Microgrid and Energy Management

Renewable and Sustainable Energy Reviews
10 (2006) 78–127
www.elsevier.com/locate/rser
Can microgrids make a major contribution to UK

energy supply?
S. Abu-Sharkha,1, R.J. Arnolde,1, J. Kohlerd,1, R. Lia,1,
T. Markvarta,*,1, J.N. Rossb,1, K. Steemersc,1, P. Wilsonb,1, R. Yaoc,1
a
School of Engineering Sciences, University of Southampton, Highfield, Southampton SO17 1BJ, UK
b
School of Electronics and Computer Science, University of Southampton, Southampton, UK
c
Martin Centre for Architectural and Urban Studies, University of Cambridge, Cambridge, UK
d
Department of Applied Economics, University of Cambridge, Cambridge, UK
e
Siemens plc, Manchester, UK
Received 14 September 2004; accepted 16 September 2004
Abstract
Almost all the electricity currently produced in the UK is generated as part of a centralised power
system designed around large fossil fuel or nuclear power stations. This power system is robust and
reliable but the efficiency of power generation is low, resulting in large quantities of waste heat. The
principal aim of this paper is to investigate an alternative concept: the energy production by small
scale generators in close proximity to the energy users, integrated into microgrids.
Microgrids—de-centralised electricity generation combined with on-site production of heat—bear
the promise of substantial environmental benefits, brought about by a higher energy efficiency and by
facilitating the integration of renewable sources such as photovoltaic arrays or wind turbines. By
virtue of good match between generation and load, microgrids have a low impact on the electricity
network, despite a potentially significant level of generation by intermittent energy sources. The
paper discusses the technical and economic issues associated with this novel concept, giving an
overview of the generator technologies, the current regulatory framework in the UK, and the barriers
that have to be overcome if microgrids are to make a major contribution to the UK energy supply.
The focus of this study is a microgrid of domestic users powered by small Combined Heat and
Power generators and photovoltaics. Focusing on the energy balance between the generation and
load, it is found that the optimum combination of the generators in the microgrid- consisting of
* Corresponding author. Tel.: C44 238 059 3783; fax: C44 238 059 3016.
E-mail address: [email protected] (T. Markvart).
1
The authors are also members of the Tyndall Centre for Climate Change Research, UK.
1364-0321/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.rser.2004.09.013
S. Abu-Sharkh et al. / Renewable and Sustainable Energy Reviews 10 (2006) 78–127 79
around 1.4 kWp PV array per household and 45% household ownership of micro-CHP generators-
will maintain energy balance on a yearly basis if supplemented by energy storage of 2.7 kWh per
household.
We find that there is no fundamental technological reason why microgrids cannot contribute an
appreciable part of the UK energy demand. Indeed, an estimate of cost indicates that the microgrids
considered in this study would supply electricity at a cost comparable with the present electricity
supply if the current support mechanisms for photovoltaics were maintained.
Combining photovoltaics and micro-CHP and a small battery requirement gives a microgrid that is
independent of the national electricity network. In the short term, this has particular benefits for
remote communities but more wide-ranging possibilities open up in the medium to long term.
Microgrids could meet the need to replace current generation nuclear and coal fired power stations,
greatly reducing the demand on the transmission and distribution network.
q 2005 Elsevier Ltd. All rights reserved.
Contents
1. Introduction and background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 80

1.1. Some history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 80
1.2. Distributed generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 81
1.3. Why microgrids? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 82
1.4. Issues to consider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 83
1.5. Structure of the paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 84
2. Microgrids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2.1. The microgrid concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2.2. The relationship between the microgrid and a local electricity utility . . . . . . . . . . . 86
2.3. Internal control of a microgrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
2.3.1. Power balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
2.3.2. Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2.3.3. Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2.3.4. Power quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2.4. Energy balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
2.5. Energy storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3. Generators and loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

3.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.2. Photovoltaic arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.3. Fossil fuel power generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.3.1. Combined heat and power generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.3.2. Reciprocating internal combustion engines . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.3.3. Micro-turbines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.3.4. Stirling engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
3.3.5. Fuel cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
3.4. Electrical power generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
3.4.1. Inverters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.4.2. Utility integration issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
3.5. The domestic load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
80 S. Abu-Sharkh et al. / Renewable and Sustainable Energy Reviews 10 (2006) 78–127
4. Modelling energy consumption in buildings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

4.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.2. Domestic electricity load profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.3. Heating load profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4.4. Domestic hot water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.5. Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.6. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
5. An example of sizing: microgrids powered by photovoltaics and micro-CHP . . . . . . . . . 105

5.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.2. The methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.3. Daily energy balance in a microgrid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
5.4. Hourly energy balance Energy storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
5.5. Security of supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.5.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.5.2. Photovoltaic generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
5.5.3. Micro-CHP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5.6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
6. Economic analysis of microgrids. Regulatory issues . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

6.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
6.2. Micro-generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6.2.1. General overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
6.2.2. Summary of regulatory issues and current regulation activities . . . . . . . . . . . 119
6.3. Microgrids: the context of current government thinking in the UK . . . . . . . . . . . . . 119
6.3.1. Registered power zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
6.3.2. The view of the electricity industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
6.4. Economic analysis of microgrids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
6.5. Where do we go from here? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
1. Introduction and background
1.1. Some history
Public electricity supply in the UK began in 1881, but it was not until 1926 that the future
direction of growth had been established. The generation and bulk transmission of
electricity as well as its control were vested in one utility. This developed to take the form of
large (up to 2000 MW) usually coal fired power stations feeding a high voltage (up to
400 kV) interconnected transmission network (the Grid). Twelve area-based utilities
bought the electricity in bulk, distributed it via lower voltage feeders and sold it to
Fig. 1. A typical electricity transmission and distributions system, illustrated on the example of the UK public
electricity supply. Most generators considered in this paper would be connected at 415/230 V voltage (adapted
from R. Cochrane, Power to the People, CEGB/Newness Books, 1985).
consumers at various voltages, from 33 kV to 240 V Fig. 1. This situation prevailed in

several forms until 1989, the most prominent feature being the control of generation, bulk
transmission planning and the bulk supply tariff being under the control of one utility.
The Electricity Act of 1989 both split the industry into many separate parts and
privatised it. Also introduced was an Office of Electricity Regulation (OFFER) with a
primary responsibility for promoting competition and protect the consumer. The gener-
ating stations were privatised and sold their electricity on the open market. The market was
facilitated by a transmission network (privately owned and managed) and a number of
Distribution Network Operators (DNOs). None of these ‘network owning’ companies
bought or sold electricity—their revenue came from the transmission or distribution of
electricity.
Electricity Suppliers bought from the generators and resold to customers, the price
being controlled by the market place. This situation existed, again in several forms, until
today, when the New Electricity Trading Act (NETA) has altered pricing and charging
methods.
1.2. Distributed generation
The use of renewable energy sources for the generation of electricity is seen as one of
the important ways of reducing carbon dioxide emissions. Whilst some of these sources
can produce large power outputs in single power stations (hydropower or geothermal
power, for example) the majority are relatively small in size. This means such generators
are more conveniently (and cheaply) connected at lower voltages within the distribution
system. It was never envisaged that this system would be required to support the
connection of generation.
A similar situation exists with combined heat and power (CHP) units which produce
both electricity and heat. In the UK, almost 50% of the primary energy consumption is
used to provide heating and hot water in buildings [1] and the aim of CHP is to supply this
low-grade heat alongside electricity generation. The advantages of a high overall
efficiency of energy production which are thus attained must be offset against the necessity
to operate smaller units close to the consumer, and usually need to operate a district
heating system to distribute the heat. The term ‘distributed generation’ is used to describe
these generators (typically small renewables and CHP, but also other on-site electricity
generators) connected to the distribution system. A synonymous term also used is
‘embedded generation’, describing generators embedded within the distribution system.
Connection of generation in this way poses many technical, commercial and safety issues;
all of which must be tackled in order to allow a wide penetration of renewable generation.
In addition to the problems posed by distributed generation, the use of renewable
sources and CHP usually adds more specific issues related to the actual method of
generation used. An example that can be used to illustrate this point is photovoltaic
(PV) generation. There is no generation at night, which is of course predictable but in
addition, the sun can be obscured by cloud cover on a random basis. A similar
situation exists with CHP which is normally controlled to supply heat, with electricity
production as a by-product. These phenomena change in all time scales leading to a
constantly changing electrical output. As a source of electrical energy this has some
disadvantages which, in the present format, have to be addressed by the operators of
the public electricity supply.
Consumers have become used to electrical power available on demand. They do not
need to structure their load pattern, the entire responsibility for matching power and
demand is placed upon the utilities, which must have enough generation available at all
times. With more creative thinking about the way energy is supplied, used and controlled it
may be possible to satisfy the demand for energy, but accommodate the fluctuating
resources which are a feature particularly of renewable energy sources. This may be
possible by ensuring a satisfactory mixture of sources and loads to enable the demand and
supply to match.
1.3. Why microgrids?
A microgrid is a small-scale power supply network that is designed to provide power

for a small community. The definition of the ‘small community’ will be discussed in more
detail in Section 2, but may range from a typical housing estate, isolated rural
communities, to mixed suburban environments, academic or public communities such as
universities or schools, to commercial areas, industrial sites and trading estates, or
municipal regions. The key concept that differentiates this approach from a conventional
power utility is that the power generators are small (often referred to as micro-generators,
of a similar size as the loads within the microgrid), they are distributed and located in close
proximity to the energy users. The generators, and possibly also loads, are controlled to
achieve a local energy and power balance.
In the commercial world, these developments in distributed generation have been
compared with the changes currently taking place in the telephone industry [2] and it has
been estimated that, within a decade, the market for such equipment will exceed $60
billion a year. The US Electric Power Research Institute [3] and ABB even see the
emergence of ‘virtual utilities’ which, by analogy with the Internet, will allow intelligent
metering and switching and result in reduced environmental impact, greater system
reliability and lower operating cost. The motivation behind this study lies in the potential
of the microgrid concept to deliver a significant reduction of CO2 emissions, for the
following reasons:
† The use of both electricity and heat permitted by the close proximity of the generator to
the user, thereby increasing the overall energy efficiency.
† Significant environmental benefits made possible by the use of low or zero emission
generators including PV arrays and fuel cells.
† Low impact on the electricity network, by virtue of good match between generation
and load, despite a potentially significant level of generation by intermittent energy
sources.
1.4. Issues to consider
Although examples of networks similar to microgrids exist, there are technical and
regulatory issues that need to be considered before this concept can be applied on a
wider scale. The principal issue to consider is how closely the energy supply (both
electricity and heat) within the microgrid can satisfy the local loads. The answer to this
question will help decide how the microgrid interacts with the main utility, and the nature
of the connection to be determined. Indeed, it may even be desirable in some
circumstances for the microgrid to be disconnected from the utility, and operate as
‘stand-alone’. The issues that must be resolved to permit this type of operation include:
(i) Precise energy and power balance within the microgrid, on a time scale ranging
from milliseconds to years. Over the short time scale, the power balance is linked
to the question of control; over longer time scales, one needs to consider the
relationship between energy supply, demand and storage. Similar arguments are
used to design stand-alone power supplies, for example, photovoltaic or hybrid
systems which power remote equipment or serve isolated rural communities across
the world.
(ii) The nature of connection with the main utility (the ‘grid connection’). An
arrangement which would permit the microgrid operator the choice to operate in
the ‘grid connected’ or ‘stand alone’ mode is an uncharted territory for
conventional power utility engineers, and issues remain both at the technical and
regulatory level.
(iii) Energy storage. The conventional utility supply operates on the principle that power
is generated when it is required. Energy storage introduces a novel component in
a utility supply and broadens the design criteria. On a quantitative level, the size of
the energy store is intimately linked to the energy balance and to the required security
of supply provided by the microgrid.
(iv) Demand management. The temporal mismatch between generation and load can
be alleviated by managing the demand. The shifting of load facilitates achieving
the energy balance and helps reduce the size of energy storage. Whilst
experience exists of demand-side management at industrial level and lessons can
be learned from concepts such as storage heating, demand management at the
domestic level is attracting much interest in the research community [4] but
further experience is needed before routine applications become commonplace.
(v) Seasonal match between generation and load. Energy storage and demand
management can be effective to achieve energy balance at the diurnal time scale.
A sufficient energy must be available from the generators to ensure energy balance
over longer time scales if a microgrid powered by renewable or other intermittent
energy sources such as micro-CHP is to be capable of stand-alone operation. This
can usually be achieved only by a diversity of generation methods appropriate to
the load.
1.5. Structure of the paper
This paper addresses the technical and economic issues associated with the
integration of small generators into microgrids, serving to depict the essential features
of de-centralised electricity generation combined with on-site production of heat. An
overview of the field is given, including the technology and the economic and
regulatory framework. The technical aspects are discussed with focus on the
generators and load profiles, the utility connection and the control of the microgrid.
The regulatory framework is reviewed based on the current position of the UK
Government and views of the industry.
A principal novel contribution of the project is a model which allows a detailed analysis
of the microgrid in terms of the relationship between the energy supply, demand and
storage. It is shown by using the example of a microgrid consisting of domestic dwellings
in the UK powered by domestic micro-CHP generators and PV arrays how an optimum
mix between these generators can achieve a good seasonal balance between energy
production and demand, both for electricity and heat. Energy storage provides a bridge
between the daily generation and usage profiles. Although electricity storage in
rechargeable batteries is the most immediate option, more innovative solutions may be
developed in the future to make use of the effective thermal storage (hot water in the hot
water tank, heat retained in the fabric of the building). The management of demand
(for example, washing machines, tumble driers and dish washers) can be used to reduce
this storage significantly.
Economics of the microgrids is then considered in the light of the technology
analysis developed in the project. Several scenarios are identified for possible future
developments. The paper also raises some economic and institutional questions that
will need to be overcome if microgrids are to be deployed on a large scale in the UK.
2. Microgrids
2.1. The microgrid concept
The microgrid is a concept based around the assumption of a cluster of electrical and
thermal loads together with small scale sources of electrical power and heat [5]. The power
sources will generally be mixed, including renewable sources such as photovoltaic or wind
generators together with fossil fuelled generators meeting local heating requirements and
generating electricity (cogeneration). The connection between this network and the wider
electrical power network will be through a well defined and controlled interface. The
microgrid is responsible for servicing the needs of its consumers, ensuring a quality of
supply and possibly controlling some of the non-critical loads. The interface with the local
electricity utility will be one of exchanging power so that the microgrid will look like a
well behaved load or generator.
A potential advantage of this approach is that it should facilitate more imaginative
schemes for meeting the local requirements in a flexible manner with the small scale
generators and consumers closely integrated. For example, at the electrical utility scale,
load control to help match short term supply and demand has proved very hard to
implement. Although, the ownership and operation issues for the microgrid concept have
yet to be addressed, one possible way forward might be for a microgrid as intrinsically a
local ‘co-operative’ venture. In such systems, the consumers may also be the suppliers and
so a more imaginative approach to load control may be possible in the joint interests of
cost and efficiency. In particular, the microgrid will need to encourage consumers to
participate in small scale cogeneration, photovoltaics, and other renewable energy
schemes. Metering and charging arrangements within the microgrid would be agreed
locally and would have to reflect the market for power within the microgrid. The microgrid
concept is made possible by the recent advances in small scale, reliable generators, power
electronics and digital controllers that make it possible to reverse the trend to large scale
generation and bulk supply.
The key feature of the microgrid is that there should be local electricity generation
that matches the power requirements in the microgrid. There are various types of
generator that may be considered. Photovoltaic generators are attractive if the
environment is primarily residential since they may be incorporated into buildings in
an unobtrusive manner. Very small scale cogeneration schemes can be based on gas
boilers for central heating and domestic hot water. Possible technologies for this are fuel
cells or Stirling Engine powered generators. For a microgrid including commercial or
light industrial premises than larger cogeneration schemes based on gas turbines or other
prime movers may become appropriate.
Energy storage will probably be required to accommodate the variations of
available generation and power demand. Short term storage of electrical power will be
necessary to help accommodate the rapid fluctuations of load or generation that may
be anticipated on a relatively small network. Over longer time scales, energy
management made possible by storage can be used to make the most efficient use of
photovoltaic generation or the electricity produced by micro-CHP. Some energy
storage may be possible in the form of domestic hot water or as part of space heating.
The microgrid may exist as a remote power system in regions where utility supply is
not available. It may, on the other hand, be embedded in a larger electrical utility—
this would be the typical situation, for example, in the UK with a mature utility
power system. In this instance, two control issues need to be addressed: the
relationship between the microgrid and the local utility, and the question of internal
control. These topics are the subject of Sections 2.2 and 2.3.
2.2. The relationship between the microgrid and a local electricity utility
The intention is that the microgrid be self-sufficient, but for security of supply and
flexibility it would almost certainly be connected to the local electrical utility
network, or even to adjacent microgrids. These links may be bi-directional enabling
the import or export of electricity, or, depending on commercial considerations, it
might just be a unidirectional flow of power. From the point of view of the microgrid,
the utility connection might be viewed just as another generator or load.
This raises the question as to whether or not the microgrid should be linked to
other networks over a synchronous alternating current (AC) connection. The
advantage of a synchronous link would be its simplicity, requiring only an electrical
interconnection, circuit breakers and probably a transformer. Lasseter [5] has
considered this possibility and shown that in principle it should be possible to run
a microgrid with minimal central control of local generation which is able to operate
connected to the utility, or, in the event of loss of the connection, move smoothly into
stand-alone or island operation with no loss of power to the microgrid. What is
perhaps less clear is how the synchronous connection would be re-synchronised once
the utility was ready to re-establish the connection.
The alternative approach would be an asynchronous connection using a direct current
(DC) coupled electronic power converter. This might be bi-directional, enabling import
and export of power or simply a device to import power when local resources were
inadequate. An advantage of this approach is that it isolates the microgrid from the utility
as regards reactive power, load balance, etc. Only power is exchanged with the utility, the
microgrid is entirely responsible for maintaining the power quality (frequency, voltage
and supplying reactive power and harmonics) within its area.
With an asynchronous link the microgrid might be unusual that all its power will
be supplied through electronic inverters. Some generators, such as photovoltaic cells
are intrinsically sources of DC and hence need inversion to connect them to an AC
network. Others, for example, microturbines or Stirling engines may generate AC but
are not well suited to operate a synchronous generator because the frequency is
unsuitable or variable. Voltage source inverters with suitable control schemes will be
required to permit stable operation of the network with many small generators
attached. Fortunately, advances in power electronics and digital controllers mean that
sophisticated control strategies are possible and the cost need not be excessive. Which
of these approaches is more appropriate may well depend on the size of the
microgrid. It may also depend on the regulatory environment governing the
interchange of power between the microgrid and the utility.
2.3. Internal control of a microgrid
There are many commercial and political issues concerned with control; however,
the technical problems of a microgrid must be managed, for the concept to become a
reality. The control of a microgrid is intimately tied with the energy and power
balance in the microgrid, and the question of energy storage. There are three main
parameters—frequency, voltage and power quality—that must be considered and
controlled to acceptable standards whilst the power and energy balance is maintained.
2.3.1. Power balance

A power system usually contains no significant energy storage; the generated
and dissipated power must, therefore, be constantly kept in balance. There are specific
circumstances in which energy storage is significant, and these are discussed in
Section 2.5. This power balance must be maintained on a cycle-by-cycle basis if the
system is to maintain its frequency. Too much generation and the system accelerates, to
little and it slows; neither situation is acceptable. The permissible frequency deviation is
defined by Statute, and in the UK it is the responsibility for the NGC to ensure that this
deviation is not exceeded. Since the whole of the UK is run as one synchronous system,
any new generator means the disconnection of another or a rise in load, if the system
frequency is to remain constant. Power balance in a microgrid is, therefore, essential for
frequency control.
In a microgrid, frequency stability becomes critical; therefore, control is a major
concern. There are a number of techniques used to restore the power balance and hence
correct the frequency: the use of load shedding, increase in primary generation and
recovery of stored energy. All of these are available within a microgrid, but because the
system is small the problem is much more difficult to manage to the same standard as is
normal in a utility system.
Short-term storage of energy is needed to cope with the fluctuations in power
demand or accommodate the sudden loss of some generation. A microgrid with many
small generators will not be an intrinsically stiff system, unlike a national
interconnected utility. The small generators will neither store significant energy in
their mechanical inertia, nor will they necessarily respond quickly to sudden changes
of load. Short term storage, probably distributed with the generators, will permit the
inverters to follow the rapidly changing demand while giving time for the generators
to respond, or extra generation to be brought on line or for generators to be closed
down. This same storage could be used to help accommodate the diurnal variation of
demand.
There are two related issues, firstly quite small power imbalances will produce large
frequency excursions and secondly they will happen much more quickly. The first issue
may also be an advantage for a microgrid since small energy stores will have significant
effects. The second issue means that stored energy recovery must be fast and precise. Since
the most probable store, in the near future is likely to be a battery with an inverter, this does
not pose an insurmountable problem; such a system is quite fast enough to ensure adequate
frequency control.
2.3.2. Frequency
The UK power system operates at 50 Hz and there are obvious advantages in adopting
this frequency, whether there is to be a synchronous connection or not. The frequency
limits are laid down by law and are relatively tight though not to the same standards as
some other power systems. There is, however, no reason to adopt these standards and some
relaxation could be possible (in a non-synchronous system) if it were advantageous. It is
doubtful, however, that limits larger than G0.5 Hz could be acceptable. Frequency
therefore must be controlled to within these limits.
The ‘normal’ method of frequency control in the UK power system is by control of the
rotational speed of the synchronous machines supplying the power. Within a large
interconnected system, with many synchronous generators, no single machine can control
the frequency, there being a flow of ‘synchronising power’ into any machine that is slowed
in order to keep it in synchronism. There needs to be a large power imbalance to alter the
speed and hence frequency of the system, of the order of 5000 MW per Hz for the UK.
The fewer the number of machines, the less stiff the system and frequency control
becomes a technical issue. Machines in such a system must be able to respond quickly to
load variations in order to preserve the power balance at all times. This means rapid
detection of frequency change and fast, accurate control of load generation, or both.
Not all renewable generators are synchronous machines: wind turbines are often
induction generators and photovoltaic arrays connect to the system through inverters.
These two require very different frequency and load control in order to satisfactorily
operate in a system. Inverters can be used to control frequency since the inverter frequency
can be controlled independently of load. However, inverters do not behave as rotational
synchronous generators and require different philosophies.
2.3.3. Voltage
The system voltage within a large multi generator system is controlled by initially the
voltage of the machines but also by the reactive flow. In general, the reactive balance
becomes more critical in a smaller system. For example, all reactive demand must be
supplied from one generator in a single machine system. This is not strictly true, but adds
significantly to cost and control problems if reactive demand has to be compensated by
extra static plant.
A conventional distribution system is usually a feeder network, and there is little
interconnection. Voltage drop along feeders becomes an issue, as it will vary with load and
distance along the feeder. This dictates that any simple microgrid will have to be either
small to be satisfactory or be specially designed as an interconnected network.
The voltage and its limits at consumer’s terminals are specified by law, but they are
reasonably wide. With proper design, production of the correct voltage should not be an
insurmountable problem.
2.3.4. Power quality

Control of power quality will be the biggest issue for a microgrid. Voltage dips, flickers,
interruptions, harmonics, dc levels, etc. will all be more critical in a small system with few
generators. There will need to be a critical appraisal of both the effects and consequences
of relaxing and/or enforcing standards in this area.
As has been discussed by Venkataramanan and Illindala, the distributed generation

within the microgrid could enable better control of power quality [6]. With electrical
storage together with the distributed generation power quality could be maintained in
much the same way as is achieved by Uninterruptible Power Supply (UPS) systems. The
electronic inverters can not only supply power at the fundamental frequency, but also
generate reactive power to supply the needs of reactive loads, cope with unbalanced loads
and generate the harmonic currents needed to supply non-linear loads.
2.4. Energy balance
Little significance is usually attached to the concept of energy balance in a conventional

system: the solution is just to add more fuel over time. A microgrid which contains a high
proportion of intermittent energy sources—be it renewables such as PV or wind, or energy
sources controlled for other purposes such as micro-CHP—is not able to do this. The
energy available to the system is finite and depends on matters that cannot usually be
controlled or even predicted with any certainty.
If such a source is to be used and achieve levels of reliability similar to those of
conventional plant, energy storage is essential. It is also clear that, as the diversity of the
generation methods in any system is reduced, the role of energy storage becomes more
dominant. It, therefore, appears desirable that any microgrid should employ more than one
method of generation as well as some form of energy store.
The diversity of generation methods is particularly important if the microgrid is to
operate stand-alone. The microgrid must then contain sufficient generation capacity and
type that can supply adequate amounts of energy with sufficient reliability. Photovoltaic
arrays, for example, are a reliable power source during the summer months. They combine
well with generators such as micro-CHP which generate most power in winter to provide
heat for domestic dwellings. This point will be taken up further in Section 5 which
examines the sizing of generators and storage.
2.5. Energy storage
There is no economic general purpose method for the storage of electricity per se in the
quantities required for public utility use. There are of course methods involving capacitors
and super conducting magnets; both of which are technically complex and with present
knowledge, rather expensive, but nevertheless used in specific situations. Because the
direct storage of electricity is not very practical, the storage of energy by other methods,
for later use in electricity generation is employed. These are many and varied, depending
upon the situation and the purpose for which the electricity is to be used.
It is likely that a microgrid will rely on chemical energy storage in the form of electric
batteries. In the simplest of systems this will mean lead acid cells, which are well
developed, available, predictable and robust. For more sophisticated applications, redox
batteries are becoming available, and development will continue. In critical situations,
where cost is not an issue, the application of super conducting energy storage has been
used. Again, continued development is expected to both reduce costs and to increase
reliability. Over shorter periods of time, the use flywheels may be appropriate (Table 1).
Table 1
Possible energy storage technologies for a microgrid (adapted from www.energystorage.com)
Device Discharge time Power Comments

Batteries (lead Hours Up to tens of Technologies range from highly developed to
acid, NaS, MW research stage. A wide range of uses, lifetimes
Li-ion, NiCd) and costs
Regenerative Hours Hundreds of kW Some flow batteries used commercially; others still
fuel cells (flow to about 10 MW in demonstration stage
batteries)
Flywheels Seconds to min- Ten to hundreds Little maintenance, long life (10 s of thousands of
utes of kW cycles) and environmentally inert material. Can
bridge the gap between short and long term storage
Supercapacitors Seconds 100 kW–1 MW Lower energy density than lead acid batteries but
long life (10 s of thousands cycles); fast charge and
discharge capability
SMES Seconds to min- 10–100 MW Storage of electrical energy-no conversion
utes required; virtually no losses. Cryogenic hardware
required. Low volume production at present
Hydrogen A wide range of options, as part of the hydrogen economy. See, for example, Tyndall project
IT1.26 at www.tyndall.ac.uk for more detail.
The more common forms of large-scale utility storage (hydropower or even compressed air) are probably too
large and too site specific to be appropriate for most microgrid applications.
The calculation of battery size (energy), and inverter rating (power), will depend on the
size of the loads and generators within the microgrid, as well as its topography. As an
alternative to storing energy, the shedding of load is more likely to be used in a microgrid,
rather than a large scale public utility, because it is easier to identify those loads which are
least critical. Where cogeneration is used, some of this energy storage may well be in the
form of heat. This storage could be in the form of domestic hot water or stored for use in
space heating. Innovative control strategies can be developed to make use of this storage
and, if necessary, the plant may be run to meet the electrical load when there is no demand
for thermal energy.
3. Generators and loads
3.1. Introduction
The wide range of potential generators within the microgrid share one common
feature: they are small, and comparable in size to the loads within the microgrid. It is
also likely that several different types of generators will be present, adding a diversity
of generation to the diversity of loads. This issue is particularly important when
considering the seasonal aspects of the energy balance in the microgrid (discussed in
Section 5).
The focus of this study is microgrids comprising of domestic CHP units and
photovoltaic array. A broad overview of these generators is given in this section, as
appropriate to the energy balance purposes of this work. Also included is a brief overview
of a typical domestic load, and the effective averaging that occurs when the load of several
households is combined. The connection of small generators within the microgrid will be
an important integration issue, and the related utility connection aspects (recently aired
internationally with focus on small photovoltaic generators) are reviewed in Section 3.5.
3.2. Photovoltaic arrays
The photovoltaic array consists of one or several modules mounted on support

structures and electrically connected to form a DC power producing unit. The modules are
most commonly based on crystalline silicon solar cells although thin film modules using
amorphous silicon, cadmium telluride or copper indium diselenide (CIS) have also
recently become available. The array is connected to the utility distribution network
through an inverter. A comprehensive overview of photovoltaic power generation
technologies, including the system aspects, is available from two recent Handbooks on this
subject [7,8].
To maintain an optimum point on the power characteristic of the array, the DC interface
of the inverter incorporates a maximum power point tracker. The output from the PV array
is then closely related to its rated power Po and to the solar irradiance. The expected daily
energy produced by a PV array can be estimated from the expression
Ed Z Po ðPSHÞ (1)
where PSH—Peak Solar Hours—is equal to the mean daily solar radiation in kWh/m2 at
the site of the installation. Table 2 shows the typical values of mean daily solar radiation
for several cites with different climates. A more accurate estimate can be obtained by
correcting (1) for the inclination of the panels and reflection from the ground. The
reduction in cell efficiency on account elevated operating temperatures of the solar cells
and system losses should also be allowed for (see, for example, Chapter IIIa in Ref. [7]).
Photovoltaic arrays integrated in the roofs and facades of buildings are becoming
an increasingly common method of power generation within the urban environment.
Large-scale government support programs in Japan and Germany are driving the
industrial production but many other countries (including UK) provide subsidies for
Table 2
Examples of the mean daily solar radiation (in kWh/m2, equal to the Peak Solar Hours) for various locations in the world
(see Part1 in Ref. 27)
Mean daily radiation (kWh/m2)

Stockholm 59.35 N 2.52
London 51.52 N 2.55
Freiburg 48.00 N 3.04
Geneva 46.25 N 3.31
Nice 43.65 N 4.03
Oviedo 43.35 N 3.18
Porto 41.13 N 4.26
Heraklion 35.33 N 4.44
Sde Boker 30.90 N 5.70
Ilorin 8.58 N 4.19
the installation of photovoltaic generation. The perception of solar cells as a building

component has an impact on non-technical aspects of photovoltaics, for example, on
the required pay-back times for photovoltaic generators. This issue will be taken up
again in Section 6.
3.3. Fossil fuel power generators
3.3.1. Combined heat and power generators

Within the context of a microgrid all fuel powered generators of electricity may be
assumed to be included within a combined heat and power (CHP), or cogeneration,
scheme to ensure maximum energy efficiency. When connected to a power utility it is
usually assumed that in order to gain the maximum benefit these generators will run
primarily to match the thermal load with electricity as a by-product. In the context of
a microgrid this will not necessarily be the case since the generation may be required
to meet the electrical load. This raises two issues: does the ratio of electrical to
thermal load match the generator and is thermal storage possible, perhaps as domestic
hot water. There may also be a requirement that some generators are equipped with
alternative cooling arrangements so that the microgrid is able to meet the electrical
demand using fossil fuelled generators even if the thermal load is not required.
Reciprocating gas or oil fuelled engines are probably the best established prime movers
for small to medium scale CHP systems although micro-turbines or Stirling cycle external
combustion engines are alternatives that are being developed and trailed. An alternative
approach is to replace the heat engine prime mover with a fuel cell which is able to convert
the energy from the fuel directly to electricity while also providing a significant thermal
output (see Table 3).
3.3.2. Reciprocating internal combustion engines

These engines can broadly be split into two groups, compression ignition (diesel)
engines and spark ignition engines. Compression ignition engines run from diesel fuel
or heavy oil. They are well developed having been widely for many years for small
scale electricity generation, standby generators, etc. The principal concerns with
engines of this type is that they need heavy oil as a fuel and generate significant
pollution from oxides of nitrogen and particulates. Spark ignition engines may run
from a variety of fuels, but for domestic generation natural gas is likely to be most
attractive where it is available. Low frequency acoustic noise is also an issue for
internal combustion engines that are to be used in a domestic environment. Combined
heat and power systems based on these technologies and suitable for the domestic
scale are available from a range of suppliers [9].
An attractive feature of these power sources is that they can start-up quickly and
respond rapidly to changes in load.
3.3.3. Micro-turbines
Small scale gas turbines are being developed for small scale generation, although much
of the effort has been aimed at generators in the region of 30–100 kW (electric),
significantly larger than the size envisaged in this study. Smaller generators of order
Table 3
Fossil fuel power sources for a microgrid
Prime Fuel Ratio of Advantages Disadvantages Commercial

mover thermal to status
electrical
power
Internal Diesel, 2–3 Well established Noise, emission of Available now
combustion heavy oil, technology, good NOX and particu-
engine natural gas load following, lates esp. from
rapid start-up, compression
maintenance skills ignition engines
readily available
Micro-turbine Natural gas w2 Lower noise than High speed and Developmental
internal combustion temperature in small sizes,
engine, efficient requires sophisti- larger ratings
combustion with cated materials, available
low emissions of may be difficult to
NOX, fast start and achieve efficiency
good load following in small sizes
Sterling engine Natural gas, 3–6 Low noise, efficient Load following less Prototype sys-
other fuels combustion and low good than for tems being sub-
possible emissions, robust, internal combustion jected to large
low maintenance engine scale tests
Fuel cell Natural gas, 1.5–2 Only moving parts Relatively complex Developmental
hydrogen for auxiliary sys- because of need to
tems (pumps/fans), reform fuel if natu-
have the potential ral gas is used, slow
for high electrical start-up especially
efficiency for high tempera-
ture cells
2–5 kW have been demonstrated [10], but there are significant challenges in producing a
system which is economically viable. The main attractions of the gas turbine approach are
efficient combustion, flexibility of fuel and potentially low noise and vibration. There are,
however, significant engineering challenges from the high temperature and high speed
required for the turbine.
There is also significant activity related to larger microturbines with power ratings
between 25 and 250 kW, in which a permanent magnet generator is directly coupled to a
small gas turbine. A recuperator is normally used to recover some of the exhaust heat and
boost efficiency to around 30–35%.
Leading companies in the area include Bowman Power Systems Ltd (35–60 kW),
Capstone (28–60 kW), and Turbogenset (60 kW?), Elliots (80 kW) and Pratt and Whitney
(80 kW). A good overview of the state of the art is presented in books written by Hamilton
[11] and Moore [12] and a paper by Nicholas [13]. Davis [14,15] makes the case for using
the technology in a Microgrid system.
Tests on actual systems produced by Bowman and Capstone [16,17] revealed that these
systems are still not very reliable. In general, the technology is viewed as promising, but it
is still in its infancy with many teething problems.
3.3.4. Stirling engines

The Stirling Engine has recently attracted attention as a power source for small
generators. The attraction of this prime mover is that as an external combustion engine
almost any heat source may be used (e.g. solar power has received considerable attention
with possible applications in space). The external source of heat means that for
conventional fuels the burner may be optimised to achieve efficient combustion and low
pollution.
The Stirling Engine has a thermodynamic cycle that requires a relatively complex
mechanical system usually involving two pistons, or a piston and a displacer. Two types of
engine are recognised: the kinematic engine which uses a conventional crank mechanism
for the pistons and produces a rotary motion and the free piston engine in which the piston
and displacer oscillate linearly on springs. To produce electricity the kinematic engine will
drive a conventional rotary generator. The free piston engine generally requires a linear
generator. Both types of Stirling engine are being developed for domestic scale CHP
systems [9].
Stirling engines have the advantages of potentially high thermodynamic efficiency (the
Stirling cycle is reversible), continuous external combustion and low noise. However, the
complexity of the motion, the need for a thermal store (regenerator) and difficulties with
lubricating the piston and displacer are challenges to the engineer. As an external
combustion engine the thermal capacity of the system will limit the speed at which it can
start-up and the speed with which it can follow variations in load.
3.3.5. Fuel cells

Fuel cells generate electricity directly from the reaction between fuel and oxidant
without the need for a heat engine and a mechanical system. A number of fuel cell types
have been developed, some require high temperatures to operate, e.g. the solid oxide fuel
cell (SOFC) requires a temperature of 800 8C, others require lower temperatures, e.g. the
proton exchange membrane (PEM) cell requires a temperature of up to 90 8C. Operation is
complicated by the need to convert hydrocarbon fuels into hydrogen in a reformer before
the fuel can react in the cell.
A domestic scale CHP system based on a PEM fuel cell has been demonstrated [18].
This system operated at a power level of 4 kW electrical power and 6.8 kW thermal. This
paper highlights some of the problems associated with current fuel cell technology,
particularly the complexity and inefficiency associated with the fuel processor required to
generate hydrogen from natural gas. The fuel cell also had a poor load following capability
and was reinforced with battery storage to allow the system to follow the load and
accommodate short duration peak loads.
3.4. Electrical power generation
Internal combustion (IC) engines and kinematic Sterling engines may use conventional
rotary generators. With available micro-CHP systems based on IC engines, induction
generators seem a popular choice [9]. While these are attractive on grounds of simplicity,
cost, low maintenance and ease of starting up they are more appropriate for use connected
to a stiff utility network rather than a microgrid. Synchronous generators may be more
appropriate in this case.
All the other sources will need electronic conversion to achieve a conventional
50/60 Hz supply. Micro-turbines operate at high rotational speed giving a high frequency
from the generator which must be rectified and converted electronically to AC. Free piston
Sterling engines most simply use a linear generator. The frequency of generation depends
on the resonant frequency of the piston and displacer systems in the engine. Hence, this
type of generator is unlikely to be suitable for synchronous generation. Fuel cells are
intrinsically generators of DC and so will require a power inverter.
It is clear that power electronics will play a key role in a microgrid to provide efficient
conversion from the generator to link it to the network. Even for conventional rotary
generators it may be preferred to use an asynchronous or induction generator (with a
suitable inverter) and then use a DC linked inverter to connect the power source to
the network. The power electronics can also be used to control the generation of reactive
power to meet the needs of the load.
3.4.1. Inverters
Inverters for utility connection can be broadly classified into two types: single-phase
inverters and three-phase inverters. Detection of islanding (see Section 3.4.2) is much
easier in a three-phase than a single-phase inverter, although inverters that are rated at a
power below 5 kW are mostly connected to single-phase networks.
Most modern inverters which are used for connection of small generators employ
controlled power switches (e.g. MOSFET, IGBT), and generally use pulse width
modulation (PWM) control signals for producing an AC output. Previous thyristor based
systems were turned-off using the ‘zero crossing’ of alternating current from the mains. In
either case, inverters need to synchronise with the utility network and cease to energise the
utility network on islanding.
Fig. 2 shows a schematic structure of a self-commutated PV inverter with a low
frequency transformer to provide isolation from the grid. In order to reduce the weight
and the price, a high frequency transformer is also sometimes used but it is uncertain
whether DC injection can be maintained sufficiently low to satisfy the most stringent
regulations.
Fig. 2. Schematic diagram of an inverter.

3.4.2. Utility integration issues

The connection of micro-generators to the public supply system has been an issue under
much discussion over the recent years, and the progress made in resolving the technical
and legal issues is likely to play a significant role in the development of a similar
framework for the microgrid.
Although there are differences in these grid-connection procedures presently applied in
different countries and by different utility companies, there are many common attributes.
The network operator will generally require that the connection of a generator conforms to
the relevant codes of practice and engineering recommendations, particularly with respect
to safety. There must be adequate protection for both the supply network and the inverter.
The power quality will also have to be sufficient not to affect adversely the utility
equipment and other users connected to the network.
The main technical barrier for utilities has been the issue of ‘islanding’ where the power
to a local area of the electricity network could potentially be maintained live by a
distributed generator, even where the main supply is switched-off or lost in a fault.
Alternatively, islanding can be considered as the condition in which a portion of the utility
system, which contains both load and generation, is isolated from the remainder of the
utility system and continues to operate.
The commercial terms for payment for exported energy are contained in the Tariff
Agreement which varies from country to country and electricity company to electricity
company. In a number of countries (for example, Germany, Spain and Switzerland), the
utilities are obliged by law to pay a premium price for PV electricity. Further information
on year-by-year changes is available from an annual survey published by IEA Task 1.
One forum which has been involved in the study of these problems is the Task 5 of
the Photovoltaic Power Systems (PVPS) Programme of the International Energy Agency.
Table 4
Summary of the current technical and regulatory framework for the connection of small generators in the UK
Authorisation procedure Installations need to be agreed with the local Distribution Network Operator
(DNO). For larger units the DNO may require on-site commissioning tests and may
also want the opportunity to witness the commissioning tests. However, the use of a
type approved inverter for smaller units greatly simplifies the process. A connection
(operating) agreement also needs to be in place with the local DNO and an
appropriate supply (tariff) agreement with an electricity supplier if any export
settlement is required
Engineering Guidelines The original Engineering Recommendation governing embedded recommendation
G59 and G59/1 was supplemented by G77 and G77/1 (2002) which addresses single
phase generators up to 5 kVA. G77 has been superseded recently by a broader based
G83. The mandatory protection features in G77 and G83 include disconnection of
the inverter based on over and under voltage, over and under frequency, and a
recognised loss of mains technique, such as vector shift or frequency shift. Active
techniques that distort the waveform beyond harmonic limits or that inject current
pulses are not allowed
Legal and tariff situation Electricity suppliers are not obliged to buy electricity produced by a small
distributed generator. Tariffs paid for electricity fed into the grid vary with supplier,
however, the rate is normally less than that for imported units. Some suppliers now
offer the same price. A separate meter for energy fed into the grid is required
A number of reports have been published and Engineering Guidelines have been prepared
to facilitate these procedures. The general technical, legal and commercial aspects of
regulations in the UK is summarised in Table 4. The situation in several other countries
can be found in the IEA Report [19]. A possible connection scheme for photovoltaic
generators is shown in Fig. 3.
Fig. 3. A diagram showing the connection of a small generator in a domestic installation on the example of a PV
system ([7], Chapter IIIc-1). (a) The generator. (b) Connection scheme of the meters and consumer units suitable
for countries (notably Spain) where all PV power is exported to the network. (c) An alternative scheme where
only the excess power not used on-site is exported to the network. The case (c) is common, for example, in the
UK, and a PV generation meter is then often installed on the AC side of the inverter. Combined export/import
meters are now also available.
Fig. 4. The typical electric load profile of a household over two days.
3.5. The domestic load
The profile of electricity consumption and heating requirements of households are

discussed in detail in Section 4. These smooth profiles are obtained by averaging over
many households, and are applicable to sufficiently large microgrids. For smaller
networks, peaks due to individual appliances will appear in the profile, presenting a
challenge to the control of the power balance in the microgrid. Typical load profiles which
Fig. 5. The domestic load profile averaged over 100 winter days, compared with the standard winter domestic
profile produced by the Electricity Association.
are likely to arise have been studied by monitoring the actual consumption of a household
over the period of several months.
We have seen in Section 2 that the power profile of the demand is an important
consideration for the control of the microgrid. Fig. 4 shows the electric power consumed
by this household over a typical day. To examine the averaging or ‘diversity’ effect, Fig. 5
shows the average load profile averaged over 100 days. Whilst only an approximate
representation of the load profile of 100 households during one day, this profile is now seen
to be relatively smooth, and approaching the standard domestic load profile quoted by
the former UK Electricity Association. The electricity consumption by the electric
shower—which represents the largest load (7 kW) of this household—remains visible, and
a larger average (corresponding to a larger microgrid) is needed for it to ‘blend in’ to
obtain a smooth profile overall.
In a microgrid with energy storage, short-term fluctuations in energy balance are
compensated by the battery. It will be seen in Section 5 that the natural design procedure
for microgrids which contain photovoltaic arrays provides for sufficient storage to cover
variation in the hourly demand during one day. The fluctuations in the daily energy
consumption then become important and, as seen in Fig. 5, these are appreciable at the
single household level. The total fluctuations in daily electricity consumption (Fig. 6) in a
larger microgrid, however, will be less significant. Adopting a similar approach as for the
power fluctuations, an estimate has been obtained by taking the average of the total energy
demand over a number of days, and considering the relative fluctuations d, where
hEn2 i K hEn i2
d2 Z (3.2)
hEn i2
where En is the energy consumption over n days. It is seen from Fig. 7 that d can be
represented by the square root dependence
Fig. 6. The daily household load as a function of time.

1 0:46
d y pffiffiffiffiffiffiffiffiffiffiffiffiffiffi Z pffiffiffi (3.3)
4:655n n
as one would expect from the law of large numbers for n independent loads. Although
further work is needed to substantiate the validity of (3.3) in general, it is seen that
the random variations in the electricity demand in a microgrid with over hundred houses
are small, at the level of a few percent. Further work is also required, to characterise
regular variations over, for example, weekends and public holidays, which have not been
singled out here for a special analysis.
Fig. 7. The inverse square of the relative deviation as a function of the number of days of load, approximating the
size of the microgrid.
4. Modelling energy consumption in buildings
4.1. Introduction
This Section focuses on the modelling of the load profile as a function of time for the
housing sector. This has been grouped into two categories:
1. Domestic appliance electricity consumption load profile.

2. Space heating and hot water load profile.
The electricity profile is used to validate the methodology as a comparison is available

with the average domestic electricity profile, available from the former Electricity Asso-
ciation. The validated methodology is then used to produce a heating energy load profile
(space heat and hot water) which forms the basis of the sizing calculations in Section 5.
The energy consumption by domestic appliances has a high correlation to people’s
customs and may be slightly influenced by season. The electricity load profile depends
Table 5
Domestic appliances
Brown goods Electric consumer goods-TVs, VCRs, music centres and satellites and cable TV
equipment
Cold appliances Refrigerators, freezers and combined fridge-freezers
Cooking appliances Electric ovens, electric hobs, kettles, microwaves and small cooking appliance
Washing machines
Wet appliances Tumble dryer
Dishwashers
Cleaner Vacuums
Miscellaneous 1 Irons, electric showers, PCs and other office equipment
Lighting Lights
Miscellaneous 2 Central heating pumps
mainly on the household size and occupancy pattern. Sufficient daylight in summer was
considered when calculating electricity lighting load. Space heating load profile mainly
depends on the weather conditions, building characteristics, occupancy pattern, income
level, etc. Domestic hot water depends on the household size and on the occupancy pattern.
The United Kingdom has a population of 57 million (1996) and a total of 23.5 million
households. The model of electricity load profile is based on the household size of three—
the average household size. [20] Space heating load calculation is based on a typical UK
three-bedroom house.
4.2. Domestic electricity load profile
The domestic appliances are listed in Table 5. Lighting and central heating pump
electricity consumption varies in winter and summer. The common UK occupancy
scenarios are shown in Table 6. Table 7 indicates the lights on/off pattern for each
occupancy scenarios in summer and winter. The statistical data from UK Electricity
Association Load Research have been analysed, and the national average electricity load
profiles for the different household size have been generated. The load profile for the urban
group can be calculated using the equation:
X 5
Loadurban Z Fi Loadi (4.1)
iZ1
Fi is the percentage of UK national household in the size category (see Table 8), iZ1–5
is the UK household size, and Loadi is the hourly load profile of each size of household.
Table 6
The occupancy styles for a three-person family
Scenarios Type Unoccupied period

Scenarios 1 Part time working morning session 9:00–13:00
Scenarios 2 Full time working 9:00–18:00
Scenarios 3 Part time working 9:00–16:00
Scenarios 4 Not working N/A
Scenarios 5 Part time working afternoon session 1/2 13:00–18:00
Table 7
The electric lighting patterns
Hour Winter Summer

S1 S2 S3 S4 S5 S1 S2 S3 S4 S5
1 Off Off Off Off Off Off Off Off Off Off
7 On On On On On Off Off Off Off Off
9 Off Off Off On On Off Off Off Off Off
16 On Off On On Off Off Off Off Off Off
17 On Off On On Off Off Off Off Off Off
20 On On On On On On On On On On
4.3. Heating load profile
The hourly heating load profile for the UK coldest month (January) was simulated using
the thermal simulation tool developed at the Martin Centre. The simulation for different
house types (see Table 9) has been performed. The building information is listed in
Table 10. The heating load profile is shown in Fig. 8.
Table 8
Percentage of UK household size
1 person 2 person 3 person 4 person Five person

26% 39% 29% 5% 1%
Table 9
Typical 3-bedroom house types
3-Bed flat 3-Bed semi-detached 3-Bed detached 3-Bed mid-terrace

Table 10
Building information used in the modelling
Climate UK Kew 67 Solar: beam, horizontal diffuse Air temperature

Dimension Length, 5 m; width, 8 m; height, 2.7 m for a house (2 storey). Length, 10 m;
width, 8 m; height 2.7 m for a flat
Orientation South–north
Windows South window 6.5 m2; north window, 5.4 m2 (double glazing)
Thermal mass Medium
4.4. Domestic hot water
The domestic hot water (DHW) system is usually designed to provide approximately
50 l per head per day. The energy used for domestic hot water depends on many factors
such as the required water temperature (currently, the average is 50 8C), the amount
required per person, and the household size. The load profile will depend on the operating
schedule, which depends on the family occupancy pattern. The method is similar to the
electricity load profile calculation. The average daily DHW for a 3-person household has
been calculated and is shown in Fig. 9.
4.5. Validation
The space heating load profile was produced using the thermal model developed at the
Martin Centre. The model has been validated by the simulation software Espr [21].
There is a relatively close agreement between the modelled and statistical data for the
3-person profile (see Fig. 10). The correlation coefficient for these two profiles is 0.7. The
electricity profile for a 3-person household has been compared with the UK national
statistical data (see Fig. 11). From the figure we can see that the electricity consumption
Fig. 8. The heating load profile in winter.

Fig. 9. Daily DHW demand of a 3-person household.
given in the national statistical data is higher than the modelled one. This is probably
because the national statistical data include a proportion of electricity heating during the
night. However, the modelled data are just related to the domestic appliance.
4.6. Discussion
The method of predicting the electricity profile is based on five approximate scenarios.
Since the different houses have different daily load shapes, the peaks in the aggregate load
shape will not be so pronounced as for each individual house. When the five approximate
scenarios are aggregated, the load profile shape appears as a relatively smooth curve
throughout the day. When compared with the UK national statistical data, this averaged
curve shows a good agreement. The creation of a set of demand models corresponding to
actual pattern using stochastic techniques should be the subject of a future study.
Fig. 10. Correlation between the model and national statistical data for a 3-person household.
Fig. 11. Comparison of the modelled electricity load profile and the national statistical one.
The space heating load profile is the result of a simulation result for a 3-bedroom
house/flat specified by a typical building information and UK Kew 67 climate data. The
model itself can simulate different sizes and types of buildings.
5. An example of sizing: microgrids powered by photovoltaics and micro-CHP
5.1. Introduction
If operating stand-alone or isolated from the main public network, the energy generated
within the microgrid must equal the consumption. This energy balance can be considered
over a range of time scales. The energy balance over very short time scales—the power
balance—was discussed as part of the microgrid control in Section 2. This Section
considers the energy balance over longer periods of time: the hourly energy balance during
one day, and the daily balance over the period of one year. The requirements of energy
balance over these two disparate periods of time have different impacts on the design and
configuration of the microgrid. Central to this analysis is the issue of energy storage which
needs to be provided to bridge any imbalance between the generation and load, and the
necessity that sufficient energy be available over time scales which cannot be bridged by
storage.
As an illustration let us suppose that there is a systematic imbalance between the energy
generated and consumed in the microgrid during one day. Over a period of weeks or
months, this imbalance will increase by a sizeable factor, producing a large energy deficit.
To cover this deficit, the energy storage would need to be unrealistically large. For this
reason, the arguments developed in this Section will be based on the requirement that the
mean daily imbalance in the microgrid is zero. We shall show in Sections 5.2 and 5.3 that
this requirement can be achieved by an appropriate composition of generators in the
microgrid—for example, by a suitable ‘mix’ of micro-CHP and photovoltaic arrays.
It is unrealistic to expect an exact hourly balance between energy produced by these

sources and the user demand, and this needs to be supplied by energy storage. The size
of the battery can be estimated from the average generation profile during the day
(Section 5.4).
5.2. The methodology
On a daily basis, the energy balance in the microgrid powered by several generators can
be written as
X
D% En (5.1)
n
where D is the daily electricity demand and En is the energy produced by generator n. For
renewable generators, the energy production is closely linked to resource availability, and
can be approximately assumed to be equal to the product of the daily resource energy Rn
and the rated power Pn the generator:2
E n Z P n Rn (5.2)
Although approximate, Eq. (5.2) has had much success in practical use [22]. For an
improved accuracy, the efficiency Pn should allow for the dependence on operating
conditions.
Eq. (5.2) can be illustrated on the example of the daily output from the PV generator. In
this instance, the resource energy is often quoted as Peak Solar Hours (see Section 3.2) and
the efficiency is equal to the peak power of the array under the irradiance of 1 kW/m2 at
standard conditions. In the case of wind turbines, the rated power is a suitable average of
the product of the power coefficient and the swept area by the rotor; the resource energy R
is the mean daily wind energy per unit area.
A different reasoning needs to be applied to micro-CHP. Let us suppose that a single
type of micro-CHP generator is employed which produces electricity and heat with
efficiencies he and hh, respectively. The overall energy efficiency is thus heChh. Each
micro-CHP generator supplies heat to the household where it is installed but feeds the
electricity to the network of the microgrid. If—in households where it is installed—the
micro-CHP generator is the only source of heat for space heating and hot water, it must
cover the total thermal energy demand Q of the household. The electricity produced by
this generator is thus heQ/hh. Hence:
he
EZ Q (5.3)
hh
One can, therefore, write (5.1) in the concise form
X
D% Pn Rn (5.4)
n
2
For some generators (for example, wind turbines), ‘rated power’ has a somewhat different meaning than in the
power generation industry. See below for further details.
where the ‘driving forces’ R are the renewable resources or heat demand in the case of
micro-CHP, and P is the rated power in the case of renewable generators and the ratio of
efficiencies in the case of micro-CHP.
If sufficient energy is to be available for stand-alone operation, Eq. (5.4) must apply at
any time of the year. In practice, it is usually applied to seasonal or monthly averages of
R’s. The system of Eq. (5.4) then represents a set of equations which, subject to economic
constraints, defines the composition of generators in the microgrid.
5.3. Daily energy balance in a microgrid
The above formalism has been used in the past in the design of hybrid photovoltaic/
wind energy systems [23], and will be applied here to microgrids consisting of
photovoltaic and micro-CHP. For two types of generators, Eq. (5.4) have a simple
graphical solution for the rated powers PPV and PCHP, as shown in Fig. 12. Here, the points
in the Cartesian plane corresponds to the configurations (PPV, PCHP) of the hybrid system;
those that satisfy the inequalities (5.4) lie in the shaded area of the plane, bounded by two
lines determined by the two inequalities in (5.4).
It follows from the linear programming theory that the optimum configuration lies in
one of the vertices of this region. Thus, there are three possible solutions: a micro-CHP
unit or PV array alone, and a combined system. We shall see that the combined system is
likely to be the preferred solution as the PV array size would need to be very large to
supply all the electrical demand on a winter day. Similarly, the micro-CHP electrical
output (and, therefore, the overall size) would need to be large if it were to supply all the
electricity during summer when the heat requirement is low. We thus have a convenient
methodology which can be used to determine the required complement of generators in
the microgrid.
Fig. 12. A graphical solution of the energy balance method for a microgrid consisting of PV arrays and a micro-CHP.
To this end let us suppose for simplicity, that the microgrid employs a single type of
micro-CHP generator. Eq. (5.4) then becomes
X h X
D% ðPSHÞ PPV C e Q (5.5)
hh
P
where PPV denotes the total size of the PV array in the microgrid, and the sums in the
second term on the right hand side goes over all households with micro-CHP generators.
Dividing (5.5) by the total number of households N
he
d% ðPSHÞPPV C Qn (5.6)
hh
where d is the average electricity demand of a household, PPV is the average size of the PV
array, n is the fraction of households with micro-CHP generator, and Q now refers to the
average full heating load.
To obtain numerical estimates, let us consider Eq. (5.6) for the summer and winter
seasons only. Denoting by QHW the average domestic daily hot water demand and by QSH
the space heating demand on a cold winter day, we then have:
he he
ds % ðPSHÞs PPV þ Q n dw % ðPSHÞw PPV þ ðQ þ QSH Þn (5.7)
hh HW hh HW
To obtain numerical estimates, the ratio he/hh of the electrical to thermal power
was taken as l/3 as might be expected for a typical thermal engine (see Section 3.3).
As fuel cells become available with time, this ratio might increase to 0.5–0.6. The
first units on the markets, however, are likely to be configured for the ratio of
electrical to heat demand of a typical household rather than of the microgrid, with a
considerably lower ratio he/hh. The effect of temperature on the output from the PV
array as well as any inefficiencies in system operation have not been explicitly
highlighted in the analysis but can be included without difficulty in the definition of
the rated power of PV array, as discussed in Section 3. Mean daily solar radiation
values from the European Solar Radiation Atlas [24] have been used. The input data
are summarised in Table 11.
Table 11
Data used in the modelling
Quantity Value Abstract Units Remarks

QHW 8.6 kWh Martin Centre model (see Section 4)
QSH 58 KWh Martin Centre model (see Section 4)
dS 8.3 KWh Electricity Association data
dW 11.5 KWh Electricity Association data
hc 0.2 Estimate based on Section 3
hh 0.6 Estimate based on Section 3
(PSH)s 5.1 KWh/m2 South of England; panel inclination 408 facing
South [24]
(PSH)w 1.1 KWh/m2 South of England; panel inclination 408 facing
South [24]
Table 12
Results of the study giving the sizes of the PV and micro-CHP generation capacity
PV array per household (kWp) n (micro-CHP generators per household)

1.37 0.45
The results are shown in Table 12 and the full range of microgrid configurations
determined from (5.7) depicted in Fig. 13. It is seen that the optimum size of a PV
generator suitable for an average house in combination with micro-CHP is about 1.4 kWp.
It is also interesting to note that, as long as the heat output from the micro-CHP generator
is sufficiently large to provide the heating needs of the house, it does not enter into the
configuration of the microgrid. The relevant quantities are the efficiencies he and hh and
the ‘penetration ratio’ n. For a typical housing estate, n equal to about 0.5 is sufficient to
supply all the electric power of the microgrid. Clearly, this is determined by the ratio of
the winter heat demand (67 kWh/day) to electricity demand (11.5 kWh/day) of the typical
house. Allowing for the ratio of the heat to electrical efficiencies of the micro-CHP, we
obtain a roughly twice the daily electrical energy output than is consumed in the house. In
a microgrid this electricity can be distributed among the users in the microgrid. If the
micro-CHP units are installed individually without integration, the surplus electricity must
be exported to the utility. Alternatively, a micro-CHP unit with a lower value he/hh should
be installed which does not benefit from the optimum efficiency of electricity generation.
For a larger n, of course, the power can be exported to the public network or a lower
number of units can be installed. When fuels cells become commercially available, for
example, the ratio of households that need to install this type of micro-CHP can be reduced
to about 15%.
Fig. 13. Possible microgrid configurations for the parameters shown in Table 5.1. Configurations in the shaded
region generate sufficient energy to supply load throughout the year. The configuration shown by point represents
the optimum system.
Fig. 14. The daily generation and load profile during a typical winter day.
5.4. Hourly energy balance. Energy storage
We now turn to the energy balance during one day. The basis for this analysis will be
average profiles of the electricity demand available from the former Electricity Association
and average generation profiles by the micro-CHP and the photovoltaic array. The former is
assumed to be proportional to the average heat demand Q and given by n (he/hh) Q, where Q
consists of space heat and hot water demand, as described in detail in Section 4. The
irradiance during a typical day with daily total solar irradiation shown in Table 11 was
obtained from the measured data at the STaR Facility, University of Southampton [25].
The results are summarised in Figs. 14–17. Power generation and demand profiles
over typical days in summer and winter are shown in Figs. 14 and 15, using system
Fig. 15. The daily generation and load profile during a typical summer day.
Fig. 16. The cumulative energy balance per household during a typical winter day, illustrated on the energy stored
in the battery during the day. This graph shows that a battery of about 2.7 kWh is needed to cover the mismatch
between generation and load.
configurations discussed in Section 5.3. Fig. 16 shows the cumulative energy balance
during a typical cold winter day, plotted as the energy that would need to be stored to
maintain electricity supply. It is seen that, in this case, a relatively modest energy storage
of approximately 2.7 kWh would be needed. Fig. 17 shows the cumulative energy
balance for a summer day. It is seen that the requirements for energy storage estimated
for winter and summer days are almost identical.
Fig. 17. The cumulative energy balance per household during a typical summer day. It is seen that a similar
battery size (2.7 kWh) will cover the mismatch between generation as on a winter day.
5.5. Security of supply
5.5.1. Introduction
The security of supply in a traditional power system is determined principally by the
likelihood of equipment failure. In microgrids which are powered by renewable sources,
resource availability becomes an additional issue.
The close proximity of generators and load in a microgrid should improve both
efficiency and reliability since there are fewer items of plant between source and load.
Also, since there is expected to be a relatively large number of small generators this should
help improve the reliability. Since there are unlikely to be overhead transmission lines,
lightening and storm damage should not be a problem. It is also worth noting that the area
affected by a failure will be limited to the microgrid if it is internal and the microgrid may
be isolated from failure outside. This avoids the type of problem seen recently in London
where a single failure propagated across the system causing large scale blackout, with
even more dramatic failures in North America. The diversity and number of generators
should also make for improved reliability. The storage could also help cope with short
term problems. There is uncertainty over the reliability of small machines as compared
with large ones, but this may be offset by time to repair.
The principal consideration security of supply of a system designed by the average
daily energy balance considered in Sections 5.2–5.4 is the likelihood of energy supply
exceeding the load; an energy deficit can come from insufficient generation or from
excessive demand. In a system that relies solely on the renewable energy resource, a
high security of supply can only be achieved by installing a larger array and/or
energy storage. The operator of the microgrid, however, has also other the options. It
may be possible to make an arrangement with the local utility to trade the surplus and
deficit of electricity to provide a back-up supply. In stand-alone operation, the micro-
CHP units can be run over above the times necessary to provide the heat. To examine
the extra cost, it is necessary to determine the energy deficit which is likely to be
experienced with the two power sources considered in this Section. The deviation in
demand from the average value was considered in Section 3.5. In a reasonably large
microgrid (over a hundred houses or so), the energy deficit due to random daily
fluctuations should be small (of the order of a percent or two of the total CHP
generated electricity) and can be neglected.
As already noted in Section 3.5, however, a more accurate analysis is needed to
examine separately regular events such as weekends and public holidays where correlated
deviations from average occur in both the heat requirement and electricity consumption
due to specific occupation patterns.
5.5.2. Photovoltaic generation

Solar radiation is the principal source of energy for the microgrid outside the heating
season. Rather than consider the complexities of the full probability distribution of daily
solar radiation, a sufficiently accurate estimate can be obtained by using the following
expression for the expected annual energy deficit DEPV based on Gaussian statistics:
sG
DEPV Z E (5.8)
2Gd PV
where sG is the standard deviation of the distribution from the average daily solar radiation
value Gd, and EPV is the photovoltaic energy produced.
Using the solar radiation data for the period from the beginning of May to the end of
September, the standard deviation sG has been found to be
sG Z 0:40Gd (5.9)
The result (5.9) has been obtained using the measured 1997–1999 solar radiation data at
Southampton [25]. The frequency distribution of the mean daily solar radiation data is
shown in Fig. 18.
5.5.3. Micro-CHP
In addition to the occupancy pattern which is discussed further in Sections 5.4 and
5.5.1, the principal reason for fluctuation in the daily energy produced by the micro-CHP
are likely to be fluctuations in temperature. A simple estimate of the energy deficit can be
obtained by assuming that the energy required to heat the house is proportional to the
difference in temperatures between the inside (Ti) and outside (To) the house. Assuming
again Gaussian statistics, this leads to the expression for the expected annual electrical
energy deficit DECHP
sT he
DECHP Z Q (5.10)
2fhTi i K hTo ig hh SH
Fig. 18. The probability distribution of daily solar radiation (in kWh/m2) in Southampton (UK) from beginning of
May to the end of September during the period 1997–1999. Data from STaR Facility [25].
Fig. 19. The probability distribution of the average daily December temperatures (for Southampton, UK, in
degrees Celsius) during the period 1997–1999. Data from STaR Facility [25].
where sT is the standard deviation from the average daily temperature and QSH is the
space heat requirement. Eq. (5.10) should be considered on a monthly basis before the
results for ECHP are combined. For simplicity, the value of DECHP from an analysis of
December data for Southampton for the period 1997–1999 [25], with the result
sT Z 3:5C hTo ðDecÞi Z 6:12C (5.11)
The frequency distribution of the average daily temperatures at Southampton is
shown in Fig. 19.
5.6. Conclusions
We have shown, by considering the energy balance in a microgrid on typical summer

and winter days, that a photovoltaic array of about 1.5 kWp on every roof combined with a
micro-CHP unit in every other house, will provide sufficient electricity to satisfy the
demand of the households in microgrid. Energy storage of about 2.7 kWh per household is
also needed to make this electricity available at the appropriate time of day. The micro-
CHP ownership can be reduced to some 15% when fuel cells become available as a
domestic CHP generator.
Although the size of the required energy storage is not large (it corresponds to three
or four typical traction batteries) it can be reduced further by demand management. For
example, by setting the washing machines, tumble driers and dish washers to run
during the early morning or afternoon generation peak on a winter say, the required
energy storage can be reduced by about 1.4 kWh. A similar result can be achieved by
switching on non-critical loads during the PV generation peak on a summer day.
A further reduction in the battery size (by some 1 kWh) is possible by making use of
the energy stored in the hot water tank and in the fabric of the building. It follows from
the discussion in Section 2, however, that the battery capacity cannot be reduced
without limit as some storage is likely to be needed to allow satisfactory control of the
microgrid. Notwithstanding, no attempt has been made in this study to estimate this
minimum storage requirement.
A microgrid powered by intermittent sources also has to make provision for below-
average generation. For photovoltaics and micro-CHP, an estimate made in Section 5.5
shows that about 15% of additional electrical energy—which may come from the
supplementary generation by micro-CHP or from the utility supply—needs to be provided
for this purpose.
These factors have impact on the economics of the microgrids which is considered in
Section 6.
6. Economic analysis of microgrids. Regulatory issues
A few definitions
† A ‘Deep Connection’ charge is when the connection charge includes all the associated
costs of connection including any costs incurred at remote locations or at higher voltage
levels.
† A ‘Shallow Connection’ charge is when the connection charge includes the cost of
connecting to the nearest appropriate point in the network (i.e. does not include remote
costs or cost incurred at higher voltages).
† A ‘Shallowish Connection’ charge will include some element as yet undecided of the
cost of reinforcement incurred. These definitions may be more explicitly defined, or
altered, following more detailed financial analysis of the potential impact.
† Black Start: restart time from fault or emergency shutdown.
† CI: Customer Interruptions
† CML: Customer Minutes Lost
† DG: Distributed generation or ‘embedded’ generation:
† DGCG: UK government Distributed Generation Coordination Group (www:distrib-
uted-generation:gov:uk)
† DNO: Distribution Network Operator
† DUoS: Distribution Use of System Charges
† ESQCR: Electricity Safety, Quality and Continuity Regulations 2002.
† Ofgem: The Office of Gas and Electricity Markets
† DTI: Department of Trade and Industry
6.1. Introduction
This section reviews the still limited literature on the economics of microgrids.
Microgrids have a significant potential to reduce GHG emissions from household energy
use and could therefore make a major contribution to reducing GHG emissions. Much of
the debate in the UK has concentrated on micro-CHP installations, especially since the
formation of the DGCG. This section considers the implications of the microgrids project
for the potential for widespread adoption of microgrid technologies in the UK. Barriers
that have to be overcome if microgrids are to make a major contribution to the UK energy
system are identified. It will briefly review the discussion about micro-CHP and the
economics of distributed generation in the UK. It will then consider the economics of
microgrids in the light of the findings of the microgrids project and raise some economic
and institutional questions that will have to be overcome if microgrids are to be deployed
on a large scale in the UK.
6.2. Micro-generation
6.2.1. General overview

The main technology under current consideration is small gas fired Stirling engines,
sized to provide the required heat output, but which also provide electricity output in a
domestic scale CHP installation. The heat requirements for a typical household imply that
there will be a surplus of electricity, which is commonly assumed to be sold into the national
electricity grid. The DGCG has identified considerable potential for micro-generation [26]:
more than 1 million household hot water boilers are replaced each year in the UK, which
could be replaced by CHP. They suggest a realistic potential of 400,000 micro-CHP units by
2010 and up to 10 million by 2020. There is also substantial potential in SMEs. Crozier-Cole
and Jones [27] report 17 million gas-fired central heating boilers in Great Britain, with 1.3
million boilers sold/year, also indicating a large potential market for micro-CHP.
The prospects for photovoltaics are more difficult to assess. Based on the market data
over the last 10 years which show an approximate 30% increase/year and a 0.8 progress
ratio (20% reduction in cost per capacity doubling) [28,29] in 6 years there would be a
500% increase in installations, which could be expected to lead to a 40–50% reduction in
costs of the PV installation. Given favourable tax treatment and regulatory regime, this
rate of cost reduction could dramatically improve the economic position of PV cells in just
a few years time.
The DGCG has produced a series of studies of various aspects of distributed or embedded
generation. DGCG considers charging principles [30]. They identify five possible options
for future charging regimes for connection and use of distribution systems.
Option 1. Status Quo (The Reference Case)

† Generators continue to pay ‘deep’ connection charges (including micro and
small generators, with exemptions for ‘larger generators’)
† Demand continues to pay ‘shallowish’ connection charges
† All other reinforcement costs are met through DUoS, which is paid solely by
demand (customers)
Option 2. Shallow generator connection, all reinforcement costs paid by load customers.
Option 3. Shallow generator connection cost, reinforcement costs being shared by all
parties.
Option 4. Shallowish generator connection cost, reinforcement costs being shared by all
parties.
Option 5. Shallowish connection charge for smaller generators, Site specific charges for
larger generators.
Options 2–5 would require fundamental changes to the regulatory pricing structure for
connection of embedded generators and a clear commitment from Ofgem as to how they
will treat the funding in the current and future ‘price control’ periods. Currently, embedded
generation has to pay ‘deep’ connection charges, which disadvantages embedded
generation in comparison with large scale generators, who only pay a ‘shallow’ connection
charge. Ofgem has now decided to proceed with ‘shallowish charging’ which includes
payment for connection and reinforcement at one voltage level above that of connection.
DGCG looks at options for micro-scale generation [31]. It finds that domestic and other
micro-scale generators would like a simplified charging mechanism and simple technical
requirements for connection that allows them to connect to the networks easily and quickly,
with transparent charging and payment systems. It suggests that the freedom of choice to
become a customer-generator may itself be of great value to the customer. Key issues are:
† meter arrangements for measuring the domestic and other micro-scale generation output,
export and import units and maximum demand;
† technical requirements for connection to the distribution network to enable ‘parallel’
operation;
† the connection charge;
† tariff mechanisms (via metering, profiles and fixed charges) for use of the transmission
and the distribution system, selling domestic and other microscale generation exports
and buying imported electricity.
Current standard consumption profiles do not cater for non-half-hourly metered

generation and so either additional profiles or a cheap and appropriate form of half-hourly
metering is required. A further problem is that the majority of domestic and other
microscale generation would currently be ineligible for ROCs. It is now likely that micro-
generators using PV and other renewables will be eligible.
There are also problems with the charging structure for distribution. The DNOs major
revenue driver is from the DUoS charge on each unit distributed for suppliers. Large
generators do not currently pay DUoS on their exports, whereas micro-generators would
have to pay this. One possibility would be for a DNO to set an annual capacity charge
(often called an entry charge in discussions) ACC to provide locational signals for demand
and generation within a distribution network area. Combined with ACC charge could be a
Generator DUoS charge (not currently paid on exports) to be levied by DNOs to support
the cost of providing the distribution network to customers with other micro-scale
generation to enable units to be exported (the generator DUoS is also part of Ofgem’s
latest proposals for the next Distribution Price Control). These charges could be levied on
suppliers who would pass them on to customers (in some form of their choosing). Ten
possible options for metering are identified; as combinations of three different meter
arrangements: single direction meter, bi-direction meter, import and export metering with
existing load profiles or separate import and export profiles.
Choudhury and Andrews also consider payment mechanisms for micro-generation
[32]. They conclude that half-hourly metering can be excluded as a route to a final
payment mechanism. It is currently far more expensive than other viable options; it does
not seem to deliver benefits to the system or to the micro-generator that can justify the cost
and there are no industry participants calling for half-hourly metering to be mandatory for
micro-generation. They consider that the main realistic options for discussion are: to allow
micro-generators to maintain their existing single direction meters with a generation
profile or to insist on the installation of import–export multi-register meters.
DGCG looks at the business environment and market based opportunities for
resolving some of the issues, which could co-exist alongside the existing RPI-X
regulatory mechanism [33]. It argues that there is uncertainty of the regulatory process. It
would be beneficial to provide DNOs with a clear long term approach to: measurement
of the relative performance of DNOs; treatment of capital and operational expenditure
and the relationship between them; performance based regulation. For treatment of DNO
business costs, the regulatory regime currently (in 2000) agrees a cash settlement
covering both capital and operating expenditure. There are merits of moving from an
asset based regulatory framework to one which takes output performance into account.
Opportunities for DNOs to provide extra services and hence obtain extra revenues might
include:
(a) extra security

(b) shortening post fault restoration times
(c) more stable voltage (reduces fluctuations)
(d) higher fault level (dip reduction).
Another important set of possibilities is the creation of a local ancillary services market.
Ancillary services can be defined as:
1. Security Services
† Provision of network capacity
† Provision of customer service improvements in terms of CI, CML, Worst served
customer and Transient interruptions
† Provision of Power Quality improvement in terms of voltage flicker, harmonics and
protection operating times
† Provision of voltage support
2. Ancillary Services
† Provision of Reactive Power as a tradable commodity (i.e. not necessarily for
voltage support locally)
† Provision of Frequency Response
† Provision of Reserve
† Provision of Black Start
Crozier-Cole and Jones [27] consider the potential market for micro-CHP in the UK.
They identify the following critical issues that will have to be addressed if micro-CHP is to
be adopted on a large scale in the UK:
† Grid connection procedures and standards appropriate for a mass market.

† Product testing and field trials covering representative UK housing stock and
occupancy types.
† Preparation of the delivery chain-skills for installation and service; actions to prepare
UK workforce, e.g. an Installer Accreditation scheme
† Introduction of simplified metering, settlement and trading procedures to obtain
maximum value from micro-CHP generation.
† Review of 28-day utility rule as a potential barrier to utility led Energy Service
Companies (this rule effectively prevents energy providers signing long term contracts
with retail customers).
6.2.2. Summary of regulatory issues and current regulation activities

It is apparent from the above survey of DGCG reports that, quite apart from the costs of
installation of micro-generation, the electricity market in the UK is organised such as to
present considerable barriers to the large scale adoption of micro-generation. Some key
issues are summarised in DCGC [26]:
† Deep connection charges—Micro-generation currently has to pay deep connection

charges, which large scale generators do not have to pay.
† High cost of participation in electricity markets—the current arrangements under
NETA require considerable expense and expertise to make contracts to supply
electricity to the grid. Currently, it is necessary to bilaterally agree a new contract if
generation is to be installed. This penalises micro-CHP in particular, if an individual
household is installing a micro-CHP unit. What is required is a standardised generation
connection and Use of Supply agreement, that comes into force automatically upon
connection.
† The network is built for uni-directional flows, with little active management—there are
no simple arrangements for bi-direction metering or demand and supply.
† Lack of incentive for DNOs to connect small generators.
There are also engineering requirements, which have been addressed by the DGCG:
† G83 Connection Standard is now defined.

† A suitable interpretation of the Wiring Regulations has been developed.
† ESQCRs have been defined.
† Metering-BSC modification defined.
Ofgem is currently considering changes to the regulatory regime to address some of the
charging issues. Ofgem is committed to moving from a deep to shallower charging regime,
with shallow charging at EHV and shallowish at lower voltages. They are also considering
price control mechanisms to provide more suitable incentives for DNOs [34].
6.3. Microgrids: the context of current government thinking in the UK
6.3.1. Registered power zones

The UK Office of Gas and Electricity Markets (Ofgem) in their documents [35] set out a
proposal for ‘registered power zones’. This proposal follows extensive consultation within
the Industry and forms part of the innovation initiatives flowing from the ongoing work to
encourage both distributed generation and the use of ‘renewable’ sources.
Whilst a registered power zone (RPZ) is as yet not defined in any detail, it is clear that
Ofgem are considering something that must be innovative. The intention is for a DNO to
register a defined zone in which innovative plant and systems may be used either to
generate or control power flow. The financial incentives to set up and run such a zone are
very attractive. Thus Ofgem hopes to persuade the DNOs to set up and operate such zones.
The outstanding issues, on which Ofgem are inviting views concern the application of
industry standards particularly those pertaining to quality standard at consumers terminals.
Derogation of some standards is likely to encourage innovation and so Ofgem is actively
considering a move in this direction.
A report [36], in July 2003 set out a number of both short tem and longer term solutions
to some of the connection problems. It is likely that any move to use these methods to
establish a registered power zone would succeed in attracting the attractive connection
terms.
The solutions considered by the report concentrate on preserving the network and many
ideas are explored. Only in one area is ‘network splitting’ considered and the conclusions
are not very positive. This was because it was only considered as a method of reducing
fault level in an existing network. Power quality and short-term interruptions were seen as
the disadvantages of a smaller network.
It is clear that the Working Group did not see any real benefits for a separated or loosely
coupled microgrid. Areas of investigation were identified where work is required to
identify the balance between reduction in fault levels and impact on customers.
6.3.2. The view of the electricity industry

The concept of a microgrid is not under active consideration by DNO’s for two major
reasons:
† It is not in the interest of a DNO.

† The perceived benefits are not easily defined.
A DNO perceives its best interest is served by preserving its own network and hence the
RPZ concept is acceptable. The RPZ allows financial incentives and may allow relaxation
of ‘power quality’ standards, both seen as an attractive way forward. The notion of RPZs
has not—quite deliberately—been defined very precisely. It is possible, however, to
interpret the rules in a way that will allow the concept of a microgrid.
A microgrid may be of any size, as may an RPZ; hence a small RPZ could be a large
microgrid. The degree of ‘connection’ to the utility network is also undefined and can
therefore encompass the loose coupling envisaged of a microgrid. A DNO owned and
operated microgrid is, therefore, entirely possible whilst not being seen as threatening
the network.
Whilst not seen as a primary initiative, a microgrid could operate within an RPZ if it
can be seen as possessing advantages. It is thus the advantages, technical, political and
commercial, which must be carefully set out as the argument for the setting up of a
microgrid. The only other alternative is the promotion of a completely new, private
and independent microgrid, which whilst possessing some advantages is unlikely to attract
DTI incentive funding. It is likely therefore that the thinking on microgrids should take
two directions:
† As a very specific RPZ.

† A quite small network (up to several megawatts).
6.4. Economic analysis of microgrids
There are several potential economic benefits of microgrids:
† Reduced transmission and distribution costs and energy losses.

† Potentially total higher energy efficiency.
† The small scale of individual investments reduces capital exposure and risk, by closely
matching capacity increases to growth in demand.
† The low capital cost potentially enables low-cost entry into a competitive market.
The micro-generators considered in this paper also provide savings through the
higher efficiency of energy provision. When integrated into a microgrid, the generated
electricity can be shared among the consumers, obviating the necessity to export power
to the public network at a lower price. There are other potential benefits for the
electricity system in security and ancillary services from distributed generation,
outlined from DGCG [33] above.
As with most other renewable energy sources, the cost of electricity generated by
photovoltaic generators is dominated by the initial capital outlay. The cost to consumer is
therefore dependent heavily on the financing terms, and the requirements of short term
repayment schedules put these energy sources at a disadvantage. In particular, the
financing of photovoltaics considered here reflects the increasingly common perception of
photovoltaic arrays as a building component and can therefore be paid for over the same
term, by a 25 year mortgage. An indication that longer term investment is possible comes
from a number similar projects in the UK (for example, the Woking Energy System and
the Southampton Geothermal Project) as well as the experience with building integrated
photovoltaics in Germany and Japan.
It is instructive to compare the different integration strategies for micro-generation in
terms of three ‘Scenarios’. Scenario 1 considers micro-CHP units installed individually,
without integration into microgrids; there is no photovoltaics and no energy storage.
Scenario 2 describes microgrids consisting of micro-CHP and battery only, designed to
supply all heat and electricity in winter. In summer, the household buys the electricity
shortfall from the main utility. Finally, scenario 3 describes the full microgrid consisting
of micro-CHP, PV and storage, and capable of stand-alone operation, as analysed in
Section 5.
Electricity generation, purchase and export in the three scenarios is summarised in
Table 13. The amount of energy exported in Scenario 1 has been estimated from the
comparison of the micro-CHP generation and demand in Section 5. The total household
consumption in all three cases is assumed to be equal to the UK average of 3300 kWh per
Table 13
Electricity generation, purchase and export, summarised in the form of three Scenarios (in kWh per household per
year)
Scenario 1 2 3
Micro-CHP generation With heat recovery 2700 2064 2064
Without heat 0 0 505
recovery
PV generation 0 0 1236
Electricity export 900 0 0
Electricity purchase 1500 1236 0
annum, although the electricity generated by the household may exceed this value. In
scenario 1, this is due to the fact that there is no energy storage to accommodate the
mismatch between generation and load. In scenario 3, on the other hand, electricity
additional to the expected values has to be generated to provide the required security of
supply. Both these surpluses can be removed by installing extra storage or extra
generation, at an additional cost.
Scenario 1 describes the situation when DCHP units enter the marketplace, which is
considered here to occur in the near term. The results are broadly consistent with the
findings of the DGCG and EA Technology reports [37,38]. When considering the cost of
the micro-CHP it is usual to include only the additional cost to installing an efficient
central heating boiler. If a net metering arrangement can be negotiated with the electricity
supply utility, the saving in electricity can reach 50% of the electricity bill [37],
representing a simple payback period of some 4 years.
If the electricity supply company is less forthcoming and offers a tariff in keeping with
the bulk electricity price, the reduction in household electricity costs become less
advantageous to the householder. This extends the payback period to some 5–6 years. The
latter situation is likely to prevail once DCHP generation becomes established in the
market and rigorous economics is applied to the electricity trading arrangements.
At this point, there is a likely to be little incentive on the part of the householders to take
the next step and integrate micro-CHP units into microgrids. This is because the electricity
cost to the householder under the microgrid scenario 2 is likely to be similar as under
scenario 1 with no net metering where no action needs to be taken by the householder aside
from the purchase and installation of the micro-CHP units.
The economic balance is likely to change as the penetration of micro-CHP (or PV
generators) approaches the limit which can be tolerated by the distribution system. As
the costs incurred by the electricity distribution companies to connect intermittent
embedded generators rise, the economic environment is likely to tilt towards the
microgrid scenarios 2 and 3. Scenarios 3 represents the more environmental alternative
but are only likely to compete as a serious alternative if PV prices fall, or a government
subsidy is available. We estimate that a reduction in PV cost of some 50% is needed to
make the electricity cost in the full microgrid (scenario 3) competitive with the current
electricity cost.
The above analysis applies to countries (such as UK) which have a comprehensive
electricity supply network. In the developing countries with a poor electricity supply
system, or in rural locations in countries with a low population density (such as Australia)
the microgrid scenarios may be the preferred solution at a much earlier stage. Many
examples already exist of single stand-alone PV, wind or hybrid systems, or small
networks of these generators which are already now cheaper than extending the electricity
supply network.
6.5. Where do we go from here?
The analysis undertaken in this paper has shown that the electricity demand in a
microgrid can be supplied by micro-CHP generators with penetration ratio of one for
every second household, together with a photovoltaic array of about 1.5 kWp and energy
storage corresponding to about 2.7 kWh per household (four industrial lead-acid
batteries). This would then deliver an energy service independent of the grid network. As
has been indicated, the economics of central heating boiler replacement with micro-CHP
units is already quite favourable. If an installation and maintenance infrastructure can be
developed, there should be a mass market for micro-CHP, which will both bring down
costs and raise the public profile of these technologies. This would require firms with
expertise in installation and maintenance, together with large scale manufacture and sales
activities. These considerations suggest that a new type of firm—an energy service
company with expertise in energy saving measures, supply and demand contracting,
renewables and microgrids may be necessary to exploit and develop the new market
structures.
A supportive regulatory structure is also essential. The ‘public profile’ is very
important. Historically, new technologies have always faced economic and institutional
barriers and only some technologies ‘take-off’. If there is a strong reason for society to
pursue microgrids, these issues have to be addressed and institutions/economics changed
through policy to make it happen. This requires not only a supportive public environment,
but active policy and commercial support. Given that climate change is now a major
element of energy policy, this can in principle happen, but a rapidly expanding market
would also speed up the necessary process of regulatory change. New technology that
exploits information technology for, e.g. more complex metering such as net metering and
automated real time control of domestic energy systems will have to be made available at a
commercial scale. This is discussed in Choudhury and Andrews [32]. These technologies
are available, but have not been applied on a large scale.
It is also important to note that the literature surveyed above concentrates on the current
energy system. Microgrids, however, open up possibilities for dramatic change in the
structure of the energy system. Active demand management in response to electricity price
signals could not only significantly reduce the peak demands on the network but could also
lead to large scale energy savings, introducing a culture favourable to energy efficiency in
households.
The present analysis, combining PV and micro-CHP and a very small battery
requirement, gives a microgrid that is independent of the national electricity network.
This has particular benefits for remote communities. In particular, there would be no
further requirement to provide transmission cables. This would remove distribution costs
outside the microgrid. It would also have the further benefit in areas of natural beauty of
enabling the removal of unsightly power cables. These factors would increase the social
benefit of installing microgrids. This opens up dramatic possibilities. The medium term
requirement to replace current generation nuclear and coal fired power stations could be
met by this technology. This would require the combination to match the consumer’s
expectations for supply security with the same low level of interruptions as the grid.
There could also be a greatly reduced demand on the transmission and distribution
network, which would also lead to large savings in operation and investment in the long
run. Microgrids could also be combined with other energy saving measures and
appliances, further reducing domestic and office energy demand and GHG emissions.
Overall, the conclusion is that microgrids do have real potential to make a major
contribution to reducing GHG emissions from buildings, which in the UK are now 50%
of final energy demand. This will only happen if there are major changes to the electricity
market and regulatory structure.
7. Conclusions
The paper has examined the energy production by small scale generators in close
proximity to the energy users, integrated into microgrids. A review of the current status,
including an overview of distributed generation has been given, highlighting the views of
the electricity and gas regulator (Ofgem) and the electricity industry, and giving a
description of relevant regulatory concepts which may evolve into microgrids.
The issues which underpin the concept of microgrid have been examined, including the
integral control of the microgrid and the power balance, focusing on the frequency and
voltage control as well as the power quality. No less important is the relationship between
the microgrid and the local utility, when present, and the control of the interface and of the
power flow.
Components of the microgrid have been discussed with emphasis on the generators
(fossil fuel generators as well as photovoltaic arrays, fuel cells and inverters), typical
domestic load profiles and the importance of diversity, and a discussion of the progress
made in resolving the technical and regulatory issues associated with the utility integration
issues of small generators.
The quantitative contribution of the paper rests on the analysis of the energy balance in
the microgrid. Whilst the power balance is essential for the control of the microgrid,
energy storage at the diurnal time scale will be needed to compensate for the mismatch
between generation and demand, and to make the most efficient use of the renewable
generation and the electricity produced by micro-CHP More fundamentally, an
appropriate diversity of generation methods needs to be employed to satisfy the energy
demand and permit stand-alone operation, if required, at any time of the year.
Numerical estimates have been made focusing on the microgrid of domestic users
powered by small fossil-fuel generators and photovoltaic arrays. A model of the energy
consumption in buildings was used to produce typical profiles for space heating, hot water
and the electricity consumption. Based on the energy balance between the generation and
load, methodology has been developed to determine the optimum combination of the
generators and energy storage in the microgrid. To our knowledge, this is the first study
which highlights the importance of energy balance in the design of microgrids. The
principal results include the determination of the optimum micro-generation capacity,
consisting of 1.37 kWp PV array per household and 45% household ownership of micro-
CHP generators. This is the minimum size which can maintain energy balance on a yearly
basis if supplemented by energy storage of 2.7 kWh per household. The required micro-
CHP ownership might be reduced to some 15% when fuel cells become available as a
domestic CHP generator.
The economics of the microgrids has been discussed against the background of the
current regulatory and economic framework for distributed generation. The potential
future development of microgrids is discussed by means of three scenarios which, in broad
terms, describe the likely stages of deployment to their full potential. We find that there is
no fundamental technological reason why microgrids cannot contribute an appreciable
part of the UK energy demand. Indeed, an estimate of cost indicates that the microgrid
considered in this project would supply electricity at a cost comparable with the present
unit cost if the current support mechanisms for photovoltaics were maintained.
Although the present paper concentrates on the comparison with current energy
system, microgrids open up possibilities for dramatic change in the structure of the
energy system. Combining PV and micro-CHP and a very small battery requirement,
gives a microgrid that is independent of the national electricity network. In the short
term, this has particular benefits for remote communities or areas of natural beauty.
More dramatic possibilities open up in the medium to long term. The requirement to
replace current generation nuclear and coal fired power stations could be met by this
technology, greatly reducing also the demand on the transmission and distribution
network. Microgrids provide a direct facility to implement other energy saving
measures and appliances, further reducing domestic and office energy demand and
GHG emissions.
Overall, the principal conclusion is that microgrids do have real potential to make a
major contribution to reducing GHG emissions from buildings. This will only happen if
there are major changes to the electricity market and regulatory structure. The policy
question is therefore, how to initiate these changes?
Acknowledgements
Work reported in this paper was performed on project IT1.33 and sponsored by the
Tyndall Centre for Climate Change Research UK (see www.tyndall.ac.uk). We are
grateful to Harry Barnes, John Page, Geoff Levermore, Andy Wright, Sukumar Natarajan,
Alan Ingham, Alan Collinson and Tony Lakin for many stimulating discussions on the
subject of this paper.
References
[1] Digest of UK energy statistics, HMSO; 2003.

[2] The dawn of micropower, The economist; August 5th 2000, p. 99.
[3] Powering progress: electricity technology roadmap, EPRI; 1999.

[4] See, for example, Implementing agreement on demand side management technologies and programmes.
Annual Report of the International Energy Agency; January 2004.
[5] Lasseter RH. Microgrids. Proc IEEE Power Eng Soc Winter Meeting 2002;1:305–8.
[6] Venkataramanan G, Illindala M. Microgrids and sensitive loads. Proc IEEE Power Eng Soc Winter Meeting
2002;1:316–22.
[7] Markvart T, Castañer L. Practical handbook of photovoltaics: fundamentals and applications. Oxford:
Elsevier; 2003.
[8] Luque A, Hegedus S. Handbook of photovoltaic science and technology. Chichester: Wiley; 2003.
[9] d’Accadia MD, Sasso M, Sibilio S, Vanoli L. Micro-combined heat and power in residential and light
commercial applications. Appl Therm Eng 2003;23(10):1247–59.
[10] McDonald CF, Rogers C. The ubiquitous personal turbine—a power vision for the 21st century. J Eng Gas
Turb Power, B 2002;835–44.
[11] Hamilton S. The handbook of microturbine generators.: Penwell; 2003.
[12] Moore MJ. Microturbine generators.: American Society of Mechanical Engineering; 2002.
[13] Nicholas DK, Loving KP. Assessment of microturbine generators. IEEE Power Eng Soc Gen Meeting 2003;
4:2314–5.
[14] Davis MD. Mini gas turbines and high speed generators, a preferred choice for servicing large commercial
customers and microgrids. Part I—generating systems. IEEE Power Eng Soc Summer Meeting 2002;2:669–76.
[15] Davis M. Mini gas turbines and high speed generators, a preferred choice for servicing large commercial
customers and microgrids. Part II—Microgrids. IEEE Power Eng Soc Summer Meeting 2002;2:682–7.
[16] Hamilton S. Microturbine generators, distributed generation project report, California Energy Commission;
Sept 1999.
[17] Yinger R. Behaviour of capstone and honeywell microturbine generators during load changes, California
Energy Commission 2001.
[18] Gigliucci G, Petruzzi L, Cerelli E, Garzisi A, la Mendola A. Demonstration of a residential CHP system
based on PEM fuel cells. J Power Sources 2004;131:62–8.
[19] Trends in photovoltaic applications in selected IEA countries between 1992 and 2000. IEA report PVPS T1-
10:2001, International Energy Agency; 2001.
[20] Mansouri I, Newborough M, Probert D. Energy consumption in UK households: impact of domestic
electrical appliance. Appl Energy 1996;54(3):211–85.
[21] Yao D, Baker N, McEvoy M. A simplified thermal resistance network model for building thermal
simulation, Sim2002. Proceedings of eSim, Montreal, Canada; September 11–13th, 2002, p. 135–40.
[22] See, for example, the recommended sizing practice for photovoltaic systems at www.bpsolar.com.
[23] Markvart T. Sizing of hybrid photovoltaic/wind energy systems. Solar Energy 1996;57:277–81.
[24] European solar radiation atlas. Cologne: Verlag TUV Rheinland; 1984.
[25] Photovoltaic system for test and development. ETSU Report S/P2/00242/REP; 2001.
[26] Distributed Generation Coordination Group website www.distributed-generation.org.uk.
[27] Crozier-Cole T, Jones G. The potential market for micro CHP in the UK. Energy for sustainable
development Ltd for energy savings trust report P00548; 2002.
[28] Anderson D. The economics of photovoltaics. In: Archer M, Hill R, editors. Clean electricity from
photovoltaics. Singapore: World Scientific/Imperial College Press; 2001.
[29] Maycock P. Photovoltaic markets, chapter V-3. In: Markvart T, Castañer L, editors. Practical handbook of
photovoltaics: fundamentals and applications. Amsterdam: Elsevier; 2003.
[30] Charging Principles DTI/OFGEM Embedded Generation Working Group WP4 Rapporteur Contribution
(www.distributed-generation.org.uk).
[31] Options For Domestic and Other Micro-Scale Generation DTI/OFGEM Embedded Generation Working
Group WP5 Rapporteur Contribution (www.distributed-generation.org.uk).
[32] Choudhury W, Andrews S. Payment mechanisms for micro-generation. Report number: K/EL/0282/REP,
URN 02/1146, Ilex Energy Consulting for DGCG; 2002.
[33] Future Network Design. Management and Business Environment Generation DTI/OFGEM Embedded
Generation Working Group WP7 Rapporteur Contribution (www.distributed-generation.org.uk).
[34] Minutes of 10th DGCG COORDINATING GROUP; 24th Jan 2004.
[35] Electricity distribution price control review policy document, UK Office of Gas and Electricity Markets,
March 2003; Regulatory Impact Assessment for distributed generation and structure of distribution charges,
UK Office of Gas and Electricity Markets, March 2004.
[36] Collinson A, Dai F, Bedoes A, Crabtree J. Solutions for the connection and operation of distributed
generation. EA technology report URN 03/1195 for DTI; 2003.
[37] Second annual report of the distributed generation co-ordinating group (2003). DTI March, 2004
(www.distributed-generation.gov.uk).
[38] Harrison J, Redford S. Domestic CHP: what are the benefits?. EA Technology report to the UK Energy
Saving Trust; 2001.

Microgrid and Energy Management

Uploaded by

Copyright:

Available Formats

Microgrid and Energy Management

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Microgrid and Energy Management

Uploaded by

Copyright:

Available Formats

Renewable and Sustainable Energy Reviews

Can microgrids make a major contribution to UK

1. Introduction and background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 80

3. Generators and loads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

4. Modelling energy consumption in buildings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

5. An example of sizing: microgrids powered by photovoltaics and micro-CHP . . . . . . . . . 105

6. Economic analysis of microgrids. Regulatory issues . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

1. Introduction and background

1.1. Some history

consumers at various voltages, from 33 kV to 240 V Fig. 1. This situation prevailed in

1.2. Distributed generation

1.3. Why microgrids?

A microgrid is a small-scale power supply network that is designed to provide power

1.4. Issues to consider

1.5. Structure of the paper

2.1. The microgrid concept

2.3. Internal control of a microgrid

2.3.1. Power balance

2.3.4. Power quality

As has been discussed by Venkataramanan and Illindala, the distributed generation

2.4. Energy balance

Little significance is usually attached to the concept of energy balance in a conventional

2.5. Energy storage

Device Discharge time Power Comments

3. Generators and loads

3.2. Photovoltaic arrays

The photovoltaic array consists of one or several modules mounted on support

Mean daily radiation (kWh/m2)

the installation of photovoltaic generation. The perception of solar cells as a building

3.3. Fossil fuel power generators

3.3.1. Combined heat and power generators

3.3.2. Reciprocating internal combustion engines

Prime Fuel Ratio of Advantages Disadvantages Commercial

3.3.4. Stirling engines

3.3.5. Fuel cells

3.4. Electrical power generation

Fig. 2. Schematic diagram of an inverter.

3.4.2. Utility integration issues

3.5. The domestic load

The profile of electricity consumption and heating requirements of households are

Fig. 6. The daily household load as a function of time.

4. Modelling energy consumption in buildings

1. Domestic appliance electricity consumption load profile.

The electricity profile is used to validate the methodology as a comparison is available

4.2. Domestic electricity load profile

Scenarios Type Unoccupied period

Hour Winter Summer

4.3. Heating load profile

1 person 2 person 3 person 4 person Five person

3-Bed flat 3-Bed semi-detached 3-Bed detached 3-Bed mid-terrace

Climate UK Kew 67 Solar: beam, horizontal diffuse Air temperature

4.4. Domestic hot water

Fig. 8. The heating load profile in winter.

Fig. 9. Daily DHW demand of a 3-person household.

5. An example of sizing: microgrids powered by photovoltaics and micro-CHP

It is unrealistic to expect an exact hourly balance between energy produced by these

5.2. The methodology

5.3. Daily energy balance in a microgrid