ARTICLE IN PRESS

Ocean Engineering ] (]]]]) ]]]]]]

Contents lists available at ScienceDirect

Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng

Application and costbenet analysis of solar hybrid power installation on merchant marine vessels
Alexandros Glykas a,, George Papaioannou b, Stylianos Perissakis b
a b

University of the Aegean, Department of Shipping Trade and Transport, 8 Korai Str. Chios 82100, Chios, Greece National Technical University of Athens, Greece

a r t i c l e in fo
Article history: Received 10 October 2008 Accepted 31 January 2010 Keywords: Photovoltaic Green energy Bulk carriers Ship

abstract
Photovoltaic systems are renewable energy sources with various applications and their implementations in energy production and saving are veried. Installing those systems onto merchant marine vessels could prove to be an efcient way of minimizing fuel costs and simultaneously protecting the environment by reducing signicant carbon dioxide emissions. This paper examines the feasibility of installing solar panels onto vessels and also calculated the payback period from the adopted investment with respect to fuel oil savings. Thus, the two important parameters incorporated in the parametric analysis are the solar radiation density and the fuel cost. In order to calculate the energy production of the solar installation systems, the globe is divided in six different zones, according to solar radiation density (Stackhouse and Whitlock, 2008). For one square meter of the considered solar panels the peak output power is taken equal to 130 W (Kagaraki, 2001). The payback period of the investment depends greatly on the fuel prices. For a reasonable fuel price annual increase at about 1015% the estimated payback period varies from 16 to 27 years. The more the fuel oil increases, the methodology reveals that the payback period converges to a minimum of 10 years. When using any storage media such as hydrogen, the methodology shows that the payback period increases and this depends on the proportion of the energy stored and from the storage media itself. & 2010 Elsevier Ltd. All rights reserved.

1. Introduction Nowadays the pollution caused by the marine industry is calculated to be double to what initially was assumed. The annual carbon dioxide emissions from tanker vessels and bulk carriers equal the emissions from the whole USA. Marine industry has major responsibilities for the Greenhouse effect and measures against profuse pollution should be taken (Papalambopoulos Michail and Glykas Alexandros, 2008). The presented paper is based on the concept of hybrid power systems. The term hybrid power system (Hybrid Power Systems, 2008) is used to describe any power system with more than one type of generator (usually a conventional generator powered by a diesel or gas engine) and a renewable energy source such as a photovoltaic (PV), wind, or hydroelectric power system. There are thousands of those systems in use today. They range in size from a few tens of watts to tens of kilowatts. The smaller systems are mostly on remote residences where homeowners add a few PV modules to their existing Genset to reduce the noise and inconvenience of having the generator running all the time.

Should there be multiple generators, a smart controller is required. The ultimate smart controller is a human; however many hybrids are installed in remote locations which rely on electronic controllers. Based on parameters, such as the load demand, generator status, and battery state of charge, the controllers should act to keep power owing to the load as well as to protect the equipment. Performing this with efciency is not an easy task due to the fact that most hybrid controllers are custom designed. Except from the generators and the alternative energy sources, hybrid power systems also consist of an inverter (often called a power conditioning unit or PCU), which is necessary to drive AC loads from a battery or a photovoltaic array. A battery charger/ rectier is required if batteries are to be charged from an engine generator or the alternative energy source and nally there is the balance of systems (BOS).

2. Installation of photovoltaic systems on vessels The main idea is to examine whether an implementation of solar energy on motor vessels would be costbenet and would actually assist in reducing fuel oil consumption along with the emissions of CO2.

Corresponding author. Tel.: + 30 2271035272: fax: + 30 2271035299.

E-mail address: aglykas@aegean.gr (A. Glykas). 0029-8018/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.oceaneng.2010.01.019

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
2 A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]]

Installation of photovoltaic systems on merchant marine vessels is innovative not only conceptwise but also because the PV systems have to be adopted within the abnormal and noncontinues geometric area of the vessel. Within this paper the costbenet analysis is examined and results are extracted with reference to the applicability and viability of PV systems. The parameters incorporated in the analysis are the solar radiation density (SRD) and the fuel cost increase (FCI) during the last decades. Data for SRD have been extracted from statistical data (Stackhouse and Whitlock, 2008). All nancial and commercial values used for items such as PV systems, Inverters, Fuel oil, etc. are reecting current market prices (June 2008).

dirtiness of its upper surface. The dirtier the surface is, the lower the performance is. The dirtiness performance fraction was taken to be equal to 0.93 since it is assumed that the developed system will operate at the sea, which is considered to be a clean environment.

4. Solar radiation density The solar radiation density depends on the geographical latitude and the season. The most determinant factor is the geographical latitude. The closer to the equator, the higher the solar radiation density (Stackhouse and Whitlock, 2008). With respect to the effect of seasons, it was found that during the summer, solar density reaches higher rates compared to the winter. For merchant marine vessels there is no constant reference point since they always operate within different zones worldwide. Calculation of the solar radiation density referring to vessel operation is a complicated analysis, contrary to inland implementation where such information is easy to obtain from statistical data. All the same, for solar installations on merchant vessels operating worldwide, the statistical data required include a wide range of areas. Such data were collected from NASAs statistical tables (Stackhouse and Whitlock, 2008), where there was a detailed record on a daily basis for the solar radiation density, for every latitude and longitude and from year 1983 to 2005. From all those records an average was calculated for different latitudes. The globe was divided into 6 zones. Each zone has a range of 301 in latitude. The zones were divided amongst the latitude instead of the longitude since the variation of the solar radiation density depends on the latitude. Hence the three developed zones are the following:

3. Average PV performance, heating and dirtiness performance fraction The performance of the P/V panels uctuates between 10% and 22% (22% is mostly found in laboratory applications). The P/V panels, based on polycrystalline silicon, in commercial usage can perform around 1315% (Kagaraki, 2001). In this project the performance considered is 13% (the lower possible). The heating performance coefcient (Kagaraki, 2001) is an index that shows the connection between temperature and P/V performance. The higher the temperature on the P/V panels, the lower the performance of the system. In room temperature the heating performance fraction equals 1 and for every 1 1C temperature increase, it is reduced by 0.005. Graph 1 shows this correlation. In the presented analysis the heating performance fraction was taken to be equal to 0.9 with the assumption that the maximum developed temperature on the panel surface would not exceed 70 1C at the sea environment. The dirtiness performance coefcient (Kagaraki, 2001) shows the connection between the performance of the P/V panel and the

1.2 1 0.8 0.6 0.4 0.2 0 10 Temperature (C) 100

performance

 zone 1 (0291 north),  zone 2 (30591 north) and  zone 3 (60891 north).
Similarly for the south hemisphere there is

 zone 1 (1291 south),  zone 2 (30591 south) and nally  zone 3 (60901 south).
Fig. 1 presents the division of the globe in all those zones.

Graph 1. Correlation between the performance of P/V panel and the temperature which evolves on it. The temperature axis is on logarithmic scale.

Fig. 1. Different zones according to different latitudes.

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]] 3

The solar radiation density units are shown in Eq. (1).

pzone_i : CkWh=m2 day

Table 1 shows the solar radiation density for each zone.

5. Calculation of solar installation The peak power of a solar plant with reference to the daily demand is given from Eq. (2) as PkWh : EkWh=day 1kW=m2 pkWh=m2 daysy sr 2

solar panels from 500 to 10,000 m2. Taking into account the market value of the P/V panels (World Solar Energy News, 2008), an estimation of the installation cost can be made. Furthermore, with respect to current (5/7/2008) oil prices (ISO 8217, 2005) and the energy connection between oil and kWh (The Complete Guide to the Bunker Market, 2008) a costbenet study is applicable (on the basis of the production of energy kWh per day). The peak power for a solar plant depends from the available surface and the average performance of the P/V panels. Eq. (2) can be also expressed as a function of the surface S and the average performance of the solar panel (n) as follows: PS : 1kW=m2 Sn 3

where P is the power that the whole installation can give; p the solar radiation density, which depends on mostly the latitude of the area; E the energy produced every day in kWh. It depends on the latitude and surface covered by solar panels; sy the performance fraction of solar panels because of the heating. The higher the temperature, the lower the efciency of the P/V; sr the performance fraction because of the dirtiness of the upper cover of the solar panels; 1 denotes 1 Sun. It is the base reference for solar radiation density. An average performance of 13% means that 1 m2 of surface may produce 130 W of power, for solar radiation density 1 kW per square meter (1 Sun). Another very important aspect of the installation is the available surface. On a vessel the available surface for solar panels installation may vary, depending on the dimensions and type of the vessel. An average available surface for solar panels installation on bulk carriers has a range from 500 m2 in small vessels to 20,000 m2 in bigger ones (Robert Scott, 1993). This study will focus on area sizes from 500 to 10,000 m2 since 90% of the world commercial eet of bulk carriers has a deck area of less than 10,000 m2 (Clarkson Research Studies, 2008). An arithmetic calculation of the total energy that may be produced per day in each zone will follow, for surfaces covered by
Table 1 Solar radiation density for each zone. Zone (i) 1 2 3 1 2 3 Latitude (deg.) 029 3059 6089 129 3059 6090 C North North North South South South 5.610 3.720 2.339 5.703 3.646 2.739

where P(S) is the peak power for a solar plant; n the average performance of the solar panel (13%); and S the surface covered by solar panels (50010,000 m2). The amount of energy production depends on the area where the vessel is located and therefore it is dependent on the solar radiation density p, the total covered surface with solar panels, the dirtiness of the upper surface of the panel (sr =0.93) and the temperature that is evolved on the solar panel (sy = 0.90). The resulting energy produced every day in kWh is given as ES; p : pzone sy sr 1kW=m2 Sn 1kW=m2 4

Graph 2 shows the differences between the 6 zones in energy production per day. It is noticeable that the symmetricals to the equator zones have very similar energy production capacities. Table 2 shows the daily and annual energy production for each zone in connection with the total surface of panels installed.

6. Costbenet study Other alternative energy methods are still very expensive regarding development and implementation on merchant vessels. A costbenet study is adopted and presented herewith in order to obtain the cost inclusive of raw materials and installation of P/V systems on merchant vessels. Additionally the internal rate of return of such investment is calculated along with the required payback period. The cost for a Solar energy is taken equal to 4.82 USD per Watt (installation and wiring included) (World Solar Energy News, 2008) or 4820.00 USD per kW peak. All the P/V systems produce a constant output DC voltage (Kagaraki, 2001). The magnitude of such DC voltage depends on the way the panels have been connected amongst each other during installation. However all merchant vessels use AC voltage of either 380 or

7,000.00 zone 1 6,000.00 5,000.00 kWh/day 4,000.00 3,000.00 2,000.00 1,000.00 0.00 0 1000 2000 3000 4000 5000 6000 7000 available surface (m2) 8000 9000 10000 zone -1 zone 2 zone -2 zone 3 zone -3

Graph 2. Energy production per day in the 6 zones.

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
4 A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]]

Table 2 Energy production in each zone and for different surfaces of solar panels. Surface of panels (m2) Zone 3 Solar radiation density p = 2.34 (kWh/m2 per day) kWh/day 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10,000 127 254 382 509 636 763 891 1018 1145 1272 1400 1527 1654 1781 1909 2036 2163 2290 2418 2545 kWh/year 46,442 92,884 139,325 185,767 232,209 278,651 325,092 371,534 417,976 464,418 510,859 557,301 603,743 650,185 696,626 743,068 789,510 835,952 882,393 928,835 Zone 2 Solar radiation density p = 3.72 (kWh/m2 per day) kWh/day 202 405 607 810 1012 1214 1417 1619 1822 2024 2226 2429 2631 2834 3036 3238 3441 3643 3846 4048 kWh/year 73,878 147,756 221,634 295,512 369,389 443,267 517,145 591,023 664,901 738,779 812,657 886,535 960,412 1,034,290 1,108,168 1,182,046 1,255,924 1,329,802 1,403,680 1,477,558 Zone 1 Solar radiation density p = 5.61 (kWh/m2 per day) kWh/day 305 610 916 1221 1526 1831 2136 2442 2747 3052 3357 3663 3968 4273 4578 4883 5189 5494 5799 6104 kWh/year 111,402 222,805 334,207 445,610 557,012 668,414 779,817 891,219 1,002,622 1,114,024 1,225,426 1,336,829 1,448,231 1,559,634 1,671,036 1,782,438 1,893,841 2,005,243 2,116,646 2,228,048 Zone 1 Solar radiation density p = 5.70 (kWh/m2 per day) kWh/day 310 621 931 1241 1551 1862 2172 2482 2792 3103 3413 3723 4033 4344 4654 4964 5274 5585 5895 6205 kWh/year 113,244 226,489 339,733 452,978 566,222 679,467 792,711 905,955 1,019,200 1,132,444 1,245,689 1,358,933 1,472,177 1,585,422 1,698,666 1,811,911 1,925,155 2,038,400 2,151,644 2,264,888 Zone 2 Solar radiation density p = 3.65 (kWh/m2 per day) kWh/day 198 397 595 793 992 1190 1389 1587 1785 1984 2182 2380 2579 2777 2975 3174 3372 3570 3769 3967 kWh/year 72,401 144,801 217,202 289,602 362,003 434,403 506,804 579,205 651,605 724,006 796,406 868,807 941,208 1,013,608 1,086,009 1,158,409 1,230,810 1,303,210 1,375,611 1,448,012 Zone 3 Solar radiation density p = 2.74 (kWh/m2 per day) kWh/day 149 298 447 596 745 894 1043 1192 1341 1490 1639 1788 1937 2086 2235 2384 2533 2682 2831 2980 kWh/year 54,393 108,785 163,178 217,570 271,963 326,355 380,748 435,141 489,533 543,926 598,318 652,711 707,103 761,496 815,889 870,281 924,674 979,066 1,033,459 1,087,852

Electrolysis

Storage media for Hydrogen

Hydrogen usage

P/V panels DC OUTPUT Control Electronics

DC to AC Inverter

AC Load

Fig. 2. Sketch of an electrical installation of P/V systems.

440 V (Robert Scott, 1993). Hence, in order to supply the vessel using solar energy, inverters are a prerequisite. A simple installation scheme of P/V systems is shown in Fig. 2. Further to US statistics an average cost of inverters per continuous watt is considered equal to 0.721 USD or 721.00 USD/kW (Kapros, 2006). Total cost of the investment (solar panels and inverters included) is shown as follows: Cinvestment : Csolar_panels Cinverters The cost for the whole solar installation is Csolar_panels : PSCpanels_kWp 6 5

The cost of panels per kWp is equal to Cpanels_kWp : 4820USD=kW The cost for inverters for the whole solar installation is Cinverters : Cinverters_per_kW PS The cost of inverters per kWp is taken equal to Cinverters_per_kW : 721USD=kW 7:1 7 6:1

Table 3 shows the maximum output power that a P/V system may produce according to the available surface and represents the respective cost for solar panels and inverters. In order to appraise this long-term investment, NPV (Net Present Value) will be used as an index. Net Present Value is

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]] 5

Table 3 Power and cost estimation. Total surface of solar panels (m2) 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10,000 kWp 65 130 195 260 325 390 455 520 585 650 715 780 845 910 975 1040 1105 1170 1235 1,300 Cost for solar panels (USD) 313,300 626,600 939,900 1,253,200 1,566,500 1,879,800 2,193,100 2,506,400 2,819,700 3,133,000 3,446,300 3,759,600 4,072,900 4,386,200 4,699,500 5,012,800 5,326,100 5,639,400 5,952,700 6,266,000 Cost for inverters (USD) 46,865 93,730 140,595 187,460 234,325 281,190 328,055 374,920 421,785 468,650 515,515 562,380 609,245 656,110 702,975 749,840 796,705 843,570 890,435 937,300 Total cost (cost of panels + cost of inverters) (USD) 360,165 720,330 1,080,495 1,440,660 1,800,825 2,160,990 2,521,155 2,881,320 3,241,485 3,601,650 3,961,815 4,321,980 4,682,145 5,042,310 5,402,475 5,762,640 6,122,805 6,482,970 6,843,135 7,203,300

where C is the cost of the investment; Fk the net cash ow (the amount of cash) at time k; k the time of the cash ow; N the total time of the investment; r the discount rate (the rate of return that could be earned on an investment in the nancial markets with small risk)3.5%; SVN the salvage value of the investment at the end of the total time N of the project. Salvage value of the investment is considered equal to zero. It is prudent to accept this salvage value since the photovoltaic systems have no commercial value after 2530 years of operation (Kagaraki, 2001). Furthermore, the removal and relocation cost of the photovoltaic system and associated apparatus will be subjected to enormous costs. The net cash ow each year is increased by x% (depending on the increase of the fuel price). The net cash ow is actually the prot that the solar installation can provide, because of the energy saved. This energy would otherwise be produced by conventional fuel oil. Most diesel generators installed on Vessels examined in this research use either 180 or 380 cst fuel oil (ISO 8217, 2005). All engines also burn MDO (Marine diesel Oil) but only as a startup fuel or during vessel maneuvering, which is a small percentage with respect to the main operational fuel. The prices for those mixtures per metric tn on 06/07/08 (The Complete Guide to the Bunker Market, 2008) were 790.00 and 740.00 $/tn, respectively. Furthermore the energy exerted from fuel oil according to Department for Business Enterprise and Regulatory Transport, 2008 is equal to 1 kWh 0:000082943 MT of fuel oil also 1 kg fuel oil 43:4 MJ of energy 1 MJ 0:2778 kWh 10a 10b 9

USD/barrel

dened as follows (Xatziargiriou, 2005): " # N X Fk SVN NPV C k 1 rN k 1 1 r

80 70 60 50 40 30 20 10 0 1905 -10

1932

1960

1987 Year

2014

2042

y = 0.0017x - 32.631 R2 = 0.6278

Graph 3. Oil prices in USD per barrel since 1946.

Considering from (8) that C180 cst_per_tn : 790 USDC380 cst_per_tn : 740 USD We obtain the cost for each kWh for fuel 180 cst as Cper_kWh : C180 cst_per_tn USD -Cper_kWh 0:066 tn kWh We obtain the cost for each kWh for fuel 380 cst as 12

C380 cst_per_tn USD -Cper_kWh 0:061 13 tn kWh In order to calculate the prot of the investment, it is useful to observe the pattern of the fuel oil prices. For that reason statistical data (Dow Jones & Company, 2006) for oil prices from 1946 to 2006 were obtained. Graph 3 represents all those data. It is clearly shown that the price of oil has a tendency to increase. The annual increase in oil prices from the beginning of 1990 until the beginning of 2009 is calculated to be equal to 15.44%. However no forecast can be accurate as far as the future prices are concerned. The factors that affect oil prices are innumerable and no prediction can be effective enough. For an annual increase of x%, the price of 1 kWh after k years can be given from the following equation: Cper_kWh : Cper_kWh k : Cper_kWh fuel_price_increase 1k 14

From relations (9), (10a), (10b) we may compute the current cost for production of 1 kWh from diesel generators. From relation (9) we obtain tn : 1 kWh 0:000082943 11

The net cash ow (or prot) can be calculated as the product of the energy produced every year (differs among zones and vessels) to the cost of kWh produced by conventional fuel. The prot depends on the price of kWh, the solar radiation density

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
6 A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]]

Table 4 Payback period and IRR for a variety of energy price increases for 180 cst. Price increase of Zone 1 energy (USD/kWh) (%) Payback IRR period (years) (%) 5 10 15 20 25 30 35 40 45 50 37.08 22.41 16.90 13.87 11.92 10.54 9.51 8.70 8.06 7.52 0.10 4.87 9.63 14.40 19.17 23.93 28.70 33.47 38.23 43.00 Zone 2 Payback IRR period (years) (%) 50.33 27.68 20.25 16.35 13.91 12.21 10.95 9.98 9.21 8.57 2.43 2.21 6.86 11.51 16.15 20.80 25.44 30.09 34.74 39.38 Zone 3 Payback IRR period (years) (%) 68.79 34.13 24.22 19.26 16.22 14.14 12.62 11.46 10.53 9.78 5.03 0.51 4.01 8.54 13.06 17.58 22.10 26.63 31.15 35.67 Zone 1 Payback IRR period (years) (%) 36.61 22.21 16.77 13.78 11.84 10.48 9.45 8.65 8.01 7.48 0.21 4.98 9.75 14.52 19.29 24.07 28.84 33.61 38.38 43.15 Zone 2 Payback IRR period (years) (%) 51.05 27.95 20.41 16.48 14.00 12.29 11.02 10.05 9.26 8.62 2.55 2.09 6.73 11.37 16.01 20.65 25.29 29.93 34.57 39.21 Zone 3 IRR Payback IRR period (years) (%) 62.10 31.88 22.85 18.26 15.43 13.48 12.05 10.95 10.08 9.37 4.17 0.39 4.95 9.52 14.08 18.64 23.21 27.77 32.33 36.90

Table 5 Payback period and IRR for a variety of energy price increases for 380 cst. Price increase of energy (USD/kWh) (%) Zone 1 Payback IRR period (years) (%) 5 10 15 20 25 30 35 40 45 50 37.08 22.41 16.90 13.87 11.92 10.54 9.51 8.70 8.06 7.52 0.32 4.43 9.17 13.92 18.67 23.41 28.16 32.91 37.65 42.40 Zone 2 Payback IRR period (years) (%) 50.33 27.68 20.25 16.35 13.91 12.21 10.95 9.98 9.21 8.57 2.81 1.81 6.44 11.07 15.70 20.33 24.95 29.58 34.21 38.84 Zone 3 Payback IRR period (years) (%) 68.79 34.13 24.22 19.26 16.22 14.14 12.62 11.46 10.53 9.78 5.38 0.87 3.63 8.14 12.65 17.15 21.66 26.16 30.67 35.17 Zone 1 Payback IRR period (years) (%) 36.61 22.21 16.77 13.78 11.84 10.48 9.45 8.65 8.01 7.48 0.21 4.54 9.29 14.04 18.79 23.54 28.30 33.05 37.80 42.55 Zone 2 Payback IRR period (years) (%) 51.05 27.95 20.41 16.48 14.00 12.29 11.02 10.05 9.26 8.62 2.93 1.69 6.32 10.94 15.56 20.18 24.81 29.43 34.05 38.67 Zone 3 Payback IRR period (years) (%) 62.10 31.88 22.85 18.26 15.43 13.48 12.05 10.95 10.08 9.37 4.53 0.02 4.56 9.11 13.65 18.20 22.75 27.29 31.84 36.38

(p(kWh/m2 per day)) and the total surface (S) covered by solar panels as shown Profitk; S; p : ES; pCper_kWh k 15

Finally the Net Present Value in the year k, for surface S and with solar radiation density p is given as follows: NPVk; S; p : Cinvestment
k X Profitk; S; p k1

1 rk

16

The analysis showed that, contrary to the NPV, the Internal Rate of Return (IRR) and the Discount Payback period (DPB) are independent of the P/V surface coverage and are highly dependent on the solar radiation density, ergo the zone in which the vessel is sailing. Tables 4 and 5 illustrate the DPB and IRR for a variety of fuel price increases each year (550%) in each zone. Table 4 is for fuel product 180 cst and Table 5 for fuel oil type 380 cst. Associated with the tables are Graphs 4 and 5. Solving Eq. (8), for zero NPV and salvage value, we can derive the formula of DPB as follows: DPB ln1C=ADrx=1 x ln1 x=1 r 17

where PB is the payback period; r is taken equal to 3.5%; prot(k) equals to AB(k); A the constant equal to the maximum energy capacity of each zone for a specic installed P/V area; B(k) equals to D(1+ x)k; D the constant equal to the initial cost of one kWh as per Eqs. (12) or (13); x the increase in energy cost equal to 5%, 10%, 15%, y, 50%; k the time of the cash ow. One may see that regardless of the sailing zone and also as the cost of energy increases the payback period converges to a

minimum value of about 10 years. Therefore should the fuel price increases more than 50% on an annual base the payback period will not decrease signicantly. Taking into account the fact that the energy cost increase uctuates below 20% on an annual basis, then the zones parameter inuence on the payback period is signicant. Thus should a vessel sail mostly around zones 1 and 1, then the investment of P/V systems is more protable. The installed solar panels will produce energy, which may be directly consumed for vessel needs. The vessel may reduce the fuel oil consumption via the amount of the energy that solar panels produce (depending of course on the surface that is covered by solar panels and the weather conditions). However there is a strong possibility that solar energy production, in some cases, exceeds the momentary need of energy. In that case a storage media could be introduced. The most common way to store electric energy is the use of batteries. However, batteries are expensive, require to be replaced almost every 6 years (depending on the use) and after their stultication they need special treatment in order to avoid pollution of the environment. Although renewable energy systems are more benign to the environment (Hybrid Booklet and Hybrid Power SystemsIssues & Answers, 2008) the use of electrochemical batteries is a drawback. Batteries are potentially hazardous to the environment because most contain toxic materials like lead, lithium, cadmium, and various acids. Installations must be in a well ventilated area and include methods to contain any electrolyte leakage from potential cell ruptures. These ruptures may be caused by an explosion of hydrogen gas that is generated during charging of common lead-acid batteries. Finally, any evaluation of system costs must include the cost of disposing of the batteries in an

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]] 7

environmentally responsible way (Hybrid Booklet and Hybrid Power SystemsIssues & Answers, 2008). An alternative way to store energy is by electrolysis of water, through which hydrogen is produced. After the production of hydrogen a store media like a tank is required. All the stored hydrogen can be either used in fuel cells or burned in an internal combustion engine. Either one of the two solutions is a decision of investment strategies. However modifying a diesel engine to burn hydrogen is nancially more benecial in comparison to the installation of fuel cells systems.

Because hydrogen is the lightest of all gases, it is easier to migrate through small spaces. Hence about 10% of gaseous hydrogen is lost during distribution. This amount is much less for liquid hydrogen (about 6%). The efciency of hydrogen production is calculated as follows (Bossel, 2003): Electrolysis coefcient: 65% Compression coefcient: 90% Distribution coefcient: 90% Overall efciency: 52.65%

7. Efciency of hydrogen procedures In order to produce 1 kg of H2 158 MJ of energy is required, or equivalently 43.889 kWh (Papageorgiou, 2004). The energy efciency of water electrolysis varies. The efciency is a measure of what fraction of electrical energy used is actually contained within the hydrogen. Some of the electrical energy is converted to heat, a useless by-product. Some reports quote efciencies between 50% and 70%. This efciency is based on the Lower Heating Value of Hydrogen. The Lower Heating Value of Hydrogen is total thermal energy released when hydrogen is combusted minus the latent heat of vaporization of the water. This does not represent the total amount of energy within the hydrogen; hence, the efciency is lower than a more strict denition. After the production of hydrogen gas, compression or liquefaction has to take place. The efciency of compression is about 90% and that of liquefaction about 65% (Bossel, 2003). Then the hydrogen has to be transferred to the storage media and burned in the internal combustion engine (generator).

8. Fuel cells Installation of fuel cell technology on merchant marine vessels is a very cost demanding solution. The implementation of Fuel Cells on merchant marine vessels requires the installation of expensive equipment and also hydrogen at up to 99.9999% purity. Furthermore fuel cells cannot operate on a blend of hydrogen and conventional fossil fuel, which is the essence of making a market transition from todays fossil fuel based economy to tomorrows hydrogen economy.

9. Operation of internal combustion engine on hydrogen In order to use hydrogen directly in an internal combustion engine it is necessary to use a spark plug or glow plug (Saravanan and Nagarajan, 2008). This is attributed to hydrogens selfignition temperature, which is 858 K. Hence the use of hydrogen as a sole fuel for diesel engines is impropriate. An alternative method is to combine hydrogen with another pilot fuel for ignition. This can be achieved by enrichment or induction. Hydrogen can be mixed with air prior or in the combustion chamber and result in a complete combustion. Because of the gaseous nature of hydrogen, hydrogen and air combination can be achieved easily and the resulting mixture reveals a good homogeneity. Hydrogen-enriched engines produce approximately the same brake power and higher thermal efciency compared to diesel engines over the entire range of operation. However there are some limits concerning the amount of hydrogen used in the hydrogen-enriched engines. When injecting hydrogen more than certain quantity, knocking problems appears. According to some experiments (Saravanan and Nagarajan, 2008) should the hydrogen exceed 30% in the mixture unsmooth combustion will take place. On the other hand, when increasing

80 70 Payback Period 60 50 40 30 20 10 0 0% 10% 20% 30% 40% Energy Price Increase 50% 60% Zone 1 Zone -1 Zone 2 Zone -2 Zone 3 Zone -3

Graph 4. Payback period and energy price increase correlation for 180 cst.

40 35

80 70 Payback Period 60 50 40 30 20 10 0 0% 10% 20% 30% 40% Energy Price Increase 50% 60% SEC,(MJ/kWh) Zone 1 Zone -1 Zone 2 Zone -2 Zone 3 Zone -3

30 25 20 15 10 5 0 0 10 20 30

100%DIESEL 10%hydrogen 20%hydrogen 30%hydrogen 50%hydrogen 70%hydrogen 90%hydrogen

Knock limit

60 70 40 50 BRAKELOAD,(%)

80

90

100

Graph 5. Payback period and energy price increase correlation for 380 cst.

Graph 6. Specic energy consumption for different loads. Source: [16].

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
8 A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]]

Fig. 3. Combustion chamber for hydrogen and gasoline. Source: [17].

60 55 50 45 40 35 30 25 20 15 10 5 0 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% immediate consumption
5% price increase 25% price increase 20% price increase 10% price increase 30% price increase 15% price increase 35% price increase

Graph 7. Hydrogen storage and payback period correlation for zone 1.

the percentage of hydrogen mixed with air, the specic energy consumption reaches lower values. At the highest diesel ow rate, thermal efciency is found to be the same as that of diesel engines. Hence, the overall behavior of the engine is similar to that of a diesel engine. Graph 6 depicts the variation of specic energy consumption with brake power for different levels of hydrogen enrichment. SEC decreases with an increase in hydrogen; the reduction is more prominent at higher loads. The lowest specic energy consumption is obtained for 90% hydrogen enrichment. But with 90% hydrogen enrichment the maximum load that can be carried out is 65%. For 30% hydrogen enrichment at full load, SEC is 12.9 MJ/kWh compared to diesel where it is 14.5 MJ/kWh. Therefore the fuel consumption can be reduced by 11%. Hydrogen displaces more of the combustion chamber than liquid fuel (College of the Desert, 2001), because it is a gaseous fuel. Consequently less of the combustion chamber can be occupied by air. More than 30% of the combustion chamber is displaced by hydrogen at stoichiometric conditions, compared to that of 12% for gasoline. Fig. 3 compares combustion chamber volumes and energy content for gasoline and hydrogen fueled engines.

The power output of hydrogen fueled engines, uctuates from 85% (intake manifold) to 120% (high pressure injection) compared to a gasoline/diesel engine. As already mentioned, premature ignition is a problem that is mainly related to hydrogen fueled engines rather than other IC engines. Thats because of hydrogens lower ignition energy, wider ammability range and shorter quenching distance. An effective way to eliminate or at least reduce pre-ignition is by adapting or re-designing the fuel delivery system. There are two methods of delivering fuel to a hydrogen engine (College of the Desert, 2001). One is by a carburetor and the other is by an injection system. When using a central injection system the hydrogen supply method need not be as high as for other methods. Furthermore central injection and carburetors are already used on gasoline/diesel engines, so only few modications are needed to convert a standard gasoline/diesel engine to a hydrogen fueled engine. The problem with central injection is that it is more susceptible to irregular combustion due to pre-ignition and backre. The greater amount of hydrogen/air mixture within the intake manifold compounds the effects of pre-ignition.

payback period (years)

10. Hydrogen storage potentials The amount of hydrogen that can be stored depends on the energy production and the load. When the load is low or when energy production reaches its peak, there is an excess of energy that must be stored. However the amount of energy to be stored cannot be estimated. Furthermore because of the losses during the electrolysis and storage procedures it is not very protable to store the energy. The more the energy is stored, the longer the payback period of the investment becomes. Some cases have been studied as far as storage concerns (10%, 20%, 30%, 40%, y, 90% storage potentials) with the price increase of fuel oil also being included. For all those cases the payback period has been calculated and presented in Graphs 712. In the graphs the horizontal axis (immediate consumption) indicates that x% of produced solar energy goes directly to the vessels energy demands. That means that the rest (100% x%) of the energy is used for hydrogen production.

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]] 9

75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 0% 10% 20% 30% 40% 50% 60% 70% immediate consumption 10% price increase 30% price increase 80% 90% 100%

payback period (years)

5% price increase 25% price increase 20% price increase

15% price increase 35% price increase

Graph 8. Hydrogen storage and payback period correlation for zone 2.

100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

payback period (years)

0%

10%

20%

30%

40% 50% 60% 70% immediate consumption 10% price increase 30% price increase

80%

90%

100%

5% price increase 25% price increase 20% price increase

15% price increase 35% price increase

Graph 9. Hydrogen storage and payback period correlation for zone 3.

60 55 50 45 40 35 30 25 20 15 10 5 0 0% 10% 20% 30% 40% 50% 60% 70% immediate consumption 10% price increase 30% price increase 80% 90% 100%

payback period (years)

5% price increase 25% price increase 20% price increase

15% price increase 35% price increase

Graph 10. Hydrogen storage and payback period correlation for zone 1.

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
10 A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]]

80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 0% 10% 20% 30% 40% 50% 60% 70% immediate consumption 10% price increase 30% price increase 80% 90% 100%

payback period (years)

5% price increase 25% price increase 20% price increase

15% price increase 35% price increase

Graph 11. Hydrogen storage and payback period correlation for zone 2.

90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 0% 10% 20% 30% 40% 50% 60% 70% immediate consumption 10% price increase 30% price increase 80% 90% 100%

payback period (years)

5% price increase 25% price increase 20% price increase

15% price increase 35% price increase

Graph 12. Hydrogen storage and payback period correlation for zone 3.

11. Installation of solar panels The way the solar panels could be installed is demonstrated herewith. The inclination of the solar panel surface is highly associated with the geographical latitude. As it is already mentioned, such an approach on vessels has no practical application, because there is no stable reference point. In order to maximize the solar plant efciency, all solar panels should be placed above the deck of the vessel at a zero angle. Zero angle in the solar panels will successfully capture most of the solar radiation from all directions. Fig. 4 shows how solar panels could be installed on a bulk carrier.

12. Conclusions The costbenet analysis revealed that installation of P/V systems on merchant marine vessels highly depends on the annual average increase of the fuel oil. The higher the increase, the smaller the payback period of the investment.

Fig. 4. Installation of solar panels on the deck of a bulk carrier.

The cost of the P/V investment should be included in the cost of a vessel construction in order to reduce the investment cost. Modications on existing vessels will increase the investment cost.

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019

ARTICLE IN PRESS
A. Glykas et al. / Ocean Engineering ] (]]]]) ]]]]]] 11

P/V systems efciency depends on solar radiation. Statistical data showed that zones that are close to the equator receive greater amount of solar radiation density. Hence, for merchant marine vessels sailing in those zones, a potential P/V investment is much more efcient compared to other more distant zones, since the solar energy production reaches higher values. Another interesting point is that, contrary to the annual increase rate of the fuel prices, the payback period converges to a minimum of about 10 years. A potential use of hydrogen technology as an energy storage media makes the investment less appealing. The main reason for that is the incorporated reduced coefcient rate (52.65%). Therefore, the more the solar energy used for the production and storage of hydrogen, the less protable the investment becomes. Photovoltaic systems are a highly appreciated renewable energy source. Unfortunately the technology for such systems is still very expensive with complicated implementations. References
Bossel, U.L.F., 2003. Efciency of Hydrogen Fuel Cells, DieselSOFC-Hybrid and Battery Electric Vehicles, European Fuel Cell Forum. Morgenacherstrasse 2F, CH-5452 Oberrohrdorf. Clarkson Research Studies, 2008. Clarkson Research Studies (various issues) Shipping Intelligence Weekly. College of the Desert, 2001. Hydrogen Use in Internal Combustion Engines, Module 3.

Department for Business Enterprise and Regulatory Transport, 2008. Available from /http://www.berr.gov.uk/energy/statistics/source/index.htmlS. Dow Jones & Company, 2006. Spot Oil Price: West Texas Intermediate, Wall Street Journal, Date range 1946-01-01 to 2008-06-01. Hybrid Booklet, Hybrid Power SystemsIssues & Answers, 2008. Photovoltaic Systems Research & Development, Sandia National Laboratories. Available from /http://photovoltaics.sandia.gov/docs/Hybook.htmlS. ISO 8217, 2005. Fuel quality standard for RESIDUAL marine fuels. Available from /http://www.bunkerworld.com/technical/iso8217_res.htmS. Kagaraki, K., 2001. Photovoltaic Technology. Symmetria Publications, ISBN: 978-960-266-183-3, pp. 52129. Kapros, P., 2006. Energy Economics. National Technical University of Athens. Papageorgiou, C., 2004. Solar turbine power stations with oating solar chimneys. In: IASTED Proceedings of Power and Energy Systems, EuroPES Conference, Rhodes, Greece, July 2004, pp. 151158. Papalambopoulos Michail, Glykas Alexandros, 2008, Renewable energy implementation on merchant marine vessels. Wind turbines and photovoltaic systems. Unpublished Dissertation No. 333.79, University of the Aegean, Department of Shipping Trade and Transport. Robert Scott, 1993. Standard Ship Designs (Bulk Carriers and Tanker). Fairplay Publications Ltd, ISBN: 0-905045-72-6. Saravanan, N., Nagarajan, G., 2008. An experimental investigation of hydrogenenriched air induction in a diesel engine system. International Journal of Hydrogen Energy 33 (6), 17691775. Stackhouse, P., Whitlock, C., 2008. Surface Meteorology and Solar Energy: A Renewable Energy Resource Web Site (release 6.0) sponsored by NASAs Earth Science Enterprise Program. Available from /http://eosweb.larc.nasa.gov/sse/S. The Complete Guide to the Bunker Market, 2008. Available from /http://www. bunkerworld.com/markets/prices/S. World Solar Energy News, 2008. Solar Electricity Prices. Available from /http:// www.solarbuzz.comS. Xatziargiriou, N., 2005. Investment Evaluation of Aeolian Parks. National Technical University of Athens.

Please cite this article as: Glykas, A., et al., Application and costbenet analysis of solar hybrid power installation on merchant marine vessels. Ocean Engineering (2010), doi:10.1016/j.oceaneng.2010.01.019