2.19

Stand-Alone, Hybrid Systems

KA Kavadias, Technological Education Institute of Piraeus, Athens, Greece
© 2012 Elsevier Ltd. All rights reserved.

2.19.1
2.19.2
2.19.3
2.19.4
2.19.4.1
2.19.4.2
2.19.4.3
2.19.4.4
2.19.5
2.19.5.1
2.19.5.2
2.19.5.3
2.19.5.4
2.19.5.5
2.19.6
2.19.6.1
2.19.6.2
2.19.7
References
Further Reading

Introduction
Historical Development of Wind Stand-Alone Energy Systems
Contribution of Wind in Stand-Alone Energy Systems

System Configuration
Wind Turbine Generator
Storage System Unit
Complementary Electric Generator Unit
Auxiliary Electronic Equipment
Stand-Alone Hybrid Systems Configurations
Stand-Alone Wind Power Systems
Stand-Alone Wind–Diesel Power Systems
Stand-Alone Wind–Photovoltaic Power Systems
Stand-Alone Wind–Hydro Power Systems
Stand-Alone Wind–Hydrogen Power Systems
Energy Storage in Wind Stand-Alone Energy Systems
Design Parameters of Energy Storage Systems
Short Description of Energy Storage Technologies
Design, Simulation, and Evaluation Software Tools for Wind-Based Hybrid Energy Systems

Glossary
Hybrid power system A power system which uses
multiple generation sources by incorporating different
components such as generators, storage medium, power
conditioning and system control in order to supply power
to a remote consumer.
Loss of load hours (LOLH) Power reliability factor
indicating the number of load failures in which the load
demand exceeds the power supply on hourly based
simulations.

623

624


626

629

631

632

632

632

632

633

636

638

643

645

647

649

649


651

653

655


Loss of load probability (LOLP) Power reliability factor
indicating the probability that instantaneous power
demand will exceed the respective power supply for the
time period analyzed.
Loss of power supply probability (LOPSP) Power
reliability factor indicating the probability of insufficient
power supply for a given period of time.
Stand alone energy system An electricity system which
operates independently of the electricity transmission and
distribution network or is not connected to it at all.

2.19.1 Introduction
Energy is indispensable for sustainable development and poverty reduction. At present, there are 1.6 billion people in the world,
mostly in rural areas, who have no access to electricity. Another 2.5 billion people still rely on traditional fuels such as wood, dung, and
agricultural residues to meet their daily heating and cooking needs, this, however, having serious impacts on the local environment
and on people’s health [1, 2]. Apart from the Third World and many of the developing countries that face serious problems of
insufficient electrical network infrastructure, isolated electricity consumers who lack direct access to electrical networks and have
limited political influence may be encountered in many regions of the developed countries as well. In this context, stand-alone wind
power systems, which have already been in use for hundreds of years, have proven to be a reliable and environmentally friendly
technological solution for the electrification of remote consumers in areas with moderate or high wind potential.
Stand-alone electrical energy systems constitute the first applications of the implementation of renewable energy sources (RES). The
first attempts at generating electricity from wind energy were directed toward providing energy independently in remote areas where there

was no connection to the grid [3]. By using a small wind turbine of only a few hundreds of watts and a storage medium, it was possible to
cover the modest needs of electrical energy (in most cases only lighting). Since then, the term ‘stand-alone’ has defined an electricity system
that operates independently of the electricity transmission and distribution network or is not connected to it at all. The aim of such systems
is to meet load demand in a direct way, keeping power generation and consumption as close as possible. In recent years, the term ‘stand­
alone’ has mainly been used to describe power systems of up to tens of kilowatts, which mainly refer to domestic wind systems [4].
Further, hybrid power systems suggest a concept the roots of which can be traced back many years, describing a power system
that uses multiple generation sources by incorporating different components such as generators, storage medium, power

Comprehensive Renewable Energy, Volume 2

doi:10.1016/B978-0-08-087872-0.00222-5

623


624

Stand-Alone, Hybrid Systems

conditioning, and system control in order to supply power to remote communities [5]. Wind hybrid power systems usually
combine two or more forms of energy, resulting in a more efficient system overall. In the basic form of wind hybrid systems, a
wind turbine is combined with a small diesel engine (hybrid stand-alone system), or is connected to the local diesel power station as
in the case of isolated power stations (autonomous hybrid system) such as in remote islands. Contemporary small-scale wind
hybrid energy systems are stand-alone systems which usually comprise a wind turbine, and a photovoltaic (PV) generator and the
corresponding energy storage system. In a stand-alone installation that consists only of a wind power energy system, on a short-term
basis the wind turbine energy production should meet the power demand of the consumer. Therefore, at any given moment there
must be a balance between the energy production of the wind turbine and the respective load demand. In order to attain power
balance, either the wind turbine should be controlled accordingly or the power consumption should be adapted to the output of the
wind turbine. Hybrid energy systems are used in order to tackle the fundamental technical problems that arise from the dependence
of the remote consumer on the stochastic energy yield of a wind turbine [6].

Stand-alone hybrid energy systems are being used in a wide range of applications worldwide. As already made clear, such applications
normally concern remote consumers that either do not have the choice of grid connection – as in isolated small islands – or live in entirely
remote locations, far away from the nearest electrical grid, resulting in a grid connection cost that is extremely high. In the aforementioned
cases, installation of a properly sized wind hybrid energy system may sufficiently fulfill the energy demand. The most common applications
of stand-alone hybrid energy systems include winter or summer shelters, isolated farms, grid-isolated communities, telecommunication
stations, small desalination systems, water pumping installations, electrification of lighthouses, and even far-off road lighting.
As already mentioned, the most common application of hybrid stand-alone power systems concerns satisfaction of the domestic
electrical needs of consumers in remote locations where the available wind potential – either alone or combined with the local solar
potential – is exploited. Such installations are always supported by energy storage systems, which are essential to ensure energy
autonomy. Another significant parameter that should be taken into consideration is the load demand profile, which in combination
with the renewable energy potential of the area constitutes the essential input data for the sizing and cost estimation of wind-based
stand-alone systems. In any case, energy security in stand-alone systems is accomplished with the inclusion of a conventional power
generator (i.e., diesel generator) which, besides guaranteeing 100% energy demand satisfaction, could also contribute to lowering
the size of the renewable energy devices and energy storage system components.
Remote islands can also be considered as application areas of stand-alone hybrid systems, because in most cases their weak
microgrid operates on expensive fuel and therefore the exploitation of renewable energy sources is considered as vital for energy
cost reduction. In such isolated communities, however, certain problems need to be considered, mainly deriving from the
minimum permissible contribution of renewable energy sources in the local microgrid [7]. In order to fulfill such constraints,
serious research efforts have been recorded during the last 15 years – and are still ongoing – achieving even 90% RES contribution in
small island grids by implementation of RES-based hybrid energy solutions.
Regarding the applications of wind-based hybrid power systems in the telecommunication sector, RES have been identified as an
energy solution for minimizing the operating expenses [8]. Windy areas such as coastal locations and hills, where, in many cases, the
telecommunication masts are installed, are ideal for wind stand-alone systems that are capable of minimizing the fuel consumption
of diesel generators and thereby the operational cost of the installations. Furthermore, RES exploitation can be achieved with the
installation of a suitable storage system, which will absorb residual renewable energy and return it back for consumption when
lower renewable energy production is encountered. Most telecommunication transmitters require air-conditioning services during
the summer periods, which makes the combination of wind energy generation with photovoltaic systems’ energy ideal, because
during summer the availability of solar energy is quite high.
In combination with covering the electrification needs of remote or isolated consumers, stand-alone renewable energy systems can also
contribute to the satisfaction of potable water needs through small desalination systems [9]. Potable water shortage is often encountered in

remote island regions, where, in cases where wind conditions in the area are favorable, wind-based stand-alone systems can be used in
water desalination plants for the production of potable water. The main constraints on the use of wind energy in such systems is the
nonsteady power supply, which forces the desalination plants to operate in suboptimal conditions. In order to overcome this undesirable
way of operation, considerable energy storage capacity is necessary to support the installation. Furthermore, the contribution of
photovoltaic energy could be essential in many cases for realizing uninterruptible power supply to the desalination plant [10].
Direct utilization of the mechanical energy produced by the wind turbine (shaft power), without converting it to electricity, is possible
in water pumping systems [11]. Wind-powered water pumps operate either in a direct manner, that is, directly attached to the turbine’s
shaft, or through electricity generated by a typical small wind turbine. Wind energy has been in use for centuries for water pumping, and
even nowadays there is a large number of installations worldwide. In cases where a high starting torque is necessary, a large number of
blades are used similar to the older wind turbines. In cases where the operation of a mechanical drive is used for pumping water, the
placement of the wind turbine is restricted to be near the water reservoir, whereas in cases where wind electricity is supplied to the water
pump, the wind turbine may be placed far away from the water reservoir for maximizing wind energy exploitation [12]. Potential
applications of wind water pumping systems include domestic water supply, community water supply, cattle watering, and irrigation [13].

2.19.2 Historical Development of Wind Stand-Alone Energy Systems
Wind energy has been exploited for grinding grain or pumping water for at least 3000 years [6]. The use of wind turbines for
electricity generation can be traced back to the nineteenth century when an experimental wind turbine (Figure 1)[14] driving a


Stand-Alone, Hybrid Systems

625

Figure 1 The first electricity-producing wind turbine, installed in 1891 [14].

dynamo was built by Poul La Cour in 1891 in Denmark [3]. The remarkable fact is that La Cour at once tackled the problem of
energy storage. He used the direct current (DC) generated by his wind turbine for electrolysis and stored the hydrogen gas that was
produced, establishing in this way the first wind stand-alone energy system.
Based on La Cour’s wind turbine model, by 1908 the Lykkegard company had built 72 electricity-generating wind turbines with
power output ranging between 10 and 35 kW, which were used to supply energy to rural settlements. For much of the twentieth

century, wind turbines were being used to charge storage batteries which then were used to operate small appliances [15]. The
interest in electricity generation by means of wind power during the wind turbine evolution had seen some fluctuations following
the diesel fuel cost fluctuations. In periods when the fuel prices were rising, such as during World War I and World War II, the
interest in wind power was growing. Back then, the subject of environmental protection had not yet arisen and thus there was no
association with energy production. Despite the reduced interest in supplying wind energy to electricity networks, wind turbine
manufacturers continued their efforts in building wind turbines for stand-alone applications. In 1922 in the United States,
Marcellus and Joseph Jacobs developed small wind turbines which became known as ‘wind chargers’ which were used for recharging
batteries for power supply of rural settlements and remote houses. In Germany until the 1930s, a total of 3600 American wind
turbines were built under license, and most of them were used for pumping water but some of them were modified for electricity
generation. The first wind turbine feeding a local grid was installed in 1931 in the USSR in Balaklava. The electricity generated was
fed into a small grid which was supplied by a 20 MW steam power station. The idea of using wind turbine electricity for supplying
a grid network was also supported by Hermann Honneff in 1932, whose vision was a five-rotor wind turbine of 20 MW that
was to generate electricity in combination with conventional power plants [16]. Accordingly, in the United States in 1941 the
world’s first large wind turbine was installed in Vermont [17, 18].
Until the ‘oil price shock’ of 1973, the extremely low prices of conventional fuels held back the development of the wind energy
sector, as the investors were not highly motivated to invest new funds in order to overcome the numerous technical problems and faults
that had been encountered during the more practical operation of large wind turbines. Therefore, all these years stand-alone wind power
installations continued to be the main application of wind power. After the energy crisis in 1973, the interest in wind turbine technology
was rekindled giving a significant boost to the wind technology evolution. Based on the traditional models of three-bladed rotors and
grid-connected induction generators that dominated large-scale wind turbine models, Danish companies, which were active in
agricultural machinery, began building small turbines to sell them to private owners or agricultural holdings (Figure 2). These small

Figure 2 Remote installation of a small wind turbine at a private electricity user’s holding in Denmark (1985). Photo by Rüth found in Reference [3].


626

Stand-Alone, Hybrid Systems

Small wind turbine manufacturers (2009)


18 countries
with 4 or less
companies;
31

Sweden; 5
Spain; 5
Netherlands; 6

USA; 95
Germany; 16

China; 19

UK; 22

Canada; 24

Japan; 29

Figure 3 Global distribution of small wind turbine manufacturers (number of companies per country) [20].

wind turbines were also used by consumer associations to cover the electricity needs of small communities. The installation of small
wind turbines was supported by the Danish government through financing and legal regulations, and at the same time the pattern of
rural Danish settlements with its many single farms generally favored the decentralized installation of wind turbines. By 2001, about
150 000 Danish households were registered as owners of shares in wind turbines [19].
According to the most recent data available, there is a growing interest in the small wind turbine (those of rated capacity
< 100 kW) industry. In the United States, the small wind turbine market grew by 15% in 2009 pushing their total installed capacity
to 100 MW. It is worth mentioning that almost 50% of the small wind turbines were installed during the last 3 years of the industry’s

80 years of history. In this context, the global sales of small wind turbines for the year 2009 were 15 500 units for off-grid
connections with a total power of 7600 kW. Another 5200 units were sold for on-grid connection with a total capacity of
34 400 kW. Regarding the different size range of small wind turbines and their use, 100% of the wind turbines with rated power
up to 1 kW, 10% of those with rated power 1–10 kW, and < 1% of those with rated power 11–100 kW are used for off-grid
applications. According to Figure 3 [20], about 40% of global small wind turbine manufacturers are located in the United States,
25% in Europe, 10% in Asia, and 25% in the rest of the world, indicating the worldwide interest in small wind turbine installations.

2.19.3 Contribution of Wind in Stand-Alone Energy Systems
Stand-alone power systems are mainly used in cases where there is no grid electricity available or the cost of connection to the local
electricity grid is prohibitive. Given that the minimum grid extension cost for low-voltage lines exceeds 10 000 € km−1 of grid line – a
value which increases in cases of difficult access situations [21–23] – and that the already high cost of fuels increases even more with
the remoteness of the location [24], remote consumers should try to exploit all alternative choices that are available in their area. In
such cases, renewable energy sources can provide the necessary electricity and thermal energy at a cost competitive to the
corresponding electricity cost of the local network.
In the case of grid-connected wind parks, the area in which the park will be installed is selected according to the available wind
speed values. In these cases, measurements are taken for quite a long period of time (i.e., at least 12 months) in order to evaluate the
wind potential of the area. On the other hand, in the case of wind stand-alone systems, the area of installation is already given, and
the owners are not willing to make time-consuming wind speed measurements for the estimation of the wind potential. Therefore, the
decision of whether to install a wind stand-alone system is taken in accordance with physical indications such as bending of trees and
existence of old windmills in the area, as well as on the basis of the local habitants’ experience concerning wind patterns in the area.
In this context, the annual energy yield of a wind turbine depends on the operational characteristics of the wind turbine and the
available wind potential in the area. The wind potential of an area could be described to a good approximation by the Weibull
function given in eqn [1].
f ðuw Þ ¼

k � uw �k − 1 −
e
C C

� u �k

w

C

½1Š


Stand-Alone, Hybrid Systems

627

where

f(uw) is the Weibull distribution function;

k is the shape factor;

C is the scale factor; and

uw is the wind speed

The quality of the wind potential of the area depends, according to the Weibull distribution, on the scaling factor, C, which is

proportional to the mean annual wind speed and the shape factor, k, which depends on how widely wind speeds are spread around

the mean wind speed value (inversely proportional to standard deviation).

Accordingly, with respect to the wind frequency distribution, the annual power production of a typical wind turbine for different
wind speeds can be calculated as
1

Pw ¼ ρAuw 3 ηwt Cp
2

½2Š

where
ρ is the air density (kg m−3);
A is the swept area of the rotor (m2);
ηwt is the total electromechanical efficiency of the electrical and mechanical components of the wind turbine; and
CP is the power coefficient of the wind turbine’s rotor.
For a wind turbine with a typical (simplified) wind power curve (Figure 4), the energy yield for different wind potential areas is presented
in Figure 5. According to Figure 5, the energy production of the wind turbine strongly depends more on the scale factor, C, which is
related to the mean annual wind speed and less on the shape factor, k, which is related to the standard deviation of the wind speeds.
Further, if one considers a wind turbine that operates according to a typical power curve, the estimated energy performance per
kilowatt of nominal wind turbine power is presented in Figure 6. According to Figure 6, the energy production potential of a small
wind turbine can reach up to 4200 kWh kW−1 for areas with a mean annual wind speed of 7 m s−1.
As already mentioned before, the power curve of the wind turbine is of great importance, as it describes the expected power
generation for each wind speed at the hub height. Therefore, the annual energy yield of a commercial wind turbine is expected to be
different from the one estimated using the typical power curve. Figure 7 presents the percentage of annual energy yield per kilowatt
of rated power of several commercial small wind turbines of different sizes relative to the typical power curve turbine. According to
Figure 7, the expected deviation between wind turbines with different power curves can be as high as 25%. The data were collected
from the official websites of the manufacturers [25–27] and the wind turbine database of Soft Energy Applications and
Environmental Protection (SEA&ENVIPRO) Laboratory of TEI of Piraeus [28].
The energy production of the previous figures presented refers to the capability of the wind turbine to produce energy in areas
with specific wind potentials. In stand-alone systems, the wind power generation does not always match the power demand of the
consumer. Therefore, not all wind energy generated will be absorbed by the demand. The maximum wind energy absorption
depends on the wind turbine energy production based on the wind potential of the area and the wind turbine selected, as well as on
the corresponding power demand in accordance to the load profile of the consumer. Thus, in order to estimate the wind energy
absorption rate of a stand-alone power system the parameters that should be considered are


Non-dimensionalized wind turbine power curve
1.2
1.0

P/P0

0.8
0.6
0.4
0.2
0.0
0

5

Figure 4 Typical nondimensional wind turbine power curve.

10
15
Wind speed (m s−1)

20

25


628

Stand-Alone, Hybrid Systems


Annual energy yield of 5 kW wind turbine
in different wind potential areas
25 000

Energy yield (kWh)

20 000

k = 1.0
k = 1.5
k = 2.0
k = 2.5

15 000

10 000

5 000

3.0

3.5

4.0

4.5

5.0

5.5


6.0

6.5

7.0

7.5

8.0

Scaling factor "C"
Figure 5 Estimated annual energy yield of a 5 kW wind turbine based on typical power curve, installed in different wind potential areas.

Energy performance of typical wind turbine
Annual energy perfomance (kWh kW–1)

6 000
5 000
4 000
3 000
2 000
1 000

1

2

3
4

5
6
Mean annual wind speed (m s–1)

7

8

Figure 6 Mean annual expected energy production of a typical wind turbine at different wind potential areas (shape factor k = 2.0).

•
•
•
•

the wind potential of the area,
the type and the size of the wind turbine selected,
the energy demand profile of the consumer, and
the storage system size.

Kaldellis and Vlachos [29] presented a detailed case study indicating the influence of the above-mentioned parameters. Based on a
rural household load profile of a remote consumer [30] with an annual energy consumption of 4750 kWh, they estimated different
optimum wind stand-alone configurations in respect of zero load rejections, for an area with a mean annual wind speed of 5.6 m s−1
(Kea island in Greece).
According to their results, presented in Figure 8, one can clearly see that the wind energy utilized by the consumption is between
35% and 50% of the total wind energy produced during 1 year of operation.
A wind turbine in combination with a PV power station and with the appropriate storage capacity can satisfy the energy demand
of stand-alone power systems. The addition of a PV power station to a wind stand-alone system can significantly increase the
system’s reliability and its ability to cover the electricity needs of a remote consumer setting the diesel fuel generator as an emergency
backup unit [31].

In Figure 9, the possible combinations of wind turbine, photovoltaic station, and storage medium size are presented [31]. Each
proposed configuration is selected based on the zero-load-rejection condition and is capable of covering the annual needs
(4750 kWh) of a remote consumer on an hourly-basis calculation. According to Figure 9, the storage capacity strongly depends


Annual energy yield related to typical wind turbine

Stand-Alone, Hybrid Systems

629

Commercial small wind turbines' performance
related to typical one
140%
120%
100%
80%
60%
40%
20%
0%
Typical
power
curve
(5 kW)

Sudwind Aeolos-H AOC DANmark Aeolos-H Aeroman Aerostar HSW 30 UGE-4k
N 1245 (1 kW)
15/50
11

(5 kW) 14.8–33
18
(30 kW) (4 kW)
(45 kW)
(50 kW) (20 kW)
(33 kW) (36 kW)
Commercial small wind turbines

Figure 7 Small commercial wind turbines’ annual energy performance relative to the wind turbine with a typical power curve at an area of mean annual
wind speed of 5 m s−1 and shape factor k = 2.0.

KEA hybrid system-wind energy distribution
60%

No diesel oil
100 kg y–1 diesel oil

50%

100 kg y–1 diesel oil

40%

30%

20%

10%

0%


Rejected energy

System loss

System consumption

Figure 8 Application results of different wind stand-alone system configurations in the island of Kea in Greece capable of covering the energy demand of
a rural consumer.

on the photovoltaic rating and wind turbine size selected. It should be noted that the figure refers to an area with a mean annual
wind speed of 5.5 m s−1 and an annual solar potential of 1650 kWh m−2.

2.19.4 System Configuration
Stand-alone hybrid energy systems have emerged as one of the most promising ways to handle the electrification requirements of
numerous isolated consumers worldwide, including houses in the country, remote farms, shelters, telecommunication stations, small
islands, lighthouses, and so on. A typical hybrid energy system combines two or more electricity generation units, which in some cases,
where high RES quality exists, can be based purely on RES, along with the appropriate energy storage system and the corresponding
auxiliary electronic devices. Typical stand-alone hybrid systems also utilize a small thermal power unit, which is used as a backup
power system in cases when the RES power generator units along with the storage system cannot fulfill the energy demand.


630

Stand-Alone, Hybrid Systems

Energy autonomy configuration of wind-photovoltaic
stand-alone system
25 000
Wind turbine's rated power

No wind turbine
1 kW
5 kW
7.5 kW
10 kW
15 kW

Storage capacity (Ah)

20 000

15 000

10 000

5 000

1 000

2 000

3 000

4 000

5 000

6 000

7 000


8 000

9 000 10 000

Photovoltaic station rated power (kW)
Figure 9 Energy autonomous wind–photovoltaic stand-alone system configurations in the island of Kea in Greece capable of covering the energy
demand of a rural consumer.

UPS

AC current flow
DC current flow

Wind turbine

Data monitoring

AC/DC
rectifier

In case of AC wind
turbine

Wind turbine’s
charge controller

DC
loads


DC
switchboard
PV array

PV charge
controller

Inverter
AC
loads
AC
switchboard

Diesel
generator
Battery
bank

Overall
management
control

Figure 10 Typical hybrid RES-based stand-alone system [5, 32].

Figure 10 [5, 32] presents the most common configuration of a small-scale RES hybrid stand-alone energy system which
comprises a wind energy converter, a photovoltaic power station, a diesel generator used as a backup power provider, a battery
bank unit for storing residual energy, and auxiliary electronic equipment which includes charge controllers, an inverter and the
corresponding switchboards for alternative current (AC) and DC loads. Contemporary stand-alone systems also include an
overall system management unit capable of controling the power flow according to the instantaneous power generation and
power demand.

Several wind-based hybrid energy configurations can be found in the international literature, which incorporate different
combinations of renewable energy power production units, conventional electricity generation systems, and storage system


Stand-Alone, Hybrid Systems

631

configurations. The most well-known systems, apart from the one presented in Figure 10 are the following (the corresponding
indicative references are given in brackets):
•
•
•
•

wind–diesel systems [33–35]
wind–hydro installations [36–39]
wind–biomass-based installations [40, 41]
wind–hydrogen/fuel cell hybrid energy systems [42, 43].

Besides, different system configurations exist with regard to the energy storage medium used in each of the above combinations.

2.19.4.1

Wind Turbine Generator

The wind turbines used in stand-alone systems are often in the range of up to 50 kW [4], as the term ‘stand-alone system’ usually
indicates small electricity systems up to the scale of a small community. The amount of power a turbine will produce depends
primarily on the diameter of its rotor, as it is the rotor diameter that determines the quantity of wind intercepted by the turbine.
Small wind turbines in comparison to large wind turbines operate at a higher rotational speed for the same wind speed; have a tail

for the orientation of the nacelle; have significantly smaller tower heights and therefore experience lower average wind speeds;
and comprise simpler and cheaper safety systems to withstand high wind speeds. As far as the overspeed protection of small wind
turbines is concerned, the most common safety mechanism is the turbine pitch-up or tilt-up and furling. Pitching is more common
on very small wind turbines.
Because wind speeds increase with height, the turbines are mounted on a tower. In general, the higher the tower, the more power
the wind system can produce. The tower also keeps the turbine above the air turbulence that may exist close to the ground because of
obstructions such as hills, buildings, and trees. Note that relatively small investments in increasing the tower height can yield high
rates of return in terms of increased power production.
The generator of a small wind turbine is one of the most important parts of the structure and strongly influences the energy
performance and the reliability of the wind turbine. Most small wind turbines use permanent-magnet generators, which do not
require external excitation. They are simple to use, as they need only a rectifier to produce the DC voltage required for a battery, but,
on the other hand, the magnets are easily broken and many are sensitive to temperature changes. Similar to permanent-magnet
generators are synchronous generators, which need a field current charge to produce the magnetic field, thus reducing their
efficiency. The permanent-magnet generators give satisfactory performance if they are connected through a rectifier to the batteries,
whereas in case they are connected directly to an AC load with a constant frequency, the speed of the wind turbine should be
constant. The electrical output from the generator is usually three-phase AC with a variable voltage and frequency. The correspond­
ing current is converted to DC using a rectifier and then to a fixed voltage and frequency as required in ordinary
electricity-consuming appliances.
Regarding the wind turbine installation, there are two basic types of towers: self-supported (free standing) and guyed. Most
home wind power systems use a guyed tower. Guyed towers, which are the least expensive, can consist of lattice sections, pipes, or
tubing depending on the design, and are supported by guy wires (see Figure 11). They are easier to install than self-supported
towers; however, because the guy radius must be between one-half and three-quarters of the tower height, guyed towers require
considerable space to install them. Tilt-down towers (which can be either self-supported or guyed) are more expensive, but they
offer the consumer an easy way to perform maintenance on smaller, light-weight turbines, usually 5 kW or less. Lattice towers are
easier to transport but tend to have a lower service life than pole towers. Tubular towers require smaller foundations but are usually
heavier than the other types, thus increasing the purchase and transportation costs. In cases of wind turbines installed near the sea,
hot-dipped galvanized tubular towers should be considered. According to Wood [44], the optimum tower height for a small wind
turbine is typically 18–33 m depending on the turbine size and wind potential of the area.

Guyed

Figure 11 Basic types of wind turbine towers.

Self-supported

Lattice


632

2.19.4.2

Stand-Alone, Hybrid Systems

Storage System Unit

Owing to the stochastic behavior of wind, wind generation cannot always provide a firm capacity to an autonomous electrical power
system [45]. In addition, the implied fluctuations can – in some cases – cause problems related to stability, harmonics, or flicker. An
energy storage system, when sized appropriately, can match the highly variable wind power production to a generally variable system
demand, remarkably limiting the energy production cost (e.g., by generating capacity savings). In this context, the critical parameters
concerning the storage systems potentially used in a wind hybrid installation include lifetime expectancy, energy efficiency, depth of
discharge, and the initial and operational cost. A short description of the most common storage systems is given in Section 2.19.6.

2.19.4.3

Complementary Electric Generator Unit

The quantity of energy that a wind power generator can produce strongly depends on the available wind potential at the installation
area. Although the total annual energy production might seem enough to cover the corresponding electrical energy needs, satisfaction
of the load demand by the energy produced should be examined at least on an hour-by-hour basis. The duration of calm spells is an
important parameter that influences the decision about the choice of components and the size of a wind hybrid installation for a

stand-alone system to provide constant electricity for consumption. There could be situations where, although the calculated annual
energy produced seems enough to cover a consumer’s power needs, long calm spells could cause a load failure. In order to confront
such situations, larger storage systems, which significantly increase the initial cost of the plant, are usually considered.
An interesting option is the installation of an additional, independent electric power generator, which reinforces the electricity
production system. Several studies have shown that a wind turbine in combination with a secondary power generator, which could
also be based on renewable energy sources (e.g., photovoltaics) or could be conventional fuel-based generators (e.g., diesel or gas),
can limit the energy storage system size and in many cases reduces energy production costs.
Another quite interesting option is the combination of the energy storage system with an alternative electric power generator
unit. An example of such an installation is the combination of a wind turbine coupled with an appropriate hydrogen production
system based on electrolysis, to be used as energy storage, and a fuel cell unit that uses the hydrogen produced and stored during
low energy demand to produce electric power during low or very high wind speeds. An additional advantage of the specific
installation is the opportunity to use the hydrogen produced as a fuel in appropriate devices (taking advantage of the heat produced
by its combustion) or even as a fuel for hydrogen cars. Of course, the currently low energy efficiency of the cycle, that is, from
hydrogen production to the final power production by the fuel cell, should also be taken into consideration [43].
Another option is the combination of a wind electric power generator with a small pump-hydro unit in which water is pumped
from a lower water reservoir to a higher water reservoir during low energy demand situations, and returned through the hydro
turbine to the lower reservoir during low or very high wind speeds [46]. In such installations, the water stored could also be used
to cover any water supply needs.

2.19.4.4

Auxiliary Electronic Equipment

The auxiliary electronic equipment needed to support a stand-alone wind hybrid system depends on the application type. For
example, the parts required for a wind turbine coupled with a pump-hydro storage system will be very different from those needed
for a wind–diesel hybrid system. Most manufacturers provide system packages that include all the necessary parts of the system.
Stand-alone systems, which in most cases are combined with batteries, also need a charge controller to prevent the batteries from
overcharging or overdischarging. Small wind turbines generate DC electricity. When using standard appliances that use conven­
tional AC current, an inverter to convert the DC electricity from the batteries to AC is necessary.
The controller of the system ensures that there is a current limit in order to protect the generator and also to monitor the battery

condition to avoid overcharged conditions. In addition, it could be used as a primary overspeed protection system by reducing the
blade speed. Most wind turbine controllers have a current limit so as to protect the generator by limiting the power output of the
wind turbine. In this way, overheating of the generator can be avoided, protecting it from possible insulation and wires melting.
Such control is essential as the generator in small wind turbines is usually air-cooled; therefore, the generator’s current capability
depends on air temperature, wind speed, and the thermal resistance from the wires to the air but also the heat loss of the generator.
The inverter is an essential part of the auxiliary electronic equipment, as it produces the correct voltage and frequency output
required by the load. Contemporary inverters also monitor the battery level, thus protecting the battery when the depth of discharge
has been reached. Some inverters produce a nearly square output, which is likely to cause more electromagnetic interference.

2.19.5 Stand-Alone Hybrid Systems Configurations
In this section, the most common commercial stand-alone hybrid system configurations are presented. In order for the reader
to have a comprehensive view of the opportunities given by the different combinations, scientific research results of the
SEA&ENVIPRO Lab are used for the sizing integration of each configuration. Those results refer to case studies in the isolated
Aegean Sea islands in Greece, which possess high renewable energy source potential. However, for forming an integrated concept,
Section 2.19.7 presents other optimization tools that could also be used for the sizing of stand-alone hybrid power systems.


Stand-Alone, Hybrid Systems

2.19.5.1

633

Stand-Alone Wind Power Systems

A stand-alone wind power system is one of the most interesting and environmentally friendly technological solutions for the electrification
of remote consumer premises or even entire rural areas. A properly sized wind turbine can exploit the available medium–high wind
potential (which is necessary for the implementation of such systems) for producing useful electrical energy. Most wind-only systems are
small sized; therefore, the wind turbines used in such systems are up to a few kilowatts in size. An energy storage system is also necessary and
should be properly sized in order to be able to match the electricity demand of the consumer with the stochastic behavior of wind.

More precisely, a complete stand-alone wind power system includes a wind turbine, an appropriate energy storage system, and
the electronic devices used both to control the battery operation (AC/DC rectifier, battery charge controller) and to guarantee
high-quality electricity for the consumer. Besides, an uninterruptible power supply (UPS) and a DC/AC inverter should be used at
the outlet of the system if AC output is required (see Figure 12).
The reliability of stand-alone wind power systems strongly depends on the wind turbine’s generator and control system as both
have significant influence on the overall safety and functionality of the system. Of course, the energy reliability of the system’s
performance also depends on the existence of an energy storage bank, which in most applications refers to a lead–acid battery bank.
More precisely, a stand-alone wind power system is composed of
• a wind converter of rated power Po and a given power curve P = Pw(uw) for standard-day conditions;
• an appropriate storage system for ho hours of autonomy, or, equivalently, with total capacity Qmax, operation voltage U, and
maximum discharge capacity Qmin (or an equivalent maximum depth of discharge DODL);
• an AC/DC rectifier of Pr kW and operation voltage values UAC/UDC;

uw (m s–1)
ηinv

Ew (kWh)

Pw (kW)

t (h)

Ed (kWh)

+

–θ (°C)
a

t (h)


uw (m s–1)

t (h)

P/Pinv

00

24

t (h)

ρa (kg m–3)

t (h)

Wind turbine
UPS

Charge
controller

AC/DC
rectifier

AC
load

Inverter

Control
panel

Battery

Q

U (Volts)
13
12
11
10
0

Figure 12 Wind-only stand-alone system configuration.

25

50

%
Depth of
discharge

(Ah)

t


634


Stand-Alone, Hybrid Systems

• a DC/DC charge controller of rated power Pc, charge rate Rch, and charging voltage UCC;
• a UPS of PUPS kW, frequency 50 Hz, autonomy time Δt,″ and operation voltage 230/400 V (or 60 Hz, 120/208 V in the United
States); and
• a DC/AC inverter of maximum power Pinv, capable of meeting the consumption at peak load demand, frequency 50 Hz, and
operational voltage 230/400 V (or 60 Hz, 120/208 V in the United States).
During the long-term operation of the stand-alone wind power system under discussion, the following operating modes may appear:
1. The power demand, Pd, of the consumption is less than the power output, Pw, of the wind turbine including any possible transfer
and transformation losses.
a) In that case, the energy surplus, ΔP = Pw – Pd, is available for being stored in the storage system via the rectifier and the battery
charge controller.
b) The available capacity for charging, Qch = Qmax·DOD (where DOD is the current depth of discharge), of the storage system is
compared to the energy surplus (also taking into consideration the corresponding energy losses during charging) and the
possible energy surplus is stored.
c) If the storage system is full, the residual energy is forwarded to low-priority loads.
2. The power demand, Pd, is higher than the power output of the wind turbine, Pw.
a) In that case, the state of charge, Q = Qmax·(1 − DOD), of the storage system is checked. The available storage capacity in the
storage system (taking into consideration the energy losses of the discharging process and the energy transformation losses in
the inverter) is compared to the energy deficit, ΔP, and is used to cover the load demand.
b) If the available stored energy is less than the energy deficit or if the storage system is empty, then an energy management plan
should be applied, otherwise the load will be rejected.
3. There is no available wind energy either because of low or very high wind speeds or because the machine is not available owing to
technical reasons. In that case, the energy demand will absorb the available energy from the storage system. Correspondingly to
the previous operating mode, load rejection will take place if the available stored energy is less than the power demand or the
batteries are empty and no load management plan is applied.
The sizing procedure is an important task, as the size of the wind turbine and most importantly the size of the storage system – which
presents high variable cost owing to the required replacement of the batteries during the service life of the wind turbine – may be
significantly lower if an optimization procedure is applied. In order to prove the importance of a wind-based stand-alone system

sizing, a representative case study will be presented based on Kaldellis’ [33] results. The case study concerns the electricity demand
fulfillment of a typical remote consumer (4–6 member families) by the exploitation of the available wind potential of the area. Thus, a
small wind turbine with the corresponding battery size and the necessary auxiliary equipment are used. The annual peak load is set at
3.5 kW, whereas the weekly electricity consumption varies between 80 and 100 kWh. Accordingly, the monthly electrical energy
consumption varies between 300 and 430 kWh. The area of the installation is a small Aegean Sea island (Kithnos) with an outstanding
wind potential, as in several locations in the island the annual mean wind speed approaches 7 m s−1 at 10 m height.
In a simplified approach for the sizing of a stand-alone wind-based power system, the rated power, Po, of the wind turbine ranges
as follows:
Pmin ¼

Etot
Etot
≤ Po ≤
¼ Pmax
Δt CF
Δt CF ηÃ

½3Š

where Etot is the consumer’s electricity requirements (increased by 20% to take into account potential future energy consumption
increase over the system’s lifetime) for the period Δt (e.g., 1 year), CF is the capacity factor of the installation for the same time
period, and η* is the energy transformation coefficient (round-trip efficiency), expressing the portion of the wind energy produced
and stored via the storage system, which is finally forwarded to the consumption. It should also be taken into consideration that the
power output of the proposed wind turbine should be high enough to face the maximum (peak) load demand, Pp, of the system.
The capacity factor is the product of the installation’s technical availability, Δ, and the mean power coefficient, ω, that is,
CF ¼ Δω

½4Š

with the mean power coefficient, ω, being calculated as

uf

ω ¼ ∫uC

Pw ðuw Þ
f ðuw Þdu
Po

½5Š

with uc and uf being the corresponding cut-in and cut-out wind speed, respectively, of the wind turbine examined, while Pw(uw) is
the corresponding power versus wind speed, uw, curve and f(uw) is the wind speed probability density function at hub height,
describing the local wind potential for the time period Δt.
The battery storage capacity is estimated on the basis of the hours of autonomy required for the uninterruptible operation of the
system. In wind-based applications, the hours of autonomy are calculated taking into consideration the hours of calm spells that


Stand-Alone, Hybrid Systems

635

Optimum system sizing

Storage capacity (Ah)

55 000
50 000

Zero load rejections


45 000

100 load rejections

40 000

Simplified calculation

35 000
30 000
25 000
20 000
15 000
10 000
5 000
3

5

7
9
11
Wind turbine nominal power (kW)

13

15

Figure 13 Optimum system sizing of a wind stand-alone power system.


appear, in accordance with the wind potential of the area and the maximum permissible – from the consumer – load rejections.
Therefore, the maximum battery capacity in ampere hours is given as
Qmax ¼

Etot ho
Δt ηss DODL U

½6Š

According to eqn [6], the storage system size is determined by the autonomy hours, ho, of the system; the total energy demand,
Etot, for a period Δt; the efficiency of the storage system, ηss; the maximum permitted depth of discharge, DODL; and the battery
operation voltage, U. In any case, the battery capacity, Q, varies between Qmin and Qmax, with
Qmin ¼ DODL Qmax

½7Š

where the DODL value is strongly related to the life duration (operational cycles, nc) of the batteries; for example,
DODL nc ≈ 1500 to 1800

½8Š

Load rejections per year (h)

By applying the above calculations to the specific case study, the requested nominal power of the wind turbine is estimated at
9 kW and the maximum battery capacity required is 7200 Ah at 24 V. A more accurate system sizing requires detailed wind speed
data along with the ambient conditions data. The numerical algorithm WINDREMOTE was developed by SEA&ENVIPRO Lab in
order to confront similar problems by carrying out the necessary parametrical analysis on an hourly energy production–demand
basis. The corresponding numerical algorithm is executed for each pair of Po and Qmax values in the range defined by the user on the
basis of the permitted load rejection number. The battery size is increased; the calculation is repeated; and the size combinations of
the components of the installation that satisfy the user restrictions are recorded. Calculation results are presented in Figure 13,

which depicts different size combinations for obtaining different load rejection levels [30].
The results of Figure 13 denote the significance of an analytical approach to the stand-alone system sizing. The system defined by
the simplified calculation is much smaller than the ones required for zero or even 100 h load rejections during 1 year of operation. The
choice of the simplified calculation system would result in more than 300 h load rejections. The significance of restricting load
rejections is given in Figure 14, where the number of load rejections per year (in hours) is given as a function of no-energy fulfillment
cost (in Euro/hour) [47]. According to Figure 14, high numbers of load rejections are possible only for low no-energy fulfillment cost.

No-energy fulfilment cost on load rejections

500
400
300
200
100
0
0

5

10

15

20

25

30

No-energy fulfilment cost ( h−1)

Figure 14 Impact on the non-energy fulfillment cost of the annual number of load rejections for a typical wind stand-alone power system.


636

2.19.5.2

Stand-Alone, Hybrid Systems

Stand-Alone Wind–Diesel Power Systems

In medium- or low-wind potential areas, large dimensions of wind-only stand-alone system are required; thus the corresponding
initial installation cost in some cases becomes almost prohibitive. For that reason, most remote consumers cover their electricity
demand using small oil-fired diesel-electrical generators, with low initial installation cost but very high operational cost. In this
context, implementation of the stand-alone wind power system may lead to the respective wind–diesel power system. The addition
of a diesel engine can contribute to the reliability and energy security of the system as it guarantees 100% energy demand fulfillment
under all circumstances on the condition that sufficient fuel backup exists. The schematic diagram of a typical wind–diesel hybrid
power system is presented in Figure 15.
More precisely, a typical wind–diesel hybrid system includes
• a wind converter of rated power Po and a given power curve P = Pw(uw) for standard-day conditions;
• a small internal combustion engine of P* kW, capable of meeting the consumption peak load demand Pp (i.e., P* ≥ Pp), presenting
a typical specific fuel consumption (SFC) curve versus loading of the engine, that is, SFC = SFC(P/P*);
• a lead–acid battery storage system for ho hours of autonomy, or, equivalently, with a total capacity of Qmax, operation voltage U,
and maximum discharge capacity Qmin (or equivalent maximum depth of discharge DODL);
• an AC/DC rectifier of Pr kW and operation voltage values UAC/UDC;
• a DC/DC charge controller of rated power Pc, charge rate Rch, and charging voltage UCC;
• a UPS of PUPS kW, frequency 50 Hz, autonomy time Δt and operation voltage 230/400 V (or 60 Hz, 120/208 V in the Unite States); and
• a DC/AC inverter of maximum power Pinv capable of meeting the consumption peak load demand, frequency 50 Hz, and
operational voltage 230/400 V (or 60 Hz, 120/208 V in the United States).
During the operation of a wind–diesel system, the following energy-production scenarios exist:

1. The power demand, Pd, is less than the power output, Pw, of the wind turbine (Pw > Pd). In that case, the energy surplus
(ΔP = Pw – Pd) is stored via the rectifier and the battery charge controller. If the battery is fully charged (Q = Qmax), the residual
energy is forwarded to low-priority loads.
uw (m s−1)
ηinv

Ew (kWh)

Pw (kW)

t (h)

Ed (kWh)

⊕

θa (°C)

t (h)

uw (m s−1)

t (h)

P/Pinv

00

24


t (h)

–3

ρa (kg m )

t (h)

Wind turbine
UPS

Charge
controller

AC/DC
rectifier

Diesel

AC
load

Inverter
Control
panel

SFC

Battery
1.0


P/P*

U (Volts)
13
12
11
10

0

25

Q

50

%

Depth of
discharge

Figure 15 Typical wind–diesel hybrid power system configuration.

(Ah)

t


Stand-Alone, Hybrid Systems


637

2. The power demand is higher than the wind turbine power output (Pw < Pd), for example, low wind speed, machine nonavailability.
a) In such situations, the energy deficit (ΔP = Pd – Pw) is covered by the batteries via the battery charge controller and the DC/AC
inverter, under the precondition that the corresponding battery depth of discharge DOD is lower than a given limit DODL,
that is, DOD < DODL.
b) If this precondition is not fulfilled (i.e., DOD≈DODL), then the energy deficit is covered by the diesel generator at the expense
of the existing oil reserves.
c) In the case of no further oil reserves, the energy deficit (ΔP = Pd – Pw) is covered by the batteries via the battery charge
controller and the DC/AC inverter, violating the first-degree battery protection precondition, that is, accepting DOD values
higher than DODL. However, if the battery’s maximum permitted depth of discharge is exceeded, load rejection takes place.
Further to the sizing procedure presented in the wind stand-alone section, the rated power, P*, of the engine should be capable of
meeting the consumption peak load demand, Pp, increased by an appropriate safety coefficient, SF; hence,
PÃ ≥Pp ð1 þ SFÞ

½9Š

Therefore, close attention should be paid on selecting an appropriate SFC of the diesel engine (Figure 16) especially under
partial loading (P ≠ P*) of the engine, in order to minimize the corresponding fuel consumption. It should be noted that for short
periods at zero load, diesel generator fuel consumption is almost 30% of the corresponding fuel consumption at rated power. On
top of this, it is recommended that diesel engine operation below 30% of full load for long periods be avoided, in order to prevent
serious maintenance problems, like chemical corrosion and glazing.
Further, the maximum annual fuel consumption of the installation implies the theoretical case that the energy consumption is
fulfilled solely by the diesel generator; hence, the corresponding annual fuel consumption is limited by
Mf ≤

Etot
ηd ⋅ Hu


½10Š

where ηd is the mean annual generator’s electrical efficiency and Hu is the fuel’s specific calorific value.
For the operation of a wind–diesel power system, there are two basic operational strategies possible to be adopted. The first
strategy concerns the operation of the diesel engine on a continuous basis. In that case, the diesel fuel consumption is significant and
the operational cost remains high, although one of the main purposes of installing a hybrid wind–diesel system is the reduction of
operational cost. Furthermore, in this particular operational strategy, a minimum load on the diesel engine should be considered in
order to minimize the respective fuel consumption (Figure 16). The second operating strategy concerns the operation of the diesel
engine on an intermittent basis. In this case, the diesel engine operates only if the wind turbine and the energy storage system cannot
fulfill the load demand. During the second operating strategy, limitations on the frequency of on/off cycling of the diesel engine as
also the wear of the diesel generator caused by frequent switching should be considered.
Advanced numerical simulation models can be applied to specific load demand profiles and the corresponding wind potential
of the area of the installation, for estimating the different size combinations of the wind–diesel hybrid power system. The final
decision on which of the configurations will be installed is in most cases made on the basis of either the minimum installation cost
or the minimum long-term energy production cost.
Diesel generator's specific fuel consumption curve

Specific fuel consumption (gr/kWh)

300
290
280
270
260
250
240
230
220
210
200

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Relative power (P/P*)
Figure 16 Typical SFC curve of diesel–electric generators (P/P*).

1.0

1.1

1.2

1.3


638

Stand-Alone, Hybrid Systems


Energy autonomy configurations for a winddiesel hybrid system
50 000
45 000

Battery capacity (Ah)

40 000
35 000
30 000

1500 kg yr–1
500 kg yr–1
250 kg yr–1
50 kg yr–1
25 kg yr–1
0 kg yr–1
30 000
50 000
70 000

25 000
20 000
15 000
10 000
5000
0
0

2000


4000

6000

8000
10 000 12 000 14 000
Wind turbine rated power (W)

16 000

18 000

20 000

Figure 17 Energy autonomous configurations for a wind–diesel hybrid system in Kea island capable of satisfying the energy needs of a typical
household.

The numerical code WIND-DIESEL developed by the SEA&ENVIPRO Lab is capable of estimating the appropriate configurations
of a wind–diesel hybrid system. There are three governing parameters defined by the WIND-DIESEL algorithm: the rated power,
Po, of the wind turbine that should be used, the maximum necessary capacity, Qmax, of the storage system, and the annual fuel
consumption, Mf. More specifically, given the Mf value – for each Po and Qmax pair – the algorithm is executed for all the selected time
periods (e.g., for 1 month, 6 months, 1 year or even for many years) and emphasis is laid on obtaining zero-load rejection operation.
For every (Po, Qmax, Mf) combination ensuring the energy autonomy of the remote system, energy production and demand balance
details are available along with the corresponding time-dependent battery depth of discharge and the time evolution of diesel oil
consumption.
The algorithm was applied to a typical household of 4750 kWh annual energy demand and 3.5 kW annual peak load [29]. The
corresponding wind–diesel system was sized for being installed in a small Greek island (Kea) which possesses low wind potential as
compared to the wind potential of most of the Aegean Sea islands. The mean annual wind speed in the island is 5.6 m s−1, and the
maximum calm spells appearing during a 3 year measurement period are 158 h.
The results of the WIND-DIESEL algorithm application are presented in Figure 17 for various annual diesel oil consumption

levels. More precisely, each curve corresponds to a given diesel oil annual consumption, whereas the x-axis describes the wind
turbine’s rated power and the y-axis, the corresponding battery capacity. According to the results provided, there is a considerable
battery capacity reduction by accepting a minimum (25 kg yr−1) diesel oil consumption, representing ∼1% of the annual fuel
consumption of diesel-only systems. A significant reduction in battery capacity is also come upon by accepting 250 kg yr−1 diesel oil
consumption.
The optimum configuration is subsequently predicted on the basis of the minimum long-term cost. In Figure 17, the
constant-initial cost curves are drawn without including the annual diesel oil consumption (dotted lines). If one examines the
long-term cost of the configuration, then the optimum diesel oil contribution can be realized in order to minimize the system’s life
cycle cost (Figure 18).

2.19.5.3

Stand-Alone Wind–Photovoltaic Power Systems

Wind-driven stand-alone systems have proved to be a reliable energy solution, capable of satisfying the electrification needs of
numerous remote consumers around the globe, especially in cases where the local wind energy potential is medium to high. Even
so, where local conditions are not as favorable, oversizing of the wind turbine and the excessive energy storage capacity required
discourage consumers from proceeding with RES installations [48]. On the other hand, there are several areas on our planet where
one may encounter both abundant availability of solar energy and high or medium-high wind energy potential, the combination of
which may substantially reduce the energy storage requirements of the traditional wind-based stand-alone systems. Solar and wind
energy availability vary greatly over time and therefore normally cannot match the time variation of the load demand if operating
independently. In this context, both photovoltaic and wind energy stand-alone systems require oversized storage capacity in order
to fulfill the energy demand of remote consumers. The complementary nature of wind and solar (Figure 19) can smooth out the
variation in the energy production of the system leading to a significant decrement of the energy storage requirements.


Stand-Alone, Hybrid Systems

639


20-Year cost of a wind-diesel hybrid system
140

20-year cost ( 1000)

120
100
80
60
40
20
0
0

200

400

600

800

1000

1200

1400

1600


1800

2000

Annual fuel mass consumption (kg yr–1)
Figure 18 The minimum 20 year cost of a wind–diesel hybrid system installed in Kea island.

RES potential characteristics of an area in Crete island
9

25
Solar energy
Wind speed
20

7
6

15

5
4

10

3
2

Wind speed (m s−1)


Solar energy (kWh m−2)

8

5

1
0

0
0

50

100

150
200
Day of the year

250

300

350

Figure 19 Solar energy and mean wind speed variation in Crete island.

A typical wind–PV hybrid system (Figure 20) consists of a small wind turbine, a photovoltaic generator, and an appropriate
storage system along with the corresponding electronic equipment. More precisely, it involves

• a wind converter of rated power Po and a given power curve P = Pw(uw) for standard-day conditions;
• a photovoltaic array of z panels (with a maximum/peak power P+ of each panel) properly connected to feed the charge controller
to the required voltage;
• a lead-acid battery storage system for ho hours of autonomy, or, equivalently, with a total capacity of Qmax, operation voltage
U, and maximum discharge capacity Qmin (or an equivalent maximum depth of discharge DODL);
• an AC/DC rectifier of Pr kW and operation voltage values UAC/UDC;
• a DC/DC charge controller of rated power Pc, charge rate Rch, and charging voltage UCC;
• a UPS of PUPS kW, frequency 50 Hz, autonomy time Δ′t and operation voltage 230/400 V (or 60 Hz, 120/208 V in the United
States) (The utilization of this device is optional, and it is used to protect the installation from any unexpected electricity
production fluctuations owing to the stochastic behavior of wind); and
• a DC/AC inverter of maximum power Pinv capable of meeting the consumption peak load demand, frequency 50 Hz, and
operational voltage 230/400 V (or 60 Hz, 120/208 V in the United States).


640

Stand-Alone, Hybrid Systems

uw (m–1 s)

t (h)
⊕

θa (°C)

ηinv

Ew (kWh)

Pw (kW)


uw (m–1 s)

Ed (kWh)

t (h)

P/Pinv

t (h)

00

24

t (h)

–3
ρa (kg m )

t (h)

Wind turbine
UPS

Charge
controller

AC/DC
Rectifier


AC
load

Inverter
Control
panel

Battery

PV array

U

(Volts)

Q

13
12
11
10

%
0

25

–2


I (A)
θa (°C) ⊕

θc (°C)

t (h)

–2
(W m )

ηinv

EPV(kWh)

⊕

t (h)

t

50

Depth of
discharge

GT (W m )

t (h)

η/η


(Ah)

Ek (kWh)

⊕ uPV
t (h)

N

P/Pinv

00

24 t

(h)

θc (°C)

Figure 20 Typical wind–photovoltaic hybrid system configuration.

During the long-term operation of a stand-alone wind–PV hybrid system the following situations may arise:
1. The power demand, Pd, of the consumption is less than the power output, Pw, of the wind turbine (Pw > Pd). In that case, the
energy surplus (ΔP = Pw − Pd) is stored via the rectifier and the battery charge controller along with the energy production of the
photovoltaic generator, PPV. If the battery is fully charged (Q = Qmax), the residual energy is forwarded to low-priority loads.
2. The power demand is more than the power output of the wind turbine (Pw < Pd) but less than the sum of the powers of the
photovoltaic station and the wind converter, that is, Pw + PPV > Pd. In that case, the extra load demand is covered by the photovoltaic
station via the DC/AC inverter. Any energy surplus from the photovoltaic station is stored in the battery via the charge controller. If
the battery is fully charged (Q = Qmax), the residual energy is forwarded again to low-priority loads.

3. The power demand is more than the combined power output of the two renewable stations, that is, Pw + PPV < Pd, where
Pw + PPV ≠ 0. In such situations, the energy deficit (ΔP = Pd − (Pw + PPV)) is covered by the batteries via the charge controller and
the DC/AC inverter under the condition Q > Qmin. During this operational condition, special emphasis is laid on the manage­
ment plan for the three electricity production subsystems.
4. There is no renewable energy production (e.g., low wind speed or machine not available, and zero solar irradiance), that is,
Pw + PPV = 0. In that case, all the energy demand is covered by the energy storage subsystem under the condition Q > Qmin.


Stand-Alone, Hybrid Systems

641

Photovoltaic module operation curves
3.5
3.0

1000 W m–2
800 W m–2

Current I (A)

2.5

600 W m–2
400 W m–2

2.0


200 W m–2


1.5
1.0
0.5
0.0
0

5


15



10

20

25

Voltage (V)

Figure 21 Operation curves of a photovoltaic module for different solar radiation values.

5. In cases (3) and (4) above, when the battery capacity is near the bottom limit, an electricity demand management plan should be
applied; otherwise the load would be rejected.
In order to maximize the energy security of the system, a diesel engine could be added to act as a backup energy source in the
extreme case that no renewable energy production is available and at the same time no energy is available in the storage system.
The sizing procedure of a wind–PV hybrid system is much more complicated than the above-mentioned stand-alone systems, as
one has to match the stochastic wind generation and the fluctuating PV generation with the time distribution of load demand.

Generally, the PV systems comprise an array of PV modules that produce electricity, taking advantage of the existing solar radiation,
according to the PV operational curves (Figure 21). The number, z, of PV panels is bounded as follows:
Etot
Etot
≤ z P þ ≤
8760 CFPV ηÃ
8760 CFPV

½11Š

According to the charge controller voltage, Ucc, and the photovoltaic panels’ operation voltage, UPV, the necessary number of
photovoltaic panels, z2, connected in series is estimated as:
z2 ¼

Ucc
UPV

½12Š

Therefore, ‘z1’ parallel strings of panels are required for the installation:
z1 ¼

z
z2

½13Š

Note that P+ is the nominal power of the module, CFPV is the photovoltaic installation’s capacity factor, and η* is the
corresponding energy transformation coefficient, given that the PV production is not rectified.
In order to investigate the potential configurations of stand-alone wind–PV hybrid power systems, a more detailed sizing

procedure should be followed. More precisely, given that analytical data of the operational characteristics exist, integrated computa­
tional algorithms could be used for the estimation of the most appropriate system configuration to be chosen. The analytical system
sizing can considerably increase the stand-alone system’s reliability and decrease the installation cost and, furthermore, the long-term
energy production cost. The main inputs required for an analytical sizing procedure are
• detailed wind speed, uw, measurements at hub height over a given time period (e.g., 1 year);
• detailed solar radiation, G, measurements over a given time period (e.g., 1 year) usually at a horizontal plane;
• ambient temperature, θa, and pressure data for the entire period analyzed;
• operational characteristics of the wind turbine (at standard-day conditions);
• operational characteristics (current, voltage) of the PV modules selected;
• operational characteristics of all the other electronic devices of the installation; and
• the electricity consumption profile on an hourly basis, being also dependent on the period of the year analyzed (winter, summer,
or other).
An example will be presented based on the research of Kaldellis et al. [31], which strengthens the necessity of the analytical hybrid
system dimension estimation in order to guarantee energy autonomy of a typical remote consumer. In this study, the numerical


642

Stand-Alone, Hybrid Systems

algorithm WT-PV, developed by SEA&ENVIPRO Lab, was used. This algorithm estimates the combination of the required wind
turbine size, Po; the number of photovoltaic panels, z, needed; and the corresponding battery capacity, Qmax, that will guarantee
system energy autonomy for a given period of time. WT-PV is based on the following steps:
1. For the region analyzed, the wind turbine rated power, Po, is selected, taking values from a specific numerical range defined by
the user.
2. Accordingly, the number, z, of PV panels, each with a peak power P+, is determined, based on the respective operational
characteristics.
3. A battery capacity is selected, starting from a minimum value, while a maximum battery capacity limit also exists. Battery capacity
range can vary according to the user’s definitions. For every time point of a given time period, the wind energy, Pw, produced by
the wind turbine and the energy yield, PPV, of the photovoltaic generator are estimated, taking into account the existing wind

speed, the available solar radiation, the ambient temperature and pressure, the selected wind turbine power curve, and the power
curve of the photovoltaic panels (see Figure 21).
4. The wind energy production is compared with the consumer’s energy demand, Pd. If an energy surplus occurs (Pw > Pd), the
energy surplus along with the energy produced by the PV generator is stored in the battery system and a new time point is
analyzed. Otherwise, the algorithm proceeds to the next step.
5. If (Pw < Pd), the energy deficit (Pd – Pw) is covered by the photovoltaic generator production. Any energy surplus is stored to the
batteries and a new time point is analyzed. If this is not the case, the algorithm proceeds to the next step.
6. The energy deficit (Pd – Pw – PPV) is finally covered by the energy storage system, if the batteries are not near the lower capacity
permitted limit (Q > Qmin). Accordingly, the algorithm is repeated from step (4). In case the battery is practically empty, the
battery size is increased by a given quantity, provided the maximum battery capacity limit is not exceeded. Then the complete
analysis is repeated from step (3). If the maximum battery size is reached, the number of photovoltaic panels is increased and the
algorithm proceeds to step (2). In case the maximum available number of photovoltaic panels is reached, a new wind turbine
rated power is selected and the algorithm restarts.
After the analysis is completed, the distribution Qmax = Qmax(Po,z) is predicted, taking into account that every set of Qmax, Po, and z
guarantees the energy autonomy of the remote consumer for the entire period analyzed. The optimum configuration may be
subsequently predicted on the basis of an appropriate criterion, for example, the minimum initial cost.
A case study in the island of Zakynthos in Greece presented by Zafirakis et al. [49] is selected for the implementation of the above
algorithm. The wind–PV hybrid system should be able to cover the electricity needs of a typical remote consumer with a given
consumption profile based on his seasonal electricity needs. In this study, the peak power demand of the remote consumer does not
exceed 3.5 kW, whereas the annual energy consumption reaches ∼4750 kWh. The configuration was designed to be installed in
Zakynthos island, which possesses medium solar energy potential (1500 kWh m−2 yr−1) and medium wind potential with mean
annual wind speed values reaching 6 m s−1. Figure 22 presents the different configurations that are capable of guaranteeing 100%
energy autonomy totally based on renewable energy sources for 1 year of operation. Scenarios of wind–battery and PV–battery were
also included by the authors for comparison. According to their results, a stand-alone PV system would require at least 1250 Ah storage
capacity combined with more than 10 kW of PV installation. On the other hand, the wind stand-alone system requires significantly
higher battery capacities, up to 6000 Ah, for the same wind turbine size (15 kW). The combination of both renewable energy sources
could decrease the storage system requirements as well as the necessary size of the wind turbine and the photovoltaic installation.

Stand-alone configurations solutions for Zakynthos island
20 000


Wind-only
10 panels
20 panels
50 panels
75 panels
100 panels
PV-only

18 000
Battery capacity (Ah)

16 000
14 000
12 000
10 000
8000
6000
4000
2000
0
0

1500 3000 4500 6000 7500 9000 10 500 12 000 13 500 15 000
Wind power (Watts) (or PV power (Watts) for the PV-only solution)

Figure 22 Variation of a wind–PV hybrid system dimensions in Zakynthos island.


Stand-Alone, Hybrid Systems


643

Stand-alone configurations electricity production cost for Zakynthos island
Electricity production cost (Euro kWh−1)

2.5

Wind-only

10 panels

20 panels

50 panels

2.1

75 panels

100 panels

1.9

PV-only

2.3

1.7
1.5

1.3
1.1
0.9
0.7
0.5
0

1500 3000 4500 6000 7500 9000 10 500 12 000 13 500 15 000
Wind power (Watts) (or PV power (Watts) for the PV-only solution)

Figure 23 Electricity production cost of wind–PV stand-alone configurations for Zakynthos island.

As already mentioned, selection of the most appropriate configuration is normally decided on an economic basis: either the
minimization of the initial installation cost or the optimization of the electricity production cost. Figure 23 presents the electricity
production cost for the different configurations of Figure 22. According to Figure 23, the minimum electricity production cost
ranges between 0.6 € kWh−1 for a 500 W wind turbine and 5 kW of PV, and 1.2 € kWh−1 for a PV-only installation of 7 kW. It is also
important to note that for high wind turbine rated power (15 kW) the electricity production cost presents very narrow variations
converging at about 1.1 € kWh−1.

2.19.5.4

Stand-Alone Wind–Hydro Power Systems

Complementarity of renewable energy sources can also be exploited in wind–hydro power systems, which are based on the
exploitation of both wind potential and hydraulic power in order to enhance the reliability, energy quality, and stand-alone system
performance. In addition, the water storage capability of the hydroelectric system can significantly limit the intermittence of wind
power generation. Thus, a stand-alone wind–hydro power system does not essentially refer to the independent production of
electricity by a hydro power installation or a wind turbine, both of which supply energy to a remote consumer. The wind–hydro
concept mainly refers to the integration of a wind power installation with a pumped hydro storage (PHS) system that will be able to
absorb the residual wind energy during low-power demand periods and return it for consumption when wind power cannot satisfy

the demand. The implementation of wind power generation with PHS is targeting mainly the range of isolated communities in
remote islands with no connection to any mainland grid rather than single consumers as indicated for the systems described in the
previous sections.
The integration of wind power with PHS has been investigated for at least 20 years by numerous researchers [37, 46, 50, 51].
Most of the cases analyzed refer to isolated islands with the target of minimizing the conventional fuel energy consumption and
eliminating the negative environmental impacts. Combined wind–hydro energy stations can contribute to the maximum RES
penetration into the load demand, which, according to research results, can even exceed 90%.
A typical wind–hydro power system capable of fulfilling the energy needs of an isolated community is presented in Figure 24.
More precisely, the hybrid system consists of
•
•
•
•

one or more wind turbines,
a small hydroelectric power plant,
a water pump station, and
two or more water reservoirs at elevations h1 and h2 (h1 > h2) working in a closed circuit along with the corresponding pipelines.

The hybrid wind–hydro power plant is usually supplemented by an existing autonomous power station (APS) which usually
comprises conventional internal combustion engines. The main objective of the wind–hydro station is the fulfillment of the energy
demand by increasing the renewable energy source absorption and reducing the operation time of the local APS.
The sizing procedure of the wind–hydro power system includes sizing of the wind turbine and the hydro turbine, as well as the
determination of the exact location, volume, and geometry of the water reservoirs along with the determination of the rated power
and operational range of the water pumps and the water piping system dimensions (diameter, length).
More precisely, the rated power of the water pumps may be determined by the maximum power of the wind turbines, as the
water pump must have the capability to absorb the maximum power output of the wind turbines, whereas in the case of large-scale
systems, the rated power of the pump depends on the frequency distribution of the wind park’s energy surplus; that is,



644

Stand-Alone, Hybrid Systems

Upper reservoir
(h1)

APS

Energy
consumption

Wind park
(zWTs)

Water

pump

station

Lower reservoir (h2)

Reversible

hydraulic

machines



Figure 24 Combined wind–hydro installation for remote communities.

Ppump ¼

ρw g H V_
ηp ηel

½14Š

where Ppump is the power required by the water pumps; H the pump head; V_ the volume flow rate; ηp the pump efficiency; ηel the
electrical efficiency of the system; ρw the density of the water; and g the acceleration due to gravity.The static head, H, of the pump
must satisfy the expression
H ≥ ðh1 − h2 Þ þ δ Hf ¼ ðh1 − h2 Þ þ Kp V_

2

½15Š

where δHf is the total hydraulic losses, both lengthwise and local, when the water reservoir is used for energy storage and Kp is the
friction losses factor.
It should be noted that H and ηp depend on the operational characteristics of the selected pump.
The nominal power of the hydro installation results from the precondition that it covers the peak power demand of the system
each time examined, with an optional future increase (of 20%). The exit power is given as
_ ηH η′el
Ph ¼ ρw g H′ V′

½16Š

where V_ ′ is the flow rate of the turbine; H′ the hydro turbine head, ηH the turbine efficiency, and η′el the electrical efficiency of the
system.

In addition, the following equation is also valid:
_
H′≤ ðh1 − h2 Þ − δ H′ f ¼ ðh1 − h2 Þ−KH V′

2

½17Š

where h is the hydrostatic head and δHf′ is the total hydraulic losses, both lengthwise and local, when the water circuit is used for
energy production.
Note that H′ and ηH depend on the operational characteristics of the hydro turbine selected.
The dimensions of the upper water reservoir are defined by the available hydrostatic head, which depends on the relative
elevation between the upper and the lower water reservoirs, and by the required levels of energy autonomy for the system. For
example, by selecting d0 days of energy autonomy, the useful volume Vo of the water reservoir is given as
Vo ¼

Etot 24d0
¼ Vmax − Vmin
Δt ηΗ ηel ρw g H′

½18Š

where Etot is the total energy demand for the time duration of analysis, Δt, in hours (e.g., 8760 h for 1 year); and Vmax and Vmin are
the maximum and minimum storage capacity, respectively, of the upper water reservoir.
During a long-term energy balance analysis of a wind–hydro power system operation, the following operational situations
may arise:
1. The wind power produced is in excess of the energy demand of the system.
a) In that case, the energy surplus is stored through operation of the water pumping system in the upper reservoir.



Stand-Alone, Hybrid Systems

645

Renewable energy sources penetration in Karpathos island
100

RES Penetration (%)

95
90
85
80
do = 1.0
do = 1.5
do = 2.0
do = 2.5
do = 3.0

75
70
65
60
12

13

14

15


16

17

18

19

20

21

22

23

24

25

26

27

28

Number of WTs (z)
Figure 25 Renewable energy sources penetration capability using the wind–hydro solution in the autonomous electrical system of Karpathos island.


b) In case the upper reservoir is full, the energy surplus is forwarded to other alternative uses, such as a water desalination plant.
2. The electrical power demand is higher than the wind park output.
a) In that case, the hydro turbines cover the power deficit.
b) In case the upper reservoir is almost empty, the internal combustion engines of the APS take over the power deficit, under a
scheduled operational plan.
For estimating the optimum wind–hydro configuration, advanced numerical algorithms should be used, to analytically simulate
the operation of different system size combinations. By applying an analytical simulation procedure, Kaldellis and Kavadias [52]
presented interesting results regarding the renewable energy possibilities in the electrification of remote islands. The study, which is
presented here, concerned a medium-sized Aegean Sea island (Karpathos), and the basic scope was the maximization of RES
penetration. The annual energy production of the local APS of the island was estimated at 24 400 MWh and the peak-load demand
at ∼6500 kW, whereas the corresponding minimum value was 1400 kW. The island has a very high wind potential, as the long-term
annual mean wind speed approaches 9.6 m s−1, at 10 m. According to their results, remarkable renewable energy penetration can be
achieved (Figure 25) by increasing the number of wind turbines used and the size of the water reservoirs selected through the
parameter do which represents the number of days of energy storage autonomy.
Another interesting optimization approach for the economy enhancement of large wind–hydro installations concerns a planned
hydro power production under a pattern of guaranteed energy by the hydro power system on a daily basis during the peak load
demand hours. In this way, high energy-purchase prices can be realized by selling power to the local autonomous grid during the
peak load demand hours [46]. Of course, in case the water stored in the upper reservoir is not enough for the fulfillment of the
condition of guaranteed energy delivered to the local grid, the water pump absorbs the required energy from the grid during
low-demand periods when the energy purchased price from the local grid is low.

2.19.5.5

Stand-Alone Wind–Hydrogen Power Systems

As already mentioned in Section 2.19.2, the first wind turbine installed in Denmark in 1891 was generating DC, which was used for
electrolysis to produce hydrogen. The hydrogen produced was used for gas lighting and later on for autogenous welding [3, 53].
Since then wind energy was scarcely used for hydrogen production, and only during the last few decades did wind–hydrogen
stand-alone systems become a reality. In this context, such configurations, when in stand-alone system mode, exploit the energy
surplus of the wind turbine to produce hydrogen, provided of course that the electricity demand of the consumption side has been

satisfied. Hydrogen as an energy carrier can be stored to overcome the daily and seasonal discrepancies between energy source
availability and demand [54]. In cases of load deficit or no available wind, the hydrogen is fed to a fuel cell device to produce
electricity and satisfy the demand.
There are several issues that should be taken into consideration when using wind energy for the production of hydrogen.
More precisely, direct coupling of an electrolyzer with a wind turbine denotes intermittent operation and highly variable
power output, which could cause the electrolyzer to operate at very low power rate resulting in the mixing of H2 and O2,
which at such load levels permeate through the electrolyzer. Also, intermittent power makes the electrolyzer operate at
temperatures lower than the respective nominal, as some amount of time is required for the electrolyzer to reach its normal
operating temperature [55]. These problems in stand-alone power systems can be eliminated by the use of either a
complementary renewable power source or a diesel generator, which may fill the power gaps during the electrolyzer’s


646

Stand-Alone, Hybrid Systems

Electrolyser's operation curves

31
20 °C
30 °C
40 °C
50 °C
60 °C

29

Voltage (V)

27

25
23
21
19
0

5

10

15
Current (A)

20

25

30

Figure 26 Operational characteristics of electrolyzers for different power levels and temperatures, based on Reference [55].

operation [56]. In this case, an extended economic analysis should be undertaken in order for the system to be proved not
only energy-efficient but also cost-effective (Figure 26).
The use of an electrolyzer–fuel cell set as a storage option for wind stand-alone systems is still in its initial stage, although both
electrolyzer and fuel cell technologies have achieved considerable progress during the last decades. The main drawback of such
configurations is the low energy efficiency of the charge–discharge cycle (round-trip efficiency), which is estimated to be between
30% and 40%. The use of advanced electrolyzers can raise the efficiency even up to 60%, but, on the other hand, increased purchase
cost should be taken into consideration.
A very interesting and viable application of wind–hydrogen stand-alone systems could be in configurations where a bulk energy
storage system is necessary in order to absorb the rejected wind energy in remote electricity grids, similar to the case of wind–hydro

power applications. Islands where usually weak electrical grids exist often possess significantly high renewable energy potential.
However, the local autonomous electrical networks in most cases are unable to absorb the renewable energy produced. The other
barriers against renewable energy penetration in islands are caused mainly by the significant difference between energy production
and demand. As a result, significant amounts of renewable energy are rejected by the local grids. Furthermore, one should not
neglect the cost of electricity production in remote electrical grids, where in most cases the electricity production cost could even be
4 times the corresponding selling price. By storing the excess wind energy and using it during peak load demands, that is, when there
is no wind available, renewable energy penetration limits may be bypassed and the disturbance of the local grid stability may be
avoided. The use of the stored energy at peak demand can also improve the operation of the autonomous power stations and reduce
their operational cost.
A typical wind–hydrogen configuration capable of fulfilling the energy needs of an isolated community is presented in
Figure 27. More precisely, the hybrid system consists of
• one or more wind turbines;
• a water purification unit to improve the quality of the water used;
• a water storage tank to ensure that the process has adequate water in storage in case the water supply system is interrupted;
• an electrolyte solution (in alkaline systems);
• a hydrogen generation unit consisting of an electrolysis stack, a gas purification module, a dryer, and a heat removal system;
• a hydrogen storage medium (It should be noted that in order to fill the hydrogen tank, a compressor may be required if the
electrolyzer is not designed to provide high pressure [57]);
• a fuel cell electricity generation unit; and
• a power conditioning and control unit.
During the operation of a wind–hydrogen energy production installation, the following operational situations may arise:
1. The wind power produced is in excess of the energy demand of the system.
a) In that case, the hydrogen production unit absorbs the energy surplus to produce and store hydrogen.
b) In case the available wind energy is more than the hydrogen production unit’s power capacity or less than the minimum
power required, the surplus is transferred to low-priority loads.


Stand-Alone, Hybrid Systems

647


UPS
Wind
turbine
Electricity
consumption
Rectifier

Inverter

Fuel cell
syetem
Compressor unit

Control
panel
Water flow
Hydrogen flow
DC current
AC current

Feed water
Electrolysis
unit

Hydrogen
storage

Figure 27 Integrated wind–hydrogen stand-alone installation.


2. The electrical power demand is more than the wind park output.
a) In that case, the fuel cell unit covers the power deficit.
b) In case there is not enough stored hydrogen, the internal combustion engines of the APS cover the power deficit, under a
scheduled operation plan.
An alternative operation mode of a wind–hydrogen stand-alone system was suggested by Ntziachristos et al. [58]. In their study on a
hybrid wind–fuel-cell power station, they consider that the electrolyzer should always remain in operation, and in cases where the
wind turbine is not in operation, an internal loop should provide the electrolyzer with electrical power at standby levels from the
fuel cell unit in order to avoid intermittent operation. In addition, in such a case, the wind turbine’s electricity production is
primarily supplied to the electrolyzer for producing and storing hydrogen. The analysis was based on a case study for the Karpathos
island in Greece. Karpathos Island is a medium-size island in Greece with an autonomous electricity grid and possesses excellent
wind potential. The simulation procedure developed by Ntziachristos et al. [58] introduced a level of hybridization, which indicates
the ratio of wind energy delivered directly to the local grid to the energy delivered to the grid from the fuel cell. Given that the fuel
cell provides constant power to the grid, the level of hybridization also indicates the variation over the constant power delivered to
the grid. According to the results of their study, depending on the wind turbine selected and the preferred hybridization ratio,
different electrolyzer rates and storage tank capacities can be realized.
In an attempt to define the energy production cost of a wind–hydrogen power system, Kavadias et al. [59] investigated the option
of installing electrolyzers in order to absorb the wind energy rejected by a remote local electricity network in the island of Crete in
Greece. Crete island is of great importance, as it has the largest autonomous electrical network in Greece. Although the island
possesses excellent wind potential and faces substantial energy demand fulfillment, wind energy cannot be fully exploited owing to
the local electrical grid instability barriers [60]. By an analysis made by Kaldellis et al. [60], it was estimated that in a 25 MW wind
park almost 10% of the annual production was rejected leading to an average income loss of 25 000 € MW−1 for the wind park
owners on the island [61]. According to their results, an optimum configuration could be achieved on the basis of minimum
hydrogen production cost, which depends on the wind energy purchase price (Figure 28).

2.19.6 Energy Storage in Wind Stand-Alone Energy Systems
In long-term operation, dependence on the wind turbine energy production leads to the question of security of supply (firm power).
Because of the stochastic nature of wind, independent and firm power cannot be realized by means of wind turbines alone. It
requires a holistic supply concept which includes at least an energy storage system. All efforts in striving for an autonomous energy
system with the aid of renewable energy sources always end up with a requirement for energy storage. The search for cost-effective
energy storage is a theme pervading the whole range of these technologies. To exaggerate, one might say that as soon as an

economically viable solution for storing energy has been found, all energy problems concerning the utilization of renewable energy
sources can be solved [3].