Energy 35 (2010) 3245e3255

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Performance analysis of hybrid photovoltaic/diesel energy system under Malaysian conditions

K.Y. Lau, M.F.M. Yousof, S.N.M. Arshad, M. Anwari*, A.H.M. Yatim
Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor, Malaysia

a r t i c l e i n f o
Article history: Received 29 October 2009 Received in revised form 5 April 2010 Accepted 7 April 2010 Available online 15 May 2010 Keywords: Hybrid PV/diesel system Standalone diesel system Performance analysis HOMER software

a b s t r a c t
Standalone diesel generating system utilized in remote areas has long been practiced in Malaysia. Due to highly uctuating diesel price, such a system is seemed to be uneconomical, especially in the long run if the supply of electricity for rural areas solely depends on such diesel generating system. This paper would analyze the potential use of hybrid photovoltaic (PV)/diesel energy system in remote locations. National Renewable Energy Laboratorys (NREL) HOMER software was used to perform the technoeconomic feasibility of hybrid PV/diesel energy system. The investigation demonstrated the impact of PV penetration and battery storage on energy production, cost of energy and number of operational hours of diesel generators for the given hybrid congurations. Emphasis has also been placed on percentage fuel savings and reduction in carbon emissions of different hybrid systems. At the end of this paper, suitability of utilizing hybrid PV/diesel energy system over standalone diesel system would be discussed mainly based on different solar irradiances and diesel prices. 2010 Elsevier Ltd. All rights reserved.

1. Introduction In Malaysia, there are abundances of remote villages that are located far away from utility grid. These areas normally lack in the supply of electricity since it is impractical to extend the utility grid to these dispersed populated areas that are usually located in rugged terrains. As a result, such villages are normally powered by standalone diesel generators to cope for the demand of electricity. However, current increase in global fuel price has drawn serious attention of using renewable energy sources in these remote locations. Since these remote areas largely depend on diesel consumption for electricity supply, increase in global fuel prices will have great impact to these societies. Therefore, the use of renewable energy in these locations would be of great benet, especially in reducing the dependence on such highly uctuating diesel price. Solar energy is one of the in-exhaustible energy sources available for the implementation of renewable energy system in remote areas. It has been pursued by a number of countries with monthly average daily solar radiation in the range of 3e6 kWh/m2 in an effort to reduce their dependence on fossil fuels [1e4]. Malaysia, being gifted with abundance of solar radiation, has a wide potential

* Corresponding author. Tel.: 60 75535235; fax: 60 75566272. E-mail address: makbul@ieee.org (M. Anwari). 0360-5442/$ e see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.energy.2010.04.008

of solar energy applications to meet the electricity demand of remote villages. Therefore, integration of solar photovoltaic with readily available standalone diesel generators (or generally known as the hybrid PV/diesel system) has seen potential application in such remote areas. The use of hybrid PV/diesel system comes with various advantages. Among them are improved reliability, reduced emissions and pollutions, reduced cost and more efcient use of power. The maintenance cost involved for PV system is negligibly small, and can be assumed to be maintenance-free after its installation. Unlike standalone diesel generator, it has to be maintained accordingly, such as replacing diesel or components exhausted during operation to ensure reliable supply of the generator. Apart from that, the use of PV also allows it to be easily expanded to meet the growing energy needs. However, since the use of PV is still new, its initial cost, especially its installation cost is quite high. Another drawback is that PV is sunshine-dependent and its output does not match the load demand on 24-hour basis. Luiz Carlos Guedes Valente et al. [5] performed an economic analysis on hybrid PV/diesel system and demonstrated that the system has advantages over standalone diesel system. With cost analysis over a 20-year period, hybrid system was proven to reduce fuel consumption, operation and maintenance costs while improving the quality of service. This is exceptionally true for small villages with up to 100 families. The application of hybrid PV/diesel system has seen its successful implementation in Malaysia with the Langkawi Cable

3246

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

standalone diesel system was discussed mainly based on different solar irradiances and diesel prices.

2. HOMER software HOMER is a computer model that simplies the task of evaluating design options for both off-grid and grid-connected power systems for remote, stand-alone and distributed generation (DG) applications [9]. It has been developed by United State (US) National Renewable Energy Laboratory (NREL) since 1993. It is developed specically to meet the needs of renewable energy industrys system analysis and optimization. There are three main tasks that can be performed by HOMER: simulation, optimization and sensitivity analysis. In the simulation process, HOMER models a system and determines its technical feasibility and life cycle. In the optimization process, HOMER performs simulation on different system congurations to come out with the optimal selection. In the sensitivity analysis process, HOMER performs multiple optimizations under a range of inputs to account for uncertainty in the model inputs. Detailed description on HOMER software can be found in [8].

Fig. 1. Daily load prole.

Car Resort Facilities Project [6]. The hybrid system consists of diesel generators with electronic control system, lead-acid battery system, solar PV, inverter module and system controller with remote monitoring capability. The project has been proven successful in offering solution to off-grid power generation in terms of reduced operation, maintenance and logistics problem and cost, providing 24-hours reliable supply at an effective cost as well as preserving the nature. With such a good experience in hybrid PV/diesel system installation, the use of such system, especially in remote areas should gain wider consideration in Malaysia. Another successful implementation of hybrid PV/diesel project in Malaysia is described in [7]. In the paper, the authors conducted studies on the alternative energy design scheme for an Information and Communication Technologies (ICT) Telecenter. The authors remarked that hybrid PV/diesel energy system was more practical than standalone diesel generator. This has yet proven economically the potential use of hybrid PV/diesel energy system in Malaysia. According to Farret et al. [8], there are four major aspects related to distributed generation which involve the use of renewable energy. They are the hardware and control, effect on the grid, interconnection standards and economic evaluation. Since hybrid PV/diesel system is a standalone system which does not involved interconnection to the grid or with other renewable energy sources, this report will seek to analyze the potential use of hybrid PV/diesel system with and without battery to determine its suitability in remote areas, in the perspective of hardware and economical analysis. HOMER software has been used to perform the technoeconomic feasibility of hybrid PV/diesel energy system. The investigation demonstrated the impact of PV penetration and battery storage on energy production, cost of energy, number of operational hours of diesel generators for a given hybrid congurations. Emphasis has also been placed on percentage fuel savings and reduction in carbon emissions of different hybrid systems. The suitability of the hybrid PV/diesel energy system over the

3. Background information 3.1. User load It was assumed that the remote residential area consisted of a total of 40 houses. Each house required loads of 2 kW peak. Therefore, 40 houses would require a maximum of 80 kW peak demand, approximately. The daily load prole of a typical remote inhabitant area is as shown in Fig. 1. It can be noticed that load requirement varies throughout the day, with the maximum demand occurs at night. Since the area investigated was a purely residential area, most of the users will not be at home in the morning or afternoon. However, at noon, family members would be at home for lunch and rest, which caused the load demand to increase. Load requirements further changes according to each month. This is shown in Fig. 2. It was assumed that the hottest month occurs between May to August. Therefore, the load requirements would be high for those few months. However, from November to January, more cloudy days are expected and the weather would be cooler. Thus, it was assumed that those few months would require less electricity demand, especially for cooling purpose (e.g. fan). A random variability factor was given to HOMER software in order to estimate differences that may be encountered each day when using the load prole. They are known as day-to-day variability and time-step-to-time-step variability, with each approximated to be around 2% respectively. Based on all of the assumption made, the energy demand required by the remote inhabitant area, as simulated by HOMER software, was estimated to be about 1156 kWh/day (or 421.94 MWh/year).

Fig. 2. Monthly load prole.

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

3247

Fig. 3. Solar radiation data.

3.2. Solar radiation From the data provided by Malaysian Metrological Department [10], the solar radiation data for the selected remote area was estimated to range between 4.8 kWh/m2 and 6.1 kWh/m2. This is illustrated in Fig. 3. It can be noticed that more solar irradiance can be expected from the month of May to August while less solar irradiance is to be expected from November to December. This correlates to the load requirements as discussed earlier (refer Fig. 2). The scaled annual average of the solar radiation was estimated to be 5.51 kWh/m2/day. The HOMER software can generate the clearness index from the solar radiation data obtained from [10], according to the latitude of the place that has been chosen. In this analysis, the latitude was selected to be 1280 North, according to data provided by [11]. It should be highlighted that this latitude was chosen just for simulation purposes, whereby it was assumed that a remote location exist in such an area. With this information, if the solar radiation data is not available, clearness index can also be used to generate the solar radiation data. Therefore, either the clearness index or the solar radiation data can be used to represent the solar resource input, as long as the data of latitude is available to the HOMER software. 3.3. Diesel Currently available diesel price in Malaysia is RM 1.70 per liter. This value, when converted into US dollar ($), would be about $0.49 per liter. Unfortunately, according to AsiaOneNews [12], the price of

diesel can be unreasonably high in rural areas. Checks found that diesel is priced at about RM 7.00 ($ 2.03) per liter in many places in the Ulu Baram district, some 200 km inland from Miri, Sarawak. Therefore, in this simulation, diesel price was varied to determine its effect on the system, since current global fuel shortage is also causing the potential increase in diesel price. 4. Design specication In a hybrid PV/diesel energy system, there are four main components to be considered. They are the generators, PV modules, batteries and converter.
Table 1 Data for selected components. Description PV Size Capital cost/replacement cost Operating and maintenance cost Lifetime Storage battery Type of battery Nominal voltage (2 batteries per string) Nominal capacity State of charge Nominal energy capacity of each battery Capital cost Replacement cost Operating and maintenance cost Inverter Size Capital cost Operating and maintenance cost Lifetime Efciency Diesel generators Number of generators Size Capital cost Replacement cost Operating and maintenance cost Data 60 kW $ 5600/kW $ 0/year 25 years Surette 6CS25P 6 V (12 V) 1156 Ah 40% 6.94 kWh $ 1100 $ 1000 $ 10/year 60 kW $ 900/kW $ 0/year 15 years 90% 2 50 kW each $ 500/kW each $ 400/kWh each $ 0.025/hour/kW each

Fig. 4. Hybrid PV/diesel energy system.

3248

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

In order to meet the user AC load prole as discussed previously, the following design specications for each of the component are provided. 4.1. PV module Solar energy is used as the baseload power source. In an isolated system, the renewable energy contribution of 50% is considered to be high. Such a system might be very difcult to control while maintaining a stable voltage and frequency. The level of renewable energy penetration in hybrid systems (deployed around the world) is generally in the range of 11e25% [13]. The designed PV array size was 60 kW. This amount would be enough to cater for the load in the day (about 50 kW). The excess power generated would be used to charge the battery bank. Each of the proposed PV modules is rated at 50 W, with a 12 V nominal voltage. In order to obtain a total generation of 60 kW, 1200 modules were stacked together. Since the area of each module would be 0.451 m2, the total modules resulted in an area of 540.652 m2. It should be highlighted that this PV array would only generate electricity at day time, from 6 a.m. to 6 p.m. At night, there is no electricity generated. Therefore, the output from solar would be 0 W. At night, either the battery or the generators will take over the task. For economical analysis, it was assumed that each kW of PV module would cost $ 5600. The cost of replacement was assumed to be the same as the initial cost. Operating and maintenance cost was assumed to be zero since it is negligibly small. 4.2. Storage battery The storage battery chosen was Surrette 6CS25P. These batteries were congured such that each string consisted of two batteries,

Fig. 5. Conguration of hybrid PV/diesel energy system.

For the purpose of simulation, the hybrid PV/diesel energy system was designed in such a way that it consisted of two diesel generators, one photovoltaic array, one inverter and a battery bank (optional), to supply to the AC load. This is illustrated in Fig. 4.

Fig. 6. Comparison between different energy systems.

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

3249

Fig. 7. Total NPC for standalone diesel system.

with a total of six strings. This means the total batteries used were 12 units, with a bus nominal voltage of 12 V. This battery bank is capable of providing about one hour of electricity. From the datasheet given by HOMER software, the minimum state of charge of the battery is 40%. Its round trip efciency is 80%. The batterys capital cost was assumed to be $ 1100; replacement cost $ 1000; operating and maintenance cost $ 10/year. 4.3. Diesel generator Diesel generator is usually sized to meet the peak demand of the power [14]. Since the peak demand of the residential area is 84 kW, two diesel generators were chosen, each with 50 kW capacity, which resulted in a total capacity of 100 kW. The excess 16 kW from the generators will cover the spinning reserve of about 19%. This would cater for the additional loads in the future. The generators mainly operate at night since solar power is not available at this time, and the load demand is higher at night. In the day time, one of the generators would operate if the solar power and the battery could not meet the load demand. The initial cost and the replacement cost of 50 kW generator was assumed to be $ 25,000 and $ 20,000 respectively. For operating and maintenance cost, it was assumed to be quite high, with $ 1.25/hour. This is because the area considered is a remote area. Therefore, difculty arises in transportation problem when maintenance is required, which would indirectly add up the cost. Each of

the generators was assumed to have a lifetime of 15,000 operating hours. 4.4. Inverter The inverter was rated based on the selected PV array. Since 60 kW output would be generated from PV, the inverter was rated at 60 kW to fully supply the power from PV. However, it is assumed that the inverter has an efciency of 90%. Therefore, the supplied power would be less than 60 kW. The initial cost of the inverter is assumed to be $ 900, which is the same as the replacement cost. There was no operating and maintenance cost estimated. A brief summary on the data for each of the selected components is provided in Table 1. 5. Operating strategies The designed system was assumed to operate load following dispatch strategies. This means that only PV array will charge the battery bank. Generators would not charge the battery bank. They would only generate power to serve the required load. The conguration of the hybrid PV/diesel system is as shown in Fig. 5. The PV was used as the base load supply which produced DC power. It was then converted into AC source by using an inverter. Since the PV will charge the battery bank, this happens when there is extra power after meeting the demand of the end user load. If the

Fig. 8. Monthly average electric production for standalone diesel system.

3250

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

Fig. 9. Total NPC for hybrid PV/diesel system without battery.

PV cannot meet the demand, the battery bank will not be charged, but being discharged to cater for the demand. Generator 1 and Generator 2 will operate if both PV and battery bank cannot meet the demand. It should be highlighted that Generator 1 will be more frequently operated. Meanwhile, Generator 2 mainly operates at night since solar energy is unavailable at night. In other words, during the day, the PV and Generator 1 will operate. At night, Generator 1 and Generator 2 will operate. The operating reserve as a percentage of hourly load was 10%. Meanwhile, the operating reserve as a percentage of solar power output was 25%. Operating reserve is the safety margin that helps ensure reliability of the supply despite variability in electric load and the solar power supply. For an example, if the load at an hour is 50 kW and the PV output is 30 kW, this means that the operating reserve would be 5 kW 7.5 kW 12.5 kW. The diesel generator must therefore provide 20 kW of electricity plus 12.5 kW of operating reserve. This means that the capacity of the operating generator must be at least 32.5 kW. 6. Results and discussions Simulation was performed by comparing the use of standalone diesel system, hybrid PV/diesel system without battery, and hybrid

PV/diesel system with battery. It was done based on a projection period of 25 years and 6% annual real interest rate. The PV capacity was varied from 0 kW to 100 kW while the battery storage was varied from 12 units to 36 units for comparison purpose. The result of the simulation is shown as in Fig. 6. It should be highlighted that the 6% annual real interest rate was applied to the investment cost as well as the diesel price. Since HOMER assumes all prices escalate at the same rate, it is not possible to model the escalation of diesel price at different rate. Therefore, calculations are based on current diesel prices and do not reect the effects of possible further increases of the diesel prices. It is possible, however, to explore the effects of an escalating diesel price by doing a sensitivity analysis on the diesel price. It should be noted that the hybrid diesel/battery system is presented in Fig. 6. However, since the load following dispatch strategy was used, the generator will not charge the battery, except that at the beginning of the simulation, the software might assume that battery was available. But the battery will not be charged after being discharged for just one single cycle. Therefore, this combination will not be discussed, since its conguration is somehow the same as the diesel only system. However, it appeared in the simulation since HOMER software will calculate for each of the possible combinations.

Fig. 10. Monthly average electric production for hybrid PV/diesel system without battery.

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

3251

Fig. 11. Total NPC for hybrid PV/diesel system with battery.

6.1. Standalone diesel system From HOMER simulation, the use of standalone diesel system was the cheapest, with the total Net Present Cost (NPC) of $1,482,455. This was according to the global solar irradiance of 5.51 kWh/m2/d and diesel price of $ 0.49/L. Detailed breakdown of the NPC is shown in Fig. 7. It can be noticed that the largest portion of NPC came from generator 1, since it operates most of the time. Monthly average electric production is shown in Fig. 8. It can be noticed that Generator 1 will operate the most, followed by Generator 2. This means that Generator 1 served as the base load. If Generator 1 could not meet the load demand, Generator 2 would operate to cover for the inadequacy. In other words, Generator 1 is always ON (1 start/year, according to the simulation). Generator 2, on the other hand, will be turned ON and OFF depending on the load demand (847 starts/year, according to the simulation). It could be further noticed that the monthly average electric production match the monthly load prole as given in Fig. 2. As mentioned earlier, the demand for electricity is the highest from May to August, which resulted in the higher electric generation for those months. Meanwhile, for period between November to

January, the demand of load is lower. This resulted in lower generation of electricity, as shown in Fig. 8. From HOMER simulation, if only generators are used to supply for all the loads, Generator 1 would produce electricity of 344,493 kWh/year (82%), while Generator 2 would produce electricity of 77,447 kWh/year (18%). This gives the total electricity generation of 421,940 kWh/year (100%) to meet the load demand. Since solar photovoltaic is not being considered, there was no electricity from the PV. It should be noted that the cost of energy (COE) for diesel only system was $ 0.275/kWh. To determine the feasibility of hybrid PV/diesel installation, two types of congurations was analyzed, one without storage element (battery) and one with storage element (battery) respectively. 6.2. Hybrid PV/diesel system without battery As highlighted by Shaahid et al. [13], the level of renewable energy penetration in hybrid systems (deployed around the world) is generally in the range of 11e25%. From the simulation, the proposed hybrid PV/diesel system (one unit of 60 kW PV array, two units of 50 kW diesel generator, without battery) satised the PV

Fig. 12. Monthly average electric production for hybrid PV/diesel system with battery.

3252

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

Fig. 13. State of charge of battery.

penetration of 22%. The utilization of PV array size more than 60 kW is out of consideration since it would result in higher values of the total NPC as well as the COE (as compared to 60 kW PV array). In addition, higher contribution of renewable energy fraction might result in difculties in control while maintaining stable voltage and frequency [13]. On the other hand, reducing the PV array size would result in higher dependence of diesel generators. This is not favorable since the objective is to reduce the dependence on diesel. Therefore, the use of 60 kW PV array is justied. By using the proposed hybrid PV/diesel system without battery (one unit of 60 kW PV array, two units of 50 kW diesel generator, without battery), the total NPC was $ 1,669,299. This combination was the most expensive among the 22% renewable energy fraction. One of the main reasons is because the power generated by PV is not being fully utilized. The excess solar power, which could have been used to charge a battery (which was not available in this case), was considered a loss. As a result, when the PV cannot meet the demand, there was no storage element that existed to cover the increased demand. The generator will have to be operated in order to cope for the demand. Both the NPC for Generator 1 and Generator 2 have been reduced (as compared to standalone diesel system) since PV is available to supply the load. However, the initial PV cost is quite high, which mainly contribute to the extremely high total NPC cost (refer Fig. 9). The monthly average electric production is shown in Fig. 10. For Generator 1, no signicant reduction can be noticed as compared to the standalone diesel system. Electric production from Generator 2 has somehow been reduced due to the existence of PV. The PV array provided energy of 96,512 kWh/year (22%). In other words, PV penetration was 22%. Generator 1 provides energy of 292,127 kWh/ year (66%). Generator 2 provides energy of 51,871 kWh/year (12%). This means that Generator 1 still operates the most, followed by the PV and then Generator 2. Comparing this system with the

previously analyzed standalone diesel system, the electricity output of Generator 1 has reduced for about 15%. Meanwhile, electricity output from Generator 2 decreased by 33%. This means that the dependence on diesel has also been reduced. The cost of energy for this type of system is $ 0.309/kWh. Meanwhile, its operating cost would be $ 96,164/year. Since battery was not available, the operating hours of Generator 1 was the same as standalone diesel system. The use of PV caused the operating hour of Generator 2 to reduce to 3126 hours (4826 hours when using standalone diesel system). This resulted in a total saving of diesel, for approximately 16.5% as compared to standalone diesel system. However, as mentioned earlier, this type of conguration does not take into account of energy storage. The electricity production generally depends on either PV or the generator to supply the load. If PV is insufcient or unavailable, the generator will take over the task. From renewable energy viewpoint, battery storage is important to ensure that the excess of power produced from PV can be stored for later use. This would greatly optimize the system. The following discussion will focus on the hybrid PV/diesel system with battery. 6.3. Hybrid PV/diesel system with battery From the simulation result shown in Fig. 6, it can be noticed that the higher number of batteries would increase the total cost of the system. Therefore, 12 units of batteries (1 h of autonomy) were considered to be sufcient. Considering the use of hybrid PV/diesel system with battery (one unit of 60 kW PV array, two units of 50 kW diesel generator, with 12 units of battery), it can be noticed that operating cost of the system has been reduced to $ 89,170 per year. This was the cheapest among the different congurations proposed. This is a huge reduction as compared to the standalone diesel system. However, the total NPC was still high, which was

Fig. 14. Comparison between different energy systems with high diesel price.

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

3253

Fig. 15. Total NPC for hybrid PV/diesel system with battery under high diesel price.

about $ 1,593,086. This is due to the extremely high initial capital cost for PV installation, which will not occur if a standalone diesel system is used. This is shown in Fig. 11. Despite Generator 1s NPC still being high, it is lower if compared to the standalone diesel system. The same principle is applied to Generator 2. However, the use of the PV, inverter and batteries has added up the total NPC of the system. The monthly electrical energy production for hybrid PV/diesel system with battery is shown in Fig. 12. It can be noticed that PV has taken over the Generator 1s function as the supply to the base load. Since battery storage capacity was quite small (only 1 hour of autonomy supply), its portion can hardly be noticed as shown in Fig. 12. The PV array provided energy of 96,512 kWh/year (22%).

Generator 1 provides energy of 307,023 kWh/year (70%). Generator 2 provides energy of 32,086 kWh/year (7%). The batteries provide energy of 2301 kWh/year (less than 1%) due to its low energy storage. Up to this point, it should be highlighted that most of the load demand occur at night time, which is when the PV is not available. At this time, the battery will function to supply the load, since PV is not available. Unfortunately, the battery storage can only provide up to 1 hour of supply. After that, generators will operate to supply the load. With the availability of PV, Generator 1 does not need to operate nonstop throughout the year. It will only operate when the PV and batteries cannot meet the load demand. From the simulation, the

Fig. 16. Total NPC for standalone diesel system under high diesel price.

3254

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

Fig. 17. Optimal system type.

start cycle for Generator 1 is 98 starts/year. For Generator 2, it had 405 starts/year. The state of charge of the battery bank is shown in Fig. 13. Since May to August has the highest demand of energy, the battery can hardly maintain its charge. It should be noted that the minimum battery charge need to be maintained at 40%, as provided by the specication and should not fall to less than 40%. The use of battery allows the reduction of Generator 1 and Generator 2 operating hours. The operating hours for Generator 1 has been reduced to 8545 h. For Generator 2, it has been reduced to 1807 h. This shows that battery has signicant contribution in the long run. As compared with the hybrid PV/diesel system without battery, electricity generation from Generator 2 has been reduced signicantly for this type of system. At the same time, dependence on diesel was found to be much lower. The total consumption of diesel has been reduced by about 21%. Besides that, excess electricity from PV has also reduced from 9910 kWh/year (without battery) to 3999 kWh/year (with battery).

From this simulation, it can be seen that although the initial capital cost of the hybrid PV/diesel energy system was extremely high ($ 453,200) as compared to standalone diesel system ($ 50,000), it can be compensated within the projected lifetime of 25 years, with the annual real interest rate of 6%. Therefore, for high price of diesel, it was proven that the use of hybrid PV/diesel system with battery is the best solution in remote areas.

6.5. Sensitivity analysis As highlighted previously, the total NPC of hybrid PV/diesel system with battery was still higher as compared to standalone diesel system, provided if diesel price is low enough. However, if the diesel price increases considerably, the total NPC of hybrid PV/ diesel system with battery would be the lowest among all other systems. For this reason, the value of global solar irradiance and diesel price has been varied in order to determine the suitability of the implementation of the different types of energy systems. Fig. 17 shows the appropriate implementation of those three generating systems under different global solar irradiance and diesel price. The optimized option would be such that if the diesel price is less than $1.05/L, the use of the standalone diesel system would still be the cheapest among all. The choice of the hybrid PV/ diesel system is only feasible if the diesel price is more than $1.05/L. On the other hand, it can also be noticed that the use of battery in hybrid PV/diesel system does not always yield the cheapest solution as compared to the one without battery. In other words, the

6.4. High price of diesel As highlighted previously, the price of diesel can be unreasonably high in rural areas. Checks found that diesel is priced at about RM 7.00 ($ 2.03) per liter in very remote places where the cost of transportation of goods and fuel is very expensive such as in the Ulu Baram district, some 200 km inland from Miri, Sarawak. Therefore, this section seeks to discuss the effect of such high diesel price on the utilization of hybrid PV/diesel system. As shown in Fig. 14, the use of hybrid PV/diesel system with battery has beneted signicantly when the cost of the diesel is as high as $ 2.03 per liter, with the same solar irradiance of 5.51 kWh/ m2/day as analyzed previously. In such a circumstance, the use of hybrid PV/diesel with battery was found to have the least total NPC, approximately $ 4,292,632. Its operating cost would be about $ 300,346/year with its COE of $ 0.796/kWh. Fig. 15 shows breakdown of the NPC. However, if the standalone diesel system was used to supply the energy, the total NPC would yield up to $ 4,628,908 (refer Fig. 16), which is 7.8% more compared to the use of hybrid PV/diesel system with battery. Besides that, the cost of energy would also be higher, about $ 0.858/kWh, which is about $ 0.09/kWh more expensive than the hybrid system.

Fig. 18. Pollutants emissions for standalone diesel system.

K.Y. Lau et al. / Energy 35 (2010) 3245e3255

3255

system with battery might not signicantly reduce the total NPC and COE, it has been able to cut down the dependence on diesel. In addition, it also helps to reduce pollutants, such as carbon emission, thus reducing the green house effect. On the other hand, it was also proven that the use of hybrid PV/diesel system with battery would be more economical if the price of diesel increased signicantly. With a projection period of 25 years and 6% annual real interest rate, it was found that the use of hybrid PV/diesel system with battery could achieve signicantly lower NPC and COE as compared to a standalone diesel system. As a conclusion, the hybrid PV/diesel system has potential use in remote areas, especially in replacing or upgrading existing standalone diesel systems in Malaysia.
Fig. 19. Pollutants emissions for hybrid PV/diesel system with battery.

References use of battery would become the best option when the diesel price is more than $1.6/L, or when the global solar irradiance is higher. This is best explained in the shaded area as shown in Fig. 17. 6.6. Harmful gas emissions Since the use of diesel generally causes air pollution, in this section, a comparison between standalone diesel system and hybrid PV/diesel system with battery is done, in terms of harmful gas emissions. It should be noted that the analysis assumed no penalty cost to be imposed for the pollutant. However, if non-zero cost penalty is imposed for a particular pollutant, HOMER will add the cost to the total annual cost of the system. As shown in Fig. 18, for standalone diesel system, the total emissions of pollutants were 432,259 kg/year. For the hybrid PV/diesel system with battery, total emissions were 342,246 kg/year (refer Fig. 19). This has shown considerable reduction in emission with the introduction of PV system. 7. Conclusion From HOMER software simulation, it has been demonstrated that the use of hybrid PV/diesel system with battery (one unit of 60 kW PV array, two units of 50 kW diesel generator, with 12 units of battery) can signicantly reduce the dependence on solely available diesel resource. Although utilization of hybrid PV/diesel
[1] Post HN, Thomas MG. Photovoltaic systems for current and future applications. Solar Energy 1988;41(5):465e73. [2] Mahmoud M. Experience results and techno-economic feasibility of using photovoltaic generators instead of diesel motors for water pumping from rural desert wells in Jordan. IEE Proc 1990;37(6(Pt C)):391e4. [3] Frank W, Terrence G. Fundamental characteristics and economic operation of photovoltaic solar collectors. Solar Energy Technology, ASME 1992;13:75e9. [4] Douglas B. The essence way renewable energy system. Solar Today 1997;May/ June:16e9. [5] Guedes Valente Luiz Carlos, Anibal De Almeida Silvio Carlos. Economic analysis of a diesel/photovoltaic hybrid for decentralized power generation in Northern Brazil. Energy 1998;23(4):317e23. [6] Sopian K, Othman MY, Rahman MAA. Performance of a photovoltaic diesel hybrid system in Malaysia. ISESCO Sci Technol Vision 2005;1:37e9. [7] Abdullah MO, Yung VC, Anyi M, Othman AK, Ab. Hamid KB, Tarawe J. Review and comparison study of hybrid diesel/solar/hydro/fuel cell energy schemes for a rural ICT Telecenter. Energy; 2009. doi:10.1016/j.energy.2009.10.035. [8] Farret FA, Simoes MG. Integration of alternative sources of energy. New Jersey: John Wiley and Sons Inc; 2006. [9] United States (US) National Renewable Energy Laboratorys (NREL) HOMER Software. See also, https://analysis.nrel.gov/homer. [10] Malaysian Metrological Department. Monthly weather bulletin. See also, http://www.met.gov.my/index.php?optioncom_content&taskview&id846 &Itemid1586. [11] Maps of World. Malaysia Latitude and Longitude. See also, http://www. mapsofworld.com/lat_long/malaysia-lat-long.html. [12] AsiaOneNews. Paying high prices for diesel. See also, http://news.asiaone. com/News/AsiaOne%2BNews/Malaysia/Story/A1Story20090415-135337.html. [13] Shaahid SM, El-Amin I. Techno-economic evaluation of off-grid hybrid photovoltaic-diesel-battery power systems for rural electrication in Saudi Arabia e a way forward for sustainable development. Renew Sustain Energy Rev 2009;13:625e33. [14] Nfah EM, Ngumdam JM, Tchinda R. Modelling of solar/diesel/battery hybrid power systems for far-north Cameroon. Renew Energy 2007;32:823e44.