Economic Analysis Of Wind-powered Desalination - 4119

Desalination 137 (2001) 259–265

Economic analysis of wind-powered desalination L. García-Rodrígueza*, V. Romero-Terneroa, C. Gómez-Camachob a

Dpto. Física Fundamental y Experimental, Universidad de La Laguna, c/Astrofísico Francisco Sánchez s/n 38204 La Laguna (Tenerife), Canary Islands, Spain Tel. +34 (922) 318303; Fax +34 (22) 318228; email: [email protected] b Dpto. Igeniería Energética y Mecánica de Fluidos, Universidad de Sevilla, ESI, Camino de los Descubrimientos s/n., 41002 Sevilla, Spain Received 26 July 2000; accepted 14 August 2000

Abstract Wind-powered desalination is one of the most promising uses of renewable energies for seawater desalination. The influence of the main parameters on the levelized cost of fresh water was analyzed: climatic conditions, nominal power of the wind turbine, salt concentration of seawater or brackish water, design arrangement, operating conditions, plant capacity, cost of reverse osmosis modules and cost of wind turbines. In addition, the competitiveness of wind power vs. conventional energy in reverse osmosis plants was studied. Results obtained are useful, not only to quantify the influence of the parameters studied, but also to system design and to evaluate the economic perspectives of this technology. Keywords: Economics; Wind-powered desalination

1. Introduction Nowadays there are several seawater or brackish water desalination plants driven by wind power [1–6]. The optimized design has been studied by Kiranoudis et al. [7]. In Spain there are two wind-powered reverse osmosis (RO) systems, one of them in Fuerteventura Island and the other one in the Gran Canaria Island. The technological improvements of RO and wind *Corresponding author.

power systems achieved over the last decades make a promising future of wind-powered RO desalination possible. This paper presents an economic analysis of this technology. The influence of the main parameters on the levelized cost of fresh water was analyzed: climatic conditions, salt concentration of seawater or brackish water, design arrangement, operating conditions, plant capacity, nominal power of the wind turbine, cost of RO modules and cost of wind turbines.

Presented at the conference on Desalination Strategies in South Mediterranean Countries, Cooperation between Mediterranean Countries of Europe and the Southern Rim of the Mediterranean, sponsored by the European Desalination Society and Ecole Nationale d’Ingenieurs de Tunis, September 11–13, 2000, Jerba, Tunisia. 0011-9164/01/$– See front matter © 2001 Elsevier Science B.V. All rights reserved

260

L. García-Rodríguez et al. / Desalination 137 (2001) 259–265

2. Wind-powered desalination The RO module may consume either electric or mechanical power. Then a wind-powered turbine may drive a RO plant. Normally, electric power delivered from a wind farm is used; nevertheless, the direct use of mechanical power is also possible.

In this paper we considered three windpowered turbines, NM600/43, NM600/48 and NM750/44, manufactured by the NEG-Micon Danish Co. (see Table 1). The two first have 600 kW of nominal power. The NM600/48 is a suitable wind turbine for low wind speed due to its rotor diameter, greater than that of the

Table 1 Wind turbine items Wind turbine

NM600/43

NM600/48

NM750/44

Nominal power, kW Diameter, m Cost range, $/kW Representative cost, $/kW

600 43 900–1200 1050

600 48 +10–20% NM600/43 + 15 % NM600/43

750 44 900–1000 950

Table 2 Wind-powered reverse osmosis plant Wind-powered RO RO plant: Plant capacity, m3/d Specific energy consumption, kWh/m3 Membrane replacement cost, c$/m3 Investments cost, $/m3/d Chemicals, c$/m3 Availability, % O&M, 200–3000 m3/d, c$/m3 O&M, 3000 m3/d, c$/m3 Conventional energy cost, c$/kWh Wind energy resources: Weibull distribution shape parameter Annual average wind speed at hub height, m/s Wind farm: Wind turbine Investment cost, $/kW Availability, % O&M, c$/kWh Economical parameters: Real discount rate, % Life time, y Construction time, y

Range of values

Representative values

200–3000 3.5–6.5 3–10 2400–1400 4–0 85–95 60–20 15–25 6–10

200–3000 5.0 6 1400 $/m3-day 6 90 20 20 6

1.5–2.5 5–10

2.0 7

NM600/43,48,750/44 900–1200 85–95 0.5–1.5

NM600/43 1050 90 1.0

0–10 — —

5 20 1

L. García-Rodríguez et al. / Desalination 137 (2001) 259–265

261

NM600/43. NM750/44 has 750 kW of nominal power. With regard to a RO plant, Table 2 shows its main characteristics.

pump to the RO modules; (4) the properties of the membranes. • The economic and financial parameters.

3. Levelized cost of fresh water

3.1. Climatic conditions

This section presents a detailed study of the influence of different parameters on the levelized cost (LC) of fresh water for a wind-powered RO plant. The main parameters that influence the fresh water LC are: • plant capacity. • climatic conditions, the characteristic of wind turbines and the energy requirement of the RO plant that define the size of the wind farm required for a given annual production of fresh water. • energy requirement of the desalination plant, which is determined by (1) the salt concentration of the seawater or brackish water supply; (2) the coupling of the wind-powered turbine and the RO module, since the RO plant may consume either electric or mechanical power; (3) the coupling of a recovery

The wind power available in a given place is defined by the annual average of the wind speed at the hub height (VM) and the shape parameter (k). Two different plant capacities, 200 m3/d and 3000 m3/d, were selected to study the influence of the available wind power on the LC of fresh water. Fig. 1 shows that the LC goes up as the VM goes down. In addition, the greater the VM, the lower the change of LC. If VM is lower than 6 m/s, a raise of 1 m/s of VM results in a 10 c$ of LC decrease. Nevertheless, if the VM is greater than 8 m/s, the rate of decrease of the LC is about 2.5 c$·m-1·s. Otherwise, the influence of k on the LC goes down with VM, having a low effect for VM greater than 6 m/s. If VM is greater than 8 m/s, the LC slightly decreases with k, and if the VM is lower than 7 m/s, the LC slightly increases with k.

Fig. 1. Effect of k and VM on the levelized cost of fresh water. Wind turbine: NW600/43. Plant capacities: (a) 200 m3/d; (b) 3000 m3/d. Remaining parameters of wind-powered RO are the representative values given in Table 2.

262

L. García-Rodríguez et al. / Desalination 137 (2001) 259–265

3.2. Selection of the wind-powered turbine With regard to the comparison between NM600/43, NM600/48 and NM750/44, results showed that: • If the NM600/48 and NM750/44 wind turbines are compared with NM600/43, the wind speed parameters that make equivalent the LC of fresh water are not dependent on the plant capacity. • Otherwise, if k is greater than 1.5, the LC of NM600/48 wind turbines is lower than that of NM600/43 for a VM lower than 7 m/s, 6 m/s and 5 m/s when NM600/48 costs are, respectively, 10%, 15% and 20% greater than NM600/43. • Alternatively, if k is lower than 2.4, the LC of NM600/48 wind turbines is lower than that of NM600/43 for a VM lower than 9.5 m/s, 7.8 m/s and 6.4 m/s when NM600/48 costs are, respectively, 10%, 15% and 20% greater than NM600/43. • NM750/44 results in a LC of fresh water lower than that of NM600/43 if the VM >8 m/s

for k = 1.5, or VM >10 m/s for k = 2.5. Fig. 2 gives the most representative figures. 3.3. Design arrangement and operating conditions The energy requirements of a RO plant not only depend on the salt concentration of the saline water supply but also on the status of the technology. As a practical case, Glueckstern [8] gave an energy consumption of RO seawater plants in Israel, 4.5–6 kWh/m3 for current plants and 3.5–5.0 kWh/m3 in advanced technology systems. Different aspects may reduce the LC of fresh water by reducing the energy requirements as follows: (1) The future improvement of coupling RO modules to wind-powered turbines since the RO plant may use mechanical power directly instead of electric power; (2) The improvement of the energy recovery in RO plants; and (3) The development of membrane technology. Fig. 3 shows the effect of the specific consumption of the RO plant on the LC of the product.

Fig. 2. Comparative study of different wind turbines. Symbols LC600/43, LC600/48 and LC750/44 represent the LC of fresh water of wind turbines NM600/43, NM600/48 and NM750/44, respectively.

L. García-Rodríguez et al. / Desalination 137 (2001) 259–265

Fig. 3. Effect of plant capacity and the specific energy consumption on the LC of fresh water.

3.4. Future perspectives of the technology

263

Fig. 4. Levelized cost of fresh water. Horizontal and vertical axes represent plant capacity and availability of the RO plant, respectively.

3.4.2. Future perspectives of RO desalination technology

The influence of the availability on the LC of fresh water is given in Fig .4 where the real discount rate is 5%, and the remaining parameters are the representative ones given in Table 2. From Fig. 4 it may be concluded that the LC of fresh water increases 8.5% and 10% for 200 m3/d and 3000 m3/d plants, respectively, when the availability decreases from 95% to 85%. Results showed that the fresh water LC is a linear function of the costs of membrane replacement and O&M. On the one hand, the LC decreases 1.1 c$/m3 as the O&M decreases 1 c$/m3. On the other hand, the LC decreases 0.33 c$/m3 as the cost of membrane replacement decreases 1c$/m3.

The technological improvements of RO desalination technology mainly depend on membrane development. Any membrane improvement may result in longer lasting of membranes, less operation and maintenance, greater availability of the plant and lower costs of investment and membrane replacement.

3.4.3. Influence of the real discount rate The influence of the real discount rate on the competitiveness of wind-powered desalination is given in Fig. 5. This figure represents the difference between the LC obtained for a windpowered RO plant (LCwe) and a conventional

3.4.1. Future perspectives of wind-powered technology The influence of possible future decreasing of the wind farm costs was investigated for 200– 3000 m3/d of plant capacity, k = 2; VM = 7 m/s; and the real discount rate, 5%. Results show that in the range of values studied: • The LC of fresh water increases 4.5 c$/m3 per 1 c$/kWh of increasing the O&M costs. • The LC of fresh water increases 1.7 c$/m3 as investment cost of the wind farm increases 100 $/kW.

264

L. García-Rodríguez et al. / Desalination 137 (2001) 259–265

energy driven one (LCce). Results are not dependent on the plant capacity. Otherwise, the competitiveness of wind power against conventional energy in RO desalination may be represented by the quotient LCwe/LCce. Fig. 6 shows the influence of perspectives of the real discount rate and the cost of conventional energy on this competitiveness. 4. Results The influence on product levelized cost of the parameters studied is summarized in Table 3.

Table 3 Influence on levelized cost of product Fig. 5. Influence of the real discount rate on the difference of LCwe–LCce for different VMs. The remaining para-meters of wind-powered RO are the representative ones given in Table 2.

Parameter RO plant: Plant capacity Specific energy consumption Availability Membrane replacement costs O&M (without energy consumption) Wind energy resources: Average wind speed Weibull distribution shape parameter Wind farm: Wind farm cost (turbine and O&M) Wind turbine model Economics: Real discount rate

Effect on the LC 5 3–4 3 2 4

4 2–4

2–3 2–3 4–5

5: very high, 4: high, 3: medium, 2: low, 1: very low.

5. Conclusions Fig. 6. Influence of the real discount rate and the cost of conventional energy on the quotient Lcwe/LCce. The remaining parameters of wind-powered RO are the representatives ones given in Table 2.

The technological improvements of RO and wind-powered systems achieved during the last decades make a promising future possible for

L. García-Rodríguez et al. / Desalination 137 (2001) 259–265

wind-powered RO desalination since the parameter that could change in the future has a high effect on the LC. The results obtained are useful, not only to quantify the influence of the parameters studied, but also to design systems and to evaluate the economic perspectives of this technology.

[3]

[4]

[5]

Acknowlegedment This study was supported financially by the Canary Autonomous Government and La Laguna University.

[6]

References

[7]

[1] J.A. Carta and R. Calero, Era Solar, 60 (1995) 5 (in Spanish). [2] R. Coutelle, Seawater desalination by wind-powered mechanical vapour compression plants, Seminar on New Technologies for the Use of Renewable Ener-

[8]

265

gies in Water Desalination, Centre for Renewable Energy Sources,. Athens, 1991. H. Ehmann and M. Cendagorta, Proc., Mediterranean Conference on Renewable Energy Sources for Water Production, Santorini, 1996, pp. 84–87. C. Kostopoulos, Proc., Mediterranean Conference on Renewable Energy Sources for Water Production, Santorini, 1996, pp. 20–25. A. Maurel, Desalination by reverse osmosis using renewable energies (solar-wind) Cadarche Centre Experiment, Seminar on New Technologies for the Use of Renewable Energies in Water Desalination, Centre for Renewable Energy Sources, Athens, 1991. M. Stahl, Small wind-powered RO seawater desalination plant design, erection and operation experience, Seminar on New Technologies for the Use of Renewable Energies in Water Desalination, Centre for Renewable Energy Sources, Athens, 1991. C.T. Kiranoudis, N.G. Voros and Z.B. Maroulis, Desalination, 109 (1997) 195. P. Glueckstern, International Workshop on Desalination Technologies for Small and Medium Size Plants with Low Environmental Impact, Rome, 1998, Academia Nazionale Delle Scienze detta Dei XL (ed.), Rome, 1999, pp. 15–40.