IecWP Gridintegrationlargecapacity LR en

stability power integration battery energy-generation flywheel decarbonization energy
solar variability unpredictability dependency photovoltaic UHVDC control pumped

bility assessment planning storage technologies generation forecast operational enhan
and response modeling electrical energy storage alternating current contingency analys
ent induction generator transmission network energy efficiency heat recovery electrotec
ysis strategy electricity market renewables energy access grid operator grid infrastructure
pact coordination
energy market operators transmission planning body challenges st
ogen policy-makers regulators industry research community energy access grid arch
-fuel-cost energy cost interconnection power fluctuations black-out geopolitics distributio
ADA generation portfolio grid operation hydro variable generation steam solar concentrat
intermittency location dependent grid owner wind speed supply and demand voltage
cast weather dispatchable load nuclear transmission capacity sunlight load management
rol frequency regulation spinning reserve black-start capacity Smart Grid remote lo
rtainty deterministic complexity micro-grid power output power conversion steam
mal energy solar tower parabolic through linear Fresnel reflector system parabolic dish
casting spatial aggregation system security reliability monitoring co-generation
uations power exchange real-time fault ride-through capability frequency drop grid-friendl
ronics ramping curtailment pitch regulation inertial response dynamic modeling
-shifting AMI electric vehicles cyber-security grid stability power integration
gy-generation flywheel decarbonization energy security wind solar variability unpredi
endency photovoltaic UHVDC control pumped storage - flexibility assessment planning
nologies generation forecast operational enhancement demand response modeling electri
ge alternating current contingency analysis direct current induction generator transmissio
nergy efficiency heat recovery electrotechnology analysis strategy electricity market ren
gy access grid operator grid infrastructure synthesis impact coordination energy market o
mission planning body challenges standards hydrogen - policy-makers regulators industry
munity energy access grid architecture zero-fuel-cost energy cost interconnection
uations black-out geopolitics distribution utility SCADA generation portfolio grid operatio
iable generation steam solar concentrating power PV intermittency location dependent g
nd speed supply and demand voltage support forecast weather dispatchable load
mission capacity sunlight load management voltage control frequency regulation spinning
k-start capacity Smart Grid remote locations uncertainty deterministic complexity mi
er output power conversion steam turbine thermal energy solar tower parabolic throug
nel reflector system parabolic dish power forecasting spatial aggregation
security r
White system
paper
itoring co-generation balancing fluctuations power exchange real-time fault ride-through c
uency drop grid-friendly power electronics ramping curtailment pitch regulation inertial r
mic modeling UHVAC load-shifting AMI electric vehicles cyber-security
Grid integration of large-capacity

Renewable Energy sources and use of
large-capacity Electrical Energy Storage
October 2012
y
Executive summ ar
The present White Paper is the third in a series

whose purpose is to ensure that the IEC can
continue to contribute with its standards and
conformity assessment services to the solution
of global problems in electrotechnology. The
White Papers are developed by the IEC MSB
(Market Strategy Board), responsible for
analyzing and understanding the IECs market
so as to prepare the IEC strategically to face
the future.
The proportion of renewable energies (RE) is
called upon to increase in all major electricity
markets. The reasons for this are not examined
closely here, since they have been fully treated
elsewhere. This paper explores what is needed
to integrate large quantities of renewables
into existing electricity grids, given various
characteristics and difficulties which necessarily
accompany such a change. Section 2 examines
these characteristics, describes the difficulties
and analyzes the consequent challenges for
grid operators as well as for producers of
electricity, both renewable and conventional.
Section 3 shows todays methods and
responses to the challenges. These are
extensive and applied widely and professionally;
the section nevertheless concludes that they
will not suffice as the proportion of renewables
grows to 15%, 25% or even 35% of the energy
in some grids.
Thus section 4, one of the two core chapters of
the paper, covers all the research, investment
and other tools without which large-scale
renewables cannot be successfully integrated.
These range from what renewable generation
needs to provide in order to be accepted,
through all the control and infrastructure the grid
itself needs in order to cope, to the realization
that conventional generation facilities also need

to contribute significantly to make the whole
exercise a success.
The second core part of the paper is section 5
on electrical energy storage (EES), and it
extensively uses the results of the IEC White
Paper on this subject published in 2011. It turns
out that the various challenges and difficulties
covered in section 2, and even more the
avenues for the future sketched out in section 4,
either depend on the use of storage or at least
can benefit from it. Section 5 therefore outlines
its use and usefulness for the integration of
renewables and concludes in harmony with
the White Paper on storage that significant
developments are required in this area as well.
The following section, section 6, briefly surveys
the contribution that standards already make
and can make in the future to solving the issues
covered elsewhere.
Section 7 starts with a brief conclusion. Its thrust
is that the electricity community knows in broad
outline what will be needed to integrate largescale renewables, but that many elements are
not yet in place and much effort will be required.
There follow recommendations addressed to
the IECs partners, in both the public and private
sector, and to the IECs own structures. The
IEC MSB believes that future implementation
of these recommendations is the factor which
will constitute the greatest added value of the
present White Paper.
This White Paper was written by a project team under

the MSB, in particular the experts of the State Grid
Corporation of China (CN) and RASEI (the Renewable
and Sustainable Energy Institute) in the University of
Colorado at Boulder and NREL (US).
List of abbreviations
Section 1 Introduction
Section 2 RE generation: the present, the future and the integration

challenges 13
2.1 Drivers of RE development

2.1.1 Decarbonization

2.1.2 Energy security

2.1.3 Expanding energy access
14
14
15
16
18
19
2.2 Present status of RE generation and future projections

2.2.1 Wind energy
2.2.2 Solar energy
2.3 RE grid integration challenges
2.3.1 Non-controllable variability
2.3.2 Partial unpredictability
2.3.3 Locational dependency
22
25
28
29
30
Section 3 Present: state of the art in integrating large-capacity RE
31
3.1 General
32
3.2 RE generation technology

3.2.1 Wind power generation
3.2.2 PV power generation
3.2.3 Concentrated solar power generation
32
32
38
40
3.3 Transmission technology

3.3.1 AC transmission
3.3.2 VSC-HVDC transmission
43
43
45
3.4 Operational technologies and practices

3.4.1 Power forecasting
3.4.2 Operational practices
46
46
50
Section 4 Future: technical solutions for integrating

more large-capacity RE
57
4.1 General
58
59
59
4.2 Grid-friendly RE generation

4.2.1 Need for grid-friendly RE generation
C O N T E N TS
4.2.2 Advanced characteristics of RE generating units and plants

4.2.3 Centralized control of an RE plant cluster
4.2.4 Improvements in modelling RE generation
59
61
62
4.3 Improved flexibility in conventional generation

4.3.1 Need for more flexibility in conventional generation
4.3.2 Assessment of generation flexibility
4.3.3 Generation planning for both adequate capacity and adequate flexibility
63
63
63
65
4.4 Transmission expansion

4.4.1 Needs for transmission expansion
4.4.2 Application of new transmission technologies
4.4.3 Developments in transmission planning
67
67
67
72
4.5 Operational enhancement

4.5.1 Need for operational enhancement
4.5.2 More accurate RE power forecasts
4.5.3 Enhancement of operational tools and practices
74
74
75
76
4.6 Demand response

4.6.1 Demand response applications for RE integration
4.6.2 Demand response practices and trends
4.6.3 Technologies supporting more demand response
79
79
80
82
4.7 Summary
83
Section 5 Application of large-capacity EES to support RE integration
85
5.1 General
86
5.2 Promising large-capacity EES technologies
86
5.3 Roles of EES in RE integration

5.3.1 Grid-side roles of EES
5.3.2 Generation-side roles of EES
5.3.3 Demand-side roles of EES
88
88
92
95
5.4 Technology needs of large-capacity EES applications
96
5.5 Summary
98
Section 6 Standards for large-capacity RE integration
99
6.1 General
100
6.2 Present situation
100
6.3 Future needs
103
Section 7 Conclusions and recommendations
105
7.1 Conclusions
106
7.2 Recommendations addressed to policy-makers and regulators

Recommendation 7.2.1 Coordinating all actors
Recommendation 7.2.2 Single framework for connecting and controlling renewables
Recommendation 7.2.3 Regulations to enable integration
106
107
107
107
7.3 Recommendations addressed to utilities, industry and research

Recommendation 7.3.1 Enhanced transmission as a precondition for renewables
Recommendation 7.3.2 Stochastic forecasting
Recommendation 7.3.3 Research for forecasting and complex modelling
Recommendation 7.3.4 Research for cluster connection and control
107
107
107
108
108
Recommendation 7.3.5 Research into EES

Recommendation 7.3.6 Forecasting the demand side
7.4 Recommendations addressed to the IEC and its committees
Recommendation 7.4.1 Technical contribution to the RE integration framework
Recommendation 7.4.2 Rapid progress in RE integration standards
Recommendation 7.4.3 Synergy with industry associations on RE integration
108
108
108
108
108
108
References 109
on s
L is t o f a b b r e v i at i
Technical and scientific terms
AC
AGC
AMI
BMS
CA
CAAGR
CAES
CECRE
CSC-HVDC
CSP
CSR
DC
DFIG
DLC
DR
DSA
EEE
EES
ELCC
EMS
EUE
EV
EVPP
FACTS
FES
FiT
GEMAS
GHG
HVAC
HVDC
IGBT
IRRE
LA
LCC-HVDC
LFP
LFR
Li-ion
LOLE
LVRT
MTDC
NaS
NGCC
Alternating current
Automatic generation control
Advanced metering infrastructure
Battery management system
Contingency analysis
Compound average annual growth rate
Compressed air energy storage
(Spanish for) Renewable energy power control centre
Current source converter HVDC
Concentrated solar power
Controllable shunt reactor
Direct current
Doubly fed induction generator
Double layer capacitor
Demand response
Dynamic security analysis
Electrical energy efficiency
Electrical energy storage
Effective load carrying capacity
Energy management system
Expected unserved energy
Electric vehicle
Electric vehicle virtual power plant
Flexible AC transmission system
Flywheel energy storage
Feed-in tariff
(Spanish for) Maximum admissible wind power generation system
Greenhouse gas
High voltage alternating current
High voltage direct current
Insulated gate bipolar transistor
Insufficient ramping resource expectation
Lead acid
Line commutated converter HVDC
Lithium iron phosphate (LiFePO4 )
Linear Fresnel reflector
Lithium ion
Loss of load expectation
Low voltage ride through
Multi-terminal DC
Sodium sulphur
Natural gas combined cycle
NWP
Numerical weather prediction
PCS
Power conversion system
PHS
Pumped hydro storage
PIRP
Participating intermittent resource program
PV Photovoltaic
RE
Renewable energy/ies
RFB
Redox flow battery
RMSE
Root mean square error
SCADA
Supervisory control and data acquisition
SCED
Security constrained economic dispatch
SCGT
Simple cycle gas turbine
SCIG
Squirrel cage induction generator
SMES
Superconducting magnetic energy storage
Synthetic natural gas
SNG
Static synchronous compensator
STATCOM
Static var compensator
SVC
Thyristor controlled series compensator
TCSC
Transient stability analysis
TSA
Unit commitment
UC
Ultra-high voltage AC
UHVAC
Ultra-high voltage DC
UHVDC
V2G Vehicle-to-grid
Virtual power plant
VPP
Vanadium redox flow battery
VRFB
Voltage stability analysis
VSA
Voltage source converter HVDC
VSC-HVDC
Wind power plant
WPP
Wound rotor induction generator
WRIG
WSAT
Wind security assessment tool
Wind turbine
WT
Wind turbine generator
WTG

Organizations, institutions and companies
AESO
AQSIQ
BPA
BCTC
CAB
CAISO
CanWEA
CEPRI
CSPG
EPE
Alberta Electric System Operator

Administration of Quality Supervision, Inspection and Quarantine (of China)
Bonneville Power Authority
British Columbia Transmission Corporation
Conformity Assessment Board (of IEC)
California Independent System Operator
Canadian Wind Energy Association
China Electric Power Research Institute
China Southern Power Grid
Energy Research Corporation (of Brazil)
on s
L is t o f a b b r e v i at i
EWEA
FERC
GIVAR
IEA
IEC
IEEE
IOU
IPCC
ISO
ISO
IVGTF
JWD
MSB
NDRC
NEA
NERC
OECD
ONS
NYISO
PES
PMA
REE
RTO
SAC
SGCC
SMB
TC
TEPCO
TSC
TSO
UWIG
WAPA
WECC
European Wind Energy Association

Federal Energy Regulatory Commission (of US)
Grid Integration of Variable Renewables Project (of IEA)
International Energy Agency
International Electrotechnical Commission
Institute of Electrical and Electronics Engineers
Investor-owned utility
Intergovernmental Panel on Climate Change
International Organization for Standardization
Independent system operator
Integration of Variable Generation Task Force (of NERC)
Japan Wind Development Co.
Market Strategy Board (of IEC)
National Development and Reform Commission (of China)
National Energy Administration (of China)
North American Electric Reliability Corporation
Organisation for Economic Co-operation and Development
The Operator of the National Electricity System (of Brazil)
New York Independent System Operator
Power & Energy Society (of IEEE)
Power marketing administration
Red Elctrica de Espaa
Regional transmission organization
Standardization Administration of China
State Grid Corporation of China
Standardization Management Board (of IEC)
Technical Committee (of IEC)
Tokyo Electric Power Company
TSO Security Cooperation
Transmission system operator
Utility Wing Integration Group
Western Area Power Administration (of US)
Western Electricity Coordinating Council
SECTION 1
Introduction
10
S E C T I O N 1
Introduction
This report discusses the challenges of

synthesizing the development and operation of
RE and EES resources with the planning and
operation of the rest of the power grid, including
existing generation resources, customer
requirements and the transmission system itself.
The generation of electricity from RE sources
includes technologies such as hydropower,
wind power, solar power, tidal and wave power,
geothermal power, and power from renewable
biomass. Wind and solar power are the focus
of this report, for two reasons. First, they are
among the renewable generation types wind,
solar, and wave that are subject to natural
variability in their energy sources. This variability
creates distinct challenges for integration into the
larger power system, namely nondispatchability.
Secondly, wind and solar are relatively mature for
use in large capacities and in wide areas, and so
have a significant impact on the power grid that
is likely to increase over time.
Integration of RE is a poly-nodal problem
involving multiple decision-makers at a variety
of spatial and temporal scales and widely
varying degrees of coordination. These
decision-makers include operators of RE and
energy storage resources, grid operators,
energy market operators and transmission
planning bodies. As such, grid integration is
not performed by any one entity in the power
system, but instead involves the actions of a
variety of entities, some highly coordinated and
others discrete. The burgeoning development
of smart grids adds still more tools, options and
players to the mix. Many of these actors engage
with various technology standards, practices,
procedures and policies for the operation of
individual generators, RE clusters, substations,
and the broader electrical energy system.
This report, produced by the International

Electrotechnical Commission (IEC) Market
Strategy Board (MSB), is the third in a series of
MSB White Papers which already includes:
1) Coping with the Energy Challenge (September
2010), hereafter referred to as the MSB EEE
Report;
2) E
lectrical Energy Storage (December 2011),
hereafter referred to as the MSB EES
Report.
The reports primary goal is to provide a
comprehensive, global view on the state
of the art and future directions for grid
integration of large-capacity RE sources
and the application of large-capacity energy
storage for that purpose. It is directed towards
the IECs partners worldwide, as well as to
its own Standardization Management Board
(SMB) and Conformity Assessment Board
(CAB), such that they may act to support
grid integration efforts around the world and
provide guidance to the electric utility industry
and policy-makers.
The report is divided into seven sections:
(1) Introduction; (2) RE generation: the present,
the future and the integration challenges;
(3) Present: state of the art in integrating largecapacity RE; (4) Future: technical solutions
for integrating more large-capacity RE;
(5) Application of large-capacity EES to support
RE integration; (6) Standards for large-capacity
RE integration; and (7) Conclusions and
recommendations. Sections 1 and 2 provide
background information about the report, the
state of RE generation, and the challenges of
integrating RE sources into the grid. Section 3
describes the key practices and technologies

presently involved in grid integration of largecapacity RE. Section 4, the heart of the report,
explores the future technology and practice
needs of the grid as RE penetration increases.
Section 5 discusses the role of energy storage
as a supportive technology for grid integration
of RE, and maps its supportive roles onto the
general needs identified in Section 4. Section 6
connects the information from the previous
five sections with relevant IEC standards

activities, and identifies future standards
needs. Section 7 provides a strategic overview
and recommendations to the relevant policymakers, regulators, power utilities, industry
and research communities, as well as the
IECs own committees.
11
SECTION 2
RE generation:
the present, the future and
the integration challenges
14
S E C T I O N 2
d the integration challenges
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
2.1 Drivers of RE development

RE is a growing component of electricity grids
around the world due to its contributions to
(1) energy system decarbonization, (2) longterm energy security, and (3)expansion of
energy access to new energy consumers in the
developing world. As stated in the MSB EEE
Report:
In short: the challenge is ensuring energy
availability and preserving the environment. The
key elements are the following:
1) Stabilizing climate impact from fossil fuel use
2) Meeting the energy demand of a growing
population
3) Bringing electricity to the 1.6 B people
without access
4) Ensuring stable and secure energy access
for all nations
5) Transporting electricity long distances from
where it is generated to where it is used.
[msb10]
RE is implicated in all of these elements, and is
critical to transforming energy grids to meet the
environmental, economic and social challenges
of the future. Globally, REs share of electricity
generation will increase substantially over the
next two decades and beyond. Indeed, this is
already occurring: governmental action at the
international, national and subnational levels has
created a wide variety of laws and policies to
promote RE development. These include:
carbon taxes: taxation of greenhouse gas
emissions, so as to internalize the climatedisruption costs of fossil-fuel use;
cap-and-trade systems: provision of
tradable annual emissions allowances to
greenhouse gas emitters coupled with
reduction in the quantities of allowances
issued each year;
RE goals: mandates requiring load-serving

entities to source a specified proportion of
energy sold from renewable sources;
feed-in tariffs (FiTs): guaranteed wholesale
prices for RE coupled with a requirement
that load-serving entities take renewable
power whenever it is available;
tax credits: credits against taxable income
for generation or installation of RE;
the development of smart grids: advances
in the architecture, functionality and
regulation of electricity grids so as to
enable higher penetrations of RE; and
removal of long-standing fossil fuel
subsidies.
We will discuss the major public policy drivers
behind RE development in turn.
2.1.1 Decarbonization
The need to address global climate change,
a worldwide environmental phenomenon that
will affect everyone on the planet, is the most
public driving force for RE deployment. The
Intergovernmental Panel on Climate Change
(IPCC), the worlds leading authority on climate
change science, states in its Synthesis Report
to the Fourth Assessment Report that warming
of the climate system is unequivocal, as is
now evident from observations of increases
in global average air and ocean temperatures,
widespread melting of snow and ice and rising
global average sea level, and that most of the
global average warming over the past 50 years
is very likely due to anthropogenic greenhouse
gas (GHG) increases and it is likely that there is a
discernible human-induced warming averaged
over each continent (except Antarctica).
The MSB EEE Report notes that CO2 emissions

related to energy use account for 70% of total
GHG emissions, and that emissions related
to electricity generation approach half of that
[msb10]. Consequently, governments have
enacted policies to curb GHG emissions from
the power sector. Because electricity generated
from RE produces no GHG emissions,
increasing penetrations of RE onto the electrical
grid contribute to a decarbonization of the
electricity system: a reduction in GHGs emitted
per unit of energy produced. Energy system
decarbonization in turn slows the increase in
concentrations of GHGs in the atmosphere and
thereby mitigates the resultant radiative forcing
of the climate system.
In recent years, the ostensible progress of climate
change policies has stalled at the international
level, with a lack of hard commitments to
emission reductions from some large emitters.
Nevertheless, many countries have developed
incremental policies to promote RE development
in the absence of full international agreement. For
example, nearly 30 states in the USA have enacted
their own RE goals in the absence of federal
action; Germany has long used aggressive feed-in
tariff requirements that oblige power companies
to purchase renewably-generated energy at
fixed rates; and China has set a capacity goal of
150GW-180GW of wind power and 20GW of
solar photovoltaic (PV) power for 2020. These
goals and policies will result in significant growth in
RE that will affect the operation of the power grid.
average energy costs tend to decline over time

for renewable generation, as variable costs are
limited to operations and maintenance and do not
include fuel. Secondly, RE assets are insulated
from fluctuations in fossil fuel prices, which are
historically volatile and subject to geopolitical
disruptions. Coal, gas and oil-fired generation
costs, in contrast, increase when the cost of the
relevant fuel increases. Figure 2-1 depicts the
International Energy Agencys (IEA) projections1
for the share of world electricity generation by
fuel up to 2035, and shows a displacement of
coal and oil-based generations shares by wind,
biomass and other renewables as governments
continue to promote RE.
Because fossil fuel supplies are both unevenly
distributed and ultimately exhaustible, many
countries have identified a long-term energysecurity proposition in gradually decreasing
dependence on them in the production of electricity.
In comparison to fossil resources, renewable
resources are better distributed throughout the
world and do not diminish as they are used. A
countrys investment in RE results in a zero-fuelcost generation resource that is domestically
located. Thus even countries with substantial fossil
fuel resources, such as China, have set aggressive
wind power targets. And despite a recent boom
in natural gas production in the USA, states have
made no indication of any intent to remove RE
goals. RE can also prove useful for short-term
energy security concerns. Many electric utilities
have diversified their generation mixes with
renewables so as to hedge against volatile fossil
fuel prices on the oil, gas and coal markets.
2.1.2 Energy security

Driven by the wind, the sun and the waves, RE
has no fuel costs. This zero-fuel-cost aspect
of RE manifests itself in two benefits. First,
These projections come from IEAs New Policies

Scenario, the centrepiece of IEAs analysis in its
World Energy Outlook. The new policies scenario
accounts for future policy developments that drive
world energy sources toward greater sustainability.
15
16
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
2009
20 043 TWh
2020
27 881 TWh
2035
36 250 TWh
0%
20 %
Coal
Gas
40 %
Oil
Nuclear
60 %
Biomass
Hydro
80 %
Wind
100 %
Other renewables
Figure 2-1 Share of world electricity generation by fuel in IEAs New Policies Scenario
[weo11]
2.1.3 Expanding energy access

Energy demand in developing countries is
growing rapidly (see Figure 2-2). IEAs New
Policies Scenario projects electricity demand in
non-OECD countries to increase at a compound
average annual growth rate (CAAGR) of 3.5%
to 2035. Total non-OECD electricity demand
nearly triples from 8 000 TWh in 2009 to almost
20 000 TWh by 2035 (see Figure2-3). Asian
electricity demand grows the most rapidly,
with a 4.2% CAAGR in the same period. In
addition to the needs outlined in the previous
subsections for cleaner energy and more
secure energy, the world simply needs more
energy as more people in the developing world
gain access to it.
As global energy demand increases, RE

provides one means among many of adding
energy assets to the system alongside growth
of other resources. IEAs New Policies Scenario
projects a near tripling of global use of RE, from
3 900 TWh in 2009 to 11100 TWh in 2035,
and growth in renewables accounts for nearly
half of the total increase in generation by 2035.
Indeed, under this scenario, a full third of global
electricity generation will be supplied by RE
(including hydroelectricity) by 2035. Figure
2-4 provides a breakdown of incremental
renewables growth by technology. Note the
large increase in wind power.
Mtoe
18 000
Other OECD
16 000
European Union
14 000
United States
12 000
Other non-OECD
10 000
Middle East
8 000
India
6 000
China
4 000
Inter-regional
(bunkers)
2 000
0
1990
2000
2010
2020
2030
2035
Figure 2-2 World primary energy demand by region in IEAs New Policies Scenario
[weo11]
Figure 2-3 Electricity demand by region in IEA's WEO 2011 Scenario (TWh)
[weo11]
17
TWh
18
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
8 000
46 %
7 000
44 %
Solar PV
6 000
42 %
Biomass and waste
5 000
40 %
Hydro
4 000
38 %
3 000
36 %
2 000
34 %
1 000
32 %
Other renewables
Wind
Share of renewables
in total increase in
generation (right axis)
30 %
2015
2020
2025
2030
2035
Figure 2-4 Incremental global renewables-based electricity generation relative to 2009 by

technology in IEAs New Policies Scenario
[weo11]
While a world where a majority of electricity

generation is based on renewable sources
is far beyond the horizon, it is clear that the
confluence of government policy, utility planning
and global demand growth has the potential
to increase penetrations of RE substantially
on electricity grids worldwide. This shift in
generation portfolios will have profound effects
on the operation of the grid, which will in turn
affect the operation of RE resources themselves
as well as the operation of other resources and
equipment connected to the grid.
2.2 Present status of RE

generation and future
projections
At 39 02 TWh, RE accounted for 19.46% of
the worlds electricity generation in 2009.
Hydroelectricity, by far the largest contributor
among the renewables, accounted for over
83% of that share. Biomass, wind and solar PV
combined accounted for only 15% of the global
RE contribution, or 2.9% of world electricity

generation [weo11]. Thus, while RE as a whole
comprises a substantial portion of global
electricity generation, the proportion of RE that
comes from variable sources such as wind
and solar is still relatively limited. Consequently
most power system operators to date have had
relatively small amounts of variable generation
to integrate. As we examine here, however, the
situation may change materially by 2035.
The IEA projects that global electricity production
from renewables (including hydroelectricity)
will grow to 8108 TWh by 2025, an over100% increase from 2009. By 2035, that figure
rises to 11100 TWh, as illustrated in Figure
2-5. These estimates are based on IEAs New
Policies Scenario, which takes into account
recently announced commitments and plans,
even if they are yet to be formally adopted
and implemented [weo11]. The New Policies
Scenario is the central scenario for IEAs World
Energy Outlook 2011, and assumes a global
CO2 price of 30USD to 45 USD per tonne.
11 100
12 000
9 540
10 000
8 108
6 712
TWh
8 000
5 394
6 000
3 902
4 000
2 317
2 000
0
2009
2015
2020
2025
2030
2035
3 000
2 000
1 000
0
1990
2009
Brazil
Latin
Africa
Middle East
India
China
Russia
Japan
OECD
2015
US
TWh
1990
Country/Region
2020
2025
2030
2035
Figures 2-5 and 2-6 RE generation globally and by country/region to 2035

[weo11]
Figure 2-6 displays the projected growth of RE

generation by region or country [weo11]. Notably,
while the USA, OECD Europe, Latin America
and China have relatively similar numbers
in 2009, the growth rates are dramatically
different. Chinas growth substantially outpaces
OECD Europes, and OECD Europes growth
substantially outpaces that of the USA and
Latin America. Africa and the Middle East
see relatively little growth in renewables. India
exhibits an aggressive growth rate, but begins
2009 with smaller numbers than other regions,
and so does not see the same degree of
absolute growth as neighbouring China.
It is important to note that charts in this
subsection referring to present RE capacities
reflect the state of the market in 2009, which
is the most recent year for which present data
is available from the IEA. However, RE capacity
has already expanded substantially since then,
with some notable developments in 2010. We

discuss these developments in the text when
relevant.
2.2.1 Wind energy

Wind energy plants around the world produced
273 TWh of electricity in 2009, from an estimated
installed capacity of 159 GW. IEA's estimates
of 2009 wind energy generation and capacity
by region and country are provided in Figures
2-7 and 2-8 [weo11]. Wind power developments
in 2010 have been substantial: China installed
over 16GW of new wind capacity in 2010,
bringing its total to 42GW. This exceeded the
US 2010 total of 40GW, and made China the
world leader in wind capacity for the first time.
Europe installed nearly 10GW of wind in 2010,
bringing its total capacity to 86GW, over half of
which is located in Germany and Spain [smp11].
19
Africa
Latin America
1
Brazil
0
Middle East
0
India
11
2
China
Brazil
26
Japan
35
Russia
76
US
2
Africa
0
India
Japan
18
Middle East
0
China
Russia
OECD Europe
27
Latin America
GW
74
80
70
60
50
40
30
20
10
0
OECD Europe
135
US
TWh
160
140
120
100
80
60
40
20
0
Country/Region
Country/Region
Figures 2-7 and 2-8 Wind energy generation and capacity by country/region in 2009
[weo11]
3 000
1 724
GWh
835
2035
2030
2025
2020
2035
2030
2025
2020
2015
2009
1990
159
2015
200
4
0
397
400
273
500
582
600
2009
1 000
752
800
1 282
1 500
921
1 000
2 182
2 000
1 102
1 200
2 703
2 500
TWh
20
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
Figures 2-9 and 2-10 Global wind energy generation and capacity projections to 2035
[weo11]
IEAs New Policies Scenario projects 1282TWh

of annual wind-generated electricity globally by
2020 [weo11], a 369% increase from 2009. By
2030 that figure reaches 2182 TWh, a neardoubling of the 2020 estimate over the course
of a decade, as shown in Figure 2-9 [weo11].

In terms of capacity, IEA projects growth from
159GW in 2009 to 582GW in 2020, reaching
1102GW by 2035, as shown in Figure 2-10
[weo11].
1990
2009
2015
Brazil
Latin
Africa
Middle East
India
China
Russia
Japan
OECD
2020
US
TWh
900
800
700
600
500
400
300
200
100
0
2025
2030
2035
Country/Region
Figure 2-11 Wind energy generation to 2035 by region/country

[weo11]
Wind capacity growth over this period is

dominated overwhelmingly by China, OECD
Europe and the USA, as shown in Figure 2-11.
Indeed, while the current disparity between
these countries and the rest of the world in wind
capacity is stark, it is dwarfed by future growth
estimates, by which the leaders will outpace the
others by orders of magnitude. OECD Europe
and China maintain growth in lockstep through
2035, leaving the USA somewhat behind, though
still a major player. It is also apparent that Latin
Americas growth in renewables overall does not
translate to a significant growth in wind.
Regionally, the OECD European countries
together show the strongest wind growth, slightly
outpacing China. 76GW of European wind
power produced 135TWh of electricity in 2009
[weo11]. Germany, Spain, Italy and France are
the major contributors to wind energy capacity
in this region [gwe10]. In Europe, the majority of
wind farms developed during the past ten years
have been onshore and small-capacity. With
many wind-rich areas now thoroughly exploited,
European wind developers are turning their
attention to large-capacity offshore wind farms
with centralized integration to the power grid.
By 2020, IEA projects wind capacity of 209GW

and 449TWh of generation in Europe. By 2030,
capacity reaches 289GW and generation
reaches 675TWh [weo11]. Germany has set a
target of 45.75GW of wind capacity for 2020
[ger10], and Spain a target of 38GW [esp10].
These plans contribute substantially to Europes
regional estimate, particularly in the next decade.
If we examine single countries rather than
regions, China is the worlds tour de force in
wind power development. 26GW of wind power
supplied 27TWh of electricity in China in 2009,
ranking it third globally in wind capacity. A year
later, China had jumped into first place with a
total of 42GW in 2010 [smp11] [weo11]. China
is set to lead the world in wind generation and
wind capacity by 2035. The IEA predicts China
will produce 388TWh of electricity from wind in
2020, and the National Energy Administration
(NEA) of China has set a target of 150-180GW of
wind capacity by the same date [sgc12], which
matches IEAs estimate of Chinas installed wind
capacity of 180GW. By 2030, IEA projects that
China will reach 280GW of wind capacity, just
behind estimates for the combined European
countries [weo11].
21
22
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
US wind capacity stood at 35GW in 2009,

generating 74TWh of electricity [weo11]. Most
of US wind capacity is concentrated in the
states of Texas, Iowa, California, Michigan
and Washington, and is onshore [wpa12].
As a result of declining energy demand, an
economic recession and a precipitous drop in
North American natural gas prices, the USA
did not keep pace with Europe and China in
2010, installing only 5GW to Europes 10GW
and Chinas16 GW. Still, the USA is expected to
remain a significant player in wind. IEA projects
that US wind generation will grow to 165TWh
by 2015, more than double its 2009 value. By
2030, the capacity grows to 388TWh from
151GW [weo11].
Japans 2GW of wind capacity produced
3TWh of electricity in 2009 [weo11]. IEA
estimates Japanese wind capacity to grow to
7GW by 2020, producing 18TWh of electricity,
and to 15GW by 2030, producing 41TWh of
electricity [weo11]. Though these numbers are
dwarfed by those from geographically larger
regions such as China, OECD Europe and the
USA, it is worth noting that the expected rate
of increase of wind generation and capacity on
the Japanese grid is dramatic: generation is
expected to grow by 650% between 2009 and
2030 under the IEA's New Policies Scenario.
The figures above do not differentiate between
onshore and offshore wind. However, the sorts
of integration challenges presented may differ
between onshore and offshore wind projects,
specifically with regard to the need for special
transmission technologies for offshore plants. We
therefore briefly examine the offshore segment
of the wind market, which at present exists
almost entirely in Europe, with a few projects in
China. Europes offshore wind capacity stood
at 4GW at the end of 2011, with an additional

6GW under construction at the time and 17GW
consented to by EU member states [ewe11]. The
majority of these projects are in the UK, Denmark
and Germany, with some projects in Belgium,
the Netherlands and Sweden. The European
Wind Energy Association (EWEA), an industry
association, projects that Europe will have 40GW
of offshore wind by 2020 producing 148TWh of
energy, and 150GW producing 562TWh by 2030.
While industry estimates must be taken with the
proverbial grain of salt, these numbers at least
plausibly harmonize with IEAs OECD European
wind (off- and onshore) projections of 209GW by
2020 and 298GW by 2030. EWEA itself identifies
the availability of high voltage direct current
transmission (HVDC) as a critical bottleneck for
the development of offshore wind in Europe.
2.2.2 Solar energy

Grid-relevant solar energy technologies can be
divided into two types: PV and concentrated
solar power (CSP). PV generates electricity
directly, converting sunlight to electricity
through a semiconductor such as silicon. CSP
technologies produce electricity by reflecting
and concentrating sunlight onto a fluid, which
then heats and boils water, the steam from which
then drives a turbine that produces electricity.
Presently, CSP has a lower contribution to RE
production than solar PV. We will discuss each
market in turn, beginning with the larger PV
market.
Solar PV generated 20TWh of electricity from
22GW of global capacity in 2009 (see Figures2-12
and 2-13) [weo11]. The OECD Europe region far
surpassed all other regions in both capacity
and generation, despite its relatively weak solar
resource. This apparent discrepancy is explained

by highly favourable policy environments for solar
PV in many European countries.
several decades. The IEA projects solar PV

generation of 230TWh from 184GW of capacity
in 2020, an over 1000% generation increase from
2009. By 2030, those figures reach 551TWh and
385GW, more than double the 2020 estimates.
Figures 2-14 and 2-15 display IEA's projections
for solar PV energy production to 2035 [weo11].
16
14
12
10
8
6
4
2
0
14
GW
TWh
Though solar PV capacity is many times smaller

than wind capacity at present, it is expected to
grow at a faster pace than wind over the next
2
US
OECD
Europe
17
18
16
14
12
10
8
6
4
2
0
Japan
US
Country/Region
OECD
Europe
Japan
Country/Region
Figures 2-12 and 2-13 Solar PV energy generation and capacity in 2009 by country/region
[weo11]
800
741
700
551
TWh
600
500
369
400
300
230
200
100
126
0
20
1990
2009
TWh
0
2015
2020
2025
2030
2035
180
160
140
120
100
80
60
1990
2009
2015
2020
40
20
0
2025
US
OECD Japan
Europe
Russia
China
India
Middle
East
Africa
Latin
Brazil
America
2030
2035
Country/Region
Figures 2-14 and 2-15 Energy generation from solar PV globally and by country/region
[weo11]
23
24
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
In the OECD Europe region, solar PV produced

14TWh of electricity from 17GW of solar
PV capacity in 2009 [weo11]. Favourable
government policies and pricing have led to
higher penetrations, particularly in Spain, Italy
and Germany. In Germany, the government has
opted for a feed-in tariff, in which the utilities
pay the owner of a solar PV system a set tariff
for renewable power over a period of time
[obo12]. Consequently, solar PV provided 3%
of the total power in Germany in 2011 [eck11].
Germany led the world in PV capacity in 2009
with 9785MW. Spains 2009 capacity figure,
at 3386MW, was lower but still substantial in
comparison to other countries [epa10]. Italy
has ramped up solar PV capacity dramatically
since then, reaching 12750MW and producing
10TWh of energy in 2011 [gse12].
IEA projects 90TWh from 84GW of OECD
European capacity by 2020 and 139TWh from
115GW by 2030 [weo11]. Germany expects
its solar PV capacity to reach 52GW by 2020
[ger10], and Spain estimates 8.4GW by the
same year [esp10]. It is worth noting that
Europes generation capacity factors (the ratio of
energy generated from a given unit of power) for
solar PV are lower than those for the USA. This
disparity is explained by differences in the quality
of the resource: the USA receives much more
sunlight than Europe. Nevertheless, Europes
policy environment provides substantially more
support to solar power, particularly in Germany
and Spain, than does the US policy environment,
explaining the capacity estimate differences as
well as the ultimately higher generation estimates
for Europe.
US solar PV generated 2TWh of electricity from
2GW of capacity in 2009 [weo11]. IEA estimates
US solar PV generation at 38TWh from 25GW
of capacity in 2020 and 81TWh from 50GW of

capacity in 2030. [weo11] Note that the 2030
estimate for US solar PV capacity is roughly a
third of estimated US wind power capacity in
the same year.
Japan generated 3TWh of its electricity from
solar PV sources in 2009 from 3GW of capacity
[weo11]. By 2010, Japan had increased its
solar PV capacity to 3.6GW. This increase
is attributable to a subsidy programme for
residential PV system installations and another
programme to purchase surplus PV power from
small systems at double the retail electricity
price [yam11]. IEA projects 18TWh of electricity
from 17GW of Japanese solar PV by 2020, and
32TWh from 28GW by 2030 [weo11].
China did not produce any significant amounts
of electricity from solar PV in 2009, but
that is changing rapidly, as it has become a
manufacturing leader in the technology. IEA
projects that China will produce 29TWh from
20GW of solar PV by 2020, and 89TWh from
58GW by 2030 [weo11]. This places China
behind the USA in solar PV generation in
2020, but ahead of it by 2030 [weo11]. Chinas
National Development and Reform Commission
has set targets for China to achieve 10GW
of solar capacity in 2015, and 50GW of solar
capacity installed by 2020 [won11].
CSPs market is much smaller than wind
power or solar PV, and it is less challenging
to integrate into the power system due to its
thermal aspects, which reduce variability in
output. CSP produced 1TWh of electricity in
2009 from a global capacity of 1GW, located
primarily in the USA, though Spain has since
taken the lead [weo11].
400
307
TWh
300
167
200
100
1
1990
2009
52
24
92
0
2015
2020
2025
2030
2035
Figure 2-16 Global CSP energy generation to 2035

[weo11]
CSP generation estimates are lower than those

for PV, but exhibit similar strength in growth
rates. IEA projects 52TWh of CSP-generated
energy from 14GW of capacity in 2020, and
167TWh from 45GW in 2030. Figure 2-16
displays IEAs projections for global CSP
generation to 2035 [weo11].
Spain led the world in 2010 in CSP capacity at
over 632MW. Spanish CSP capacity grew by
400MW in 2010 due to a Royal Decree from the
Spanish government that provided incentives
for solar energy. In 2011, it began construction
on nearly 1 GW of additional CSP capacity
[rep11]. IEA projects 14 TWh of electricity from
4 GW of CSP sources in OECD Europe by 2020.
In 2030, that rises to 36 TWh from 10 GW. The
Spanish government, however, estimates that
Spain alone will install 5 GW of CSP to produce
15.35 TWh by 2020, more than IEAs projection
for all of Europe [esp10].
IEA projections for US CSP closely track those
for OECD Europe, with 14 TWh from 4 GW in
2020, and 30 TWh from 8 GW in 2030.
2.3 RE grid integration challenges

Wind and solar generation both experience
intermittency, a combination of non-controllable
variability and partial unpredictability, and
depend on resources that are locationdependent [per11]. These three distinct aspects,
explained below, each create distinct challenges
for generation owners and grid operators in
integrating wind and solar generation.
Non-controllable variability: Wind and
solar output varies in a way that generation
operators cannot control, because wind
speeds and available sunlight may vary
from moment to moment, affecting
moment-to-moment power output. This
fluctuation in power output results in the
need for additional energy to balance
supply and demand on the grid on an
instantaneous basis, as well as ancillary
services such as frequency regulation
and voltage support. Figure 2-17 provides
a graphical example of hourly wind power
variability.
25
26
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
Figure 2-17 Hourly wind power output on 29 different days in April 2005 at the Tehachapi
wind plant in California
[haw06]
Partial unpredictability: The availability

of wind and sunlight is partially
unpredictable. A wind turbine may only
produce electricity when the wind is
blowing, and solar PV systems require the
presence of sunlight in order to operate.
Figure 2-18 shows how actual wind power
can differ from forecasts, even when
multiple forecast scenarios are considered.
Unpredictability can be managed through
improved
weather
and
generation
forecasting technologies, the maintenance
of reserves that stand ready to provide
additional power when RE generation
produces less energy than predicted, and
the availability of dispatchable load to soak
up excess power when RE generation
produces more energy than predicted.
Location dependence: The best wind

and solar resources are based in specific
locations and, unlike coal, gas, oil or uranium,
cannot be transported to a generation site
that is grid-optimal. Generation must be colocated with the resource itself, and often
these locations are far from the places
where the power will ultimately be used.
New transmission capacity is often required
to connect wind and solar resources to
the rest of the grid. Transmission costs
are especially important for offshore wind
resources, and such lines often necessitate
the use of special technologies not found
in land-based transmission lines. The global
map in Figure 2-19 displays the latest data
on mean land-based wind speeds around
the world.
30000
Wind power production [MW]
25000
20000
15000
10000
5000
0
1
Scenario 1
Scenario 6
11
13
15
17
Scenario 2
Expected Value
19
21
23
Scenario 3
Realized Value
25
27
Scenario 4
29 Forecast
31
33hour35[h]
Scenario 5
Figure 2-18 Example of a day-ahead forecast scenario tree for the wind power forecast for
the PJM region of the United States
[mei10]
Figure 2-19 Global mean wind speed at 80m altitude

[tie12]
27
28
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
Because the presence of wind and sunlight are

both temporally and spatially outside human
control, integrating wind and solar generation
resources into the electricity grid involves
managing other controllable operations that may
affect many other parts of the grid, including
conventional generation. These operations
and activities occur along a multitude of time
scales, from seconds to years, and include new
dispatch strategies for rampable generation
resources, load management, provision of
ancillary services for frequency and voltage
control, expansion of transmission capacity,
utilization of energy storage technologies, and
linking of grid operator dispatch planning with
weather and resource forecasting [per11].
The essential insight to integration of variable
RE is that its variability imposes the need for
greater flexibility on the rest of the grid, from
other (controllable) generators to transmission
capacity to loads. Discussion of variable
generation operation alone is insufficient to
describe the full impact of high penetrations
of RE on power system operation. Thus this
report explores RE integration from both a plant
operator and a system operator perspective,
so as to identify the full range of operations
involved.
On the seconds to minutes time scale, grid

operators must deal with fluctuations in frequency
and voltage on the transmission system that, if left
unchecked, would damage the system as well as
equipment on it. To do so, operators may order
generators to inject power (active or reactive) into
the grid not for sale to consumers, but in order to
balance the actual and forecasted generation of
power, which is necessary to maintain frequency
and voltage on the grid. These ancillary services go
by a plethora of names and specific descriptions.
Typical services for an impressionistic overview
include:
frequency regulation: occurs on a
seconds-to-minutes basis, and is done
through automatic generation control
(AGC) signals to generators;
spinning reserves: generators available to
provide power typically within 10 minutes.
These reserves are used when another
generator on the system goes down or
deactivates unexpectedly;
non-spinning reserves: these generators
serve the same function as spinning
reserves, but have a slower response time;
voltage support: generators used for
reactive power to raise voltage when
necessary;
black-start capacity: generators available
to re-start the power system in case of a
cascading black-out.
2.3.1 Non-controllable variability

Variability in the context of wind and solar
resources refers to the fact that their output is
not constant. It is distinct from unpredictability,
which we discuss in the following section. Even
if operators could predict the output of wind
and solar plants perfectly, that output would still
be variable, and pose specific challenges to the
grid operator, which we introduce here [per11].
Additionally, grid operators must track loads

demand for electricity on the consumption
side of the grid and ensure that generation
matches load at all times. This load following
function becomes particularly important at
times of day when demand for electricity
increases substantially, such as morning, a hot
afternoon, or evening. Load following may be
provided through a class of ancillary service or
through a fast energy market, depending on

the system operator.
These functions are not new. Grid operators
have been regulating frequency and voltage,
maintaining reserves and following shifts in load
since the development of the electricity grid.
This is because loads themselves are variable,
and even conventional, controllable generation
experiences problems and cannot perform as
scheduled all of the time. Consumers demand
electricity in ways that, while predictable, are not
controllable and have some degree of variability.
Thus wind and solar generation does not introduce
entirely novel problems with which operators have
never grappled. Indeed, at low penetrations, the
integration challenges are primarily device and
local-grid specific, such as subsynchronous
resonance and harmonics, which the turbine itself
may cause. These issues are explained in greater
detail in section 3, and their solutions tend to be
device-specific rather than grid-level.
However, high penetrations of wind and solar
generation will add more variability to the energy
system than grid operators have traditionally
managed in the past, and thus increase demand
for ancillary services and balancing energy overall.
It is more difficult, and sometimes impossible, to
manage such challenges at the device level, and
so grid-level actions, technologies and strategies
are often needed. Wind and solar resources in
sufficient amounts may also complicate load
following functions when large demand shifts
coincide with weather events that alter power
output from wind or solar resources. Grid
operators located in more remote regions and
serving smaller loads may have less flexibility
to provide ancillary services and load following
than their larger counterparts. Compounding
matters, plentiful RE resources are often located
in these remote locations. The IEA and other

bodies have recommended consolidation of grid
operators, in order to integrate RE sources over
larger areas and so reduce the variance of the
power produced, as well as easing of market
restrictions on sales of ancillary services as a
solution to this problem [iea09].
2.3.2 Partial unpredictability

Partial unpredictability, also called uncertainty,
is distinct from variability. The variability of wind
and solar generation is ever-present, a result of
reliance on the ever-changing wind and sun, and
affects the system at the moment-to-moment
time scale as a cloud passes over a PV plant
or the wind drops. Partial unpredictability, on
the other hand, refers to our inability to predict
with exactness whether the wind and sun will be
generally available for energy production an hour
or a day from now. This hour-to-day uncertainty
is significant because grid operators manage
the great majority of energy on the grid through
unit commitment, the process of scheduling
generation in advance, generally hours to a full
day ahead of time, in order to meet the expected
load. When actual production does not match
the forecast, the grid operator must balance the
difference. RE generation increases the cost of
this function by increasing the spread between
predicted and supplied energy, a cost that is
ultimately borne by consumers.
Unit commitment at present is largely
deterministic, meaning that once a generator is
scheduled to run, its full capacity is expected
to be available for use. This practice reflects
the relative predictability and controllability of
traditional coal, gas and hydropower generation
resources. Operators ensure the availability of
29
30
S E C T I O N 2
an
re
tu
fu
e
th
t,
en
es
pr
e
th
:
RE generation
reserves generators that withhold the supply of

energy and so stand ready to balance the system
in an emergency so as to protect against a
potential transmission line or generator outage.
But the process of unit commitment and the
calculation of reserves needed to ensure reliability
becomes more complex when dealing with
stochastic (uncertain) generation, whose output
at the committed time carries some degree of
uncertainty. Forecasting technologies aim to
predict weather and thus generation output
from wind and solar resources at various timescales more accurately, and communicate those
predictions to grid operators in a manner that
allows the operator to more effectively schedule
and dispatch resources. Properly anticipating
wind and solar output levels allows the operator
to modify the scheduling of other generators so as
to more optimally utilize all assets under the grid
operators purview. The operator must, for example,
ensure that reserves are available not only to cover
transmission line or generator outages, but also
to respond to still unanticipated changes in wind
and solar output. Assisting the operator in this
process are advanced unit commitment methods,
which aim to prepare the system for multiple
potential and uncertain outcomes that cannot
be predicted by the forecasting technologies.
Unlike deterministic unit commitment processes,
advanced unit commitment methods must take
into account the stochastic nature of wind and
solar generation and their relative concentration
on the system in recommending the scheduling of
other resources. Ultimately, the goal of advanced
unit commitment is to cost-effectively maintain
sufficient flexibility on the system, such that the
integration of RE resources neither exposes the
system to unacceptable reliability risks nor overschedules reserves in a way that unnecessarily
burns fuel and emits pollution.
2.3.3 Locational dependency

Far removed from the day-to-day management
of the grid is its long-term planning specifically
the siting and utilization of new transmission
lines. Here RE generation plays a significant role
and introduces new challenges. Because wind
and solar resources are often located in remote
locations, far from load centres, developing
sufficient transmission to move RE to markets
is critical to their integration.
Transmission planning processes are highly varied,
and tend to be influenced by regional politics.
For example, a transmission line may provide
capacity for energy produced in one country or
state, passed through another, and consumed
in yet another. These disparities in generation
capacity, transmission location and load size
between locations can make the development
of transmission for RE contentious and complex,
particularly with respect to cost allocation.
Because new transmission lines built out to RE
generation resources will carry primarily renewably
generated, variable and partially unpredictable
electricity, technical needs arise regarding the
transmission technology to be used.
On the other hand, distributed energy resources
provide for an alternative vision of the future grid,
where energy is generated and used locally on
a micro-grid, avoiding the cost of line losses
and the high capital cost of transmission lines.
In such a schema, the electricity grid could be
conceptualized as a collection of independent
micro-grids with vastly reduced long-distance
energy transmission needs.
SECTION 3
Present: state of the art in
integrating large-capacity RE
32
SECTION 3
egrating large -capacity RE
Present: state of the ar t in int
3.1 General
transmission technology and the operational

technology and practices.
Although on a system-wide level RE power

plants generate electricity just like any other
power plants, RE power has quite distinctive
characteristics in generation, transmission
and operation technology when compared to
conventional generation. Understanding these
distinctive characteristics and their interaction
with the other parts of the power system is the
basis for the integration of large-capacity RE
power in the grid.
3.2 RE generation technology

3.2.1 Wind power generation
1) Basics of wind power generation
Wind turbine generators (WTGs) extract energy
from wind and convert it into electricity via an
aerodynamic rotor, which is connected by a
transmission system to an electric generator
[iea11]. Todays mainstream WTGs have three
blades rotating on a horizontal axis, upwind of
the tower (see Figures 3-1 and 3-2). Two-blade
WTGs (see Figure 3-3) and vertical-axis WTGs
(see Figure 3-4) are also available.
In this chapter, the state of the art of the

technologies and practices related to largecapacity RE integration is described to facilitate
the understanding of their interaction with the
power grid. This discussion is further divided
into the RE generation technology itself, the
Figure 3-1 A standard WTG with three blades and horizontal axis
(SGCC)
How Wind Power Works Horizontal-axis Turbine

Rotor Blade
Tower
Nacelle
Hub
Low-speed High-speed
Shaft
Shaft
Rotor Hub
Gearbox
Figure 3-4 Vertical-axis turbine

(Xu Ji)
Transformer
Brake
Brake
Generator
Figure 3-2 Structure diagram of a WTG

(HowStuffworks)
In general, a WTG can begin to produce power

in winds of about 3 m/s and reach its maximum
output around 10m/s to 13m/s. Power output
from a WTG increases by the third power of
wind speed, i.e. a 10% increase in wind speed
increases available energy by 33
%, and is
directly proportional to the rotor-swept area (the
area swept by the rotating blades). Power output
can be controlled both by rotating the nacelle
horizontally (yawing) to adapt to changes in
wind direction, and rotating the blades around
their long axes (pitching) to adapt to changes in
wind strength.
The capacity of WTGs has doubled
approximately every five years, but a slowdown
in this rate is likely for onshore applications due
to transport, weight and installation constraints.
Typical commercial WTGs at present have a
capacity of 1.5MW-3 MW; larger ones can
reach 5MW-6MW, with a rotor diameter of up
to 126 metres [iea11].
Figure 3-3 Two-blade WTG

(Ming Yang)
Since a single WTG has limited capacity, much

less than a conventional power generator, a wind
power plant (WPP, usually called wind farm)
33
34
SECTION 3
normally consists of many WTGs connected

together by overhead lines or cables. Their
power output is collected and transmitted to
the grid through an alternating current (AC) or
direct current (DC) line, after voltage step-up

at the substation in the WPP (see Figure 3-5).
Some WPPs now have a capacity comparable
to that of conventional power generators.
Power Grid
High voltage busbar
Step-up
transformer
Low voltage busbar
Overhead
line or cable
Wind
turbine
Figure 3-5 Structure diagram of a wind farm

(SGCC)
2) T
ypes and characteristics of major
WTGs
WTGs currently in operation mainly include
the following four types. Each type has some
unique characteristics due to its features in
topology [ewe05] [ner09] [sge11].
Type 1 Fixed speed induction generator:
Introduced and widely used in the 1980s, type 1
WTGs (see Figure 3-6) are based on a squirrel

cage induction generator (SCIG). They can only
slightly vary their rotation speed (1%-2%), i.e.
they are almost fixed speed, thus their output
fluctuates as wind speed varies. To alleviate
this problem, a double-speed version of the
type 1 WTG was developed. Original type 1 WTGs
have few control options but passive stall. Active
stall designs can be adopted in improved type 1
WTGs, where the blades can pitch towards stall by

a control system. Because induction generators
absorb a lot of reactive power when generating
active power, type 1 WTGs are generally equipped

with reactive power compensators.
Squirrel cage
induction generator
Transformer
Grid
Gear box
Reactive power
compensator
Figure 3-6 Topology of a fixed speed induction generator
(SGCC)
Type 2 Variable-slip induction generator:

Introduced in the 1980s and 1990s, type 2 WTGs
(see Figure 3-7) are equipped with a wound rotor
induction generator (WRIG). Power electronics
are applied to control the magnitude of the WRIGs
rotor current, which allows a speed variation of
10% up and down, improving power quality and

reducing the mechanical loading of the turbine
components. Type 2 WTGs are equipped with an
active pitch control system. Like type 1 WTGs,
type 2 WTGs are also generally equipped with
reactive power compensators.
Wound rotor
induction generator
Transformer
Grid
Gear box
Rectifier
IGBT R Control
Reactive power
compensator
Figure 3-7 Topology of a variable-slip induction generator
(SGCC)
35
36
SECTION 3
Type 3 Doubly fed induction generator

(DFIG): As the most popular WTGs at present,
type 3 WTGs (see Figure 3-8) combine the
advantages of previous WTG designs with
advances in power electronics. The WRIGs rotor
is connected to the grid through a back-to-back
insulated gate bipolar transistor (IGBT) power
converter that controls both the magnitude and
frequency of the rotor current. Up to about 40%
of the power output goes through the converter
to the grid, the rest directly to the grid. This

design can provide approximately 40% up and
down speed variation, maximizing wind energy
capture. The converter provides decoupled
control of active and reactive power, enabling
flexible voltage control without additional
reactive power compensation, as well as fast
voltage recovery and voltage ride-through.
Pitch control is also incorporated.
Wound rotor
induction generator
Transformer
Grid
Gear box
Back-to-Back IGBT
Power converter
Figure 3-8 Topology of a doubly fed induction generator

(SGCC)
Type 4 Full-power conversion WTG: In a

type 4 WTG (see Figure 3-9), the stator of the
generator is connected to the grid via a fullpower back-to-back IGBT power converter,
which means all the power output goes to the
grid through the converter. The generator may
be a synchronous generator with wound rotors,
a permanent magnet generator or a SCIG. The
gear box may be classical (drive-train), a low
speed step-up one (half direct-drive), or there

may even be no gear box (direct drive). A type
4 WTG has similar characteristics to type 3 and,
since it is completely decoupled from the grid,
it can provide an even wider range of speed
variation as well as reactive power and voltage
control capability. In addition, its output current
can be modulated to zero, thereby limiting the
short-circuit current contribution to the grid.
Synchronous
or induction
generator
Back-to-Back IGBT
Power converter
Transformer
Grid
Gear box
Figure 3-9 Topology of a full-power conversion WTG

(SGCC)
3) Offshore wind power generation

Offshore sites generally have better wind
resources than onshore sites, so WTGs
installed in offshore sites can achieve
significantly more full-load hours. Offshore
wind farm development can also relax many
constraints faced by onshore wind farms, such
as transport and land occupation. Todays
offshore WTGs are essentially large onshore
ones with, for example, enhanced corrosion
protection. A specific offshore wind power
industry is developing and a specific offshore
supply chain is emerging [iea11]. A number of
WTG manufacturers have now developed large
WTGs with a capacity of more than 3MW for
the offshore wind power market, and some
have been installed for trials. Future WTGs will
be even larger and specially tailored for offshore
applications. Offshore wind farms will also
be large in scale compared to onshore wind
farms. For instance, several planned projects
in the North Sea and the East China Sea have
capacities of well over 1000MW.
However, a number of issues make the
construction of offshore wind farms more
challenging and costly than onshore wind

farms, such as support structures for the large
turbines, difficult weather conditions, lengthy
and costly transmission cables and extended
transmission systems, high wind speed and
deep water, environmental issues and the
impact on other marine stakeholders [iea11].
In spite of this, offshore wind power is growing
and will grow more rapidly in the future with
improvements in the technology.
Figure 3-10 The Shanghai Donghaidaqiao

offshore wind farm
(SGCC)
37
38
SECTION 3
3.2.2 PV power generation
PV cells are connected in series and in parallel

to one another to form PV modules with
capacities of typically 50W-200W. PV modules,
combined with a set of additional applicationdependent system components (e.g. inverters,
sun trackers, batteries, electrical components
and mounting systems), form a PV system,
which is highly modular, with capacities ranging
from a few watts to tens of megawatts [iea11].
Large utility-scale PV systems are usually called
PV power stations. The structure diagram of a
PV power station is given in Figure 3-11.
1) Basics of PV power generation

Made of semiconductor materials, a photovoltaic
(PV) cell directly converts solar energy into
direct-current electricity. When sunlight shines
on an individual PV cell, the energy that the
cell absorbs from the sunlight is transferred to
electrons in the atoms of the semiconductor
material. These energized electrons then
become a part of the electrical current in the
circuit, generating electricity.
Power grid
High voltage busbar
Step-up
transformer
Lower voltage busbar
Inverter
AC
DC
Inverter
AC
DC
Inverter
AC
DC
Cable
PV panel
Figure 3-11 Structure diagram of a PV power station
2) T
ypes of PV power generation
technologies
attention. Although thin-film PV generally has a

lower efficiency than silicon PV (around 11%),
it is less expensive and less energy-intensive to
manufacture and is also more flexible for versatile
applications. Concentrating PV, in which sunlight
is concentrated and strengthened by a lens
before it reaches the PV cells, is on the edge of
entering full market deployment. Concentrating
PV can reach an efficiency of up to 40%. Other
technologies, such as organic PV cells, are still in
the research phase [hts09] [iea11].
According to the materials and design, current

PV power generation technologies can be
classified into crystalline silicon, thin-film and
concentrating PV (see Figures 3-12 and 3-13).
Crystalline silicon PV is currently the bestestablished PV technology, with an energy
conversion efficiency of up to 20
%. More
recently, thin-film PV, which can also use nonsilicon semiconductor materials, is gaining
(a) A-Si Thin-film PV
(b) Polycrystalline silicon PV
Figure 3-12 Thin-film and silicon PV
Figure 3-13 Concentrating PV

(SGCC)
39
Compared to CSP described below, PV power

generation (except for concentrating PV) has
the advantage that it can use not only direct
sunlight but also the diffuse component of
sunlight to generate electricity, which allows
its effective deployment in many more regions.
Compared to wind power generation, PV power
generation is less challenging for grid integration
because sunlight is more predictable than wind.
Up to now, the comparatively high cost of PV
power generation has been the main barrier to
its mass deployment [iea11].
3.2.3 Concentrated solar power

generation
1) Basics of CSP generation
CSP generation, also known as solar thermal
power generation, is much like conventional
thermal power generation that converts thermal
energy into electricity, but differs in how the
thermal energy is obtained. CSP plants use
various mirror configurations (with a sun tracking
system) to reflect and concentrate direct-beam
n
io
at
di
ra
m
ea
rb
la
so
and flexible active and reactive power control

capabilities. However, since there is no rotating
component, PV systems cannot supply inertia
support to the power system.
ct
One of the key components of PV systems

is the inverter. DC output from PV systems is
changed into AC by inverters. The performance
of the inverter is especially important for gridconnected PV power plants, since it directly
influences whether the PV power plant can
meet the requirements of grid operation. Most
inverters have low voltage ride through (LVRT)
sunlight to heat the working fluid flows (such as

air, water, oil or molten salt) in the receivers to a
high temperature, thus converting solar energy
into thermal energy (see Figure 3-14).
re
3) Characteristics of PV power generation
di
40
SECTION 3
receiver
reflector
Figure 3-14 Principle of obtaining thermal

energy in CSP
[dlr05]
Using a generator driven by a steam turbine, gas

turbine or heat engine, this thermal energy is
then converted into electricity. CSP plants may
also be equipped with thermal energy storage
systems for operating during cloudy periods or
at night. A typical stand-alone CSP plant has a
configuration such as that in Figure 3-15.
Collector Field
Turbine
Generator
Storage
Tank
Grid
Solar Steam
Generator
Condenser
Heat Transfer Fluid

Steam
Figure 3-15 Stand-alone CSP plant

[gre12]
A CSP plant can also be designed as a hybrid

system that uses fossil fuel to supplement the
thermal output during low solar radiation periods
(see Figure 3-16), making the plant output
more stable and dispatchable. It can also be

integrated with a conventional combined-cycle
plant to improve energy efficiency.
Turbine / Generator
Collector Field
Storage
Tank
Solar Steam
Generator
Fuel
Grid
Boiler
HTF
Heater
Fuel
Condenser
Heat Transfer Fluid (HTF)

Steam
Figure 3-16 Hybrid CSP plant

[gre12]
41
42
SECTION 3
(a) Parabolic trough system

(b) Linear Fresnel reflector (LFR)
(c) Solar tower

(d) Parabolic dish
Figure 3-17 Types of CSP technologies

[dlr05]
2) Types of CSP generation technologies

According to how the solar energy is collected,
present CSP technologies can be classified into
four major types [dlr05] [hts09] [iea11]: parabolic
trough, linear Fresnel reflector, solar tower and
parabolic dish systems (see Figure 3-17).
Parabolic trough systems use long
rows of parabolic mirrors to reflect and
concentrate sunlight beams onto a linear
receiver tube that contains the working
fluids
Linear Fresnel reflector systems use

long rows of flat or slightly curved mirrors
to concentrate the sunlight beams onto
a downward-facing linear receiver tube
fixed in space above the mirrors
Solar tower systems, also known as
central receiver systems, use numerous
large, flat mirrors to concentrate sunlight
beams onto a receiver at the top of a
tower
Parabolic dish systems, also known as
dish/engine systems, concentrate sunlight
beams onto the focal point of a single

dish, where the receiver and an engine/
generator (such as a Stirling machine or a
micro-turbine) are installed
air) is an effective alternative, but with higher

costs and lower operating efficiencies.
3.3 Transmission technology

The capacity of parabolic trough, linear Fresnel
reflector and solar tower systems can reach
hundreds of megawatts, while the capacity of
dish/engine systems is generally in the 10kW
range, suitable only for distributed generation.
The solar-to-electricity efficiency of parabolic
trough systems has been demonstrated to be
about 10%-15%, with a perspective of reaching
about 18% in the medium term.
3) Characteristics of CSP generation
CSP is advantageous in that it offers a utilityscale, dispatchable and firm RE option. With the
inherent energy storage capability in the form
of heat that can be used to generate electricity
up to hours later, and with further support from
additional thermal storage systems or a hybrid
system design, CSP plants can continue to
produce stable electricity even when clouds
block the sun or the sun sets. So it is much less
challenging to integrate CSP than wind or PV
generation into power systems.
Although CSP has better performance for grid
integration, the relatively immature technology
and the high cost are currently limiting its largescale deployment.
One disadvantage of CSP is that it requires
strong direct sunlight. As a result, adequate CSP
resources are generally limited to semi-arid, hot
regions or deserts. At the same time, like other
thermal power generation plants, CSP requires
water for cooling and condensation, which
means a tremendous challenge in exploiting
CSP resources in arid regions. Dry cooling (with
Large-capacity RE generation plants are

usually far from load centres, and they therefore
need long-distance power transmission. Up to
now, AC transmission has been used for largecapacity RE power transmission, and voltage
source converter high voltage DC (VSC-HVDC)
transmission has been used for offshore wind
power integration. Examples will be given in
this section. Ultra-high voltage AC (UHVAC) and
current source converter HV/UHVDC (CSCHV/UHVDC) are planned to be used in largecapacity RE power transmission; these will be
described in section 4.
3.3.1 AC transmission
AC transmission is a mature technology.
The capacity of an AC transmission line is
proportional to the square of the voltage level
and inversely proportional to the impedance of
the line, which increases with the transmission
distance. To achieve a large increase in the
transmission capacity of long-distance AC
lines, a natural way is to raise the voltage level.
The approximate transmission capacities and
distances of different voltage-level AC lines
are listed in Table 3-1 [sgc07]. For small-tomedium scale RE power plants, transmission
lines below 330kV are usually used. For largescale, long-distance RE power, transmission
lines above 500kV are usually needed. Some
examples of AC transmission above 500kV
for RE integration in China and the USA are
described below.
43
44
SECTION 3
Table 3-1 Typical transmission capacities and distances of different voltage-level AC lines
Voltage level (kV)
Capacity (MW)
Distance (km)
110
10-50
50-150
220
100-500
100-300
330
200-800
200-600
500
1000-1500
150-850
765
2000-2500
Above 500
the interconnection of the Xinjiang power grid

and the Northwest power grid as well as for the
transmission of the phase I Jiuquan wind power
base. Jiuquan wind power base phase I has an
installed capacity of 5160MW. A small portion
is locally consumed, while most is transmitted
to the load centre of the Northwest power grid
(see Figure 3-18). A second 750kV transmission
corridor is now under construction. 500kV
transmission lines have also been built for the
Jilin wind power base in Northeast China.
1) USA
Currently, three major 500kV transmission
projects are under construction or have been
approved in Southern California for RE power
transmission [ete11], a summary of which is
given in Table 3-2.
2) China
In November 2010, a 2398km double-circuit
750kV transmission line was commissioned for
Table 3-2 Three AC transmission projects in California for RE power transmission
Project
RE integration
capacity provided
by project (MW)
Transmission
line length
(km)
Purpose
Sunrise Power Link
1700
196
Enhance grid connection between

Imperial Valley and San Diego,
transmitting wind, solar and
geothermal power from Imperial Valley
and Arizona to San Diego
Tehachapi
Renewable
Transmission Project
5850
530
Transmitting wind and solar power in

Tehachapi area
270
Transmitting wind and solar power

in the Blythe area between California
and Arizona
Colorado River
Devers Valley (CDV)
4700
Jiuquan wind power base
Ov
er
Xinjiang
10
00
km
Load centre
Figure 3-18 Transmission of the phase I Jiuquan wind power base, Northwest China
(SGCC)
3.3.2 VSC-HVDC transmission

IGBT-based VSC-HVDC differs from the
conventional thyristor-based current source
converter HVDC (CSC-HVDC, also known
as line commutated converter HVDC (LCCHVDC)) in that it is self-commutated via control
circuits driven by pulse-width modulation, while
CSC-HVDC is line-commutated, i.e. switched
off when the thyristor is reversely-biased from
the AC voltage [wei11]. Compared to CSCHVDC, VSC-HVDC offers among others the
following major advantages [vsc10]:
1) It can rapidly control both real and reactive
power, independently, within its rated MVA
capacity. As a result, it can transmit very low
and even zero active power, which is suited to
the frequent and wide-range output fluctuation
of RE generation, while CSC-HVDC is limited by
minimum startup power. VSC-HVDC terminals
can generate or absorb a given amount of
reactive power as instructed or according to
the voltage level of the connected AC grid,
providing excellent voltage support, while CSCHVDC terminals always absorb reactive power
when working, requiring large amounts of
reactive power compensation.
2) It can rapidly reverse the reactive power

direction merely by reversing the current
direction, without changing the voltage polarity
as CSC-HVDC requires. It therefore needs no
changes in the topology and strategy of the
converter station control system.
3) It does not require support from the connected
AC grid for commutation as CSC-HVDC does,
and can therefore be connected to weak or even
passive AC grids, while CSC-HVDC requires the
connected AC grid to be sufficiently strong.
Due to these advantages, VSC-HVDC is desirable
for RE integration, especially for offshore wind
power, as well as for grid interconnection and
power supply to isolated systems and to crowded
metropolitan areas. The fast development of
RE has led to a rapid increase in VSC-HVDC
projects in recent years. Some VSC-HVDC
projects commissioned for RE integration are
listed in Table 3-3. There were also 12 VSCHVDC transmission projects under construction
all over the world at the end of October, 2011.
From 2009 onwards, the total capacity of VSCHVDC projects under construction in the world
has reached 10 GW, which is four times higher
than that of the projects built before 2009.
45
46
SECTION 3
Table 3-3 VSC-HVDC projects commissioned for RE integration

Year
Project
Country
MW / kV / km
Remarks
1999
Gotland
Sweden
50 / 80 / 70
Onshore wind power integration

(voltage support)
2000
Tjreborg
Denmark
7.2 / 9 / 4.3
Onshore wind power integration

(testing for offshore wind power
integration)
2009
Nord E.ON 1
Germany
400 / 150 / 203
Offshore wind power integration
2011
Shanghai Nanhui
China
18 / 30 / 10
Offshore wind power integration
3.4 Operational technologies

and practices
This section focuses on operational technologies
and practices related to wind power, since wind
power generation is currently the most widely
deployed large-capacity RE generation and has
significant impacts on power system operations.
Much has been done to address these impacts
by researchers and grid operators worldwide.
Similar technologies and practices can also be
used for PV and solar thermal power generation.
For the operation of power systems with high
penetration of large-capacity RE generation, RE
power forecasting is critical for grid operators
to carry out operational planning studies and
ensure that adequate resources are available
for managing the variability of RE output.
3.4.1 Power forecasting

1) Forecasting methods
Based on the time scale of the forecast, wind
power forecasting can be classified as ultrashort-term forecasting, short-term forecasting
and medium/long-term forecasting. Short-term

forecasting is currently the most widely used,
with a time scale up to 48-72h. The principle
of short-term wind power forecasting is shown
in Figure 3-19. Present methods for shortterm wind power forecasting generally include
physical methods, statistical methods, and a
hybrid of the two.
Physical methods start with a numerical
weather prediction (NWP) model, which
provides the expected wind speed and
direction at a future point in time. Further
steps include the application of the NWP
model results to the wind farm site, the
conversion of the local wind speed to
power, and the further application of the
forecast to a whole region [ewe05].
Statistical methods first establish the
relationship between the historical NWP
data and the historical power output data
of wind farms via one or more learning
algorithms, and then predict the wind farm
power output based on this relationship.
The advantages and disadvantages of the two
types of method are shown in Table 3-4.
Numerical weather
prediction (NWP)
Forecast program
Power output
forecast
Figure 3-19 Principle of short-term wind power forecasting

(SGCC)
Table 3-4 Advantages and disadvantages of two forecasting methods

Methods
Advantages
Disadvantages
Physical
methods
Require no historical power output

data for the wind farms; suitable for
new wind farms.
Based on detailed analysis of each
atmospheric process; the forecasting
model can be optimized to obtain
more accurate forecasts.
Very sensitive to systematic

errors caused by incorrect initial
information.
Statistical
methods
High accuracy can be achieved if

high-quality data can be obtained.
Self-adjustment can be made to give
appropriate output even if the input is
not included in the training set.
Require a great deal of highly

consistent historical data.
Work like a black box, difficult
to understand the learning and
decision-making process and
optimize the model.
Some of the major short-term wind power

forecasting programs available on the market
are listed in Table 3-5. Although the current
programs are already able to provide valuable

forecast results, further improvement is eagerly
expected [ewe05].
47
48
SECTION 3
Table 3-5 Overview of short-term wind power forecasting programs
Program
Developer
Methods
Country
Operational
since
Prediktor
RISO National Laboratory

(DK)
Physical
Spain, Denmark,
Ireland, Germany, USA
1994
WPPT
IMM, Technical University

of Denmark
Statistical
Denmark (East and

West)
1994
Previento
University of Oldenburg
and Energy & Meteo
System (DE)
Physical
Germany
2002
AWPPS (MoreCare)
Armines/Ecole des Mines

de Paris (F)
Statistical,
Fuzzy-ANN
Ireland, Crete (Greece),

Madeira (Portugal)
1998,2002
RAL (MoreCare)
RAL (UK)
Statistical
Ireland
--
Sipreolico
University of Carlos III,

Madrid; Red Elctrica de
Espaa
Statistical
4 GW, Spain
2002
Local PredRegioPred
CENER (ES)
Physical
Spain
2001
Casandra
Gamesa (ES)
Physical
Spain, Portugal, USA
2003
GH Forecaster
Garrad Hassan (UK)
eWind
Physical and Spain, Ireland, UK, USA,

Statistical
Australia
2004
TrueWind (USA)
Physical and
Statistical
Spain (represented
though Meteosim), USA
1998
HIRPOM
University College
Cork, Ireland; Danish
Meteorological Institute
Statistical
Under development
--
AWPT
ISET (DE)
Statistical,
ANN
15 GW, Germany
2001
AleaWind
Aleasoft (ES)
Statistical
Spain
2004
Scirocco
Aeolis (NL)
Physical
Netherlands, Spain
2004
Meteologica
MBB
Physical
Spain
2004
Meteotemp
No specific model name
Physical
Spain
2004
WPFS
CEPRI
China
2009
Physical and
Statistical
2) Forecast accuracy
The accuracy of wind power forecasting can

be measured by different indices. Root mean
square error (RMSE) normalized to the installed
wind power is most commonly used. Regardless
of the forecasting method used, the forecast
error (RMSE) for a single wind farm is between
10% and 20% of the installed wind power
capacity for a horizon of 36 hours [ewe05].
Error reduction
0.8
0.6
0.4
0.2
0
500
1 000
1 500
2 000
Region size (km)
Figure 3-20 Decrease of wind forecast error

for aggregated wind power production
Spatial aggregation greatly reduces forecast

errors, just as it reduces variability. As seen in
Figure3-20, the level of accuracy improves as
the size of the region considered for forecasting
increases. This is due to spatial smoothing
effects. The error reduction shown along the
vertical axis in Figure3-20 is the ratio between
the RMSE of a regional prediction and the RMSE
of forecasting for a single site, and is based on the
measured power production of 40 wind farms in
Germany. Aggregation over a 750km region may
reduce forecasting error by about 50% [eis11]. If
forecasting is scaled up to apply to the aggregated
[hol09] (energy & meteo systems)
wind power of a large area, the forecast error

(RMSE) can drop to below 10% [ewe05].
On the other hand, the forecast error increases
as the time horizon of the forecast increases,
as shown in Figure 3-21, which also shows the
improvement in forecasting techniques over the
years. Generally, the error for 1-2 hours ahead
for a single farm forecast can be about 5%-7%,
and can reach 20% for day-ahead forecasts.
Mean absolute error/mean production (%)
WIND FORECAST EVOLUTION 2005-2008 ( DATA FROM R ED ELECTRICA ESPANA)
40
35
30
25
20
15
2005
2006
2007
2008
10
5
0
0 1
10
13
16
19
22
25
28
31
34
37
40
43
46
49
Hours
Figure 3-21 Wind forecast error as a percentage of wind production, as a function

of the time forward from the present
[uei10]
49
50
SECTION 3
3.4.2 Operational practices
The requirement is also emphasized for

WPPs to be equipped with control and
monitoring systems; these serve to enhance
their controllability and provide operational
information to grid operators. To date, all
grid-connected WPPs have been equipped
with control and monitoring systems that can
communicate with the dispatching centres in
real time.
Due to the differences in RE generation

development, the conventional generation fleet,
grid structure as well as market and institutional
environments, the operational practices
related to RE integration are quite different in
different countries or regions. Some operational
practices related to wind power integration in
major countries are briefly introduced below.
Wind power forecasting systems have been

deployed in several regional or provincial
dispatching centres in the State Grid Corporation
of China (SGCC), but only about 30 WPPs have
established a plant-level wind power forecasting
system, a situation needing to be improved by
more stringent management. A SGCC NWP
centre has also been established in the China
Electric Power Research Institute (CEPRI) to
provide weather information needed by the
wind power forecasting systems. Based on
wind power forecasting at different time scales,
a wind power optimal dispatching decision
support system had been developed and put
into operation in the dispatching centre of the
Jilin provincial power grid (see Figure 3-23).
1) China
According to the Renewable Energy Law and
related regulations, wind power (and other
renewables) should be given first priority in
generation scheduling and dispatching under
normal power system operating conditions. To
meet this requirement, grid operators consider
predicted wind power generation in the mid-tolong term, day-ahead and intra-day operation
planning processes and fully exploit flexibility
from conventional power plants, as well as the
capacity of inter-grid tie-lines to accommodate
the maximum wind power while maintaining
system security and reliability (see Figure 3-22)
[sgc11] [wei11].
Wind power forecast

Mid-long term
Forecast
Wind/Conventional power plant dispatch

Entire grid
Day-ahead
forecast
Entire grid
Intra-day
forecast
Entire grid
and WPP
and WPP
Wind energy
scheduling
Operation planning,
Unit commitment
Day-ahead
scheduling of
WPPs
Day-ahead scheduling
of conventional plants
Real-time
control of WPPs
Real-time control of
conventional plants
Figure 3-22 System operation in China taking wind power into account
[sgc11]
AVC
AGC
WP information
input
Day-head & real

time WP plan
WP forecasting
WPP control
D5000 dispatching platform
Security analysis
regarding WP
Analysis and display

of WP operation data
Figure 3-23 The wind power optimal dispatching decision support system in Jilin
[sgc11]
2) Denmark
Energinet.dk, the transmission system operator
(TSO) of Denmark, mainly uses two wind power
forecast tools: one external and one internal. The
external forecast tool gives a 0-12 hour forecast
every 5 minutes and a 0-48 hour forecast every
hour. The input to the forecasting includes four
different numerical weather predictions. The
internal forecast tool includes a 0-6 hour shortterm forecast and a 12-36 hour day-ahead
forecast.
A well-functioning power market and an
automatic dispatching system have been
established throughout the north European
power system including Denmark, through
which the hydropower in Norway and Sweden
is often used to balance the wind power in

Denmark. Figure 3-24 shows the monitoring
curves of power generation and exchange,
where line A is total power production based
on individual 5-minute schedules from all
power plants, lines B are exchange plans with
neighbouring grids (Western part of Denmark
(DK1) to SvK in Sweden, DK1 to Tennet in
Germany, DK1 to ENDK in Denmark and DK1 to
SN in Norway, from top to bottom). In general,
the Nordic Power System can be balanced by
trading: the additional electricity produced by
WPPs is purchased at a low price and sold to
the other grids at a higher price. In addition
co-generation plants and electric boilers play
an important role in winter in balancing the
fluctuations caused by large-scale wind power.
51
52
SECTION 3
Figure 3-24 Monitoring power generation and power exchange with neighbouring grids
[sad11]
3) Germany
Three of the four TSOs in Germany (50Hertz,
Tennet, Amprion and ENBW) have established
wind power forecasting systems, covering
approximately 98
% of Germanys wind
power capacity. For example, the Amprion
company has set up a Front Office to manage
the operation of the forecasting system and
optimize the forecast results by a combination
tool (see Figure 3-25). Amprion shares wind
power forecasting data (day-ahead and ultrashort-term) with other members of TSC (TSO
Security Cooperation).
In September 2004, the German Renewable
Energy Sources Act amendment introduced
a new mechanism for wind power balancing,
which requires each system operator to
contribute to balancing the whole country's
wind power output in proportion to the size of
its regional grid.
This mechanism allocates wind power and the

associated fluctuation to each system operator
in real time. It is more equitable in the distribution
of balance services and related costs. The four
system operators have developed a real-time
wind power monitoring system to determine
the wind power balancing capacity that every
system operator should be responsible for.
In fact, as concerns wind power, the regional
grids have integrated into a single large grid in
Germany.
In certain areas, the wind power output will
sometimes exceed the transmission capacity of
the grid, and the grid managers have a legal
obligation to increase power transmission
capacity. In the interim period before the
additional capacity objectives are achieved, the
grid managers can require the wind farms to
reduce their power output or shut down some
of the wind turbines.
Wind power Prediction - Combination tool
Weighting of the forecats

and combination
Classification of the
weather situation
Optimal forecast of wind power

for specific weather situation
Figure 3-25 Schematic of a combination tool for wind power forecast

(Amprion Company)
4) Japan
Japans power system consists of two parts, the
Western 60Hz network and the Eastern 50Hz
network, which are interconnected through
frequency converter stations. The Eastern
network consists of three utilities: the Hokkaido
Electric Power Company (EP), Tohoku EP
and Tokyo Electric Power Company (TEPCO).
Hokkaido EP is interconnected with Tohoku EP
through DC submarine cables, and Tohoku EP
and TEPCO are connected by a 500kV doublecircuit AC transmission line (see Figure 3-26).
The areas suitable for wind power generation
are limited by geographical factors, and they are
unfortunately concentrated in areas whose system
capacities are comparatively small. Most wind
resources are located in the smaller Hokkaido
and Tohoku EP areas, where peak demand is
around 5GW and 15GW respectively. In the
area of TEPCO, which has an electricity demand
of 60GW including the Tokyo metropolitan area,
there are few suitable locations for wind power.
This situation poses a challenge for wind power
integration, since the amount of RE that can be

integrated into a grid depends on the capacity of
the network where the RE is located.
To enhance wind power integration, one option
is to use inter-ties between utilities efficiently
to balance wind power generation output. In
2010, collaboration schemes were launched by
the three utilities for timely sharing of the wind
power forecasts in the Hokkaido and Tohoku
EP areas, and closely monitoring the power flow
of all the inter-ties from Hokkaido EP to TEPCO.
If there is room for extra power flow in the interties and other grid conditions permit, wind
power generated in the Hokkaido and Tohoku
EP areas can be transmitted to TEPCO, with
the output of the thermal power plants being
reduced accordingly to absorb the incoming
wind power. These collaboration schemes will
improve the utilities capability to accommodate
wind power, but the capability is limited by the
capacity of the inter-ties. When the capacity of
the inter-ties is insufficient, excess wind power
generation has to be curtailed.
53
54
SECTION 3
Figure 3-26 Three utilities in the eastern half of Japan
5) Spain
Red Elctrica de Espaa (REE), the only grid
company in Spain, has established the world's
first RE power control centre (CECRE in
Spanish), which is responsible for scheduling
and controlling national RE generation.
Wind farms are also required to set up real-time
control centres, and the control centres of the
wind farms with an installed capacity greater
than 10MW must connect directly to CECRE,
as shown in Figure 3-27. These control centres
must be able to control wind power output
according to setting values issued by CECRE
within 15 minutes at any time. For power system
security reasons, if necessary, REE has the
right to reduce the wind power output.
CECRE carries out three-phase permanent

fault simulation on the 70 different substations
buses every 20 minutes, using an assessment
programme named GEMAS (Maximum
Admissible Wind Power Generation System).
On one hand, it can determine the maximum
RE which the entire system can accommodate
while still guaranteeing system security. On the
other hand, using an optimization method, it
can calculate the maximum output of each wind
farm. The resulting setting thus calculated is
sent by CECRE to every control centre. Figure
3-28 shows typical scheduled and actual wind
power output curves in a day.
VSAT
HUB
Point
to
point
CECRE
Tele
control
Wind Farm
Control
center
ADSL
Special channel
Figure 3-27 Interconnection between wind farms and CECRE

(REE)
18:08 set point

change 2311
19:07 End
18:52 new set

point 1223
18:17 new set

point 118
18:27 new set

point 812
Figure 3-28 Scheduled and actual daily curves of wind power output
6) USA
Due to different grid composition, rules and wind
power penetration, the various regional grids of
the USA have developed widely varying wind
power scheduling, dispatch and operational
mechanisms. For example, the California
Independent System Operator (CAISO), as
a leader in state-of-the-art mechanisms,
has developed a Participating Intermittent
Resource Program (PIRP) that allows individual
wind facilities to self-schedule according to

shared forecasting technologies. The New York
Independent System Operator (NYISO) requires
wind farms to behave like conventional power
in order to participate in real-time electricity
markets: in certain circumstances, the wind
farm must reduce output power or be fined if
it exceeds the value specified in scheduling
instructions.
55
56
SECTION 3
In the western portions of the USA outside

CAISO, which follow reliability standards
set by the Western Electricity Coordinating
Council (WECC), an even greater degree
of diversity is present. Because of a lack of
organized, open electricity markets, the WECC
is primarily managed by vertically-integrated
investor-owned utilities (IOUs), federally owned
power marketing administrations (PMAs),
and municipal and rural utilities that primarily
purchase power at wholesale to serve local
loads. Each utility has its own scheduling
and dispatch procedures according to
the peculiarities of its generation mix, and
energy trading between utilities is difficult
and dominated by long-term power-purchase
agreements. Nevertheless, many utilities
integrate renewables into day-ahead, hourahead, and real-time scheduling procedures,

much like the open markets of the Eastern
Interconnect and the CAISO. Rule changes are
underway in the WECC, and utilities with higher
proportions of renewable power are pressing
for more robust energy imbalance markets in
order to ease RE integration costs. Though it
is the least-organized of US power markets,
the WECC is of critical importance to the USA
due to its high resource concentrations of RE
and the presence of very large federal PMAs
such as the Western Area Power Administration
(WAPA) and the Bonneville Power Authority
(BPA), which are under the direction of the US
Department of Energy.
SECTION 4
Future: technical solutions
for integrating more
large-capacity RE
58
S E C T I O N 4
more large -capacity RE
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
4.1 General
RE generation on the power system, enabling it

to contribute to system reliability and stability by
improving its design and control technologies.
However, the amount of flexibility which can be
achieved by this approach is limited. Flexibility
from conventional generation is currently the
major source of power system flexibility, and is
generally referred to as generation flexibility.
As we have seen, integrating more largecapacity RE into the grid brings variability and
uncertainty. At the same time, there will continue
to be unexpected disturbances stemming from
load variation, grid faults and conventional
generation outages. Worldwide studies and
experience in recent years have shown that new
technical solutions are needed to address this
conjunction of difficulties. The new solutions
will include new technologies, methods and
practices, applied in order to provide more
flexibility and improve the efficiency of power
systems, constantly balancing generation and
load. Only this will make the power systems
reliable and maintain security of supply, i.e.
avoid any interruption in the supply of power.
Flexibility can also be achieved fromthe load

side through demand response, and from
energy storagethat can act as either generation
or load (see Figure 4-1). In addition to adding
new flexibility, existing flexibility can be better
exploited by operational enhancementwithin
a balancing area, and can be shared in wider
geographic footprints by cooperation between,
or consolidation of, smaller balancing areas,
supported by transmission expansion.
The required power system flexibility can be

achieved on the generation side, from both
RE generation and conventional generation. It
should first be pursued using grid-friendly RE
generation. This means mitigating the impacts of
The application of energy storage to support

large-capacity RE integration will be discussed
separately in section 5, while the other solutions
mentioned will be discussed in this section.
EES
RE output fluctuation
Demand response
Grid faults
Grid -friendly RE generation
Conventional generation outages
Conventional generation
flexibility
Load variation
System
flexibility
Variability and
uncertainty
Figure 4-1 System flexibility vs. variability and uncertainty

(SGCC)
4.2 Grid-friendly RE generation

4.2.1 Need for grid-friendly RE
generation
At the beginning of its development RE
generation technology focused more on tapping
the maximum power from RE resources. It
neglected to make any contribution to power
system reliability and stability and in the absence
of standards and incentives was not designed
to operate in a coordinated fashion with the rest
of the system. As long as RE penetration is low
this is manageable and can be accepted by
power system operators. But as RE generation
penetration grows, and especially as the
capacity of RE power plants becomes larger
and larger, this will have a serious impact on
system operation.
For example, in 2011 several serious wind
turbine disconnection incidents occurred
in China due to the lack of fault ride-through
capability (see 4.2.2), including one at the
Jiuquan wind power base on February 24,
which led to the disconnection of 598 wind
turbines from the grid (840.43MW of power
output) and a significant frequency drop.
Therefore it is becoming increasingly important
that RE generation should play a greater role
in helping to maintain system reliability and
stability, and this may be increasingly required
by interconnection standards. Technologies
have been developed and are continuously
improving at the generating unit, plant and
plant cluster level to make RE generation more
predictable, controllable and dispatchable, or in
other words more grid-friendly.
4.2.2 Advanced characteristics of RE

generating units and plants
Development of power electronics and
mechanical engineering technologies, as well
as the design of proper control strategies,
have enabled wind generating units to possess
performance comparable or even superior
to those of conventional thermal or hydro
generating units. Some advanced operational
capabilities of wind generating units and the
methods to achieve them are cited below from
[ner09] and [bac11] with minor modifications.
Most of these capabilities can also be achieved
for solar PV generating units since they share
many technical characteristics with wind turbine
generators, especially the inverter-based ones.
1) Voltage/Var control and regulation
Reactive power support and power factor control
can be provided either through a built-in capability
(available for wind generating units types 3 and 4)
or through a combination of switched capacitor
banks and power electronic based transmission
technologies such as static var compensator
(SVC) and static synchronous compensator
(STATCOM) (flexible AC transmission system
(FACTS) equipment as discussed in 4.4.2,
applicable to all types of wind generating units).
2) Fault ride-through
The ability is needed to survive (ride through)
specific low and high voltage/frequency ranges
and durations caused by faults or disturbances
in the power system. Voltage ride-through can be
achieved with all modern wind generating units,
mainly through modifications to the controls. In
some cases, with older type 1 or type 2 wind
59
60
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
6) Short-circuit current control
generating units at weak short-circuit nodes in

the transmission system, there may be a need
for additional transmission equipment.
All inverter-based variable generators have a builtin capability to limit the fault current to a level that
does not exceed 150% of the full load current.
3) A
ctive power control, ramping and
curtailment
For grid integration it is important to view RE

generation at the plant level. An RE power plant
is not just a simple collection of RE generating
units, but is also supported by many other
components and systems so as to function like
a conventional power plant. Based on advanced
generating units, these characteristics can also
be achieved at the plant level. Plant-level reactive
power compensation, accurate RE generation
output forecasting, the presence of monitoring,
control and data communication systems, as well
as properly designed relay protection schemes all
help to improve the predictability, controllability
and dispatchability of RE power plants. An
example of the typical structure of a grid-friendly
wind power plant is given in Figure 4-2 [sge11].
This can be achieved through unit control

mechanisms for wind turbine units with activestall or pitch control, or discrete tripping of units.
4) Primary frequency regulation
Primary frequency regulation can be supplied
by all units that are equipped with some form of
pitch regulation (i.e. active-stall or pitch control).
5) Inertial response
Inertial response is inherent in type 1 and type 2
units, and can be achieved through supplemental
controls in the converter to emulate inertial
behaviour for type 3 and type 4 units.
Power grid
Dispatch Centre
High voltage busbar
SCADA
Step-up
transformer
Wind farm controller
Reactive power /
Voltage control
Low voltage busbar

Communication
module
Active power /
Frequency control
Fault ride-through
control
Other control
Reactive
power
compensation
Command, Data
Measurement
Power
Communication
module
Figure 4-2 Typical structure of a grid-friendly wind power plant

(SGCC)
4.2.3 Centralized control of an RE plant

cluster
power control as an example [pei12]: with

multiple wind plants having requirements to
provide reactive power and control grid voltages,
problems have emerged when the voltage and
reactive power controllers in nearby plants have
been incompatible, resulting in poor voltage
control and counterproductive reactive power
flows. There have been numerous reports of
one or more plants at maximum reactive power
output while neighbouring plants are absorbing
reactive power. To avoid this kind of situation,
centralized cluster control is an attractive solution.
A centralized cluster control system, configured
in a multi-layer structure such as that depicted in
Figure 4-4, can be used to coordinate the active
and reactive power control of a RE power plant
cluster, just as cascaded hydro power plants on
a river are controlled. Such a centralized cluster
control system, the so-called Large Wind Power
Plants Cluster Active Power Intelligent Control
System, has been developed and deployed at
the Jiuquan Wind Power Base [sgc11].
The areas with the best RE resources may see

the development of RE power bases with many
RE power plants located adjacent to each other.
A schematic diagram of the Jiuquan Wind Power
Base in Northwest China is given in Figure 4-3
as an example. It is organized in several layers:
arrays of wind turbine generators form a wind
power plant, several plants are connected to
a 330kV collection substation through 35kV
lines, and several collection substations are
further connected to a final 750kV integration
substation through 330kV transmission lines
of tens of kilometres. In other words, a largecapacity RE power base may consist of several
RE plant clusters.
The control of the RE power plants in such
a cluster must be coordinated if operational
problems are to be avoided. Taking reactive
200 MW
200 MW
200 MW
200 MW
35 kV line
200 MW
330 kV line
#4
200 MW
750 kV line
300 MW
200 MW
#3
Wind power plant
#5
200MW
200 MW
200 MW
200 MW
#2
750kV
Jiuquan
substation
200 MW
#6
#1
#7
330 kV
substation
200 MW
750 kV Anxi
substation
#10
#8
200 MW
200 MW
#9
200 MW
280 km
Figure 4-3 Schematic diagram of the Jiuquan Wind Power Base, Gansu, China
(SGCC)
61
62
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
Dispatching centre
Cluster control
station #1
Cluster control
substation #1
Plant
executive
station #1
Plant
executive
station #N
Cluster control
station #2
Cluster control
substation #N
Plant
executive
station #1
Plant
executive
station #N
Cluster control
substation #1
Plant
executive
station #1
Plant
executive
station #N
Cluster control
substation #N
Plant
executive
station #1
Plant
executive
station #N
Figure 4-4 RE power plant cluster control system structure

(SGCC)
4.2.4 Improvements in modelling RE

generation
In order to represent accurately the static,
dynamic as well as short-circuit performance
of RE generators, plants and clusters and their
impacts on power systems, appropriate models
of the different types of RE generation are
needed.
Up to now, most RE generation modelling efforts
have been carried out at the generator level
[aba11] [bac11] [mma11]. Although RE generator
manufacturers have long been developing and
refining models to improve their generator
design these are typically user-written
proprietary models on commercial software
platforms it has now been recognized that
publicly available industry-standard models,
similar to those available for conventional
generation, are needed for planning studies.

Significant progress has been made towards
industry-standard models by the task forces or
working groups under WECC, the Institute of
Electrical and Electronics Engineers (IEEE), the
Utility Wing Integration Group (UWIG) and the
North American Electric Reliability Corporation
(NERC), with the development and distribution
of generic dynamic models for wind turbine
generators. These are models with general
model structures that allow for representation
of wind turbine generators of the same type,
regardless of manufacturer, by selecting
appropriate model parameters. WECC and
IEC groups are moving forwards to refine
and validate these generic models, for which
collaboration between the power engineering
community and wind turbine manufacturers is
necessary. Efforts to develop generic models
for PV have also begun. However, regarding RE
generator modelling, a great deal of research

still needs to be carried out, e.g. to determine
the short-circuit contribution of different types
of RE generators, and to represent the more
advanced and rapidly developing control
characteristics of RE generation, such as those
described in 4.2.2.
IEEE and IEC groups have started to expand
efforts in generic dynamic modelling for
entire wind power plants, including control
and auxiliary equipment [mfc11]. Modelling of
large-scale wind power plants is particularly
challenging due to the fact that they may
consist of tens or even hundreds of individual
generators, distributed over the large footprints
of the plants, and the requirement of being
represented as a single equivalent unit in power
system simulations for simplicity. Techniques
are needed to aggregate the individual
generator models while considering the spatial
and temporal interaction among the generators.
Techniques are also needed to represent all the
cables connecting all the generators as a single
equivalent circuit, and to test the validity of the
equivalent model as a whole. Extending this
modelling work to huge clusters of wind power
plants, such as the 10GW-scale wind power
bases in China, will be even more difficult.
4.3 Improved flexibility in

conventional generation
4.3.1 Need for more flexibility in
conventional generation
Although system flexibility can also be achieved
through demand response and energy storage
as well as through grid-friendly RE generation,
it may take decades for low-cost, large-scale
storage to mature (with the exception of the

widely used but topographically restricted
pumped hydro storage), and demand response
as a flexibility resource is still in its infancy.
Flexibility in conventional power plants (generally
referred to as generation flexibility) is therefore
the major source of power system flexibility, and
the need for increased generation flexibility is
central to the challenges posed by intermittent
renewables [eis11]. The need for generation
flexibility differs in different countries and regions
due to their different existing generation mix.
Generally, hydro power plants with a reservoir
are most flexible, followed by gas-fired power
plants, then coal-fired power plants and
last nuclear power plants, but it is difficult to
measure flexibility so as to allow use of the
measurements in generation planning.
4.3.2 Assessment of generation flexibility

1) Flexibility of conventional power plants
The relevant technical characteristics of
conventional power plants that contribute to
operational flexibility are the cycling, ramping and
partial loading capabilities. Cycling capability
is a generators ability for frequent and fast
startup or shutdown. Ramping capability is the
rate at which a generator can change its output,
either upwards or downwards. Partial loading
capability is the minimum output level at which
a generator can be stably operated, as well as
the efficiency when operating at different output
levels lower than rated capacity. Due to the
differences in design and technology, different
types of power plants have different levels of
operational flexibility, implying varying impacts
on equipment health, service life, operational
costs and GHG emissions.
63
64
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
Hydro power plants with reservoirs are very

flexible in operation. They can be easily and
quickly started up or shut down with little impact
on equipment health or service life. They can start
from off-duty status and ramp to full-loading status
in several minutes, and change their capacity
by 50% in fractions of a minute. Their minimum
output can be very low, which means a regulation
range of near 100% capacity. However, output
and hence the flexibility contribution of hydro
power plants can be influenced by seasonal
factors (drought and flooding seasons) as well
as non-electricity-sector requirements such as
flood and drought control or environmental and
navigational considerations.
Gas-fired power plants, including natural
gas combined cycle (NGCC) and simple cycle
gas turbine (SCGT) plants, are also flexible
generation resources. Current designs of
SCGTs in operation are able to ramp to 100MW150MW in 10 minutes, and NGCC plants in
60 to 80 minutes. New NGCC designs with an
increased focus on the ability to operate in a
system with a large proportion of intermittent
renewables are expected to produce 150MW

in 10 minutes and to ramp to full load in 30
minutes. The ramping rate of NGCC and SCGT
plants is generally up to 8% per minute. NGCC
plants can reduce their output to 80
% of
capacity with only a minimal heat rate penalty,
but with increasing efficiency losses at lower
outputs [eis11].
Coal-fired power plants have limited
operational flexibility. They are not suitable for
frequent cycling since they may take several
hours to start up or shut down. Their ramping
rates are generally only 1%-3% per minute;
new large-capacity generators can reach 5%
per minute. Their minimum output is generally
70%-80% of capacity; new large-capacity or
specially designed peaking generators can go
down to 50% or lower, but lower output will lead
to higher fuel consumption per kWh generated.
An example is given in Table 4-1 [npg10]. The
flexibility of specific coal-fired power plants
can be influenced by their size, age, operating
pressure, fuel quality and the control systems
installed [eis11].
Table 4-1 Standard coal consumption of typical 600MW and 1000MW generators
at different loading levels
Loading level
(% of rated capacity)
Standard coal consumption (g/kWh)

1000MW ultra-supercritical
generators
600MW ultra-supercritical
generators
40%
306
311
50%
298
303
75%
287
290
100%
282
284
Nuclear power plants also have limited

operational flexibility and are less used for ramping
for economic reasons. Relatively new nuclear
reactors can down-ramp 20% of total output within
an hour, but require six to eight hours to ramp
up to full load. Nuclear plant ramping operations
are not fully automated, creating opportunities
for operator error. Although operational costs for
nuclear plants running at partial load versus full
load do not significantly differ, they need to run
as base load units at high output levels to recover
their high capital costs [eis11].
With more large-capacity RE generation being
integrated into power systems, not only the
flexible hydro and gas-fired power plants are
required to adjust their output more frequently,
the coal-fired and nuclear power plants
traditionally operated mainly as base load units
are also shifting to more flexible operation
paradigms. Much effort is needed to better
understand the impact of flexible operation
on the technical, economic and environmental
performance of conventional power plants,
especially thermal power plants, hence to
support the improvement in equipment design,
retrofit and daily operation, as well as the design
of necessary financial incentives.
2) Measuring generation flexibility
The need for more generation flexibility is
well recognized. However, there is currently
a lack of metrics, methods and tools to
measure the flexibility and its related cost of
provision, determine the flexibility required,
and optimize the generation resources to
provide the required flexibility with minimal
cost, all of which would be needed to support
the planning and operation of power systems
with more large-capacity RE. It is much more
difficult to measure generation flexibility than to
measure the adequacy of generation capacity;

the former is more complicated and requires
more data and detailed modelling [trp11].
Collaboration between grid operators and
the research community is desirable to foster
efforts to develop more rigorous methods and
tools to accurately quantify, monitor and assess
the degree of flexibility in a given power system
for a given level of RE penetration [sad11].
The research community has started to
respond to the need for power system flexibility
measurement. The emphasis is on generation
flexibility, currently the major flexibility resource, but
efforts also include using demand response and
energy storage for flexibility. The Integration of
Variable Generation Task Force (IVGTF) under the
NERC in 2010 released a special report on the
flexibility requirements and metrics for variable
generation [ner10], which discussed the metrics
and approaches for flexibility measurement. The
IEA Grid Integration of Variable Renewables
(GIVAR) project has proposed a semi-quantitative
method supported by qualitatively described
limiting constraints on flexible resources to
assess system flexibility, and a subsequent
project specially targeting the quantification of
system flexibility is expected [fat11]. The Alberta
Electric System Operator (AESO) has developed
a dispatch decision support tool, incorporating
wind power forecasting, to measure system
flexibility and help operators to manage the
increased variability and uncertainty [pao11].
4.3.3 Generation planning for both

adequate capacity and adequate
flexibility
In power systems with higher variability and
uncertainty due to the integration of more
65
66
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
large-capacity RE, it is becoming increasingly

important for the generation planning process
to consider not only the adequacy of generation
capacity, but also the adequacy of generation
flexibility. Simply having a specified number of
megawatts of capacity may not be adequate
for system security if that capacity is not flexible
enough to respond to system variability [aba11].
of other types of generation in the calculation,

the accuracy of approximate methods and the
requirement of high-quality wind, solar and
synchronized demand data for the ELCC method
to produce reliable results [bac11] [cvw11].
2) G
eneration planning considering
flexibility at the system level
In order to consider flexibility requirements, a
new paradigm of generation planning is needed.
One example is the approach proposed in
[trp11], adding a flexibility assessment stage into
the planning process, as shown in Figure 4-5.
1) P
lanning for adequate generation
capacity
For capacity planning, the calculation of the
capacity value (also known as capacity credit)
of RE generation has been recognized as a major
issue. The capacity value of a generation resource
can be defined as the amount of additional load
that can be served due to its addition while
maintaining existing levels of reliability. The
calculation of the capacity value of RE generation
is much more difficult than that of conventional
generation due to its inherent variability and
uncertainty. Research on this issue has received
much attention. Both the IEEE Power & Energy
Society (PES) and the NERC IVGTF have
established task forces to study it. A rigourous
effective load-carrying capacity (ELCC) method
based on reliability simulation as well as several
approximate methods have been proposed for
wind power capacity value calculation, and are
being extended to solar power. But there remain a
number of problems, including the representation
Capacity
adequacy
assessment
It starts from the relatively simple traditional

capacity adequacy calculation, using metrics
such as loss of load expectation (LOLE) and
expected unserved energy (EUE), to form an
initial generation portfolio, and then moves to the
more data-intensive system flexibility assessment
and production cost simulation stages. Both of
these stages involve computation-intensive unit
commitment and economic dispatch studies.
This three-stage approach is designed to reflect
the aims of the planning process, i.e. to provide a
reliable, operable, least-cost generation portfolio
for the given study period. If excessive or
inadequate flexibility is observed, the generation
portfolio or unit commitment model can be
altered and the whole process iterated, in order
to find a better generation portfolio that meets
System flexibility
assessment
Production cost
simulation
Generation
portfolio
If portfolio is unreliable in operation
Figure 4-5 Generation planning process incorporating flexibility assessment

[trp11]
the dual aims of minimizing cost and meeting a

specified flexibility standard.
Again, the system flexibility assessment is
possible only when metrics measuring flexibility
in the planning context are available, which
is currently not the case. A good start is the
metric named insufficient ramping resource
expectation (IRRE) developed by the IEA, which
applies principles similar to those underlying the
LOLE metric for capacity adequacy calculation.
3) G
eneration planning considering
flexibility at resource site level
Besides incorporating flexibility considerations
into generation planning at the system level,
special designs can also be used at resource
site level to exploit the flexibility of conventional
generation in order to compensate for the
variability and uncertainty of RE.
Such an approach is intended for some wind
power bases in China, where the so-called
wind-fire bundling design has been proposed
(see Figures 4-9 and 4-10): coal-fired power
plants are planned near a wind power base
and will share its transmission lines. Wind
power output fluctuations can be smoothed
out by adjusting coal-fired production; thus
the utilization of the transmission line can be
improved as compared to transmitting only
wind power and the average cost of electricity
at the receiving end can be reduced [sgc11].
4.4 Transmission expansion

4.4.1 Needs for transmission expansion
Accommodation
of
large-capacity
RE
generation needs large-scale transmission
grid expansion and reinforcement for the

following reasons:
1) Grid expansion and inter-regional connection
are needed to transmit the energy generated by
large-capacity RE sources, which are generally
located far from load centres and the existing
grid.
2) Through grid expansion, the geographic
diversity of RE generation can be exploited
to smooth out their aggregated variability
and uncertainty and to reduce the RE power
forecast error.
3) Grid expansion and reinforcement can
support interconnection between balancing
areas, hence facilitating their cooperation or
consolidation to share flexibility resources.
As shown in [smp11], transmission for RE
integration is an issue everywhere for both
technical and regulatory reasons. An inability to
solve the transmission problem will jeopardize
the achievement of RE goals, and drive up the
cost to reliably integrate these new resources
into power systems.
4.4.2 Application of new transmission

technologies
1) H
igher voltage level AC transmission:
UHVAC
UHVAC transmission lines with rated voltage
levels of 1150kV or 1000kV were built and
commissioned by the former Soviet Union and
Japan in the 1980s and 1990s, but then operated
at a 500kV voltage level for practical reasons.
China is now leading the research and application
of 1000kV UHVAC transmission, which seems a
desirable technology to meet the need for largescale, long-distance power transmission from the
67
68
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
large coal, hydro, wind and solar energy bases in

the northern and western regions to the central and
eastern regions with huge and still fast-growing
electricity demand. The first 640km single-circuit
Jindongnan-Nanyang-Jingmen 1000kV UHVAC
test and demonstration transmission project was
commissioned in 2009, the reinforcement which
followed was completed at the end of 2011, and
it can now transmit 5000MW of power stably.
Some UHVAC projects have been planned to form
part of the large-capacity, long-distance wind
power transmission installation from the WestInner Mongolia and Hebei wind power base in
Northern China [sgc11]. UHVAC has also attracted
attention from other emerging large economies
facing similar energy delivery challenges, such as
India [tea07].
A single-circuit 1000kV AC line can transmit
4000MW to 5000MW over an economic
distance of 1000km to 1500km. Compared
to 500kV AC transmission, 1000kV AC
transmission has many advantages in improving
transmission capacity and distance, reducing
power loss, reducing land use and saving cost,
as shown in Table 4-2 [sin06] [tgy06]. For RE
energy transmission, UHVAC is mainly suitable
for transmitting power from on-shore RE plants
using overhead lines.
2) More flexible AC transmission: FACTS
Based on advanced power electronic technologies
and innovative designs, FACTS equipment can
be applied to improve the capacity, stability and

flexibility of AC transmission, making it more
capable of transmitting large-capacity RE. For
example, thyristor controlled series compensators
(TCSCs) can be installed in transmission lines to
reduce electrical distance, increase damping and
mitigate system oscillation; SVC, STATCOM and
controllable shunt reactors (CSRs) can be shuntinstalled on substation buses to solve the reactive
power compensation and voltage control problems
which are common in RE integration due to their
output fluctuation. SVCs or STATCOMs may
also be used to improve the performance of RE
power plants to meet integration requirements on
reactive power and voltage control, while keeping
the design of RE generators relatively simple.
It was noticed when studying the integration of
the Jiuquan wind power base in Northwest China
that the large wind power output fluctuations
would cause large voltage fluctuations in the
whole network, especially at the nodes in the
transmission corridor. To stabilize voltage, many
voltage-regulating devices must act frequently
and in a well-coordinated way, which is harmful
for the equipment and difficult to implement. In
order to solve this problem, a 750kV CSR was
installed on the Dunhuang substations 750kV
bus (see Figure 4-6), with field testing completed
at the end of 2011. Developed by CEPRI, it has a
capacity of 300Mvar and 4 controllable grades:
100%, 75%, 50% and 25%; the adjustment
range is 25%-100%.
Table 4-2 Comparison between 1000kV and 500kV AC transmission: gain factors
Item
Typical
capacity
Economic
distance
1000kV/500kV
4~5
1.5~2
Power Loss
Corridor width
per kW
Comprehensive
construction
cost per kW
1/4
1/3
<3/4
Figure 4-6 750kV CSR at the Dunhuang substation, NW China

(SGCC)
3) H
igher voltage level DC transmission:
UHVDC
longer-distance power delivery. Again, China is

leading in the application of ultra-high voltage DC
(UHVDC) transmission. The 1907km, 6400MW,
800kV UHVDC demonstration project from
Xiangjiaba to Shanghai was commissioned in
April 2010. SGCC has planned to build more than
ten 800kV or 1100kV lines to transmit power
from the wind power bases in Xinjiang, Gansu and
Inner Mongolia to the Northeast-Central China
load centres, ranging in length from 1600km to
2700km [sgc11], as shown in Figure 4-7.
CSC-HVDC is a conventional HVDC transmission

technology that is relatively mature and has long
been used for long-distance, large-capacity
power transmission without midway drop points,
as well as for the interconnection of asynchronous
power networks. Compared to AC transmission,
it has advantages such as lower loss, lower
line cost, narrower corridor and rapid power
control capabilities. Like AC transmission, DC
transmission is also progressing in the direction
of ultra-high voltage levels for larger-capacity and
Mengdong
Xinjiang
Mengxi
Jiuquan
North China Grid
Northwest China Grid
Central China Grid
East China Grid
800 kV-DC
1000 kV-DC
1100 kV-DC
Figure 4-7 UHVDC transmission plan for some 10GW-level wind power bases in China
(SGCC)
69
70
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
However, there are still some problems in

the use of CSC-HVDC or UHVDC to transmit
RE [hsh12]. For example, when HVDC lines
are used to transmit only wind power to load
centres (see Figure 4-8), not only the low
utilization rate problem mentioned in 4.3.3, but
also the minimum startup power of the HVDC
lines and problems in frequency stability and
voltage stability require more research.
but if wind speeds change rapidly frequency

is difficult to control. Without strong voltage
support, outages on the sending end may
lead to voltage collapse and disconnection of
wind generators, depending on the electrical
distance between coal and wind power plants.
With voltage support from the local AC grid
(see Figure 4-10), the stability problems can be
mitigated, but attention should be paid to the
impact of wind power fluctuation on the local
grid.
The wind-fire bundling design (see Figure 4-9)

can withstand common wind disturbances,
DC
AC
HVDC
AC
DC
Wind power base
Load centre
Figure 4-8 Design for transmitting wind power only with HVDC
(SGCC)
Coal power base

DC
AC
HVDC
AC
DC
Load centre
Wind power base
Figure 4-9 Wind-fire bundling design with HVDC transmission

(SGCC)
Coal power base
DC
AC
HVDC
AC
DC
Load centre
Wind power base
Local grid
Figure 4-10 Wind-fire bundling design with HVDC transmission and local grid
(SGCC)
4) M ore flexible DC transmission: from

VSC-HVDC to MTDC and DC grids
The major advantages of VSC-HVDC as
compared to conventional CSC-HVDC described
in 3.3.2 make it not only suitable for application
in RE integration, but also more convenient to
form multi-terminal DC (MTDC). Three or more
converter stations are linked to each other with
DC lines, each interacting with an AC grid, which
facilitates flexible multi-grid interconnection and
even DC grids. These will be useful in future RE
integration where multiple resource sites and
multiple receiving ends are involved.
The preference for VSC-HVDC for relatively
long-distance offshore wind power integration is
presented in Figure 4-11. It is clear that for short
distances and relatively low power high voltage
alternating current (HVAC) technology will
suffice. As power as well as distance increases,
HVAC should be augmented with FACTS devices
in order to compensate for HVAC losses and to
provide stability support. In order to connect
remote offshore wind power plants to a grid,
VSC-HVDC technology is preferred.
In order to tap large quantities of offshore wind

power from the North Sea, a transnational
offshore grid based on multi-terminal VSCHVDC has been proposed. A simulation
benchmark test system has been developed to
facilitate detailed investigations into operational
questions, as shown in Figure 4-12. Study
results show that extensive coordination of
control systems is essential, both in order to
avoid unwanted DC loop flows and to find the
location of the optimal slack node in the system,
which also depends on the selection criteria
chosen [coc11]. Multi-terminal VSC-HVDC has
also been adopted in the design of the Tres
Amigas Super Station project proposed in the
USA to interconnect its Eastern, Western and
Texas power grids, as well as to fully exploit the
abundant RE resources near the project site
[ete11] [tgs12]. In EUs Super Grid plan, MTDC
and DC grids based on VSC-HVDC technology
will also play an important role.
600 MW
Onshore
Grid NOR
NOR
60
0
40
1,
0
1,0
MW
Hub NL
295 MW
300 MW
468 MW
282 MW
303 MW
5
,0
0
Onshore
Grid DE
Onshore
Grid NL
DK
0 MW
00
7
Onshore
Grid DK
200 MW
ra
ob
210 MW
700 MW
Hub DE
MW
0M
Hub DK
NordLink 1,40
1,000
0M
DE
NL
1,648 MW
VSC Offshore Grid

VSC Country Interconnectors
ac Offshore Grid
Figure 4-11 Transmission technologies for

integrating offshore wind power
Figure 4-12 Benchmark test system for the

European transnational offshore grid
[ern11]
[coc11]
71
72
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
To date, VSC-HVDC has not reached voltage or

capacity as high as CSC-HVDC, and VSC-HVDC
terminals also suffer from higher power loss.
VSC-HVDC transmission with voltage levels less
than 150kV and capacity less than 200MW
is relatively mature. VSC-HVDC transmission
projects with voltage levels of 350kV and
capacity of 400MW have been commissioned,
but they are still much smaller than recent
800kV, 6400MW CSC-HVDC projects.
The power loss in a VSC-HVDC converter
station is typically 1
%-6
% currently, much
higher than the 0.5%-1% loss of a CSC-HVDC
converter station. Predominantly VSC-HVDC
installations use cables as the transmission
medium, although a 350kV VSC-HVDC using
overhead lines has been commissioned by ABB
in Namibia at the Caprivi Link. One of the limits
imposed on the power transfer capability of a
VSC-based HVDC system lies in the capability
of DC power cables. Predominantly DC power
cables are used with VSC-HVDC connections
to help ease right-of-way discussions, and
they are the only option for offshore VSCHVDC connections. However, with the option of
overhead lines for VSC-HVDC, power levels can
increase as VSC technology advances. Finally,
since most VSC-HVDC projects have been
operating for only a few years, the overall longterm security and reliability of VSC-HVDC also
remain to be proved over time, as with any new
type of technology.
4.4.3 Developments in transmission

planning
1) C
urrent practice in transmission
planning
Up to now, transmission planning has
mainly been based on expert judgment
and deterministic simulations based on

mathematical models of the power network
and its components. For one scenario of load
forecast and generation portfolio for the year
studied, transmission alternatives are first
proposed by planning experts, then extensive
simulations are performed, including power
flow, stability and short-circuit studies, among
others, under typical normal and contingency
situations (generally worst cases), to verify
whether the deterministic planning criteria
can be met. If not, modifications are made to
the alternatives and simulations are re-run.
For technically viable alternatives, economic
comparisons may be made to find the best
one. In some cases, in order to account for
uncertainty, several scenarios of load forecast
and generation portfolio may be considered as
a so-called sensitivity analysis or scenario
method.
2) T
owards probabilistic transmission
planning
The necessity of probabilistic transmission
planning has been acknowledged in the
research and industry community, following the
realization that deterministic planning methods
may not be able to reflect the probabilistic
nature of outage and system parameters, and
that the widely used N-1 security criterion
may be insufficient to capture the real worst
case, for which a corresponding risk analysis
is necessary. As the separation of generation
and transmission planning resulting from power
market restructuring has made deterministic
transmission planning less meaningful, the
increasing uncertainty introduced by largecapacity RE integration is making the need for
probabilistic transmission planning more urgent
[bac11] [mit11] [pts11].
Rather than replacing traditional deterministic

transmission planning, probabilistic planning is
called to complement it by adding probabilistic
planning criteria and evaluation in the planning
processes, as shown in the conceptual flow
chart in Figure 4-13.
Probabilistic planning criteria, tools
techniques
have
been
developed
probabilistic transmission planning over the

past several decades; however, they will require
critical review for completeness and applicability
before they can become accepted by industry.
More research on additional probabilistic
planning techniques is required. High-quality,
high-resolution datasets of RE generation are
also necessary; they are currently very limited
and difficult to obtain [ner09].
and
for
Initial 7 alternatives
Environmental, societal,
and political assessment
5 alternatives
Deterministic technical
analysis, including N-1
principal
3 alternatives
Probabilistic economic
analysis
Probabilistic reliability
evaluation
The best alternative
Figure 4-13 Conceptual flow chart of probabilistic transmission planning

[pts11]
73
4.5 Operational enhancement
and an application layer energy management

system (EMS). The SCADA system covers most
of the spread-out elements in a power system,
with sensors to monitor their operational
conditions and report them to the operations
centre through communications channels.
The EMS residing in the operations centre
exploits the information collected by the SCADA
system to analyze the situation and reveal any
problems in the power system, make security
and economic dispatch and control decisions,
and send real-time commands to control the
relevant system elements through the SCADA
system. Since modern power systems rely
heavily on computerized communications and
control for operations, they have evolved into
cyber-physical systems.
4.5.1 Need for operational enhancement

When resource capacity and flexibility as well as
transmission availability have been determined
by the planning process, it is operations
responsibility to manoeuver all the system
capabilities to cope with the variability and
uncertainty resulting from the integration of largecapacity RE generation. The operations process,
as a broad concept distinct from planning, can
be further divided into scheduling, dispatch and
control processes (see Figure 4-14).
Future
The operation of modern power systems is

supported by a physical layer supervisory
control and data acquisition (SCADA) system
What: Generation and transmission planning

Who: System planners
Time frame : Years into future
Planning
What: Unit commitment, Maintenance scheduling, etc.

Who: Operational planners
Time frame: Hours to days, Seasonal, Annual
Dispatch
Control
Operation
Scheduling
Real time
74
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
What: Economic dispatch, Contingency analysis, etc.

Who: System operators
Time frame: Minutes to hours
What: Automatic frequency/voltage control,
relay protection, load shedding, etc.
Who: Automatic equipment and control systems
Time frame: Seconds to minutes
Figure 4-14 Typical processes of power system planning and operation

(SGCC)
Historically, various operational tools (EMS

applications) have been developed and deployed
to successfully address the existing variability
and uncertainty in power systems. As variability
and uncertainty increase substantially due to
the integration of large-capacity RE generation,
enhancements have to be made to major
operational tools as well as some operational
practices; the development and incorporation
of more accurate RE power forecasting is critical
to most of these enhancements. The integration
of large-capacity RE generation may also pose
more challenges to cyber-security.
4.5.2 More accurate RE power forecasts

As described in section 3.4.1, many wind
power forecast methods and programs have
been developed in the last two decades, and
the forecast accuracy has been significantly
improved, especially for short-term forecasts.
However, the forecast accuracy is still low as
compared to load demand forecasts, especially
for day-ahead and longer time horizons. In
addition, the need to forecast significant weather
events and provide probabilistic information
along with forecast results is not well addressed.
Listed below are some directions for improving
forecast accuracy and the value of forecasting
in operations [atp11] [ner09] [pei12] [sad11].
1) Model and data improvement
Improvements in atmospheric observation and
numerical weather prediction models are critical
for improving RE power forecast accuracy; here,
a promising avenue is collaboration among
related sectors at national and international
levels to improve boundary-layer weather
forecasts. Collection and processing of high
quality meteorological and electrical data from

RE power plants, both historical and real-time,
is also essential, and for this purpose fourdimensional data assimilation technology may
play an important role.
2) C entralized forecast and ensemble
forecast
Centralized forecasting at system level can
improve forecast accuracy as compared to
plant-level forecasts, thanks to the spatial
smoothing effect mentioned in section 3.4.1.
It is also beneficial for error reduction that the
single centralized forecasting system should
receive input data from several commercial
forecast providers and combine them to form a
single ensemble forecast.
3) H igh-resolution plant-level forecast
and nodal injection forecast
While centralized forecasting is the best
approach for system-level forecasts, highresolution separate forecasts at different RE
power plants are also very important for RE
generation dispatch, and for determining the
power injected into each delivery node in the
power system for managing transmission
congestion.
4) R
amp events forecast and situational
awareness
Forecasting of ramp events, sudden and large
RE generation output changes caused by severe
weather events, is of great importance to provide
situational awareness to grid operators and help
decision-making. However, the definition and
forecast methods of ramp events need more
research before ramp event forecasts can be
integrated into forecasting products.
75
76
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
5) Human forecast
System operators may become good human
forecasters after accumulating years of
experience. They sometimes outperform
advanced forecasting tools. More research
is needed into how best to combine human
forecasts with computed forecasts.
6) Probabilistic forecast
By providing not only the value but also the
probability of expected RE power production
or ramp events, probabilistic forecasts could
become very valuable for system operations.
4.5.3 Enhancement of operational tools
and practices
Power system operation is the field that sees
the most prominent and direct impacts of
large-capacity RE integration. Many aspects of
enhancements to operational tools and practices
have been discussed in the literature [coc11]
[ner09] [opp11] [sad11] [sin12]. Briefly cited below
are some major enhancements related to EMS
applications. The underlying principle behind
these developments is to improve operators
situational awareness by evaluating potential
events and their impacts, and to provide operators
with guidance on possible mitigating measures.
1) Unit commitment
Detailed unit commitment (UC) studies are
normally conducted one day ahead, to determine
what amount and types of conventional
generation units should be available at what
time to ensure the desired generation capacity,
and also enough flexibility to address system
variability and uncertainty. With increased levels
of RE generation it is recognized that UC will
not be effective unless the RE power forecast

is taken into account, and that it should be run
more frequently, say every 4 to 6 hours or at
even shorter intervals, or alternatively each time
a new RE power forecast is provided.
2) Contingency analysis
Contingency analysis (CA) assesses the impacts
of potential contingencies, which are normally
outages of different grid components, under
certain operating conditions. With increased levels
of RE generation, the contingency set must be
augmented to include those extreme ramp events
which are due to RE and outages of RE power
plants. Information from the RE power forecast is
needed to determine potential contingencies.
3) Online dynamic security analysis
Online dynamic security analysis (DSA) has
been implemented in many power system
control centres to help operators make the
right operational decisions. It evaluates system
security and stability limits based on near-realtime network topology and operating conditions,
such as the transmission line thermal limits and
the system voltage/transient/frequency stability.
By incorporating the RE power forecast and
modelling, an online DSA system can be adapted
to cope with the operational risks and challenges
resulting from increased levels of RE generation.
As an example of such a tool, the wind security
assessment tool (WSAT) implemented in
EirGrids control centre (see Figure 4-15) is used
to assess the instantaneous secure amount
of wind generation based on voltage stability
analysis (VSA) and transient stability analysis
(TSA) under normal operating conditions and
credible faults.
Results display
Near real-time
wind limits
Other
information
Security
violation details
Input Data
Real-time
system
snapshot
WSAT Manager
Python interface
Wind
forecast
Real-time
contingencies
Security assessment
Online
VSAT
Online
TSAT
Study
WSAT
Figure 4-15 Wind Security Assessment Tool (WSAT) structure

[sin12]
4) Security constrained economic dispatch

Security constrained economic dispatch
(SCED) determines how to dispatch generators
to produce electricity at the lowest cost subject
to reliability requirements and operational limits
on generation and transmission facilities. It
is now evolving to accommodate two major
directions of power system development: the
increase in high penetration RE generation
and the development of demand response
and smart grid applications (see 4.6). As
described in 4.3, more flexibility is needed
from conventional generation to integrate high
levels of large-capacity RE generation, and the
increased ramping and cycling requirements of
thermal power plants will lead to a significant
increase in their operational costs. These costs
have to be better understood before they can
be taken into account to improve the economic

dispatch process.
5) Automatic generation control
AGC is a centralized system designed to
ensure real-time generation/load balance
and frequency stability, by regulating the
power output of selected generation units
and exchanging power on tie-lines between
different power systems or control areas.
Existing AGC algorithms need to be modified
so that, based on RE power forecasts, they
may address the variability and uncertainty
of RE generation. There is also an increasing
requirement to regulate the output of RE
power plants in addition to that of conventional
generation units, based on the advanced
control capabilities and the monitoring, control
77
78
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
and data communication systems for RE power

plants described in section 4.2.
6) S
tochastic operations and risk-based
decision making
To address the increased uncertainty,
including that related to RE power forecasts,
many researchers believe that future EMS
applications should make greater use of
stochastic modelling techniques. Stochastic
UC, stochastic SCED and stochastic optimal
power flow, for example, should be feasible
by taking advantage of the greater computing
power now available. Risk-based decision
making techniques are also needed to improve
the current deterministic and binary decisionmaking process; for this, research on how to
quantify the relevant operational risks and the
severity of contingencies such as extreme ramp
events is critical.
7) Security and defence generally
Security in DSA and SCED (see above)
mainly refers to the physical aspects of power
system security, or security of supply. As a
cyber-physical system, the power system also
faces cyber-security challenges, including
the reliability of the communications systems
serving the power system, and protection of
critical information related to power system
monitoring and control as well as confidential
customer information [top12]. Failures in
cyber-security, especially those caused by
malicious cyber attacks on the control system,
may damage power system elements and
endanger the physical power systems security
of supply. Since large-capacity RE power
plants are usually remotely located and consist
of many widely-distributed, small-capacity
generating units, the cyber-security of their

control systems may require more attention.
Although substantial efforts have been made
by some organizations to address control
system security, such as the power system data
communication standard protocols developed
by IEC Technical Committee (TC)57 and widely
used in power equipment, SCADA and EMS,
cyber vulnerability is still a salient problem,
and is becoming even more complex with the
development of smart grids [iig12] [sic12].
Moreover, as power systems, meteorological
systems, communications networks, water,
commerce, etc., the so-called critical
infrastructures,
become
more
closely
integrated, it becomes increasingly important
that the security protocols in one sector are
considered within the broader context of the
security protocols in connected sectors, as
well as the security needs of the country and
region. This issue involves harmonization of
cyber-security policies both vertically (e.g.
from system operation down to individual
wind turbine control) and horizontally (e.g.
from the power grid to emergency services
and telecommunications). With regard to RE
integration, this would suggest a need for
integrated security policies between weather
forecast systems and power system operation,
specifically dispatch. For example, a highly
secure power grid system with high RE
concentrations could still be quite vulnerable
to an attack that targets the countrys weather
forecasting service, either disrupting forecasting
or providing false forecast data. A grid operator
who relied on such data might find himself in
serious trouble, beyond simple variations in
forecast.
4.6 Demand response

4.6.1 Demand response applications for
RE integration
Demand response (DR), the development and
extension of traditional demand-side management
or load management practices, is recognized as
a key application of the smart grid. Currently the
USA is far in the lead in research and application
of DR, with Europe, China and other countries
catching up [jef11]. The US Federal Energy
Regulatory Commissions (FERC) definition of
DR is: Changes in electric use by demand-side
resources from their normal consumption patterns
in response to changes in the price of electricity,
or to incentive payments designed to induce lower
electricity use at times of high wholesale market
prices or when system reliability is jeopardized
[adr11]. While this definition covers well the
current DR practices in the USA, it may not be
able to reflect DRs evolving capabilities, including
those expected to support RE integration.
RE generation, particularly for wind power

generation, which often exhibits inversepeaking characteristics: generating more
power during off-peak periods and less power
during peak demand periods. At off-peak
periods, conventional generation plants are
often already reduced to their minimum output,
and the insufficiency in demand will make wind
power curtailment inevitable, reducing wind
power plants capacity utilization efficiency and
preventing the replacement of fossil-fuelled
power generation for emission reduction.
Shifting load to off-peak periods also brings
additional energy efficiency and cost benefits
for customers. For example, Denmark has
implemented zero- and negative-spot electricity
prices to encourage customers to use electricity
during hours of excess wind generation, and
has planned for increased integration between
the transport, heating and electricity sectors
to find new sources of off-peak load for wind
energy [coc11].
2) Demand-side balancing services
As RE penetration rises, DRs value as an additional

source of power system flexibility to compensate
for the variability and uncertainty of RE generation
will increase. For example, REE (Spain) created
a demand-side management department in
2007 in order to promote demand management
mechanisms such as interruptible service, electric
vehicle integration, smart metering and time-ofuse tariffs. These mechanisms, used together,
are expected to enhance the ability of the system
to integrate increasing quantities of RE [coc11].
DR can help RE integration in two main ways,
load shifting and balancing.
Fast-acting DR can be deployed to help balance

generation and load in real time. Loads can
be aggregated and directed to respond very
quickly and therefore be capable of following
the fast ramps of RE generation, reducing the
need for ramping capability from conventional
generation [ded10]. Some system operators
are already using DR to counter down-ramps
of RE generation, showing flexibility potentially
equal to conventional generation options. But
different types of load have different response
capabilities and different costs of response,
and more research is needed to identify the
1) Load shifting
true aggregate value and capability of DR in this

area [ner09].
DR can be deployed to transfer a part of the

load to off-peak periods to absorb excess
79
80
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
4.6.2 Demand response practices and

trends
signals or neighbourhood-comparative data,

may prove useful in the future. Reactive
programmes can further be divided into
wholesale programmes administered by
independent system operators (ISOs) and
regional transmission organizations (RTOs), and
retail programmes that present customers with
retail prices carefully determined by specific
time-varying pricing structures.
The practices of DR vary not only among

different countries due to differences in
electricity markets, technological development
and goals for deploying DR, but also inside the
boundaries of certain countries. The leading
DR programmes in the USA can generally be
classified into two categories: dispatchable
programmes and reactive programmes [mit11].
Each of the dispatchable, wholesale and retail

reactive programme categories consists of
several types of programmes, as shown in
Figure 4-16. According to a recent survey [adr11],
they respectively contributed to approximately
62%, 27% and 8% of the nations potential
peak load reduction (aggregate load enrolled
in DR programmes). Although dispatchable
programmes still represent a high proportion of
enrolment, the advent of smart metres, which
allow for two-way information flows between
customers and utilities, has enabled the growth of
reactive programmes and increased engagement
with residential and commercial customers.
a) Dispatchable programmes, also known

as load management or control programmes,
allow direct control of load responses by the
grid operator or a third-party aggregator. An
incentive is often offered to customers in return
for participation.
b) Reactive programmes rely on customers
voluntary responses to a variety of signals
communicated to them. The most common
signal used at present is price, although other
types of information, such as environmental
(Dispatchable programmes)
Emergency payment programmes (25 %)
Load control
Interruptible tariff programmes (20 %)
programmes
(62 %)
Direct load control programs (17 %)
Demand
response
Capacity market programmes
Wholesale
programmes
(27 %)
Energy price programmes

Ancillary service programmes
Critical peak pricing programmes
Retail
programmes
(8 % )
Peak time rebate programmes

Real-time pricing programmes
(Reactive programmes)
Time-of-use prices programmes
Figure 4-16 Types of DR programmes in the USA

(SGCC)
Demand
response
System
reliability
System
efficiency
System
flexibility
Emergency
use
Daily use
Real-time
use
Large
customers
Small customers
(with aggregators)
Downward
only
Downward or
upward
Objective
Frequency
of calling
Participants
Direction
Figure 4-17 Trends in DR development (extending rather than replacing)

(SGCC)
From the practices in the USA some trends in

DR development can be identified, as illustrated
in Figure 4-17. The objective of deploying DR
has been to go further than simply improving
system reliability, and extend towards improving
system efficiency and then system flexibility.
The frequency with which DR is called upon
has been extending from emergency use to
daily use and further to real-time use. The
participants in DR have been extending from
large industrial and commercial customers to
smaller commercial and residential customers,
with more aggregators acting as intermediaries
between utility or grid operators and individual
customers. Finally, DR has been extending
from one direction to both directions: from
downwards, reducing load only, to both
upwards and downwards, either increasing or
reducing load as required.
1) DR for improving system reliability
The earliest and most commonly practiced
DR focuses on system reliability. A customer,
often a large industrial facility, agrees to reduce
load to guarantee system reliability under peak

demand conditions or other emergency system
events, and is paid an incentive for doing so.
Since they are designed for emergency use,
these DR programmes are infrequently called
upon.
2) DR for improving system efficiency
More recently the focus of DR has been
increasingly placed on system efficiency. Many
DR programmes have begun to focus on noncrisis peak shaving flattening load curves
to improve the efficiency of long-term power
system capacity use, since the generation,
transmission and distribution capacity of a
power system is sized to meet the expected
peak demand.
3) DR for improving system flexibility
This emerging application of DR is very important
for supporting RE integration, as mentioned in
4.6.1. For this purpose, automated and fastacting dispatchable programmes are more
effective and reactive programmes, particularly
81
82
S E C T I O N 4
ing
rat
eg
int
r
fo
s
ion
lut
so
l
ica
Future: techn
price-mediated retail programmes, may be

less effective. Changes to market rules and
reliability standards are also required in some
regions, to allow DR to participate in providing
balancing services [mit11]. This application of
DR is further envisioned in [ded10] as demand
dispatch, which is expected to perform many of
the same ancillary services currently provided
by conventional power plants. Potential loads
that are suitable for demand dispatch are
those that, when remotely controlled, would go
largely unnoticed by the customers, such as
electric hot water heaters, heating-ventilationair conditioning systems and electric vehicles.
and other aspects, and utilities to receive timeof-use data that reveals how much energy
customers use at any given time [mit11].
4.6.3 Technologies supporting more

demand response
3) Electric vehicles
Although the success of DR programmes

depends to a very large extent on effective
commercial arrangements (including rate
structures and pricing schemes) and on an
accurate evaluation of cost-effectiveness, some
new technologies are physically essential for
DR to function or to function better. We discuss
these briefly here.
1) A
dvanced metering infrastructure
technology
Advanced
metering
infrastructure
(AMI)
technology, commonly known as smart
metering, permits fine-grained communication
of system conditions to customers and finegrained measurement of customer responses
via two-way communications between the
customer and the utility. It is the technical
foundation for engaging more DR, especially
from smaller customers in the future. It allows
customers to receive information signals from
utilities involving price, environmental impact
2) Behind-the-meter technologies
AMI and home area networks also enable the
use of a host of consumer-side technologies
for building or home energy management, such
as controllable appliances, monitoring and
analysis of energy use, and price-responsive
thermostats. These technologies can enable
smaller commercial and residential customers
to respond more actively to price or other
supply-side signals. However, currently they are
relatively immature and costly [mit11].
Plug-in electric vehicles (EVs) are being

promoted in a variety of useful roles. Not only
are EVs low-emission, but they also have the
potential to function as storage facilities from
which energy can be dispatched to the grid or
the home; they can be a dispatchable, nighttime load to soak up excess wind energy; and
they can provide balancing energy and ancillary
services for RE integration.
4) Cyber-security solutions
Since DR involves the coordinated reaction or
control of a large number of loads using very
many communications messages, there are
serious concerns with the cyber-security risk.
First, private customer information such as that
concerning living habits might be vulnerable;
this concern has led to public opposition to AMI
and DR in some places. Secondly, information
can be intentionally modified or faked to gain
a financial advantage by fraud. Even more
seriously from the system operation point of
view, there are concerns that DR could be
manipulated so as to damage power system

stability and security of supply, for example a
large amount of load maliciously driven up or
down. Therefore the success of DR depends
on managing these risks through development
of technologies, standards, policies and laws.
The cyber-security solution must be an integral
part of any DR programme from the beginning.
4.7 Summary
1) In order to address the increased variability
and uncertainty brought about by integrating
higher levels of large-capacity RE, the power
system must become more flexible so as
to maintain a constant balance between
generation and load.
2) P
ower system flexibility can be achieved from
the generation side (both RE generation and
conventional generation), from the load side,
and through EES acting as either generation
or load. It can be better exploited if system
operating technologies and practices are
improved, and based on control shared over
wider geographic areas with the support of
transmission expansion.
3) R
E generation can be made more predictable,
controllable and dispatchable, or in other
words more grid-friendly, by improving the
design, operation and modelling technology
at the generating unit, plant and plant cluster
level.
4) F
lexibility in conventional generation is the
major source of power system flexibility
currently and for the foreseeable future.
Future generation planning should consider
both capacity and flexibility. Different kinds
of power plants have different degrees of

flexibility, but it is difficult to quantify their
flexibility and that of the overall system.
5) H
igher-voltage-level transmission and the
power-electronics-based FACTS and DC
transmission technologies are paving the
way for the transmission expansion needed
everywhere for accommodating more largecapacity RE generation. The development of
probabilistic transmission planning methods
is also desirable for the more uncertain future.
6) Improvements in operational technologies
and practices should be made at each
stage in power system operation, namely in
scheduling, dispatch and control. Of these,
the development of more accurate RE
generation forecasting and its incorporation
into the scheduling and dispatch tools is the
most important.
7) D
emand response, supported by new
smart grid, smart building and smart home
technologies, is a promising source of power
system flexibility in the future, but is still in
its infancy. The rate at which it will mature
and be widely applied depends heavily on
an understanding of customer behaviour
underlying the load demand, as well as on
institutional and commercial innovations.
8) T
he integration of large-capacity RE and
the application of demand response and
other smart grid technologies will bring more
challenges in cyber-security. Harmonization
of cyber-security solutions is required
both vertically within the power sector and
horizontally across sectors such as power,
communications and weather forecast
systems.
83
SECTION 5
Application of large-capacity
EES to support RE integration
86
S E C T I O N 5
EES to suppor t RE integration
Application of large -capacity
5.1 General
through reserves; and smoothing output from

individual RE plants. We do not examine energy
storage technologies themselves in great detail
here, as the MSB EES Report [msb11] has
already done so. Nor do we cover applications of
energy storage for purposes not directly related
to RE integration. Please refer to the MSB EES
Report for a detailed and thorough discussion
of all relevant energy storage technologies and
their entire scope of use.
This section examines the many uses of largecapacity energy storage to meet grid needs in
the integration of large-capacity RE. Section 4
identified the need for greater flexibility in
power systems as RE penetrations rise, and
divided sources of flexibility into grid-friendly
RE generation, generation flexibility, demand
response, system operation, and transmission
expansion. Here we revisit the first three
concepts with an emphasis on energy storage
as a means of providing generation flexibility for
the grid, RE generation flexibility, and flexibility
through demand response via electric vehicles.
5.2 Promising large-capacity EES

technologies
The universe of energy storage applications
maps closely to the challenges of integrating RE
into the grid. In the same way that RE integration
creates needs at a variety of temporal scales,
different types of energy storage are suited
to different discharge times, from seconds to
seasons. The tremendous application range of
storage is shown in Figure 5-1.
Energy storage, due to its tremendous range

of uses and configurations, may assist RE
integration in any number of ways. These uses
include, inter alia, matching generation to loads
through time-shifting; balancing the grid through
ancillary services, load-following, and loadlevelling; managing uncertainty in RE generation
Figure 5-1 Comparison of rated power, energy content and discharge time of different EES technologies
[msb11]
The suitability of an energy storage resource for

a particular discharge time-frame is determined
by its power density and energy density. Power
density refers to the energy storage technologys
ability to provide instantaneous power. A higher
power density indicates that the technology can
discharge large amounts of power on demand.
Energy density refers to the ability of the
technology to provide continuous energy over a
period of time. A high energy density indicates
that the technology can discharge energy
for long periods. Generally, energy storage
technologies with the highest power densities
tend to have the lower energy densities; they
can discharge enormous amounts of power, but
only for a short time. Likewise, technologies with
the highest energy densities tend to have lower
power densities; they can discharge energy for a
long time, but cannot provide massive amounts
of power immediately. This quality gives rise to
a division of energy storage technologies into
categories based on discharge times. While the
categories are general and nearly always admit
of exceptions, they are useful in conceptualizing
how many roles storage can play with respect
to renewables integration.
Short discharge time resources discharge
for seconds or minutes, and have an energy-topower ratio (kWh/kW) of less than 1. Examples
include double layer capacitors (DLCs),
superconducting magnetic energy storage
(SMES), and flywheels (FES). These resources
can provide instantaneous frequency regulation
services to the grid that mitigate the impact of
REs uncontrollable variability.
Medium discharge time resources discharge
for minutes to hours, and have an energy-topower ratio of between 1 and 10. This category
is dominated by batteries, namely lead acid (LA),
lithium ion (Li-ion), and sodium sulphur (NaS),

though flywheels may also be used. Medium
discharge time resources are useful for power
quality and reliability, power balancing and loadfollowing, reserves, consumer-side time-shifting,
and generation-side output smoothing. Moreover,
specific batteries may be designed so as to
optimize for power density or energy density. As
such, they are relevant to both the uncontrollable
variability and partial unpredictability that RE
generation brings to the grid.
Medium-to-long discharge time resources
discharge for hours to days, and have energy-topower ratios of between 5 and 30. They include
pumped hydro storage (PHS), compressed air
energy storage (CAES), and redox flow batteries
(RFBs). RFBs are particularly flexible in their
design, as designers may independently scale
the batterys power density and energy density by
adjusting the size of the cell stacks or the volume
of electrolytes, respectively. Technologies in this
category are useful primarily for load-following
and time-shifting, and can assist RE integration
by hedging against weather uncertainties and
solving diurnal mismatch of wind generation and
peak loads.
Long discharge time resources may
discharge for days to months, and have
energy-to-power ratios of over 10. They include
hydrogen and synthetic natural gas (SNG).
Technologies in this category are thought to
be useful for seasonal time-shifting, and due
to their expense and inefficiency will likely see
deployment only when RE penetrations are
very large. For example, large amounts of solar
power on the grid will produce large amounts of
energy in the summer months, but significantly
less in the winter. Storing excess generation in
the summer as hydrogen or SNG and converting
87
88
S E C T I O N 5
it back to electricity in the winter would allow

a time-shift of generation from one season
to the next. Such technologies can assist RE
integration in the long term by deferring the need
for transmission expansion and interconnection
that arises due to the locational dependency of
renewable resources.
predictions about exactly what the distribution

of uses will or ought to be for EES in assisting
RE integration, and instead simply present all of
the potential uses from a variety of perspectives.
The actual use of EES in various countries in
the future will vary significantly depending on
government policies, utility strategies, social
and cultural factors, and the peculiarities of
each particular grid.
5.3 Roles of EES in RE integration

5.3.1 Grid-side roles of EES
The widest range of uses for EES lies in services
to the grid operator in providing generation
flexibility. These services also represent from
the grid operators perspective the optimal
use of storage as a tool to mitigate variability
and uncertainty for an entire grid, rather than
for specific loads or generation assets. The
optimality arises from the fact that integration
of large amounts of wind and solar energy
over large geographic areas results in lower
net variability and output uncertainty than the
integration of a single RE plant, and so the need
for services overall is reduced.
Nevertheless, it is simplistic to expect that this
will be the only use of energy storage for RE
integration that emerges in future grids. Indeed,
the grid operators is not the only perspective
that is important or relevant. Individual RE
generators or plants facing specific incentive
policies or isolated grids may find it in their best
interests to co-locate generation and storage
to level output prior to grid integration. On the
demand side, expanded use of electric vehicles
may provide substantial aggregate energy
storage to the grid even if the storage resource
itself appears sub-optimal to the grid operator.
We avoid making any specific judgments or
Table 5-1 describes various grid-side roles of

energy storage and their relevance to largecapacity RE integration challenges, along with
some examples of EES technologies currently
in use. These examples are impressionistic: the
suitability of an EES technology for a particular
use is highly context-dependent and will vary
according to the needs of the grid operator and
the specific design of the EES.
1) G rid-side EES case study: The national
wind power, solar power, energy storage
and transmission demonstration
project, Zhangbei, China
The national wind power, solar power, energy
storage and transmission demonstration project
is co-sponsored by the Ministry of Finance,
the Ministry of Science and Technology, the
National Energy Bureau and SGCC. The project
is located in North Zhangjiakou. The wind and
solar resources are rich, but the local load is
small and the installation is far away from the
Beijing-Tianjin-Tangshan load centre. Thus
the energy must be transmitted to the load
centre by a high-voltage and long-distance
transmission network. This project exemplifies
the basic characteristics of RE development
in China, and is a typical project for studying
the problem of accommodating large-scale
renewable power.
Table 5-1 Grid-side roles of EES

[msb11] [tre10]
Time scale(s)
Description
Benefits to RE integration
Examples of EES
technologies
Hours to days
EES allows storage of offpeak energy and release

during high-demand
period
A solution to diurnal
generation cycles that do not
match load cycles
NaS batteries,
CAES, PHS, RFB
Seasonal shifting
Months
EES stores energy

for months at a time,
releasing it at times of the
year when RE output is
typically lower
Allows use of renewablygenerated energy yearround, reducing reliance

on traditional generation
in seasons with, e.g., low
sunlight
Hydrogen, SNG
Load following / Ramping
Minutes to
hours
EES follows hourly

changes in demand
throughout the day
May mitigate partial

unpredictability in RE output
during critical load times
Batteries, flywheels,
PHS, CAES, RFB
< 1 second
Provision of reactive
power to the grid to
handle voltage spikes,
sags and harmonics
Mitigates voltage instability

and harmonics caused
or exacerbated by
uncontrollable variability of
RE generation
LA batteries, NaS
batteries, flywheels,
RFB
Seconds to
minutes
A fast-response increase
or decrease in energy
output to stabilize
frequency
Mitigates uncontrollable
moment-to-moment
variability in RE generation
output
Li-ion batteries, NaS

batteries, flywheels,
PHS (with advanced
variable speed
control)
~10 Minutes
A fast-response
increase or decrease in
energy output to cover
a contingency, e.g.
generator failure
Mitigates partial
unpredictability of RE
generation output, providing
(or removing) energy when
the RE resource does not
perform as expected
PHS, flywheels,
batteries
Minutes to
hours
A slower response
resource that comes
online to replace a
spinning reserve
Provides firm power in the

event of an especially severe
and long-lasting drop in RE
output. Use for RE integration
is expected to be infrequent
and low-value
PHS
Efficient use of
transmission network
Minutes to
hours
EES can help grid

operators defer
transmission system
upgrades through timeshifting and more efficient
operating reserves
Reduced transmission
costs, mitigates locational
dependency challenges of
RE generation
Li-ion
Isolated grid support
Seconds to
hours
EES can assist in the

integration of RE on small
power grids, such as
those in use on islands
Time-shifting and powerquality applications to

mitigate variability and
unpredictability of RE
generation
LA batteries
Minutes to
hours
EES may be used to

re-start the power
system in the event of a
catastrophic failure
No specific benefit accrues

to RE integration, but storage
resources may nonetheless
provide black start capability
to the grid
Role
Time shifting / Arbitrage /
Load levelling
Power quality
and stability
Operating reserves
Frequency regulation
Spinning Reserves
Supplemental reserves
Emergency power supply

/ Black start
LA batteries
89
90
S E C T I O N 5
The planned capacity of the project is 500MW

wind power, 100MW PV power and 110MW
energy storage. Phase I of the project, which
was completed in 2011, consists of 100MW
wind power, 40MW PV power and 20MW
energy storage. In order to test the performance
of different types of battery storage, three
types of battery storage are used in the 20MW
energy storage station: 14MW of lithium iron
phosphate (LiFePO4, LFP) batteries, 4MW of
NaS batteries and 2MW of vanadium redox
flow batteries (VRFBs).
system has contributed to making the wind

farm and PV station more grid-friendly.
2) G rid-side EES case study: Battery
storage in Southern China
A demonstration lithium iron phosphate battery
storage station has been built in Shenzhen for
both commercial and research purposes. The
storage station was planned to have a capacity
of 10MW/40MWh, divided equally into two
phases. 4MW/16MWh of the phase I project
has already been put into operation, and of the
rest 1MW/4MWh will be installed in 2012. It is
managed by the Peak/Frequency Regulation and
Generation Company, a subordinate company of
The architecture of the phase I project is shown

in Figure 5-2. Through a panoramic intelligent
optimal control system, panoramic monitoring,
intelligent optimization, comprehensive control
and smooth mode-switching between wind,
solar and storage, the project has met targets
of output smoothing, schedule following, load
levelling and frequency regulation. The storage
220 kV Smart
substation
the China Southern Power Grid (CSPG) which

is responsible for the construction, operation,
maintenance and management of the peak/
frequency regulating power plants in CSPG,
including several hydro power plants and all
pumped-hydro storage power plants.
Strong smart grid
Legends:
Storage
AC 220 kV
AC 35 kV
Control signal
Solar
Wind +solar
AC 35 kV
AC
AC
DC
100 MW wind farm
Wind
40 MW PV station
DC
Two-way
conversion
20 MW storage
Wind +solar +
storage
7 operation modes of
wind-solar-storage
combination
Wind +storage
Solar+storage
Panoramic intelligent optimal control

system can realize the panoramic
monitoring and intelligent optimization of
the wind farm, PV station and storage
system according to the dispatch schedule ,
wind power forecast and solar power
forecast . It can also automatically
configure and seamlessly switch from one
operation mode to another .
Panoramic intelligent optimal control system
Figure 5-2 Architecture of the national wind power, solar power, energy storage and transmission
demonstration project, Phase I
(SGCC)
Battery units are connected at the 10kV bus

of the 110kV Biling substation via 10/0.4kV
transformers, as shown in Figure 5-3. A battery
unit consists of battery, power conversion
system (PCS) and battery management system
(BMS). The storage station can work in several
modes. It can adjust its output according to a
scheduled curve given by the dispatch centre

or a fixed curve for load levelling, or participate
in advanced regulation and backup services.
For example, the load forecast curve at the
Biling substation of one day with and without
load shifting by the storage station is shown in
Figure 5-4.
10 kV
MV Switch
MV Switch
1,250 kVA
1,250 kVA
LV Switch
1,250 kVA
LV Switch
0.4kV
Storage Unit
(PCS
Battery
BMS)
MV Switch
LV Switch
0.4kV
Storage Unit
(PCS
Battery
BMS)
Storage Unit
(PCS
Battery
BMS)
Storage Unit
(PCS
Battery
BMS)
0.4kV
Storage Unit
(PCS
Battery
BMS)
Storage Unit
(PCS
Battery
BMS)
Figure 5-3 Architecture of the Biling energy storage station

(BYD)
50
45
40
Daily load forecast
35
Power
(MW)
Load forecast after

load shifting
30
25
10 0
:0
0
12
:0
0
14
:0
0
16
:0
18 0
:0
0
20
:0
22 0
:0
0
:0
8
:0
6
:0
4
:0
2
:0
20
Time
Figure 5-4 A daily load forecast curve at the Biling substation with/without storage
(BYD)
91
92
S E C T I O N 5
3) G
rid-side EES case study: Multiapplication use of pumped hydro
storage in Taiwan
Though PHS is historically associated with
time-shifting, newer units offer a considerably
expanded range of operation, including the use
of PHS for operating reserves such as frequency
regulation and spinning reserves. The Taiwan
Power System, for example, contains ten PHS
units: four 250MW units located at the Ming-Hu
hydro plant and six 267MW units located at the
automatic generation control of the turbine

speed and a reduction in system operating
costs by replacing peaking generators. During
off-peak periods, PHS units operate in pumping
mode to consume surplus energy, and also
offer emergency reserves by way of PHS loadshedding. That is, if a large generator fails in an
off-peak period, the PHSs load from pumping
may be immediately shed to stabilize system
frequency [rvp08].
Ming-Tan hydro plant.
5.3.2 Generation-side roles of EES
Because Taiwans grid is isolated, it must

guarantee sufficient operating reserves to
maintain system frequency without loadshedding in the event of the failure of its
largest generating unit. In Taiwans case, this
largest unit is 950MW, a substantial portion of
Taiwans 34630MW of capacity. Taiwan thus
uses PHS units in daily operations for both
Operators of RE generation plants may use

energy storage technologies to assist in the
integration of a particular plant, or of several
plants that feed into the same substation. Using
the terminology of section 4, EES used in this
fashion serves to improve the grid-friendliness
of RE generation itself. It is important to
understand that generation-side use of energy
time-shifting and operating reserve functions

simultaneously. During peak demand periods,
the PHS units are operated in generating mode,
supplying both frequency regulation through
storage is not simply a shift in ownership of

the storage resource, but an entirely different
role for storage from that envisioned by gridside use of EES. Rather than using EES as a
tool to balance an entire power grid, an RE
generation plant may use EES to provide
integration applications prior to grid integration,
either at the plant or substation level. While the
technical requirements of generation-side EES
applications are similar to those of grid-side
EES, greater flexibility is required of generation
side EES facilities, because a single RE plant
exhibits greater variability and uncertainty than
many RE plants aggregated on the same grid.
This means that dedicating EES facilities to
specific RE generation results in proportionately
higher costs than using EES to balance net
variability and uncertainty on the grid. For
isolated and geographically-constrained grids,
Figure 5-5 A PHS plant in Taiwan

[tem12]
however, co-location of RE generation and EES

may be an attractive option, as balancing such
grids through interregional trading, conventional
backup capacity or demand-side management
is more challenging than for larger and more
interconnected grids.
Essentially, generation-side use of EES aims
to transform an uncontrollably variable and
partially unpredictable resource into a controlled
and predictable one it turns RE generation
into something that looks very much like
conventional energy generation. Such an RE
generation resource is said to be dispatchable. It
may also play a role in effectively utilizing limited
transmission capacity, particularly where the
RE generation is located on an isolated or weak
grid. Generation-side uses of EES include:
its generator produces it, and stands

ready to dispatch energy to the grid
when needed. This can make RE output
both predictable to grid operators and
co-temporal to demand. Time shifting
functions require EES facilities to store
large quantities of energy for significant
periods of time, from hours to days. NaS
batteries exemplify the qualities needed
for this function: they may store relatively
large amounts of energy efficiently for
hours at a time as well as ramp quickly, as
shown in Figure 5-6. Storage efficiency is
very important for economical operation
of time shifting, as an inefficient storage
facility will lose significant portions of the
energy it time-shifts. Efficiency varies
greatly by EES technology and also by the
Time shifting. The dedicated energy

storage facility stores energy whenever
operation profile, as is covered in detail in

the MSB EES Report.
Large capacity Long hours
System output [kW]

100 000
Capacitor
fly wheel
NAS battery
10 000
500 kW
Li-ion battery
stationary
More than 10 000 kW

several hours
1 000
100
Lead-acid battery
10
Electric vehicle
Mobile PC
Li-ion battery
4
Time [hours]
Figure 5-6 Comparison of battery output and duration

(NGK)
93
94
S E C T I O N 5
Output smoothing/flattening. Even

when RE generation is producing energy at
a time when it is needed, the EES resource
may be used to smooth out fluctuations in
frequency and voltage that result from the
inherently variable nature of RE generation.
Smoothing functions require ramping
capability the ability to rapidly change
power output or uptake in order to regulate
the output of the RE plant. When RE
output spikes, the EES technology must
be capable of storing the excess energy
quickly. Conversely, when output suddenly
drops, the storage system must be able
to release energy quickly to provide extra
power, keeping the plant output stable.
The necessary function of storage facilities
varies according to the requirements.
In some cases just smoothing output is
satisfactory, but in other cases output
is required to be kept at the fixed values.
Output smoothing at the plant level reduces
the need for power quality and ancillary
services on the grid itself.
Transmission utilization efficiency.
Because RE generation is location-
dependent, sufficient transmission may

not be available to move energy to loads. It
is often the case that transmission may be
available, but it may be heavily congested.
Generation-side EES resources may allow
for more efficient use of transmission
capacity by allowing an RE generation
facility to wait to use the transmission line
until congestion has cleared.
1) Case study of EES support of RE plant
integration in Japan
In 2008, Japan Wind Development Co. (JWD)
began operating the first commercial Wind
and NAS Battery Hybrid System (see Figures
5-7 and 5-8). This plant consists of 51MW
(1500kW 34 units) of wind turbines and
34MW (2000kW 17 units) of NAS batteries.
The NAS battery application regulates the output
of the plant to produce more electricity during
high demand (price) periods, and less during
low demand (price) periods. Output can also be
reduced when system conditions require. JWD
has operated its wind and EES technologies in
combination according to plan for 3 years.
Figure 5-7 JWDs wind and NAS battery hybrid system in Japan
(NGK)
Compensation for intermittent renewable energy
Power
Power
Charge
Discharge
Wind power
Smoothing and energy

shifting by NAS
Flat operation
Shifting a firm capacity to higher demand period
Power
PV output
Solar power
Energy shift
Figure 5-8 Compensation for intermittent RE

(NGK)
5.3.3 Demand-side roles of EES

Energy storage has a number of applications
for energy consumers; time-shifting to reduce
consumption of grid electricity at peak times,
firm power for off-grid, renewably-powered
homes or critical industrial applications, and
emergency power supply are a few examples.
These applications, however, are related more
to the needs of the consumer than to solving
particular challenges related to the integration
of large-capacity RE. In seeking demand-side
EES technologies that directly relate to largecapacity RE integration, only one critical type
emerges: electric vehicles.
EVs are significant to RE integration because
of the potential for aggregation. While a single
EV can store a relatively small amount of
energy, many EVs all plugged into the grid at
the same time may someday be operated as
a single large energy-storage device, or virtual
power plant (VPP). As such an electric vehicle
virtual power plant (EVPP) may provide both
time-shifting and other energy applications to
store RE at times of low demand and release

it to meet peak demand, as well as operating
reserves such as frequency regulation service,
increasing quantities of which are needed as
more variable RE generation is added to a
system [mev09]. Such functions are referred to
as vehicle-to-grid (V2G) systems.
EVPPs providing V2G services must satisfy the
requirements of both vehicle owners and grid
operators. By aggregating individual vehicles into
a single controllable EES resource, an EVPP can
potentially achieve this balancing act, bidding and
providing ancillary services at all times without
locking a vehicle owner into a charging station
from which she or he cannot depart at will. Thus
the vehicle owner is not inconvenienced and
the grid operator may treat the EVPP as though
it were a conventional provider of ancillary
services [tec10]. Aggregation also allows for the
creation of a large enough virtual facility to meet
the capacity requirements of many ancillary
service markets, which are often too high for an
individual EV to satisfy.
95
96
S E C T I O N 5
Figure 5-9 Bornholms distribution grid and power plant locations

[evf10]
EVPPs are still conceptual in nature, and involve

significant complexities that are beyond the
scope of this report. A number of modelling
efforts are presently examining EVPP feasibility
and architecture. One of the more robust and REintegration relevant modelling efforts is located on
the Danish island of Bornholm, which relies heavily
on wind turbines with 30MW of wind capacity that
services 22% of the islands load. The EDISON
project on Bornholm aims to coordinate charging
and discharging of EVs so as to optimize utilization
of wind energy on the islands grid. Successfully
implementing V2G functionality so as to support
grid integration of RE will require a host of new
standards and grid codes [evf10] [gcr09].
do not change considerably when EES is used

to support RE integration. Consequently, we
do not cover those issues again here. Rather,
this section focuses on a larger-scale inquiry
as to the role of EES in assisting RE integration
on a specific electricity grid. That is, how can
a prospective operator of an EES facility be it
grid-side, generation-side, or demand-side VPP
determine whether and how to design, place and
utilize the facility? What type of storage to install,
and with what capacity, depends on how that
storage will be used. How it is used depends on
where it is located and the specifics of the power
system that it serves. Where it is located depends
on both what type of storage is being installed
and on the siting of other future generation.
5.4 Technology needs of largecapacity EES applications
EES technologies tend to be flexible; they can

provide multiple services within a number of
time scales, as explained above. A PHS plant
with a variable-speed turbine, for example, may
participate in both time-shifting functions and
frequency regulation. Its design specifications
The interconnection and facility management

needs of EES technologies are discussed at
length in the MSB EES Report, and these needs
are likely to change depending on the proportion

of operations that fall under time-shifting and
the proportion of operations that fall under
regulation. Those proportions, in turn, depend
on the generation profiles and interconnection
status of electricity grids, which will change
substantially over the coming decades.
Because an EES technology can draw on a
number of value streams which themselves
may be met by other means, there is a need for
optimization of storage placement and use within
the context of the power system as a whole, both
today and into the future. Research on storage
optimization is already under way, but there
is relatively little convergence or organization
of results at the moment. In 2009 alone, over
500 published articles applied optimization
algorithms to RE in some way, but a review of
the literature demonstrates the experimental
and highly varied topics and approaches of the
researchers [oma11]. An optimal optimization
strategy that is ready for global, commercialscale use has yet to emerge for low-carbon
energy system planning.
In addition to their present lack of convergence,
an impressionistic review of several popular
storage-specific optimization studies reveals
a focus on optimizing economic dispatch
decisions for a specific type of storage facility
on a known grid configuration, often a simplified
or isolated grid [ede10] [mlt10] [ops08] [sco12]
[vce03]. These studies treat the installation of
an EES facility as an a priori decision. In other
words, they assume a storage facility, and then
go about determining its optimal operation.
This is not to say that such optimization studies
are not necessary and useful they are most
certainly both, particularly to grid operators
working with existing or planned EES facilities.
But there is a need for peer-reviewed,

consensus-supported optimization tools at
a higher level of abstraction. Such tools could
provide commercial-level decision support
for the planning of storage on the grid at a
pre-installation stage from a more global
perspective, and across larger time scales.
Indeed, when considering the long-term needs
of global electricity grids as they accommodate
ever-increasing penetrations of variable RE, the
decision of when and where to place certain
kinds of energy storage in the first place is a
primary question. The HOMER optimization
model for distributed power, which allows a user
to evaluate economic and technical feasibility
for a wide range of remote, stand-alone and
distributed generation applications, represents a
potential starting point for developing such tools
[hey12].
Based on the topics discussed throughout this
paper, an EES planning and decision tool for
utilities and facility developers might consider
the following factors, both at present and in
future scenarios:
amounts and net variability of RE
generation on the grid;
interconnectivity of the grid to other grids,
and balancing capabilities between them;
conventional backup capacity available
and desirable;
demand-side management applications
and capabilities;
system costs or market prices for
operating
reserves,
power
quality
services, and balancing energy;
time-shifting/arbitrage potential in relevant
energy markets; and
technological capabilities and flexibility of
various EES technologies.
97
98
S E C T I O N 5
Such an analysis would provide a clearer picture

of the need for EES on a particular grid as
compared to other solutions to RE integration,
as well as a sense of its likely future uses. This
knowledge in turn may inform utility or facilityowner decisions about where to place EES, in
what amounts, and how to use the technology.
5.5 Summary
EES may serve as a source of flexibility for the
integration of RE in a wide variety of ways, from
improving the grid-friendliness of RE generation
itself through increasing generation flexibility
to providing demand response from electric
vehicles. These represent the near-term uses
of energy storage as one means among many
of providing system flexibility. In the medium
term, energy storage may allow, through both
balancing and time-shifting functions, for more

effective and full utilization of transmission
lines and thus assist in transmission expansion
and siting to RE resource areas. In the longer
term, energy storage may influence energy
system planning in unique and profound ways.
Large-scale, long-term energy storage such as
hydrogen and synthetic natural gas may provide
a means of storing seasonally-produced RE
for months or years and thus serve the need
for dispatchable and controllable generation
that is currently met through fossil fuels. The
cost of such storage is currently considered
prohibitively expensive and the energy
penalties too high by many system operators
and governments. Advances in technology and
shifts in the politics of energy may be necessary
before such a future becomes likely.
SECTION 6
Standards for large-capacity
RE integration
100
SECTION 6
RE integration
Standards for large -capacity
6.1 General
Besides improvements in the technologies,
methods and operational practices described
in sections 4 and 5, improvement in standards
is another important aspect of supporting the
integration of more large-capacity RE generation
while maintaining power system reliability
and stability. Many device-level standards
have already been developed, such as the
IEC61400 series on wind turbines developed
by IEC TC88 and the IEC60904 series on
PV devices developed by IEC TC82. These
standards are very important for promoting
the development of wind and solar PV power
generation technologies. A new TC, IEC TC117,
was also established in 2011 for solar thermal
power plants.
But for grid integration, more relevant are the
system-level integration standards prescribing
the performance of RE power plants and
their interaction with the power system, such
as the requirements for the interconnection,
design, modelling, testing, monitoring, control
and operation of RE power plants. Since solar
thermal power plants use steam-turbine-driven
synchronous generators and standards for
them can therefore easily be adapted from

those for conventional thermal power plants,
this section focuses on integration standards
for large-capacity wind and PV power plants.
6.2 Present situation

Currently, RE integration standards mainly
exist at the national level or grid company
level. Based on experiences and lessons
learned from the past and from other countries
[pei12], many countries or grid companies
have been updating their general grid codes,
or developing separate standards documents
such as requirements or guidelines, to meet the
demands of fast-growing wind and PV power
generation. Some standards for wind power
plant interconnection in some major countries
are listed in Table 6-1. Some of the grid codes
also include requirements for PV power
integration, since they are intended to address
the interconnection of all kinds of generation as
well as loads (i.e. all customers of the grid). The
major integration standards for wind and PV
power in China are listed in Table 6-2, including
national, industry-wide and grid company level
standards.
Table 6-1 Some wind power plant interconnection standards in some major countries
Country
Issued by/in
Numbering or
version
Title
ONS/2008
Brazilian Grid Codes

( Procedimentos de Rede )
EPE/2009
Guidelines for wind power generation

expansion in Brazil
Brazil
Manitoba Hydro/2009
Version 2
Transmission provider requirements for the

connection of power plants to the HydroQubec transmission system
Hydro-Qubec/2009
Canada
BCTC/2008
Transmission system interconnection

requirements
Revision 0
CanWEA/2006
60kV to 500kV technical interconnection

requirements for power generators
CanWEA base code
AESO Alberta/2004
Revision 0
Wind power facility technical requirements
Denmark
Elkraft System and

Eltra/2004
Regulation TF 3.2.5
Technical regulation for the properties and

the regulation of wind turbines connected to
grids with voltages above 100kV
Germany
E.ON Netz/2006
Ireland
EirGrid / 2009
Grid code high and extra high voltage

Version 3.4
EirGrid grid code
Ministry of Economy,
Trade and Industry/2004
Guidelines of technical requirements for

system interconnection for maintaining power
quality
Japan Electric
Association/2010
Grid interconnection code
Ministry of Economy,
Trade and Industry/2009
Ministerial ordinance setting technical

standards concerning wind power generation
facilities
Japan Electric
Association/2001
Wind turbine generator code
Spain
REE/2006
P. O. 12.3
Installations connected to a power

transmission system and generating
equipment: minimum design requirements,
equipment, operations, commissioning and
safety
UK
National Grid Electricity

Transmission plc/2010
Issue 4 Revision 5
The grid code
USA
FERC/2005
RM05-4-001;
Order No. 661-A
Interconnection for wind energy
Japan
101
102
SECTION 6
RE integration
Table 6-2 Major integration standards for wind and PV power in China
Issued by
Numbering
Title
Notes
AQSIQ
GB/T 19963-2011
Technical rules for connecting wind farm

to power system
National standard, replacing

GB/Z 19963-2005
NEA
NB/T 31003-2011
Design regulations for large-scale wind

power connecting to the system
Industry-wide standard
NDRC
DL/T 5383-2007
Technical specification for wind power

plant design
Industry-wide standard
SGCC
Q/GDW 392-2009
Technical rules for connecting wind farm

into power grid
Grid company standard
SGCC
Q/GDW 432-2010
Specification for wind power dispatch

and operation management
SGCC
Q/GDW 588-2011
Functional specification for wind power

forecasting
SGCC
Q/GDW xxx-201x
Technical rules for wind farm reactive

power configuration and voltage control
Grid company standard,

in process of approval
SGCC
Q/GDW xxx-201x
Standard for wind farm dispatch and

operation information exchange

SGCC
Q/GDW xxx-201x
Procedure for wind turbine grid

compatibility testing

AQSIQ, SAC
GB/Z 19964-2005
Technical rules for connecting PV power

station to power system
National standard,
under revision
SGCC
Q/GDW 480-2010
Technical rules for PV power station

connected to power grid
SGCC
Q/GDW 618-2011
Test procedures for PV power station

connected to power grid
SGCC
Q/GDW xxx-201x
Technical specification for PV power

station power forecast

Since their intention is to solve similar problems,

the contents of integration standards in different
countries or grid companies are often similar.
For example, most of the wind power plant
interconnection standards contain requirements
for the following aspects:
(1) Voltage range for continuous operation
(2) Frequency range for continuous operation
(3) Active power set point and ramp rate control
(4) Reactive power (power factor) control and

voltage regulation
(5) LVRT
(6) Power quality, e.g. flicker, harmonics, voltage
fluctuation.
However, owing to the differences among the
countries and grid companies and how their
grids are managed, as well as the different
6.3 Future needs
features and development stages of RE

power generation, these integration standards
may also differ in much of their contents and
especially in the specific values of certain
requirements. For example, the newly updated
GB/T 19963-2011 (replacing GB/Z 199632005) in China, which was issued at the end of
2011 and entered into force on 1 June 2012, also
sets out requirements on the number of circuits
connecting power plants to the grid, generation
forecast and report, configuration of reactive
power compensation, provision of simulation
models and parameters, communication with
the grid operator, and provision of field test
reports. To take the low-voltage ride-through
requirement as a specific example, Figure 6-1
shows the differences among major countries
or grid companies. In addition to the LVRT
requirement, some countries or grid companies
(e.g. E.ON Netz in Germany) also require wind
power plants to provide reactive power during
a fault period to help system voltage recovery.
In order to support the integration of more largecapacity RE generation, much standardization

effort is needed. Since it would not be useful to
discuss detailed technical requirements here,
some important general considerations are
given below.
1) Standards should be developed and kept
continuously updated to reflect advances in
RE generation technology and encourage
RE to become more grid-friendly, with
performance comparable or even superior
to that of conventional generators. For
example, the ability of wind power plants
to provide zero-voltage ride-through is
becoming an industry norm, high-voltage
ride-through is under discussion, and inertial
response may also be required in the near
future [pei12] [sra12].
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.2
0.2
0.4
0.6 0.8
1.0
1.2 1.4
1.6
1.8 2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
Time (s)
ELTRA & ELKRAFT
E.ON Type 1
FERC
REE
China
Figure 6-1 Differences in LVRT requirements in major countries or grid companies

(SGCC)
103
104
SECTION 6
RE integration
2) Interconnection standards should place

performance requirements at the plant level
or at the point of interconnection rather than
interfere in how these requirements are met
within the RE power plant. RE power plants
should be treated as closely as possible in
the same way as conventional power plants
for equity and simplicity, while appreciating
the unique features of RE power generation.
3)
Interconnection standards should also
consider anticipated as well as existing
conditions, aggregate impacts, and the
effects of displacing conventional generation
by RE generation. It is difficult to modify
requirements on existing facilities after the
fact [opp11].
interconnection
standards,
4)
Besides
standards or best-practice documents are
also needed in the whole planning, design,
commissioning and operation process for
RE integration, such as modelling, testing,
communications,
monitoring,
control,
generation
forecast,
scheduling
and
dispatch. For example, the Typical Design of
Wind Farm Electrical Systems [tpd11] issued
by SGCC in August 2011 has provided
modular design guidance for wind power
plants in China to meet the interconnection
performance requirements in an efficient
and cost-effective way. The IEEE PES Wind
and Solar Plant Collector System Design
Working Group is also planning to initiate
standards-making activities on wind power
plant collector system design [aba11]. In this
work, the collector system of a wind power
plant is informally defined as everything in
the power plant that is not a wind turbine
generator, which is similar to the concept of
electrical system used by SGCC.
5) Early experience in distributed RE integration

may be helpful for developing standards for
large-capacity RE integration, but largecapacity RE integration differs in many
respects from distributed RE integration
and should be treated very differently. One
example is that the current standards for
relatively small-capacity PV power plants
may not be applicable to large-capacity
desert PV power plants. Another example is
the conflict between the LVRT requirements
of FERC Order NO. 661-A and IEEE
Standard 1547 for distributed resources
interconnection [bac11].
6) In addition to continuing the development
of device-level standards, the IEC should
make an effort to develop system-level,
performance-oriented
RE
integration
standards, based on relevant national,
regional or grid company standards.
Although the challenges and practices of
RE integration differ substantially among
different countries, there are many common
issues and interests. To enable this to
happen, platforms for worldwide research,
discussion and exchange of experiences
are needed. To facilitate communication,
developing a common language and
terminology for RE integration might be a
good starting point.
7) Other, supporting standards are also needed
in related technologies, such as MTDC
and DC grids as well as demand response
[des10], but they are only indirectly relevant.
The IEC and other standardization bodies in
some cases already have groups addressing
them.
SECTION 7
Conclusions and
recommendations
106
SECTION 7
tions
Conclusions and recommenda
7.1 Conclusions
Renewable energies, driven by climate change,
fuel security and other motives, will be providing
more and more of our electricity in the future.
They represent an opportunity and a risk. The
opportunity is not the subject of the present
paper; it is assumed simply that excellent reasons
exist for the share of renewables in the energy mix
to grow considerably, and that they will therefore
do so. The risk stems from characteristics of
certain renewables which make them difficult to
incorporate into our current electricity system.
It is only the renewables (and their large-scale
use) presenting that risk which are dealt with
here, for together with many others it is the IECs
responsibility to help the world community cope
with the risk. The renewable energies in question
are wind and solar both photovoltaic and
thermal and the risk is that if they are present on
a large scale their variability and unpredictability
will prevent the correct functioning of the whole
electricity supply grid.
We have seen that the more renewables we
feed into the grid, the more difficult the grid and
its electrical properties will be to control and to
operate efficiently. The risks include frequency
and voltage fluctuations and outages, as well as
major inefficiencies and waste. Much is already
known and done to stay in control, but it is
not enough for the 15%, 25% or even 35% of
variable renewables some grids will contain over
the next decades. Section 4 shows that gridfriendly renewable generation and renewablefriendly grids are both needed, and suggests
some methods for achieving them. These
include improved forecasting of the likely energy
available, flexibility and reserves to guarantee
supply and the grids electrical characteristics,
information and fast reactions to enable
constant control, and enhanced transmission

capability to adjust the grid without wasting
energy. A constant in many of the methods is
that the availability of large-scale EES will make
them easier to apply, so the lessons from the
IECs preceding White Paper on that subject
have been very useful in the current one.
Two fundamental conclusions may be drawn.
First, we understand, to a certain extent, what will
be needed to cope with large-scale renewables
in the grid but we do not yet have what we
need. Very considerable efforts will be needed
to obtain it, whether it is knowledge, practical
experience, tools, guidance or investment.
Secondly, neither theoretical knowledge nor
practical experience is enough if it is applied by
just those who know, or just those who have the
experience, separately in their own domains.
That is happening today, and it will obviously
not be able to cope with the increase in
renewables. Instead, it will be required to attack
the problem together, across borders and
areas of responsibility, basing the solutions on
common research, tools and infrastructure, and
in particular on common rules and international
standards. The problem is too complex for any
other approach to work.
7.2 Recommendations addressed

to policy-makers and
regulators
In addition to the recommendations below,
those already formulated in the two previous IEC
White Papers remain relevant for the present
case, in particular Recommendations 5.5.1,
5.5.4, 5.5.5, 5.6.2 and 5.7.2 from the MSB EES
Report, and Recommendations 8.2.1, 8.2.3,
8.2.4 and 8.2.9 from the MSB EEE Report.
Recommendation 7.2.1 Coordinating all

actors
Since integrating large-scale RE requires
many actions at different timescales, levels
of control and points in the generation life
cycle and the tools and infrastructure must
be provided by many actors both public and
private, the IEC recommends governments
and
intergovernmental
organizations
to
take responsibility for uniting all the relevant
stakeholders in a single effort to set the rules,
develop the standards and take the decisions
needed.
Recommendation 7.2.2 Single
framework for connecting and controlling
renewables
The interdependence of the different parts of
any grid with a high proportion of renewables,
such as the renewable sources themselves, the
control centres at various levels and central and
distributed storage, requires one framework
into which the connection rules, pricing
and investment incentives and operational
standards will all fit. The IEC recommends that
such a framework, which will be simultaneously
technical and policy-related and must leave the
necessary room for different policies in different
economies, be worked out internationally under
governmental leadership.
Recommendation 7.2.3 Regulations to
enable integration
The IEC recommends regulators to frame
connection rules and incentives (in pricing and
for investments) in harmony with the framework
called for in Recommendation 7.2.2, so that
solving every different aspect of the problems
of integrating renewables may be encouraged
rather than obstructed by regulations. In
particular regulators should encourage the

implementation of larger balancing areas (fullyconnected grids under central control), without
neglecting local power quality, so as to enable
the concerted operators to reduce average
variability in generation. They should also set
up stable and predictable financial incentives
which make the best technical and public-policy
solutions simultaneously the most attractive
ones financially.

to utilities, industry and
research
Recommendation 7.3.1 Enhanced
transmission as a precondition for
renewables
The IEC recommends that transmission
infrastructures
should
be
developed
appropriately and in time, in cooperation
between utilities and renewable generation
developers, well in advance of any steep rise
in the proportion of renewables. In most cases
the integration of renewables cannot take place
without a corresponding enhancement. UHVAC
and (U)HVDC techniques, where feasible, have
an increasingly wide application.
Recommendation 7.3.2 Stochastic
forecasting
The IEC recommends significant effort to
be put into developing and operating with
stochastic forecasting techniques in addition to
deterministic algorithms, despite their novelty
in the historical context. When combined with
the ability to react fast to forecast errors, they
promise better optimization of the entire park of
generation resources.
107
108
SECTION 7
tions
Conclusions and recommenda
Recommendation 7.3.3 Research for

forecasting and complex modelling
The IEC recommends industry, utilities and
research institutions to develop the renewable
integration scenarios sketched in the present
paper, in particular for forecasting and for
modelling grid behaviour with a view to control
algorithms, and push research forward rapidly
so that experience can be gained and the
algorithms refined.
Recommendation 7.3.4 Research for
cluster connection and control
The IEC recommends industry and researchers
to develop the electronics and the techniques
for active power/frequency control, reactive
power/voltage control and multilevel control for

to the IEC and its committees
Recommendation 7.4.1 Technical
contribution to the RE integration
framework
The MSB recommends the IEC to take an active
part in the development of the framework called
for in Recommendation 7.2.2, cooperating with
governments and international bodies and
taking responsibility for the technical portions.
Recommendation 7.4.2 Rapid progress
in RE integration standards
the whole of a large RE plant cluster (generation

units, plants, substations and the cluster).
The MSB recommends the SMB to implement

the list of future needs given in section 6.3 of
the present paper, paying particular attention to
the harmonization of already existing national or
regional standards under the framework.
Recommendation 7.3.5 Research into

EES
Recommendation 7.4.3 Synergy with

industry associations on RE integration
The IEC recommends industry, research

institutes and utilities to put significant effort into
developing EES so as to support the integration
of large RE systems into electric grids.
The MSB recommends the SMB to encourage

TCs to follow developments at the global
industry level. Many industry associations are
active in this area and produce studies and
position papers which contribute certain views
of the problems. Standardization efforts should
take account of these efforts.
Recommendation 7.3.6 Forecasting the

demand side
The IEC recommends industry, research
institutes and utilities to develop models and
forecasting techniques for the demand side
in order to develop more reliable dispatching
programmes.
References
110
References
[aba11]
Richard Piwko, et al.: A Blast of Activity: Wind Power at the IEEE Power & Energy Society,
IEEE Power & Energy Magazine, 9(6), p26-35, Nov/Dec 2011.
[adr11]
Federal Energy Regulatory Commission: 2010 Assessment of Demand Response and

Advanced Metering, p21, Staff Report, Feb 2011.
[atp11]
Mark Ahlstrom, et al.: Atmospheric Pressure: Weather, Wind Forecasting, and Energy Market
Operations, IEEE Power & Energy Magazine, 9(6), p97-107, Nov/Dec 2011.
[bac11]
Mark G. Lauby, et al.: Balancing Act: NERCs Integration of Variable Generation Task Force
Plans for a Less Predictable Future, IEEE Power & Energy Magazine, 9(6), p75-85, Nov/Dec
2011.
[coc11]
Hannele Holttinen, et al.: Currents of Change: European Experience and Perspectives with
High Wind Penetration Levels, IEEE Power & Energy Magazine, 9(6), p47-59, Nov/Dec 2011.
[cvw11]
Task Force on the Capacity Value of Wind Power, IEEE Power and Energy Society, et al.:
Capacity Value of Wind Power, IEEE Transactions on Power Systems, 26(2), p564-572, May
2011.
[ded10]
Alec Brooks, et al.: Demand Dispatch: Using Real-Time Control of Demand to Help Balance
Generation and Load, IEEE Power & Energy Magazine, 8(3), p20-29, May/June 2010.
[des10]
Scott Coe, et al.: Demanding Standards: Developing Uniformity in Wholesale Demand

Response Communications to Enhance Industry Growth, IEEE Power & Energy Magazine,
8(3), p55-59, May/June 2010.
[dlr05]
German Aerospace Center, et al: Concentrating Solar Power for the Mediterranean Region,
Report, April 2005.
[eck11]
V. Eckert: http://af.reuters.com/article/commoditiesNews/idAFL6E7NT1WK20111229?page
Number=2&virtualBrandChannel=0, Article 2011, Accessed 29 March 2012.
[ede10]
S. You, et al.: Economic Dispatch of Electric Energy Storage with Multi-service Provision,
IPEC 2010.
[eis11]
Douglas Todd: Managing Large-Scale Penetration of Intermittent Renewables, MITEI

Symposium on Managing Large-Scale Penetration of Intermittent Renewables, Cambridge/
U.S.A, 20 April 2011.
[epa10]
European Photovoltaic Association: Global Market Outlook for Photovoltaics until 2014,
Report, May 2010.
[ern11]
Ernst Scholtz: Grid Integration of Renewables: Challenges and Technologies, MITEI

U.S.A, 20 April 2011.
[esp10]
Gobierno de Espaa: Spain's National Renewable Energy Action Plan, Report, 2010.
[ete11]
Yi Zhang, et al.: Overview of the Practice of Renewable Generation Integration and

Transmission Planning in North America, Energy Technology and Economics, 2011, 23(8),
p1-7, 23, in Chinese.
[evf10]
C. Binding, et al.: Electric Vehicle Fleet Integration in the Danish EDISON Project A Virtual
Power Plant on the Island of Bornholm, IBM Research Report RZ 3761 (2010) # 99771.
[ewe05]
The European Wind Energy Association: Large Scale Integration of Wind Energy in The
European Power Supply: Analysis, Issues and Recommendations, Report, Dec 2005.
[ewe11]
The European Wind Energy Association: Wind in Our Sails: The coming of Europes Offshore
Wind Energy Industry, Report, 2011.
[fat11]
Aidan Tuohy, et al. Flexibility Assessment Tool: IEA Grid Integration of Variable Renewables
Project, IEEE Power & Energy Society General Meeting Power System Flexibility Panel, 2011.
[gcr09]
EDISON Consortium: Report: WP2.7 Grid Codes and Regulation Related to EVs, Report,
2009.
[ger10]
Federal Republic of Germany: National Renewable Energy Action Plan in accordance with
Directive 2009/28/EC on the promotion of the use of energy from renewable sources, Report,
2010.
[gre12] Webpage: http://www.greenrhinoenergy.com/solar/technologies/cst_systems.php,

Accessed 27 May 2012.
[gse12]
GSE Group: Osservatorio Statistico, http://www.gse.it/it/Dati%20e%20Bilanci/

Osservatorio%20statistico/Pages/default.aspx, Accessed 18 July 2012.
[gwe10]
Global Wind Energy Council: Global Wind Energy Outlook 2010, Report, October 2010.
[haw06]
D. Hawkins, et al.: Evolving Role of Wind Forecasting in Market Operation at the CAISO,
Power Systems Conference and Exposition, pp. 234-238, Atlanta/U.S.A, 29 October-1
November 2006.
[hey12]
Homer Energy: http://www.homerenergy.com, Accessed 23 July 2012.
[hol09]
H. Hollttinen, et al. Design and operation of power systems with large amounts of wind
power, Final report, IEA Wind Task 25, 2009.
[hsh12]
Hong Shen: Transmission of large scale wind power in China. IEC Workshop on Grid
Integration of Large-capacity Renewable Energy and Large-capacity Energy Storage,
University of Colorado at Boulder, March 2012.
[hts09]
Benjamin Kroposki, et al. Harnessing the Sun: An Overview of Solar Technologies, IEEE
Power & Energy Magazine, 7(3), p22-33, May/June 2009.
[iea09]
S. Inage: Prospects for Large-Scale Energy Storage in Decarbonized Power Grids,

International Energy Agency Working Paper, 2009.
[iea11]
Adam Brown, et al: Renewable Energy: Markets and Prospects by Technology, International
Energy Agency Information Paper, 2011.
[iig12]
Chen-Ching Liu, et al.: Intruders in the Grid: Cyber-vulnerability and Mitigation Studies Using
a SCADA Test Bed. IEEE Power & Energy Magazine, 10(1), p58-66, Jan/Feb 2012.
[jef11]
Jeff St. John: Is Europe Ready for Automated Demand Response?, Greentechmedia, 7 Oct 2011.
[mei10]
P. Meibom, et al.: Advanced Unit Commitment Strategies in the United States Eastern
Interconnection, 9th Annual International Workshop on Large-Scale Integration of Wind
Power into Power Systems as well as on Transmission Networks for Offshore Wind Power
Plants, Quebec/Canada, 18-19 October 2010.
111
112
References
[mev09]
M. Caramanis, et al.: Management of Electric Vehicle Charging to Mitigate Renewable

Generation Intermittency and Distribution Network Congestion, Joint 48th IEEE Conference
on Decision and Control and 28th Chinese Control Conference, 2009.
[mfc11]
Robert Zavadil, et al. Models for Change: Ensuring System Reliability in the Age of Large
Wind Power Plants, IEEE Power & Energy Magazine, 9(6), p88-90, Nov/Dec 2011.
[mit11]
Massachusetts Institute of Technology: The Future of The Electric Grid, Report, 2011.
[mlt10]
G. Corso: Multi-objective long term optimal dispatch of Distributed Energy Resources in

micro-grids, UPEC 2010.
[mma11]
Abraham Ellis, et al.: Model Makers: Simulating PV Systems in Transmission and Distribution
Planning, IEEE Power & Energy Magazine, 9(3), p55-61, May/June 2011.
[msb10]
International Electrotechnical Commission Market Strategy Board: Coping with the Energy
Challenge: The IECs role from 2010 to 2030, White Paper, September 2010.
[msb11]
International Electrotechnical Commission Market Strategy Board: Electrical Energy Storage,

White Paper, December 2011.
[ner09]
North American Electric Reliability Corporation, Integration of Variable Generation Task

Force: Accommodating High Levels of Variable Generation, Report, April 2009.
[ner10]

Force: Flexibility Requirements and Metrics for Variable Generation, Report, Aug 2010.
[npg10]
Northwest Power Grid, State Grid Corporation of China: Power Demand Forecast, Power
and Electricity Balancing, and Load Following Capability Analysis of the Northwest Power
Grid in the 12th Five-year Plan Period, Report, 2010, in Chinese.
[obo12]
1BOG Blog, How Solar Feed-in Tariffs (FiTs) Work, http://solarfinancing.1bog.org/feed-intariffs/, Accessed 13 Feb 2012.
[oma11]
R. Banos, et al.: Optimization methods applied to renewable and sustainable energy: A

Review, Renewable and Sustainable Energy Reviews 15 (2011) p1753-1766.
[opp11]

Force: Operating Practices, Procedures, and Tools, Report, Mar 2011.
[ops08]
P.D. Brown: Optimization of Pumped Storage Capacity in an Isolated Power System with
Large Renewable Penetration, IEEE Transactions on Power Systems 23 (2008), p523-531.
[pao11]
John H Kehler, et al.: Planning and Operational Considerations for Power System Flexibility,
IEEE PES General Meeting Power System Flexibility Panel, 2011.
[pei12]
Richard Piwko, et al.: Penetrating Insights: Lessons Learned from Large-Scale Wind Power
Integration, IEEE Power & Energy Magazine, 10(2), p44-52, Mar/Apr 2012.
[per11]
I. Perez-Arriaga: Managing Large Scale Penetration of Intermittent Renewables, MITEI

U.S.A, 20 April 2011.
[pts11]
Wenuan Li: Probabilistic Transmission System Planning, Book, IEEE-Press, 2011.
[rep11]
Renewable Energy Policy Network for the 21st Century: Renewables 2011 Global Status
Report, Report, July 2011.
[rvp08]
C.C. Wu, et al.: Role and Value of Pumped Storage Units in an Ancillary Services Market
for Isolated Power Systems Simulation in the Taiwan Power System, IEEE Transactions of
Industry Applications, 44(6), Nov/Dec 2008.
[sad11]
Lawrence E. Jones: Strategies and Decision Support System for Integrating Variable
Resources in Control Centers for Reliable Grid Operations: Global Best Practices, Examples
of Excellence and Lessons Learned, Report, Alstom Grid Inc., 2011.
[sco12]
J. Hill, et al.: System constraints on optimal dispatch schedule for battery storage systems,
IEEE PES Innovative Smart Grid Technologies 2012.
[sgc07] Webpage: http://www.sgcc.com.cn/dlkp/dlhb/kpzs/sbdssjbzs/61089.shtml, Accessed 11

March 2012.
[sgc11]
State Grid Corporation of China: Promoting Wind Power Development in China, White Paper,
Apr 2011.
[sgc12]
State Grid Corporation of China: http://www.sgcc.com.cn/ywlm/mediacenter/

corporatenews/04/245999.shtml, Article 2011, Accessed 29 March 2012.
[sge11]
State Grid Energy Research Institute, Vestas Wind Energy Technology (China): Integrated
Solution Strategies for Coordinated Wind Power and Grid Development-International
Experiences and China Practices, Report, Oct 2011.
[sic12]
Julie Hull, et al.: Staying in control: Cybersecurity and the Modern Electric Grid, IEEE Power
& Energy Magazine, 10(1), p41-48, Jan/Feb 2012.
[sin06] Webpage: http://finance.sina.com.cn/roll/20061201/09091074144.shtml, Accessed 15 April

2012.
[sin12]
Ivan M. Dudurych, et al.: Safety in Numbers: Online Security Analysis of Power Grids with
High Wind Penetration, IEEE Power & Energy Magazine, 10(2), p62-70, Mar/Apr 2012.
[smp11]
J. Smith, B. Parsons: Wind Integration: Much Has Changed in Two Years, IEEE Power &Energy
Magazine, 9(6), p18-25, Nov/Dec 2011.
[sra12]

Force: Special Reliability Assessment: Interconnection Requirements for Variable Generation,
Report, Mar 2012.
[tea07]
Mukhopadhyay S.: Towards electricity for all: transmission planning for the growth of India's
power system, IEEE Power & Energy Magazine, 5(3), p71-78, May/June 2007.
[tec10]
C. Quinn, D. Zimmerle, T.H. Bradley: The effect of communication architecture on the

availability, reliability, and economics of plug-in hybrid electric vehicle-to-grid ancillary
services, Journal of Power Sciences 195 (2010), p1500-1509.
[tem12]
Taiwan Electrical and Mechanical Engineering Services (TEMES): www.temes.com.tw/

Item03_EN.htm, Accessed 23 July 2012.
113
114
References
[tgs12]
Mark Reynolds, et al.: The Golden Spike: Advanced Power Electronics Enables Renewable
Development Across NERC Regions, IEEE Power & Energy Magazine, 10(2), p71-78, Mar/
Apr 2012.
[tgy06] Webpage: http://www.sgcc.com.cn/ztzl/tgyzl/tgyzs/default.shtml, Accessed 15 April 2012.

[tie12] Webpage: http://www.3tier.com/en/, Accessed 29 March 2012.
[top12]
Daniel E. Nordell: Terms of Protection: The Many Faces of Smart Grid Security, IEEE Power
& Energy Magazine, 10(1), p18-23, Jan/Feb 2012.
[tpd11]
State Grid Corporation of China: Typical Design of Wind Farm Electrical Systems. China
Electric Power Press, 2011, Book in Chinese.
[tre10]
P. Denholm, E. Ela, B. Kirby, M. Milligan: The Role of Energy Storage with Renewable
Electricity Generation, U.S. National Renewable Energy Laboratory Technical Report NREL/
TP-6A2-47187, Report, 2010.
[trp11]
Eamonn Lannoye, et al.: The Role of Power System Flexibility in Generation Planning, IEEE
PES General Meeting Power System Flexibility Panel, 2011.
[uei10]
The Union of the Electricity Industry: Integrating Intermittent Renewable Sources in the EU
Electricity System by 2020: Challenges and Solutions, Report, 2010.
[vce03]
G. Bathurst, G. Strbac: Value of Combining Energy Storage and Wind in Short-term Energy
and Balancing Markets, Electric Power Systems Research 67 (2003), p1-8.
[vsc10]
Guangfu Tang: VSC-HVDC Based High Voltage DC Transmission Technology, Book, China
Electric Power Press, 2010, in Chinese.
[wei11]
Liping Jiang, et al. Wind Energy in China: Status and Prospects, IEEE Power & Energy
Magazine, 9(6), p36-46, Nov/Dec 2011.
[weo11]
International Energy Agency: World Energy Outlook 2011 OECD/IEA 2011.
[won11]
F. Wong, C. Yixin: http://af.reuters.com/article/energyOilNews/idAFL3E7G5546 20110506,

Article 2011, Accessed 29 March 2012.
[wpa12]
Webpage Wind Powering America: http://www.windpoweringamerica.gov/wind_ installed_

capacity.asp, Accessed 29 March 2012.
[yam11]
M. Yamamoto, O. Ikki: Nation Survey Report of PV Power Applications in Japan, Report,

International Energy Agency, 17 June 2011.
International
Electrotechnical
Commission
3, rue de Varemb
P.O. Box 131
CH-1211 Geneva 20
Switzerland
Tel.: +41 22 919 02 11
Fax: +41 22 919 03 00
info@iec.ch
www.iec.ch
ISBN 978-2-8322-0340-8
CHF 50.-
Copyright IEC 2012. All rights reserved.

IecWP Gridintegrationlargecapacity LR en

Uploaded by

Copyright:

Available Formats

IecWP Gridintegrationlargecapacity LR en

Uploaded by

Document Information

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

IecWP Gridintegrationlargecapacity LR en

Uploaded by

Copyright:

Available Formats

stability power integration battery energy-generation flywheel decarbonization energy

solar variability unpredictability dependency photovoltaic UHVDC control pumped

Grid integration of large-capacity

The present White Paper is the third in a series

that conventional generation facilities also need

This White Paper was written by a project team under

Section 2 RE generation: the present, the future and the integration

2.2 Present status of RE generation and future projections

Section 3 Present: state of the art in integrating large-capacity RE

3.2 RE generation technology

3.3 Transmission technology

3.4 Operational technologies and practices

Section 4 Future: technical solutions for integrating

4.2 Grid-friendly RE generation

4.2.2 Advanced characteristics of RE generating units and plants

4.3 Improved flexibility in conventional generation

4.4 Transmission expansion

4.5 Operational enhancement

4.6 Demand response

Section 5 Application of large-capacity EES to support RE integration

5.2 Promising large-capacity EES technologies

5.3 Roles of EES in RE integration

5.4 Technology needs of large-capacity EES applications

Section 6 Standards for large-capacity RE integration

6.2 Present situation

6.3 Future needs

Section 7 Conclusions and recommendations

7.2 Recommendations addressed to policy-makers and regulators

7.3 Recommendations addressed to utilities, industry and research

Recommendation 7.3.5 Research into EES

Technical and scientific terms

Organizations, institutions and companies

Alberta Electric System Operator

European Wind Energy Association

This report discusses the challenges of

This report, produced by the International

describes the key practices and technologies

five sections with relevant IEC standards

2.1 Drivers of RE development

RE goals: mandates requiring load-serving

The MSB EEE Report notes that CO2 emissions

average energy costs tend to decline over time

2.1.2 Energy security

These projections come from IEAs New Policies

2.1.3 Expanding energy access

As global energy demand increases, RE

Biomass and waste

Figure 2-4 Incremental global renewables-based electricity generation relative to 2009 by

While a world where a majority of electricity

2.2 Present status of RE

RE contribution, or 2.9% of world electricity

Figures 2-5 and 2-6 RE generation globally and by country/region to 2035

Figure 2-6 displays the projected growth of RE

with some notable developments in 2010. We

2.2.1 Wind energy

IEAs New Policies Scenario projects 1282TWh

of a decade, as shown in Figure 2-9 [weo11].

Figure 2-11 Wind energy generation to 2035 by region/country

Wind capacity growth over this period is