The economic feasibility of small solar MED seawater desalination plants for remote arid areas

ELSEVIER Desalination 134 (2001) 173-186

DESALINATION

www.elsevier.com/locate/desal

The economic feasibility of small solar MED seawater desalination plants for remote arid areas

Ali M. E1-Nashar

Research Center, Abu Dhabi Water and Electricity Authority, P.O. Box 41375, Abu Dhabi, UAE Tel. +971 (2) 5081-510; Fax + 971 (2) 4434-906; e-mail: elnashar(~emirates.net.ae

Received 26 September 2000; accepted 10 October 2000

Abstract

The objective of this paper is to compare the economics of using solar energy to operate small multiple effect seawater distillation systems in remote areas with the conventional method of using fossil fuels. The particular multiple effect system used is an advanced horizontal tube, falling film system called "multiple effect slack", MES, in which the pumping energy.requirement is relatively low compared with the horizontal in-line system. Three system configurations were investigated: (1) conventional system using a steam generator to provide steam for the MES evaporator and a diesel generator to provide pumping power, (2) solar-assisted system which uses solar thermal collectors to provide hot water (instead of steam) for the evaporator and a diesel generator for pumping power, and (3) solar stand-alone system which uses solar thermal collectors for the evaporator heat requirement and a solar PV array to provide electrical energy for pumping. At the present time, solar energy cannot compete favorably with fossil energy particularly under the present international market prices of crude oil. However, in many remote sunny areas of the world where the real cost of fossil energy can be very high, the use of solar energy can be an attractive alternative. Two important cost parameters affect the relative economics of solar energy vis-a-vis conventional (fossil) energy: the collector cost in $ per square meter and the cost of diesel oil in $ per Giga Joule. Solar energy becomes more competitive as the local cost of procuring conventional fuel increases and as the collector cost decreases. The water cost from a solar thermal-diesel-MES system (Configuration #2) can be seen to approach the water cost from a steam generator-diesel-MES system (Configuration #1) when the collector cost drops to 200 $/m 2 and diesel oil cost at the remote site reaches 50 $/GJ. Using a 100% solar system (Configuration #3) with solar thermal and solar PV collectors being utilized, the economics was seen to improve in favor of the solar system. Even when diesel fuel can be procured at 10 $/GJ at the remote site, the cost of water from the solar system can be seen to approach that from a conventional plant when thermal collectors costing 200 $/m 2 are used. The cost of water from the solar system was shown to be always less than that fi'om a conventional system which uses diesel oil procured at the high price of 50 $/GJ but always higher than water produced from a conventional system using diesel oil at the low price of 10 $/GJ.

Keywords: Solar desalination; Seawater distillation; Multiple effect stack; Economics of solar desalination; Conventional vs. solar desalination; Desalination in remote areas

Presented at the International Conference on Seawater Desalination Technologies on the Threshold of the New Millennium, Kuwait, 4-7 November 2000.

0011-9164/01/$-- See front matter © 2001 Elsevier Science B.V. All fights reserved PII: SO011-9164(01)00124-2

174 A.M. El-Nashar / Desalination 134 (2001) 173-186

1. Introduction Many remote areas of the world such as

coastal desert areas in the Middle East or some Mediterranean and Caribbean islands are suffering from acute shortage of drinking water. Drinking water for these locations can be hauled in by tankers or barges, or produced by small conventional desalination units using the available saline water. The transportation of water by tankers or barges involves a lot of expenses and is fraught with logistical problems which can make fresh water not only very expensive when available but its supply is susceptible to being frequently interrupted. The use of small conventional desalination units that utilize fossil fuel, such as diesel oil, as the energy supply can suffer from the same procurement problems that are encountered with transporting fresh water, namely transportation expenses and supply reliability.

Some of the remote areas are blessed with abundant solar radiation which can be used as an energy source for small desalination units to provide a reliable drinking water supply for the inhabitants of the remote areas. Recently, considerable attention has been given to the use of solar energy as an energy source for desalination because of the high cost of fossil fuel in remote areas, difficulties in obtaining it, interest in reducing air pollution and the lack of an electrical power source in remote areas.

Desalination of seawater and brackish water is one of the ways for meeting future fresh water demand. Unfortunately, conventional desalination technology is fairly well established and some of the processes may be considered quite mature although there is still considerable scope for improvement and innovation. Conventional desalination processes are energy intensive and one of the major cost item in operating expenses of any conventional desalination plant is the energy cost. Thus, one of the major concerns about using desalination as a means of supplying fresh water to remote communities is the cost of energy.

Apart from energy cost implications, there are environmental concerns with regard to the effects of using conventional energy sources. In recent years, it has become clear that environmental pollution caused by the release of green house gases resulting from burning fossil fuels is responsible for ozone depletion and atmospheric warming. The need to control atmospheric emissions of greenhouse and other gases and substances will increasingly need to be based on growing reliance on renewable sources of energy.

In this paper, the economics of a solar-assisted and solar stand-alone multiple-effect stack (MES) distillation systems for seawater desalination are compared with that of a conventional system consisting of an MES unit operated in association with a steam generator and a diesel generator. The comparison is based on the economic and meteorological environments prevailing in Abu Dhabi, UAE.

2. System configurations

Three system configurations are considered for the current economic study: • A conventional distillation system consisting

of an MES evaporator supplied by steam from a low-pressure steam generator with the pumping power supplied by a diesel generator as shown schematically in Fig. l a (Configu- ration 1).

• A solar-assisted system consisting of an MES evaporator supplied by thermal energy in the form of hot water from either high-efficiency fiat plate collectors or evacuated tube collectors with pumping power supplied by a diesel generator as shown in Fig. 1 b (Configuration 2 ).

• A solar stand-alone system consisting of an MES evaporator supplied by thermal energy from fiat plate or evacuated tube collectors, as in Configuration2, with pumping power supplied by a solar PV system as shown in Fig. 1 c (Configuration 3).

A.M. El-Nashar / Desalination 134 (2001) 173-186 175

(a) Configuration 1

crude oil I .~ Boiler

diesel oil __J Diesel r I Generator

(b) Configuration 2

steam

condensate

electricity

distillate

MES Evaporator

Solar Thermal Collectors

diesel oil~Diese ! ']Generator

seawater

~ brine blowdown

hot water I °twater .ea, h

Accamo~ l ~ warm water lator ~ . ~ ' ~

. . . . . . . . . . . . . . . . . . 7g;21 electricity |

MEB Evaporator

seawater

(e) Configuration 3 distillate brine blowdown

Solar Thermal Collectors

Solar PV Collectors

Fig. 1. System configurations.

hot water~ ~ hit water LTrm Heat warm water A~:rUmU

wat~ MES

Evaporator

seawater

. . . . - ' ; 3 - - / I I °'~r~°~ electrici nanenes Inverter distillate brine biowdown


The system labeled as "Configuration 1" needs two types of fossil fuel, namely crude oil for the steam generator and diesel oil for the diesel generator. The thermal energy required by this system is supplied by the steam generator. Several pumps are required to force the different streams to flow through the system and to inject different chemicals into these streams to ensure safe and reliable operation. The major pumps required by the evaporator are the seawater intake pump, feed water pump, distillate (product water) pump, brine blowdown pump, drain pump, antiscalant dosing pump (for feedwater stream), NaCIO dosing pumps (for seawater intake and distillate streams), NaHCO3 dosing pump (for distillate stream) and CaCI2 dosing pump (for distillate stream). Steam jet ejectors that are supplied by steam from the steam generator are used to create and maintain the vacuum inside the evaporator.

The system labeled as "Configuration 2" needs only one type of fossil fuel, namely, diesel oil for the diesel generator. The thermal energy required by the MES evaporator can be supplied in the form of hot water generated by either flat plate or evacuated tube collectors. Double glazed, well designed flat plate collectors can produce hot water at about 80°C with reasonable efficiency while evacuated tube collectors can easily produce water at more than 95°C with good efficiency albeit higher collector cost. The pumping power for this system is more than that for the conventional system because at least three additional pumps need be incorporated: (1) mechanical vacuum pump, (2) heat collecting pump and (3) heating water pump. The mechanical vacuum pump replaces the steam ejector in the conventional system since no steam is available in the current system. The heat collecting pump is used to circulate the collector fluid (water) through the collector field to enhance solar heat collection. The heating water pump is used to draw hot water from the heat accumulator and supply it to the first effect of the MES

evaporator to initiate seawater boiling in that effect. The capacity of the diesel generator for this system is therefore larger than that of the conventional system.

The system labeled "Configuration 3" utilizes solar energy to satisfy both thermal and electrical energy demands of the evaporator. The electrical power requirements of this system are essentially the same as that of "Configuration 2". However, the power requirement is supplied by an array of solar PV cells instead of a diesel generator. An inverter is used to change the DC output into AC electricity required by the pump motors.

3. Design and sizing considerations

3.1. Thermal energy requirement

The thermal energy requirement of the MES evaporator, Qe~, depends essentially on the evaporator's rated capacity as well as its performance ratio. The performance ratio for the MES evaporator depends essentially on the number of effects and vary according to the following relation [1,2]:

PR = -0.809 + 0.932N - 0.0091N (1)

Qev can be obtained from Eq. (2)

L Q,~ = M, - - (2)

PR

where Md is the rated (design) capacity and L is an average latent heat of vaporization.

3.2. Pumping requirements

The electrical power requirements of the MES evaporator depends on the means of achieving the vacuum inside the evaporator. If a vacuum pump is used, then its power consumption has to be added to the power consumption of the other pumps to obtain the total pumping power


requirement. If vacuum is created with a steam jet ejector, then the total pumping power will only be those needed by the pumps. The power requirements of the vacuum pump and other evaporator pumps [1] are obtained from the following relation which are based on manufacturer's data:

radiation intercepted by the absorber plates of the collector. This efficiency, referred to as rk, is essentially dependent on a parameter x defined a s

0.5 (v, + To x - ( 6 )

1,

Pv~ = 1.42 + 0.0395M~ °3 (3)

Per = 4.36 + 0 . 0 9 6 5 M d - 2.2 (10-5)M,~ (4)

where the power is in kW and Md is in m3/d. Per represents the summation of the evaporator pumping power requirements without the vacuum pump.

In addition to the pumping power requirements given above we have to add the power required to circulate the water through the solar collectors. Based on data obtained from Abu Dhabi solar plant, the heat collecting flow rate per m s of collector absorber area is about 45 kg/h m s which is assumed to be maintained for the current study. Based on a pressure drop per collector panel of about 1 mH~O, and assuming a pump efficiency of 70%, the rated power consumption of the heat collecting pump (in kW) can be obtained from the equation

Pc = 2.691.(10-2 )(0.045A~ ) 1'3 (5 )

where Ac is the collector area in m 2.

3. 3. Collector performance

Evacuated tube collectors [1] are assumed to be the collector of choice when collector outlet water temperatures in excess of 80°C are considered. For lower water temperatures, flat plate collectors can be used. The performance of a collector is determined by its efficiency defined as the ratio of the amount of heat collected by the collector fluid (water) to the amount of solar

where T~ is the collector inlet water temperature, T2 is the outlet water temperature and I is the solar radiation on the absorber area. For evacuated tube collectors of the type used at the solar plant (manufactured by Sanyo), the collector efficiency can be expressed by the following formula

r L = 0.84 - 2.6x - 1.92x 2 (7)

where x is in units of °Chm2kca1-1. For flat plate collectors, the collector effici-

ency is assumed to be

r/¢ = 0.76 - 4.36x (8)

3. 4. Sizing thermal collector area

The optimum thermal collector area required for a particular evaporator rated capacity was defined as the collector area corresponding to the minimum cost of water produced. This optimum area is specific to both weather and economic conditions at the locality where the plant is to be constructed. By using the available weather and economic data for Abu Dhabi city, as an example, it was possible to simulate the performance of a solar desalination plant using a computer program and obtain the optimum area for different plant capacities for this location. For example, for an evaporator capacity of 120 m3/d (identical to the capacity of Abu Dhabi solar plant), the relative water cost for different collector areas is as shown in Fig. 2. The minimum water cost can be seen to correspond to a collector area of about


2.5

o 2 u

~ 1 . 5

o > 1

0.5

e N/Tb=30180 4 - - N/Tb:24/71

~ - - dL - - ,Nrrb:18/63

500 1000 1500 2000 2500 Collector area, nl 2

Fig. 2. Cost of water produced by a solar MES plant having a rated capacity of 130 mVd for different collector a r e , a s .

1,500 m s for an evaporator having N = 30 effects and designed for a maximum brine temperature Tb = 80°C.

3.5. Sizing the PV system

Sizing the elements of the PV system [3] means determining the PV collector area and the amount of energy storage capacity that are needed so that the system can provide the amount of electrical power that is required by the evaporator and the solar thermal collector system on a 24-hour per day basis.

Methods of sizing the stand-alone PV system usually treat the PV array and the battery storage independently. The array size [3,4] is determined by requiring that the average daily array output to match the average daily load for the worst case month of the year. A safety factor is introduced to account for degradation in army output due to dirt accumulation, line losses and general uncertain- ties about "average" outputs and inputs.

The amount of storage capacity is usually fixed by estimating the number of "no-sun days", the number of consecutive days during which overcast conditions prevail, liable to be expe- rienced at the site. If a high degree of system reliability is demanded, the usable daily battery capacity, measured in Ah is fixed by the maxi-

mum daily load, again measured in Ah. The number of no-sun days times the maximum daily load gives the system's required total usable battery capacity.

3.5.1. Sizing the PV array

A PV module, and hence a PV array, is rated according to its maximum power. The maximum power is the power the module will produce at certain reference conditions of module temperature and solar radiation intensity. For example, a 30W module has a maximum power output of 30W at a standard cell temperature of 25°C and solar radiation intensity of 1.0 kW/mL As a rule of thumb, the maximum power decreases by 5% for every 10°C temperature rise [4]. The net power delivered to the load will be less than the power produced due to losses in the inverter, battery and lines. Assuming a 15% loss in power due to temperature rise and 10% loss due to the other factors, we can estimate the amount of energy produced in a typical day (averaging 7.4 full sunshine hours per day) by one module as 7.4 x 0.75 x 30 = 166.5 Wh.

3.5.2. Sizing battery storage capacity

The first step in sizing of the battery required is to calculate the daily load, L, in terms of Ah and the system voltage, V. The load will determine the battery capacity while the system voltage will determine the number of cells to be connected in series. The total usable battery capacity (TUC), can be obtained by multiplying the load, L, by the number of "no-sun" days, N which depends on both weather conditions at the plant site and the required system reliability:

TUC=(L~(N) (9)

Once the system TUC has been determined, the system's required rated battery capacity, C, can be calculated. The calculation procedure

A.M. EI-Nashar / Desalination 134 (2001) 173-186 179

differes according to the battery manufacturer and the following parameters are taken into consideration in this procedure: • Daily depth-of-discharge (DOD) • Maximum allowable depth-of-discharge

(MDOD) • Rates of charge and discharge • Operating temperature • Desired capacity remaining in the battery at

end of its useful design life

Details of the calculation procedure are available in [3].

4. Economic ground rules and water cost optimization

4.1. Capital equipment cost

4.1.1. Capital cost of MES evaporator

The cost of MES evaporator is based on bud- get offers from different manufacturers and on cost information available in the open literature [1,5]. The capital cost was found to depend on the design capacity, number of effects and maximum brine temperature according to the correlation:

t 2,500) t. 8 ) t,r )

where Cev - evaporator capital cost, $ (1999 prices); M a - rated (design) capacity, mVd; N - number of effects; Tb - design maximum brine temperature, °C.

4.1.2. Capital cost o f solar thermal collectors

Solar collectors used for this application should be capable of producing hot water at a temperature ranging between 70-90°C. Evacuated tube collectors and high-efficiency fiat plate collectors can be used to produce hot water at a temperature in excess of 80°C. The specific cost

of the solar collectors are assumed to vary between 2005/m 2 (flat plate collectors) to 500 $/m 2 (evacuated tube collectors). The cost is assumed to include both the solar collector proper as well as the support structure, piping, valves, etc.

4.1.3. Capital cost of P V electricity generating system

The main components of the photovoltaic (PV) system are a PV array, a bank of batteries and an inverter. The cost of the PV array- including support structure- is assumed to be in the range $6.0-10.0 per peak watt which is based on recent quotations for commercial-grade polycrystalline silicon cell material. Regular 12 V lead-acid batteries with a capacity of 96 Ah are considered for this application with an estimated cost of $55.0 each.

4.1.4. Capital cost of heat accumulator

The heat accumulator is assumed to be a thermally-stratified vertical cylindrical tank made of mild steel with a thick layer of fiber glass insulation to reduce heat loss. The tank is designed to operate at atmospheric pressure and is provided with a pressure relief valve as a safety measure. Hot water from the collector field is supplied to the tank at the top via a special water distribution grid which ensures that hot water diffuses slowly through the surrounding water without causing too much turbulence in order to enhance thermal stratification throughout the tank.

The specific capital cost of the heat accumulator as obtained from manufacturer's data is correlated by the following relation

z x - 0 . 4 6 / I A " ~

j where c,t - specific cost of storage capacity, $/m 3 (1999 price); M~t = storage capacity, m 3.

180 A.M. EI-Nashar / Desalination 134 (2001) 173-186

4.1.5. Capital cost of steam generator

Low pressure and low capacity steam gene- rators are required to supply the MES evaporator with the low-temperature thermal energy neces- sary to drive the unit. The capacity of the steam generator depends on the capacity of the MES unit as well as its performance ratio. For a unit having a PR = 13 and producing 200m3/d at design conditions, requires only 0.6 ton/h of low- pressure steam.

A fire tube packaged boiler producing steam at 10 bar and having an efficiency (LHV) of 86% is considered appropriate. The capital cost of such a boiler, Cb ($) is obtained from [6] and adjusted to the 1999 cost level using the Marshall and Swift Equipment Cost Index. The resulting correlation is shown below:

Csg=10,824(S) °87 +6,973 0 . 1 5 < S < 1 5 (12)

• Fuel cost, • Cost of chemicals, • Cost of spare parts, • Salaries for O&M personnel.

4.2.1. Fuel cost

For the stand-alone solar desalination plant option (Configuration3), no fossil fuels are needed. For Configuration2 (solar thermal collectors and diesel generator) only diesel oil is required whereas for Configuration 1 (steam generator and diesel generator) crude oil is required for the steam generator and diesel oil is required for the diesel generator. As mentioned before, the cost of diesel oil is assumed to be 5 times as much as that of crude oil. In remote locations, the cost of diesel oil is assumed to vary between 10-20 $/GJ.

where Csg- cost of boiler, $; S - steam generating capacity, ton/h.

4.1.6. Capital cost of diesel generator

The electrical demand of the desalination plant can be supplied by a diesel generator whose capacity will obviously depend on the capacity of the plant itself. The capital cost of the diesel generator is obtained from the following relation which is based on recent commercial bids:

c,,, = 50,800( 0) °549' (13)

where Cag is the cost in $ and P is the rated capacity in kW.

4.2.2. Cost of chemicals

The chemicals used, their dosing rate and their unit cost are shown in Table 1. The unit costs are those prevailing in Abu Dhabi, UAE during 1999.

Table 1 Cost of chemicals used in MES desalination plants

Name of chemical Dose, Cost, Flow stream ppm (S/ton

NaCIO 14 284 seawater Belgard EV 10 1486 feedwater antiscalant NaI--ICO3 23 270 distillate NaCIO 3 284 distillate CaCI2 18 210 distillate Anti-corrosion 5000 2430 chemical (e.g. Nalcool)

4. 2. Operating costs

The main components of the operating costs include:

4.2. 3. Cost of spare parts

Annual cost of spare parts is assumed to be 2% of the capital cost for the evaporator, steam


generator and diesel engine. The cost of spare parts for the thermal and PV collectors is assumed to be negligible compared to the spare parts cost of other equipment. This has been validated by 15 years of operational experience of the Abu Dhabi solar plant.

4.2. 4. Salaries for O & M personnel

For plants of Configuration 1 and 2 where a diesel engine and/or steam generator are in operation, the operating staff are: 1 plant engineer, 2 technicians and 1 skilled laborer. For plants of Configuration 3 which are solar stand-alone systems, only 1 plant engineer and 1 skilled laborer are assumed to be needed for plant operation. The annual salary for each operating staff is assumed as shown in Table 2.

The major economic parameters used in the current study are summarized in Table 3.

Table 2 Annual salaries for plant operation and maintenance personnel

O&M personnel Annual salary, $

Plant engineer 44,600 Technician 33,150 Skilled laborer 9,780

4. 3. Optimum cost o f water for each plant con- f~urat ion

The optimum (minimum) water cost for each plant configuration was estimated. The water cost in $/m 3 of fresh water depends on the plant capital cost, O&M costs (which includes fuel costs for plants that use steam generator and/or diesel generator) and the annual amount of fresh water produced. The capital cost depends on plant configuration and design parameters such as evaporator capacity, design maximum brine temperature, number of effects, solar collector area, type of solar collector (whether flat plate or evacuated tube type), and capacity of heat accumulator. The operating cost depends also on plant configuration, plant capacity, number of effects, collector area, heat accumulator capacity and cost of fuel and chemicals used.

The annual amount of fresh water produced by a solar desalination system (Configuration 2 and 3) depends on the evaporator design capacity, design maximum brine temperature, number of effects, solar collector area, and heat accumulator capacity. This amount is obtained from the output of program "SOLDES"[1,7]. The output of this program is correlated with the primary design variables using the least square technique. Thus for a plant having a design capacity of 500 mVd, for example, the correlation connecting the annual fresh water production , M(m3), to the collector area, Ac, the number of effects, N, and the design maximum brine temperature, Tb is given below:

Table 3 Major economic parameters used in the study

Cost of collectors, $/m 2 200-400 Cost of PV system, $/Wp 6-10 Cost of crude oil, $/GJ 2-4 Cost of diesel oil, $/GJ 10-20 Plant life time, y 20 Interest rate, % 8t Fuel cost escalation rate, %/y 3

M = ( - 57,3 04.9 + 49.5475A c -0.00235542)+

(- 190~122.4 + 16,142.3N- 331.81N2)+ (14)

(99,400-135(~r b -Tb 2)

where Ac - solar collector area, m 2 ranging between 2,500o12,500 m2; N - number of effects of evaporator, ranging between 10-30; T b - design maximum brine temperature, °C, ranging between 60-80°C.


Thus for Configuration 2, for example, the optimum cost of water, c f , was estimated by minimizing the following objective function

)l+ (15) Cw- M . f ~

where CRF is the capital recovery factor.

* Min(c ) c w = w (16)

Subject to the following constraints:

2,500 --- Ac (m s) = 12,500 10 = N = 30 60 = Tb (°C) = 80

The optimum water cost of a plant of a given capacity is the minimum water cost which can be achieved by the plant when its design parameters are optimized. The optimization process was carried out using the mathematical software Eureka version 1.

5 . R e s u l t s

The conventional system is that which uses a steam generator to produce steam for the MES evaporator and uses a diesel generator to produce power to run the pumps. This is Configuration 1 (see Fig. 1). The optimum water cost for this system is shown in Fig. 3. In this system the fuel used by the steam generator is assumed to be crude oil with a cost assumed to be equal to one- fifth of that of diesel oil. As can be expected, the optimum water cost for this system can be seen to increase steadily as the fuel cost increases.

The optimum number of effects and the corresponding performance ratio for two different fuel costs for the conventional system (Configu- ration 1) are shown in Figs. 4 and 5. It can be seen that the optimum number of effects and the corresponding performance ratio increase as the

"E

| "6

o

100 200 400 $00 800 t000

Plant capacity, m 31d

Fig. 3. Optimum water cost from conventional MES evaporator using a steam generator and diesel generator (Configuration 1).

26

15

10 E = 5 Z

0

~ , d i e s e l fuel cost = 10 $/GJ . . . . . . diesel fuel cost = 20 $/GJ

0 200 400 600 800 1000

Plant capacity, m31d

Fig. 4. Effect of diesel fuel cost on the optimum number of effects of a conventional MED plant.

fuel cost increases. With higher performance ratio, the fuel consumption decreases thus reducing the fuel cost per unit distillate produced. The increase in the performance ratio causes also an increase in the capital cost of the MED plant which partly offsets the benefit gained from savings in fuel costs.

A.M. EI-Nashar ~Desalination 134 (2001) 173-186 183

16

14

~ 12

• 10

g. 4 2

0

--dJeselfuelcost = i0 $~3J

. . . . . . d~sel~elcost= 20 Sj(3J

0 200 400 600 800 1000

Plant capacity, m3/d

Fig. 5. Effect of diesel fuel cost on the optimum performance ratio of a conventional MED plant.

The optimum (minimum) cost of water produced by the solar thermal-diesel-MES system (Configuration 2) depends mainly on the collector cost and the cost of diesel oil. The effect of collector cost on the optimum water cost for an evaporator capacity of 500mVd is shown in Fig. 6 for assumed diesel oil costs of $10/GJ and $20/GJ. The cost of water can be seen to increase linearly with the collector cost. At the lowest collector cost of 200 S/m s , the cost of water varies between 5-5.5 $/m 3 and at the high collector cost of 400 S/m s , the water cost varies between 6.2- 6.7 $/m 3.

7 -

5"

| , ,63 ;o= 0 t

0

200

ipmmmB~l DOlla~mO~mml~laav

~ d i e s e l fuel cost = 10 $/GJ

. . . . . . diesel fuel cost = 20 $/GJ

Plant capacity = 500 m=day , u

300 4O0 Collector cost, $1m z

Fig. 6. Effect of collector cost on optimum water cost for two values of diesel fuel cost- solar thermal-diesel-MES system (Configuration 2).

Fig. 7 shows the effect of plant capacity and collector cost on the water cost for the same system (Configuration 2) for a diesel fuel cost of 10 $/GJ. The drop in the cost of water with plant capacity is the direct influence of the economy of scale for the MES evaporator and diesel generator. Plants with a capacity of 100m3/d has a water cost varying between 8.3-9.3 $/m 3 while those with a capacity of 1000m3/d has water costs varying in the range 3.4--4.4 $/m 3.

10 9

~e 85

1 0

collector cost = 200 $/m2 ~..~,.~.. ~ - - - - collector cost = 300 $/m2

. . . . . . collector cost = 400 $/m2

" " L ' . L . . . . .

Diesel fuel cost = 10 $1GJ

200 400 600 800 t000

Plant capacRy, m3/d

Fig. 7. Effect of plant capacity and collector cost on the cost of water for the solar collector-diesel generator-MES plant (Configuration 2).

Fig. 8 shows the effect of the cost of PV array on the water cost for Configuration 3. By com- paring the water cost from this system with that of Configuration 2, it can be seen that the two cost figures are close to each other for small plant capacities. For capacities in the higher end of the range it can be seen that water cost from Configuration 3 is higher than the corresponding one from Configuration 2.

Fig. 9 shows the effect of the collector cost and the PV cost on the water cost for a plant having a capacity of 500mVd using Configuration 3. It can be seen that water cost is sensitive to both collector cost and PV cost. Increasing the PV cost from 6$/Wp to 10$/Wp increases the water cost by about 15% whereas doubling the collector cost from 200 $/m 2 to 400 $/m 2 increases the water cost by about 19%.


10' '~E 9' ~ 8 '

" 7 '

6'

~ 4 '

0 2 ' t '

~ PV cost = 6 $NVp - . " - ~ . . . . . . PV cost = 8 $NVp ~ . , ~ . . . . PV cost = 10 $NVp

200

Collector cost = 300 $/m 2

i

400 600 800 1000 Plant capacity, m3/d

therefore only influenced by the price of fuel. The cost of water from the stand-alone solar system (Configuration 3), on the other hand, is seen to vary between 8.65/m 3 and 9.95/m ~ depending on the price of the PV system cost. The cost of water from the conventional system varies from 7.3 $/m 3 to 8.3 $/m 3 depending on the prevailing cost of fuel; it can be seen to approach that from the solar system at high fuel costs when the price ofPV pannels is set at 6 $/Wp.

Fig. 8. Effect of plant capacity and cost of PV array on the water cost (Configuration 3).

8

7

1

__ 1 ~ ..__ 3 3 .__.. ~ : -.. 1 : __--- ~ l .--. : / .". 3 ~ .". " "."

- - PV cost = 6 $/Wp

. . . . . . PV cost = 8 SNVp

. . . . PV cost = 10 $/Wp

Plant capacity = 500 m3/d

200 250 300 350 400

Collector cost, $1m2

Fig. 9. Effect of collector cost and PV array cost on water cost (Configuration 3).

A comparison between the cost of water from Configuration 2 (solar thermal-diesel-MES system) and the corresponding cost from the conventional system (Configuration 1) is depicted in Fig. 10. It can be seen that the conventional system produces water at a lower price than the solar system but the cost difference gets smaller at lower collector cost and higher fuel cost.

A comparison between water cost from the solar thermal-PV-MES system (Configuration 3) and the conventional system (Configuration 1) is shown in Fig. 11 for a plant having a capacity of 100 mVd. In this figure, the cost of diesel fuel is plotted as abscissa and the water cost as ordinate. Obviously, the water cost from the conventional system is independent of collector cost and is

l o . . . . . . . . _ _ _ _ .

9 i:__--'-----.-. . - . - 7 . . - . . - . - . . . .

6 ' conventional (Confio. 2)

~ s . (Config. 1)

Plant caoscitv = 100 m 3/d 8 3 '

. . . . . . Collector cost = 200 $/m2

2 ' - - - - - - Collector cost = 300 $/m2

1 . . . . Collector cost = 400 $/m2

Conventional 0

10 12 14 16 18 20

Cost of diesel fuel, $/GJ

Fig. 10. Effect of collector cost on the cost of water - comparison between solar thermal-diesel-MES system (Configuration 2) with the conventional boiler-diesel- MES system (Configuration 1).

10

m o

i .

J

. . . . . . . ~ m . . . . . . . . . . . . . . . . . . . . . . .

Solar thermal-PV-MES system, Configuration 3

Collector cost = 300 S/m =

Plant capacity = 100 m~/d

8

7

6

5

4

3

2

1

0

i0

i

12

- - ~ - - PV cost = 6 $/Wp . . . . . . PV cost = 8 $/Wp . . . . PV cost = 10 $/Wp

Convetional ,

14 16 18 20

D ~selfuelcoa~ $~3J

Fig. ! 1. Comparison between the cost of water from solar thermal-PV-MES system (Configuration 3) wilh boiler- dicsel-MES (conventional) system (Configuration 1).


A comparison between the capital cost of each of the three systems is shown in Fig. 12 which clearly indicates the high capital cost of the two solar systems (Configurations 2 and 3) in relation to the conventional system (Configuration 1). This is due to the fact that, under the present economic conditions, the capital cost of the solar collector field is substantially higher than the cost of a steam generator having the same heat capacity, and that the cost of the PV system is much higher than the cost of a diesel generator having the same power output.

14"

~ 1 2 -

~,7t0 -

~0- ~4- ~2 2

0 0

~ . C o n f i g . 1 , , ~, ° "

. . . . . . Config. 2 . P

. . . . Confio. 3 o

• p ,

w

200 400 600 800 1000 Plant capacity, m3/d

Fig. 12. Comparison between the capital cost of the three configurations (Assumptions: collector cost = 300 $/m 2, diesel fuel cost = 10 S/G J, PV cost = 6 S/G J).

The present worth of the life-cycle fuel cost for the three systems is shown in Fig. 13. Obvi- ously, Configuration 3 has a zero fuel consumption since it is a stand-alone solar system thus has a zero fuel cost.

6. Concluding remarks

The present study investigated the economics of small-scale, multiple-effect, stack-type distillation plants for seawater desalination using either conventional energy sources or solar energy sources. The plants are to be considered for the coastal remote areas which are characterized by very high fuel costs and the unavailability of electricity from the national electrical grid. Three

4 . 5 " Config. 1

- - Config. 2

3 .5- ~ , - Config. 3 ~ ,, ,I

_u2.5~ ~ " ' * " ,* ' "

~ 1 t , e ' '

" Jo .5

0 , , - , . . . . . . .

200 400 600 800 1000

Plant capacity, m3/d

Fig. 13. Life-cycle fuel cost for each configuration (Assumptions: diesel fuel cost= 20 $/GJ, fuel cost escalation assumed=3%, interest rate = 8% and life cycle = 20 years).

systems configurations were investigated: (1) conventional system using a steam generator and a diesel generator, (2) solar-assisted system in which thermal energy required by the evaporator is provided by solar thermal collectors and the pumping power provided by a diesel generator, and (3) solar system in which thermal energy is supplied by solar thermal collectors and the pumping power by solar PV collectors.

At the present time, solar energy cannot compete favorably with fossil energy particularly under the present international market prices of crude oil. However, in many remote sunny areas of the world where the real cost of fossil energy can be very high, the use of solar energy can be an attractive alternative.

Two important cost parameters affect the relative economics of solar energy vis-&-vis conventional (fossil) energy: the collector cost in $/m 2 and the cost of diesel oil in $/GJ. Solar energy becomes more competitive as the local cost of procuring conventional fuel increases and as the collector cost decreases. The water cost from a solar thermal-diesel-MES system (Con- figuration 2) can be seen to approach the water cost from a steam generator-diesel-MES system


(Configuration 1) when the collector cost drops to 200 S/m 2 and diesel oil cost at the remote site reaches 50 $/GJ.

Using a 100% solar system (Configuration 3) with solar thermal and solar PV collectors being utilized, the economics was seen to improve in favor of the solar system. Even when diesel fuel can be procured at 10 $/GJ at the remote site, the cost o f water from the solar system can be seen to approach that from a conventional plant when thermal collectors costing 200 S/m s are used. The cost o f water from the solar system was shown to be always less than that from a conventional system which uses diesel oil procured at the high price of 50$/GJ but always higher than water produced from a conventional system using diesel oil at the low price of 10 $/GJ.

7. Symbols

r/c - - Collector efficiency A+ - - Collector area, m 2 C - - Rated battery capacity, Ah Cag - - Capital cost o f diesel generator, $ tag - - Specific cost of diesel generator, $/kW Cev - - Capital cost of MED evaporator, $ Ce+ - - Specific cost of evaporator, $/m3d -1 Csg - - Capital cost o f steam generator, $ c++g - - S p e c i f i c cost of steam generator,

S/ton d -1 c+t - - Specific cost of heat accumulator, $/m 3 Cw - - Cost o f water, $/m 3 /t - - S o l a r radiation intensity on collector

surface, kW/m 2 L - - Latent heat of vaporization, kJ/kg M - - Annual distillate production, m3/y Ma - - Rated (design) capacity of evaporator,

m3/d

M+< - - S t o r a g e capacity of heat accumulator, m 3

N - - Number of effects of MES evaporator and number of no-sun days

P - - Pumping power requirement, kW Pc - - P u m p i n g power of heat collecting

pump, kW P+ - - Pumping power of evaporator, kW PR - - Performance ratio Pvac - - Power of vacuum pump, kW Qe~ - - Thermal energy required by evaporator,

kJ/s T1 - - Collector fluid inlet temperature, °C T2 - - Collector fluid outlet temperature, °C Ta - - Ambient temperature, °C Tb - - Design maximum brine temperature, °C TUC - - Usable battery capacity, Ah x - - Solar collector parameter, °C m2/kW

References

[1] Research and development cooperation on solar energy desalination plant, ENAA & WED, Final Report, 1986.

[2] W.T. Hanbury and T. Hodgidess, Desalination Technology 95", Porthan Ltd., Glasgow, UK, 1995.

[3] Applications of photovoltaics, Mobil Solar Energy Corporation, Report, 1985.

[4] Photovoltalcs, electricity from the sun, Report, Solarpac Energy Systems Inc., Mississauga, Ontario, Canada, 1985.

[5] G. Fosselard and K. Wangnick, Desalination, 76 (1989) 215.

[6] M.S. Peters and K.D. Timmerhaus, Plant Design and Economics for Chemical Engineers. 3rd ed., McGraw-Hill International, 1981.

[7] El-Nashar, Computer simulation of the performance of a solar desalination plant, Solar Energy, 44(4) 099o).

Documents

The economic feasibility of small solar MED seawater desalination plants for remote arid areas