Module- IV - Types of Electrical storage systems

4.1 Double-layer capacitors (DLC)

Electrochemical double-layer capacitors (DLC), also known as super capacitors, are a technology which has been
known for 60 years. They fill the gap between classical capacitors used in electronics and general batteries,
because of their nearly unlimited cycle stability as well as extremely high power capability and their many orders
of magnitude higher energy storage capability when compared to traditional capacitors. This technology still
exhibits a large development potential that could lead to much greater capacitance and energy density than
conventional capacitors, thus enabling compact designs.

The two main features are the extremely high capacitance values, of the order of many thousand farads, and the
possibility of very fast charges and discharges due to extraordinarily low inner resistance which are features not
available with conventional batteries.

Still other advantages are durability, high reliability, no maintenance, long lifetime and operation over a wide
temperature range and in diverse environments (hot, cold and moist). The lifetime reaches one million cycles (or
ten years of operation) without any degradation, except for the solvent used in the capacitors whose disadvantage is
that it deteriorates in 5 or 6 years irrespective of the number of cycles. They are environmentally friendly and
easily recycled or neutralized. The efficiency is typically around 90
% and discharge times are in the range of seconds to hours.

They can reach a specific power density which is about ten times higher than that of conventional batteries (only
very-high-power lithium batteries can reach nearly the same specific power density), but their specific energy
density is about ten times lower.

Because of their properties, DLCs are suited especially to applications with a large number of short
charge/discharge cycles, where their high performance characteristics can be used. DLCs are not suitable for the
storage of energy over longer periods of time, because of their high self-discharge rate, their low energy density
and high investment costs.

Since about 1980 they have been widely applied in consumer electronics and power electronics. A DLC is also
ideally suited as a UPS to bridge short voltage failures. A new application could be the electric vehicle, where they
could be used as a buffer system for the acceleration process and regenerative braking [7].

4.2 Superconducting magnetic energy storage (SMES)

Superconducting magnetic energy storage (SMES) systems work according to an electrodynamic principle. The
energy is stored in the magnetic field created by the flow of direct current in a superconducting coil, which is kept
below its superconducting critical temperature. 100 years ago at the discovery of superconductivity a temperature of
about 4 °K was needed. Much research and some luck has now produced superconducting materials with higher
critical temperatures. Today materials are available which can function at around 100 °K. The main component of
this storage system is a coil made of superconducting material. Additional components include power conditioning
equipment and a cryogenically cooled refrigeration system.

The main advantage of SMES is the very quick response time: the requested power is available
almost instantaneously. Moreover the system is characterized by its high overall round-trip efficiency (85
% - 90 %) and the very high power output which can be provided for a short period of time. There are no moving
parts in the main portion of SMES, but the overall reliability depends crucially on the refrigeration system. In
principle the energy can be stored indefinitely as long as the cooling system is operational, but longer storage times
are limited by the energy demand of the refrigeration system.

1
Large SMES systems with more than 10 MW power are mainly used in particle detectors for high-energy physics
experiments and nuclear fusion. To date a few, rather small SMES products are commercially available; these are
mainly used for power quality control in manufacturing plants such as microchip fabrication facilities [17 ].

4.3 Thermal storage systems

Thermal (energy) storage systems store available heat by different means in an insulated repository for later use
in different industrial and residential applications, such as space heating or cooling, hot water production or
electricity generation. Thermal storage systems are deployed to overcome the mismatch between demand and
supply of thermal energy and thus they are important for the integration of renewable energy sources.

Thermal storage can be subdivided into different technologies: storage of sensible heat, storage of latent heat, and
thermo-chemical adand absorption storage [18 ]. The storage of sensible heat is one of the best-known and most
widespread technologies, with the domestic hot water tank as an example. The storage medium may be a liquid
such as water or thermo-oil, or a solid such as concrete or the ground.
Thermal energy is stored solely
through a change of temperature of the storage medium. The capacity of a storage system is defined by the
specific heat capacity and the mass of the medium used.

Latent heat storage is accomplished by using phase change materials (PCMs) as storage media. There are organic
(paraffins) and inorganic PCMs (salt hydrates) available for such storage systems. Latent heat is the energy
exchanged during a phase change such as the melting of ice. It is also called “hidden” heat, because there is no
change of temperature during energy transfer. The best-known latent heat – or cold – storage method is the ice
cooler, which uses ice in an insulated box or room to keep food cool during hot days. Currently most PCMs use the
solid-liquid phase change, such as molten salts as a thermal storage medium for concentrated solar power (CSP)
plants [19 ]. The advantage of latent heat storage is its capacity to store large amounts of energy in a small volume
and with a minimal temperature change, which allows efficient heat transfer.

Sorption (adsorption, absorption) storage systems work as thermo-chemical heat pumps under vacuum conditions
and have a more complex design. Heat from a high-temperature source heats up an adsorbent (e.g. silica gel or
zeolite), and vapour (working fluid, e.g. water) is desorbed from this adsorbent and condensed in a condenser at low
temperatures. The heat of condensation is withdrawn from the system. The dried adsorbent and the separated
working fluid can be stored as long as desired. During the discharging process the working fluid takes up low-
temperature heat in an evaporator. Subsequently, the vapour of the working fluid adsorbs on the adsorbent and heat
of adsorption is released at high temperatures [20 ]. Depending on the adsorbent/working fluid pair the temperature
level of the released heat can be up to 200 °C [18 ] and the energy density is up to three times higher than that of
sensible heat storage with water. However, sorption storage systems are more expensive due to their complexity.

In the context of ESE, it is mainly sensible/latent heat storage systems which are important. CSP plants primarily
produce heat, and this can be stored easily before conversion to electricity and thus provide dispatchable electrical
energy. State-of-the-art technology is a two-tank system for solar tower plants, with one single molten salt as heat
transfer fluid and storage medium [21 ]. The molten salt is heated by solar radiation and then transported to the hot
salt storage tank. To produce electricity the hot salt passes through a steam generator which powers a steam turbine.
Subsequently, the cold salt (still molten) is stored in a second tank before it is pumped to the solar tower again. The
main disadvantages are the risk of liquid salt freezing at low temperatures and the risk of salt decomposition at
higher temperatures. In solar trough plants a dual-medium storage system with an intermediate oil/salt heat
exchanger is preferred [21]. Typical salt mixtures such as Na-K-NO3 have freezing temperatures > 200 °C, and
storage materials and containment require a higher volume than storage systems for solar tower plants. The two-
tank indirect system is being deployed in “Andasol 1-3”, three 50 MW parabolic trough plants in southern Spain,
and is planned for Abengoa Solar’s 280 MW Solana plant in Arizona. Apart from sensible heat

2
storage systems for CSP, latent heat storage is under development by a German-Spanish consortium – including
DLR and Endesa – at Endesa’s Litoral Power Plant in Carboneras, Spain. The storage system at the pilot facility is
based on sodium nitrate, has a capacity of 700 kWh and works at a temperature of 305 °C [22 ].

In adiabatic CAES the heat released during compression of the air may be stored in large solid or liquid sensible
heat storage systems. Various R&D projects are exploring this technology [23 ] [24], but so far there are no
adiabatic CAES plants in operation. As solid materials concrete, cast iron or even a rock bed can be employed. For
liquid systems different concepts with a combination of nitrate salts and oil are in discussion. The round-trip
efficiency is expected to be over 70 % [25].

Of particular relevance is whether a pressurized tank is needed for the thermal storage, or if a non- pressurized
compartment can be used. In liquid systems, a heat exchanger can be used to avoid the need for a large pressurized
tank for the liquid, but the heat exchanger means additional costs and increases the complexity. A dual-media
approach (salt and oil) must be used to cover the temperature range from 50 °C to 650 °C [24]. Direct contact
between the pressurized air and the storage medium in a solid thermal storage system has the advantage of a high
surface area for heat transfer. The storage material is generally cheap, but the pressurized container costs are
greater.

4.4 Standards for ESE

For mature ESE systems such as PHS, LA, NiCd, NiMH and Li-ion various standards exist. The standards cover
technical features, testing and system integration. For the other technologies there are only a few standards, covering
special topics. Up to now no general, technology-independent standard for ESE integration into a utility or a stand-
alone grid has been developed. A standard is planned for rechargeable batteries of any chemistry.

Standardization topics for ESE include:

• terminology
• basic characteristics of ESE components and systems, especially definitions and measuring methods for
comparison and technical evaluation
- capacity, power, discharge time, lifetime, standard ESE unit sizes
• communication between components
- protocols, security
• interconnection requirements
- power quality, voltage tolerances, frequency, synchronization, metering
• safety: electrical, mechanical, etc.
• testing
• guides for implementation.

4.5 Technical comparison of ESE Technologies

The previous sections have shown that a wide range of different technologies exists to store electrical energy.
Different applications with different requirements demand different features from ESE. Hence a comprehensive
comparison and assessment of all storage technologies is rather ambitious, but in Figure 4-1 a general overview of
ESE is given. In this double logarithmic chart the rated power (W) is plotted against the energy content (Wh) of
ESE systems. The nominal discharge time at rated power can also be seen, covering a range from seconds to
months. Figure 4-1 comprises not only the application areas of today’s ESE systems but also the predicted range in
future applications. Not all ESE systems are commercially available in the ranges shown at present, but all are
expected to become important. Most of the technologies could be implemented with even larger power output and
energy capacity, as all systems have a modular design, or could at least be doubled (apart from PHS and some
restrictions for underground storage of H2, SNG and CAES). If a larger power range or higher energy capacity is
not realized, it will be mainly for economic reasons (cost per kW and cost per kWh, respectively).

3
On the basis of Figure 4-1 ESE technologies can be categorized as being suitable for applications with:

• Short discharge time (seconds to minutes): double-layer capacitors (DLC), superconducting magnetic energy
storage (SMES) and flywheels (FES). The energy-to-power ratio is less than 1 (e.g. a capacity of less than 1 kWh
for a system with a power of 1 kW).

• Medium discharge time (minutes to hours): flywheel energy storage (FES) and – for larger capacities –
electrochemical ESE, which is the dominant technology: lead-acid (LA), Lithium ion (Li-ion) and sodium sulfur
(NaS) batteries. The technical features of the different electrochemical techniques are relatively similar. They have
advantages in the kW - MW and kWh - MWh range when compared to other technologies. Typical discharge times
are up to several hours, with an energy-to-power ratio of between 1 and 10 (e.g. between 1 kWh and 10 kWh for a 1
kW system). Batteries can be tailored to the needs of an application: tradeoffs may be made for high energy or high
power density, fast charging behavior or long life, etc.

• Long discharge time (days to months): hydrogen (H2) and synthetic natural gas (SNG). For these ESE systems
the energyto- power ratio is considerably greater than 10.

Pumped hydro storage (PHS), compressed air energy storage (CAES) and redox flow batteries are situated between
storage systems for medium and long discharge times. Like H 2 and SNG systems, these ESE technologies have
external storage tanks. But the energy densities are rather low, which limits the energy-to-power ratio to values
between approximately 5 and 30.

Figure 4.1. Comparison of rated power, energy content and

discharge time of different energy storage technologies

In Figure 4-2 the power density (per unit volume, not weight) of different ESE technologies is plotted versus the
energy density. The higher the power and energy density, the lower the required volume for the storage system.
Highly compact ESE technologies suitable for mobile applications can be found at

4
the top right. Large area and volume-consuming storage systems are located at the bottom left. Here it is again clear
that PHS, CAES and flow batteries have a low energy density compared to other storage technologies. SMES, DLC
and FES have high power densities but low energy densities. Li-ion has both a high energy density and high power
density, which explains the broad range of applications where Li-ion is currently deployed.

NaS and NaNiCl have higher energy densities in comparison to the mature battery types such as LA and NiCd, but
their power density is lower in comparison to NiMH and Li-ion. Metal air cells have the highest potential in terms
of energy density. Flow batteries have a high potential for larger battery systems (MW/MWh) but have only
moderate energy densities. The main advantage of H2 and SNG is the high energy density, superior to all other
storage systems.

Figure 4.2. Comparison of power density and energy density (in

relation to volume) of energy storage technologies

Figure 4.3 summarizes the maturity of the storage technologies discussed. The state of the art for each ESE
technology is plotted versus the power range. Thus the suitability for different applications of the available
technologies covered can be compared.

Clearly PHS, CAES, H2 and SNG are the only storage technologies available for high power ranges and energy
capacities, although energy density is rather low for PHS and CAES. Large power ranges are feasible as these
ESE systems use the turbines and compressors familiar from other power generation plants. However, only PHS
is mature and available. Restrictions in locations (topography) and land

5
consumption are a more severe limit for this technology than the characteristic of low energy density (although the
two may be linked in some cases). Figure 2-11 shows a lack of immediately deployable storage systems in the range
from 10 MW to some hundreds of MW. Diabatic CAES is well-developed but adiabatic CAES is yet to be
demonstrated. Single components of H2 and SNG storage systems are available and in some cases have been used in
industrial applications for decades. However, such storage systems become viable and economically reasonable only
if the grids have to carry and distribute large amounts of volatile electricity from REs. The first demonstration and
pilot plants are currently under construction (e.g. in Europe).

From the technical comparison it can be concluded that a single universal storage technology superior to all other
storage systems does not exist. Today and in the future different types of ESE will be necessary to suit all the
applications described in section 1. Bearing in mind the findings from Figures 4-1 and 4-2, Figure 4.3 suggests the
following conclusions.

1) ESE systems for short and medium discharge times cover wide ranges of rated power and energy density.
Several mature ESE technologies, in particular FES, DLC and battery systems, can be used in these ranges.

2) PHS is the only currently feasible large capacity ESE for medium discharge times; further development in CAES
is expected. Suitable locations for large PHS and CAES systems are topographically limited. An increase in the
capacity of other ESE systems, and control and integration of dispersed ESE systems (see section 3.3), will be
required for medium-duration use.

3) For long discharge times, days to months, and huge capacities (GWh - TWh), no ESE technologies have so
far been put into practical operation. New ESE technologies such as H 2 and SNG have to be developed.

Figure 4.3. Maturity and state of the art of storage systems for
electrical energy

6
 Module – V: Applications of Energy Storage systems

In this section an overview of the markets for ESE is given by describing existing ESE application cases.
Applications for conventional electric utilities and consumers are presented as well as near-future use cases,
concentrating on storage applications in combination with renewable energy generation.

1. Present status of applications

In this section, those cases are described which have already been implemented by electric utilities and
consumers. These are respectively time shift and investment deferral for the former, and emergency supply and
power quality for the latter.

2. Utility use (conventional power generation, grid operation & service)

1) Reduce total generation costs by using pumped hydroelectricity for time shifting, which stores electricity
during off-peak times and provides electricity during peak hours. 2) Maintain power quality, voltage and
frequency, by supplying/absorbing power from/into ESE when necessary.
3) Postpone investment needed by mitigating network congestion through peak shift.
4) Provide stable power for off-grid systems (isolated networks).
5) Provide emergency power supply.

Utility use of pumped hydro storage for time shift and power quality

Pumped hydro storage (PHS) has historically been used by electric utilities to reduce total generation cost by time-
shifting and to control grid frequency. There are many PHS facilities in different countries, and they have the largest
proportion of total storage capacity worldwide. A conventional installation cannot function as a frequency controller
while pumping, butan advanced variable-speed-control PHS (Figure 3-
1) can do so by varying the rotational speed of the motor.

Figure 5.1. Variable-speed PHS operated by TEPCO

Utility use of compressed air energy storage for time shift and power quality

Today only two diabatic compressed air energy storage (CAES) power plants are in operation worldwide. In 1978
the first CAES power plant was built in Huntorf, Germany (Figure 3-2). It works as a diabatic CAES plant with a
round-trip efficiency of roughly 41 % [25 ]. It consists of a low-pressure and high- pressure compressor with
intercooler, two salt caverns (2 x 155 000 m³ usable volume, 46 - 72 bar pressure range), a motor-generator (60 MW
charging, 321 MW discharging) and a high-pressure (inlet

7
conditions: 41 bar, 490 °C) and low-pressure turbine (13 bar, 945 °C). The second CAES plant is in McIntosh
(Alabama, USA) and was commissioned in 1991. It has a net electrical output of 110 MW and is also based on a
diabatic CAES process, but additionally a recuperator is used to recover heat from the exhaust at the outlet of the
gas turbine. Therefore a higher round trip efficiency of 54 % can be achieved. Both systems use off-peak electricity
for air compression and are operated for peak levelling on a daily basis.

Worldwide several CAES plants are under development and construction. In Germany for example a small
adiabatic CAES plant is scheduled for demonstration in 2016 (project ADELE), which will achieve a higher
efficiency in comparison to a diabatic CAES [23 ].

Figure 5.2. CASE plant in Huntorf.

Utility’s more efficient use of the power Network

As one of the examples of ESE for utilities, a Li-ion battery can provide the benefit of more efficient use of the
power network.

In 2009 the US companies AES Energy Storage and A123 Systems installed a 12 MW, 3 MWh Li-ion battery at
AES Gener's Los Andes substation in the Atacama Desert, Chile (Figure 3-3). The battery helps the system operator
manage fluctuations in demand, delivering frequency regulation in a less expensive and more responsive manner
than transmission line upgrades. In addition, because the project replaces unpaid reserve from the power plant, AES
Gener will receive payment for its full output capacity by selling directly to the electric grid.

8
 Figure 5.3. Li-ion battery supplying up to 12 MW of power at Los Andes substation in
Chile

Utility’s emergency power supply

Important facilities, such as power stations, substations and telecommunication stations, need power sources for their
control installations with high power quality and reliability, since these are the very facilities which are most needed
for power in the case of an interruption. ESE systems for this application are mostly DC sources and supported by
batteries. Historically lead acid batteries have been used for this purpose.

Utility’s off-grid systems (isolated grids)

In the case where a utility company supplies electricity in a small power grid, for example on an island, the power
output from small-capacity generators such as diesel and renewable energy must also match with the power
demand. On Hachijo-jima (island), where about 8 000 people live, TEPCO uses NaS batteries with diesel
generators and a wind power station to meet the varying demand. For off-grid photovoltaic systems in the power
range (50 W -) 1 kW - 500 kW lead acid batteries for ESE are commonly used.

3 Consumer use (uninterruptable power supply for large consumers)

1) Suppress peak demand and use cheaper electricity during peak periods, i.e. save cost by buying off- peak
electricity and storing it in ESE. The result is load leveling by timeshifting. 2) Secure a reliable and higher-quality
power supply for important factories and commercial facilities.

Example: consumers’ use of NaS batteries

Figure 5-4 shows the applications of NaS batteries installed in the world with their respective power capacities. The
systems used exclusively for load levelling (LL) account for almost half the total, and installations for load levelling
with the additional functions of emergency power supply or stand-by power supply represent another 20 % each.
However, the need for storage linked to renewable energy, as explained in section 3.2, is growing.

Figure 5-5 shows the locations of NaS batteries installed in the TEPCO service area; the average capacity per
location is about 2 MW. The majority of batteries are installed in large factories (64 %), but there are some in large
commercial buildings (19 %) as well as in water supply/sewerage systems and schools/ research institutes (12 %
together).

9
4 ESE installed capacity worldwide

Figure 3-6 shows the installed capacity of ESE systems used in electricity grids. Pumped hydro storage (PHS)
power plants, with over 127 GW, represent 99 %, and this is about 3 % of global generation capacity. The
secondlargest ESE in installed capacity is CAES, but there are only two systems in operation. The third most
widely-used ESE is the NaS battery. As of the end of September 2010, NaS systems were installed and
operational in 223 locations in, for example, Japan, Germany, France, USA and UAE (total: 316 MW). However,
a large quantity of other ESE is expected to be installed given the emerging market needs for different
applications, as shown in the next section.

Figure 5.4. NAS Battery applications and installed

capacities

Figure 5.5. Locations of NaS systems in

service area

10
 Figure 5.6. Worldwide installed storage capacity for electrical
energy

5. New trends in applications

Five new trends in ESE applications are described: renewable energy, smart grids, smart microgrids, smart
houses and electric vehicles. Current use cases of these applications include experimental equipment and plans.

5.1 Renewable energy generation

In order to solve global environmental problems, renewable energies such as solar and wind will be widely
used. This means that the future energy supply will be influenced by fluctuating renewable energy sources –
electricity production will follow weather conditions and the surplus and deficit in energy need to be balanced.
One of the main functions of energy storage, to match the supply and demand of energy (called time shifting),
is essential for large and small-scale applications. In the following, we show two cases classified by their size:
kWh class and MWh class.

Besides time shifting with energy storage, there are also other ways of matching supply and demand. With a
reinforced power grid, regional overproduction can be compensated for by energy transmission to temporarily less
productive areas. The amount of energy storage can also be reduced by overinstallation of renewable energy
generators. With this approach even weakly producing periods are adequate for the load expected.

A further option is so-called demand-side management (described under Smart Grid in section 3.2.2), where users
are encouraged to shift their consumption of electricity towards periods when surplus energy from renewables is
available.

These balancing methods not requiring ESE need to be considered for a proper forecast of the market potential
for ESE.

Decentralized storage systems for increased self-consumption of PV energy (kWh class)

With the increasing number of installed PV systems, the low-voltage grid is reaching its performance limit. In
Germany, the EEG (Renewable Energies Law) guarantESE, for a period of 20 years, a feed-in tariff for every kWh
produced and a fixed tariff for every kWh produced and self-consumed. To encourage

11
operators of decentralized systems, the price for self-consumed PV energy is higher. Therefore self- consumption of
power will become an important option for private households with PV facilities, especially as the price of
electricity increases.

Figure 5-7 shows an example of system design. To measure the amount of energy consumed or fed into the grid two
meters are needed. One meter measures the energy generated by the PV system. The other meter works
bidirectionally and measures the energy obtained from or supplied to the grid. The generated energy that is not
immediately consumed is stored in the battery.

Figure 5.7. Photovoltaic system designed for energy self-

consumption

In order to examine how much electricity can be self-supplied from PV, the results from a simulation for a typical
household in Madrid may be of interest [26]. The total consumption of the household over one year is about 3 400
kWh. The aim is to use as much energy internally as possible, with a 10.7 kW PV generator and a 6 kWh lithium
ion storage system. Figure 5-8 shows the electricity consumption of the household over a year. Regardless of the
time of energy production, the storage provides the energy generated by the PV generator to electrical appliances.
Supply and demand can be adjusted to each other. The integrated storage system is designed to cover 100 % of the
demand with the energy generated by the PV system during the summer. During the rest of the year a little
additional energy has to be purchased from the grid.

Figure 5. 8. Consumption of a
typical household with a
storage system: energy
consumed from the grid and
from the photovoltaic system.

12
To provide a consumer-friendly system at low cost, maintenance cost in particular needs to be low and the most
important factor for stationary batteries is still the price per kWh. Currently for this application lead acid
batteries are the most common technology because of the low investment costs. Lithium ion batteries are
generally better in efficiency and in the number of cycles, but they have much higher investment costs. NaNiCl
batteries are also an option for this application, but they need daily cycling to avoid additional heating.

Smoothing out for wind (and PV) energy (MWh class)

The Japan Wind Development Co. Ltd. Has constructed a wind power generation facility equipped with a battery in
Aomori, Japan (Futamata wind power plant, shown in Figures 3-9 and 3-10). This facility consists of 51 MW of
wind turbines (1 500 kW x 34 units) and 34 MW of NaS batteries (2 000 kW x 17 units). By using the NaS battery,
the total power output of this facility is smoothed and peak output is controlled to be no greater than 40 MW.
Operation started in June 2008.

Figure 5.9. General view of the Futamata wind power plant in

Japan

Figure 5.10. NaS battery unit – 34 MW in

Japan

13
Figure 5-11 shows an example of output from this facility. The electric power sales plan is predetermined one day
before. In order to achieve this plan, the NaS battery system controls charging or discharging in accordance with the
output of wind power generation. This facility meets the technical requirements of the local utility company to
connect to the grid.

Figure 5.11. Example operational results of constant output control over 8 hours
(NGK)
5.2 Smart Grid

Today’s grids are generally based on large central power plants connected to high-voltage transmission systems
that supply power to medium and low-voltage distribution systems. The power flow is in one direction only: from
the power stations, via the transmission and distribution grid, to the final consumers. Dispatching of power and
network control is typically conducted by centralized facilities and there is little or no consumer participation.

For the future distribution system, grids will become more active and will have to accommodate bi- directional
power flows and an increasing transmission of information. Some of the electricity generated by large conventional
plants will be displaced by the integration of renewable energy sources. An increasing number of PV, biomass and
on-shore wind generators will feed into the medium and low voltage grid. Conventional electricity systems must be
transformed in the framework of a market model in which generation is dispatched according to market forces and
the grid control center undertakes an overall supervisory role (active power balancing and ancillary services such as
voltage control).

The Smart Grid concept (Figure 5-12) is proposed as one of the measures to solve problems in such
a system. The Smart Grid is expected to control the demand side as well as the generation side, so that the overall
power system can be more efficiently and rationally operated. The Smart Grid includes many technologies such as
IT and communications, control technologies and ESE. Examples of ESE-relevant applications in the Smart Grid
are given below.

14
Figure 5.12. The Smart
GrGrid

1) Penetration of renewable energy requires more frequency control capability in the power system. ESE can be
used to enhance the capability through the control of charging and discharging from network operators, so that the
imbalance between power consumption and generation is lessened.

2) In some cases, ESE can reduce investment in power system infrastructure such as transformers, transmission
lines and distribution lines through load levelling in certain areas at times of peak demand. ESE for this purpose
may also be used to enhance frequency control capability.

3) A further option is so-called demand-side management, involving smart grids and residential users. With
intelligent consumption management and economic incentives consumers can be encouraged to shift their energy
buying towards periods when surplus power is available. Users may accomplish this shift by changing when they
need electricity, by buying and storing electricity for later use when they do not need it, or both.

Electrochemical storage types used in smart grids are basically lead acid and NaS batteries, and in some cases also
Li-ion batteries. For this application redox flow batteries also have potential because of their independent ratio of
power and energy, leading to cost-efficient storage solutions.

5.3 Smart Microgrid

A smart factory, smart building, smart hospital, smart store or another intermediate-level grid with ESE may be
treated as a “Smart Microgrid” 8. For flexibility in resisting outages caused by disasters it is very important to
deploy Smart Microgrids, that is, distributed smart power sources, as an element in constructing smart grids.

ESE is an essential component of a Smart Microgrid, which should be scalable, autonomous and ready to cooperate
with other grids. The architecture for the Smart Microgrid should have a single controller and

15
should be scalable with respect to ESE, i.e. it should adjust smoothly to the expansion and shrinkage of ESE
(battery) capacities according to the application in for example a factory, a building, a hospital or a store. The
microgrid and ESE should in general be connected to the network; even if a particular Smart Microgrid is not
connected to a grid, for example in the case of an isolated island, it should still have similar possibilities of
intelligent adjustment, because an isolated Smart Microgrid can also expand or shrink. Figure 5-13 shows a
schematic of a scalable architecture.

In Annex B two examples are given, a factory and a store, which have fairly different sizes of batteries, but with
controllers in common. Microgrids controlled in this way have the features of connecting and adjusting to the
main grid intelligently, showing and using the input and output status of batteries, and controlling power
smoothly in an emergency (including isolating the microgrid from the main grid if needed). These are the
characteristics needed in Smart Microgrids, regardless of ESE scale or applications.

Figure 5.13. Scalable architecture for energy storage applications in a smart microgrid
(Sanyo)

5.4 Smart House

The concept of the Smart House is proposed in order to use energy more efficiently, economically and reliably
in residential areas. ESE technologies are expected to play an important role.

1) The consumer cost of electricity consists of a demand charge (kW) and an energy charge (kWh). Load levelling
by ESE can suppress the peak demand; however, charge/discharge loss will simultaneously increase the amount of
electricity consumed. Consumers may be able to reduce electricity costs by optimizing ESE operation.

2) Some consumers prefer to use their own renewable energy sources. ESE can reduce the mismatch between
their power demand and their own power generation.

16
3) In specific situations such as interruption of power supply, most on-site renewable generators have problems in
isolated operation because of the uncontrollable generation output. ESE may be a solution.

Figure 5-14 schematically represents the smart house, and Figure 5-15 maps a possible energy architecture for it. In
smart houses mainly lead acid systems are used currently, but in the future Li-ion or NaNiCl batteries in particular
may be installed because of their high cycle lifetime and their ability to deliver high peak power.

Figure 5.14. The Smart House

Figure 5.15. Future home energy network in a smart

Ghouse

17
5.5 Electric vehicles

Electric vehicles (EVs) were first developed in the 19th century but, since vehicles with conventional combustion
engines are much cheaper and have other advantages such as an adequate driving range of around 500 km, electric
vehicles have not been introduced in large quantities to the market. The main obstacle for building electricity-driven
vehicles has been the storage of energy in batteries. Due to their low capacity it has not been possible to achieve
driving ranges that would be accepted by the consumer. The emerging development of battery technology in recent
years presents new possibilities, with batteries displaying increased energy densities.

In the transitional period of the next few years, mainly hybrid cars will come onto the market. They combine an
internal combustion engine with an electric motor, so that one system is able to compensate for the disadvantages
of the other. An example is the low efficiency in partial-load states of an internal combustion engine, which can be
compensated for by the electric motor. Electric drive-trains are particularly well suited to road vehicles due to their
precise response behavior, their high efficiency and the relatively simple handling of the energy storage. In spite of
the advantages of electric motors, the combination of an electric drive-train with an internal combustion engine is
reasonable. That is because electricity storage for driving ranges of up to 500 km, which are achieved by
conventional drive-trains (and petrol tanks), are not feasible today.

Hybrid classes and vehicle batteries

Generally the different hybrid vehicles are classified by their integrated functions, as shown in Figure 5-
16. The power demand on the battery increases with additional integrated functions. The more functions are
integrated in the vehicle, the higher the potential of fuel savings and therefore the reduction of carbon dioxide
emissions. While vehicles up to the full hybrid level have already entered the market, plug-in hybrids and pure
electric vehicles are not yet established in larger quantities.

Figure 5.16. Hybrid classes sorted by electrical power and functional range,
against stage of development

Regarding energy storage for vehicles, today lead acid batteries are commonly used in microhybrids. In
combination with a double-layer capacitor there might also be options for their use in mild or full hybrids, but
since technically better solutions are available and economically feasible they will not play any role in the future.

18
NiMH batteries are mainly used in hybrid vehicles because their system is well-engineered and, compared to Li-ion
batteries, they are actually more favorable especially due to safety issues. Good cycle stability in low states of
charge which often appear in hybrid cars is characteristic for these batteries. All Toyota hybrid vehicles use a NiMH
battery with 1.3 kWh and 40 kW. Toyota has sold in total about 3 million hybrid vehicles with this battery; this
means the total storage volume sold is about 4 GWh and 120 GW.

A major problem of this technology is the limited potential for further technical or economic improvements.
With lithium ion batteries becoming technically more favorable and having significant potential for cost
reduction there does not seem to be a medium-term future for NiMH batteries.

Lithium batteries are ideally suited for automotive use, for both electric vehicles and hybrid electric vehicles. For
the hybrid vehicles a good choice might be the lithium-titanate battery because of its high cycle stability and power
density. With rising battery capacities for more advanced hybrid types, the relatively low energy density of the
lithium-titanate batteries has a bigger effect on the total car weight that results in a higher energy demand. Therefore
lithium-iron-phosphate and especially lithium- NMC batteries with high energy densities are preferred for plug-in-
hybrids and pure electric cars – for the latter the driving range is the most
important criterion.

An alternative battery technology for pure electric cars is the high-temperature sodium-nickel- chloride battery
(also called ZEBRA battery). It has a huge self-discharge rate of about 10 % per day in stand-by status from
having to keep the battery at a high temperature. Therefore these NaNiCl batteries are preferred for fleet vehicles
such as buses, where they are in permanent operation and no additional battery heating is usually necessary.

5.6 Management and control hierarchy of storage systems

In this section the concepts of the management and control of storage systems are introduced. While it is essential
to have local management for the safe and reliable operation of the storage facilities, it is equally important to have
a coordinated control with other components in the grid when grid-wide applications are desired. The purpose of
this section is to help readers visualize
the components and their interactions for some of the applications described in this paper. Many storage systems are
connected to the grid via power electronics components, including the converter which modulates the waveforms of
current and voltage to a level that can be fed into or taken from the grid directly. Sometimes the converter is
connected to a transformer before the grid connection in order to provide the required voltage. The converter is
managed by a controller which defines the set-points of the storage system. These set-points can be expressed as the
magnitude of active and reactive power, P and
Q. Such a controller may also be called control electronics – a controller in this context is simply a representation of the
place where intelligence for decision-making is applied.

5.6.1 Internal configuration of battery storage systems

Complex storage systems consisting of batteries are equipped with a Battery Management System (BMS) which
monitors and controls the charge and discharge processes of the cells or modules of the batteries. This is necessary
in order to safeguard the lifetime and ensure safe operation of the batteries. The diagram in Figure 5-17 shows a
possible realization of the internal control architecture for a battery storage system. It should be noted that for bulk
energy storage it is very likely that there is a more refined hierarchy for the BMS, which involves a master control
module coordinating the charging and discharging of the slave control modules. It is possible that the batteries and
converters are from two different manufacturers, and therefore compatibility and interoperability of the two systems
regarding both communication and electrical connections is imperative.

19
 Table 5.1. Differences between hybrid and electric vehicle
powertrains

Figure 5.17. A possible realization of internal control architecture for a battery storage system
(ABB)

5.6.2 External connection of ESE systems

The P and Q set-points for an ESE for a certain application can be set locally or remotely, depending on the control
scheme implemented. The control scheme should in turn be determined by the application. More precisely, the
application determines the algorithmic and input/output requirements for the ESE system. For instance, an
application which requires simple logic using only local measurements can have the set-points determined locally
through the storage controller. An example of such an application is load levelling, which only needs to know the
loading conditions of the local equipment (e.g. lines, transformers) next to which the ESE is installed. The same
applies for applications which have predetermined set- points that do not change during operation. However, set-
points for applications which require dynamic adaptation to the network operational environment and much remote
data or measurements might be better determined by a remote controller which can gather these remote inputs more
efficiently. One example of such an application is wind power smoothing, which uses wind output forecasts as well
as measurements from the wind farm as inputs.

Another example is energy time-shifting, making use of dynamic market prices. A generalized setup with remotely
determined set-points is shown in Figure 5-18. Batteries and the BMS are replaced by the

20
 “Energy Storage Medium”, to represent any storage technologies including the necessary energy
conversion subsystem.

Figure 5.18. A Control hierarchy involving remote data/measurements

(ABB)

5.6.3 Aggregating ESE systems and distributed generation (Virtual Power Plant)

The control hierarchy can be further generalized to include other storage systems or devices connected to the grid,
illustrated in Figure 5-19. This diagram represents an aggregation of ESE systems and DGs (Distributed
Generators) which can behave like one entity, a so-called “VPP with ESE” in this example. VPP stands for
Virtual Power Plant which, according to one definition, is the technology to aggregate power production from a
cluster of grid-connected distributed generation sources via smart grid technology, by a centralized controller
which can be hosted in a network control center or a major substation. The integration of distributed energy
storage systems at different locations of the grid will further enhance the capabilities of the VPP. It should be
noted that in the figure the communication and electrical infrastructures are highly simplified in order to show the
general concept but not the details.

A concrete example of an implementation based on aggregated energy storage systems using batteries is given in
the following section on “battery SCADA”.

Figure 5.19. A generalized control concept for aggregated energy systems and
DGs

21
5.6.4 “Battery SCADA” – aggregation of many dispersed batteries

As progress is realized in battery capabilities and costs, many batteries will be installed both by consumers and in
the grid, with large cumulative capacity and correspondingly large effects. Most will be small battery storage
systems, dispersed in location and used locally. However, if they are gathered into a virtual assembly and controlled
centrally, they may also be used for many utility applications, such as load frequency control, load levelling and
control of transmission power flow. To implement such uses, a group of battery manufacturers and electric utilities
in Japan is developing technologies for central control of dispersed batteries, named “Battery SCADA”. Using
Battery SCADA distributed batteries can be assembled and managed like a virtual large capacity battery, and
batteries with different specifications made by different manufacturers can be controlled and used by grid operators
in an integrated way.
Battery SCADA is shown schematically in Figure 5-20. Information from batteries on both the grid side and the
customer side is collected by Battery SCADA, processed, and transmitted to the control center. Based on this
information and the situation of the network, the control center sends commands to Battery SCADA, which
distributes corresponding commands to each battery system.

Demonstration of this technology will start in 2012 in Yokohama City, Japan, with various types of Li-ion
batteries installed on the grid side and in consumer premises, to be controlled by the Battery SCADA.

Figure 5.20. Schematic diagram of battery

SCADA.

22
References
1. C. Dötsch: Electrical energy storage from 100 kW – State of the art technologies, fields of use. 2nd
, 2nd Int. Renewable Energy Storage Conference, Bonn/Germany, 22 Nov 2007.

2. T. Fujihara, et al.: Development of pump turbine for seawater pumped-storage power plant,
Hitachi Review 47 (5), pp. 199-202, 1998.

3. M. Nakhamkin: Novel Compressed Air Energy Storage Concepts Developed by ESPC, ESEAT, May
2007.

4. VDE - ETG Energy Storage Task Force: Energy storage in power supply systems with a high share of
renewable energy sources Significance - state of the art - need for action, Report, Dec 2008.

5. T. Smolinka, et al.: Stand und Entwicklungspotenzial der Speichertechniken für Elektroenergie –

Ableitung von Anforderungen an und Auswirkungen auf die Investitionsgüterindustrie, BMWi-
uftragsstudie 08/28, 2009.

6. M. Dahlen, et al: Nickel Batteries, INVESTIRE 2003, [per04].

7. B. Espinar, D. Mayer: The role of energy storage for mini-grid stabilization, Report, IEAPVPS T11-
0X:2011, 2011.

8. B. Worth: Metal/Air, INVESTIRE 2002, [per04].

9. T. B. Atwater and Arthur Dobley, Metal/Air batteries, Lindens Handbook of Batteries, 2011, ISBN 978-0-
07-162421-X.

10. M. Kawashima: Overview of Electric Power Storage, Internal paper of Tepco, 2011.

11. A. Jossen: Redox-Flow Batterien – Ein System zur Langzeitspeicherung, Forum Elektrische
Energiespeicher - Netzoptimierung bei regenerativer Stromerzeugung, Nürnberg/Germany, 6 Dec 2007.

12. P. Patel: [Link] Article, May

2010, Accessed 10 April 2011.

13. Webpage Enertrag: [Link] html,

Accessed: 8 Aug 2011.

14. E. Kuhnhenn, J. Ecke: Power-to-gas Stromspeicher, Gasproduktion, Biomethan oder flexible Last,
DVGW Energie / Wasserpraxis 7/8, 2011.

15. Michael Sterner: Bioenergy and renewable power methane in integrated 100% renewable energy systems -
Limiting global warming by transforming energy systems, Dissertation, University Kassel, July 2009.

16. M. Waidhas: Elektrolysetechnologie aus Sicht eines Großunternehmens, NOW Workshop Stand und
Entwicklungspotenzial der Wasserelektrolyse, Berlin/Germany, 9 May 2011.

17. International Energy Agency: Prospects for Large Scale Energy Storage in Decarbonized Grids,
Report, 2009.

23
18. P. Schossig: Thermal Energy Storage, 3rd International Renewable Energy Storage Conference,
Berlin/Germany, 24-25 November 2008.

19. P. Fairley: [Link] to- start-

up, Article 2008, Accessed 27 July 2011.

20. D. Jähnig, et al.: Thermo-chemical storage for solar space heating in a single-family house, 10th
International Conference on Thermal Energy Storage: Ecostock 2006, 31 May - 2 June 2006, New Jersey,
USA.

21. R. Tamme: Development of Storage Systems for SP Plants, DG TREN - DG RTD Consultative Seminar
“Concentrating Solar Power”, Brussels/Belgium, 27 June 2006.

22. Webpage CSP Today: [Link]

generationparabolic-troughs-what-does-endesa-dlr-have-store,Accessed 4 Aug 2011.

23. Webpage RWE ADELE:

[Link]/web/cms/en/365478/rwe/innovations/powergeneration/energy-
storage/compressed-air-energy-storage/project-adele, Accessed 8 April 2011.

24. C. Bullough: Advanced Adiabatic Compressed Air Energy Storage for the Integration of Wind Energy
European, Wind Energy Conference and Exhibition, London, 22-25 Nov 2004.

25. P. Radgen: 30 Years Compressed Air Energy Storage Plant Huntorf - Experiences andOutlook,
3rd International Renewable Energy Storage Conference, Berlin/Germany, 24-25 November 2008.

26. Simon Schwunk: Battery systems for storing renewable energy, Report, Fraunhofer-Institutfür Solare
Energie, April 2011.

24