Sizing the power producing units

In the case of Crete it was decided to exploit the biomass potential that is available in the island which especially comes from agricultural wastes. The annual energy amount of these wastes is 7,532Tcal/year (14.3TWh/year) [1]. In the hypothetical scenario the existing diesel generators were modified to biomass plants where electricity and heat where produced at the same time with an efficiency of 80%. The heating needs where estimated to be at 2.7-3 TWh/year. From the bibliography [2] a house in Crete needs annualy 1,100 hours of heating while every hour 14Mcal should be supplied to heat a house. The existing diesel generator plants are close enough to the two biggest cities of Crete (Chania and Heraklio) so the heating that is produced can be delivered quite efficient to those cities. In order to cover the heating demands in the other five big towns (Lasithi, Agios Nikolas, Sitia, Rethimno and Malia) three biomass plants are developed as it is shown in the picture below (square dot orange lines). The purpose also for the development of these biomass plants is to cover the energy demands for the next 20 years with a 4% annual increase of energy demand.

In order to cover the electricity peak demands and the transportation needs 15 wind farms are developed especially in the east and centre parts of the island. At about 115∙10⁶ Nm³/year of hydrogen were needed to cover the transportation needs for 2003.

The annual gasoline fuel consumption in Crete is estimated to be 125,000 tn/year and the average fuel consumption is 0.09lt/km/car. The mean gasoline density is 0.73 gr/ml.

Annual Gasoline Volume = 125,000,000kg / 0.73kg/lt =170,000,000 lt/year

Annual Kilometres driven 170,000,000 lt/year / 0.09 lt/km/car = 1.888∙10⁹ Km/year

The average consumption of a fuel cell car is 0.06018 Nm³/km [6].

Annual Hydrogen Consumption = 115∙10⁶ Nm³/year

Taking into account the aforementioned increase in the demand, the demand at the end of the life span of our installation would be:

Demand = [115∙10⁶ Nm³/year]∙[1.04²⁰] = 197,118,761 Nm³

Taking into account that for our electolyser 4kWh are needed to produce 1Nm³ of hydrogen, we can work out the number of kWh needed in our electrolyser.

Energy needs for the Electrolyser (E) = Demand∙4 KWh/Nm³ = 197,118,761 Nm³∙4KWh/Nm³ = 788,475,044 kWh

Using the RETScreen software for Crete, we can see that the delivered output from 250 1.3 MW rated Nordex wind turbines is 1,001,076 MWh. Calculating the losses from the rectifiers the total amount of energy provided to the electrolyser is:

(E) = 1,001,076 · 0.95 = 951,022 MWh = 951,022,000 kWh

This number is greater than our actual needs but we have to take into account the random nature of the wind to slightly oversize our system as well as the increase in energy needs that might be higher than 4% each year. Doing that, we are on the safe side of our calculations. Hence, wind farms consist of 250 Nordex wind turbines with a nominal power of 1.3MW. Additionally power will be supplied to the electrolysers from the biomass plants, when power from wind farms is not enought for the hydrogen production. In the hypothetical scenario all transportation medias use hydrogen as a fuel which is produced mainly from the wind farms. In the proposed sites the wind potential is quite high with mean annual speeds more than 7.5m/sec (Centre of Renewable Energy Sources). In most of these sites Wind Farms are already been developed or it is proposed to be developed in the near future. For our estimates though, we used a 7 m/sec wind speed which we thought is a good figure for the sum of our sites and would also lead us to conclusions on the safe side of the calculations. The topology of the sites is also considered to be the proper for the development of such plants. Considering the geomorphology of the sites it was decided to develop 15 wind farms which was the minimum number of wind farms that could be build 250 wind turbines. On the other hand that number is large enough to guarantee security of supply (not much production lost in case that there is no wind on a specific sites). The lines of the high and medium voltage level are near that sites so the connection to the main electricity grid is easy, quick and costs less. It was also considered to have 8 new transformers near the wind farms in order to transform the voltage into the desirable level.

The electrolysers that produce the hydrogen that is used to transportation are being build near the wind farms and in that case we have less energy and fund loses. The nominal capacity of the electrolysers is 75MW and it is distributed to 8 plants. The size of the electrolysers was estimated in order to cover the transportation demands for the next 20 years with an annual increase of 4%. Also in the proposed sites it was considered the existence of suitable water that the electrolysers can use in order to produce hydrogen. In addition a pipe line system of 500km length, is being build where the hydrogen can be transferred to main hydrogen stations where storage tanks, of 200 tonnes total capacity, are available and from there smaller hydrogen stations are supplied by lorries.

A demonstration of how the numbers for the storage tanks were calculated (methodology) can be found in the case of Karpathos - Kasos bellow.

	The proposed electricity system in Crete
(Click image to enlarge)

 

Sizing the power producing units

The annual electricity demand at the Karpathos-Kasos complex is 24,369 MWh/year. The highest recorded demand peak was 6.5 MW. In order to cover that demand we decided to use 12 Nordex N54/1000 1MW wind turbines.

Those will be split in 3 wind farms, 2 of which will be situated at the western part of Karpathos and one of those will be situated at Kasos Island. The selected spots for the wind farms have an average wind speed of 7 m/s at 10 m height and are found on the wind potential map of the islands provided by the Centre of Renewable Energy Sources in Greece. The total energy yield of the wind turbines is estimated to be 37,275MWh using RETScreen. The system is also expected to cope with the increasing energy demand of the two small islands and to provide hydrogen for transportation.

The fuel cell plant will have a nominal power of 6MW in order to cover the highest demand in low wind situations but also to be able to cope with future –increased- energy demand.

Calculating transportation fuel needs

The total number of kilometres per year run in the islands is 4,378,900 km. This was calculated assuming 1 car per 5 inhabitants in Karpathos and 1 car per 7 inhabitants in Kasos. The annual mileages were taken 3,500 km and 1,500 km respectively. Those numbers were taken from the Statistical Service together with the help of inhabitants of those areas. The average consumption of a fuel cell car worked out in Nm³/km as taken from reference [6] is found to be 0.06018 Nm^3/km. So, the number of Nm³ needed for transportation can be worked out as follows:

V = 4,378,900∙0.06018 = 263522.2 Nm³/yr

 Assuming an electrolyser that uses 4 kWh to produce 1 Nm³ of hydrogen we can calculated the amount of energy per year (in kWh) needed to provide us with enough hydrogen to cover all transportation needs. Hence in our case it is:

E = V∙4 = 1,054 MWh.

Electrolyser – Hydrogen Storage Tank

The size of our storage tank is calculated so that there is enough reserve to cover the islands’ needs both in electricity and transportation for 5 days of no wind situation in the worst period (summer period). In the summer period the daily demand is around 40% higher than in the winter. The size of the tank is hence calculated to be 31,750 kg. Calculations were made as followed:

On a typical summer day the energy demand is:

E = 94,278 kWh

Hence for a five day period the demand is:

E = 94,278 ∙ 5 = 471,390 kWh

This means that the fuel cell power plant would have to be provided with hydrogen of:

LHV = E/η = 471,390/0.45 = 1,047,533 kWh

Or in MJ:

LHV = 3,771,220 MJ

Converting into Nm³ ( LHV of hydrogen is 10.8 MJ/Nm³⁾ we get:

V_e = LHV/10.8 = 349178 Nm³

The number of total Nm³ per year for transportation was calculated above. For the transportation needs of five days then we get:

V_t = (V/365) ∙ 5 = 3,610 Nm³

Hence the total volume needed is:

V_total = 349,178 + 3,610 = 352,788 Nm3

Since 1Nm³ of hydrogen weights 0.09 kgs the total weight of the hydrogen needed to be stored is:

W = V_total ∙ 0.09 = 31,750.9 kg

For construction purposes a 37,500 kg storage tank was proposed

An electrolyser of 2 MW nominal power is assumed for our case study. Our excess energy –had it been uniformly distributed- could be converted in hydrogen using a 700 kW electrolyser. The energy delivered though, is very far from being uniform and without detailed demand and weather data and simulation methods a detailed sizing cannot be made. We assumed though that a 2 MW electrolyser would be adequate for our system.

 

	The proposed electricity system in Karpathos and Kasos
(Click image to enlarge)

 

Maintenance tasks (annual and periodic)

Except for the usual annual maintenance tasks, some extra periodic maintenance tasks have to be made. For the wind farms, drive train maintenance works has to be carried out every 10 years while blades replacement is scheduled in 15 years time.

In the case of the fuel cell plant, fuel cell stack replacements have to be carried out. We estimated that the stacks should be changed every 4 years. Our calculations are based on the fact that PEM fuel cells when used in stationary applications have a life of 40,000 hours (approximately 4.6 years) and if operated continuously at ¾ of their nominal power they can reach 48,000 hours of operation (around 5.5 years). Switching on and off of a fuel cell can greatly reduce its lifetime (in fuel cell cars lifetime is around 2,000 hours). Since we will be switching our fuel cells on and off sometimes (but not causing the same fatigue they would go under if they were in a FC car) we assumed that a 4 year stack replacement would be reasonable.