An Investigation Into Economic Desalination Using Solar/Wind Energy System

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

733

AN INVESTIGATION INTO ECONOMIC DESALINATION USING SOLAR / WIND ENERGY SYSTEM Ahmed Hossam-Eldin 1 and Ahmed Ismaiel 2 1

Dr.,

2

Eng.

Electrical Engineering Department, Faculty of Engineering Alexandria University, Alexandria, Egypt

ABSTRACT Desalination has now been practiced on a large scale for more than 50 years. A desalination system driven by a hybrid power supply system (HYB) consisting of renewable energy source (RES) is described. The system is considered to be an attractive solution to provide water and electricity in certain arid areas, where water scarcity is severe and without grid of electricity. Reverse osmosis (RO) desalination process was chosen since is a fairly mature technology. Further improvements are likely to be incremental in nature, unless design improvements allow major savings in capital costs for different capacities. A unit of RO was studied with different power supply sources for both seawater and brackish water. The HYB power source showed to be the most promising configuration for desalination using renewable energy technologies. It showed to be the most competitive from the cost point of view and the superiority from the point of view of environmental pollution. In this study a HYB-RO computer program has been developed on the above basis to choose the proper size and to calculate the cost per unit. It showed to be very effective in designing such versatile system. It can be used for the design of hybrid power supply system to drive a reverse osmosis unit for the desalination of brackish water and seawater. It is used to assist the designer to develop an optimal configuration for the hybrid power supply system as well as the optimal water production. Cost analysis was investigated and it was concluded that the developed technique brings a moderate cost compared with the traditional desalination methods. Keywords: Renewable energy; Reverse osmosis; Desalination; Arid areas

INTRODUCTION Water shortages affect 88 developing countries that are about one half of the world’s population. Arab Region has serious water shortage problems due to the fact that most of the Arab countries are located in arid and semi-arid zones known for their scanty annual rainfall, very high rates of evaporation and consequently extremely insufficient water resources. Sustainable management of water resources is great because water scarcity is becoming more and more a development constraint impeding the economic growth of many countries in the Arab region [1].

734

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

Arid areas, with even, few inhabitants need a continual source of fresh water especially for pilot investigations and army camps. It is difficult for the utility network of electricity to reach these areas [2,3,4]. It is also difficult to transport fuel to diesel engine sets of large sizes. For these reasons it was essential to find a solution for such a problem. It was the Egyptian future vision in the field of desalination to be based on renewable energy [5]. Renewable energy produced by wind turbines (W/T) and photovoltaic cells (PV) seems to be a good solution for such projects. This solution is dependent upon the meteorological data in these areas. W/T can be used effectively when mean wind speed is suitable [2,3], while PV can be very useful in areas where high solar radiation is abundant allover the whole year [3]. The environmental benefits can be added to the advantages of these methods.

METHODOLOGY A program in (MATLAB) language simulates the performance of a hybrid power supply system, combining W/T and PV arrays, which provides the required power to a reverse osmosis plant was developed. It calculates the size of both PV arrays and W/TS. It also calculates the cost of fresh water produced in Egyptian pounds (L.E.). The approach provides a good mean to examine several scenarios for the power supply system, i.e. wind/PV, PV/backup diesel, wind/backup diesel, for the location and number of inhabitant under study. The approach was repeated for sea and brackish water (from digged wells). Figure (1) illustrates a typical HYB-RO system configuration, which was considered in our study. Figure (2) shows a simple Reverse Osmosis plant.

Photovoltaic Generator

Wind Generator

Charge Controller

Inverter

Charge Controller

Backup Diesel Generator

Auxiliary Loads

Battery System

RO Unit -HPP -BP

Converter

Figure (1) HYB-RO system configuration

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

735

HPP

Figure (2) Simple Reverse Osmosis Plant

1. Determination of the water requirements The water requirements, means the daily water demand (dwd) from the RO plant in m3/day. This can be calculated from the following equation. (dwd) = population (inhabitant) * (dwd / inhabitant)

(1)

A daily water demand of 50 liters/inhabitant is assumed to be satisfactory in this study for arid areas.

2. Determination of the feed water data It has to be determined which type of water is available in this area (brackish water or sea water). The feed water quality is to be indicated such as salt density index (SDI) and salinity or total dissolved solids (TDS) in parts per million (ppm) as well as temperature in degree Celsius, pH factor, etc… . As far as RO desalination process is considered, reverse osmosis system analysis program (ROSA) software from Film Tec [6] is used to give complete design of the plant with respect to the type of feed water and to determine the type of membranes suitable for this purpose.

3. Determination of the energy requirements The energy requirement from a hybrid energy system is equal to the energy required to drive the RO unit and the auxiliary loads. The following steps are to be taken before determination of the size of the energy source and its components. 3.1 Energy to drive RO plant - Recovery ratio – The recovery ratio R of the system is defined as follows: R= MP / MF

(2)

736

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

where:

MF

is the feed flow rate (m3 s-1) and

MP

is the product flow rate (m3 s-1).

- Brine flow rate – The brine flow rate is calculated by overall material balance as: MB =MF – MP

(3)

where: MB is brine flow rate (m3 s-1). - Booster pump – The total power required to run a booster pump (PBP) is given by: PBP = ρ *g*H* MF /η BP where:

(4)

ρ is the feed water density at 25°C (kg m-3), H is the intake manometric height (m) and equal to static head (m) + friction head losses in pipeline (m) [7]. g is the acceleration due to gravity (9.81 m s-2) and η BP is the booster pump efficiency.

- High-pressure pump – The power required to run a high-pressure pump (PHPP) is given as: PHPP = prF* MF /η HPP where:

(5)

prF is the feed pressure in (bar) and η HPP is the high- pressure pump efficiency.

- Energy recovered – The energy recovered by the energy-recovery turbine (ER) is calculated as: ER = prB * MB * η T where:

(6)

prB is the brine pressure in (bar) and η T is the turbine efficiency.

It has to be noted that no recovery was considered in our study. 3.2 Auxiliary RO unit loads Several dosing pumps will be used in the pretreatment and post treatment sections. Although the dosing pumps consume insignificant power compared to the main pumps in the plant, their contribution to the load is considered to be 1 kW. - Annual energy requirement from hybrid energy system - (AN.E.HYB.) is given as: AN.E.HYB. = [8760 * ( PBP+PHPP+ Auxiliary loads )] in kWh

(7)

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

737

In the following sections, we will discuss the different types of energy sources.

4. Wind generator To determine the wind turbine rated power, the suitable (W/T) model has to be chosen according to the site meteorological data. The NORDEX model is the most suitable for Egyptian climate (mean wind speed in between 6.5-7 m/s) [8]. Assuming an average yearly operational time of 8 months/annum, then The annual (W/T) energy is equal to [8*30*24* (W/T) rated power] kWh

(8)

5. Photovoltaic generator Determination of the power requirements from (PV) modules is also done according to the site meteorological data. The type BP Solar, BP5200 module was chosen because its nominal power is relatively high [9]. The required numbers of (PV) modules is given as: (PV)modules= [the power requirements for RO plant/module mpppower]

(9)

where: Module mpp is the maximum power point of the PV modules. Assuming an average daily operation time of 7 hours and thinking of no black out days during this average period, then The annual (PV) energy is [7 * 365*(PV)modules* module mpppower] kWh (10)

6. Battery system The daily shining time is actually more than 7 hours especially during summer time. In all cases, there is surplus power from the renewable energy sources. This power has to be stored in battery system (BATT.) to allow for supplementary power in case of any failure. This would maintain a high quality of electric power to supply the desalination plant. To model a battery system the following steps are to be taken: Assuming charging losses = Discharging losses = 10% of BATT. Power [10], then the output power from BATT. must be equal to [(RO)unit load + charging or discharging losses]. This means that, output power from BATT. in kW =1.1 * (RO)unit load

kW

(11)

BATT Net capacity=[1.1 * (RO)unit load (kW)] during one hour kWh

(12)

Required number of batteries Nbatt..= BATT.Net capacity/single battery capacity (13)

738

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

7. Diesel generator Determination of the suitable diesel generator set (GEN) which can drive the RO plant is evaluated according to the (RO)unit load (kW). Obviously a single diesel generator set is available in arid area. The annual fuel consumption rate in (liter) is equal to [GEN.fuel consumption rate in (liter/kWh) * GEN. nominal power (kW)*Annual GEN. operational time (h)] (14)

ECONOMIC ANALYSIS 1. Wind turbine The capital cost of large wind turbines (>100 kW) is about L.E.7000/kW. Smaller wind turbines (W/T) of 20–60 kW will cost around L.E. 14000/kW, while very small W/T (between 1 and 15 kW) will cost about L.E.17500–21000/kW. Other initial costs include the transportation and installation of the turbine, which also vary with the size of the turbine and the installation location. The operation and maintenance (O&M) costs of a W/T are consisted of the cost of spare parts and W/T service charges. Typical O&M costs are around 3–5% of the equipment capital cost. They are around 1.5–2% for large W/T. The lifetime of a W/T is around 20 years [3].

2. Photovoltaic modules The capital cost of the PV modules is around L.E.33000/kW. Other initial costs, the transportation and installation, are usually around 8–15 % of the capital cost. The maintenance cost of the PV system is around 1% of the capital cost per year. PV modules have a lifetime of about 20 years [3].

3. Battery system The capital cost of the batteries is around L.E.1190/kWh. Their maintenance cost is negligible. The lifetime of the batteries mainly depends on their operation and maintenance and is usually 2-3 years [10].

4. Diesel generator This has a relatively low initial capital cost, around L.E.2800/kW. The (O&M) costs of a diesel engine mainly consist of the fuel cost to run the engine and service charges. Diesel engine lifetime is about 10 years [3].

5. RO plant The capital cost of a brackish water RO unit ranges between L.E.2205/m3/day and L.E.12250/m3/day, while for seawater RO it is L.E.7210– 36400/ m3/day. Small units

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

739

with a capacity of 0.5 m3/h cost more than the large units with a capacity of 15 m3/h. In general, the lifetime of RO unit is 20 years.

6. Levelized cost of water It is the cost of 1 cubic meter of fresh water produced in Egyptian pound (L.E.), this can be calculated as: Cost / unit = [total cost /annual production of water], for a lifetime of 20 year

(15)

CASE STUDIES Sixteen (16) simulation case studies for brackish-water and 16 for seawater are carried out to evaluate the performance of HYB-RO model. Assuming a daily water demand of 50 liter per inhabitant, the software package ROSA was used for designing the RO plant for the following 2 cases (each case consists of 4 different populations):

1. Case 1 1. Population = 1000 i.e. (RO plant capacity = 50 m3/d), brackish water source. 2. Population = 1500 i.e. (RO plant capacity = 75 m3/d), brackish water source. 3. Population = 2000 i.e. (RO plant capacity =100 m3/d), brackish water source. 4. Population = 5000 i.e. (RO plant capacity =250 m3/d), brackish water source.

2. Case 2 1. Population = 1000 i.e. (RO plant capacity = 50 m3/d), for seawater source. 2. Population = 1500 i.e. (RO plant capacity = 75 m3/d), for seawater source. 3. Population = 2000 i.e. (RO plant capacity =100 m3/d), for seawater source. 4. Population = 5000 i.e. (RO plant capacity =250 m3/d), for seawater source. Several HYB-RO simulation runs have to be carried out to evolve an optimal power supply system to run each of the above 8 RO units. The following 4 hybrid energy system design cases were examined for this purpose: a. b. c. d.

(W/T+PV+BATT) energy system. (PV+BATT+GEN) energy system. (W/T+BATT+GEN) energy system. (GEN, only) energy system.

740

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

RESULTS AND DISCUSSIONS 1. Brackish water 1.1. (w/t + pv + batt.) hybrid sys. W/T

Population 1000 1500 2000 5000

No. 1 1 1 1

kW 3 6 6 20

PV module No. kW 12 2.04 24 4.08 24 4.08 71 12.07

BATT No. kWh 18 3.23 25 4.5 31 5.63 103 19.06

Cost / unit (L.E.) 5.06 5.21 3.77 3.25

1.2. (pv + batt. + diesel gen.) hybrid sys. Population 1000 1500 2000 5000

PV module No. kW 22 3.74 27 4.59 36 6.12 112 19.04

BATT No. kWh 18 3.23 25 4.5 31 5.63 103 19.06

GEN (kW)

Cost / unit (L.E.)

4 6 6 20

5.06 4.95 3.67 3.23

GEN (kW)

Cost / unit (L.E.)

4 6 6 20

5.19 4.87 3.51 2.95

1.3. (w/t + batt. + diesel gen.) hybrid sys. Population 1000 1500 2000 5000

W/T No. 1 1 1 1

kW 6 6 6 20

BATT No. kWh 18 3.23 25 4.5 31 5.63 103 19.06

1.4. (Diesel gen. Only) sys. Population

GEN (kW)

Cost / unit (L.E.)

1000 1500 2000 5000

4 6 6 20

4.74 4.7 3.38 2.84

Figure (3) Shows relationship between brackish water desalination cost in Egyptian pound and RO unit capacity (population) for different hybrid energy systems.

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

741

From the earlier results it is clear that for 250 m3/d (population = 1500) produced from brackish water the cost/unit using wind turbine is almost the same as that for diesel generator set with advantages of no pollution. 5.5

cost / unit (L.E./cubic meter)

5

[w/t+pv+batt.] hyb. sys. [pv+batt.+diesel gen.] hyb. sys. [pv+batt.+diesel gen.] hyb. sys. [diesel gen. only.] sys.

4.5

4

3.5

3

2.5

1

1.5

2

2.5

3

3.5

4

4.5

5

population*1000

Figure (3) brackish water desalination cost vs. population

2. Sea water 2.1. (w/t + pv + batt.) hybrid sys. Population 1000 1500 2000 5000

W/T No. 1 1 1 2

kW 20 30 55 80

PV module No. kW 59 10.03 89 15.13 118 20.06 236 40.12

BATT No. kWh 116 21.53 151 27.98 232 43.11 440 81.76

Cost / unit (L.E.) 13.43 13.51 13.25 8.72

2.2. ( pv + batt. + diesel gen.) hybrid sys. Population 1000 1500 2000 5000

PV module No. kW 128 21.76 165 28.05 265 45.04 486 82.62

BATT No. kWh 116 21.53 151 27.98 232 43.11 440 81.76

GEN (kW) 20 50 50 75

Cost / unit (L.E.) 13.9 14.04 13.34 9.09

742

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

2.3. (w/t + batt. + diesel gen.) hybrid sys. Population

W/T No. 1 1 1 2

1000 1500 2000 5000

kW 20 30 55 80

BATT No. kWh 116 21.53 151 27.98 232 43.11 440 81.76

GEN (kW) 20 50 50 75

Cost / unit (L.E.) 12.17 12.25 11.99 7.71

2.4. (Diesel gen. Only) sys. Population

GEN (kW)

1000 1500 2000 5000

20 50 50 75

Cost / unit (L.E.) 11.58 12.54 11.05 7.26

Figure (4) Shows relationship between seawater desalination cost in Egyptian pound and RO unit capacity (population) with different hybrid energy systems.

15 14

[w/t+pv+batt.] hyb. sys. [pv+batt.+diesel gen.] hyb. sys. [pv+batt.+diesel gen.] hyb. sys. [diesel gen. only] sys.

cost/unit (L.E./cubic meter)

13

12

11 10

9

8

7 1

2

3

4

5

population*1000

Figure (4) Seawater desalination cost vs. population

6

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

743

From the earlier results it can be seen that for 75 m3/d (population = 1500) produced from sea water the cost/unit using wind turbine is cheaper than using diesel generator set. This makes using renewable energy systems superior.

CONCLUSIONS This study provides the following important conclusions:
The present cost / unit using brackish-water RO unit is roughly one half that of seawater RO unit.
As the rate of RO unit capacity is increased the cost of water produced is reduced.
When PV arrays are used the cost / unit is relatively increased, this is due to the fact that PV modules have high capital cost.
It is concluded that the cost / unit by means of W/T-RO system is competitive to the diesel-RO system which has the lowest value (provided that diesel fuel is abundant at that site).
Renewable energy sources are good means of electric power supply to RO desalination plants. It is superior to other means from the point of view of clean energy.

REFERENCES [1] Radwan Al-Weshah, “The role of UNESCO in sustainable water resources management in the Arab World”, Desalination, 152, (2002), 1–13 [2] Middle East Desalination Research Center Series of R&D Reports, Project: 97AS-008a “Hybrid Desalination Systems – Effective Integration of Membrane / Thermal Desalination and Power Technology”, December 2001. [3] Middle East Desalination Research Center Series of R&D Reports, Project: 97AS-006a “Matching Renewable Energy with Desalination Plants”, September, 2001. [4] Fath H. E., “Desalination Technology”, Eldar-Elgameia, Alexandria, Egypt, 2001. [5] Mohmoud Abou Zaid, “Desalination in Egypt between the Past and Future Prospects”, the Newsletter of the Middle East Desalination Research Center, Issue 9, March 2000. [6] FilmTec. ROSA, Reverse Osmosis System Analysis Program, Version 5.3, (May 2003). [7] Khurmi R. S., “A Textbook of Hydraulics, Fluid Mechanics and Hydraulic Machines”, S. Chand, New Delhi, India, 1998. [8] Report on “Wind Energy”, New and Renewable Energy Authority (NREA), Ministry of Electricity & Energy, Egypt, 2003. [9] “BP 5170 module”, BP Solar global marketing, 2002, http://www.bp solar.com.

744

Tenth International Water Technology Conference, IWTC10 2006, Alexandria, Egypt

[10] Stevens J. W. and Corey G. P., “A Study of Lead-Acid Battery Efficiency Near Top of Charge and the Impact on PV System Design”, Sandia National Laboratories, Albuquerque, New Mexico. [11] Buros, O. K., “The ABCs of Desalting”, Second Edition, the International Desalination Association (IDA); sponsored by the Saline Water Conversion Corporation (SWCC). [12] The International Renewable Energy Magazine (REFOCUS), “Desalination Powered by Renewable Energy Sources”, July/Aug., 2001. [13] Frerris L. L., "Wind Energy Conversion Systems", Prentice-Hall, 1990. [14] Manwell, J. F., A. Rogers, G. Hayman, C. Avelar and J. McGowan, “Hybrid2-A Hybrid System Simulation Model”, Theory Manual, University of Massachusetts, USA, 1998.