Energy Storage

Dani Strickland
d.strickland@lboro.ac.uk

1
Table of Contents

1 Aims and Objectives........................................................................................................................ 3

2 Introduction to Energy Storage....................................................................................................... 4
2.1 Background and definitions .................................................................................................... 4
2.2 History of Energy Storage ....................................................................................................... 6
2.3 Different Energy Storage types ............................................................................................... 6
2.4 Ragone plots ........................................................................................................................... 8
2.5 Summary ............................................................................................................................... 17
3 Energy storage system safety ....................................................................................................... 18
3.1 Summary ............................................................................................................................... 26
4 Components of an energy storage system ................................................................................... 27
4.1 Groundworks ........................................................................................................................ 27
4.2 Grid connection..................................................................................................................... 29
4.3 Control and communications systems .................................................................................. 31
4.4 Balance of plant .................................................................................................................... 34
4.5 Power supplies and cabling for auxiliary plant ..................................................................... 35
5 Role of energy storage .................................................................................................................. 37
5.1 Small scale – increasing self-consumption............................................................................ 37
5.2 Small scale – UPS/CPS applications....................................................................................... 43
5.3 Large scale systems – arbitrage ............................................................................................ 44
5.4 Large scale – ancillary services.............................................................................................. 47
5.4.1 STOR .............................................................................................................................. 48
5.4.2 FFR service .................................................................................................................... 49
5.4.3 EFR................................................................................................................................. 51
5.5 Large scale - Network deferral .............................................................................................. 55
5.6 Large scale – peak load/generation lopping ......................................................................... 62
5.7 Summary ............................................................................................................................... 64
6 Appendix A : Reduction potentials................................................................................................ 65

2
1 Aims and Objectives

• Describe the characteristics of different types of energy storage

• Mechanical

• Thermal

• Electrical

• Chemical

• Biological

• Understand the components of the energy storage systems

• Understand the safety implications of different types of energy storage

• Compare the performance and operation of different types of storage systems

• Describe where Energy Storage systems could be used and how to integrate them into the grid

3
2 Introduction to Energy Storage

2.1 Background and definitions

Term Meaning

Nominal Provided by the manufacturer - describes how much energy can be delivered between
capacity maximum and minimum values (where the minimum value may not be fully empty).

This may vary with external conditions (for example temperature)

This may reduce with age

Units – Joules (or variations such as kWh)

Charge rate Provided by the manufacturer - describes the rate at which the energy storage may be
“charged” by adding energy to the system

This may vary with external conditions

This may impact capacity

Units – Watts (or variations such as Joules/s )

Discharge rate Provided by the manufacturer - describes the rate at which the energy storage may be
“discharged” by providing energy to the system

As per charge rate

SOC – state of A value between 0% and 100% indicating the proportion of available energy
charge

DOD – depth of A value between 0% and 100% equal to the amount of energy that has been used.
discharge Equal to 100%-SOC

4
 Energy (J)
Max
energy

Min
energy

Min power Max Power

Power (W)

Figure 1: Ideal energy vs Power curve

Available energy is dependent on the power drawn and the energy storage system

100% SOC = 0% DOD

Energy available

Energy 0% SOC = 100% DOD

unavailable
Figure 2: Definition of terms used in these notes

Common formula from GCSE Physics

Equation (symbolic) Equation (words) Units

P=VxI power = voltage x current Watts
V=IxR voltage = current x resistance Volts
P=E/t power = energy ÷ time Watts
Q=Ixt charge = current x time Coulombs
E = VxIxt energy = voltage x current x time Joules
E=VxQ energy = voltage x charge Joules

5
Kirchhoff’s voltage law Kirchhoff’s current law

I1 I

I2
𝐼 = 𝐼1 + 𝐼2

For a closed loop series path the algebraic sum
of all the voltages around any closed loop in a
circuit is equal to zero. This is because a circuit
loop is a closed conducting path so no energy is
lost.
The total current entering a node is exactly
equal to the total current leaving the same
node. This is because it has no other place to go
as no charge is lost.

2.2 History of Energy Storage

Energy Storage has been around since the 15th Century when mechanical energy storage was first
used to provide power to watches and clocks. Since then key developments have included;

 1859 – First rechargeable battery

 1930 – First pumped hydro storage scheme
 1991 – First commercial Lithium Ion battery

Developments are still continuing and it is unlikely that research into energy storage will stop any
time soon.

2.3 Different Energy Storage types

Type Symbol used in these notes

Fossil Fuel No longer consider environmentally friendly

Mechanical
 Pumped Storage/hydroelectricity

 Flywheel

6
  Compressed air

Thermal

Chemical
 Electrochemical

 Other chemical

Electrical

Biological

Different types of Energy storage systems come in different sizes and last for different periods of
time

Figure 3: Power and Time of different Energy storage types

Energy storage devices

 Respond in different time scales

 Respond for different lengths of time
 Have different storage capacities
 Have different costs

7
In some cases it is necessary to consider more than one energy storage device in a system to meet
requirements as part of a hybrid scheme.

Figure 4: Efficiency

𝐸𝑛𝑒𝑟𝑔𝑦 𝑟𝑒𝑐𝑜𝑣𝑒𝑟𝑒𝑑
𝜂= 𝑥100%
𝐸𝑛𝑒𝑟𝑔𝑦 𝑖𝑛𝑝𝑢𝑡

2.4 Ragone plots

Consider a battery system as shown below:

Current, I
Internal
resistance,
Terminal
R
voltage, V
Cell Constant
voltage, power load,
Voc P

Figure 5: Battery with a constant power load

8
Assume

Internal Resistance= R

Initial Capacity = Qo

Constant cell voltage = Voc if Qo > Q > 0

Also called the open circuit voltage

At the battery

𝑉 = 𝑉𝑜𝑐 − 𝐼𝑅 (1)

At the load

P= 𝑉𝐼 (2)

Substituting (2) into (1) gives a quadratic in I

𝑃
V= → 𝑃 = 𝐼(𝑉𝑜𝑐 − 𝐼𝑅) → 𝑃 = 𝐼𝑉𝑜𝑐 − 𝐼 2 𝑅 → 𝐼 2 𝑅 − 𝐼𝑉𝑜𝑐 + 𝑃 = 0
𝐼

Solving for I gives

−𝑏 ± √𝑏 2 − 4𝑎𝑐
𝑥=
2𝑎

2
𝑉𝑜𝑐 ±√𝑉𝑜𝑐 − 4𝑃𝑅
𝐼=
2𝑅 2𝑅

𝑉𝑜𝑐
As P → 0, 𝐼→ 𝑜𝑟 𝐼 → 0
𝑅

• Charge (Q) is measured in Coulombs (C)

• Each electron carries a small amount of charge (1.6x10-19C)

𝑐ℎ𝑎𝑟𝑔𝑒(𝐶) = 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 (𝐴) 𝑥 𝑡𝑖𝑚𝑒(𝑠)

The battery empties after a finite amount of time

𝑄𝑜
𝑡=
𝐼

The energy from the battery

9
 𝑄𝑜 2𝑃𝑄𝑜 𝑅
𝐸 = 𝑃𝑡 = 𝑃 =
𝐼
𝑉𝑜𝑐 − √𝑉𝑜𝑐 2 − 4𝑃𝑅

This is an equation relating Energy provided by the battery with power

Maximum P at √𝑉𝑜𝑐 2 − 4𝑃𝑅=0

𝑉𝑜𝑐 2
𝑃𝑚𝑎𝑥 = 4𝑅

The equation for Energy against Power can be plotted and is known as the Ragone Plot.

• For an ideal battery with no leakage the Ragone plot looks as follows

Example

Qo=100C
Voc=10V
R = 10m

Ragone plot - ideal battery

1200

1000

800
Energy (Joules)

600

400

200

0
0.0001 0.001 0.01 0.1 1 10 100 1000 10000
Power (Watts)

Figure 6: Ragone plot for ideal battery

10
This is frequently normalised by dividing Energy by the starting energy of the system and Power by
the maximum power where:

𝐸𝑜 = 𝑄𝑜 𝑉𝑜𝑐

𝑉𝑜𝑐 2
𝑃𝑚𝑎𝑥 =
4𝑅

Giving the following curve

Normalised Ragone plot - ideal battery

1.20

1.00

0.80
Energy (E/Eo)

0.60

0.40

0.20

0.00
0.0000001 0.000001 0.00001 0.0001 0.001 0.01 0.1 1
Power (P/Pmax)

Figure 7:Normalised Ragone plot for ideal battery

The battery systems are rarely ideally and often there is an element of self leakage. This can be
represented in the model by a load across the terminals of the battery as shown below.

11
 Current, I
Internal
resistance,
R Terminal
voltage, V
Cell Constant
voltage, power load,
RL P
Voc

Figure 8: Battery with a constant power load and self leakage component

Where Leakage Resistance RL

The leakage resistance increases the current from the battery by

𝑉𝑜𝑐
𝐼𝑏 = 𝐼 + (3)
𝑅𝐿

The battery empties after a finite amount of time

𝑄𝑜
𝑡=
𝐼𝑏

Therefore the energy from the battery is:

𝑄𝑜 𝑃𝑄𝑜
𝐸 = 𝑃𝑡 = 𝑃 =
𝐼𝑏
𝑉𝑜𝑐 − √𝑉𝑜𝑐 2 − 4𝑃𝑅
𝑉
+ 𝑅𝑜𝑐
2𝑅 𝐿

2𝑅𝑃𝑄𝑜
𝐸=
2𝑅𝑉
𝑉𝑜𝑐 − √𝑉𝑜𝑐 2 − 4𝑃𝑅 + 𝑅 𝑜𝑐
𝐿

There is now an extra term due to the battery leakage. This can be re-plotted on the Ragone plot
example used before where RL=1000

12
 Ragone plot - with leakage
1200

1000

800
Energy (Joules)

600

400

200

0
0.0001 0.001 0.01 0.1 1 10 100 1000 10000
Power (Watts)

Figure 9: Ragone plot for battery with self leakage

When this is normalised the following curve is obtained

13
 Normalised Ragone plot - with leakage
1.20

1.00

0.80
Energy (E/Eo)

0.60

0.40

0.20

0.00
0.0000001 0.000001 0.00001 0.0001 0.001 0.01 0.1 1
Power (P/Pmax)

Figure 10: Normalised Ragone plot for battery with self leakage

These plots are dependent on the values of R and RL.

Figure 11: Ragone plot for battery with difference resistances

14
It is important therefore to ensure that the energy storage system is correctly designed as operating
at too low a power or too high a power results in reduced available energy.

Figure 12: Ragone plot for battery showing poor operating conditions

The Ragone plot isn’t just for battery systems but can be used for all forms of energy storage. The
easiest way to approach this is to treat the different systems as analogous and then apply the same
conditions.

15
Figure 13: Analogous physical systems

The general case is as follows:

Energy storage device

Constant
power load

Figure 14: General case

16
General form of the equations

𝛿 2𝑋 𝛿𝑋 −𝑃
𝑎 +𝑏 + 𝑐(𝑋) =
𝛿𝑡 2 𝛿𝑡 𝛿𝑋
𝛿𝑡

𝐸(𝑃) = 𝑃𝑡
And

The most straightforward case (the ideal battery ) can be found from this equation where a=0

𝛿𝑋 −𝑃 𝛿𝑄 −𝑃 −𝑃
𝑏 𝛿𝑡 + 𝑐(𝑋) = 𝛿𝑋 → 𝑅 𝛿𝑡 − 𝑉𝑜𝑐 = 𝛿𝑄 → 𝑅𝐼 − 𝑉𝑜𝑐 = 𝐼
𝛿𝑡 𝛿𝑡

The addition of an a term indicates the presences of an inductor

𝛿 2𝑋 𝛿𝑋 −𝑃 𝛿 2𝑄 𝛿𝑄 −𝑃 𝛿𝐼
𝑎 2 +𝑏 + 𝑐(𝑋) = →𝐿 2 +𝑅 − 𝑉𝑜𝑐 = → 𝐿 + 𝑅𝐼 − 𝑉𝑜𝑐
𝛿𝑡 𝛿𝑡 𝛿𝑋 𝛿𝑡 𝛿𝑡 𝛿𝑄 𝛿𝑡
𝛿𝑡 𝛿𝑡
−𝑃
=
𝐼 for a flywheel for example
A similar analogy can be used

𝛿2 𝑋 𝛿𝑋 −𝑃 𝛿2 𝜃 𝛿𝜃 −𝑃 𝛿𝜔 −𝑃
𝑎 𝛿𝑡 2 + 𝑏 𝛿𝑡 + 𝑐(𝑋) = 𝛿𝑋 → 𝐽 𝛿𝑡 2 + 𝐵𝑟 𝛿𝑡 − 𝜏 = 𝛿𝜃 →𝐽 + 𝐵𝑟 𝜔 − 𝜏 =
𝛿𝑡 𝜔
𝛿𝑡 𝛿𝑡

2.5 Summary

• The energy storage system has an efficiency based on losses, leakage and the auxiliary
power requirements

• The energy available for recovery is also based on the power that is drawn from the system

• Ragone plots can help to visualise the energy/power to ensure good energy recovery is
matched to power level

17
3 Energy storage system safety

All engineers are bound by the health and safety at work act (1974) and the electricity at work
regulations (1989). These regulations require companies and personnel to think about safe working
practices. In most cases a system should be worked on in a “dead” state. This requires the system to
be isolated, locked off and caution notices applied normally under some form of permit scheme such
as a “permit to work”. However, when considering Energy storage this is not always reasonably
practicable. In some cases fully discharging a stored energy device (eg some forms of battery) can
damage the device. This means that the stored energy can contain energy when it is being worked
on. The HSE (Health and Safety Executive) in the UK publish some guidelines on safe working
practices HSG85 that recommend good working practice. This follow a flow chart based process that
determines if it is feasible to work live.

For example, let’s consider that it is necessary to do maintenance on the contactor which connects
the energy storage system to an inverter. The process would be as shown in Figure 15. The first
decision process is listed as deciding whether to work live or dead.

Figure 15: Flow chart for safe working practice

18
Normally the contactor would be used as a point of isolation for work on the inverter and therefore,
it would normally be locked open. However, in this case, the battery is connected directly to the
contactor. It may not be possible to fully discharge the battery without causing damage to it and
therefore it is unreasonable to work dead. The risk needs to be identified and methods for
controlling these risks such that there is no risk of injury needs to be established.

Figure 16: Flow chart for safe working practice

Within an energy storage system the following types of risks as shown in Table 1 (this list is not
exhaustive and every situation is different) may exist. One of the more dangerous of these hazards is
that related to arc flash (or arcing).

Arc-flash is the light and heat produced from an electric arc with sufficient electrical energy to cause
substantial damage, harm, fire, or injury. Electrical arcs experience negative incremental resistance,

19
that is the electrical resistance decreases as the arc temperature increases. Therefore, as the arc
develops and gets hotter the resistance drops, drawing more and more current (runaway) until some
part of the system melts, trips, or evaporates, providing enough distance to break the circuit and
extinguish the arc.

Arc flash is a common risk in AC systems especially in the USA where the electricity grid tends to be
meshed and there are high fault currents available. Since large scale DC systems have been
uncommon up until recent years – there has not been so much work undertaken into these. To
calculate arc-flash in an ac system it is common to use IEEE 1584 standard. This states that it covers
208V to 15kV 3 phase ac systems. However large parts of this can be adapted for dealing with dc.

Care is needed in a dc system to account for the presence of both ac and dc. In this lecture course on
Energy storage – only dc will be considered as the more relevant and less well established. This is
used to highlight the safety considerations required for electrochemical storage but similar
considerations need to apply to all other types of storage.

Currently there is no international standard for calculating arc-flash for DC systems. A few methods
have been proposed for calculating DC Arc Flash. The main method is based on calculation of Arc
Resistance. The students should refer back to the standard for more information if required on ac
arc flash which is more established. Once the risk from the arc flash has been determined
appropriate PPE (Personal protective equipment) can be specified to help mitigate this risk. This
should be used in conjunction with other control measures such as screening of live terminals with
appropriate insulating material, no wearing of jewellery, having a second person to accompany the
first in the event of a live working situation, having a safety rescue hook available etc.

Type Danger
Electrical Accidental contact with Live Apparatus/Conductors which are operating
at Hazardous Voltages
Electrical Short circuiting live terminals or connections resulting in:
 Arcing – causing burns
 High intensity light resulting in temporary blindness
 Sparks that may result in combustion
 Molten metal being expelled – causing burns
 Chemical spill or release
Electrical Stored electrical energy which may result in Hazardous Voltages to
remain for a period after the equipment has been switched off
(batteries and capacitors).
Mechanical The release of stored mechanical energy
Mechanical Contact or entrapment with moving mechanical components
Chemical Exposure to toxic/corrosive/explosive chemicals
Chemical Chemicals and fumes from other energy sources present within the
work environment
Thermal Contact with substances or materials that contain stored thermal energy
at Hazardous Temperatures
Thermal The ignition of leaked fuel from other fuel systems
Table 1 : Example hazards present in an energy storage system

20
The Process for determining the PPE requirements needed for controlling against arc flash are as
follows:

• Determine the arc fault current

• Determine the fault current disconnection time through the protective devices

• Specify the working conditions in the fault locations

• Calculate the incident energy /Determine the flash protection boundary

• Determine the PPE requirements

Step 1: Determining the Fault current

Figure 17 shows a dc system with a battery, connected to a protection device leading to a short
circuit situation. The arc resistance is dependent on the arc current and the arc current is dependent
on the total resistance in the circuit (from ohms law). Therefore an iterative method needs to be
used to set the current then calculate the arc resistance then reset the current and so on until the
current and arc resistance converge to a constant value.

Figure 17: Battery connected dc system with a short circuit

There are two different methods of calculating the arc resistance from the arc current and the arcing
gap. These are both based on empirical test measurements done over 20 years ago. The two
methods are known as the Stoke and Oppenlander method and Paukert’s method. The first uses a
formula to calculate arc resistance while the latter uses a look up table with formula to calculate arc
resistance.

For example

Lets consider a 1000V battery with a 10mΩ internal resistance which is short circuited and has a
10mm electrode gap. The following calculations show the different methods in operation. An initial
arc current can be estimated by starting with the bolted short circuit current:

𝑉 1000
𝐼𝑎𝑟𝑐 = = = 100𝑘𝐴
𝑅𝑏 0.01

21
Figure 18: Battery connected dc system with a short circuit

Stoke and Oppenlander method:

The defined formula is

0.88
𝑅𝑎𝑟𝑐 = (20 + 0.534𝑍𝑔 )⁄𝐼𝑎𝑟𝑐

Where Zg is the arcing gap in mm

So in our example –

0.88
25.34
𝑅𝑎𝑟𝑐 = (20 + 0.534𝑍𝑔 )⁄𝐼𝑎𝑟𝑐 = = 0.001
25118

Then re-estimate the arc current

𝑉 1000
𝐼𝑎𝑟𝑐 = = = 90.8𝑘𝐴
𝑅𝑏 + 𝑅𝑎𝑟𝑐 0.011

Use this to recalculate the arc flash resistance

0.88
25.34
𝑅𝑎𝑟𝑐 = (20 + 0.534𝑍𝑔 )⁄𝐼𝑎𝑟𝑐 = = 0.0011
23073

Then re-estimate the arc current

𝑉 1000
𝐼𝑎𝑟𝑐 = = = 90.1𝑘𝐴
𝑅𝑏 + 𝑅𝑎𝑟𝑐 0.0111

Use this to recalculate the arc flash resistance

0.88
25.34
𝑅𝑎𝑟𝑐 = (20 + 0.534𝑍𝑔 )⁄𝐼𝑎𝑟𝑐 = = 0.0011
22917

22
The value of arc resistance is no longer changing significantly so the solution has converged.

Paukert method:

This uses a look up table approach, but the iterative process is similar.

Electrode Gap Arc Voltage (V) Arc Resistance (Ω)

(mm)
0.098 −0.902
1 13.04𝐼𝑎𝑟𝑐 13.04𝐼𝑎𝑟𝑐
0.211 −0.789
5 14.13𝐼𝑎𝑟𝑐 14.13𝐼𝑎𝑟𝑐
0.163 −0.837
10 16.68𝐼𝑎𝑟𝑐 16.68𝐼𝑎𝑟𝑐
0.190 −0.810
20 20.11𝐼𝑎𝑟𝑐 20.11𝐼𝑎𝑟𝑐
0.194 −0.806
50 28.35𝐼𝑎𝑟𝑐 28.35𝐼𝑎𝑟𝑐
0.241 −0.759
100 34.18𝐼𝑎𝑟𝑐 34.18𝐼𝑎𝑟𝑐
0.264 −0.736
200 52.63𝐼𝑎𝑟𝑐 52.63𝐼𝑎𝑟𝑐

For the 10mm arc gap:

−0.837
𝑅𝑎𝑟𝑐 = 16.68𝐼𝑎𝑟𝑐 = 0.0011
Arc current is therefore:

𝑉 1000
𝐼𝑎𝑟𝑐 = = = 90.1𝑘𝐴
𝑅𝑏 + 𝑅𝑎𝑟𝑐 0.0111

Leading to a re-estimation of arc resistance of:

−0.837
𝑅𝑎𝑟𝑐 = 16.68𝐼𝑎𝑟𝑐 = 0.0012

And so on

𝑉 1000
𝐼𝑎𝑟𝑐 = = = 89.3𝑘𝐴
𝑅𝑏 + 𝑅𝑎𝑟𝑐 0.0112
−0.837
𝑅𝑎𝑟𝑐 = 16.68𝐼𝑎𝑟𝑐 = 0.0012

The values between the two methods are not dissimilar.

23
Step 2 - Determine the fault current disconnection time through the protective devices

Each of the protective devices will have some form of protection curve associated with it. These may
be programmable (in the case of modern circuit breakers) or fixed (in the case of fuses) with
maximum trip times shown. The protection devices typically follow some form of inverse time curve
– so the higher the current the faster it takes to trip the device. This allows protection to be
coordinated around a system so that only the required protection operates.

Figure 19: Typical fuse trip curve

In this case – the fault current of around 90kA is assumed to trip the protection very quickly and a
value of 0.06s can be assumed for instantaneous operation.

Step 3/4 - Specify the working conditions in the fault locations and then calculate the incident
energy

It is important to specify where this fault is located as this will impact how the energy is released
from the blast. If the work is being undertaken in the open air, then the arc flash will expand as a
sphere. However, if the work is being undertaken within a panel (more likely) then the blast will be
more focussed towards the individual.

The Energy supplied by the blast is found from IEEE 1584 as (the 0.239 is a factor to turn kJ to cal)
2
𝐸𝑎𝑟𝑐 = 𝐼𝑎𝑟𝑐 𝑅𝑎𝑟𝑐 𝑡𝑎𝑟𝑐 𝐽𝑜𝑢𝑙𝑒𝑠
2
𝐸𝑎𝑟𝑐 = 0.239 𝐼𝑎𝑟𝑐 𝑅𝑎𝑟𝑐 𝑡𝑎𝑟𝑐 𝑐𝑎𝑙

24
If the work is being undertaken in air, then the energy of the arc in air is

𝐸𝑠 = 𝐸𝑎𝑟𝑐 ⁄4π𝑑2

Otherwise if the work is being undertaken on switchgear or in a panel – the following is used

𝐸1 = 𝑘(𝐸𝑎𝑟𝑐 ⁄𝑎2 + 𝑑2 )

Where the values of k and a are specified in the IEEE 1584 standard

Width Height a
Enclosure Depth mm k
mm mm mm
Panel board 305 356 191 100 0.127
LV
508 508 508 400 0.3142
Switchgear
MV
1143 762 762 950 0.416
Switchgear

Another common factor specified is the flash boundary. This is the distance at which the incident
energy equals 5 J/cm2 (1.2 cal/cm2). That is : no PPE is required outside of this boundary.

Figure 20: Arc flash boundary

Continuing the example above and assuming an arc flash current of 90kA and an arc flash resistance
of 0.0011Ω, the energy at 50cm (an arms length) can be calculated.
2
𝐸𝑎𝑟𝑐 = 0.239 𝐼𝑎𝑟𝑐 𝑅𝑎𝑟𝑐 𝑡𝑎𝑟𝑐 = 0.239𝑥90𝑥103 𝑥90𝑥103 𝑥0.0011𝑥0.06 = 127𝑘𝑐𝑎𝑙

In the open air – this equates to an energy of:

25
 127𝑥103
𝐸𝑠 = 𝐸𝑎𝑟𝑐 ⁄4π𝑑2 = (4𝑥3.14𝑥50𝑥50) = 4.0 𝑐𝑎𝑙 ⁄𝑐𝑚2

With an approach boundary of (1.2cal ) :

𝐸
𝑎𝑟𝑐 127𝑥103
√4𝜋𝐸 = √ 4𝜋1.2
= 92𝑐𝑚
𝑠

In a panel board – this equates to an energy of:

𝐸1 = 𝑘(𝐸𝑎𝑟𝑐 ⁄(𝑎2 + 𝑑2 )) = 0.127(127𝑥103 /(10x10+50x50))=6.2 Cal/𝑐𝑚2

With an approach boundary of:

𝑘𝐸𝑎𝑟𝑐 0.127𝑥127𝑥103
√ − 𝑎2 = √ − 100 = 115𝑐𝑚
𝐸𝑠 1.2

Step 5 – determine the hazard and PPE

The following table (NFPA 70E and IET Arc Flash Protection Health & Safety Briefing No. 51c) can be
used to determine the PPE

Incident Energy PPE details

2
Category 0 0 – 1.2 cal/cm 1 layer untreated cotton (covering body), polycarbonate
safety spectacles, lightweight cotton gloves
2
Category 1 1.21 – 4 cal/cm Cotton undergarments, 1-layer flame retardant (FR) work
wear, helmet, polycarbonate safety spectacles, lightweight
FR gloves
2
Category 2 4.1 – 8 cal/cm As above but with 2-layer FR outer work wear that has wrist
closures, and a full face polycarbonate visor. A FR single-layer
balaclava may also be worn
2
Category 3 8.1 – 25 cal/cm 3-layer FR outer work wear with cotton under garments and
FR shirt, a full-face hood or visor with safety spectacles,
chrome leather gauntlets

In our example, in air the hazard category is 1, whereas in a panel the hazard category is 2 and
additional flame retardant clothing would need to be worn when working on the system to prevent
risk of injury.

Note: Warning – there may be other sources of energy in a system eg capacitors in inverters or if the
system is grid connected then an infeed from the grid.

3.1 Summary

Students should understand the need for

• Safe systems of work

• Understand the risks and how to deal with them

26
 • Arc Flash calculations

• PPE requirements

4 Components of an energy storage system

A successful energy storage system is all about the fine detail, this includes the specification, design,
tendering, site preparation, installation, wiring and testing. All the components present in the
system need to be considered at all the stages.

All energy storage systems will have common elements and areas that are specific to the type of
storage that they use.

Energy storage system components

Safety rules

Groundworks
Energy storage plant and
hardware protection

Grid
connection + Auxiliary plant
metering

A control system and Power supplies for

communications auxiliary plant

Figure 21: components of an energy storage system (yellow is common components, orange is system specific)

4.1 Groundworks

The complexity of the groundworks depends on the type of energy storage scheme.

One of the most complex are those used in a pumped storage scheme

27
http://en.skmost.ru/objects/gidro/zagorskaya/ shows some example pictures (replicated here)
around the ground works in the Zagorsk Hydroelectric Pumped Storage Power Plant-2. Company
specialists were required to undertake the following activities:

• Reinforcing the existing dam of the upper accumulating reservoir.

• Building public concrete roads. At the stage of constructing the HPSPP these roads are used
as the access routes and once the construction is completed, they will become a part of the
transport infrastructure of the Zagorsk station.

• Building the joint section of the dam of the upper accumulating reservoir.

• Preparing the groundwork for the HPSPP building.

• Building delivery pipelines (4 pipes by 800 meters long and 7.5 m in diameter each).

Photo’s from the project show the scale of the work

Figure 22: ground works underway at a pumped storage scheme (http://en.skmost.ru/objects/gidro/zagorskaya/)

In many cases the requirement for ground works is much simpler. For example the ground works at
Willenhall consisted of

• Clear and level the site

• Ground radar to look for hidden cables

28
 • Design and lay concrete bases

• Design and Install Cable trenches

• Earthing

• Temporary power for welfare

Figure 23: ground works at Willenhall

4.2 Grid connection

There are cases where grid connection is not required or the energy storage medium is not
converted into electricity but used directly to produce energy in another form such as heat.
However, for the vast majority of cases – it is necessary to connect the system to the grid.

To connect to the grid it is necessary to comply with standards

• Distribution code

• G83 : connection of generation <3.68kW

• G59 : connection of generation > 3.68kW < 50MW

• G59 + Grid code: connection of generation >50MW

Generation and storage impacts the grid. Data needs to be sent to the DNO and/or TNO letting
them know about your generation

29
These codes primarily list the interface to the utility and the protection relay and settings that will be
used. On larger energy storage plant it is necessary to get these relays witnessed and injection
tested as part of the commissioning process. The metering will be a requirement of the grid
connection and the type of meter will be dependent on the usage of the energy storage system (eg
half hourly/ minutely etc)

G59 relay

Battery UPS

240V power supply to

battery

Cabling from voltage

transducers (fused)

Figure 24: G59 relay

30
 grid

G59 relay Small UPS battery

E- stop circuit

inverter

Figure 25: Schematic of the G59 system at Willenhall

4.3 Control and communications systems

Every system nowadays tends to have fairly sophisticated monitoring and control as the price of
processing power has plummeted. Each system may have different types of communication which
link back to an overall controller.

The control system will communicate with the balance of plant and also have an HMI interface to
allow the user to look at values and control setpoints

31
Figure 26: Schematic of the comms layout at Willenhall and schematic

32
Figure 27: HMI interface

The control system can become very detailed especially if it is monitoring every cell of a large unit.
For example, at Willenhall there are over 21,000 cells monitored.

In addition to the monitoring side it is necessary to have a detailed control system that runs on
automatic. This may have many influences and needs to move between all the system states safely.

Figure 28 is an example of a high level overview of a state-space diagram. In this case for a high
temperature solid oxide fuel cell system. The text in the black boxes is the state that the system is in,
for example “OFF” or it may be in “Start up” etc. The arrows indicate what needs to have happened
in order to move between states. For example, to move from “OFF” to “start up” the user is required
to press a green button. Once this is pressed the system moves into start up state. In this stage it will
start to allow gas to flow and operate a heater to get the fuel cell up to operating temperature. It
will do this by running through a sequence of events. Once the sequence of events has been
completed. The system then proceeds to “hot idle” or “steady state operation” depending whether
or not a power output has been requested by the user. At any moment if the Emergency shutdown
button is pressed or the safety system emergency shut down is instigated. The system goes into
“ESD” emergency shut down state. A shut down is a harsh and disruptive procedure so as an
alternative the system can enter abnormal operating conditions where it can try to recover and
return back to operational states. All control systems will have a similar state diagram behind them
for the different states their system can be in.

33
 Start complete and no
power requested
Purge

Start complete and

power requested

Figure 28: Example of a state diagram for a SOFC system

4.4 Balance of plant

It is not merely an issue of the main energy storage system plant – there is also the issue of the
auxiliary plant sometimes called balance of plant. This can include components such as

 Pumps,
 Fans,
 Compressors
 Valves
 Mass flow controllers
 Heaters
 Air condition systems
 Monitoring systems and the instrumentation
 Generators
 Inverters
 Motors

34
At Willenhall some of the key plant include the inverters and the air handling unit to allow these to
be properly cooled.

Figure 29: Inverters at Willenhall

In a compressed air facility the balance of plant is related more to mechanical devices

Figure 30: Schematic if a CAES [Overview of current development in electrical energy storage technologies and the
application potential in power system operation, Jing Luo et al]

4.5 Power supplies and cabling for auxiliary plant

This can be complex and requires attention to detail and wiring diagrams that are properly managed
and updated. The cabling may need to be screened to help with noise, twisted, run with special
earths, fused, sized properly and labelled. There are usually hundreds of wires going round the

35
system. The following shows an example of 415 V supplies being transformed to 120V ac and into
24Vdc – these are common voltages for instrumentation and control.

Figure 31: An example of auxiliary power supplies

36
5 Role of energy storage

Energy Storage Systems are expensive to purchase. Therefore, there needs to be a clear financial
benefit in using this on the grid system. Different suggested sources of income or usage resulting in
savings include, but are not limited to:

Small scale

• Increasing self-consumption in buildings to benefit from FIT

• Backup power eg UPS or CPS

Large scale

• Arbitrage – charging when electricity is cheap and dis-charging when it is expensive

• Ancillary services – providing frequency or voltage support to the grid

• Network deferral – providing power at peak times to avoid re-enforcing the Network

• Peak load/generation lopping – where load/or generation exceeds import/export

capacity on a line

• Other

Other services that energy storage could provide – such as voltage control, or power quality
correction are not currently subject to any income stream and therefore are not a profitable reason
to use energy storage.

The role of the energy storage also influences how the energy storage system is sized and how it is
connected to the grid system. The following subsections discuss each of these roles in turn.

5.1 Small scale – increasing self-consumption

The government, in an effort to increase the percentage of renewable generation, has offered
guaranteed payment through the use of feed-in tariff payments (or FITS) for domestic renewable
generation. These have decreased in later years. Typically, the payments are based on generation
meter readings. However, in the absence of these an import/export meter reading can be used to
infer the value of generation as shown in Figure 32. Any additional generation over load in the house
can be exported and payment for export forms another source of income. However, the payment for
export is generally less than the import payment of electricity to a consumer, so there is a financial
argument that says it is more profitable to store all the produced energy for use in the house to
offset the import costs of the electricity.

37
2016 export costs were approximately 4.85p/kWh1 while import is around 18p/kWh over the same
time period. Therefore, for each kWh of electricity stored for later use and not exported there is a
saving of around 13p/kWh.

PV

kwh
Generation
Meter

Power Distribution

Load

kwh Export Meter/

Import Meter

Grid
Figure 32: Metering for FIT on a small domestic system

There are a number of different locations for putting Energy storage into the system see Figure 33 .
Each has advantages and disadvantages. The location of the energy storage impacts the power
electronic interface to the system.

Location 1:

Shown in Figure 34.

• Any electricity losses in the battery will not be eligible for FIT payment

• Incomplete charging will impact life as battery cannot be charged from the grid

• System efficiency is higher as one dc/ac conversion

1
(http://www.fitariffs.co.uk/FITs/principles/export/)

38
 PV

- +

Battery Location 1

kwh
Generation
Meter
- +

Location 2
Battery
Power Distribution

Load

kwh Export Meter/

Import Meter - +

Location 3
Battery

Grid
Figure 33: Different locations for domestic energy storage

Location 2:

Shown in Figure 35 – but there are many variants (see Figure 36)

• Battery can be charged from grid and so can be fully charged

• Battery charging does not affect generation meter reading

• If there is no generator meter reading the FIT tariff can be “deemed” but is less likely to
work to the benefit of the customer.

• System efficiency is lower as two lots of dc/ac conversion

Location 3:

• Would require additional metering and offers no benefit so would not be used.

39
Figure 34: Location 1 – different architectures

Figure 35: Location 2 – different architectures

40
 DC Bus

PV DC Bus
Converter

Converter
Inverter

Inverter
Grid
DC Bus
Grid
Converter PV
Battery

Battery
Converter Inverter

Grid
AC Bus PV
Converter
DC Bus
PV
Battery
Inverter
Converter

Inverter
Grid
Inverter Grid
PV

Battery
Battery

Figure 36: Location 2 – variant architectures

The process is best illustrated below, where the energy management system needs to minimise the
power being imported from the grid in favour of using all the power locally.

Figure 37: Energy management

The cost of the Energy storage would offset the cost of the saving in p/kWh by not exporting excess
electricity. In terms of sizing the energy storage, there also exists, a trade-off between power and
cost. This is best illustrated below. A 5kW inverter costs approximately £1000 while a 1.5kW inverter
costs around £500.

41
Figure 38: Simulation of 5kW inverter with a 5kW PV panel

Figure 39: Simulation of 1.5kW inverter with a 5kW PV panel

The graphs show that in the first case, there is no import from the grid until about 10pm through to
7am. Whereas in the latter case there is an earlier import of up to 2kW for less than an hour. The
cost of this extra is <£90 pa which over the 5year warranty of the inverter means that it makes more
financial sense to buy a 1.5kW inverter system.

42
5.2 Small scale – UPS/CPS applications

The financial benefits in terms of UPS (uninterruptable power supply) and CPS (continuous power
supply) are more in keeping with savings against the risk of losing the electricity supply. This is often
more important in a manufacturing environment or data processing centre where loss of power and
downtime of the system can have severe financial consequences. In a domestic setting, the
implications of losing power are for example, loss of income if working from home to loss of
perishable food stuffs and the advantages of a UPS system are very situation specific.

A UPS can be classified as follows:

• Stand by UPS - Provides power when the grid goes down

• Line interactive UPS – Also conditions the power quality

These can be illustrated in Figure 40

Transfer switch

load

load

Figure 40: UPS schemes (standby – top and line interactive -bottom)

There are two main methods of connecting the line interactive UPS shown below;

Double conversion – power always go through battery

Rotary conversion – rotating machine helps to smooth out the power quality issues

43
 Bypass switch

Bypass switch load

Choke

Diesel generator
load

Flywheel
Synchronous
machine
Figure 41: line interactive UPS schemes – double conversion and rotary conversion

5.3 Large scale systems – arbitrage

Arbitrage is charging energy storage when electricity is cheap and then discharging when the price of
electricity is expensive.

The majority of wholesale trading takes place directly between buyers and sellers of electricity at a
price agreed in private. Benchmark prices for the trades are published on subscription to reference
agencies.

Short term energy trading takes place on two power exchanges in GB. Both of these publish price
data, this price is usually based on day-ahead:

44
http://www.apxgroup.com

http://www.nordpoolspot.com

An example of the long term day ahead price of electricity is shown in Figure 42

Figure 42: Day ahead electricity prices

System sell and buy prices for balancing are on the elexon website

https://www.bmreports.com/bmrs/?q=balancing/systemsellbuyprices

An example is shown below

Figure 43: Balancing prices for electricity

45
The graph shows three distinct peaks where electricity can be discharged from a battery and three
interim periods where it can be recharged as shown below.

Figure 44: Balancing prices for electricity

As the settlement periods are half hourly, we will consider a 1.2MWh battery with a round trip
efficiency of 0.83 (such that it discharges a total energy of 1MWh, but needs 1.2MWh to recharge).

Looking at the price at each settlement period the following table can be put together which looks at
the income and re-charging costs available on this day.

Settlement period Time Price Income Cost

15 7.30am £84/MWh £42

16 8am £96/MWh £48

19 9.30am £36/MWh £21.6

20 10am £36/MWh £21.6

24 12:00pm £84/MWh £42

25 12:30pm £84/MWh £42

30 3pm £36/MWh £21.6

31 3.30pm £36/MWh £21.6

34 5.30pm £110/MWh £55

35 60pm £100/MWh £50

45 10:30pm £36/MWh £21.6

46 11pm £36/MWh £21.6

Total £279 £129.6

46
The income – cost on this day is £149.40

However, if we assume that batteries cost around £1million/ MWh and that life cycles are up to
10,000 cycles then it is clear that

Total number of cycles per day = 3

Total life span of battery based on 3 cycles/day = 9 years

Total income of battery over 9 years based on this day = £490,000

This is clearly less than the battery cost and therefore is not worthwhile as a source of income on its
own.

In addition – every day is somewhat different – there are better days but also worse days

Figure 45: Balancing prices for electricity – a not very good day

5.4 Large scale – ancillary services

National grid is responsible for maintaining the frequency and voltage on the Electricity Network. To
this end they have competitive services which allow generating companies and others to bid into to
provide services at assist with this. Key services to help with frequency response include:

• Mandatory frequency response – Generators only

• STOR (short term operating reserve)

• FFR (Firm frequency response)

• EFR (Enhanced frequency response)

• Demand side response – more likely to be loads only

The last four services are all examples of where a battery may be used. The demand side response is
very similar to the FFR response but instead of increasing generation, there is a drop in load which

47
has a similar impact. In terms of an energy storage system – this can be thought of as a drop in
export of electricity or an increase in import of electricity. This work will look in more detail at STOR,
FFR and EFR based on available knowledge at this time. There are sets of detailed technical
requirements available on the National Grid website. These notes won’t look at these with any
depth but will concentrate on the application and approximate income stream based on recent
market values.

5.4.1 STOR

National Grid procures Short Term Operating Reserve (STOR) through a competitive tender process
which is conducted three times per year. This generation is instructed to turn on when there is a
shortage of generation. The generator is paid for being available (availability fee) and delivering on
request (Utilisation fee)

• Minimum 3MW for 4 hours 3 times per week

• between 07:00 and 13:30 and 16:30 and 21:00

• Prices are around £6.50/MWh availability and £62/MWh used

• In Sept 2017 330MW was contracted and total expenditure was £1.08m

Taking these values as representative for a month

• Assume 30 days of 11 hours availability of 330MW at £6.50/MWh = £700k approx

• £380000 left for Utilisation payment = approx 6000MWh of Utilisation service

• The means the units operate for around 18 hours in the month.

Let’s assume a 12MWh battery (cost £12m) providing 3MW for 4 hours

Income

• availability - 30days x 11hours x 3MWh x £6.50 = £6435

• Utilisation - 18 hours x 3MWh x £62 = £3348

• Total = £10,000

• Total cycle usage – approx 5 cycles/month

• Total life span = >100 years !

• Total income over a 10 year period => £1.2m

Similar to the arbitrage service – it is not profitable to buy and operate a battery entirely based on a
STOR service.

48
5.4.2 FFR service

This service is used to respond to large events on the grid such as a loss of generation which results
in a drop in frequency. The FFR then responds and stays on until the STOR can operate as shown
The tender tends to be based on an availability fee (when the plant is available) and a nomination
fee (when it is used)

Grid frequency (Hz)

50Hz

49.5Hz

FFR response STOR response

Time
seconds 10’s of minutes

Figure 46: Example of FFR response

The requirements for this have been changing over the last few years. Current requirements suggest
that ;

 Minimum 1MW response energy

 Respond in dynamic and non-dynamic mode

Previous requirements also used to include

 Used to be respond within 10s and stay on for 30minutes but it appears this is no longer a
requirement

Nowadays the service is more likely to be dynamic based on a frequency trigger point. As the
frequency rarely goes below 49.5, the following curve is produced by a battery operating in dynamic
mode with a tighter frequency band of between 50.04Hz and 49.96Hz.

49
The operation of the converter (orange trace), moves to positive charging mode as the grid
frequency moves above 50.04Hz (shown as upper broken green line). Vice verse as the frequency
moves below 49.96Hz.

Figure 47: dynamic frequency response

Commercial Frequency response

• Around 235 bids to provide service with 25 taken up

• Bids are a mixture of dynamic (continuously changing) and non-dynamic (triggered on set
points)

• Tenders are normally based on availability fee only. The average non-dynamic is £60/hr
based on a size offering of around 7.5MW and a mixture of time slots while the dynamic is
around £600/hr based on 1GW

• Total payment in sept was £7.47m

National Grid are not keen on using batteries for this service as they can’t be relied upon to be there
– an example of the published National Grid tender prices is shown below

50
Considering an energy storage system of 8MW which has charge for 30miniutes.

• Estimate of 3000 triggered events per year

• For a battery of 8MW with 30 min storage (4MWh at around £4m)

• Total income per year = £60*365*24 = £5.26m

• Assume each event cycles the battery – life span is just over 3 years.

• Looks to be profitable –However, National Grid have introduced the EFR service to look at
dynamically using energy storage (although this has ended up being primarily batteries) and
takes better advantage of their characteristics.

5.4.3 EFR

51
 • New service being trialled by National Grid to provide dynamic response with a tighter
frequency margin

• Payment is primarily based on an availability fee as the battery is in near continuous

operation

• Some of the challenges are state of charge management and there are penalties for non-
delivery

The battery is required to respond within a frequency/MW envelope as shown in Figure 48

Figure 48: EFR envelope

There are two services with the following reference points

As some battery and energy storage types respond more slower than others there is also the
requirement for response rates as shown below:

52
Figure 49: EFR response envelope

The most onerous part of this is the state of charge management. There are several reasons for this

The inverter is less efficient when operated at a low power level and therefore the deadband area is
not an efficient place to be charging and discharging an energy storage system. This limits control to
area B.

The figure over shows the Willenhall energy storage system operating under EFR for an injected
frequency based on National Grid supplied figures for 21st Oct 2015 (one of the more problematic
days).

53
 SOC management

Area with no SOC

Figure 50: EFR operation

As this is a new service it is difficult to gauge the market response. However National Grid published
some data based on the tender requirements put forward as part of the trial.

54
Figure 51: EFR tender

Average tender of successful plant is £9.44/MW of EFR/hr

Cost of 20minutes of 2MW of plant (0.6MWh = £600k)

Income per year = 9.44 x 2 x 8760 = £1.65m

Continual usage – life approx 10 years?

5.5 Large scale - Network deferral

The aim behind this is to use Energy storage in key places on the Network to avoid the need for
expensive Network upgrades. In addition to avoiding network upgrades there are benefits in
including energy storage including

 Helping with power quality and phase imbalance

 Helping with voltage control
 Using the batteries for other services outside of peak times

55
There have been a number of studies around this. These notes will focus on the FALCON project
results run by Western Power Distribution. This project looked at five sites with 50kW/100kWh
energy storage batteries (Sodium Nickel) and converters located at the LV substations

Figure 52: WPD falcon project

battery

RMU and local transformer Inverter and auxiliary equipment

Figure 53: WPD falcon project installation

56
 Secondary
Substation

Gridkey Power
FALCON Energy Storage Site
monitor meter

LV board
AC
Circuit
Inverter/ Battery
breaker rectifier
DC
OC G59

Cut out switch

Battery cabinet auxiliary power supply

Small power & lighting supply

Pre-existing
LV network

Figure 54: WPD falcon project schematic

These units charge or discharge the batteries at designated times or triggers:

• As specified manually;

• At specified times regardless of network conditions;

• When substation load is above a specified trigger point (peak shaving operations); or

• When grid frequency is below a specified trigger point.

In addition, the converters can also import or export reactive power, influencing voltage at the point
of connection:

• As specified manually;

• At specified times regardless of network conditions; or

• When substation voltage is above/below a specified trigger point (voltage response

operations);

Several key lessons are available from this research

 A maintenance cycle is recommended to be run at a fixed time each week.

 However, the control system decides how the cycle will proceed. This results in two key
unknowns:
o The actions taken by the control system are not pre-determined.

57
 o The maintenance cycle activity, if it occurs, is not a fixed time period but can vary
from less than an hour to many hours.
 From an availability perspective – this means the battery cannot be routinely relied upon for
service over the twelve hour period from the control system start of a maintenance cycle.

Figure 55: WPD falcon project maintenance cycle 1

Figure 56: WPD falcon project maintenance cycle 2

The discharge regime is not uniform around strings of batteries

58
The staggering at the end of the discharge cycle is indicative of sets of battery strings being turned
off by the battery management system as they reach limits.

Discharge Power by Site

0
Discharge power (kW)

-5
-10
-15
-20
-25
-30

AMBRIDGE GROVE AWA PUMPING STATION MIDDLETON

Chadds Lane combined FALCON AVENUE SPRINGFIELD
Helford Place South combined

Figure 57: WPD falcon project discharge staggering

It is also worth noting that for this chemistry of battery the charge and discharge shapes are
different

AWA PUMPING STATION MIDDLETON

40
Thousands

30

20
Power (kW)

10

-10

-20

-30

Total

Figure 58: WPD falcon project charge and discharge shapes

Discharge takes 2 hours, whereas recharging takes over six hours to broadly complete. Availability is
thus limited by the charging time required.

59
However, one of the key reasons for undertaking this research was to look into network investment
deferral by using the batteries to peak load lop. It is clear that the batteries are able to do this.

AMBRIDGE GROVE
100
Thousands

80

60
Power (kW)

40

20

-20

-40
13/09/2014 13/09/2014 04:00 13/09/2014 08:00 13/09/2014 12:00 13/09/2014 16:00 13/09/2014 20:00

Sum of Energy storage unit active power Sum of Sub active power excluding energy storage

Sum of Total mean busbar active power - inc energy storage

Figure 59: WPD falcon project peak shaving over a day

Figure 60: WPD falcon project peak shaving over a week

Peak shaving functionality (using pre-determined load threshold) has been demonstrated to
effectively limit substation load. The peak shaving function is clearly repeatable.

But, there are issues within this - Peak Shaving with threshold set too low (battery becomes fully
discharged prior to the peak demand passing)

60
 AMBRIDGE GROVE
200
Thousands
150

100
Power (kW)

50

-50

-100

Sum of Energy storage unit active power Sum of Sub active power excluding energy storage

Sum of Total mean busbar active power - inc energy storage

Figure 61: WPD falcon project peak shaving with too low a threshold

Also as time progresses the amount of energy storage needed may reduce and disappear altogether.
The following figure shows Peak load shaving same day - but three weeks later showing reduction in
required energy.

Helford Place South combined Helford Place South combined

250 250
Thousands
Thousands

200 200

150 150
Power (kW)
Power (kW)

100 100

50
50

0
0

-50
-50

Sum of Energy storage unit active power

Sum of Energy storage unit active power
Sum of Sub active power excluding energy storage
Sum of Sub active power excluding energy storage
Sum of Total mean busbar active power - inc energy storage
Sum of Total mean busbar active power - inc energy storage

Figure 62: WPD falcon project peak shaving with too low a threshold

This can be an issue. For example the load curve of the UK over the course of a year is shown in
Figure 63. Assuming that the load is approximately scaled at each substation, there are only a few
occasions in winter where this would benefit the 11kV Network. This is in itself does not make
commercial sense against the cost of substation and cable re-enforcement.

61
Figure 63: 2013 National Grid load variation

5.6 Large scale – peak load/generation lopping

This application is very similar to the Network deferral issue but is more closely tied to customer
import and export restriction limits. Normally a customer will have a fixed amount of
load/generation that they are allowed. If they want to expand or add generation then the Network
may require an upgrade. Scottish and Southern published a generation connection case studies
guide to allow approximate costs to be estimated
(https://www.ssepd.co.uk/WorkArea/DownloadAsset.aspx?id=940)

This includes a table of commonly incurred costs as replicated below and a number of case studies
which include itemised costs.

62
The notes take the figures from the case studies and consider a very approximate estimate of how
this would compare to adding energy storage over a 5 hour period to allow generation to be used
and stored and then exported later.

Case Example Indicative Battery alternative notes

connection
charge

1 10kW gen to a domestic £750 NA

customer – no re-
enforcement needed

2 10kW gen to a domestic £4450 For a 3kW average Some generation

customer, plus LV cable domestic load and a would be curtailed
upgrade 10kW PV panel and most usage
A 6kWh energy would occur in the

63
 storage system is house. A £750 fee
around £3k would also be
required
3 3MW gen to a £71,500 NA The equipment is
commercial customer – around the
Network can generation operating
accommodate so would still be
required
4 3MW gen to a £271,500 Add say £10MWh of Not worth the
commercial customer – energy storage investment
New dedicated HV switch @£10m to remove
and breaker this requirement
saving £200k on
connection
5 3MW gen to a £2.1M Add say £10MWh of Pay back possible in 5
commercial customer – energy storage years
New dedicated HV switch @£10m to remove
and breaker and primary this requirement
substation and Network saving £2M on
upgrade connection

5.7 Summary

Large scale energy storage on the grid may have issues with availability and energy level uncertainty
caused by;

• SOC control and reporting;

• Re-calibration procedures;

• Differences between strings and batteries manifesting as different discharge/charge timings;

• Different discharge/charge curve shapes.

The market is changing quickly and looks like frequency response services offer the most profitable
usage for energy storage. The most probable usage of energy storage would like come from a
mixture of uses.

It should be noted that as energy storage is added to the Network, the peak load may reduce, this
will make it easier to use base load power plant to meet load and therefore lessen the need for
expensive plant. This will reduce the cost of peak load electricity and therefore diminish the case for
energy storage.

64
6 Appendix A : Reduction potentials

Standard Cathode (Reduction) Half-Reaction Standard Reduction Potential E°

(volts)

Li+(aq) + e- ⇌ Li(s) –3.040

Ba2+ + 2e−⇌ Ba(s) –2.92
Rb+ + e- ⇌ Rb (s) -2.98
K+(aq) + e- ⇌ K(s) -2.93
Cs+(aq) + e- ⇌ Cs(s) -2.92
Ba2+(aq) + 2e- ⇌ Ba(s) -2.91
Sr2+(aq) + 2e- ⇌ Sr(s) -2.89
Ca2+(aq) + 2e- ⇌ Ca(s) –2.84
Na+(aq) + e- ⇌ Na(s) –2.713
Mg(OH)2(s) + 2e−⇌ Mg(s) + 2OH− –2.687
La3+ + 3e−⇌ La(s) –2.38
Mg2+(aq) + 2e- ⇌ Mg(s) -2.356
Ce3+ + 3e−⇌ Ce(s) –2.336
Al(OH)4− + 3e−⇌ Al(s) + 4OH− –2.310
AlF63− + 3e−⇌ Al(s) + 6F− –2.07
Be2+ + 2e−⇌ Be(s) –1.99
B(OH)4− + 3e−⇌ B(s) + 4OH− –1.811
U3+ + 3e−⇌ U(s) –1.66
Al3+(aq) + 3e- ⇌ Al(s) –1.676
SiF62− + 4e−⇌ Si(s) + 6F− –1.37
Zn(CN)42− + 2e−⇌ Zn(s) + 4CN –1.34
Zn(OH)42− + 2e−⇌ Zn(s) + 4OH− –1.285
Mn2+ + 2e−⇌ Mn(s) –1.17
V2+ + 2e−⇌ V(s) –1.13
2SO32− + 2H2O(l) + 2e− ⇌ S2O42− + 4OH− –1.13
Zn(NH3)42+ + 2e− ⇌ Zn(s) + 4NH3 –1.04
O2 (aq) + e- ⇌ O2-(aq) -1.0
Cd(CN)42− + 2e−⇌ Cd(s) + 4CN− –0.943
MoO42− + 4H2O(l) + 6e−⇌Mo(s) + 8OH− –0.913
SiO2(s) + 4H+ + 4e−⇌ Si(s) + 2H2O(l) –0.909
SO42− + H2O(l) + 2e− ⇌ SO32−+ 2OH− –0.936
Cr2+ + 2e− ⇌ Cr(s) –0.90
B(OH)3 + 3H+ + 3e− ⇌ B(s) + 3H2O(l) –0.890
2H2O(l) + 2e- ⇌ H2(g) + 2OH-(aq) -0.828
Zn2+(aq) + 2e- ⇌ Zn(s) –0.7618
Co(OH)2(s) + 2e−⇌ Co(s) + 2OH− –0.746
Cr3+(aq) + 3e- ⇌ Cr(s) –0.424
Ni(OH)2 + 2e−⇌ Ni(s) + 2OH− –0.72
Ag2S(s) + 2e−⇌ 2Ag(s) + S2− –0.71
Se(s) + 2e−⇌ Se2− –0.67 in 1 M NaOH
Cd(NH3)42+ + 2e−⇌ Cd(s) + 4NH3 –0.622
2SO32− + 3H2O(l) + 4e− ⇌S2O32− + 6OH− –0.576 in 1 M NaOH

65
Standard Cathode (Reduction) Half-Reaction Standard Reduction Potential E°
(volts)

U4+ + e− ⇌ U3+ –0.52

SiO2(s) + 8H+ + 8e− ⇌ SiH4(g) + 2H2O(l) –0.516
Sb + 3H+ + 3e− ⇌ SbH3(g) –0.510
H3PO3+ 2H+ + 2e− ⇌ H3PO2 + H2O(l) -0.50
Ni(NH3)62+ + 2e− ⇌ Ni(s) + 6NH3 –0.49
2CO2(g) + 2H+ +2e−⇌ H2C2O4 –0.481
Cr3+ + e−⇌ Cr2+ –0.424
Fe2+(aq) + 2e- ⇌ Fe(s) -0.44
S(s) + 2e−⇌ S2− –0.407
Cd2+(aq) + 2e- ⇌ Cd(s) –0.4030
Ag(NH3)2+ + e−⇌ Ag(s) + 2NH3 –0.373
Ti3+ + e−⇌ Ti2+ –0.37
PbSO4(s) + 2e−⇌ Pb(s) + SO42− –0.356
Co2+(aq) + 2e- ⇌ Co(s) –0.277
2SO42− + 4H+ + 2e− ⇌ S2O62− + 2H2O(l) –0.25
N2(g) + 5H+ + 4e−⇌ N2H5+ –0.23
H3PO4 + 2H+ + 2e−⇌ H3PO3 + H2O(l) -0.28
Ni2+(aq) + 2e- ⇌ Ni(s) –0.257
V3+ + e− ⇌ V2+ –0.255
As + 3H+ + 3e− ⇌ AsH3(g) –0.225
CO2(g) + 2H+ + 2e− ⇌ HCO2H –0.20
Mo3+ + 3e− ⇌ Mo(s) –0.2
Sn2+ + 2e− ⇌ Sn(s) –0.19 in 1 M HCl
Ti2+ + 2e−⇌ Ti(s) –0.163
MoO2(s) + 4H+ + 4e−⇌ Mo(s) + 2H2O(l) –0.152
AgI(s) + e− ⇌ Ag(s) + I− –0.152
Sn2+(aq) + 2e- ⇌ Sn(s) -0.14
Pb2+(aq) + 2e- ⇌ Pb(s) -0.126
CrO42− + 4H2O(l) + 3e− ⇌2Cr(OH)4− + 4OH− –0.13 in 1 M NaOH
WO2(s) + 4H+ + 4e−⇌ W(s) + 2H2O(l) –0.119
Se(s) + 2H+ + 2e−⇌ H2Se(g) –0.115
CO2(g) + 2H+ + 2e−⇌ CO(g) + H2O(l) –0.106
WO3(s) + 6H+ + 6e−⇌ W(s) + 3H2O(l) –0.090
Hg2I2(s) + 2e−⇌ 2Hg(l) + 2I− –0.0405
Fe3+(aq) + 3e- ⇌ Fe(s) -0.037
2H+(aq) + 2e- ⇌ H2(g) 0.00
P(s,white) + 3H+ + 3e−⇌ PH3(g) 0.06
AgBr(s) + e−⇌ Ag(s) + Br− 0.071
S4O62− + 2e−⇌2S2O32− 0.080
Co(NH3)63+ +e−⇌Co(NH3)62+ 0.1
Ru(NH3)63+ + e−⇌ Ru(s) + Ru(NH3)62+ 0.10
S(s) + 2H+ + 2e−⇌ H2S 0.144
Sn4+(aq) + 2e- ⇌ Sn2+(aq) 0.154

66
Standard Cathode (Reduction) Half-Reaction Standard Reduction Potential E°
(volts)

Cu2+(aq) + e- ⇌ Cu+(aq) 0.159

UO22+ + e− ⇌UO2+ 0.16
Co(OH)3(s) + e−⇌ Co(OH)2(s) + OH− 0.17
ClO4-(aq) + H2O(l) + 2e- ⇌ ClO3-(aq) + 2OH-(aq) 0.17
SO42− + 4H+ + 2e− ⇌ H2SO32−+ H2O(l) 0.172
BiCl4− + 3e−⇌ Bi(s) + 4Cl− 0.199
SbO+ + 2H+ + 3e−⇌ Sb(s) + H2O(l) 0.212
AgCl(s) + e- ⇌ Ag(s) + Cl-(aq) 0.2223
HCHO + 2H+ + 2e−⇌ CH3OH 0.2323
HAsO2 + 3H+ + 3e−⇌ As(s) + 2H2O(l) 0.240
Ru3+ + e−⇌ Ru2+ 0.249
IO3−+ 3H2O(l) + 6e−⇌ I− + 6OH− 0.257
Hg2Cl2(s) + 2e−⇌ 2Hg(l) + 2Cl− 0.2682
UO2+ + 4H+ + e− ⇌U4+ + 2H2O(l) 0.27
Bi3+ + 3e−⇌ Bi(s) 0.317
UO22+ + 4H+ + 2e− ⇌ U4+ + 2H2O(l) 0.327
VO2+ + 2H+ +e−⇌V3+ + H2O(l) 0.337
Cu2+(aq) + 2e- ⇌ Cu(s) 0.3419
ClO3-(aq) + H2O(l) + 2e- ⇌ ClO2-(aq) + 2OH-(aq) 0.35
Fe(CN)63− + e− ⇌ Fe(CN)64− 0.356
O2(g) + 2H2O(l) + 4e−⇌ 4OH− 0.401
ClO− + H2O(l) + e− ⇌ ½Cl2(g) + 2OH− 0.421 in 1 M NaOH
Ag2C2O4(s) + 2e−⇌ 2Ag(s) + C2O42− 0.47
Cu+(aq) + e- ⇌ Cu(s) 0.52
I2(s) + 2e- ⇌ 2I-(aq) 0.5355
I3− + 2e−⇌ 3I− 0.536
Ga3+ + 3e−⇌ Ga(s) -0.56
Cu2+ + Cl− + e−⇌ CuCl(s) 0.559
S2O62− + 4H+ + 2e−⇌ 2H2SO3 0.569
H3AsO4 + 2H+ + 2e−⇌ HAsO2 + 2H2O(l) 0.560
ClO2-(aq) + H2O(l) + 2e- ⇌ ClO-(aq) + 2OH-(aq) 0.59
MnO4− + 2H2O(l) + 3e−⇌ MnO2(s) + 4OH− 0.60
Sb2O5(s) + 6H+ + 4e−⇌ 2SbO+ + 3H2O(l) 0.605
PtCl62− + 2e−⇌ PtCl42- + 2Cl− 0.68
RuO2(s) + 4H+ + 4e−⇌ Ru(s) + 2H2O(l) 0.68
O2(g) + 2H+ + 2e−⇌ H2O2 0.695
PtCl42− + 2e−⇌ Pt(s) + 4Cl− 0.73
H2SeO3 + 4H+ + 4e−⇌ Se(s) + 3H2O(l) 0.74
Tl3+ + 3e−⇌ Tl(s) 0.742
Fe3+(aq) + e- ⇌ Fe2+(aq) 0.771
Hg22+(aq) + 2e- ⇌ 2Hg(l) 0.7960
Ag+(aq) + e- ⇌ Ag(s) 0.7996
Hg2+(aq) + 2e- ⇌ Hg(l) 0.8535

67
Standard Cathode (Reduction) Half-Reaction Standard Reduction Potential E°
(volts)

Cu2+ + I− +e−⇌ CuI(s) 0.86

Ru(CN)63− +e−⇌ Ru(s) + Ru(CN)64− 0.86
ClO− + H2O(l) + 2e−⇌ Cl− + 2OH− 0.890 in 1 M NaOH
2Hg2+(aq) + 2e- ⇌ Hg22+(aq) 0.911
HgO(s) + 2H+ + 2e−⇌ Hg(l) + H2O(l) 0.926
NO3− + 3H+ + 2e−⇌ HNO2 + H2O(l) 0.94
MnO2(s) + 4H+ + e- ⇌ Mn3+ (aq) + H2O(I) 0.95
NO3-(aq) + 4H+(aq) + 3e- ⇌ NO(g) + 2H2O(l) 0.96
HIO + H+ + 2e−⇌ I− + H2O(l) 0.985
HNO2 + H+ +e−⇌ NO(g) + H2O(l) 0.996
VO22+ + 2H+ +e−⇌ VO2+ + H2O(l) 1.000
AuCl4− + 3e−⇌ Au(s) + 4Cl− 1.002
NO2 (g) + H+ (aq) + e- ⇌ HNO2 (aq) 1.07

Br2(l) + 2e- ⇌ 2Br-(aq) 1.087

Fe(phen)63+ +e− ⇌ Fe(phen)62+ 1.147
SeO43− + 4H+ + e−⇌ H2SeO3 + H2O(l) 1.151
ClO3− + 2H+ + e−⇌ ClO2(g) + H2O 1.175
ClO3− + 3H+ + 2e− ⇌ HClO2 + H2O 1.181
IO3− + 6H+ + 5e−⇌ ½I2(s) + 3H2O(l) 1.195
Pt2+ + 2e−⇌ Pt(s) 1.2
ClO4− + 2H+ + 2e− ⇌ ClO3− + H2O 1.201
O2(g) + 4H+(aq) + 4e- ⇌ 2H2O(l) 1.229
MnO2(s) + 4H+ + 2e−⇌ Mn2+ + 2H2O(l) 1.23
Tl3+ + 2e− ⇌ Tl+ 0.77 in 1 M HCl
2HNO2 + 4H+ + 4e−⇌ N2O(g) + 3H2O(l) 1.297
HOBr + H+ + 2e−⇌ Br− + H2O(l) 1.341
Cr2O72-(aq) + 14H+(aq) + 6e- ⇌ 2Cr3+(aq) + 7H2O(l) 1.36
Cr2O72− + 14H+ + 6e− ⇌ 2Cr3+ + 7H2O(l) 1.36
Cl2(g) + 2e- ⇌2Cl-(aq) 1.396
Au3+ + 2e−⇌ Au+ 1.36
Hg2Br2(s) + 2e−⇌ 2Hg(l) + 2Br− 1.392
Ce4+(aq) + e- ⇌ Ce3+(aq) 1.44
PbO2(s) + 4H+ +2e- ⇌ Pb2+(aq) + 2H2O(l) 1.46
BrO3− + 6H+ + 6e−⇌ Br− + 3H2O 1.478
Mn3+ + e−⇌ Mn2+ 1.5
MnO4-(aq) + 8H+(aq) + 5e- ⇌ Mn2+(aq) + 4H2O(l) 1.51
BrO3− + 6H+ + 5e−⇌ ½Br2(l) + 3H2O 1.5
Au3+ + 3e−⇌ Au(s) 1.52
2NO(g) + 2H+ + 2e−⇌ N2O(g) + H2O(l) 1.59
HOBr + H+ + e−⇌ ½Br− + H2O(l) 1.604
HClO2 + 2H+ + 2e−⇌ HOCl + H2O 1.64
PbO2(s) + 4SO42− + 4H+ + 2e−⇌ PbSO4(s) + 2H2O(l) 1.690

68
Standard Cathode (Reduction) Half-Reaction Standard Reduction Potential E°
(volts)

MnO4− + 4H+ +3e−⇌ MnO2(s) + 2H2O(l) 1.70

Ce4+ + e−⇌ Ce3+ 1.72
N2O(g) + 2H+ + 2e−⇌ N2(g) + H2O(l) 1.77
H2O2(aq) + 2H+(aq) + 2e- ⇌ 2H2O(l) 1.763
Au+ + e−⇌ Au(s) 1.83
Co3+(aq) + e- ⇌ Co2+(aq) 1.92
S2O82− + 2e−⇌2SO42− 1.96
O3(g) + 2H+(aq) + 2e- ⇌ O2(g) + H2O(l) 2.07
BaO(s) + 2H+ + 2e−⇌ Ba(s) + H2O(l) 2.365
F2(g) + 2e- ⇌ 2F-(aq) 2.87
F2(g) + 2H+ + 2e−⇌ 2HF 3.053

69