5

4.0

4.5

5.0

5.5

6.0K

1,ev

a, W

/(m2

k)

Flow length, m

concentrarionK1,eva

Concentration

%

(a) BRS1.2.

0.0 0.4 0.8 1.2 1.6 2.04000

4200

4400

4600

3.5

4.0

4.5

5.0

5.5

K1,

eva,

W/(m

2k)

Flow length, m

concentrarionK1,eva

Concentration(%

)

(b) BRS1.8.

0.0 0.4 0.8 1.2 1.6 2.0 2.44000

4100

4200

4300

4400

4500

4600

4700

3.5

4.0

4.5

5.0

K1,

eva,W

/(m2

k)

Flow length, m

concentrarionK1,eva

Concentration(%

)

(c) BRS2.2.

Fig. 6. Variations of overall heat transfer coefficient along flow length in the first effect.

4 6 8 10 12 14 16 18 20 22 24 265.5

6.0

6.5

7.0

7.5

8.0

8.5

9.0

GO

R

The temperature of raw brine,

BRS1.2 BRS1.8 BRS2.2

Fig. 7. Variation of system GOR with inlet temperature of feed brine.

9.0

9.5

10.0

OR

BRS1.2

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e42 41

GOR for the seawater desalination system while varying the feedbrine flow under such contrary influences. Taking the feed brinewith inlet temperature of 25 �C as example, Fig. 8 shows the in-fluence of feed brine flow on system GOR when the amount ofvapor extraction from turbine is 50 t/h. Similar variations may alsobe found for different vapor extractions from turbine as well as

Table 3Comparison of GOR in different structure.

The structure of evaporator GOR GOR’

BRS1.2 8.84 3.82BRS1.8 8.511 3.698BRS2.2 8.38 3.63Tube and shell heat exchanger 8.33 3.61

plate evaporator configurations. Flow of feed brine should beadjusted accordingly under various operating conditions to reachthe maximum system GOR.

The influence of source vapor extracting from turbine on systemGOR is illustrated in Fig. 9. Assigning the inlet temperature of feedbrine is 25 �C and its flow rate is 350 t/h, it can be found that thesystem GOR increases nearly in proportion with the flow ofextracting vapor from turbine, indicating that the increase of hightemperature source vapor can improve the freshwater producingperformance of the system significantly. However, the increase ofGOR becomes slow down with the continuous increment of sourcevapor, implying that the capacity of MED system performanceimprovement by increasing source vapor is limited.

4. Conclusions

The physico-mathematical model is developed to describe thetransport process in multi-effect distillation seawater desalinationsystem with plate heat exchanger. The system design and off-design operating performance are analyzed, as well as comparedwith the original MED seawater desalination systemwith tube andshell (tubular) evaporator in operation. The following conclusionscan be obtained.

(1) The overall heat transfer area under the same flow of feedbrine and extracting source vapor from turbine can besignificantly reduced with plate evaporator compared withthat of tubular evaporator because the compact configurationand high surface heat transfer coefficient of plate exchanger.

250 300 350 400 4508.0

8.5

G

MB1,0, t/h

Fig. 8. Variation of system GOR with feed brine flow rate.

22 24 26 28 30 32 34 36

7.8

8.0

8.2

8.4

8.6

8.8

9.0

GO

R

D0, t/h

BRS1.2

Fig. 9. Variation of system GOR with flow rate of extracting vapor from turbine.

Z. Ge et al. / Applied Thermal Engineering 67 (2014) 35e4242

(2) The system GOR can achieve the maximum value under thesame flow of feed brine and extracting source vapor fromturbine by varying the entrainment ratio of the last effect.

(3) Different types of plate exchanger are compared for plateevaporator design. The performance of the system can beimproved by reducing the length of single plate, for morechannels with shorter plate can enhance heat transfer inevaporator and restrain the BPE impact of evaporating brine.

(4) Inlet temperature and flow of feed brine, as well as flow ofextracting source vapor from turbine can all influence theoff-design performance of MED system, of which, feed brineflow has contrast influences on system performance, leadingto the optimal GOR under certain operating conditions, andthe increase of GOR with source vapor increment is limited.

Acknowledgements

The financial support for this research project from the NationalNatural Science Foundation of China, China (Grant No. U1361108) isgratefully acknowledged.

References

[1] F. Mahbub, M.N.A. Hawlader, A.S. Mujumdar, Combined water and powerplant (CWPP) e a novel desalination technology, Desalin. Water Treat. 5(2009) 172e177.

[2] H. El-Dessouky, I. Alatiqi, H. Ettouney, Process synthesis: the multi-stage flashdesalination system, Desalination 115 (2) (1998) 155e179.

[3] A.M.S. El Din, B. Makkawi, Operation processes affecting corrosion in MSFdistillers, Desalination 115 (1) (1998) 33e37.

[4] G.P. Maheshwari, M. Alramadhan, M. Alabdulhadi, Energy requirement ofwater production in dual-purpose plants, Desalination 101 (2) (1995) 133e140.

[5] G. Fosselard, K. Wangnick, Comprehensive study on capital and operationalexpenditures for different types of seawater desalting plants (RO, MVC, ME,ME-TVC, MSF) rated between 200 m3/d and 3000 m3/d, Desalination 76 (1e3)(1989) 215e240.

[6] A. Almulla, A. Hamad, M. Gadalla, Integrating hybrid systems with existingthermal desalination plants, Desalination 174 (2005) 171e192.

[7] N.M. Wade, Distillation plant development and cost update, Desalination 136(2001) 3e12.

[8] M. Bertuzzi, M. Polla, P. Tiraboschi, Development in vertical tube evaporation,Desalination 51 (1985) 135e143.

[9] K.R. Chun, R.A. Seban, Heat transfer to evaporating liquid film, ASME J. HeatTransfer 93 (1971) 391e396.

[10] H.H. Wang, X.H. Ma, Z. Zhong, Research progress of plate evaporator, Chem.Ind. Eng. Prog. S1 (2009) 343e346.

[11] E. Mezaache, M. Daguenet, Effects of inlet conditions on firm evaporationalong an inclined plate, Sol. Energy 78 (2005) 535e542.

[12] C. Legorreta, S. Hinge, J. Tonner, A. Lovato, Plates the next breakthrough indesalination, Desalination 122 (1999) 235e246.

[13] F. Karl, V. Renaudin, D. Alonso, J. Gornut, NewMED plate desalination process:thermal performances, Desalination 166 (2004) 53e62.

[14] C. Legorreta, Heat transfer plates improve efficiency, Water Environ. Int. 8(1999) 16e17.

[15] O. Al-Hawaj, A study and comparison of plate and tubular evaporators,Desalination 125 (1999) 233e242.

[16] M.C. Chyu, A.E. Bergles, An analytical and experimental study of falling-filmevaporation on a horizontal tube, Heat Transfer 1 (1987) 109e183.

[17] S. Wu, X. Du, L. Cheng, Thermal hydraulic performance of the brine desali-nation system coupled with nuclear heating reactor, Desalination 140 (2001)297e307.

[18] Y.Y. Yan, H.C. Lio, T.F. Lin, Condensation heat transfer and pressure drop ofrefrigerant R-134a in a plate heat exchanger, Int. J. Heat Mass Transfer 42(1999) 993e1006.

[19] Y.Y. Yan, T.F. Lin, Evaporation heat transfer and pressure drop of refrigerant R-134a in a plate heat exchanger, ASME J. Heat Transfer 121 (1999) 118e127.

[20] Y.Y. Hsieh, T.F. Lin, Saturated flow boiling heat transfer and pressure drop ofrefrigerant R-410A in a vertical plate heat exchanger, Int. J. Heat Mass Transfer45 (2002) 1033e1044.

[21] H.T. El-Dessouky, H.M. Ettouney, F. Mandani, Performance of parallel feedmultiple effect evaporation system for brine desalination, Appl. Therm. Eng.20 (2000) 1679e1760.