235

MANAGEMENT OF TRANSMISSION
CAPACITY AND ACCESS:
IMPACT ON
SYSTEM DEVELOPMENT

Working Group
C1.31

August 2003
 235

MANAGEMENT OF
TRANSMISSION CAPACITY
AND ACCESS:
IMPACT ON SYSTEM
DEVELOPMENT

Working Group
C1.31
Copyright © 2002
Tout détenteur d'une publication CIGRE sur support papier ou électronique n'en possède
qu'un droit d'usage.Sont interdites,sauf accord express du CIGRE, la reproduction totale ou
partielle autre qu'à usage personnel et privé, et toute mise a disposition de tiers, dont la
diffusion sur un réseau intranet ou un réseau d'entreprise.
Copyright © 2002
Ownership of a CIGRE publication, whether in paper form or on electronic support only
infers right of use for personal purposes..Are prohibited, except if explicity agreed by CIGRE,
total or partial reproduction of the publication for use other than personal and transfer to a
third party; hence circulation on any intranet or other company network is forbidden.
 Terms of Reference
A working group was established by Study Committee 37 at its meeting in Paris in
August 2000. The group was to perform the Terms of Reference below and to
report in the last quarter of 2002.

Working Group

Colin Ray UK Convenor

Leslie Bryans UK Replacement Convenor
Luigi Salvaderi Italy
Ivar Glende Norway
Dave Barrie Australia
Ashok Manglick
Roberto Berer
Pierre Pramayon

2
Title of the group: Management of transmission capacity and
access: impact on system development.

Scope, deliverables and proposed time schedule of the Group:

Background:

Advisory Group AG37.1 "Regulation and Planning" was set up in September 1999 to
identify emerging issues related to electricity industry regulation as it impacted on system
development, in particular congestion in transmission and management of available
transmission capacity (ATC). As a result of the scoping work done by AG37.1 on this
topic, working group WG37.31 is to address the issue of evaluation and management of
transmission capacity in the current market orientated environment.

Scope:

The overall aim of WG37.31 is to describe transmission capacity, its evaluation,

management and access. Specifically, measurement or evaluation of available
transmission capacity (ATC), possible allocation methods, and advantages and
deficiencies in the evaluation and allocation methodologies.

1. Define transmission capacity in terms of:-

a) methods for calculating available transmission capacity
b) methods of allocating capacity to market players.
2. Identify the options for managing transmission capacity covering:-
a) allocation timeframe i.e. real time, day ahead, monthly or annual
b) incentives to encourage actual flow matches contracted values and
compensation arrangements where a difference occurs.
3. Assess the impact of management of transmission capacity on system
development:-
a) identify appropriate market economic investment signals in the allocation
process that support new transmission capacity.
b) what are the business risks in providing such capacity (i.e. income)?
c) what allowance should be made in planning for differences between
contracted and actual flow (i.e. capacity margin) and what information does
the planner need on who has been allocated capacity?
d) is new transmission capacity driven purely by the market investment signals
in (a) above or by a combination of security standards and market
investment signals?

Deliverables and time schedule:

Installation of the WG37.31 January 2001
Beginning of the work January 2001
Preliminary report July 2001
Intermediate report January 2002
Final Report August 2002 (CIGRE Session 2002)

3
CONTENTS

1. Transmission Capacity - What is it?

2. Transmission Capacity - How can it be allocated to market players?

3. Management of Transmission Capacity

4. System Development Issues

5. Conclusions

Attachment 1 - Assessing Volumes of Constraints Addressed Prior to Gate

Closure in a Market for Firm Access Rights
Attachment 2 - Options for a Transmission Access Regime
Attachment 3 - Simultaneous Clearing in Access Auctions and Losses
Charges
Attachment 4 - Examples of the Calculation of Imbalance Charges
Attachment 5 - Implementation of a Market in Transmission Access -
Implications for Systems

5. Appendix I - Terms and Definition

6. Appendix II - North American Electric Reliability Council Report - Framework

For determining available transfer capability

7. Appendix III - ETSO - Definitions of Transfer Capacities in Liberalised

Electricity Markets

8. Appendix IV - ETSO - Key Concepts and Definitions for Transmission Access

Products

9. Appendix V - ETSO - Co-ordinated Auctioning - A market based method for

transmission capacity allocation in meshed networks

10. Appendix VI - ETSO - Co-ordinated Use of Power Exchange for Congestion

Management

4
 WG37.31 - DRAFT REPORT
Terms used throughout this report are set out for clarity in
Appendix I.

Introduction - LIBERALISED MARKET ISSUES

The liberalised market in electricity is characterised by:

• Unbundling of production, transportation and
Stakeholders have services;
different interests and • Management roles - Transmission Operators
apply different forces. (TO)/System Operators(SO), Market Operators,
Regulators.
• A range of views, which depend upon the duties or
financial interests of these various stakeholders
(CIGRE Task Force 38.10.15 reported in detail
upon quantifying Stakeholder benefits.)

The following Report shows how different stakeholders

interests may be met in relation to Transmission Capacity.

1. TRANSMISSION CAPACITY - WHAT IS IT?

Generic Issues

1.1 Definitions
The transmission capacity can be considered as
follows:

a) maximum capacity: determined by the construction

of the transmission lines ie the conductor used and
its tension;
A System Operator's b) available capacity: determined by the network that
view of Capacity surrounds the transmission lines ie the thermal
rating may be restricted because of the voltage or
stability considerations;

c) usable capacity: determined by the way power

shares between the transmission lines and the
security standards applied ie n-1 or n-2;

d) contracted capacity: a transmission capacity at a

boundary may be determined through a contract
which may not relate directly to the assets at that
boundary .

5
 Within a large network, the useable transmission
capacity at one point may be determined by any of the
factors in a) - d). The technical transfer limit will be set
at some remote point in the network or even on an
Capacity limits are adjacent network ie a neighbouring utility. Therefore
affected by many calculation of the factors a) - d) requires information
factors in different about the entire surrounding network. Factors a) - d)
domains • have time varying values,
• are determined by the generation pattern, the
demand cycle, the power flow pattern and the
network topology (current operational
configuration and circuits that may be out of
service for maintenance or construction work).

Having determined transmission capacity, it needs to

be allocated to the many market players.

1.2 Stakeholder Views

In a market environment, transmission system users
and transmission owners/operators view transmission
capacity differently.

Users, buy or sell energy across the transmission

system.
For them, transmission capacity is their access to a
liquid energy market and enables them to
Stakeholders would generate/offtake at certain locations or to carry out a
prefer definitions transaction between two specific locations.
which suit their
interests. TO/SOs, provide the physical basis and real time co-
ordination for the energy transactions.
For them, transmission capacity comes from individual
transmission equipment whose contribution to the
overall capacity is calculated strictly according to
physical laws.

By following the physical laws, the descriptions a) to d)

above follow the TO/SO point of view.

The market may seek to adopt different definitions. For

example,
a user-oriented view of maximum capacity could be
"The maximum possible amount that can be generated
or consumed at certain locations." whereas a TO/SO-
oriented view would be conditioned by the thermal
Capacity products ratings of individual circuits.
may bridge the gap
between Such differences in the capacity definition lead to
stakeholders different types of “capacity product” to be traded in
the market and will impact on the way capacity is
managed.

6
 The use of terms may differ depending upon whether
Capacity definitions they are applied to circuits interconnecting stand alone
may need modified power systems or to circuit boundaries within a main
for different types of interconnected system.
network
Depending on the arrangements in a particular market,
the 'usable capacity' may be closely related to the
'contracted capacity'. Furthermore, in relation to
'contracted capacity' it may not just be the system
beyond the interconnector, but the interconnector itself
which is the basis of the contract.

1.3More Explanation of Physical Limits

Thermal Limits
The amount of power which can be transferred across
a boundary on the system is limited by the rating of the
individual circuits and the way in which the power
transfer is shared between them. The 'firm' thermal
Licence Standards capability is taken to be the capability which meets
are the basis for licence standards. Typically standards require supply
both design and to be maintained after the loss of two circuits resulting
operational practice from a fault during a maintenance outage. Some
standards are more onerous e.g. two
contemporaneous or simultaneous faults.. The result
is usually less than the sum of the individual ratings of
the remaining circuits. This is because one of the
remaining circuits reaches its rating limit before others.

Boundary firm capabilities are often published to

advise stakeholders and send locational development
Network transfer signals. Many organisations are required to annually
capacities are produce a Capacity Statement or a fuller document
published like a Seven Year Statement. Other standards (often
a licence requirement) may clarify the duties placed
upon a TO/SO for the design / secure operation of a
network. The following example is drawn from the
United Kingdom and shows a standard applied to
NGC.

The current standard by which the transmission

system is planned and operated ('NGC
A UK example of Transmission System Security and Quality of
how thermal limits Supply Standard, Issue 2, November 2000) is
are in licence lodged with Ofgem and referenced in the
standards Transmission Licence. The standard is commonly
referred to as the Licence Standard.

7
 That part of the Licence Standard which is
concerned with the planning of the main
interconnected transmission system (as opposed
to generation and demand connections), inter alia,
defines a thermal requirement relating to boundary
transfers between any two contiguous parts of the
system. For instance, the boundary thermal
capacity must be equal to a more than the planned
boundary transfer plus half the interconnection
allowance with any two circuits out of service. The
interconnection allowance is an allowance,
specified in the Licence standard, which takes
some account of variations in weather, plant
availability and demand forecasting error either
side of the boundary.

Voltage Capability
At times of winter peak demand it may be necessary
Network voltage to restrict power transfers to a level lower than the firm
can limit transfer thermal capability. This is to ensure that satisfactory
at times of heavy voltages can be maintained in the importing area.
load
During lighter load periods, the voltage issue may not
dominate and firm thermal capability may be applied.

Transient and Steady State Stability

Power transmission capability between two areas or
between a major generating station and the system
can also be limited by considerations of electro-
dynamic stability.

System stability Three stability regimes are usually defined:

is a crucial factor • Transient Stability following a severe
in determining disturbance, for example a network fault. The
network security key issue is usually how quickly faults can be
and therefore detected and removed from the system.
transfer limits • Dynamic stability which concerns the ability of
the active parts of the system to remain in
synchronism after the loss of a major in-feed or
load block. The key issues are the system
inertia, synchronising reactance and the
governor response of individual generating
units.
• Steady State Stability which concerns the
response to small disturbances such as the
normal random load fluctuations. The key
issues relate to any tendency to excite inter-
zone or inter-area power oscillations and
whether the system(s) can be detuned from this
natural frequency of oscillation.

8
 1.4 International / Intra-national Applications

Norway & Nordic Countries

In the Nordic electricity market the term ”boundary
transfer limit” is used to refer to transmission capacity.
Boundary Transfer Transfer limits are set currently for those boundaries
limits used; which normally determine the allowed transfers. This
term combines ”available capacity” and ”usable
n-1 criterion applied; capacity”. The n-1 criterion is normally applied.
Whether thermal rating, voltage stability or dynamic
Thermal, voltage and stability conditions are determining the transmission
dynamic issues capacity may change from boundary to boundary.
assessed When lines are out of service, the boundary transfer
limits are reduced accordingly.

There is one common and open electricity market in

the Nordic area even if the four countries still are
considered as separate control areas. Because of
One common Nordic load/frequency control, the boundary transfer limits
market - 4 countries across the national borders are not fully utilised for the
spot trade at Nord Pool (the Nordic power exchange).
The boundary transfer limits (known as the ”available
Frequency response capacities”) across national borders are reduced to
band protected in links allow a frequency regulation band. For the southern
between areas boundary between Norway and Sweden, for example,
the available capacity is normally 2000 MW and the
regulation band 150 MW. Thus, the ”available
capacity for cross border trade” is normally 1850 MW
at this boundary.

UK
The transmission system is utilised to allow generation
surpluses in one part of the country to supply load in
Intra-country example. other parts of the country where, there is a generation
deficit.

In assessing the capability of the system to meet this

System split into areas; task, the system designer splits the system into
Weakest links predominantly importing or predominantly exporting
examined to determine areas. The circuits connecting such areas together
transfer capacity. tend to constitute the weakest 'links' in the system and
thus reflect the capability of the system to accept bulk
Thermal, voltage and power transfers. These circuits which link areas
stability issues together constitute system boundaries.
examined.
Three factors can limit the capability of the
transmission system to transfer power across a
system boundary, namely thermal ratings, voltage and
stability (outlined above).

9
 North America
The North American Electric Reliability Council have
produced a framework for determining transfer
capabilities of their interconnected transmission networks
for a electricity market. These are attached in Appendix
II.

Several extracts are instructive. NAERC state the

purpose of network capacity as:
• To deliver electric power to customers;
• To provide flexibility for changing system
Capacity purposes conditions;
• To reduce the need for installed generating
capacity
• To allow economic exchange of electric power
among systems.

The document defines

• Total Transfer Capacity TTC as the amount of
electric power that can be transferred over an
interconnected transmission system in a reliable
manner while meeting all of a specific set of
defined pre- and post- contingency system
conditions.
• Transmission Reliability Margin (TRM) as that
amount of transmission transfer capacity
Definitions of ATC, necessary to ensure that the interconnected
TTC,TRM and CBM transmission network is secure under a reasonable
range of uncertainties in system conditions.
• Capacity Benefit Margin (CBM) as that amount of
transmission transfer capability reserved by load
serving entities to ensure access to generation
from interconnected systems to meet generation
reliability requirements.

Curtailability and Recallability are also defined.

Operators post Available Transfer Capacity (ATC) which

is a measure of the transfer capability remaining in the
physical transmission network for further commercial
activity over and above already committed uses.
Calculation of ATC Mathematically, ATC is defined as the TTC less the TRM
less the sum of the existing transmission commitments
(which includes retail customer service) and the CBM.

They further require harmonisation of definitions and

Boundary conditions require a range of boundary transfers to be considered in
for studies arriving at any interface ATC.

10
 2. TRANSMISSION CAPACITY - HOW CAN IT BE
ALLOCATED TO MARKET PLAYERS

There are a number of ways in which transmission

capacity can be managed in the market.

2.1 Capacity Products

Capacity can be thought of as a range of market
products rather than a single product. The main
features which distinguish capacity products are:

a) Firmness:

firm capacity: a transmission user (generation or

Transmission demand connection) can be given a firm
Capacity connection agreement i.e. there is no contractual
products are restriction on the use of the transmission system. If
based upon a a restriction occurs the user receives financial
combination of compensation;
degree of
firmness and the interruptible capacity: this provides access to the
period for which transmission system, but access can be restricted
the transfer is at any time with no financial compensation.
available.
b) Duration:

fixed long term: this provides capacity all hours, all

year;

capacity window: this provides capacity only in a

period defined in the connection agreement ie
seasonal, time of day/week/month.

2.2 Issuing Capacity to the market

There are a number of ways in which capacity can be
released or traded to the market. The principle ways
are described below:
Capacity is made
available to the a) Long Term Allocation:
market as long
term (auction or auction: capacity is offered to market players and
contracts used) sold to the highest bidder;
and short term.
contract: capacity is sold under negotiated terms to
one market player.

11
 b) Short Term Allocation

secondary trading: any unused or unsold capacity

Various market and is made available;
non-market
mechanisms exist to counter trading in energy market: re-despatch of
ensure efficient use of generation to keep flows within available
capacity day by day. transmission capacity (cost carried by TO);

market splitting: separate market and prices

established within available transmission capacity
(cost carried by market players).

2.3 Application of principles

The method adopted to manage transmission
capacity will be a combination of these various items,
you can 'pick and mix' from the various items to buy
the transmission capacity needed, for example you
can buy:

There are many a) interruptible capacity

combinations of with b) windowed profiles
arrangements for long purchased through c) auction
term and short term with d) secondary trading
arrangements
A number of combinations are possible, those chosen
or being considered by each country are shown in
Table 1 (see end of Section 2). The method chosen
may depend up on the 'product' being purchased,
these products would be:

Products will also be entry (generation)

tailored to the user's exit (demand)
primary requirements location to location
to enter, exit or transit flowgate
contract path
boundary physical path

12
 VARIOUS METHODS FOR LIMITED CAPACITY ALLOCATION & CONGESTION MANAGEMENT

Italy-F-S-A-Slo Nordic countries Germany-Denmark PJM-USA UK ETSO

Netherlands-FRG-B SMD proposed by FERC
Market Based Methods 2001 Import: NTC=AAC+ATC “Implicit Auctions” = Various time span NGC examined:- Zonal En/Ex; Nodal En-
Transaction-based AAC=long term contracts Price area Use it or lose it (UIOLI) Ex; Flowgates across
Auctioning allocated Zonal Zonal Ofgem favours full access auctions Coordinated
ATC=subdivision 50%-50% Ofgem Feb 2002 consult: auctioning under
between cross border TSOs Initial allocation of rights; secondary study
Repealed after a recourse to market (run by NGC?) Zonal
the Courts Back away from full access auction

Oct 02: NGC SO Incen. consul

Long term FTR entry/exit rights tradable
Internal Inter area congestions: to hedge short-term risk in force by
price area=implicit auction 1.4.2003.

SO capacity & buy back incentives

Capacity release: RoR >6.25%

Non Transaction-based
Counter-Trading, based on Prevailingly adopted in UK-F HVDC cable
bidded prices “emergency situations” Various time span
using Regulating mkt UIOLI
Security constrained
Central Redispatching Energy Dispatch: nodal
prices + Financial
Transmission Rights
Not Market Based
Transaction based
No more contracts on
1999 HVDC cable within
First come, first served
Nordic market
Contracts on HVDC
Pro-quota Rationing
2001 Import NTC=AAC+ATC cables Nordel-FRG,
Poland,Russia
AAC=long term contracts
allocated
ATC=subdivision 50%-50%
between cross border TSOs
“Antitrust limits” on import total
borders and on each border
Result: huge fragmentation
2002 Import NTC=AAC+ATC
AAC=long term contracts
allocated
ATC=subdivision 50%-50%
between cross border TSOs
“Antitrust limits” on import total
borders and on each border
Limits on single requests, not
differentiated among consumers
Still, excessive fragmentation.
A portion of ATC, allocated with
priority to interruptible
consumers
AAC=Already Allocated Capacity; ATC=Available Transmission Capacity; NTC=Net Transfer Capacity=Total Transfer Capacity (TTC) netted of Transmission Reliability Margin: NTC=TTC-TRM

13
 3. MANAGEMENT OF TRANSMISSION
CAPACITY

3.1 General principles

The approach to the management of transmission
capacity depends on the timeframe being considered:

a) planning:
Planners can augment The transmission planner will strive to make
the electrical system in maximum use of the transmission routes that exist.
the long run; Operators New routes will be avoided, being environmentally
must deliver the market difficult to achieve.
requirements within the
existing capability. The planner can raise thermal ratings by
reconductoring / retensioning, remove voltage or
stability constraints by use of SVC, remove stability
constraints by AVR improvements and improve
sharing across circuits by use of power flow
controllers.

b) operation:
The section below The system operator, on the other hand, has only
demonstrates how, in the useable transmission capacity to work with and
different locations, may be required to operate to a contracted position
various market and regardless of ownership and political boundaries.
non-market approaches
are used to manage
Transmission Capacity. 3.2 International / Intra-national Applications

Norway
Within the Nordic electricity market there is open and
Open access equal access to the network at all levels, i.e. nobody is
Voluntary given any priority when the capacity is a constraint in
disconnectable tariffs any part of the system. There is one exception: End-
within the market area users who are able to shift from electricity to another
fuel may be given a reduced transmission tariff
provided they accept disconnection with a defined
notice period (12 and 2 hours’ notice is presently used
in Norway).

Thus for markets designed like the Nordic electricity

market ”interruptible capacity” exists, but terms like
”firm capacity” and ”capacity window” do not have
meaning.

Interconnection with Regarding connections between the Nordic market

other markets: area and other countries (i.e. Russia, Poland and
Duration used, sold by Germany and in the future the Netherlands and
Auctions and Contracts England) the situation is different. Capacity may be
contracted to market players for different durations,
or/and auctioned at different time intervals. The

14
 capacity between West Denmark (Jutland) and
Germany has been managed by auctions
administered by Eltra and E.ON. Contracts and
auctions are also foreseen for the North Sea
Interconnector between England and Norway (NGC
and Statnett) This development is still under planning.

2 The role of Nord Pool, the Nordic power exchange, is

fundamental in the management of transmission
capacity across the national borders of the Nordic
Flow management - market area. When hourly electricity prices are set in
day ahead by Nord the day-ahead market (elspot market) the available
Pool capacities for trade between the various Nordic
countries are important. If the flows resulting from the
purchase and sales bids exceed any of the available
capacities for cross border trade, as declared by the
TSOs, a second price setting iteration is undertaken to
Market price reduce critical flows to the acceptable level. As a
iterations follow the result, the Elspot price in such situations will be
day ahead capacity different in different parts of the Nordic market area.
declaration. This For instance if the bids from the market players define
results in market a transfer need from Norway to Sweden exceeding the
price splitting. transfer capacity, the Elspot price will turn out lower in
Norway than in Sweden. This management of
transmission capacity is called the ”price area
approach”. In terminology applied within ETSO the
somewhat misleading term ”market splitting” has been
used for this approach. As mentioned, a power
exchange is a prerequisite for this capacity
management approach.

Presently Sweden and Finland constitute one Elspot

area each, Denmark two areas (East and West) and
Norway (two-four areas depending on hydrological
Within sub-markets conditions). To manage capacity constraints occurring
counter-trading and within each Elspot area the national TSO is
re-despatch are intervening by doing countertrade, i.e. a simplified
used. redespatch where generation is increased and
decreased in different parts of the system to avoid
violation of boundary transfer limits.

The financial implications of the two methods used in

Countertrades place the Nordic market are different. Countertrades are an
financial risks with expense on the TSO to achieve necessary regulation
the TSO; price area of generation due to insufficient transmission capacity.
(market splitting) With the price area approach there is no expense for
places risks with the TSO, and the burden is taken by the market
market players. players instead. However, the social cost of
insufficient capacity is then calculated. In Norway the
TSO is directed by the authorities to take into account
this kind of expected social costs due to limited

15
 transmission when measures and costs to relieve
transmission constraints are studied.

In the Nordic countries the licencing authorities have

Security not a licence not imposed any security standard. Instead reliability
standard, but a TSO and security is a responsibility for the national
responsibility transmission system operators (TSOs). In planning
n-1 planning used the n-1 criterion is normally applied, even if a trade off
pragmatic risk / reward between capacity and security is possible in principle.
used in operation. In the operation of peripheral parts of the main
Norwegian grid it is today common to make trade off
considerations between the benefits of high transfers
versus the increased probability of a regional black-
out. In different operating situations the conclusion is
operation at n-0 security. A discussion is going on
whether such trade offs also could be introduced at
the main transmission boundaries. If so, the setting of
transfer limits would be based on a probabilistic
approach (flexible transfer limits).

Italy

There are three main aspects to this market:

• import
Macro areas defined. • capability limits between macro areas
• congestion within the macro areas

New market rules were approved by the Ministry of

Industry on May, 9, 2001. The market is expected to
start at the beginning of 2003.
There are two types of markets, with some
interactions:
Energy and despatch • energy-related, run by the Market Operator
markets separate. (GME)
• despatching related, run by the ISO (GRTN)

A convention is being established for the “overlapping

issues.

Energy- related markets

Compulsory pool
One pool - 2 sided bids Day Ahead Market with System Clearing Price (DAM)
Market splitting in obtained from two sided bids. At the beginning, a one
macro areas busbar system is used, if needed. Market splitting is
used in macro areas .

Adjustment market (AM)

Short term variations Variations of the contracted energy volumes of the
DAM (Day Ahead Market).

16
 Despatching- related markets
Congestion within areas
Reserve (secondary & tertiary)
Ancilliary Balancing (real time market)
services markets
Allocation of ATC for Import Flows
This is a major problem in Italy, since the Country
imports 44.4TWh (+5,4%)out of 297.7TWh total
Energy Statistics demand in 2000. Total consumption: 278,6TWh;
207.6TWh captive; 71TWh eligible;19.2TWh losses.

Allocation of Import capacity

The Authority has the duty to establish rules; the
Italian ISO has to implement them cooperatively.

Net Transmission Capacity is limited: The allocation of

the present Net Transfer Capacity (NTC) of 6000 MW
in Winter, has been always a complex process which
Import capacity involved the Regulator and also the Courts. Long term
limited contracts signed in the past by ENEL for an Already
Allocated Capacity (AAC) of 2600 MW reduce
availability for the free market to the present value of
ATC of 6000-2600 = 3200 MW in Winter on all
borders. The following chronology details the
developments:

• Rule: 180/99 established a 50% repartition

between eligible and captive ones
• 1999: First-come, first-served
• 2000: Auction, not completed for excess of
Chronology request
demonstrates • 2001 allocation procedures
political nature of • Rule 140/00: An explicit auction procedure
capacity established, run by the ISO
allocation • Total ATC auctioned = 50% by Italian ISO;
50% by neighbouring ISOs
• Caps introduced: no more than 10% on all
borders; no more than 20% on each border
• First auction run on September for
October, November, December 2000.
• High resulting prices triggered a recourse
by some eligible consumers against the
auction procedure.
• After various legal steps, the Rule 140/00
was abolished. The Regulator issued on
December 6 ,2000 a new ruling (219/00):

¾ Total capacity assigned on a

proportional basis, free of charge

17
 ¾ 80%of ATC assigned on yearly basis,
20% on monthly basis two new caps: no
more than 5% on all borders; no more
than 10% on each border
¾ The trading of “rights of access” after
their assignment was allowed.

• First auction with the new approach

completed on December 19,2000 for 2001
allocation procedure:.
• Dramatic increase of the total number of
competitors (12 GW!! requested compared
with 1510 MW available (1200 MW yearly
and 310 for the months of January &
February)
• For the yearly capacity: 422 requests, 353
accepted, average assignment: 3.4 MW!
• The other 50% (1510 MW) to be assigned
by direct bilateral contracts Foreign
Producers-Eligible consumers
Recent actions carried on by the Italian ISO to
increase the import capacity
• Revamping of same components in
Planners have interconnections lines: + 300 MW
developed • Substitutions of conductors in two
increased capacity interconnection lines: + 200 MW
• New dc link with Greece (2002): +700 MW
• Studies for the application of FACTs: tests
possibly in 2003

2002 Allocation
The Italian network is strongly meshed on the
borders with those of France and Switzerland and
many loop flows are present. Common studies
were made in 2001 by the Italian TSO with
neighbouring TSOs of France and Switzerland,
which resulted in a common definition of the total
North-West Border. The allocation of the capacity
for the year 2002 was made, according the
Decree 301/01 of December 5th 2001 of the
Authority, with the following rules:
- The total winter NTC = 6000 MW
was subdivided in two borders:
North-West (France-Switzerland):
5400 MW and North West Border =
600 MW;
- The total NTC on the two borders
was further subdivided: -i) North
West: 2600 MW Italy-France; - 2800
MW Switzerland; - ii) North-East: -
220 MW Austria; -380 MW Slovenia.

18
 - On the North-West borders the long
term contracts were allocated, 1800
MW with France and 800 MW with
Switzerland. The remaining ATC for
the free market correspondingly
was: France-Italy= 800 MW;
Switzerland Italy=2000 MW, Austria-
Italy= 220 MW; Slovenia-Italy= 380
MW.
- An agreement was established with
France for the common allocation of
the ATC of 800 MW; for various
reasons a similar agreement was not
possible with the other countries.
Consequently, 50% of each ATC
was allocated by the Italian TSO
(GRTN) while the other 50% was
allocated directly by the foreign
TSOs.
- Correspondingly the annual capacity
allocated for the year 2002 directly
by the Italian TSO was: i) 800 MW +
1000 MW (50%) for the border with
Switzerland commonly allocated with
the French RTE; - ii) 110 MW (50%)
on the border with Austria; -iii) 190
MW (50%) with Slovenia
respectively. The allocation criteria
was pro-quota, under the “antitrust
limits” of 10% on each border and
the further constraint of at least 3
MW of request.
- Priority assignments to the
instantaneously interruptible
consumers were ruled, respectively
500 MW on the North-West border
(47 clients) out of the total ATC of
1800 MW under the responsibility of
GRTN+RTE and 100 MW (25
clients) on the North-East border out
of the total 300 MW under the
responsibility of the Italian GRTN.
- A secondary market for the
exchange of the allocated capacity
rights was also allowed
- In the allocation procedure for the
year 2003 also the DC link with
Greece, with NTC of 500 MW will be
taken into account

19
 Generally speaking, the rationale and the
results of the present allocation were still strongly
debated. On one side it was underlined by various sector
of the Industry that the “undifferentiated limits” for eligible
clients and traders caused a competitive rush between
clients, traders, brokers and Distributors with negative
effects. The results were: - contractual weakness of the
demand against the foreign producers:- absence of
contractual skinless of the minor clients:- tendence to
overcontracting in respect of the need in order to
speculate. All these concurring causes, and the related
transaction costs, caused an increase of the selling
prices, paid by the Italian consumers A suggestion is that
the “antitrust limits” should be differentiated for traders
and clients (equal to their consumption): the speculation
stemming from the reselling of not utilizable capacity and
the creation of many “sham” companies to avoid the
existing limits could be avoided. The direct assignment of
the 50% “in the hands” of the foreign TSO should also be
revised , with the aim of obtain a greater transparency
and to avoid that some foreign TSO not unbundled by the
production activity could have competitive advantages.
The absence of reciprocity between Italy and some
foreign countries, and the related impossibility of the
Italian players to enter the related markets, has altered a
level play ground.
New rules for the allocation of the ATC capacity for the
year 2003 are presently s expected soon.
Interruptible supplies
Agreements are being formed with some interruptible
and inter-TSO co- consumers aimed at “relaxing” normal capability limits.
Studies are proceeding to determine how better
operation allow better coordination with neighbouring TSOs (France,
use of network capacity Switzerland) can be achieved. This implies a common
ATC definition.

Capability limits between macro areas

The ISO is to determine and post each day maximum
capability limits between the macro areas, by taking
into account the security constraints. The areas so far
considered are :
Abroad-North;
North –North/Center;
Daily macro-limits North Center-Center,
posted Center –South,
Mainland –Sicily,
Mainland-Sardinia

20
Market operator to The Market Operator is to run a busbar system. When
run busbar system capability limits between areas are not respected,
and market splitting market splitting will be required with various SCPs in
each area.

Revenues stemming from [transits * ∆SCPs] are

accumulated by ISO in a fund, the utilization of which
will be decided by the Regulator (f.i: funding of new
Various markets - interconnections lines; rebate on transmission tariffs,
different etc)
operators The Regulator is studying a possible approach of
Fixed Transmission Rights for the allocation of the
capability between areas.

The “full zonal market model” was criticized due to the

unbalance, in presence of congestions, of higher zonal
prices paid by consumers sited in zones with
generators having high costs. A smooth transition was
deemed more equitable. Accordingly the model has
been revised in the Instruction of the Market
transmitted by the Market Operator GME to the
Ministry of the Productive Activities (Industry) on
January, 18, 2002 in order to allow - for a transient
period of three years- a nation wide unique purchase
prices, p*, paid by all the consumers. To the
Producers different zonal System Clearing Prices will
be paid. The price p* is obviously a value between the
zonal prices, but –due to the fact that the demand is
price-elastic- a feedback on the dispatched load is
present and p* it is not equal to the simple weighted
average value of the zonal prices. The ruling has
therefore entailed the need for the Italian ISO GRTN of
implementing a suitable optimisation tool which , which
– summarizing- ensures that the unique nation-wide
price p* paid by all consumers is capable of
respecting two conditions:
- a “cost recovery” rule, which
ensures that the revenues of the
generators, paid with the zonal
SCPs, will be equal to the costs
charged to the consumers paying
the unique nation-wide price p* ;
- a so called “ no surprise” rule, which
ensures that only the loads with a
bidding price higher than the
resulting unique nation-wide price
p* will be dispatched.

The method, which entails a cross subsidization between

consumers, offers the possibilities to some gaming which
can end to results “opposite” to the aim with the uniform

21
 price is targeting. The limitation of this “gaming
possibilities” is under consideration.

Congestion management - intra areas

Congestion within each area will be handled with a
specific “Congestion market”, run after the Day Ahead
Market (DAM) and the “Adjustment Market” (AM) by
the Market Operator.
The ISO is to implement needed despatching actions.

Bids will be taken from Sellers (generators willing to

Bidding Process increase production in respect contacted values in AM
and consumers willing to resell some of the contracted
energy) and Buyers (generators willing to decrease
their contracts and consumers willing to increase their
engagements) are foreseen.

Payment The payments will be made on “pay as bid” principle.

Principle - Pay as The overall cost is to be distributed on all the network
bid users through an uplift. The ISO is to handle
payments, according to rules to be established by the
Regulator.

European Transmission System Operators (ETSO)

Work proceeding In Continental Europe, ongoing efforts are strong to

on: increase cross border trade. There is work within
Pricing mechanisms ETSO to elaborate on pricing mechanisms for cross
TSO compensation border trade. TSOs are to be compensated for
Congestion
Management
transits across their systems. Management of
congestion is another topic in this context. Within
ETSO, the focus is on the following methods for
congestion management:
• Market splitting (price area approach)
• Capacity auctions
• Coordinated redespatch (coordinated between two
or more TSOs)

ETSO have agreed a number of definitions to ensure

Definitions clear communication between the member countries,
agreed these are set out in Appendix IV and V. The Appendix
IV relates to definitions for transfer capacities and the
Appendix V discusses the transmission
access/products that were described earlier in this
report in paragraph 4.

22
 The transmission capacity of interconnecting circuits that
join two systems in Europe is often allocated to users by
Highly meshed auction. Extending this to a highly meshed system
networks involving many control areas such as continental Europe
is more complex. ETSO have proposed some methods of
allocation of transmission capacity in meshed networks,
these proposals are set out in Appendix VI.

Where transmission limits the energy market causing

congestion, methods to manage such congestion need to
be established and agreed. ETSO have addressed this
issue and their findings are expressed in Appendix VII.

England and Wales

Transmission Access and Pricing

The Master Connection and Use of System Agreement
(MCUSA) and associated documentation have been
amended to form the Connection and Use of System
Code (CUSC) and associated documentation. The CUSC
Access controlled by condifies the generic provisions of the former MCUSA and
agreements Supplemental Agreements but includes new Governance
and Dispute Regulation Procedures.

Charging Statements are produced in accordance with

the requirements of the Transmission Licence. Whereas
the contractual obligation to pay charges resides within
the CUSC, the principles that underpin these charges are
contained in the Charging Statements.

It is a requirement of the Transmission Licence that NGC

charge in accordance with the above Statements. The
Statements contain sufficient detail to enable customers
Licence sets charging to make a reasonable estimate of their charges. The
rules documents are kept under constant review and any
charging disputes are referred to Ofgem.

Connection Charges
These charges relate to the costs involved in providing
Charging is split assets which afford connection to the transmissions
between : system. The charges are based on the connection assets
Connection charges provided, their capital and maintenance costs and a
Use of System Charges reasonable rate of return.

Transmission Network Use of System Charges (TNUoS)

The Transmission Network Use of System (TNUoS)
charges reflect the costs of installing, operating and
maintaining the transmission system. These activities,
which are undertaken in accordance with standards
prescribed in Condition 12 of the Transmission Licence,
provide for the bulk transfer of electricity between

23
 connection sites and also provide transmission system
security and quality of supply.

Under NGC's Licence Condition 12, NGC is required to

plan, develop and operate its transmission system in
accordance with security standards and the Grid Code.
These security standards set out the criteria and
methodologies relating to the planning and operation of
Generation Connections, Demand Connections and the
Main Interconnected Transmission System (MITS).

Specified conditions For Generation Connections, the security standards

for Generation and load specify the maximum loss of infeed power acceptable in
connections the event of certain contingencies, including busbar
failures, single and double circuit transmission failures
and generation circuit failures. For Demand Connections,
the standards specify acceptable levels of: loss of supply
capacity; overloading of transmission equipment;
deviation from voltage limits; and system stability in the
event of a set of contingencies based on transmission
circuit and busbar failures. The security standards for the
MITS, specify that the system must be able to maintain a
Planned Transfer Capability in the event of a failure of a
Security Standards double circuit overhead line, a section of busbar or the
detail simultaneous failure of up to two transmission circuits.
The security standard allows for derogation from the
deterministic standard on economic grounds or at the
request of system users. Derogation from the
deterministic standard for the MITS can be made on the
basis of a cost benefit analysis in consultation with the
Regulator.

Potential Development of Access Auctions

Consideration is being given to how new transmission
access and pricing arrangements might be developed
towards a structure consistent with the new electricity
trading arrangements (NETA). Under such
In the future allocation arrangements, participants would be required firm entry
might be in 2 stages - rights to inject electricity on to the transmission system.
wholesale, then re-sold Market mechanisms would be set up for the initial
retail. TSO can buy allocation and secondary trading of the access rights.
back or sell additional National Grid would buy-back and/or sell additional
in the retail market access rights in order to resolve transmission constraints.
Since the publication of a consultation document, the
Regulator (Ofgem) has held discussions with the industry
and has been further developing the thinking on these
issues. They have indicated that the way forward would
be outlined in a further consultation document, a
document was published in May 2001 for consultation
with market participants on the appropriate way to deal
with transmission access and losses under NETA. The

24
 document is very long and can be viewed at
More information www.ofgem.gov.uk. Attachments to this document
contributed by National Grid are relevant to this working
group and are attached as input information, they are:

Attachment 1: Assessing Volumes of Constraints

Addressed Prior to Gate Closure In a
Market for Firm Access rights
Attachment 2: Options for a Transmission Access
Regime
Attachment 3: Simultaneous Clearing in Access
Auctions and Losses Charges
Attachment 4: Examples of the Calculation of
Imbalance Charges
Attachment 5: Implementation of a Market In
Transmission Access – Implications
for Systems

North America

1. The Northeast Power Co-ordinating Council has

produced a report with a procedure for managing
See Appendix for more transmission capacity, this is attached in Appendix VII,
It is designed to help co-ordination and transparency
of transfer capacity posting across five areas. NPCC
has adopted a regional procedure to post ATC on an
interface basis, an approach that is similar to the rated
path method described in the NERC Available Transfer
Capability Definition and Determination Framework
Document. The key to success is to ensure that
Available Transfer Capacity is calculated using the
ATC to be calculated same definitions and the same calculation method by all
on the same basis by parties. It is also crucial to recognise that the evaluation
all of an ATC is an inerative process, since boundary
ATC's depend on flows on other boundaries. Thus
setting one ATC affects others. In essence, the
guideline suggests that the Areas which implement the
last changes to the transmission facilities at the
interfaces will observe restrictions on the other affected
interfaces, to the extent governed by existing operating
Capacity Benefit agreements among the concerned parties.
Margin The Capacity Benefit Margin gradually reduces from the
planning time frame to the operational time frame.
Transmission NPCC continues to apply the Transmission Reliably
Reliability Margin Margin to cover operating uncertainties including the
activation of operating reserve within an Area, and of
Data exchange Shared Operating Reserve among Areas.
Data exchange between SO's is required and specified.

25
 4. IMPACT ON SYSTEM DEVELOPMENT

4.1 Operational Time frame

Since physical networks are fixed for the purposes
Operational timeframe: of real time operation, operators are faced with
Software tools required trying to meet market expectations by a mixture of
- some well developed flow management and capacity products. Thus
real time developments are likely in the
approaches to and tools for management of flows
and management of capacity products.

Flow Management
Flow Management involves re-despatch to
optimise transfer capacities while respecting
Security constrained constraints. Dispatch constrained optimisation
Optimal power Flow tools are available commercially. They tend to be
tools based around Optimal Power Flow algorithms.
They may need to be dynamically linked to market
information if a value optimisation as opposed to
flow optimisation is required.

The difficulty is that a multi-objective optimisation

may be required. E.g. It is likely that the MO or
TO will require a flow optimisation which disturbs
the existing market nominations as little as
Multi-objective possible so as to minimise any constraint burden.
optimisation may be These form 2 objectives. There may also be
required priority trades resulting from "Take or Pay"
contracts or long term capacity contracts.

Developments will need to recognise the

complications of the interaction between market
mechanisms and flow optimisation if value
Market mechanisms optimisation is to facilitated at source, rather than
and load flows inter- placing total reliance upon a secondary market
related which may have less than perfect knowledge.

Capacity Products
It is argued here that capacity products offer some
scope for optimising the markets use of the
network, by allowing network users to pay a price,
which is acceptable for the firmness and duration
they require. In the ultimate, this would optimise
network use for each settlement period.

Several factors can be taken into account in

product creation :

26
 • The network's limitations;
Factors affecting • The market requirements
development of o Firmness
network products o Duration
• Administrative ability / cost
• Transparency of the offer
• Risk profile
o Who will carry the risk?
o How will it be managed?

Technical Limitations
Many of the above have been dealt with
extensively within the brochure.

Studies determine the network limitations and this

is a standard technical procedure. Tools and
techniques are adequate, although real time
Stability monitors may stability monitors may now be required as networks
be required are driven closer to the verge of voltage and
dynamic stability limits.

Market Requirements
Determining market requirements is market
specific. If auctions or allocation mechanisms are
over-subscribed then there is an indication that, AT
Stakeholders involved THIS TIME, there is a market requirement. There
in determining market is a stakeholder questioning and iterative process
requirements to determine what range of capacity products offer
best value to the market. This may involve
removal of some firm or full-time products and
replacement with a wider range of options.

Administrative ability /cost & Transparency

The more extensive the range of products, the
more complex the set of market rules need to be to
ensure equity and transparency. Each time there
is a change it rules for one product it may have
Capacity product knock on effects on others. E.g. some products
complexity leads to IT may be ramped up in a settlement period before
complexity thus risk others. Changes in the type or rules for products
and cost can affect the entire market system. From above,
there is a need to determine the value of products
from stakeholders and to review this periodically.
Such changes are therefore likely. Complex
systems are prone to errors are complicated to
change and IT intensive. These factors will need
to be weighed in the equation of allowable
complexity v market benefit. In the final analysis,

27
 any system too complex to be readily and
frequently audited, or which does not allow clear
market signals is unlikely to be useful. There is a
sense in which it is meaningful to fix a term within
which the market rules will remains substantially
fixed, else there is a never ending development
and testing of evolving systems with many
operational failures.

4.2 Planning
How and to what extent do planners look at
markets and market trends as the basis of
developments?

To what extent should developers, particularly

interconnectors joining two independent systems,
be left to entrepreneurs where opportunities exist?

Risk Profile
The range of products offered to the market and
their conditions of offer determine the level of risk
and the risk profile of each stakeholder.

Factors affecting risk Some factors which augment risk are:

• Reduction in network standards to increase

flows
• Market demands inhibiting network
maintenance
• Counter flows being considered according to
a superposition principle.
Super-position not a
network risk The first two factors above effect network
performance and therefore ultimately market
performance, the latter only effects market
performance.

Risk allocation is a matter of philosophy which may

depend upon the maturity and confidence of the
Risk allocation may market. In markets that are immature it may be
depend upon the stage inappropriate to force too much risk on the market
of market development players, otherwise the cost of serving the risk
unless it can be netted off, will be a multiple of that
achieved by a "perfect knowledge" market
manager. In more mature markets, the risks are
better understood and are likely to be transferable
to market players. The risk sharing may also
depend upon the need to attract players into the
market.

28
 It therefore follows that there is no uniquely good
network / market risk mechanism. In the end, most
Risks placed with those economists would concur that risks are best carried
who can control them by those who have authority to manage them. In
this way it might be argued that market risks are
carried by market players and network risks by
network managers.
Better understanding
needed There is a need to better understand the effect of
risk transfer on the efficient of capacity allocation.

4.2 Planning Timeframes

The issue for planners is best illustrated by

reference to interconnectors. Should networks
Should interconnector serve interconnector capacity, or interconnector
serve networks or visa capacity serve network need? If the former, what
versa is the requirement for interconnector capacity,
perhaps 10 years ahead, and what are the factors
which may affect this? Clearly the nature, location
and price of generation is a major factor.

Work on uncertainty Work is needed to identify how this uncertainty can

required be quantified.

Whether the planner is dealing with intra-country

ties or international interconnectors is largely
Intra-country, the same irrelevant, except that factor endowment
differences are more pronounced in the
international arena. This leads to a greater scope
for price differences over time.

In any case, there are various market players who

Planners should use will have a view, albeit biased, on the value of
balanced stakeholder additional capacity. It is the task of the planner to
network demand V attempt to determine a balanced stakeholder
price demand v price profile for removal of constraints.

Who then should carry the risk of the

developments? There are three basic options and
many variations on these:
• Network users carry the risk, which is
passed on to final customers; this is
Who carries the risk of effectively a regulated asset development;
development • The risk is carried by market players who
are prepared to fund the cost of extending
network capacity to facilitate trades;

29
 • The risk is carried by an asset developers who
needs to form a "bankable" project and obtain
funds;

Risk spread among Network User's Risk

native customers when Where the risks are beyond that which a private
too great for private developer in the asset creation market is likely to
developers take, regulators need to weigh the total benefits
from the flexibility / security as well as price offered
the capacity augmentation. In certain cases these
factors will be strong enough to allow projects to be
justified. Inter-area negotiations between SOs
MOs and Regulators will be necessary for
international augmentation.
Market plays may
ultimately pay for some Market Players Risk
links It is uncommon to find market players funding the
transfer augmentation in the first instance, although
in some cases capacity holders pay over time.

Private Developers Risk

Private asset developers need a reasonable
Contracts for Use certainty of flows of funds over a long period, if
required for "bankable they are to produce a "bankable" project. Where a
projects". Balance project is not "bankable", it could still be carried out
sheet projects may as an on-balance sheet activity, but this exposes
accept lesser risk the developer to higher financing costs and greater
management but are risks. It is usual for a developer to attempt to
more expensive to fund secure contracts for capacity in advance of the
development in order to demonstrate that the
income side risks have been managed. With other
market uncertainties, this may prove difficult. Such
contracts will certainly be required for "bankable"
projects. Balance sheet projects may be content
with projections of market trends, especially where
there is a demonstrable shortage in one area and a
demonstrable surplus in the linked area. A further
question arises as to whether there are other
developments which might strand the asset.
These could be:
• Additional network links
• A change in the generation location profile
• Loss of large load blocks from the flow-to
area.
Stranded asset creation The former is a particular risk for developers who
a real risk seek to alleviate a constraint which has occurred
because another development has been caught in
an environmental debate. If the private developer
should succeed and the original development be

30
ultimately successful, one or other may become
stranded assets.

31
5. Conclusions

• Since liberalisation and unbundling, there is a range of stakeholders

who have different expectations of network capacity.

TO / SOs feel constrained by the obligation to maintain a level of

network security. They consider the network performance against a
set of security standards, taking account of thermal, voltage and
stability considerations.

Market players expect that the network will facilitate their maximum
trade levels.

• Network capacity has often been seen as firm or pseudo-firm for each
settlement period. A range of products may offer a better opportunity
to extend effective capacity and meet market needs. Products will be
based upon a combination of firmness and duration. There is a range
of mechanisms for delivering products into the market place. The most
common mechanisms are auction and contract. Products could have a
range of life spans.

• Tools may be required to better integrate the market and system

operator principles to optimise value for stakeholders.

• There may be some limit to the complexity in capacity products which

can be safely managed.

• Work is required to determine how uncertainty in generation is to be

dealt with in planning capacity augmentation.

• Investment opportunities in capacity augmentation are linked to risk.

Investment will only be undertaken by the private sector when risks can
be quantified and managed.

32
ATTACHMENTS 1 - 5

33
Attachment 1: Assessing Volumes of Constraints Addressed Prior to Gate Closure
in a Market for Firm Access rights

The following paper describes NGC’s modelling of the volume of constraints it might be
possible to capture under a market in firm tradable access rights.

Introduction

In line with the trade-offs outlined in the main text, indicative studies have been carried out
in order to assess the effectiveness of a transmission access market at addressing
constraints prior to Gate Closure hence removing the need to address them in the
Balancing Mechanism (BM).

The studies have been limited to 6 snapshots of the system involving different
assumptions on the dispersion of generation and demand, participants' bidding strategy
and technical characteristics of the transmission network. In addition, the studies are
based on some idealised assumptions and approximations and hence their results can be
interpreted as at the 'best end' of a likely spectrum. In reality once these assumptions and
approximations are no longer valid, the effectiveness of the access market may be
reduced from results presented here.

These studies do not show the extent to which the total constraint costs seen by NGC will
be saved as a result of the new transmission access market. This will be a function of how
much cheaper constraints can be addressed in the forward access markets against being
addressed through the Balancing Mechanism. Therefore it will depend on the liquidity of
access zones, participants' valuation of risk and transparency of the market. It is
extremely difficult to estimate the effect of this without prior knowledge / data from
operation of NETA. In addition, the costs of 'constraints' addressed will depend on the
volume of access tickets sold in the Primary Auction. If a 'maximum' method is used,
participants may get rights above those that may be present without a transmission access
market and hence create additional costs to NGC of buying back rights in the secondary
markets. Conversely a 'minimum' method may depress the cost of constraints seen by
NGC to be addressed.

This note first describes the study method used in the assessment exercise. It then
describes an initial “Straw Man” design of the market based on the criteria and/or
preferences set out in Ofgem’s December Consultation Document, followed by a
discussion of the assessment results of this design.

Straw Man Model Based on the December Document

Ofgem indicated in their December Consultation Document that the transmission access
market should be a two-sided and competitive market of firm access rights with
unfacilitated trading. In response to that document, NGC constructed a Straw Man
model under these criteria in order to assess the effectiveness and to scope the
implementation of the access market.

This market model is based on an entry/exit right market design – with participants buying
rights to inject power onto the transmission system (entry rights) and rights to take power
off the system (exit rights) in given locations, both with reference to the same defined hub-

34
point. These locations are defined by zones – each entry / exit point within a given zone is
treated equally. To accommodate participation from the demand side under the restriction
of the 1998 settlement metering arrangement, the following zonal definitions were
examined:

• 12 GSP Group Areas (matching the 1998 settlement metering arrangements), and
• three different variants of 6 Supra GSP zones derived by amalgamating some of the 12
GSP groups based on the need for increased liquidity and/or defining sensible
boundary constraint numbers.

Figure 1 shows these zonal definitions which are fixed for the whole year.

Figure 1 - Transmission Access Zonal Definition Based on GSP Groups

12 GSP 6 Supra GSP [1] 6 Supra GSP [2] 6 Supra GSP [3]

The transmission access market is assumed to comprise a Primary Auction of access

rights and then secondary trading of these rights. Before the beginning of the Primary
Auction, NGC agrees with Ofgem the absolute transfer capability of each boundary
between zones, and makes these capabilities public. An auctioneer then facilitates a
Primary Auction process for zonal rights for the forthcoming year, in which participants
submit offer curves to buy or sell “bundles” of entry and exit rights within each zone.

The auctioneer operates a simultaneous clearing process across zonal access products,
which determines clearing price and optimal volumetric allocations to participants based
on their bids and the agreed transfer capabilities. The Primary Auction is subject to a set
of relative loss adjustment factors, which set a minimum differential between access
product prices to reflect the cost of unconstrained marginal losses.

Auction prices for each access right product are published. Surplus funds raised by this
auction process partially support NGC’s TNUoS revenue.

Following this Primary Auction, participants fine tune their position by trading these rights
on secondary markets (potentially down to the half hourly level). Via a Designated
Exchange, NGC participates in secondary trading to the extent that its expectations of the
inter-zonal boundary capabilities change – it is incentivised to maximise the revenue of
any release of new right, and minimise the cost of any buy-back.1 Clearly it will not be

1
In the assessment studies because it was assumed that the perfect total volume of tickets was sold in every zone, no
such buying or selling needed to be modelled. The trading amongst the participants in a zone was simulated by re-
distributing the tickets to generators strictly in order of their bid prices. This keeps the same amount of tickets in each
zone but gives a different nodal allocation.

35
possible to resolve intra-zonal constraints via this market – these will be left to the
Balancing Mechanism.

The access markets “close” at Gate Closure. The Balancing Mechanism continues as at
NETA Day 1 – effectively, NGC buys / sells a bundled access and energy product.

In each half hour, participants have a volumetric access imbalance exposure equal to the
difference of their access right holdings in a given zone (adjusted for accepted BM
bids/offers) and their metered output/demand in that zone. These imbalances are settled
at a price derived (at least in part) from closing access prices in the secondary markets.

Study Methodology

Underlying Assumptions and Approximations

The assumptions and approximations made are as follows:

• Studies are based on a DC model hence approximate the constraints.
• With the market being two-sided the studies only deal with generation bidding for
transmission access in the Primary Auction and then adjusting their position in
secondary trading (i.e. the pattern of demand and total zonal demand is assumed
constant throughout the market and in the Balancing Mechanism).
• NGC can perfectly predict the system constraints in line with the optimum nodal
solution, so that a perfect “Best Estimate” can be made on the conditions governing the
volumes of tickets available. It is further assumed that volumes of tickets sold in the
Primary Auction reflect the “Best Estimate” conditions.

Input Data

For each of the six studies, one year’s system and market operation was modelled, with
each week represented by a single demand block with its own demand level and available
generation pattern. Typical transmission availability was also included in the model.
Sensitivities on generation bids were determined with reference to recent Pool bidding
behaviour.

The temporal duration of the access rights was assumed to line up with the smallest unit of
the simulation, i.e. one week, with each unit’s rights being auctioned and traded as
separate and independent products.

Primary Auction Simulation

For each temporal unit, a nodal allocation of transmission access was first derived from an
optimisation algorithm which maximised the income from nodal auction bids while
respecting all transmission constraints. This is equivalent to assuming that a perfect “Best
Estimate” can be made on the conditions governing the volumes of rights available.
Depending on the definition of the access rights (e.g. whether entry/exit or transfer type),
this solution was converted to the appropriate volumes of rights sold, based on the
assumption that the volumes of tickets sold in the Primary Auction perfectly reflect the
“Best Estimate” conditions.

This simulation also gave the total system constraint volume.

36
Secondary Trading Simulation

In Secondary Trading, tickets of the same type can be traded amongst participants. The
System Operator would not need to do any trading as the Primary Auction has already
sold the perfect total volumes under the assumptions made for these studies. Assuming
this process to be perfectly efficient, it would result in the rights of a particular type being
held in accordance with an unconstrained merit order.

Calculation of Constraints left to be addressed in the BM

With the assumptions of a similar merit order of bids and offers in the BM and that Final
Physical Notifications exactly match access rights held, the optimum BM solution would be
the same as that in the Primary Auction simulation.2 The BM action at each node would
be defined as the difference between the rights allocated in the Secondary Trading and the
BM solution. The sum of the absolute differences at all the nodes gave the total BM action
volume to address remaining constraints.

Effectiveness Assessment

The results of the estimated level of constraints addressed before Gate Closure on the 6
snapshots are shown in the following table.

Table 1 - Volume of Constraints Addressed by A Transmission Access

Market Prior to Gate Closure (% of Total Constraint Volume)

Snapshot Total Supra Supra Supra

of Constraint 12 GSP GSP [1] GSP [2] GSP [3]
System Volume (TWh)
A 6.1 38% 29% 17% 33%
B 4.2 67% 37% 55% 54%
C 3.3 53% 35% 32% 43%
D 2.3 66% 43% 43% 60%
E 1.8 63% 39% 38% 57%
F 1.0 64% 44% 49% 49%

These results suggest that:

• Up to a maximum of two-thirds of the total constraint volume is likely to be addressed

prior to Gate Closure in an access market based on the full 12 GSP Group (REC)
zones.

• The scope for addressing constraints prior to Gate Closure would be reduced to below
half of the total constraints volume if the requirement of market liquidity leads to the use
of bigger zones (i.e. the combination of GSP groups into supra-GSP groups).
• Under certain scenarios the capture of constraints is much less than half. I.e.
significant intra-zonal constraints may be present.
2
Note that whilst this would approximately hold for volumes of constraints, it is less likely to hold for the costs as it is
highly unlikely that the bid prices in the forward contracts will be equivalent to bids/ offers in the Balancing Mechanism.
Hence the studies have been restricted to volumes addressed and not costs.

37
These results are a “top estimate” as a number of the assumptions and approximations
made in the studies, once relaxed, would be likely to bring the estimates down.

38
Attachment 2: Options for a Transmission Access Regime

Straw Man Models of Transmission access Arrangements

This note summarises three Straw Man models of transmission access market
arrangements which NGC have considered, and provides a brief analysis of the possible
advantages and disadvantages of each.

The three Straw Man models considered are:

• Straw Man 1: Zonal entry/exit market;

• Straw Man 2: Nodal entry/exit market; and

• Straw Man 3: Flowgate market with nodal participation factors. (Note: could also be
termed as a transfer rights market or a boundary rights market).

It should be noted that the descriptions of the market models below are at a high level, and
are not complete in all aspects – they are intended to provide a context to the evaluation
rather than a full description of the operation of the market.

Straw Man 1: Zonal entry/exit market

This market model is based on an entry/exit right market design – participants buy rights to
inject power onto the transmission system (entry rights) and rights to take power off the
system (exit rights) in given locations, both with reference to the same defined hub-point.
These locations are defined by zones – zonal entry / exit rights confer equal injection /
withdrawal rights to each entry / exit point in that zone. In order to resolve at least 75% of
constraints in this market model, our analysis suggests it is likely that a minimum of 24
zones would be required, with zonal boundaries being non-coincident with GSP Group
boundaries. Further analysis shows that 31 zones would be required for the transmission
access market to address about 90% of total constraint volume. These two zonal
definitions are shown below.

Figure 1 - Alternative Transmission Access Zonal Definitions

24 Zones 31 Zones

39
The table below shows the constraint capture levels of these two zonal definitions for the
six studies considered in the initial Straw Man model.

Table 1 - Volume of Constraints Prior to Gate Closure (% of Total Constraint

Volume)
Snapshot Total Constraint
of System Volume (TWh) 24 31 Zones
Zones
A 6.1 82% 93%
B 4.2 92% 95%
C 3.3 75% 92%
D 2.3 84% 97%
E 1.8 78% 95%
F 1.0 80% 88%

NGC agrees with Ofgem the absolute transfer capability of each boundary between zones,
and makes these capabilities public. An auctioneer then facilitates a primary auction
process for zonal rights for the forthcoming year, in which participants submit offer curves
to buy or sell “bundles” of entry and exit rights within each zone.

The auctioneer operates a simultaneous clearing process across zonal access products,
which determines a clearing price and optimal volumetric allocations to participants based
on their bids and the agreed transfer capabilities. The primary auction is subject to a set of
relative loss adjustment factors, which set a minimum differential between access product
prices to reflect the cost of unconstrained marginal losses.

Auction prices for each access right product are published. Surplus funds raised by this
auction process partially support NGC’s TNUoS revenue.

Following this primary auction, participants fine-tune their position by trading these rights
on secondary markets (potentially down to the half-hourly level). Via a Designated
Exchange, NGC participates in secondary trading to the extent that its expectations of the
inter-zonal boundary capabilities change – it is incentivised to maximise the revenue of
any release of new right, and minimise the cost of any buy-back. Clearly it will not be
possible to resolve intra-zonal constraints via this market – these will be left to the
Balancing Mechanism.

The access markets “close” at Gate Closure. The Balancing Mechanism continues as at
NETA Day 1 – effectively, NGC buys / sells a bundled access and energy product.

In each half hour, participants have a volumetric access imbalance exposure equal to the
difference of their access right holdings in a given zone (adjusted for accepted BM
bids/offers) and their metered output/demand in that zone. These imbalances are settled
at a price derived (at least in part) from closing access prices in the secondary markets.

As zonal boundaries are not coincident with GSP Group boundaries, a mechanism will be
needed to allocate demand to zones. At least initially, this could be achieved by simply
pro-rating GSP Group demand to the GSP nodes according to analysis of historical
demand dispersion. These nodes can then be amalgamated into the access zones.

40
Straw Man 2: Nodal entry/exit market

This market model is based on an entry/exit right market design – participants buy rights to
inject power onto the transmission system (entry rights) and rights to take power off the
system (exit rights) in given locations, with reference to a defined hub-point. These
locations are defined by nodes.

NGC agrees with Ofgem the capacity on the transmission system in the form of base
constraint data. Participants submit offer curves for the volume of entry and exit rights
which they require at each node.

Using these bids and an optimisation programme, NGC calculates the optimal allocation of
rights given the agreed limitations of the transmission system. The optimisation is subject
to the restriction of a set of minimum differentials between nodal prices set to reflect the
cost of unconstrained marginal losses.

Auction prices at each node are published. Surplus funds raised by this auction process
partially support NGC’s TNUoS revenue.

Following the primary auction, participants fine-tune their position by trading these rights
on secondary markets (potentially down to the half-hourly level). Given the bids from the
participants for increments and decrements of entry and exit right volumes at each node
on the system, NGC determines the volume of each bid to take in line with its updated
expectation of the inter-nodal capabilities. NGC’s trading would be subject to an incentive
mechanism. Following each round of secondary trading, NGC would again publish prices
at each node.

Although it is the characteristic which allows all constraints active prior to Gate Closure to
be resolved, the fact that the information on the relative effectiveness of each node with
respect to system constraints is variable and internal to NGC’s optimisation means that
participants cannot trade bilaterally (unless they are connected at the same node). The
vast majority of trades must be with NGC.

The access markets “close” at Gate Closure. The Balancing Mechanism continues as at
NETA Day 1 – effectively, NGC buys / sells a bundled access and energy product.

In each half hour, participants have a volumetric access imbalance exposure equal to the
difference of their access right holdings at a given node (adjusted for accepted BM
bids/offers) and their metered output/demand at that node. These imbalances are settled
at a price derived (at least in part) from closing access right prices in the secondary
markets.

As imbalance volumes are based on nodal injections / withdrawals, greater locational

tagging of demand than is available within the Stage 2 settlement system would be
required. As in Straw Man 1, this could be achieved by allocating GSP Group demand to
individual Grid Supply Points on the basis of factors determined by analysis of historical
demand dispersion.

Straw Man 3: Flowgate market

This market model is often also referred to as a transfer rights market or a boundary rights
market. It is based on a definition of access as the right to transport power over specified
flowgates rather than being defined in relation to injections to or withdrawals from areas
“either side” of the flowgates. Again the rights are with reference to a defined hub point.
41
In order to allow greater constraint resolution, the flowgates in question may be defined at
least in part at the circuit level rather than as boundaries (combinations of circuits).

NGC defines the flowgates which will be traded (based upon expectations of where
constraints will be active on the system), and agrees with Ofgem the volume of these
flowgate rights which will be made available.

NGC also carries out studies to determine the participation factor of each node on the
system for each defined flowgate. The participation factor for a node with respect to a
flowgate represents the fraction of each MW injected or withdrawn at the node – with
reference to a defined hub-point – which can be expected to be transported over it. NGC
publishes this participation factor information – with n nodes and k defined flowgates, this
publication would be in the form of an n x k matrix. In order to provide a firm basis for
secondary trading, these participation factors remain unchanged from the primary auction
through to imbalance settlement.

An auctioneer then facilitates a primary auction for access rights for the forthcoming year,
in which participants submit offer curves to buy and sell “bundles” of rights in each
flowgate. The auctions are independent of each other – there is no requirement for any
simultaneous clearing process. The primary auctions are subject to loss related reserve
prices, which set a minimum price for each access right to reflect the cost of unconstrained
marginal losses.

Participants determine the volumes of each of the flowgates they require on the basis of
their expected physical position at each node multiplied by the participation factor for each
node in each defined flowgate. Participants with a physical position will need to trade to a
target portfolio containing rights in a maximum of k flowgates.

Auction prices for each flowgate are published. Surplus funds raised by this auction
process partially support NGC’s TNUoS revenue.

Following this primary auction, participants fine-tune their position by trading rights on
secondary markets (potentially down to the half-hourly level). Through a Designated
Exchange, NGC releases further flowgate volumes, or buys back volumes to reflect the
evolving conditions on the transmission network, again under an incentive scheme.

Since the participation factors used in imbalance settlement are set at the primary auction
and not updated to reflect evolving system conditions, it will not be possible for the market
to efficiently resolve all constraints – some will be left to the Balancing Mechanism.
However, in contrast to Straw Man 1, it is difficult to evaluate the likely extent of constraint
resolution in this market model – such an analysis would need to map changes in network
topology to changes in participation factors to (critically) changes in the extent of efficient
constraint resolution.

The access markets “close” at Gate Closure. The Balancing Mechanism continues as at
NETA Day 1 – effectively, NGC buys / sells a bundled access and energy product.

Participants’ volumetric imbalances are calculated separately for each defined flowgate by
comparing volumetric holdings of individual flowgate rights with deemed use. Volumetric
holdings are defined for each participant as holdings of each right adjusted (using the fixed
participation factors) to reflect accepted BM offers and bids. The deemed volumetric use

42
is defined as nodal metered injections or withdrawals multiplied by the set of participation
factors for the node in the flowgate in question.

These imbalances are settled at a price derived (at least in part) from the closing prices of
each flowgate right in the secondary markets.

As imbalance volumes are based on use of flowgate rights calculated by applying

participation factors to nodal injections / withdrawals, greater locational tagging of demand
than is available within the Stage 2 settlement system will be required. At least initially,
this could be achieved by allocating GSP Group demand to individual Grid Supply Points
on the basis of factors determined by analysis of historical demand dispersion.

Evaluation of market models

In all of the above Straw Man models, the efficiency of constraint resolution will depend
upon NGC’s ability to forecast accurately system capabilities and conditions at various
points in time, and particularly at the primary auction (as this is when access rights are
initially allocated). In Straw Man 1, therefore, effectiveness will depend in part upon the
definition of the zonal boundaries and the estimation of each boundary transfer capability.
In Straw Man 2, effectiveness will depend upon the accuracy of the base data for the
optimisations. In contrast, in Straw Man 3, effectiveness will depend in part upon the initial
definition of flowgates and the estimation of participation factors.
However, in addition to the general importance of forecasting across the market models, it
is possible to identify further pros and cons of each of the models individually. In assessing
these pros and cons, we make the assumption that NGC is able to perfectly forecast
system conditions as required. Table 2 below summarises our evaluation.

Table 2 - Evaluation of Straw Man market models

Straw Man 1 Straw Man 2 Straw Man 3

Pros: Pros: Pros:
Valuation of access All constraints active Offers the possibility of
rights for participants is before Gate Closure will solving a significant
not complex be solved proportion of constraints
NGC is only involved in Valuation of access with the potential of
inter-zonal trading – no rights for participants is unfacilitated bilateral
wider facilitation of not complex trading of transmission
trading is required rights
Cons: Cons: Cons:
The number of zones (To the extent it is Efficient constraint
required to capture a perceived as a con by resolution is dependent
significant percentage of market participants) upon a degree of
constraints is likely to NGC has to facilitate stability of network
restrict liquidity. A and act as counterpart topology
number of zones would to all inter-nodal Valuation of access
have one single participant trades rights for participants is
generator. complex

43
Straw Man 1

The key advantages of this market model are:

• valuation of access rights for participants is not complex: participants are able to
value the entry / exit access rights (as a minimum) on the basis of the difference
between their expectations of the energy price at the national delivery point and their
own costs – they do not necessarily need to engage in more complex analysis of other
participants’ behaviour in the access markets to value the rights; and

• NGC is only involved in inter-zonal trading – no wider facilitation of trading is

required: for participants within the same zone, trading of entry and exit rights can take
place bilaterally, either via over-the-counter or on-exchange trading – no facilitation by
NGC is required. NGC’s trading relates solely to changes in expected transfer
capabilities between zones – for example buying back entry rights in one zone and
selling additional entry rights in another.

In contrast, the main disadvantage of this model is that the number of zones required to
capture a significant percentage of constraints is likely to restrict liquidity. As stated
above, in order to solve at least 75% of constraints, it is likely that at least 24 zones would
be needed. Since this reduces the number of different participants in each zone, it
reduces the likelihood of liquid secondary markets developing – with 24 zones, a number
might have only one generator participant.

Straw Man 2

The key advantages of this market model are that:

• all constraints active before Gate Closure will be solved: since the model is nodal
in nature and allocations are the result of an optimisation process which takes into
account changes to actual system conditions as they evolve, the market will be capable
of solving all transmission constraints which are active before Gate Closure; and

• valuation of access rights for participants is not complex: as the market is based
on entry / exit rights, as in Straw Man I, participants are able to come to a valuation of
the entry / exit rights being traded (as a minimum) on the basis of the difference
between their expectations of the energy price at the national delivery point and their
own costs.

The principal disadvantage with this market model, to the extent that it is perceived as
such by market participants, is that NGC has to facilitate and act as counterparty to all
participant trades (save for nodes where there is more than one connected participant).

Straw Man 3

The key advantage of this market model is that it offers the possibility of solving a
significant proportion of constraints whilst allowing unfacilitated bilateral trading of
individual access rights. Since flowgates can be defined in relation to individual circuits
and participation factors defined for each node in relation to each flowgate, it would
theoretically be possible to resolve all constraints (provided participation factors did not
change from the primary auction). Similar to Strawman 1, if approximations are made to

44
 the flowgates to increase liquidity or reduce complexity then some constraint capture will
be lost.

There are, however, important disadvantages of this model:

• since imbalance settlement (and hence all ex ante trading) takes place on the basis of
participation factors fixed at the time of the primary auction, trading will not take into
account changes in network topology (which result in changes to the actual
participation factors) despite the fact that they may create new constraints. As a result,
efficient constraint resolution is dependent on a degree of stability of network
topology – if topology changes frequently, the market will not be able to efficiently
resolve all the actual constraints on the network; and

• the valuation of access rights for participants is complex. In an entry/exit market

model, participants can price access rights (as a minimum) on the difference between
their expectation of the energy price and their own costs. At an aggregate level, the
same is still true under this model. Participants know that the unit price of their target
portfolio of access rights should be equal to the difference between their expectation of
the energy price and their unit costs. However, in contrast with the entry / exit model,
this information is insufficient for each participant to price every individual access right.
While a generator with a given output knows (from participation factors) the volume of
each flowgate which they require, and knows their valuation of a unit of their total target
portfolio, they cannot impute a value to each component access right in that portfolio
without further information and analysis.

Summary
Three alternative Straw Man models of the transmission access market have been
assessed and compared. The key points and important trade-offs are provided in the
following table.

Market Definition Locational Constraint NGC Need to Complexity for

model of rights resolution resolution under facilitation of solve “GSP Market
idealised trading? Group Participants to
condition metering” value the access
problem? rights
(% of total
volume)
Straw Entry / exit Zonal >75% depending Required for Yes Relatively simple
Man 1 on zonal definition inter-zonal
(>24
trading
zones)
Straw Entry / exit Nodal 100% Required for Yes Relatively simple
Man 2 inter-nodal
trading
Straw Transfer Linked to 100% None Yes Complex
Man 3 Nodes by
only with
participatio
completely stable
n factors
topology

45
Attachment 3: Simultaneous Clearing in Access Auctions and Losses Charges

Simultaneous Clearing – A Paper by NGC

Each constraint is defined as a limit on the total flow across a system boundary that
divides the system into two areas. The boundary flow is represented by the difference
between the total generation and total demand in the area which does not include the
reference point. This can be termed the constrained area. In the case of generation
exceeding demand in the constrained area, the transmission constraint sets a limit on the
excess of total entry rights over total exit rights in that area. In the case of demand
exceeding generation in the constrained area, the transmission constraint sets a limit on
the excess of total exit rights over total entry rights in that area.

Within one single auction, after receiving market participants’ bids and offers for entry and
exit rights in all the zones, a simultaneous clearing of rights over all system boundaries will
take place. This will release as many rights as possible which would maximise the total
income of the auction while respecting the transmission constraints.

An example of a two-zone system is given below to illustrate how this process works.

North

North-South
Boundary

South
(Reference Point)

Consider an auction for transmission capacity in a zone in the North of England. Suppose
there is a North-South boundary and the reference point (ie energy exchange) is located in
the South.

The clearing price in energy exchange is £20/MWh.

In North there are five generators each having 100MW capacity and all with production
cost lower than the energy exchange price. They will be prepared to bid up to the
difference between their production cost and the energy exchange price:

Production cost (Max) bid for the right

£/MWh £/MWh
G1 19 1
G2 18 2
G3 17 3
G4 16 4
G5 15 5

There are also five demand blocks in North, each sized at 100MW and having the utility
value listed below. Each block will consume only if it receives a transmission obligation
payment at least equal to the difference between its utility value and the energy exchange
price:

46
 Utility (Min) offer for the
£/MWh obligation £/MWh
D1 19 -1
D2 18 -2
D3 17 -3
D4 16 -4
D5 15 -5

Case 1 - Excluding Effect of Losses, System Constrained

The North-South boundary is constrained to a 100 MW maximum flow. The clearing

process consists of accepting those bids and offers which yield the greatest auction
income, while respecting the 100 MW constraint. Combinations which (just) satisfy the
constraint are:

Bids/offers accepted Auction income

£/hr
G5 500
G5,G4,D1 800
G5,G4,G3,D1,D2 900
G5,G4,G3,G2,D1,D2,D3 800
G5,G4,G3,G2,G1,D1,D2,D3,D4 500

The auctioneer will arrive at the third combination, with a clearing price of £3/MWh set by
G3 (highest bid taken) and D3 (lowest offer not taken).

Combining Relative Loss Adjustments to Clearing Process

A marginal charge for losses and the interaction between losses and constraints can be
automatically allowed for in a loss adjusted access rights clearing process. In this process
the “unconstrained Transmission Loss Factor (TLF)3 times energy exchange price” is
netted off all the bids and offers when they are ranked.

In the two-zone example, suppose the unconstrained TLF for the North is 8%. This gives a
marginal loss cost at 8%*£20/MWh = £1.6/MWh. The bids and offers are adjusted in the
clearing process.

Northern generation’s bids:

3
For the generation in a particular zone this is the linear sensitivity of system losses to a marginal increase in generation
in that zone which is balanced at the reference point, assuming no system constraint is present. The TLF for the demand
is equal in magnitude and opposite in sign to the corresponding generation TLF.

47
 Original bid Marginal loss adjusted bid
£/MWh £/MWh
G1 1 -0.6
G2 2 0.4
G3 3 1.4
G4 4 2.4
G5 5 3.4

Northern demand’s offers:

Original offer Marginal loss adjusted

£/MWh offer
£/MWh
D1 -1 0.6
D2 -2 -0.4
D3 -3 -1.4
D4 -4 -2.4
D5 -5 -3.4

Case 2 - Loss Adjusted Clearing, System Unconstrained

Without any constraint on the North-South transfer, all the bids and offers which have
positive TLF adjusted values are accepted, i.e. G5,G4,G3,G2 and D1. The total auction
income would be £1300/hr. The clearing price is at £2/MWh, which is set by G2 (the lowest
bid for entry accepted above marginal loss cost) and D2 (the highest offer for exit not
accepted below marginal loss cost). It is expected that if the bid and offer curves are
smooth, the clearing price for unconstrained transmission access will be at the marginal
loss cost.

Case 3 - Loss Adjusted Clearing, System Constrained

If there is a North-South boundary constraint at 100 MW, then the combinations which just
satisfy the constraint are:

Bids/offers accepted Auction Auction income TLF

income adjusted
£/hr £/hr
G5 500 340
G5,G4,D1 800 640
G5,G4,G3, D1,D2 900 740
G5,G4,G3,G2,D1,D2,D3 800 640
G5,G4,G3,G2,G1,D1,D2,D3, 500 340
D4

The auctioneer will arrive at the third combination with a clearing price of £3/MWh, which is
the same as Case 1. Cases 2 and 3 show the interaction between losses and constraints
and that the losses effect should not affect the clearing outcome where the system is
constrained.
Attachment 4: Examples of the Calculation of Imbalance Charges

48
Access Imbalance Pricing – A Paper by NGC

Overview of approach

Assume that buying an access 'ticket' imposes both a right and an obligation and that it is
impossible to monitor what a party actually has paid for a 'ticket' through all trading
mechanisms.

An imbalance price is required for two situations:

• over-run: if the market participant’s demand or generation is greater than the number
of tickets held; and
• under-run: if the market participant’s demand or generation is less than the number of
tickets held.

Imbalance prices relate to a zonal access price (ZAP) which could be calculated in any
number of ways such as average/marginal price of NGC zonal trades in the secondary
market or average/marginal price of bids accepted in the Primary Auction, or
average/marginal price of bids accepted in the Balancing Mechanism. The exact
calculation of the ZAP will need further consideration and could involve taking the
maximum or average of a number of prices. The ZAP for entry rights is the negative of the
ZAP for exit rights.

The ZAP can be adjusted up or down to form an over-run or under-run price in order to
incentivise participants to balance. Indeed, the decision may be taken to set one of the
prices to zero.

In this way, participants are purchasing both rights and obligations at the same time and
there is no need to explicitly determine individually which type they have. The sign of the
ZAP determines whether tickets relate to rights or obligations.

An imbalance liability calculation is detailed below. An allowance is made that access

prices for entry and exit could be either positive or negative is allowed for, depending on
the location of the reference point for access rights.

Calculation

There are four situations to consider. The access imbalance charge (AIC) is obtained by
multiplying the Access Imbalance Price [(1+/-x)*ZAP] by the Imbalance Volume [M-BO-
ZAT]:

Where:
x = Spread Factor
ZAP = Zonal Access Price (calculation method to be determined)
M = Zonal Metered Generation or Demand
BO = Zonal Bids/Offers accepted in the Balancing Mechanism
ZAT = Zonal Access Tickets Held
If the market participants' generation or demand is greater than the number of tickets held
(i.e. an over-run, the imbalance volume (M-BO-ZAT) is positive, then the access
imbalance charge (AIC) is as follows:

49
AIC = (1+x) * ZAP* (M - BO - ZAT) if ZAP is positive

AIC = (1-x) * ZAP * (M - BO - ZAT) If ZAP is negative

On the other hand, if market participants' actual generation or demand is less than the
number of access tickets held (i.e. an under-run and (M-BO-ZAT) is negative) the AIC
should be:

AIC = (1-x) * ZAP * (M - BO - ZAT) If ZAP is positive

AIC = (1+x) * ZAP * (M - BO - ZAT) If ZAP is negative

Example

Assume x = 0 for simplicity

Participant Imbalance Volume ZAP (2) AIP (3) AIC

(M-BO-ZAT) (1) [=(2) as x=0] [= 1*3]
Generation Positive / Over-run Positive Positive Positive
Positive / Over-run* Negative Negative Negative
Negative / Under- Positive Positive Negative
run*
Negative / Under-run Negative Negative Positive
Demand Positive / Over-run Positive Positive Positive
Positive / Over-run* Negative Negative Negative
Negative / Under- Positive Positive Negative
run*
Negative / Under-run Negative Negative Positive

*The AIP could be set to zero if a 'use it or lose it' principle was to be adopted.

Numerical Example

Consider two zones, A in which the ZAP is +3 and B in which the ZAP is –3 (as noted
above, another way to think of this is to consider the entry and exit rights respectively in
the same zone). Suppose, for example, the spread factor was set as 50% and is applied
symmetrically.

In zone A:

• the over-run price (which will be positive as it should represent a charge to participants)
should be set higher than the positive ZAP to disincentivise over-runs. The over-run
price should therefore be 4.5; and
• the under-run price (which will positive as multiplied by a negative imbalance volume it
should represent a payment to participants) should be set lower than the positive ZAP
to disincentivise under-runs. The under-run price should therefore be 1.5.

In contrast, in zone B:

50
• the over-run price (which should be negative, as it should represent a payment to
participants – they would have been paid if they had bought an access right through ex
ante trading) should be set higher than the negative ZAP to disincentivise over-runs.
The over-run price should therefore be –1.5; and

• the under-run price (which should be negative, as when multiplied by a negative

imbalance volume, it should represent a charge to participants – they would have to
pay someone to take the access right off their hands in ex ante trading) should be set
to higher than the negative ZAP, in order to disincentivise under-runs. The under-run
price should therefore be -4.5.

Pulling this together gives the following matrix of imbalance prices:

ZAP =3 ZAP negative = -3
Imbalance price = 4.5 Imbalance price = -1.5*
Over-run @
50% spread
factor
Under-run @ Imbalance price = 1.5* Imbalance price = -4.5
50 % spread
factor

• These prices could be set to zero if it were decided not to apply the spread factor
symmetrically. This would be the equivalent of a “use it or lose it” provision.

51
Attachment 5: Implementation of a Market In Transmission Access – Implications
for Systems

Introduction

This document has been prepared to brief attendees at the workshops on transmission
access issues on 7 August 2000.
It provides an outline of the likely high level systems requirements arising from the form of
market that NGC understands Ofgem intends to be established. This is provided under
the following major headings:
• background;

• assumptions for the new arrangements – a ‘straw man’;

• possible systems requirements; and

• implementation issues.

Background

The need for revised arrangements in transmission access has been known for some time,
with initial thinking on possible arrangements being undertaken at the same time as early
definition of requirements for NETA.

Ofgem has produced three documents related to transmission access:

• in its July 1999 NETA Document, Ofgem presented some initial thinking on the role of
and incentives on NGC as System Operator and the development of new transmission
access and pricing arrangements under NETA;

• the October 1999 NETA Document, published jointly by Ofgem and the DTI, discussed
respondents’ views on the initial thinking outlined in the July 1999 NETA Document;
and

• Ofgem then issued a consultation document in December 1999, in which it indicated

that its initial views were that transmission access and pricing arrangements need to
reflect the value of transmission access and therefore to reflect the locational value of
electricity.

The arrangements described in this document are based on NGC’s current understanding
of Ofgem’s views on the form of transmission access market that should be established,
as set out in the December 1999 consultation document.

NGC has been actively involved in the debate on transmission access, through
discussions with Ofgem and industry participants, and through identifying in general terms
the implications of introducing a market in transmission access rights along the lines set
out in Ofgem’s December 1999 consultation document. As part of this, NGC has made
presentations to industry groups, including presenting a consideration of alternative
approaches to implementing transmission access arrangements to the Charging Principles

52
Forum (although Ofgem’s December 1999 consultation document effectively closed down
some of these alternatives).

More recently, NGC has undertaken work to identify in more detail the likely processes
and systems that would be needed to facilitate new transmission access arrangements,
and to understand the nature and scale of work that would be required to implement those
arrangements.

The contents of this document are largely based on this recent work.

Assumptions for the new arrangements – a ‘straw man’

The possible systems requirements set out in this document are based on a number of key
assumptions which effectively define a ‘straw man’ for the transmission access market
along the lines set out in Ofgem’s December 1999 consultation document. The most
important of these assumptions are:

• The market will be for firm physical entry and exit rights; The rights would be firm in the
sense that if NGC, as the System Operator, is unable to deliver them, it would have to
buy-back the rights. The definition of firm access rights would be locational to reflect
the realities of the physical transmission system.

• The market will be two-sided. It is envisaged that both the generation and demand will
participate in the access regime. In general, generators will acquire entry rights and
suppliers (or large customers in their capacity as self-suppliers) will acquire exit rights.

• The market will be based on the use of zones based on GSP Groups. Demand side
involvement has an important consequence for the definition of zones. As imbalance
settlement will require metered volume data, zones must conform to GSP Groups as
this is the finest level of locational tagging of consumption data available without any
changes to the 1998 metering systems.

• There will be a primary auction to establish the initial allocations of access rights:
Access rights will initially be auctioned by zone and by temporal product (e.g. peak, off
peak etc) for one year. This will result in an initial allocation of rights to participants
based on a broad view of the likely capabilities of the transmission system.

• There will be unfacilitated secondary trading of transmission access rights: Subsequent

to the primary auction, both participants and NGC will wish to trade these rights via
secondary markets, which will run from the primary auction up to Gate Closure (i.e. the
opening of the Balancing Mechanism). Participants will trade to fine tune their access
position with their expected physical position in order to minimise exposure to the
access imbalance regime. Nearer real time, NGC would also participate in the
secondary access markets, releasing additional rights or buying rights back as
expectations of the transmission system’s actual capacity changed.

• There will be a bundled energy and access Balancing Mechanism. Bids and offers
accepted in the Balancing Mechanism will be deemed to include the appropriate entry
or exit access rights to avoid participants being exposed to access imbalance charges
on those volumes.

53
• There will be half-hourly zonal settlement of access imbalances. The access
imbalance settlement process will occur after the delivery half-hour is finished. This
process will compare generators’ metered output and suppliers’ metered/profiled
consumption (adjusted for BM actions accepted) with the volume of access rights held
by individual participants. Long (under-run) and short (over-run) access positions will
result in imbalance liabilities or payments.

Given the criticality of these assumptions in defining the market, the ‘straw man’ on which
this document is based should only be seen as one option for a transmission access
market.
As the design of the arrangements progresses to greater levels of detail, these
assumptions will be need to be reviewed and potentially changed. If the assumptions are
changed, the outline systems requirements, and the scope of work required to implement
those systems, are also likely to change.

Possible systems requirements

The direction set by Ofgem’s December 1999 consultation document, and the
assumptions listed above, lead to the following as a possible division of responsibilities
with respect to the operation of the transmission access arrangements:
• NGC would be responsible for agreeing the key parameters of the auction
arrangements in advance with Ofgem, for settlement of primary auction bids, for
secondary trading activity to balance the system, and then for charging adjusted
TNUoS and BSUoS tariffs;

• an Auctioneer would be responsible for offering products for sale in the primary
auction, running the system that calculates equilibrium auction prices and volumes, and
notifying participants (and potentially the market in general) of auction results;

• a Secondary Market Operator would be responsible for operating a Designated

Exchange (see below) and settling on-exchange trades; and

• a Transmission Access Registration and Settlement Agent (TARSA) would be

responsible for registering access right allocations, and for calculating and settling
imbalance liabilities with participants.

It has been assumed that there will be a Designated Exchange (as in the gas commodity
regime), to provide the market with the assurance that there will always be a transparent,
facilitated market on which NGC can make release and buy-back trades, and from which
information can be derived in a transparent manner to be used potentially in the calculation
of imbalance prices. It is assumed that the roles of Auctioneer, Secondary Market
Operator and TARSA would be undertaken by bodies contracted to NGC, but not by NGC
itself. It is therefore assumed that NGC will procure both the systems and the operational
services required for these roles.

54
The possible systems requirements for the market are summarised in Figure 1. This figure
indicates at a high-level the broad functionality that would be required in systems by each
of the roles listed above, and the key interfaces between those roles.

Figure 1: Overview of transmission access systems requirements

Auctioneer TARSA
• offer products • Register and maintain entry and
• receive bids exit access right allocations
• calculate the allocation of access • Calculate overrun and underrun
rights and equilibrium prices Volume prices and liabilities
allocations
• notify auction results • Calculate surplus funds from
• Pass auction price information to settlement
settlement • Bill or pay participants for
overruns and underruns
Accepted bids Updates to
BSUoS Trade data,
and prices including
charges
prices

NGC Secondary Market Operator

• Settle primary auction trades • Maintain participant and trade
• Update TNUoS charges with details
auction receipts and payments NGC • Operate secondary market
• Update BSUoS charges as requirements • report ‘standard’ market data,
and trades
required including pricing information
• Release and buy back access • Settle secondary trades
rights to the secondary market

Cost Estimate
At this stage, before agreement on the market design, it is not possible to estimate with
any accuracy the likely costs of implementing systems to support a transmission access
market. It is possible, however, to make rough estimates based on previous experience of
systems which have been developed in broadly comparable situations. For example, the
settlement roles and systems being developed for NETA, creation of trading exchanges
and auction mechanisms.
It must be noted that where similar systems are needed (e.g. imbalance settlement), the
scale and complexity of the projects and systems required will be of a similar magnitude to
central NETA systems, thus requiring a similar time to design, develop, test and implement
those systems. In addition, systems and operational services will be presumably be
procured through an open tendering process, using a similar approach to NETA, through
the issue of an OJEC notice inviting expressions of interest, followed by the issue of an
Invitation to Tender (ITT) to selected suppliers. Hence, the 'delivery' of the market is likely
to be an industry wide project involving many players where an overall programme
management role will be critical.

55
Figure 2 shows broad estimates of the minimum projected system costs to set up a zonal
transmission access market based on the assumptions listed above.

Figure 2

System / Process Cost Components

(real 1999/00 prices)
Primary Auction System £1.5m to £3m Zonal auction system
Ticket allocation method
Auction settlement
Publication of results
Secondary Trading £2m to £4m Includes
Mechanism release/buyback system
procurement (with
option for bi-lateral
trading)
Imbalance Settlement £10m to £20m Assume 5 of 7 NETA
roles are required (all
except metered volumes
and BM actions)
Aggregation needed at
a zonal level
Rights Registration
Agent included
Funds Administration
included
TNUoS changes £0.5m - £1.5m Includes NGC system
changes for tariff and
reconciliation
BSUoS changes £0.5m to £1m Includes NGC system
changes for algorithm
and interfacing
Overall Program £0.5m to £1.5m Managing overall
Management implementation
interfaces / legal
contracts
Total £15m to £31m

In the above, no consideration has been given to the ongoing annual costs of operation of
the access market. This would include fulfilment of the Settlement System Administrator
role, running of the Secondary Market and the running of an annual auction as well as the
trading activities of participants.
In addition, no allowance has been made for the costs involved with participants setting up
their own trading processes and systems.

56
APPENDICES I - VII

57
 Appendix I

Terms and Definitions

Total Transfer Capabilities TTC

Available Transfer Capabilities ATC

Transmission Reliability Margin TRM

Capacity Benefit Margin CBM

Open Access Same-time Information Systems OASIS

58
 Appendix II

North American Electric Reliability Council Report - Framework

For determining available transfer capability

59
 Available
Transfer Capability
Definitions and Determination

A framework for
determining available transfer capabilities
of the interconnected transmission networks
for a commercially viable
electricity market

North American Electric Reliability Council

June 1996
CONTENTS

EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Terminology Convention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
NERC Strategic Initiative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Purpose of This Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
AVAILABLE TRANSFER CAPABILITY PRINCIPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
TRANSMISSION TRANSFER CAPABILITY CONCEPTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Transfer Capability versus Transmission Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Determination of Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Limits to Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Uses of Transmission Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
TTC DEFINITION AND DETERMINATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Definition of Total Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Determination of Total Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
System Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Critical Contingencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
System Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Parallel Path Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Non-Simultaneous and Simultaneous Transfers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
TRANSMISSION TRANSFER CAPABILITY MARGINS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Technical Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Definition of Transmission Reliability Margin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Uncertainty in TTC and ATC Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Need for Operating Flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Definition of Capacity Benefit Margin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
ATC DEFINITION AND DETERMINATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Definition of Available Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Determination of Available Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Commercial Components of Available Transfer Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Curtailability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Recallability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Recallable and Non-recallable Relationships and Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Scheduled and Reserved Transmission Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Transmission Service Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Appendix A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION . . . . . . . . . . . . . . . . 22
Appendix B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION . . . . . . . . . . . . . . . . 29
Appendix C. TRANSMISSION SERVICE RESERVATIONS AND SCHEDULING . . . . . . . . . . . . . . . . 35
TRANSMISSION TRANSFER CAPABILITY TASK FORCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

NERC Available Transfer Capability Definitions and Determination

EXECUTIVE SUMMARY

This report, Available Transfer Capability Definitions and Determination, is in response to a NERC
Strategic Initiative to “develop uniform definitions for determining Available (Transmission) Transfer
Capability (ATC) and related terms that satisfy both [Federal Energy Regulatory Commission] FERC and
electric industry needs, and which are to be implemented throughout the industry.” The NERC Board of
Trustees at its May 13–14, 1996 meeting approved this report and endorsed its use by all segments of the
electric industry.

The report establishes a framework for determining ATCs of the interconnected transmission networks for
a commercially viable wholesale electricity market. The report also defines the ATC Principles under
which ATC values are to be calculated. It is non-prescriptive in that it permits individual systems, power
pools, subregions, and Regions to develop their own procedures for determining or coordinating ATCs
based on a regional or wide-area approach in accordance with the Principles defined herein. The proposed
ATC calculation framework is based on the physical and electrical characteristics and capabilities of the
interconnected networks as applicable under NERC, Regional, subregional, power pool, and individual
system reliability planning and operating policies, criteria, or guides.

This report provides an initial framework on ATC that will likely be expanded and modified as experience
is gained in its use and as more is learned about how the competitive electric power market will function.
The U.S. Federal Energy Regulatory Commission’s final rules, Orders No. 888 and No. 889 pertaining to
promoting wholesale competition through open access non-discriminatory transmission services by public
utilities and an open access same-time information system, respectively, were issued April 24, 1996. The
framework for the determination of ATC as outlined in this report is in accord with the key provisions of
these rulemakings.

ATC PRINCIPLES
The following Available Transfer Capability (ATC) Principles govern the development of the definition
and determination of ATC and related terms. All transmission provider and user entities are expected to
abide by these Principles.

1. ATC calculations must produce commercially viable results. ATCs produced by the calculations
must give a reasonable and dependable indication of transfer capabilities available to the electric
power market.

2. ATC calculations must recognize time-variant power flow conditions on the entire interconnected
transmission network. In addition, the effects of simultaneous transfers and parallel path flows
throughout the network must be addressed from a reliability viewpoint.

3. ATC calculations must recognize the dependency of ATC on the points of electric power injection,
the directions of transfers across the interconnected transmission network, and the points of power
extraction. All entities must provide sufficient information necessary for the calculation of ATC.

4. Regional or wide-area coordination is necessary to develop and post information that reasonably
reflects the ATCs of the interconnected transmission network.

Available Transfer Capability Definitions and Determination NERC 1

EXECUTIVE SUMMARY

5. ATC calculations must conform to NERC, Regional, subregional, power pool, and individual
system reliability planning and operating policies, criteria, or guides.

6. The determination of ATC must accommodate reasonable uncertainties in system conditions and
provide operating flexibility to ensure the secure operation of the interconnected network.

The calculation of transfer capability is generally based on computer simulations of the operation of the
interconnected transmission network under a specific set of assumed operating conditions. These simu-
lations are typically performed “off line,” well before the systems approach that operational state. Each
simulation represents a single “snapshot” of the operation of the interconnected network based on the
projections of many factors. As such, they are viewed as reasonable indicators of network performance and
available transfer capability.

ATC DEFINITIONS
Available Transfer Capability (ATC) is a measure of the transfer capability remaining in the physical
transmission network for further commercial activity over and above already committed uses. Mathe-
matically, ATC is defined as the Total Transfer Capability (TTC) less the Transmission Reliability Margin
(TRM), less the sum of existing transmission commitments (which includes retail customer service) and the
Capacity Benefit Margin (CBM).

Total Transfer Capability (TTC) is defined as the amount of electric power that can be trans-
ferred over the interconnected transmission network in a reliable manner while meeting all of a
specific set of defined pre- and post-contingency system conditions.

Transmission Reliability Margin (TRM) is defined as that amount of transmission transfer

capability necessary to ensure that the interconnected transmission network is secure under a
reasonable range of uncertainties in system conditions.

Capacity Benefit Margin (CBM) is defined as that amount of transmission transfer capability
reserved by load serving entities to ensure access to generation from interconnected systems to
meet generation reliability requirements.

Curtailability is defined as the right of a transmission provider to interrupt all or part of a transmission
service due to constraints that reduce the capability of the transmission network to provide that trans-
mission service. Transmission service is to be curtailed only in cases where system reliability is threatened
or emergency conditions exist.

Recallability is defined as the right of a transmission provider to interrupt all or part of a transmission
service for any reason, including economic, that is consistent with FERC policy and the transmission
provider’s transmission service tariffs or contract provisions.

Non-recallable ATC (NATC) is defined as TTC less TRM, less non-recallable reserved transmission
service (including CBM).

2 NERC Available Transfer Capability Definitions and Determination

EXECUTIVE SUMMARY

Recallable ATC (RATC) is defined as TTC less TRM, less recallable transmission service, less non-
recallable transmission service (including CBM). RATC must be considered differently in the planning
and operating horizons. In the planning horizon, the only data available are recallable and non-recallable
transmission service reservations, whereas in the operating horizon transmission schedules are known.

ATC AND RELATED TERMS

ATC and related terms are depicted graphically below. They form the basis of a transmission service
reservation system that will be used to reserve and schedule transmission services in the new, competitive
electricity market.

TTC, ATC, and Related Terms in the Transmission Service Reservation System

Available Transfer Capability Definitions and Determination NERC 3

INTRODUCTION

BACKGROUND
Available Transmission Capacity as described in the U.S. Federal Energy Regulatory Commission’s
(FERC) March 29, 1995 Notice of Proposed Rulemaking (NOPR), Docket RM95-8-000, Section III-E4f,
is a new term that has not been universally defined or used by the electric industry. The electric industry
has historically used other standard terms, techniques, and methodologies to define and calculate mean-
ingful measures of the transmission transfer capability of the interconnected transmission networks. These
terms, which include First Contingency Total Transfer Capability (FCTTC) and First Contingency Incre-
mental Transfer Capability (FCITC) as defined in NERC’s May 1995 Transmission Transfer Capability
reference document, are still applicable measures in an open transmission access environment. FERC’s
term Available Transmission Capacity and its definition and relationship to the industry’s terminology need
to be further clarified.

In its NOPR, FERC also requires that Available Transmission Capacity information be made available on a
publicly accessible Real-time Information Network (RIN). Definitions of Available Transmission Capacity
in the report of the industry’s Electronic Information Network “What” Working Group, which was filed
with FERC on October 16, 1995, are only considered to be assumptions to support the Working Group’s
effort in determining what information should be included on RINs. This report further refines those
definitions.

It must be noted early in this report that electric systems in Canada and the northern portion of Baja
California, Mexico, which are electrically interconnected with electric systems in the United States, are
active members in NERC and the Regional Councils and are committed to promoting and maintaining
interconnected electric system reliability. These non-U.S. systems are not, however, subject to FERC
jurisdiction, and the commercial aspects of the definitions contained herein are not necessarily applicable to
the operation of their internal transmission systems.

TERMINOLOGY CONVENTION
FERC used the term Available Transmission Capacity in its NOPR to label the information that is to be
made accessible to all transmission users as an indication of the available capability of the interconnected
transmission networks to support additional transmission service. To avoid confusion with individual
transmission line capacities or ratings, all references to “ATC” throughout this report will refer to Avail-
able (Transmission) Transfer Capability and its related terms as defined in this report.

NERC STRATEGIC INITIATIVE

One of several Strategic Initiatives for NERC, approved by the NERC Board of Trustees on October 3,
1995, is to “develop uniform definitions for determining Available (Transmission) Transfer Capability and
related terms that satisfy both FERC and electric industry needs, and which are to be implemented
throughout the industry.” The then existing NERC Transmission Transfer Capability Task Force, with ex-
panded membership to include all segments of the electric industry, was assigned this responsibility for
completion in May 1996.

4 NERC Available Transfer Capability Definitions and Determination

INTRODUCTION

PURPOSE OF THIS REPORT

This report is the response to NERC’s Strategic Initiative on ATC and defines ATC and related terms.
From a commercial perspective, the key element in the development of uniform definitions for transmission
transfer capability is the amount of transfer capability that is available at a given time for purchase or sale
in the electric power market under various system conditions. Open access to the transmission systems
places a new emphasis on the use of the interconnected networks. As such, future electric power transfers
are anticipated to increase over a wide range of system conditions, making the reliable operation of the
transmission networks more complex. To effectively maintain system reliability, those who calculate,
report, post, and use this information must all have the same understanding of its meaning for commercial
use. To accomplish this purpose, this report will answer the following questions:

– What is ATC?
– How does ATC relate to industry standard terminology?
– What physical factors need to be considered in determining ATC?
– What reliability issues must be considered in determining ATC?
– How is ATC calculated?
– How will ATC be commercially used?

The report establishes a framework for determining the ATCs of the interconnected transmission networks
for a commercially viable electricity market. Although the report defines the ATC Principles under which
ATCs are to be calculated, it is non-prescriptive in that it permits individual systems, power pools, sub-
regions, and Regions to develop their own procedures for determining or coordinating ATCs based on a
regional or wide-area approach in accordance with these Principles.

The report does not address transmission ownership and equity issues, nor does it address the allocation of
transmission services or ATC values. The calculation of ATC is based strictly on the physical and elec-
trical characteristics and capabilities of the interconnected networks as applicable under NERC, Regional,
subregional, power pool, and individual system reliability planning and operating policies, criteria, or
guides.

As the competitive electric power market develops, more will be learned on how these markets will function
and how the definitions of ATC will be used. This report provides an initial framework on ATC, which
will likely be expanded and modified as experience is gained in its use. The U.S. Federal Energy
Regulatory Commission’s final rules, Orders No. 888 and No. 889 pertaining to promoting wholesale
competition through open access non-discriminatory transmission services by public utilities and an open
access same-time information system, respectively, were issued April 24, 1996. The framework for the
determination of ATC as outlined in this report is in accord with the key provisions of these rulemakings.

Available Transfer Capability Definitions and Determination NERC 5

AVAILABLE TRANSFER CAPABILITY PRINCIPLES

ATC PRINCIPLES
Available Transfer Capability (ATC) is a measure of the transfer capability remaining in the physical
transmission network for further commercial activity over and above already committed uses. As a
measure bridging the technical characteristics of how interconnected transmission networks perform to
the commercial requirements associated with transmission service requests, ATC must satisfy certain
principles balancing both technical and commercial issues. ATC must accurately reflect the physical
realities of the transmission network, while not being so complicated that it unduly constrains commerce.
The following principles identify the requirements for the calculation and application of ATCs.

1. ATC calculations must produce commercially viable results. ATCs produced by the calcu-
lations must give a reasonable and dependable indication of transfer capabilities available to
the electric power market. The frequency and detail of individual ATC calculations must be
consistent with the level of commercial activity and congestion.

2. ATC calculations must recognize time-variant power flow conditions on the entire intercon-
nected transmission network. In addition, the effects of simultaneous transfers and parallel
path flows throughout the network must be addressed from a reliability viewpoint. Regardless
of the desire for commercial simplification, the laws of physics govern how the transmission
network will react to customer demand and generation supply. Electrical demand and supply
cannot, in general, be treated independently of one another. All system conditions, uses, and limits
must be considered to accurately assess the capabilities of the transmission network.

3. ATC calculations must recognize the dependency of ATC on the points of electric power
injection, the directions of transfers across the interconnected transmission network, and the
points of power extraction. All entities must provide sufficient information necessary for the
calculation of ATC. Electric power flows resulting from each power transfer use the entire
network and are not governed by the commercial terms of the transfer.

4. Regional or wide-area coordination is necessary to develop and post information that reason-
ably reflects the ATCs of the interconnected transmission network. ATC calculations must use
a regional or wide-area approach to capture the interactions of electric power flows among
individual, subregional, Regional, and multiregional systems.

5. ATC calculations must conform to NERC, Regional, subregional, power pool, and individual
system reliability planning and operating policies, criteria, or guides. Appropriate system con-
tingencies must be considered.

6. The determination of ATC must accommodate reasonable uncertainties in system conditions

and provide operating flexibility to ensure the secure operation of the interconnected net-
work. A Transmission Reliability Margin (TRM) may be necessary to apply this Principle.
Additionally, transmission capability (defined as Capacity Benefit Margin or CBM) may need to
be reserved to meet generation reliability requirements.

6 NERC Available Transfer Capability Definitions and Determination

TRANSMISSION TRANSFER CAPABILITY CONCEPTS

The key basic concepts of transmission transfer capability are described below. Numerous other terms
related to transfer capability are explored in detail in NERC’s May 1995 Transmission Transfer
Capability reference document. The concepts and terms in that document are still applicable in an open
transmission environment.

TRANSFER CAPABILITY
Transfer capability is the measure of the ability of interconnected electric systems to reliably move or
transfer power from one area to another over all transmission lines (or paths) between those areas under
specified system conditions. The units of transfer capability are in terms of electric power, generally
expressed in megawatts (MW). In this context, “area” may be an individual electric system, power pool,
control area, subregion, or NERC Region, or a portion of any of these. Transfer capability is also direc-
tional in nature. That is, the transfer capability from Area A to Area B is not generally equal to the transfer
capability from Area B to Area A.

TRANSFER CAPABILITY VERSUS TRANSMISSION CAPACITY

Electric systems throughout NERC have agreed to use common terminology to calculate and report trans-
mission transfer limits to maintain the reliability of the interconnected transmission networks. These
transfer values are called “capabilities” (differentiating them from “capacities”) because they are highly
dependent on the generation, customer demand, and transmission system conditions assumed during the
time period analyzed. The electric industry generally uses the term “capacity” as a specific limit or rating
of power system equipment. In transmission, capacity usually refers to the thermal limit or rating of a
particular transmission element or component. The ability of a single transmission line to transfer electric
power, when operated as part of the interconnected network, is a function of the physical relationship of
that line to the other elements of the transmission network.

Individual transmission line capacities or ratings cannot be added to determine the transfer capability of a
transmission path or interface (transmission circuits between two or more areas within an electric system or
between two or more systems). Such aggregated capacity values may be vastly different from the trans-
mission transfer capability of the network. Often, the aggregated capacity of the individual circuits of a
specific transmission interface between two areas of the network is greater than the actual transfer capa-
bility of that interface. In summary, the aggregated transmission line capacities of a path or interface do
not represent the transfer capabilities between two areas.

DETERMINATION OF TRANSFER CAPABILITY

The calculation of transfer capability is generally based on computer simulations of the operation of the
interconnected transmission network under a specific set of assumed operating conditions. These sim-
ulations are typically performed “off line,” well before the systems approach that operational state. Each
simulation represents a single “snapshot” of the operation of the interconnected network based on the
projections of many factors. As such, they are viewed as reasonable indicators of network performance and
available transfer capability. Among the factors considered in these simulations are:

Available Transfer Capability Definitions and Determination NERC 7

TRANSMISSION TRANSFER CAPABILITY CONCEPTS

€ Projected Customer Demands — Base case demand levels should be appropriate to the system
conditions and customer demand levels under study and may be representative of peak,
off-peak or shoulder, or light demand conditions.

€ Generation Dispatch — Utility and nonutility generators should be realistically dispatched for
the system conditions being simulated.

€ System Configuration — The base case configuration of the interconnected systems should be
representative of the conditions being simulated, including any generation and transmission
outages that are expected. The activation of any operating procedures normally expected to be
in effect should also be included in the simulations.

€ Base Scheduled Transfers — The scheduled electric power transfers that should be modeled
are those that are generally considered to be representative of the base system conditions being
analyzed and which are agreed upon by the parties involved.

€ System Contingencies — A significant number of generation and transmission system contin-

gencies should be screened, consistent with individual electric system, power pool,
subregional, and Regional planning criteria or guides, to ensure that the facility outage most
restrictive to the transfer being studied is identified and analyzed. The contingencies evaluated
may in some instances include multiple contingencies where deemed to be appropriate.

The conditions on the interconnected network continuously vary in real time. Therefore, the transfer capa-
bility of the network will also vary from one instant to the next. For this reason, transfer capability calcu-
lations may need to be updated periodically for application in the operation of the network. In addition,
depending on actual network conditions, transfer capabilities can often be higher or lower than those
determined in the off-line studies. The farther into the future that simulations are projected, the greater is
the uncertainty in assumed conditions. However, transfer capabilities determined from simulation studies
are generally viewed as reasonable indicators of actual network capability.

LIMITS TO TRANSFER CAPABILITY

The ability of interconnected transmission networks to reliably transfer electric power may be limited by
the physical and electrical characteristics of the systems including any one or more of the following:

€ Thermal Limits — Thermal limits establish the maximum amount of electrical current that a
transmission line or electrical facility can conduct over a specified time period before it
sustains permanent damage by overheating or before it violates public safety requirements.

€ Voltage Limits — System voltages and changes in voltages must be maintained within the
range of acceptable minimum and maximum limits. For example, minimum voltage limits can
establish the maximum amount of electric power that can be transferred without causing
damage to the electric system or customer facilities. A widespread collapse of system voltage
can result in a blackout of portions or all of the interconnected network.

8 NERC Available Transfer Capability Definitions and Determination

TRANSMISSION TRANSFER CAPABILITY CONCEPTS

€ Stability Limits — The transmission network must be capable of surviving disturbances

through the transient and dynamic time periods (from milliseconds to several minutes, respec-
tively) following the disturbance. All generators connected to ac interconnected transmission
systems operate in synchronism with each other at the same frequency (nominally 60 Hertz).
Immediately following a system disturbance, generators begin to oscillate relative to each
other, causing fluctuations in system frequency, line loadings, and system voltages. For the
system to be stable, the oscillations must diminish as the electric systems attain a new, stable
operating point. If a new, stable operating point is not quickly established, the generators will
likely lose synchronism with one another, and all or a portion of the interconnected electric
systems may become unstable. The results of generator instability may damage equipment and
cause uncontrolled, widespread interruption of electric supply to customers.

The limiting condition on some portions of the transmission network can shift among thermal, voltage, and
stability limits as the network operating conditions change over time. Such variations further complicate
the determination of transfer capability limits.

USES OF TRANSMISSION SYSTEMS

The interconnected transmission networks tie together major electric system facilities, generation resources,
and customer demand centers. They are planned, designed, and constructed to operate reliably within
thermal, voltage, and stability limits for the following purposes:

€ To Deliver Electric Power to Customers — Transmission networks must provide for the
reliable transfer of the electric power output from generation resources to customers under a
wide variety of operating conditions.

€ To Provide Flexibility for Changing System Conditions — Transmission capability must be

available on the interconnected network to provide flexibility to reliably handle the shift in
transmission facility loadings caused by maintenance and forced outages of generation and
transmission equipment, and a wide range of variable system conditions, such as higher than
expected customer demands, or construction delays of new facilities.

€ To Reduce the Need for Installed Generating Capacity — Transmission interconnections

between neighboring systems provide for the sharing of installed generating capacity, taking
advantage of the diversity in customer demands and generation availability over a wide area,
thereby reducing the amount of installed generating capacity necessary to meet generation
reliability requirements in each of the interconnecting systems.

€ To Allow Economic Exchange of Electric Power Among Systems — Transmission intercon-

nections between systems, coupled with internal system transmission facilities, allow for the
economic exchange of electric power among neighboring systems. Such economy transfers
help reduce the overall cost of electricity to customers.

Available Transfer Capability Definitions and Determination NERC 9

TTC DEFINITION AND DETERMINATION

DEFINITION OF TOTAL TRANSFER CAPABILITY

The Total Transfer Capability (TTC) between any two areas or across particular paths or interfaces is
direction specific and consistent with the First Contingency Total Transfer Capability (FCTTC) as defined
in NERC’s May 1995 Transmission Transfer Capability reference document.

TTC is the amount of electric power that can be transferred over the interconnected transmission network
in a reliable manner based on all of the following conditions:

1. For the existing or planned system configuration, and with normal (pre-contingency) operating
procedures in effect, all facility loadings are within normal ratings and all voltages are within
normal limits.

2. The electric systems are capable of absorbing the dynamic power swings, and remaining stable,
following a disturbance that results in the loss of any single electric system element, such as a
transmission line, transformer, or generating unit.

3. After the dynamic power swings subside following a disturbance that results in the loss of any
single electric system element as described in 2 above, and after the operation of any automatic
operating systems, but before any post-contingency operator-initiated system adjustments are
implemented, all transmission facility loadings are within emergency ratings and all voltages are
within emergency limits.

4. With reference to condition 1 above, in the case where pre-contingency facility loadings reach
normal thermal ratings at a transfer level below that at which any first contingency transfer limits
are reached, the transfer capability is defined as that transfer level at which such normal ratings are
reached.

5. In some cases, individual system, power pool, subregional, or Regional planning criteria or guides
may require consideration of specified multiple contingencies, such as the outage of transmission
circuits using common towers or rights-of-way, in the determination of transfer capability limits.
If the resulting transfer limits for these multiple contingencies are more restrictive than the single
contingency considerations described above, the more restrictive reliability criteria or guides must
be observed.

DETERMINATION OF TOTAL TRANSFER CAPABILITY

The concepts for determining transfer capability described in NERC’s Transmission Transfer Capability
reference document are still valid and do not change with the advent of open transmission access or the
need to determine ATCs. The major points contained therein are briefly outlined below.

System Conditions
Base system conditions are identified and modeled for the period being analyzed, including projected
customer demands, generation dispatch, system configuration, and base scheduled transfers. As
system conditions change, the base system conditions under which TTC is calculated may also need to
be modified.

10 NERC Available Transfer Capability Definitions and Determination

TTC DEFINITION AND DETERMINATION

Critical Contingencies
During transfer capability studies, many generation and transmission system contingencies throughout
the network are evaluated to determine which facility outages are most restrictive to the transfer being
analyzed. The types of contingencies evaluated are consistent with individual system, power pool,
subregional, and Regional planning criteria or guides. The evaluation process should include a variety
of system operating conditions because as those conditions vary, the most critical system contingencies
and their resulting limiting system elements could also vary.

System Limits
As discussed earlier, the transfer capability of the transmission network may be limited by the physical
and electrical characteristics of the systems including thermal, voltage, and stability considerations.
Once the critical contingencies are identified, their impact on the network must be evaluated to deter-
mine the most restrictive of those limitations. Therefore, the TTC becomes:

TTC = Minimum of {Thermal Limit, Voltage Limit, Stability Limit}

As system operating conditions vary, the most restrictive limit on TTC may move from one facility or
system limit to another as illustrated in Figure 1.

Figure 1: Limits to Total Transfer Capability

Available Transfer Capability Definitions and Determination NERC 11

TTC DEFINITION AND DETERMINATION

Parallel Path Flows

When electric power is transferred across the network, parallel path flows occur. This complex electric
transmission network phenomenon can affect all systems of an interconnected network, especially
those systems electrically near the transacting systems. As a result, transfer capability determinations
must be sufficient in scope to ensure that limits throughout the interconnected network are addressed.
In some cases, the parallel path flows may result in transmission limitations in systems other than the
transacting systems, which can limit the transfer capability between the two contracting areas.

Non-Simultaneous and Simultaneous Transfers

Transfer capability can be determined by simulating transfers from one area to another independently
and non-concurrently with other area transfers. These capabilities are referred to as “non-simul-
taneous” transfers. Another type of transfer capability reflects simultaneous or multiple transfers
concurrently. These capabilities are developed in a manner similar to that used for non-simultaneous
capability, except that the interdependency of transfers among the other areas is taken into account.
These interdependent capabilities are referred to as “simultaneous” transfers. No simple relationship
exists between non-simultaneous and simultaneous transfer capabilities. The simultaneous transfer
capability may be lower than the sum of the individual non-simultaneous transfer capabilities.

12 NERC Available Transfer Capability Definitions and Determination

TRANSMISSION TRANSFER CAPABILITY MARGINS

Two types of transmission transfer capability margins include:

€ Transmission Reliability Margin (TRM) — to ensure the secure operation of the interconnected
transmission network to accommodate uncertainties in system conditions.
€ Capacity Benefit Margin (CBM) — to ensure access to generation from interconnected systems to
meet generation reliability requirements.

Individual systems, power pools, subregions, and Regions should identify their TRM and CBM procedures
used to establish such transmission transfer capability margins as necessary. TRM and CBM should be
developed and applied as separate and independent components of transfer capability margin. The specific
methodologies for determining and identifying necessary margins may vary among Regions, subregions,
power pools, individual systems, and load serving entities. However, these methodologies must be well
documented and consistently applied.

TECHNICAL BASIS
Electric systems historically have recognized the need for and benefits of transfer capability margins in the
planning and operation of the interconnected transmission networks. In addition to meeting obligations for
service to native load customers and deliveries for third-party transmission users, some reserve transmis-
sion transfer capability is required to ensure that the interconnected network is secure under a wide range of
uncertain operational parameters. Also, systems have relied upon transmission import capability, through
interconnections with neighboring systems, to reduce their installed generating capacity necessary to meet
generation reliability requirements and provide reliable service to native load. With the introduction of
mandatory, non-discriminatory access, and the resulting need to identify and provide current and projected
ATCs to the competitive electric power market, a need now exists to formally address these two types of
transmission transfer capability margins.

This report provides a framework to support the development of transfer capability margin procedures.
TRM and CBM are concepts that may need to be further developed for general applicability while allowing
for tailoring to specific Regional, subregional, power pool, and individual system conditions. As these
margin concepts are developed and applied, NERC will review their implementation and consider the need
for further guidance.

DEFINITION OF TRANSMISSION RELIABILITY MARGIN

Transmission Reliability Margin (TRM) is defined as that amount of transmission transfer capability
necessary to ensure that the interconnected transmission network is secure under a reasonable range of
uncertainties in system conditions.

TRM provides a reserve of transfer capability that ensures the reliability of the interconnected transmission
network. All transmission system users benefit from the assurance that transmission services will be
reliable under a broad range of potential system conditions. TRM accounts for the inherent uncertainty in
system conditions and their associated effects on TTC and ATC calculations, and the need for operating
flexibility to ensure reliable system operation as system conditions change.

Available Transfer Capability Definitions and Determination NERC 13

TRANSMISSION TRANSFER CAPABILITY MARGINS

Uncertainty in TTC and ATC Calculations

TTC and ATC determinations depend upon a myriad of assumptions and projections of system con-
ditions, which may include such items as transmission system topology, projected customer demand
and its distribution, generation dispatch, location of future generators, future weather conditions, avail-
able transmission facilities, and existing and future electric power transactions. Such parameters are
assembled to produce a scenario to be used to project transfer capabilities under a reasonable range of
transmission contingencies as specified in Regional, subregional, power pool, and individual system
reliability operating and planning policies, criteria, or guides. Therefore, calculations of future TTCs
and ATCs must consider the inherent uncertainties in projecting such system parameters over longer
time periods. Generally, the uncertainties of TTC and ATC projections increase for longer term
projections due to greater difficulty in being able to predict the various system assumptions and
parameters over longer time periods. For instance, locations of future customer demands and
generation sources are often quite uncertain, and these parameters have a potentially large impact on
transfer capabilities. Similarly, future electric power transactions are inherently uncertain and can have
significant impacts on transmission loadings. Therefore, the amount of TRM required is time
dependent generally with a larger amount necessary for longer time projections than for near-term
conditions. TRM must also have wide-area coordination.

Need for Operating Flexibility

TTC and ATC calculations must recognize that actual system conditions may change considerably in
short periods of time due to changing operating conditions, and cannot be definitively projected without
the provision of a transfer capability margin. These operational conditions include changes in dispatch
of generating units, simultaneous transfers scheduled by other systems that impact the particular area
being studied, parallel path flows, maintenance outages, and the dynamic response of the intercon-
nected systems to contingencies (including the sudden loss of generating units).

DEFINITION OF CAPACITY BENEFIT MARGIN

Capacity Benefit Margin (CBM) is defined as that amount of transmission transfer capability reserved by
load serving entities to ensure access to generation from interconnected systems to meet generation
reliability requirements. Reservation of CBM by a load serving entity allows that entity to reduce its
installed generating capacity below that which may otherwise have been necessary without interconnections
to meet its generation reliability requirements.

The CBM is a more locally applied margin than TRM, which is more of a network margin. As such, to the
extent a load serving entity maintains policies and procedures to reserve transfer capability for generation
reliability purposes, the CBM should be included in the reserved or committed system uses in the calcu-
lation of ATC. These CBMs should continue to be a consideration in transmission system development. It
is anticipated that individual load serving entities and regional planning groups will continue to address
CBMs and that the NERC and Regional reviews of generation adequacy will continue to consider this
capability. It is also anticipated that load serving entities will develop additional procedures for reserving
transfer capability for generation capacity purposes and include these procedures in Regional planning
reviews and regulatory filings as appropriate.

14 NERC Available Transfer Capability Definitions and Determination

ATC DEFINITION AND DETERMINATION

DEFINITION OF AVAILABLE TRANSFER CAPABILITY

Available Transfer Capability (ATC) is a measure of the transfer capability remaining in the physical trans-
mission network for further commercial activity over and above already committed uses. Mathematically,
ATC is defined as the Total Transfer Capability (TTC) less the Transmission Reliability Margin (TRM),
less the sum of existing transmission commitments (which includes retail customer service) and the
Capacity Benefit Margin (CBM). ATC can be expressed as:

ATC = TTC – TRM – Existing Transmission Commitments (including CBM)

The ATC between two areas provides an indication of the amount of additional electric power that can be
transferred from one area to another for a specific time frame for a specific set of conditions. ATC can be a
very dynamic quantity because it is a function of variable and interdependent parameters. These param-
eters are highly dependent upon the conditions of the network. Consequently, ATC calculations may need
to be periodically updated. Because of the influence of conditions throughout the network, the accuracy of
the ATC calculation is highly dependent on the completeness and accuracy of available network data.

DETERMINATION OF AVAILABLE TRANSFER CAPABILITY

The determination of ATCs and the relationship of electric power transactions and associated power flows
on the transmission network are described in Appendixes A and B. The ATC calculation methodologies
described in Appendixes A and B are not intended to prescribe a specific calculation procedure nor do they
describe the only methods of calculating ATCs. Each Region, subregion, power pool, and individual
system will have to consider the ATC Principles in this report and determine the best procedure for
calculating ATCs based upon their respective circumstances.

Appendix A describes an ATC calculation approach that may be termed a “network response” method.
This method is intended to be illustrative of a procedure that is applicable in highly dense, meshed
transmission networks where customer demand, generation sources, and the transmission systems are
tightly interconnected.

Appendix B describes another ATC calculation approach that may be referred to as a “rated system path”
method. This method is intended to be illustrative of a procedure that is applicable in so-called sparse
transmission networks where the critical transmission paths between areas of the network have been
identified and rated as to their achievable transfer loading capabilities for a range of system conditions.

COMMERCIAL COMPONENTS OF AVAILABLE TRANSFER CAPABILITY

To more fully define ATC, specific commercial aspects of transmission service must be considered.
Because the terms “firm” and “non-firm” are used somewhat loosely within the electric industry, confusion
often exists when these terms are used to characterize the basic nature of transmission services. To create
reasonably consistent expectations regarding the transmission services that are being offered, the concepts
of curtailability and recallability are introduced.

Available Transfer Capability Definitions and Determination NERC 15

ATC DEFINITION AND DETERMINATION

Curtailability
Curtailability is defined as the right of a transmission provider to interrupt all or part of a transmission
service due to constraints that reduce the capability of the transmission network to provide that trans-
mission service. Transmission service is to be curtailed only in cases where system reliability is
threatened or emergency conditions exist. Curtailment procedures, terms, and conditions will be
identified in the transmission service tariffs. When such constraints no longer restrict the transfer
capability of the transmission network, the transmission service may be resumed. Curtailment does not
apply to situations in which transmission service is discontinued for economic reasons.

Recallability
Recallability is defined as the right of a transmission provider to interrupt all or part of a transmission
service for any reason, including economic, that is consistent with FERC policy and the transmission
provider’s transmission service tariffs or contract provisions.

Based on the recallability concept, two commercial applications of ATC are defined below and
depicted graphically in Figure 2. They are as follows:

Figure 2: TTC, ATC, and Related Terms in the Transmission Service Reservation System

16 NERC Available Transfer Capability Definitions and Determination

ATC DEFINITION AND DETERMINATION

€ Non-recallable Available Transfer Capability — Non-recallable ATC (NATC) is defined as

TTC less TRM, less non-recallable reserved transmission service (including CBM). NATC has
the highest priority use of the transmission network. The maximum amount of non-recallable
service that can be reserved is determined based on what the network can reliably handle under
normal operating conditions and during appropriate contingencies as defined in NERC, Region-
al, subregional, power pool, and individual system reliability operating and planning policies,
criteria, or guides. Any lower priority service can be displaced by non-recallable service that is
either new non-recallable service or non-recallable service that had been reserved but not
scheduled.

Mathematically, NATC can be expressed as:

NATC = TTC – TRM – Non-recallable Reserved Transmission Service (including CBM)

€ Recallable Available Transfer Capability — Recallable ATC (RATC) is defined as TTC less
TRM, less recallable transmission service, less non-recallable transmission service (including
CBM). Portions of the TRM may be made available by the transmission provider for recallable
use, depending on the time frame under consideration for granting additional transmission
service. To the extent load serving entities reserve transmission transfer capability for CBM,
portions of CBM may be made available for recallable use, depending on the time frame under
consideration for granting additional transmission service.

RATC has the lowest priority use on the transmission network and is recallable subject to the
notice provisions of the transmission service tariffs. Recallable reserved service may be recalled in
favor of subsequent requests for non-recallable transmission service. However, recallable reserved
service has precedence over subsequent requests for recallable transmission service, unless the
tariff or contract provisions specify otherwise. Because RATC is recallable on short notice, it can
use the transfer capability reserved for higher priority service that has been reserved but not
scheduled.

RATC must be considered differently in the planning and operating horizons. In the planning
horizon, the only data available are recallable and non-recallable transmission service reservations,
whereas in the operating horizon transmission schedules are known.

Mathematically, RATC can be expressed as:

a) Planning Horizon

RATC = TTC
– a(TRM)
– Recallable Reserved Transmission Service
– Non-recallable Reserved Transmission Service (including CBM)

where 0 < a < 1, value determined by individual transmission providers based on

network reliability concerns.

Available Transfer Capability Definitions and Determination NERC 17

ATC DEFINITION AND DETERMINATION

b) Operating Horizon

RATC = TTC
– b(TRM)
– Recallable Scheduled Transmission Service
– Non-recallable Scheduled Transmission Service (including CBM)

where 0 < b < 1, value determined by individual transmission providers based on

network reliability concerns.

NATC and RATC are depicted graphically in Figure 2. TTC, ATC, and related terms in the trans-
mission service reservation system are also shown in Figure 2. In general, the transition between the
planning and operating horizons will be a function of available information about the system, the status
of reserved and scheduled transmission services, and time considerations.

RECALLABLE AND NON-RECALLABLE RELATIONSHIPS AND PRIORITIES

The relationships and priorities of recallable and non-recallable concepts as they apply to both scheduled
and reserved transmission services are described below. In addition, the interaction between recallable and
non-recallable transmission services and the effects on ATC values are discussed and illustrated.

Scheduled and Reserved Transmission Service

Reserved transmission service constitutes a reserved portion of the transmission network transfer
capability, but the actual electric power transfer is not yet scheduled between areas. Scheduled
transmission service indicates that an electric power transfer will be occurring on the transmission
network for the time period for which the transmission service was reserved. Both terms can apply to
either recallable or non-recallable transmission service, giving the following four transmission service
terms:

– Non-recallable Reserved (NRES)

– Non-recallable Scheduled (NSCH)

– Recallable Reserved (RRES)

– Recallable Scheduled (RSCH)

The aggregate of the NSCH and RSCH must never exceed the TTC in the operational horizon.
However, in the planning horizon, individual transmission providers may allow the aggregate of the
NRES and RRES to exceed the TTC less TRM, to more fully utilize transmission assets, provided that
NRES by itself never exceeds TTC less TRM. Such over-subscription of recallable reservations must
be disclosed to the purchasers of RRES. These ATC relationships are shown in Figure 3.

18 NERC Available Transfer Capability Definitions and Determination

ATC DEFINITION AND DETERMINATION

Transfer Capabilities Transmission Services

TTC — Total Transfer Capability NRES — Non-recallable Reserved

ATC — Available Transfer Capability NSCH — Non-recallable Scheduled

RATC — Recallable ATC RRES — Recallable Reserved

NATC — Non-recallable ATC RSCH — Recallable Scheduled

Figure 3: ATC Relationships

Transmission Service Priorities

Non-recallable and recallable transmission service must adhere to a standard set of priorities
universally applied throughout the electric power market to avoid confusion. These priorities are
described below.

– Non-recallable service has priority over recallable service. Recallable transfers, reserved or
scheduled, may be recalled for non-recallable requests. Recallability will generally be applied as
needed only in areas of network constraint and not unilaterally over the entire network.

– All requests for transmission service will be evaluated in priority as established by applicable
transmission service tariffs.

– Reserved transfer capability may be used by recallable scheduled transfers, provided that those
scheduled transfers can be recalled if the reserved transfer requester wants to make use of the
reserved transfer capability.

Available Transfer Capability Definitions and Determination NERC 19

ATC DEFINITION AND DETERMINATION

Several of the possible relationships of NATCs and RATCs to the different types of transfers that have
been scheduled or reserved during a given time period are shown in Figure 4 and described below.
These concepts apply to any time during the forecast period. Therefore, no time aspect is identified.

Transfer Capabilities Transmission Services

TTC — Total Transfer Capability NRES — Non-recallable Reserved

ATC — Available Transfer Capability NSCH — Non-recallable Scheduled

RATC — Recallable ATC RRES — Recallable Reserved

NATC — Non-recallable ATC RSCH — Recallable Scheduled

Figure 4: ATC Relationships and Priorities

20 NERC Available Transfer Capability Definitions and Determination

ATC DEFINITION AND DETERMINATION

– Non-recallable scheduled (NSCH) transfers are of the highest priority (all Examples). NSCH
transfers cannot be curtailed by the transmission provider except in cases where system reliability
is threatened or an emergency exists. All NSCH transfers reduce the amount of ATC.

– Recallable ATC (RATC) can include transfer capability that is currently held by non-recallable
reserved (NRES) transfers. However, the new transfers scheduled from the RATC may have to be
interrupted if the NRES transfer requester wants to make use of the transmission network
(Example 1).

– Non-recallable ATC (NATC) cannot include transfer capability that is currently held by non-
recallable reserved (NRES) transfers because the reserved transfer would have priority over any
new non-recallable transfer (Examples 1 and 3).

– Non-recallable ATC (NATC) can include transfer capability that is currently used by recallable
scheduled (RSCH) transfers because a non-recallable transfer has priority over recallable transfers
(Example 3).

– Recallable ATC (RATC) cannot include transfer capability that is currently used by recallable
scheduled (RSCH) transfers because the scheduled transfer would have priority over any new
transfers (Examples 2 and 3).

– Both non-recallable ATC (NATC) and recallable ATC (RATC) can include recallable reserved
(RRES) transfers (all Examples). However, any new recallable transfers may have to be inter-
rupted if the RRES requester wants to make use of the transmission network (Examples 2 and 3).

The Examples in Figures 3 and 4 illustrate how ATC may be applied in the conduct of commercial
business. These definitions have no impact on the physical determination of how much additional
transfers the network can support.

Appendix C further demonstrates the interaction between recallable and non-recallable transmission
service and the effects on ATC values.

Available Transfer Capability Definitions and Determination NERC 21

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

The example in this Appendix describes an ATC calculation approach that may be termed a “network
response” method. It demonstrates the ATC Principles described in this report and the physical impacts of
electric power transfers on an interconnected transmission network. The method is intended to be illus-
trative of a procedure that is applicable in highly dense, meshed transmission networks where customer
demand, generation sources, and the transmission systems are tightly interconnected. In such networks,
transmission paths critical to a particular electric power transfer cannot generally be identified in advance.
The critical path will be very much a function of the conditions that exist at the time the transfer is
scheduled. The example does not introduce any concepts not covered in the front or main portion of this
report.

PHYSICAL SYSTEM IMPACTS OF TRANSFERS

Determination of ATC requires some translation from the area to area transactions to the resultant electric
power flows on the transmission network. This translation is done by stressing the system with appropriate
transfers under critical contingencies to determine the characteristic response of the network. These
network response characteristics, which are based on the line outage, power transfer, and outage transfer
distribution factors of NERC’s May 1995 NERC Transmission Transfer Capability reference document,
can be determined by transfer capability studies either beforehand, or on a transaction-by-transaction basis.

When electric power is transferred between two areas such as Area A to Area F in Figure A1, the entire
network responds to the transaction. The power flow on each transmission path will change in proportion
to the response of the path to the transfer. Similarly, the power flow on each path will change depending on
network topology, generation dispatches, customer demand levels, other transactions through the area, and
other transactions that the path responds to which may be scheduled between other areas.

Figure A1: Network Response Characteristics for Area A to Area F Transfers

22 NERC Available Transfer Capability Definitions and Determination

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

To illustrate this, computer simulation studies are performed to determine the transfer capability from Area
A to Area F. During that process, it is determined that 77% of electric power transfers from Area A to
Area F will flow on the transmission path between Area A and Area C (Figure A1).

Through application of those response characteristics, the impact on the path between Area A and Area C
for a 500 MW transfer from Area A to Area F is graphically described in Figure A2. In this example, a
pre-existing 160 MW power flow exists from Area A to Area C due to a generation dispatch and the
location of customer demand centers on the modeled network. When a 500 MW transfer is scheduled from
Area A to Area F, an additional 385 MW (77% of 500 MW) flows on the transmission path from Area A
to Area C, resulting in a 545 MW power flow from Area A to Area C.

Figure A2: Existing vs. Resultant Power Flows

on Path A to C for a 500 MW Transfer from Area A to Area F

To determine the ability of the network to transfer electric power from Area A to Area F, additional
potential impacts within the individual areas must also be recognized. The network responses shown in
Figure A1 must be expanded to consider possible transmission limits within each area.

The response characteristics of limiting facilities within the individual areas for an Area A to Area F
transfer are shown in Figure A3. For simplicity, the flows within each area are not shown. Rather, the
figures within each area represent the percentage of the transfer from Area A to Area F that flows on the
most limiting facility within each area. Recognition of the limiting path responses within the individual
areas for Area A to Area F transfers increases the complexity of determining the Area A to Area F ATC.

Available Transfer Capability Definitions and Determination NERC 23

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

Figure A3: Internal and Interconnection Responses to Area A to Area F Transfers

TRANSLATION OF SYSTEM IMPACTS TO ATC

The ATC of the network depends on the existing loading conditions on the limiting transmission facility,
wherever it may be, taking into account contingency criteria (i.e., outage of the most critical line or
generator or multiple lines and generators, as appropriate).

ATC is a function of how much unused or unloaded capacity is available on the most limiting transmission
facility, allowing for single and, in some cases, multiple contingencies. The translation of the unused capa-
bility of the transmission network to ATC determination for a particular direction is illustrated in Table 1,
which refers to the transmission network shown in Figure A3 for an Area A to Area F transfer. The unused
capacity of individual facilities in the transmission network, which is the difference between a facility’s
rating and its current power flow loading or its “available loading capacity,” is divided by the response
characteristic of the path facility to an Area A to Area F transfer. This procedure provides the individual
critical path ATCs (in a system or between systems) from which the ATC from Area A to Area F is then
determined by considering the most limiting path ATC. In this case, the limiting path is in Area D and the
Area A to Area F ATC is 1,200 MW.

For a different electric power transfer, a new set of network responses and a new set of available capacity
on limiting facilities would need to be determined to define the ATC for that transfer.

Electric power transfers have historically been scheduled between control areas on a contract path or area
interchange basis. However, in the determination of ATCs, the actual flows on the network must be
considered regardless of the scheduling methodology. In the preceding example, an electric power transfer
may be scheduled from Area A to Area F, using a contract path from Area A to Area C to Area F.
However, the reality of alternating current electrical systems is that the electric power would flow from
Area A to Area F over the entire network, governed by the laws of physics. The electric power flowing on
portions of the network other than the scheduled contract path is known as parallel path

24 NERC Available Transfer Capability Definitions and Determination

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

Available Loading
Capacity (ALC) System or Area A to
System Network on Limiting Path Area F
or Response Facility ATC ATC
Path (%) (MW) (MW) (MW)

Area A 2 35 1,750

Area B 4 92 2,300

Area C 12 454 3,780

Area D 15 180 1,200 1,200

Area E 8 200 2,500

Area F 5 250 5,000

A–C 77 1,000 1,300

A–D 7 157 2,240

A–B 16 440 2,750

B–D 16 512 3,200

C–E 11 198 1,800

C–D 1 18 1,800

C–F 67 1,072 1,600

D–F 22 385 1,750

E–F 11 214 1,945

Table 1: Available Transfer Capability Matrix for Transfers from Area A to Area F

flows, and can affect many systems in an interconnected network. In this particular example, the
transmission limit in Area D limits the Area A to Area F transfers to 1,200 MW.

ATC TIME VARIATION AND NETWORK DEPENDENCY

Network conditions will vary over time, changing line loading conditions, and causing the ATC of the
network to change. Also, the most limiting facility in determining the network’s ATC can change from one
time period to another, particularly in highly meshed networks. Therefore, the ATC of the network is time
dependent.

Available Transfer Capability Definitions and Determination NERC 25

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

This characteristic is illustrated conceptually in Figure A4. The first group of graphs on the left-hand side
of the figure presents the available loading capacity at different points in time (T1, T2, T3) for several lines
in an interconnected network. If an Area A to Area B transfer is to be scheduled at T1, each of the lines
(line 1 in Area A, line 3 in Area B, line 7 in Area B, and line 16 in Area D) will respond to the transfer in
accordance with its network response factor. This factor is used to determine an ATC as limited by each
individual facility. The results are shown on the middle set of diagrams of Figure A4. The ATC for the
network as a whole represents the minimum of the ATCs as defined by each facility at each time frame.
These minimum ATCs are schematically illustrated in the right side of Figure A4. As demonstrated, the
ATC is different for each time period and is determined by a different facility in each period.

Figure A4: ATC Variance

26 NERC Available Transfer Capability Definitions and Determination

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

The determination of ATC and the difference between simultaneous and non-simultaneous transfers are
demonstrated in Tables 2 and 3. These ATC demonstrations are based on the sample six system network
shown in Figure A3.

Network ALC* on Limiting Area to

Area Facility Response Facility Area ATC
(%) (MW) (MW)
Area A to Area F Transfer
D D1 15 180 1,200
Area B to Area E Transfer
B B1 5 25 500
Area E to Area A Transfer
A A2 7 103 1,470
*Available Loading Capacity

Table 2: Non-Simultaneous ATC Analyses

Table 2 presents the non-simultaneous ATC analyses for three representative transfer conditions: Area A to
Area F, Area B to Area E, and Area E to Area A. For each transfer direction, the area to area ATC is
determined by the most critical system contingency and the resultant limiting system element, varying from
500 MW for an Area B to Area E transfer (limited by line B1 in Area B) to 1,470 MW for an Area E to
Area A transfer (limited by line A2 in Area A).

Network ALC* on Limiting Area to

Area Facility Response Facility Area ATC
(%) (MW) (MW)
Area A to Area F ATC Analysis
With a Pre-existing Area B to Area E 500 MW Transfer
B B1 3.5 0 0
Area A to Area F ATC Analysis
With a Pre-existing Area B to Area E 500 MW Transfer
and a Pre-existing Area E to Area A 1,470 MW Transfer
B B1 3.5 40 1,140
*Available Loading Capacity

Table 3: Simultaneous ATC Analyses

Available Transfer Capability Definitions and Determination NERC 27

APPENDIX A. NETWORK RESPONSE METHOD FOR ATC DETERMINATION

The first section of Table 3 shows a determination of ATC for an Area A to Area F transfer, assuming that
an Area B to Area E 500 MW transfer schedule is already in effect. Under this condition, the Area A to
Area F ATC is now reduced from 1,200 MW (Table 2) to zero. This change is due to the increased loading
on line B1 due to the previously scheduled 500 MW transfer from Area B to Area E, making it the limiting
network facility. Note that the Area A to Area F transfer limiting facility was line D1 in Area D in the non-
simultaneous analysis (Table 2).

The second portion of Table 3 is another determination of ATC for an Area A to Area F transfer. In this
example, pre-existing transfers are in place from Area B to Area E of 500 MW and Area E to Area A of
1,470 MW. Under these conditions, the ATC for an Area A to Area F transfer is found to be 1,140 MW.
This transfer is a slight reduction from the 1,200 MW ATC determination in the non-simultaneous case
(Table 2), but is a significant increase from the zero ATC found in the previous case (first part of Table 3).
This increased transfer is due to the offsetting effect of the flows caused by the pre-existing Area E to Area
A transfer, which reduced the line loading on the critical facility B1, thus increasing the ATC for the Area
A to Area F transfer direction.

These examples demonstrate that the determination of ATC in a tightly interconnected network is very
much a function of system conditions that exist on the network at the time the transfer is to be scheduled.
In addition, ATC is a function of the specifics of the electric power transfer being considered in terms of its
direction, amount, and duration. To be able to properly appraise the performance of tightly interconnected
networks to support contemplated transfers (i.e., what is the ATC), a regional or wide-area approach must
be considered so that all network conditions are properly taken into account.

28 NERC Available Transfer Capability Definitions and Determination

APPENDIX B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION

OVERVIEW
The rated system path (RSP) method for ATC determination is typically used for transmission systems that
are characterized by sparse networks with customer demand and generation centers distant from one
another. Generally in this approach, paths between areas of the network are identified and appropriate
system constraints determined. ATC is computed for these identified paths and interconnections between
transmission providers.

The RSP method involves three steps: 1) determining the path’s Total Transfer Capability (TTC),
2) allocating the TTC among owners in a multi-owned path to determine the owners’ rights, and
3) calculating ATC for each right-holder by subtracting each of their uses from each of their individual
TTC rights. Wide-area coordination is achieved by developing the TTC in a manner that follows a regional
review process. This process assures individual, power pool, subregional, and Regional coordination and
the necessary consideration of the interconnection network’s constraints and conditions.

The RSP method includes a procedure for allocating TTC, and in turn ATC, among the owners of the trans-
mission path(s). It should be noted that the RSP method of allocation is not the only procedure that may be
followed in allocating transmission services.

UNSCHEDULED FLOW OR PARALLEL PATH FLOW

The RSP approach accounts for the effects of unscheduled flow (parallel path flow) on interconnected
systems through the modeling of realistic customer demand and generation patterns in advance of real-time
operations, and uses a maximum power flow test to ensure that the transfer path is capable of carrying
power flows up to its rated transfer capability or TTC.

The rating process begins by modeling the interconnected network with the actual flow that will occur on
the path and its parallel paths under realistically stressed conditions. The lines comprising the path may be
rated and operated as a single path. The network is tested under a wide range of generation, customer
demand, and facility outage conditions to determine a reliability-based TTC. When determined this way,
the TTC rating usually remains fairly constant except for system configuration changes such as a line
outage situation. To implement the RSP ATC method, consistent path rating methods and procedures must
be agreed upon and followed within the Interconnection.

Non-simultaneous ratings are normally used as the basis for calculating ATC. If, however, two rated paths
have a simultaneous effect on each other, the rating process identifies the simultaneous capabilities or
establishes nomograms that govern the simultaneous operation of the paths. Applicable operating
procedures are negotiated to ensure reliable network operation. Where simultaneous operation is necessary,
operator control is used to ensure safe and reliable operation of the transmission network.

Available Transfer Capability Definitions and Determination NERC 29

APPENDIX B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION

CAPACITY ALLOCATION
The reliability-based TTC of a transfer path (its reliability rating) is allocated among the right-holders
based upon their negotiated agreement. This determination of the property rights through the allocation
process is critical to the RSP implementation of ATC. The rights in the path are negotiated for each of the
individual transmission providers. Except for deratings based upon system operating (e.g., emergency)
conditions, these allocations become rights that the right-holder may use or resell to others as non-
recallable or recallable service.

Although the actual flows from each right-holder’s schedule will flow on all parallel lines, the advance
allocation of rights on a path makes it possible for right-holders to determine ATC and sell service within
their rights independent of others. If the rating is determined using appropriate path rating procedures,
including a maximum power flow test, the potential for adverse unscheduled power flow effects is
minimized.

In real time, neither the total of the schedules, nor the actual power flow on a path may exceed the path
TTC. Although the potential for adverse unscheduled power flow is minimized as a result of the modeling
and rating process, some acceptable or mitigatable unscheduled flows will usually occur during real-time
operation. Regions that use RSP to calculate ATC should adopt an unscheduled flow mitigation plan
which addresses such flows, if they adversely affect system operation. The adverse flows can be managed
through schedule changes, installing controllable devices such as phase shifters, or including this
uncertainty as part of the reliability margin.

ATC CALCULATION APPROACH

1. Each path for which ATC must be calculated is identified, and then a reliability-based TTC is
determined as described above. This TTC is then allocated among the owners by negotiated
agreement.

2. Deratings for outages, nomograms, maintenance, or unscheduled flow are allocated, if necessary, to
the right-holders based on prearranged agreements or tariffs.

3. Right-holders take their respective allocated shares of the TTC for a path and subtract the existing
commitments to determine the appropriate ATC.

4. Right-holders update and repost their ATC calculations as new commitments impact their ATC. A
transfer from one area to another involving several transmission owners requires locating and
reserving capacity across multiple paths and potentially multiple right-holders.

EXAMPLE OF ATC DETERMINATION

The following example illustrates the application of the RSP method for determining ATC in a sparse
network. The example transmission network is shown in Figure B1. All paths that connect the various
areas have transfer capabilities that were individually developed in coordination with all areas giving
consideration to unscheduled flow and interconnection interactions and effects. The TTCs portrayed in

30 NERC Available Transfer Capability Definitions and Determination

APPENDIX B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION

Figure B2 are shown for each path and are directional, but are not necessarily the same for each direction.

Figure B1: Sparse Network Model

Figure B2: Total Transfer Capabilities

Each path may consist of several transmission lines that can also have different owners. In this example,
the path between Areas B and D is comprised of five lines as shown in Figure B3. The TTC from Area B
to Area D is 7,500 MW and, in the reverse direction, 8,800 MW. Line 1 is owned by a single entity and
has an allocated portion of the TTC equal to 1,300 MW in either direction.

Available Transfer Capability Definitions and Determination NERC 31

APPENDIX B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION

Figure B3: TTCs of Path B to D and Allocation of TTCs to Line 1

This example reflects a snapshot in time during the planning horizon. Initial transmission service
reservations, all assumed to be non-recallable, are shown for each path in Figure B4. The corresponding
ATC for each path has been calculated by subtracting the non-recallable service from the TTC. Because all
the transmission service reservations are assumed on each path to be in one direction, the path ATC is only
calculated for that direction.

Figure B4: Initial Transmission Service Reservations

32 NERC Available Transfer Capability Definitions and Determination

APPENDIX B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION

For example, referring to Figure B4, the ATC from Area B to Area D is calculated as 7,500 MW less 4,000
MW or 3,500 MW. For line 1 of the B to D path, the ATC is 1,300 MW less 200 MW or 1,100 MW. In
the next case, as shown in Figure B5, 1,000 MW of non-recallable transmission service is acquired from
Area A to Area B to Area D. No other changes occur. The total transmission service reserved from Area A
to Area B is 1,500 MW, and the resulting ATC goes to zero. The ATC from Area B to Area D reduces to
2,500 MW (7,500 MW TTC less 5,000 MW reserved transmission service). It is assumed the 1,000 MW
of the new reserved transmission service was obtained from the owner of line 1, resulting in the total
reserved transmission service on this line being 1,200 MW. The new ATC for line 1 is 100 MW (1,300
MW TTC less 1,200 MW reserved transmission service).

Figure B5: New Transmission Service Reservation on Path A to B to D

The non-recallable transmission service reserved for a path in each direction may not exceed the path’s
transfer capability in either direction under any circumstances. These limits are consistent with NERC
Operating Policies.

Unscheduled flow may at times preclude scheduling to a path’s full transfer capability or TTC. If an
internal limit is encountered in any system as a result of the transaction from Area A to Area D, for
example in Area D, Area D’s system operator must respond to relieve the limitation such as by redis-
patching generation or using phase shifter control. An unscheduled flow mitigation plan might also be
implemented to relieve excessive unscheduled flow problems. Additional relief may be achieved by
curtailing schedules that are contributing to the unscheduled flow on the path or by increasing schedules
that would create unscheduled flow in the opposite direction. In this example, if path A to D were limiting,
unscheduled flow mitigation procedures could be implemented to initiate coordinated operation of
controllable devices such as phase-shifting transformers to relieve the limitation.

Available Transfer Capability Definitions and Determination NERC 33

APPENDIX B. RATED SYSTEM PATH METHOD FOR ATC DETERMINATION

There will probably be times in the operating horizon when the use of the transmission network results in
actual flows on a transmission path being less than the transmission scheduled on the path. During these
periods, if the transmission path is fully scheduled, additional electric power may be scheduled to Area D
from Area A by reserving transmission service over a different transmission path. In this case, trans-
mission service could be obtained from either the owners of the direct path between Area A and Area D or
the owners of the transmission system from Area A to Area C to Area D.

For the RSP method, the transmission rights to be reserved and scheduled by all transmission users are
consistent with the rating of the transmission paths. If determined through a coordinated process using
models that capture the major effects of the interconnected network, these ratings will create limits that
result in the reliable operation of the regional electric system. The owners of the transmission paths,
through a negotiated allocation process, will know their transmission service rights and the resulting use of
these rights will be consistent with the physical capability and limitations of the transmission network.
This RSP method assures efficient use and reliable operation of the interconnected transmission network.

34 NERC Available Transfer Capability Definitions and Determination

APPENDIX C. TRANSMISSION SERVICE RESERVATIONS AND SCHEDULING

OVERVIEW
The following scenarios demonstrate how the 1,200 MW ATC quantity from Area A to Area F in the
example in Appendix A may be commercially employed. The interplay between recallable and non-
recallable transmission service and the resulting effects upon calculated ATCs are demonstrated using the
equations presented in the “ATC Definition and Determination” section of this report. They clearly
demonstrate that, although both recallable and non-recallable ATC are offered simultaneously, the
combined total of recallable and non-recallable service does not exceed the TTC at any time.

For the purpose of this illustration, assume that conditions on the interconnected network are as described
in Tables 1 and 2 of Appendix A. Under this scenario, the network ATC from Area A to Area F for this
time period in the operating horizon is 1,200 MW. Also, for simplicity, assume that previous transmission
commitments are zero. Thus, TTC in the following cases is 1,200 MW. Lastly, assume that TRM is zero.
The resulting relevant simplified ATC equations for the operating horizon are:

NATC = TTC – NRES

RATC = TTC – RSCH – NSCH

The equations that describe the TTC constraints during this time frame are:

NRES < TTC

RSCH + NSCH < TTC

ATC DEMONSTRATION — SCENARIO 1

Consider the initial case identified in Figure C1 as Case 1. Reservations for 200 MW of recallable and 400
MW of non-recallable transmission service have been reserved against the 1,200 MW TTC.

Case 1 includes schedules for only 300 MW of non-recallable transmission service. Thus:

NATC = TTC – NRES

= 1,200 – 400
= 800 MW

RATC = TTC – RSCH – NSCH

= 1,200 – 0 – 300
= 900 MW

Available Transfer Capability Definitions and Determination NERC 35

APPENDIX C. TRANSMISSION SERVICE RESERVATIONS AND SCHEDULING

Transfer Capabilities Transmission Services

TTC — Total Transfer Capability NRES — Non-recallable Reserved

ATC — Available Transfer Capability NSCH — Non-recallable Scheduled

RATC — Recallable ATC RRES — Recallable Reserved

NATC — Non-recallable ATC RSCH — Recallable Scheduled

Figure C1: ATC Demonstration — Scenario 1

In Scenario 1, the transmission customer reserves an additional 100 MW of recallable transmission service
and schedules the entire 300 MW recallable reservation. The results are shown in Figure C1 as Case 2.
(Note that changed values are shown in bold italic type.) Non-recallable ATC is unchanged, but recallable
ATC is changed as follows:

RATC = TTC – RSCH – NSCH

= 1,200 – 300 – 300
= 600 MW

36 NERC Available Transfer Capability Definitions and Determination

APPENDIX C. TRANSMISSION SERVICE RESERVATIONS AND SCHEDULING

Transfer Capabilities Transmission Services

TTC — Total Transfer Capability NRES — Non-recallable Reserved

ATC — Available Transfer Capability NSCH — Non-recallable Scheduled

RATC — Recallable ATC RRES — Recallable Reserved

NATC — Non-recallable ATC RSCH — Recallable Scheduled

Figure C2: ATC Demonstration — Scenario 2

ATC DEMONSTRATION — SCENARIO 2

In Scenario 2 of Figure C2, the transmission customer reserves an additional 400 MW of non-recallable
transmission service. The results are shown in Figure C2 as Case 3. Recallable ATC is unchanged in this
scenario, but non-recallable ATC is changed as follows:

NATC = TTC – NRES

= 1,200 – 800
= 400 MW

Available Transfer Capability Definitions and Determination NERC 37

APPENDIX C. TRANSMISSION SERVICE RESERVATIONS AND SCHEDULING

Transfer Capabilities Transmission Services

TTC — Total Transfer Capability NRES — Non-recallable Reserved

ATC — Available Transfer Capability NSCH — Non-recallable Scheduled

RATC — Recallable ATC RRES — Recallable Reserved

NATC — Non-recallable ATC RSCH — Recallable Scheduled

Figure C3: ATC Demonstration — Scenario 3

ATC DEMONSTRATION — SCENARIO 3

In Scenario 3 of Figure C3, the transmission customer reserves and schedules an additional 300 MW of
non-recallable transmission service. The results are shown in Figure C3 as Case 4. In this scenario, both
recallable and non-recallable ATCs are changed as follows:

NATC = TTC – NRES RATC = TTC – RSCH – NSCH

= 1,200 – 1,100 = 1,200 – 300 – 600
= 100 MW = 300 MW

Transmission customers holding the 200 MW of recallable transmission service reservations “above the
TTC line” should be advised that they have a high probability of having their transmission service recalled.

38 NERC Available Transfer Capability Definitions and Determination

APPENDIX C. TRANSMISSION SERVICE RESERVATIONS AND SCHEDULING

Transfer Capabilities Transmission Services

TTC — Total Transfer Capability NRES — Non-recallable Reserved

ATC — Available Transfer Capability NSCH — Non-recallable Scheduled

RATC — Recallable ATC RRES — Recallable Reserved

NATC — Non-recallable ATC RSCH — Recallable Scheduled

Figure C4: ATC Demonstration — Scenario 4

ATC DEMONSTRATION — SCENARIO 4

In Scenario 4 of Figure C4, the transmission customer schedules an additional 400 MW of non-recallable
transmission service. The results are shown in Figure C4 as Case 5. Non-recallable ATC remains
unchanged at 100 MW. Unless 100 MW of recallable transmission service schedules are recalled, the total
schedules violate the TTC constraint. The transmission provider must recall 100 MW of scheduled
recallable transmission service. The recallable ATC calculation is then:

RATC = TTC – RSCH – NSCH

= 1,200 – 200 – 1,000
= 0 MW

As this demonstration has shown, recallable transmission services may be reduced as non-recallable trans-
mission services are reserved and scheduled, approaching the TTC limit.

Available Transfer Capability Definitions and Determination NERC 39

TRANSMISSION TRANSFER CAPABILITY TASK FORCE

Raymond M. Maliszewski (Chairman) Nicholas A. Brown (SPP)

Senior Vice President–System Planning Director, Engineering & Operations
American Electric Power Service Corporation Southwest Power Pool
Columbus, Ohio Little Rock, Arkansas

Garey C. Rozier (Vice Chairman) Harlow R. Peterson (WSCC)

Director of Bulk Power Supply Consultant Planning Analyst
Southern Company Services, Inc. Salt River Project
Atlanta, Georgia Phoenix, Arizona

Robert W. Cummings (ECAR) John L. Griffin (WSCC)

Manager, Transmission Services Senior Engineer/Policy Analyst
East Central Area Reliability Coordination Grid Customer Services
Agreement Pacific Gas and Electric Company
Canton, Ohio San Francisco, California

Lee E. Westbrook (ERCOT) William C. Phillips (Operating Committee

Transmission Grid Planning Manager Liaison)
TU Electric Director, Power System Operations
Dallas, Texas Entergy Services, Inc.
Pine Bluff, Arkansas
Bruce M. Balmat (MAAC)
Manager, System Services & Performance Robert Wayne Mitchell (Operating Committee
PJM Interconnection Association Liaison Alternate)
Norristown, Pennsylvania System Operations Analyst III
Entergy Services, Inc.
T.J. (Jim) Barker (MAIN) Pine Bluff, Arkansas
Manager–Electric System Engineering Department
Central Illinois Public Service Company C.M. (Marty) Mennes (NERC Electronic
Springfield, Illinois Information Network Task Force)
Director Power Supply
Terry L. Bundy (MAPP, Municipal, and Florida Power & Light Company
Transmission Dependent Utility) Miami, Florida
Manager, Power Supply Division
Lincoln Electric System Paul F. Barber (Power Marketer)
Lincoln, Nebraska Vice President, Transmission and Engineering
Citizens Lehman Power L.P.
Edward A. Schwerdt (NPCC and Canada) Boston, Massachusetts
Director of Engineering
Northeast Power Coordinating Council Trudy Utter (Independent Power Producer)
New York, New York Vice President & General Manager
Tenaska Power Services Company
W.T. (Terry) Boston (SERC and Federal) Arlington, Texas
Division Manager, Electric System Reliability
Tennessee Valley Authority Virginia C. Sulzberger (NERC Staff)
Chattanooga, Tennessee Director–Engineering
North American Electric Reliability Council
Princeton, New Jersey

40 NERC Available Transfer Capability Definitions and Determination

 Appendix III

ETSO - Definitions of Transfer Capacities in Liberalised

Electricity Markets

60
Definitions of Transfer Capacities

Definitions of Transfer Capacities in liberalised

Electricity Markets

Final Report
April 2001

April 2001 Page 1/13

Definitions of Transfer Capacities

Table of contents

SUMMARY ................................................................................................................. 3

1. INTRODUCTION .................................................................................................. 4

2. TRANSFER CAPACITY DEFINITIONS ......................................... ……………….5

3. USE OF THE DEFINITIONS ................................................................................ 8

3.1 DIRECTIONAL AND TIME DEPENDENCIES ........................................... ……………….8

3.2 PLANNING PHASE .................................................................................. ……………….9
3.3 ALLOCATION PHASES............................................................................ ……………….9

4. EXCHANGE PROGRAMMES AND PHYSICAL FLOWS .................................. 10

4.1 THE "PARALLEL FLOW" PHENOMENA ................................................ ……………….10

4.2 NOTIONS FOR PHYSICAL FLOWS....................................................... ……………….12

REFERENCES ......................................................................................................... 13

April 2001 Page 2/13

Definitions of Transfer Capacities

SUMMARY

ETSO worked out in 1999 definitions of transfer capacities that are since then used by the
European TSOs for capacity calculations on a harmonised basis. Two important notions in
these definitions are the Total Transfer Capacity TTC and the Net Transfer Capacity NTC.
The ETSO definitions are the basis for the half yearly calculation of indicative NTC values by
the TSOs that are also published on ETSO’s webside as well as for additional or more
frequent calculations needed by the TSOs for the allocation of interconnection capacities.
The practical application of the definitions raised some ambiguities in interpretation.
Therefore, ETSO worked on an improvement of those definitions that is summarised in the
present document.
The general concept of the definitions given by the notions of TTC and NTC was confirmed
and maintained. It was enlarged introducing the aspects of transfer capacity assessments for
different time frames: For the planning TTC and NTC assessment are the main objectives of
TSOs. Market actors need these values to anticipate and to plan their transactions.
During the allocation phases, that can cover according to the rules applicable at each cross-
border or “flowgate” time frames from year ahead to day ahead or even hour ahead, a set of
new notions is introduced. These are the Already Allocated Capacity AAC and the still
Available Transfer Capacity ATC.
All these values are to be interpreted in terms of exchange programmes between adjacent
areas.
In the highly meshed interconnected transmission networks in Europe programmed
exchanges and physical flows differ often considerably. The ETSO work confirmed that the
physical complexity is so big and needs always a complete view on the European load flow
scenarios that it would not be useful for market actors to try to make public additionally
physical values beside the above mentionned data. However, a separate set of notions of
physical flows was established in which every notion relates to one of the above mentionned
definitions for exchange programmes. This approach helps to easier agree between TSOs
about concrete values, to check their global consistency and to ensure, in a best way,
transparency towards decision-making or supervisory bodies like Regulators.
ETSO considers that the new set of definitions improves considerably the transparency in
this complex technical field and that the new definitions are in line with all used capacity
allocation mechanisms through out Europe. The new definitions will also allow work further
on new methods like a sophisticated auctioning mechanism.

April 2001 Page 3/13

Definitions of Transfer Capacities

1. INTRODUCTION

ETSO presented at the fourth Electricity Regulation Forum at Florence in November 1999
definitions of transfer capacities that are used between the European TSOs for capacity
calculations on a harmonised basis. They include the following notions:
Ø Net Transfer Capacity (NTC)
Ø Available Transfer Capacity (ATC)
Ø Transmission Reliability Margin (TRM)
Ø Notified Transmission Flow (NTF).

NTC and ATC are important indicators for market participants to anticipate and plan their
cross-border transactions and for the TSOs to manage these international exchanges of
electricity. ETSO has thus decided to publish twice a year in its Intranet system (www.etso-
net.org) a table “Indicative Values for Net Transfer Capacities (NTCs) in Europe” [1].
Additionally, ETSO edited in March 2000 a paper titled “Information for users” [2]. It provides
to the actors in the Internal Market of Electricity in Europe some basic explanations related to
the transfer capacity definitions and answers to frequently asked questions. Other
complementary documents were worked out and are at the disposal of the TSOs to calculate
NTC values, such as a technically oriented NTC/ATC user’s guide and a note on TRM
evaluation. All documents were presented at the different Electricity Regulation Forums at
Florence.
Given the fact that the assessment of transfer capacities in highly meshed interconnected
transmission networks like that in Europe is a very complex task and includes extensive load
flow calculations done by the TSOs, the “Information for users” tries to make this process
understandable also for non specialists.
However, as transfer capacities are one important factor that determines the possibilities of
access to market regions in Europe and thus of the international trade, it remains critical that
the TSOs do their calculations and assessments in the most transparent way, and that the
used definitions are well understood by all actors and well applicable by all decision making
bodies.
Therefore ETSO has worked since last year on improving the existing transfer capacity
definitions. This document summarises the results and provides an enlarged set of
definitions for transfer capacities. ETSO focussed its work on two aspects:
- The practical application of the existing definitions showed that they could lead to
misunderstandings, even by specialists of TSOs. The reason was that the distinction
between programmed transactions (scheduled exchanges) and physical flows was not
always clear.
- The possibilities for import/export transactions in the European transmission systems
between two countries depend on all realised transactions – also between others than
the two considered countries – due to the so called parallel flows which are the direct
consequence of physical laws of electrical flows in the interconnected networks. Thus the
maximum possible use of the capacity between two given countries depends to some
extent on all local as well as on all distant transactions, because they rely on the
European production plans and on the consumer loads.
ETSO tried first to work out new definitions in order to make the physical parallel flow effect
explicit. However, this would lead to complex notions that are not useful for the current
capacity allocation mechanisms and non transparent for market actors.
Therefore ETSO maintained the general concept for the existing transfer capacity definitions
and included the following improvements:

April 2001 Page 4/13

Definitions of Transfer Capacities

- Clear distinction between programmed values and physical flows: The new set of
definitions includes only notions that are to be interpreted as energy programmes. Market
actors are interested to plan the trade between regions or countries and do not like to
worry about load-flow problems. It is on the other hand the task of TSOs to manage
physical flows and to maintain at every moment the security in their networks. This
complex task should be handled inside the TSOs and not be an obstacle for the market.
- Improvements concerning the transfer capacity assessment methods used by the TSOs:
It is necessary to have a harmonised basic procedure between TSOs for calculating
transfer capacities. This means that the basic scenarios have to be commonly agreed
and that the calculation procedures of all TSOs are comparable. This approach helps to
easier agree between TSOs about concrete values, to check their global consistency and
to ensure in a best way transparency towards decision-making or supervisory bodies like
Regulators.
- Definitions well suited for planning of trade as well as for capacity allocation procedures:
The market needs first of all information for planning purposes. The NTC-tables
published by ETSO are a first important step. On the other hand, a clear set of definitions
is needed in order to calculate the transfer capacities (e.g. at weekly or daily time frames)
that are the basis for the different allocation procedures already implemented or on the
way to be implemented, such as auctions and market splittings.
- Applicability of the transfer capacity definitions for new allocation procedures. ETSO has
worked out in November 2000 a first vision about a future large-scale capacity allocation
scheme, so called “Co-ordinated auctioning of transmission capacity in meshed
networks” [6]. The new definitions should fit also for a future implementation of such a
mechanism.
Based on these considerations ETSO proposes to use on a European-wide level the
following transfer capacity definitions and assessment guidelines. This document replaces
the previous ETSO-paper on definition of transfer capacities [1].
Additionally ETSO will publish two other papers on this subject:
- A document on the procedures for transfer capacity assessments. This document will
give the detailed guidelines for the TSOs on how to construct realistic base cases,
and how to perform the calculations and the capacity evaluations.
- An updated version of the information for users [2].

2. TRANSFER CAPACITY DEFINITIONS

ETSO proposes to strictly distinguish between commercial and physical values. Thus, two
sets of definitions exist, one related to programme values, the other to physical flows. The
definitions that refer to programme values are presented in this document in detail. These
values are important for market actors to prepare their commercial transactions. The physical
complexity however should be dealt with by the TSOs. TSOs are responsible towards
Regulators, Authorities etc. to carry out this work in a fair and non-discriminatory matter.
Market actors should not be involved in this process.
Some basic explanations concerning the relations between transfer capacities in terms of
exchange programmes and physical flows are given in chapter 4.2.
The fundamental notions in ETSO’s transfer capacity definitions were not changed and are:

April 2001 Page 5/13

Definitions of Transfer Capacities

The Total Transfer Capacity TTC, that is the maximum exchange programme between two
areas compatible with operational security standards1 applicable at each system if future
network conditions, generation and load patterns were perfectly known in advance.

TTC is always related to a given power system scenario, i.e.: generation schedule,
consumption pattern and available network that constitute the data allowing to build up a
mathematical model of the power system (load flow equations). The solution of this model
leads to the knowledge of the voltages at the network nodes and the power flows in the
network elements which are the parameters being monitored by a TSO to assess system
security. The solution of this model is the so-called base case and is the starting point for
TTC computation. Thus evaluation of TTC between two electrical areas requires:

- To make a choice of a local power system scenario

- To define a base case2, which involves the sharing of full information amongst TSOs to
build up the global load flow model
- To apply an agreed procedure for carrying out the calculations.

As the result of this procedure, TTC equals the maximum exchange programme between the
two areas being considered, if the generation and load pattern in these areas and in other
areas strongly interconnected to these two would exactly correspond to the assumptions
made in the evaluation steps, namely the ones implicit in the base case.
The uncertainties associated to the forecast of the power system state, for a given time
period in the future, may decrease according to the selected time frame. Therefore the TTC
value may vary (i.e. may increase or decrease) when approaching the time of programme
execution as a result of a more accurate knowledge of generating unit schedules, load
pattern, network topology and tie-lines availability.

The Transmission Reliability Margin TRM which is a security margin that copes with
uncertainties on the computed TTC values arising from:
a) Unintended deviations of physical flows during operation due to the physical functioning
of load-frequency regulation
b) Emergency exchanges between TSOs to cope with unexpected unbalanced situations in
real time
c) Inaccuracies, e. g. in data collection and measurements

TRM is then associated to the real-time operation and its value is determined by each TSO,
in order to guarantee the operation security of its own power system. TRM may vary
seasonally or may be updated according to possible modifications occurred in the power
system.

1
The security standards are stated into each TSO ‘grid code’.
2
The agreed procedures for building the base cases to be used by TSOs and to evaluate the
transmission capacities are detailed in a separate ETSO document ‘Procedures for Transfer
Capacity Assessments’ [3].

April 2001 Page 6/13

Definitions of Transfer Capacities

The Net Transfer Capacity NTC that is defined as:

NTC = TTC-TRM

NTC is the maximum exchange programme between two areas compatible with security
standards applicable in both areas and taking into account the technical uncertainties on
future network conditions.

TTC, TRM and NTC may vary along different time frames (year ahead to day ahead).

NTC may be allocated in different time frames to match the need for securing longer term
trading and to provide room for shorter term trading. One may distinguish, as the result of the
allocation procedures in each allocation time frame, two notions:

The Already Allocated Capacity AAC, that is the total amount of allocated transmission
rights, whether they are capacity or exchange programmes depending on the allocation
method.

The Available Transmission Capacity ATC, that is the part of NTC that remains available,
after each phase of the allocation procedure, for further commercial activity. ATC is given by
the following equation:
ATC = NTC- AAC

AAC and ATC are thus a result of each stage of the allocation procedure.

The following figure 1 gives an overview over the transfer capacity definitions.

+
TRM

+
ATC
Transfer
+ + direction
TTC NTC
+ (I.e. A to B)
AAC

0 MW
-
AAC
- Opposite
TTC - direction
NTC
- (I.e. B to A)
ATC

-
TRM

Planning Allocation

Figure 1: Transfer capacity definitions

April 2001 Page 7/13

Definitions of Transfer Capacities

3. USE OF THE DEFINITIONS

3.1 DIRECTIONAL AND TIME DEPENDENCIES.

As noted in the previous chapter, TTC computation starts establishing a base case. This
base case will already contain exchange programmes between any pair of neighbour control
areas. These are the various transactions (long term to spot contracts) likely to exist –
according to what has been observed in the past – in the forecasted situation. In figure 2 for
a given pair of neighbour control areas, A and B, for which capacities are to be computed,
there exists in the base case a global exchange programme of magnitude BCE (Base Case
Exchange).

Time variations of
TTC, TRM and NTC

+
TRM
+
∆Emax

+ +
TTC NTC Ie.: Yearly, monthly,
daily allocations
BCE

0 MW

- - -
∆Emax TTC NTC

-
TRM

Time

Planning Allocation
phases

Figure 2: Transfer capacities in planning and for allocation phases.

From this starting situation, when computing TTC from area A to area B, generation is
stepwise increased in control area A and decreased in control area B giving rise to a power
flow from area A to area B .The shifts of generation are named in figure 2 as ∆E+ and ∆E- for
the increase and the decrease respectively. This process is carried out up to the point, where
security rules in either system A or B3 are breached (∆Emax+ / ∆Emax-). The maximum
exchange from A to B compatible with security rules without taking into account uncertainties
and inaccuracies – TTC from A to B – is then BCE + ∆Emax+. This procedure is reversed –
decrease of generation in system A and increase of generation in system B – when
computing TTC from area B to area A leading to a maximum increase of generation of

3
The breaching of security rules may happen internally in any of these two systems or in the tie lines
between them. It has to be beared in mind that the interconnector is not just the tie lines crossing
the control areas borders but any network element which has a real impact upon the real transfer
possibilities i.e., which may limit the exchange programmes.

April 2001 Page 8/13

Definitions of Transfer Capacities

∆Emax- and thus to a maximum exchange from B to A of ∆Emax- - BCE, as shown in figure
2. In a next step TRM is deducted from the TTC values for both directions resulting in the
corresponding NTC values.
The values of TTC, TRM and NTC are therefore directional. They are to be computed for a
given interconnection in both directions of the energy exchange. Generally, the values of
TTC, TRM and NTC in both directions are bound to be different. The values of AAC and ATC
are as well directional; they are nothing else that a split of NTC established through the
allocation procedure.
The scope of the transfer capacity definitions covers two sets of power system scenarios:
One associated with the planning phase and another associated with the capacity allocation
phases. The former is based on estimates of typical situations of the power system. The
latter takes into account the exchange programmes already allocated in a given time frame
(one year ahead until one day-ahead) and updates of the assumptions made on network
topology, load and generation pattern. Starting from calculations for the operational planning
and approaching the horizon of programme execution, the base cases normally vary with the
consequence that also the values of TTC, TRM and NTC may vary. Thus the transfer
capacities are also time dependent. These time dependencies are further discussed in the
following chapters.

3.2 PLANNING PHASE

For the planning phase, which is the one leading to the seasonal capacities published by
ETSO, the scenario corresponds to two typical peak periods, one in winter and the other in
summer. The corresponding base cases are built according to observed states of the power
system in the past (‘snapshots’ or recorded load flows of the power system in these
situations) and are adapted according to a set of agreed guidelines to the expected system
states corresponding to the forecasted situation.
The calculations of TTC and TRM lead then to NTC values that are indicative and non-
binding. These values are just the best estimate by TSOs, providing a signal for market
participants who should understand it as a reference value which will, sometimes, have to be
adapted when approaching the programming horizon accordingly to the prevailing system
conditions and in the extent these differ from the forecasted system conditions built in the
base cases.

3.3 ALLOCATION PHASES

Interconnection capacity is allocated in different time frames in an attempt to match the

needs of market parties for securing longer term trading and to provide room for shorter term
trading. The split of capacity into different time frames may also reflect the fact that
transmission capacities, being estimates based on power system forecasts, may vary
following unexpected changes of power flows derived from the results of each allocation
stage. The time frames considered (i.e.: year ahead to day ahead or even hours ahead) and
the split of transmission capacity between them is a matter of ‘capacity products’ and is
embedded into the allocation procedures applied by the concerned TSOs and, when
applicable, of concerned PXs.
For the allocations, which are carried out regionally by the TSOs involved in a congested flow
gate, the selected scenarios vary depending on the time frames considered. For the long-
term allocation phases (year-ahead) the scenarios may be the same as those considered for

April 2001 Page 9/13

Definitions of Transfer Capacities

the planning phase. For shorter time frames (month ahead to day ahead) the scenarios
correspond to the expected situations for peak and off-peak system conditions for the next
day or even to hourly scenarios. The corresponding base cases have to be built integrating
the results of the previous allocation phases and the expected system states corresponding
to the forecasted horizon, i.e.: Once, for example, the one year ahead allocation mechanisms
have provided access rights to a set of transactions these have to be integrated into the base
cases allowing computation of the TTC for, for example, the month ahead allocation process.
In most cases, when the allocation procedure reaches the day ahead all the exchange
programmes with allocated capacity have to be confirmed (‘use or lose it’ principle). The
confirmed programmes define CE (‘Confirmed Exchanges’) in both directions of exchanges.
Whether these CE values are to be ‘netted’ or not in order to define ATC is a matter of the
allocation procedure not of capacity definitions.

4. EXCHANGE PROGRAMMES AND PHYSICAL FLOWS

4.1 THE “PARALLEL FLOW” PHENOMENA

The above set of capacity parameters are in terms of bilateral exchange programmes
between two neighbour areas. They would be closely connected to the power flows through
the cross borders only in the ideal case of a peninsular system and its neighbour if both were
interconnected through a single tie line. However, in a widely interconnected network like for
example the UCTE network the power flow through the cross border tie lines between two
neighbour areas A and B may be interpreted as a superposition of a direct flow, which is
related to exchanges between A and B and a parallel flow, which is related to all the other
exchanges in the meshed network and to the location of generations and loads in the several
grids. Therefore there would be a parallel flow even if all the exchanges in the interconnected
system were set at zero 4.
The ‘parallel flows’ are dealt with implicitly in the capacity assessment procedure in the
sense that the base cases already contain scheduled cross-border exchanges and the
corresponding load flow situations contain the associated power flows.

Thus the figures provided about capacities for highly meshed systems are limited in scope, in
several senses:

• TTC and NTC values are computed between neighbour areas; these values are the
result of assuming that only the transactions between these two areas are modified and
the rest (‘third parties’ transactions) remain unaltered. This fact has two consequences:
- The published values cannot be used for an exact planning of transactions if these do
not correspond to generation and to consumption in the pair of control areas for which
capacities are defined. I.e.: NTCs cannot be combined to derive possibilities of
executing transactions according to a given transaction path (contract path).
- If the pattern of ‘third party transactions’ differs noticeably from that taken into
account in the forecast, TTC values may significantly differ. That may have a
important impact upon the NTC value.

4
Exchanges between A and any other system than B, between B and any other system than A,
exchanges between any other pair of control areas and the sometimes so called ‘natural flows’,
which appear even in the case of no exchanges between any pair of control areas due to the
generation and load pattern of the grids.

April 2001 Page 10/13

Definitions of Transfer Capacities

• NTC values between pairs of control areas in meshed network systems are
interdependent. For planning and for the sake of simplicity normally only one set of NTC
values, that do not reflect NTC interdependencies between several borders, is published.
In case of strong NTC interdependencies, better information can be provided by
additionally computing values of transfer capacities for groups of areas. I.e., if there is a
strong physical coupling between areas A and B regarding exchanges to area C, NTC
would be provided from area A to area C, from area B to area C and from areas A + B,
considered as a whole, to C. However, during the allocation phases the coupling between
the areas has to be respected. Allocation thus may lead to new restrictions as shown in
the following figure 3.

Area A Area B

2000 4000
NTC values
in MW

Limit A + B to C: 5000
Area C

Figure 3: Interdependencies of NTC between two araes

In the figure it is assumed that in the planning phase the NTC value between areas A and
C was assessed to 2000 MW and that idenpendently from this the calculation of NTC
between areas B and C lead to a value of 4000 MW. For planning purposes the TSOs
thus have given to market participants maximum values, not reflecting the
interdependencies between the areas. Indeed the sum of import to area C may be limited
to only 5000 MW. Then, at least during the allocations this fact has to be taken into
consideration. It is out of the scope of this document to define the criteria for the split of
this total value into the capacity for allocation from A to C and from B to C.
• Finally the NTC values itself do not provide the basis for a co-ordinated method of
allocating cross border trade over several borders in a meshed network. A vision for a co-
ordinated approach was already presented in a separate ETSO document [4]. It would
relay on the same computation principles as outlined in this paper, but the allocation of
transfer capacities would be effected on the basis of the consequences in terms of load
flows and not directly using the bilateral values of NTC. Therefore, the importance that
NTC values actually have in the transaction based concepts of the international trade in
Continental Europe will diminish.

April 2001 Page 11/13

Definitions of Transfer Capacities

4.2 NOTIONS FOR PHYSICAL FLOWS

Figure 4 provides an example of the relationship between exchange programmes and

physical flows.

Program Physical
Values Flows

TRM
+
∆ Emax
∆Fmax

TTC
NTC
TTF

BCE NTF

Inputs Load Flow Outputs

Calculation

Shift of generation
until ∆Emax+ Base Case

Check Security Constraints

Figure 4: Definitions of transfer capacities and physical flows

This relationship is established through the load flow model of the whole interconnected
system. All the terms that appear on the right hand side of this figure are net values of
physical power flows. By net values is meant the sum of individual tie line power flows. The
following terms define cross-border capacities as physical flows:

The Total Transfer Flow TTF is the net physical flow across the border associated with a
programme exchange of magnitude TTC, provided that no other exchanges have been
modified with respect to the ones existing in the base case. In this limited context – the one
which applies to TTC computation – TTF may be understood as the physically maximum
cross-border flow compatible with security standards in each control area5. TTF may be
greater or smaller than TTC.

TTF can be split into two terms:

• The Notified Transmission Flow NTF, which is the physical flow over the tie lines
between the considered areas observed in the base case prior to any generation shift
between the areas. It results from the flow originated by the base case exchange (BCE)

5
The breaching of security rules may happen internally in any of these two systems or in the tie lines
between them. It has to be beared in mind that the interconnector is not just the tie lines crossing
the control areas borders but any network element which has a real impact upon the real transfer
possibilities i.e., which may limit the exchange programmes.

April 2001 Page 12/13

Definitions of Transfer Capacities

and from the parallel flows. It is extremely difficult and often even impossible to identify
the different origins of parallel flows that lead to the NTF value and to separate them into
distinguished terms (such as loop flows, natual flows) because of non-linear physical
phenomena in the networks. On the other hand, a split of NTC is technically not
necessary for the procedure of transfer capacity assessments and would also not be
relevant for market actors.
• The physical flow ∆Fmax that is the physical flow over the tie lines between the
considered areas induced by the maximum generation shift ∆Emax.

Thus, the total transfer flow can also be expressed as:

TTF = NTF + ∆Fmax

REFERENCES:

[1] Indicative values for Net Transfer Capacities (NTC) in Europe, winter and summer,
working day, peak hours, ETSO-publication twice a year

[2] NTC an ATC in the IEM, information for user, ETSO, March 2000

[3] Procedures for Transfer Capacity Assessments, ETSO, in elaboration

[4] Co-ordinated Auctioning of Transmission Capacity in Meshed Networks, Discussion

paper, ETSO, November 2000

April 2001 Page 13/13

 Appendix IV

ETSO - Key Concepts and Definitions for Transmission Access

Products

61
Key Concepts and Definitions for Transmission Access Products

Key Concepts and Definitions for

Transmission Access Products

Final Report
April 2001

April 2001 Page 1/20

Key Concepts and Definitions for Transmission Access Products

CONTENTS

Executive Summary

Introduction

Background
The purpose of the present paper

Transmission Product Characteristics

Rights
Obligations
Location model
Path model
Location resolution
Constraint representation
Firmness
Duration

Florence Guidelines

Economic efficiency
Directional price signals
Netting
Unused capacity
Firm capacity
Transaction independence
Timing
Risk assignment
Tradability

Current Auction Schemes

Characterisitics of current schemes

Limitations of current methods

Co-ordinated Auctions

Harmonisation Requirements

Conclusions

Recommendations

April 2001 Page 2/20

Key Concepts and Definitions for Transmission Access Products

EXECUTIVE SUMMARY

This paper describes a range of possible Transmission Access Products in terms of a set of
basic characteristics. The paper is focused on congestion management methods which
involve the allocation (most likely by auction) and/or trading of these products; it does not
address market splitting, which is the subject of a separate ETSO paper.

The basic characteristics are used to identify the types of products which would be consistent
with the Congestion Management Guidelines contained within the conclusions of the
November 2000 Florence Regulatory Forum.

The same concepts are then used to characterise the transmission capacity auction
schemes which have already been introduced in the IEM area. The methods used have
been successful in rapidly introducing market-based methods into congestion management.
However, the analysis indicates that they do not comply with all of the Florence Guidelines,
and they are unlikely to be generally effective if extended in their present form to the whole of
the IEM. In particular, more sophisticated methods will be necessary where the transmission
system is highly meshed and individual auctions would become strongly interactive. A
possible method of dealing with this situation is described in a separate ETSO paper.

Subject to further consideration of the feasibility of market splitting, the paper recommends
that further regional auction schemes should be encouraged. However, a significant level of
harmonisation and coordination will be necessary, and it is proposed that ETSO should
facilitate this by providing guidance on the characteristics of products and allocation
methods.

April 2001 Page 3/20

Key Concepts and Definitions for Transmission Access Products

INTRODUCTION

Background

1. Effective congestion management is regarded as essential for successful operation of

the European Internal Electricity Market (IEM). It is therefore one of the main topics
being progressed by representatives of governments, the European Commission
(EC), regulators, transmission users, transmission system operators (TSOs) and
other interested parties through the ‘Florence process’.

2. The conclusions of the Florence Regulatory Forum held in November 2000 contain a
set of guidelines on congestion management. These require TSOs to develop
congestion management methods which meet a number of high level principles.
There is now a need to translate these into practical proposals.

3. Some practical progress has already been made. Market splitting methods have
been operating well in Scandinavia for some time. More recently, some neighbouring
TSOs have jointly auctioned capacity across a small number of national borders.
This experience indicates that both market splitting and auctions can provide effective
market-based congestion management, at least in the particular circumstances to
which they have been applied.

4. There are no plans to impose uniform market processes throughout Europe. It is

therefore assumed that pan-European congestion management will develop through
evolution and co-ordination of initially separate regional schemes. Each of these will
necessarily be designed to interface with local market arrangements.

The purpose of the present paper

5. Given the situation described above, the way forward depends on answers to
questions such as the following:
- do current congestion management methods comply with the Florence
Guidelines?
- can neighbouring congestion management schemes be made to work
together, even if they use different methods?
- what degree of harmonisation is necessary for effective co-ordination?

This paper begins to address such questions for congestion management methods
which involve the allocation (most likely by auction) and/or trading of transmission
access products. Market splitting is addressed in a separate ETSO paper.

6. The approach taken is to describe the range of possible access products in terms of
their basic characteristics. These are then used to compare currently used auction
methods with the requirements implied by the Florence Guidelines. This leads to
conclusions on the type of harmonised access products which would be both effective
and compliant with the Guidelines.

April 2001 Page 4/20

Key Concepts and Definitions for Transmission Access Products

TRANSMISSION PRODUCT CHARACTERISTICS

Rights

7. A ‘right’ enables a market participant to derive financial benefit from operating at, or
below, a particular generation, consumption or transaction level. Rights have positive
value to market participants, and will therefore attract a positive price if they are in
limited supply.

8. Transmission rights may be implemented physically, for example by requiring a user

to prove that he holds a right before scheduling a transaction through scarce
transmission capacity. Alternatively, they may be implemented financially by relieving
holders of additional congestion charges when they transmit specified amounts of
power. This document assumes that rights are implemented financially.

Obligations

9. An ‘obligation’ requires a market participant to provide a service by operating at, or

above, a given generation, consumption or transaction level. Obligations have
negative value to market participants, and they may therefore require payment.

10. A right and an obligation can be combined in the same access product. A right and
an obligation provides the TSO(s) with a measure of assurance that the market
participant will operate at a given level, rather than above or below it. The
combination of both therefore provides the most effective congestion management
product, and in particular may permit opposing flows to be ‘netted’. The market price
of the combined right and obligation will depend on its value to the marginal
participant.

Location model

11. Transmission products are most straightforwardly defined in terms of the two
locations at the ‘ends’ of an energy transaction. In general, these could be
generation, consumption or trading (eg power exchange) locations.

12. To remove the transaction dependence, all products can be referred to a common
‘hub’ location. Each product is then characterised by only one location, and
counterparty locations need not be revealed. These are referred to below as
‘entry/exit’ products.

~
~

hub

Fig 1a: Location-to-location model Fig 1b: Entry/exit model

April 2001 Page 5/20

Key Concepts and Definitions for Transmission Access Products

13. More general models might also be feasible. For example, regional hubs could be
placed at energy trading locations and used to define entry/exit products for intra-
regional transactions. Inter-regional transactions might then be supported by further
products linking the regional hubs, perhaps with reference to a single inter-regional
hub.

regional
~ hub

regional ~
hub

inter-regional hub

Fig 2: Hierarchical entry/exit model

Path model

14. To date in both Europe and the US, trades between and across transmission systems
have usually been programmed and treated commercially in ‘contract path’ terms.
This approach simply identifies a chain of contiguous transmission areas, with no
reference to the multitude of flow paths (parallel flows) which exist in practice on a
meshed system. Except for the special case of peninsular boundaries, this failure to
consider physical flows limits the usefulness of contract path methods for congestion
management.

15. In general, accurate congestion management requires physical paths to be

quantified. This may be done using standard loadflow software, and the results
expressed with sufficient accuracy in the form of a matrix of factors. It is, however, a
relatively sophisticated process which needs to be repeated at least when the system
configuration changes as a result of switching operations.

~ ~

Fig 3a: Contract path Fig 3b: Physical path

April 2001 Page 6/20

Key Concepts and Definitions for Transmission Access Products

Location resolution

16. Locations can be represented as points (nodes) on the transmission system, or as

zones. Zones could, for example, be defined as groups of nodes, control areas or
countries.

17. In continental Europe, control of the UCTE power system has been structured and
organised by decentralising the load/generation balance using the ‘control area’
concept. Control areas are managed through the use of real-time automatic devices
(load-frequency controllers) which continuously adjust the generation level inside
each area according to programmed exchanges, the actual flows on interconnectors,
and the measured frequency. The control area concept is fundamental to UCTE
system operation and security. Consistency with current UCTE operational practice
would therefore suggest that zones should be identified with control areas.

18. The choice of location resolution is a trade-off between accurate constraint

representation and access market liquidity. If transmission products are defined for
large zones, relatively few constraints are captured but the number of participants
able to trade products within each zone is relatively high. Conversely, products
related to small zones or nodes are potentially capable of fine control of congestion,
but the opportunities for trading among market participants are limited or non-
existent.

Fig 4a: Nodal resolution Fig 4b: Zonal resolution

Constraint representation

19. A constraint (congestion) can arise on a transmission system for a variety of reasons.
It is often necessary to limit physical flows to reduce the risk of thermal damage,
unacceptable voltage levels, or electrical instability. Such risks may be significant
while operating as intended, or they may appear following transmission faults
resulting in loss of equipment. A particular constraint may be a relatively permanent
feature of the system, or it may appear only temporarily as a result of particular
operational circumstances. Transmission products could in principle be defined in
terms of the detailed technical nature of each constraint, but such a scheme would be
highly complex and unworkable in practice.

20. A common simplification is to limit active power flows across ‘bottlenecks’ in the
transmission system. These are typically groups of circuits in the same electrical
vicinity, and are sometimes called ‘flowgates’, particularly in the US. In a physical
path flowgate model, the sensitivities of flowgate loadings to entry/exit flows can be
represented by linear loadflow factors.

April 2001 Page 7/20

Key Concepts and Definitions for Transmission Access Products

21. A particular application of this principle is the identification of flowgates with tie-lines
linking adjacent zones, with flowgate limits which take into account congested lines
situated inside the zones as well as on the tie-lines themselves. The physical flow
through such ‘tie-line flowgates’ is then limited to the value reached by the flow on the
tie-lines when the flow through the most congested line is at its maximum. This
approach is compatible with the ‘control area’ concept (see paragraph 17), which in
turn is usually aligned with political boundaries.

control
area

Fig 5a: Flowgates Fig 5b: Tie-line flowgates

22. A further simplification is to limit flows across a set of partitions, each of which
divides the transmission system into two separate parts. The loadflow sensitivity
factors relating to each partition are then either 0 or 1, depending on whether both
counterparties (or the entry/exit point and the hub) are on the same or opposite sides
of the partition.

Fig 5c: Partitions

April 2001 Page 8/20

Key Concepts and Definitions for Transmission Access Products

23. The choice of constraint representation is again a compromise. A representation

which reflects the technical capabilities of individual circuits allows fixed limits to be
directly applied, but is complex and difficult to reconcile with a transparent and liquid
market. The limits corresponding to more simplified flowgate or partition
representations require assumptions to be made regarding the outturn disposition of
generation and demand, and are therefore approximations. Ultimately, partition
models suffer from the same disadvantages as contract paths and market splitting,
and are probably fully relevant in practice only to special regions in Europe.

24. Rights (and obligations) can be associated directly with individual constraints,
flowgates or partitions. Market participants then have the task of assembling a
portfolio of such products for each constrained part of the transmission system, in
proportions which are informed by loadflow factors. Such ‘flowgate rights’ have been
promoted by some as an alternative to locational pricing methods used in the US.
They devolve a high level of activity to the marketplace, but are complex for market
participants to manage, especially as the loadflow factors are subject to change.
Assuming that flowgate products are combined in proportions determined by accurate
loadflow factors, the resulting ‘bundles’ are equivalent to the location-to-location and
entry/exit products described above.

Firmness

25. For transmission products to be useful for congestion management, they should, at
least on a short-term (day-ahead) basis, be regarded as firm by system users. This
implies that there is some means of enforcing or incentivising users to operate in
conformance with their access product holdings. Direct physical control by TSOs is
incompatible with unbundled, competitive electricity markets. Monitoring and
disclosure (‘name and shame’) methods are unlikely to be either popular or fully
effective. Commercial sanctions are more desirable, although they require systems
for settlement of any differences between measured generation/consumption and
access product holdings.

26. For products which combine both rights and obligations, charges (or payments) for
both underruns and overruns are necessary to provide appropriate commercial
incentives. Imbalance prices should be related to the local value of access. In this
case full netting of counter-flows is possible.

27. Transmission products can also be firm on the TSO. This takes the form of a liability
to either re-purchase rights at market price (perhaps offset by re-selling obligations)
or to fund re-dispatching operations to ‘buy out’ commitments which the TSO is
unable to honour.

28. Firmness to TSOs provides an incentive on them to accurately predict and manage
transmission capacity. However, it may also expose TSOs to considerable risks,
especially on large interconnected systems where energy schedules and parallel
flows are beyond the control, or even prior knowledge, of individual TSOs. Care must
therefore be taken to ensure that firmness on TSOs does not incentivise uneconomic
risk-averse behaviour.

April 2001 Page 9/20

Key Concepts and Definitions for Transmission Access Products

Duration

29. Each transmission access product must be associated with a specific time interval.
Near to real time, accurate control of congestion will only result if products can be
traded for time intervals similar to those used for short-term energy trading (typically
one hour or less). Initial allocation and subsequent forward trading can, however, be
conducted with products bundled into longer time periods.

30. Users in different market sectors may wish to deal in products of different lengths.
Encumbent baseload generators and large industrial customers, for example, are
likely to be interested in purchasing longer ‘strips’ of access rights than short-term
traders or generators with expensive or intermittent fuel sources.

31. For the above reasons, access markets are likely to have to accommodate products
of varying duration. New capital-intensive interconnectors are likely to require
products covering a number of years. More generally, long-term products might be
allocated in annual auctions, medium-term products in monthly auctions, and short-
term ones day-ahead. The longer-term products could be broken down by users and
traded for shorter periods, where possible down to individual market intervals of one
hour or less.

April 2001 Page 10/20

Key Concepts and Definitions for Transmission Access Products

FLORENCE GUIDELINES

32. The guidelines address various factors which relate to the choice of transmission
products for congestion management. They are as follows:

Economic efficiency

33. Guideline 2 calls for both short-term economic efficiency and efficient investment
signals.

34. The requirement for economic efficiency implies that congestion management
methods should be capable of revealing short-run marginal costs. In principle, a
transmission access market can achieve this, but conditions for success are likely to
include:

- combined rights and obligations

- a physical path model
- as far as possible, both high locational resolution and high trading liquidity
- an accurate constraint representation.

Directional price signals

35. Guideline 6 states that the price signals emerging from the congestion management
process should be ‘directional’. Guideline 31 explains that this is required for correct
treatment of counterflows on congested circuits.

36. Directionality will result as a necessary consequence of economic efficiency. It

therefore imposes no new requirements beyond those already listed above in
paragraph 34.

Netting

37. Guideline 7 states that requests to use transmission circuits in opposite directions
should be netted, and specifically that transactions relieving congestion should never
be denied. Guideline 29 recognises that safe operation of the power system should
not be compromised, and invites TSOs to propose a workable scheme.

38. A workable scheme in which flows can be safely netted requires a degree of certainty
that notified flows will materialise in practice. In terms of the product characteristics
described above, this requirement implies:

- combined rights and obligations

- a physical path model
- bi-directional firmness on users (eg both underrun and overrun imbalance
settlement).

Unused capacity

39. Guideline 8 specifies that any unused capacity must become available to other
agents, and refers in particular to the ‘use-it-or-lose-it principle’.

April 2001 Page 11/20

Key Concepts and Definitions for Transmission Access Products

40. In general, some form of underrun imbalance settlement is required to incentivise

market participants to use or trade their transmission product holdings. This provides
further confirmation that such products should be:

- combined rights and obligations

- firm on users.

In financial terms, the ‘use-it-or-lose-it principle’ is simply a special case in which the
underrun price is zero.

Firm capacity

41. Guidelines 9 and 10 indicate that products should be firm on TSOs, in the sense that
they should be responsible for re-dispatching if outturn transmission capacity is less
than that anticipated. Guidelines 10 and 26 do, however, suggest that varying
degrees of firmness may be appropriate, presumably in the form of interruptible
rights.

42. In terms of the characteristics discussed above, products should therefore be:

- normally firm on TSOs (with the possibility of some products with decreased
firmness).

Transaction independence

43. Guideline 20 express a preference for non-transaction based methods. It also

suggests that market splitting would satisfy this preference in principle, but is too
difficult to implement in the short term.

44. It is understood that this preference derives from two concerns. The first is that
transaction-based transmission access would be strongly linked to specific energy
trades, potentially destroying liquidity in both the access and energy markets. The
second is that market participants may be reluctant to entrust the identity of contract
counterparties with other parties, including TSOs.

45. All location-to-location products are transaction-based, at least in the sense that
energy volumes at both ends of the transaction must be related to the same access
product in the access settlement process (assuming commercial settlement is
required, as suggested above). Taken at face value, therefore, Guideline 20 implies
a preference for:

- an entry/exit model

46. The precise nature of this preference should, however, be examined further.
Transaction-based settlement does not imply that counterparty locations need to be
revealed to TSOs for operational purposes (except perhaps for checking and
validation of exchange programmes between control areas in the UCTE network).
Furthermore, zone-to-zone products are robust to trading between parties within
zones. For a scheme using independent settlement agents and large zones (eg
control areas or countries), an acceptable variant could be:

- a zone-to-zone model.

April 2001 Page 12/20

Key Concepts and Definitions for Transmission Access Products

Timing

47. Guideline 26 refers to composite auctions with products of varying duration.

Guideline 27 proposes a series of auctions, possibly taking place yearly, monthly,
weekly, daily and intra-daily.

48. The Guidelines therefore confirm that transmission products should be characterised
in terms of:

- a range of durations, varying from at least a year to less than one day.

Risk assignment

49. Guideline 30 states that the financial consequences of deviations from notified flows
should be borne by the responsible parties, and suggests that re-dispatching costs
could be funded by penalties on such deviations.

50. The implications of this Guideline for product definition are similar to those of the
‘economic efficiency’, ‘netting’ and ‘unused capacity’ guidelines described above.
They are:

- combined rights and obligations

- a physical path model
- as far as possible, both high locational resolution and high trading liquidity
- an accurate constraint representation
- firmness on users via underrun and overrun settlement.

Tradability

51. Guideline 33 addresses the creation of liquid energy markets, and states that auction
products should be freely tradable before notification.

52. If secondary access markets are not sufficiently liquid, the energy markets could be
impeded by the difficulty of matching energy contracts with suitable access products.
High liquidity can be obtained either by trading through TSOs acting as market-
makers, or by requiring market participants to deal directly in flowgate products.
Direct trading of entry/exit products is only possible between market participants in
the same location, and is therefore only likely to provide sufficient liquidity if locations
are defined as relatively large zones. Location-to-location products are inherently
less liquid that entry/exit ones, because they can only be traded between transactions
with the same entry and exit locations. The requirement for liquid trading therefore
favours:

- an entry/exit model
- low location resolution (ie large zones)
- trading via TSOs
or
- individual flowgate products (see paragraph 24 above).

April 2001 Page 13/20

Key Concepts and Definitions for Transmission Access Products

CURRENT AUCTION SCHEMES

Characteristics of current schemes

53. Current auction schemes within the IEM are:

• Denmark-Germany (DK-D)
• Netherlands-Germany/Belgium (NL-D/B)
• France-England (F-GB)
• France-Spain (F-ES)

The characteristics of the products sold in these auctions are shown in Table 1.

Limitations of current methods

54. Experience to date has demonstrated that cross-border capacity can be successfully
auctioned. However, an important question is whether acceptable IEM-wide
congestion management would result if current methods were simply extended to all
congested interfaces, with no particular co-ordination or harmonisation arrangements.
The following paragraphs describe the possible shortcomings of such a strategy.

55. Table 1 shows that all of the existing auction schemes are based on contract paths.
This is of little consequence for the peninsular borders (DK-D, F-GB and F-ES).
These constitute system partitions (see Fig 5c) so parallel flows are minimal or non-
existent, and physical paths coincide with contract paths.

56. The NL-D/B scheme is the first one to introduce the possibility of interactions between
multiple paths. While these interactions are expected to be manageable in this case,
extension to more central UCTE regions is likely to be more difficult. In general,
parallel flows may cause significant interactions between nominal contract paths.
There is also scope for market participants to select contract paths which they expect
to be commercially beneficial, rather than those which most closely match physical
paths. In these circumstances, the result could be poor congestion management with
potential security problems and inaccurate price messages, conflicting in particular
with Florence Guideline 2, relating to economic efficiency.

57. Although ‘use-it-or-lose-it’ provisions are incorporated in the current auctions, trading
and settlement of obligations is not well developed. As a result, the costs of capacity
under-utilisation may not be accurately assigned to market participants. Furthermore,
TSOs may lack the necessary confidence to take counterflows into account when
making capacity available. The guidelines on netting and risk assignment may
therefore not be adequately addressed.

58. All of the current auctions effectively allocate zone-to-zone products, since no
distinction is made between locations within the countries concerned. They are
therefore only able to address congestion associated with the borders themselves,
rather than with intra-zonal constraints. Zone-to-zone auctions are also transaction-
based, contrary to the preference expressed in Guideline 20 (but see paragraph 46).
Although liquidity is increased by the application to large zones, it is limited by the
need to co-locate both entry and exit parties in any trade. Guideline 33 is therefore
compromised to some extent.

April 2001 Page 14/20

Key Concepts and Definitions for Transmission Access Products

CO-ORDINATED AUCTIONS

59. ETSO has produced a separate paper entitled ‘Co-ordinated Auctioning of

Transmission Capacity in Meshed Networks’. This describes a method of allocating
transmission products in the presence of multiple constraints or congested borders,
using loadflow-derived distribution factors to take interactions into account.

60. The characteristics of this method are shown in Table 2. The loadflow factors
provide a physical path model. The constraints need not be physically located on tie-
lines, although in practice they could well be represented as tie-line flowgates (see
paragraph 21). The method places no restrictions on whether products are rights
and/or obligations, on the choice of location model or resolution, or on product
firmness or duration.

61. The proposed approach overcomes some of the shortcomings of currently used
methods. In particular, it takes parallel flows into account and eliminates exploitation
of the contract path approximation. In consequence, it should be possible to control
congestion and reveal economically efficient prices in the presence of interacting
constraints.

62. Implementation of the proposed method would require a high level of co-ordination
between TSOs. Wide-area (perhaps even pan-European) loadflow calculations
would be necessary to produce the distribution factors. Auction clearing and
subsequent trading would also need to be jointly organised, at least on a regional
basis.

April 2001 Page 15/20

Key Concepts and Definitions for Transmission Access Products

HARMONISATION REQUIREMENTS

63. The above analysis of existing auction methods suggests that satisfactory pan-
European congestion management is unlikely to result from unco-ordinated
development by individual Member States. As described above, ETSO is developing
co-ordination methods, but these require a high level of co-operation and
standardisation across borders. The following paragraphs identify the areas in which
harmonisation may be required.

64. The Florence Guidelines themselves dictate a degree of harmonisation of access

products. The characteristics which result from compliance with the Guidelines are
listed (in italics) in paragraphs 34 to 52 above.

65. In addition to harmonising the nature of the transmission products themselves, the
allocation, trading and settlement processes will also require harmonisation to enable
effective inter-working, notably between Regulators and between TSOs. Although
initially this will only be necessary on a local or regional basis, difficulties will
subsequently appear if pan-European compatibility is not considered at the outset.

66. It is of fundamental importance that a commonly agreed location model is adopted.

This is needed to relate entry and exit locations to power exchanges and other
trading hubs, and to border and/or constraint (flowgate) locations. This model must
form an unambiguous framework for posting transmission capacity forecasts,
calculating loadflow distribution factors and allocating transmission products.

67. It may be necessary to co-ordinate the timing of auctions throughout Europe, in order
to avoid discriminating between users in different locations. The efficient allocation of
products might also be facilitated if the duration of products allocated at different lead
times were harmonised.

68. In general, forecasting transmission capacities and loadflow distribution factors,

together with subsequent re-dispatching to correct forecasting errors, are all joint
TSO activities. If transmission products are to be firm on TSOs, there needs to be a
harmonised way of assigning liability for re-dispatching costs between them.

April 2001 Page 16/20

Key Concepts and Definitions for Transmission Access Products

CONCLUSIONS

69. The current strategy of auctioning access across individual borders is an efficient
means for initial implementation of market-based capacity allocation schemes.
Ultimately, however, it is unlikely to result in effective congestion management
throughout the IEM.

70. Current auction schemes are not compliant with all of the Florence Guidelines on
congestion management.

71. To ensure general effectiveness and compliance with the Guidelines, it will be
necessary to complement the current strategy by establishing:

• a harmonised definition of access rights and obligations based on physical flows

• an agreed locational model relating access points or zones to trading and

congestion locations

• effective methods of incentivising compliance with rights and obligations

• liquid trading of access rights and obligations

• methods for co-ordinating auctions, trading and settlement where congested flows
are interactive.

72. The above requirements suggest that transmission access products should be based
on:

• combined rights and obligations

• an entry/exit location model, incorporating trading hubs

• physical flow paths

• zones selected to optimise the trade-off between congestion management

effectiveness and trading liquidity

• constraint representation at the flowgate level

• firmness on both TSOs and transmission users

• a range of timescales over which transmission products can be acquired and

traded.

April 2001 Page 17/20

Key Concepts and Definitions for Transmission Access Products

RECOMMENDATIONS

73. The EC, Regulators and transmission users should be invited to agree with TSOs that
the above conclusions are valid.

74. Subject to further consideration of the feasibility of market splitting, the present
strategy of auctioning cross-border access should be encouraged, since it enables a
quick response to the challenge of allocating capacity on a market-oriented basis.
However, ETSO should complement this by providing:

• guidance on the form of transmission products which should be auctioned and

traded

• the co-ordination necessary to build an effective ‘map’ of access zones and

trading locations

• methods for co-ordinating access auctions, trading and settlement where it is

necessary to take interactions into account.

April 2001 Page 18/20

Key Concepts and Definitions for Transmission Access Products

Location Constraint
Right Obligations1 Location Path resolution representatio Firmness Duration
s model model n
TSO user

DK-D yes no zone-to-zone contract zonal boundary yes yes 1 yr, 1mth, 1 day
(UIOLI) (hourly)

NL-D/B yes no zone-to-zone contract zonal boundary yes2 yes 1yr, 1mth, 1day, (hourly)
(UIOLI)

F-GB yes no zone-to-zone contract zonal boundary no yes 3yrs, 1yr, 1 day
(DC cable) (UIOLI)

F-ES yes no zone-to-zone contract zonal boundary yes yes 1yr, 1mth, 1wk,
(planned) (UIOLI) 1 day (hourly)

1. note that ‘Use-it-or-lose-it’ (UIOLI) is not included here as an ‘obligation’

2. except for force majeure

Table 1: Current Auction Schemes

C:\Locales\cortinasdam\ETSO\NACM\Official papers\Final Papers\ETSO-NET\Key Concepts for Transmission Access Products.doc

- 19 -
Key Concepts and Definitions for Transmission Access Products

Rights Obligations Location model Path Location Constraint Duration

model resolutio representation Firmness
n
TSO user

location-to-location zonal any

yes yes1 or physical or (probably yes2 yes1 any
entry/exit nodal flowgate)

1. dependent on settlement arrangements

2. dependent on TSO regulation/incentivisation arrangements

Table 2: Co-ordinated Auctions

C:\Locales\cortinasdam\ETSO\NACM\Official papers\Final Papers\ETSO-NET\Key Concepts for Transmission Access Products.doc

- 20 -
 Appendix V

ETSO - Co-ordinated Auctioning - A market based method for

transmission capacity allocation in meshed networks

62
 Co-ordinated Auctioning
A market-based method for transmission
capacity allocation in meshed networks

Final Report
April 2001
Executive summary

The purpose of this paper is to present and propose a possibility to extend, when needed to
manage congestions, the «classical» capacity auction sale from simple (i.e. one-border
capacity sale) to more complex systems. This economically efficient method has been
implemented or is being implemented on several European borders. However, as all the
existing or considered methods up to now, it fails, when limited to a simple bilateral
implementation, to cope with the following drawbacks likely to happen when it is applied to
the different borders of a highly meshed system, involving multiple control areas, such as the
continental European grid :

§ for the market actors, the complexity of multiple auction mechanisms, together with
the risk of apparition of non-convergent interactions between separate and
successive clearing processes (if not synchronised);

§ the poor quality of the (multiple) economical signals sent to the market participants,
if the different auction systems are not combined in a way or another, whereas the
underlying «physical» power system is strongly interdependent.

The present paper provides a first vision for an allocation method that could cope with these
complex problems. The aim is first of all to show that some realistic solutions seem to exist
for capacity allocation in complex networks (i.e. most of the continental Europe borders) and
to initiate a discussion on this subject. However, a lot of questions, especially related to the
implementation issues, must still be studied in more details, and some time is necessary
before considering a practical implementation. Nevertheless, the use of some of the
described principles could be considered for capacity allocation in the year 2002 for regional
experiments.

The underlying idea of this paper is to make a proposal of a congestion management

method which provides an appropriate answer to the two following issues:

§ handling adequately the problem of allocating scarce transmission capacity in highly

meshed networks;

§ alleviating traders of the complexity of independent auctions in the main

bottlenecks.

The co-ordinated auctioning mechanism presented below is likely to ensure the feasibility of
complex bilateral cross-border trade. It does not introduce additive costs at every border and
as such avoids any inefficient pancaking effect.

It favours a clear distinction between trade and physical flows, and shows how the market
participants can be relieved from handling the complexity of the electricity physical laws,
while having transparent information on the model used to approach at best the complex
physical reality.

The main idea is that a simple representation of the meshed network effects (through load
flow factors, so-called «Power Transfer Distribution Factors (PTDF)») can take into account
the main physical interactions. Thus, a standard mathematical formulation makes it possible
to select the bids that represent the highest value for the market.

April 2001 2/22

The method presented here is an extension of bilateral explicit auction mechanisms ; in that
respect, it inherits the properties of explicit auctioning. The paper then introduces a
discussion about the compatibility between different implementation possibilities for this
extension of explicit auctioning and the different forms of markets : bilateral contracts, PXs,
etc. The conclusion is that there is no incompatibility with any chosen form, or the
combination of several: a high level of interoperability may then be achieved.

Among the properties of the proposed method, one is of great importance and should be
1
recalled here: except some very special cases of implementation : every bidder whose bid
contributes to saturate at least one bottleneck will be charged a fee., Otherwise the capacity
is free of congestion charge if the bids are not participating to an active constraint.

In reference to the Florence Guidelines, this method seems to be a valuable alternative to

market splitting.

To be implemented, the proposed method requires a high level of co-operation and co-
ordination among TSOs; of course it is more easily feasible on «regional» congestion
bottlenecks (with the involvement of the different control areas between which there are
electricity exchanges that induce the majority of the flows through the bottlenecks). The first
implementations should then be managed on such a «regional» basis, with further possible
extension if the situation asks for.

In that respect, the joined auction mechanism recently set up by four TSOs to manage the
import capacity of the Netherlands is a step forward towards improved coordination. This
area is thus a possible candidate for a first implementation of the co-ordinated auctionning
method, that takes into account network interactions.

Though requiring further work before implementation, the proposed system seems very
efficient to improve the European Power System interoperability.

1
Such as Pay-as-bid pricing, see chapter IV

April 2001 3/22

I. INTRODUCTION

Compared to other goods, the specificity of electric power is that the flow corresponding to a
given electricity exchange cannot be controlled like trucks, but is governed by immutable
physical laws. In a highly meshed network like western continental Europe’s, a single
electricity exchange between France and Italy will partly flow through France-Italy
interconnections, but also through France-Switzerland-Italy, and even France-Germany-
Austria-Slovenia-Italy, and so on.

That is why contract path mechanisms (that only allocate transmission capacity along a
single theoretical path between the generator and the consumer) do not entirely fit the needs
of the evolving European Electricity Market. Contract path mechanisms are appropriate in
longitudinal (e.g. U.S. West Coast) or two-party peninsular systems (e.g. France-Spain).
Such mechanisms are inherently bound to fail in meshed networks as soon as cross-border
power trade introduces significant swings in power flows, because they cannot account for
the physical reality of electric power transmission. This also means that transmission
capacity reservation cannot be undertaken by a single country, but has to follow a
mechanism that encompasses all the interconnected countries. In a meshed network,
transmission capacity cannot be partitioned : it is a common resource that has to be
operated jointly and in a co-ordinated way.

This paper presents the proposed principles of a mechanism that effectively implements
transmission capacity auctioning in a meshed network, supporting capacity reservation on
any given time horizon (the choice of using this mechanism for different time horizons, from
annual capacity to day-ahead or even hourly capacity is not discussed here). It requires an
efficient co-ordination between TSOs of involved countries and allows diversity of market
mechanisms as well as coexistence with bilateral contracts.

April 2001 4/22

II. PRINCIPLE

The basic idea of the proposed mechanism is to establish a clear separation between
physical flows and trade.

Energy traders are mainly interested in source and destination zones, and not in the
technical contingencies of the resulting power flows nor in the topology of the network.

It is the vocation of the TSOs to manage the complexity of the electric physical laws, while
providing the markets with sound indicators of risk of congestion, using alleviating
mechanisms both non-discriminatory and economically effective. This idea or assumption is
consistent with the orientations given by the Cross Border Tariff principles, since for the
market actors only energy input and output zones and associated hourly energy schedules
are necessary.

Consider a network divided into zones, separated by borders or interconnections, consisting

in the aggregation of multiple physical transmission lines. Instead of having an auction
market for each border, the idea is to organise co-ordinated auctions in order to allocate in
the same process the considered "interconnected" capacities in zones with strong
interactions (strongly meshed networks).

For instance, the commodity put to auction would consist in a firm transmission right-and-
obligation (firm both on TSOs and on Users2s) to transfer power from one zone to another,
each zone being a whole country or a part of it, with respect to control areas. Alternatively,
commodity put to sale could be presented as a right-and-obligation to withdraw or submit
power from one zone to a global hub chosen in one of the zones. This entry-exit alternative
can be handled in the same way than the above zone-to-zone transfer rights, providing that
netting rules are agreed (see chapter IV). We will use this entry-exit presentation in the
following rather than the zone-to-zone one, in order to establish the compatibility with the
« non transaction based » requirement expressed in Florence recommendations.

The role of the auctioning mechanism is to allocate scarce transmission capacity according
to bids by market participants which reflect the values of their individual transactions, taking
into account their contribution to the underlying network congestion. These contributions are
encapsulated in a published PTDF matrix, which simply and effectively converts transmission
rights into a contribution to interconnection flows. This matrix is built up with PTDF factors
that indicate how much a given interconnection will be loaded by the use of a transmission
right, e.g. a transaction between a given pair of zones associated with a zone-to-zone right,
or an injection/offtake of power in a certain zone associated with an entry/exit right.

The general spirit of this mechanism is to allocate the transfer capacities to get the most of
their values for the market.

As a consequence, the allocation criterion is a generalisation of the criterion used for a

«classical» auction procedure : the goal is to optimise trade under the constraints of limited
capacities, given that the different bids are ranked according to the bidding price. It can also
be interpreted as the minimisation of the burden of the infrastructure for market participants.
This burden is equal to zero if all bids are secured, which occurs when no congestion
appears. In such case, the use of the capacities is free of congestion charge for all market
participants.
2
Coordinated auctions are also possible, in case the commodity put to auction is not firm on users, but
it makes the explanation of the mechanism slightly more complex (see chapter IV).

April 2001 5/22

III. STEP-BY-STEP MECHANISM

A. Publishing
TSOs compute the PTDF matrix, giving for each type of transmission rights (e.g.
between each pair of zones) the contribution of the transaction under consideration to
the interconnection flows. The PTDF factors specify the capacity used on each
interconnection as a fraction of the volume of transmission rights associated with
particular entry and exit zones. These computations are based on one transmission
system configuration that has been commonly agreed upon for NTC computation.

For each border between neighbouring zones, linked by existing physical

interconnections, the neighbouring TSOs compute together the technical constraints
that limit the capacity. This limit, called Bottleneck Capacity (BC) in the rest of the
document, can be related to the present NTCs calculations (see appendix 2), although
it is defined as a physical flow through the tie lines. It must be stressed that these BC
limits will reflect all network security rules applied to trans-European exchanges, and
thus will hardly be comparable to the sum of the individual capacity of each tie line.

Finally, these limits may be complemented by additional transfer limits in order to avoid
situations which would turn to be over-sensitive to the risks of changes happening after
the clearing.

To ensure transparency, once aggregated between TSOs, the PTDF matrix is

published with all other relevant information, in order to provide market participants
with any useful data to anticipate their position.

B. Bidding
TSOs organise simultaneous («co-ordinated») auctions on each pair of zones. Market
participants send their bids consisting of a quantity and a price for the cross border
input (or off-take) rights they want to buy. Combined input (in one zone) and off-take
(in another zone) prices indicate the maximum value they are willing to pay for
obtaining capacity on the given zone pair, up to the requested quantity. Market
participants aiming at bilateral trade may require, if they wish, to limit their allocated
input quantity at the same level that the allocated off-take quantity, in order to get a
balanced right from zone to zone.

C. Clearing
The selected set of bids is the one that combines the best overall economic value for
the market under physical and security constraints. For example, the objective function
to be optimised may consist in the sum of the bid price multiplied by the bid quantity.
The constraints taken into account imply that the security limits are not exceeded. Bids
with identical prices for the same zone to zone relation are offered the same quantity
(pro-rata). The result is obtained by solving a standard mathematical problem called a
"Linear Program". The foreseeable size of this problem rises no special difficulty in the
application of well-known algorithms and «from-the-shelf» software. The result is then
easily auditable by Regulating Authorities.

April 2001 6/22

IV. IMPLEMENTATION ISSUES

A. Variants in the auction mechanism

As any other auction mechanism, options must be chosen between several

implementation variants, such as :

• One or several auction rounds? Several rounds may increase the

transparency for market actors, given that global results of each round are
displayed to bidders (such as : marginal price, average price of bids secured
in the round …), yet increase also the complexity ;

• Pay-as-bid or marginal pricing? This question is not totally independent from

the previous one, since it may be proved that under the assumption of perfect
information of all players the final results would be the same. It has to be
mentioned that the usual implementation of pay-as-bid pricing would breach
out the « no congestion - no payment » principle, and should then be avoided.

Anyway these variants do not prevent from implementing the co-ordination method
proposed, since they do not change either the criterion or the constraints of the
optimisation problem.

B. Firmness on TSOs :

The capacities calculated by the TSOs, then published and sold through an auction
system are subject to numerous uncertainties among which :

• "force majeure" situations, such as severe destruction of equipment with

durable consequences,

• network component outages,

• variations in the pattern of flows induced by swings in the markets

competitiveness / attractiveness, and subsequent unexpected modifications of
the parallel flows (unforeseen parallel flows).

Co-ordinated auctioning reduces significantly this last type of uncertainty, as the

swings between zones are internalised in the auction mechanism. Note that the
swings inside each zone may still induce uncertainties on cross border capacities.
The question of the firmness of commitments made by TSOs, the corresponding
financial risk (varying according to the time horizon) and the associated question of
cost recovery mechanisms remains open, knowing that several solutions are
possible :

• cost socialising, or in other words the authorisation given to TSOs to recover

these costs via the transmission tariffs,

• no guarantee, which means that the market participants will have to hedge
this risk by themselves,

April 2001 7/22

 • allocation to TSOs of a certain amount of the revenues of the capacity
auctions to cover that risk : it is then possible to incorporate to the proposed
co-ordinated auction mechanism prices associated to BC limits (for instance,
cost of forecast cross-border co-ordinated redispatching). A co-ordination
between TSOs is then necessary to organise their common liability

The principles proposed in this paper are compatible with any of the variants taken
(e.g. «use it or lose it» principle for yearly to weekly capacities, firm day-ahead
capacity). There is nevertheless a need of harmonisation of regulation rules in the
participating zones.

C. Firmness on users and counter-flows netting off :

Firmness on users generally refers to rights and-obligations products, and to

underrun and overrun costs (when the market participants deviate from the
expected program they pay a cost, related to the adjustment market) . This concept
is already widely applied at day ahead horizon where scheduling is systematically
required from users.

If the commodities put to auction are rights-and-obligations, counter-flows created

by commercial exchanges can be fully net off and the economic efficiency is at its
maximum : flows against the directions of active constraints are regarded as
creating additional capacity. Whilst the rights associated with flows in the
constrained directions will achieve positive prices, the rights in the opposite
directions, having effectively created additional capacity in the constrained direction,
would be assigned negative prices by the auction mechanism. Together with the
rewards of negative payment, holders of these rights should have obligations to
make such flows available in real time.

In this right-and-obligation case, co-ordinated auctioning will be cleared up in one

settlement, as presented in appendix 1, with a full netting off (100%). Then , the
counter-flows will see a negative price, even if their initial bid price is positive.

If rights-without-obligations are put to auction, which means the possibility for a

bidder not to use the capacity and as a consequence an additional uncertainty for
TSOs during the nomination phase, the co-ordinated mechanism must be adapted
to take in account the risk associated : for instance the margins can be enhanced
by stiffening the physical constraints, i.e. lowering the BCs values, or by adding a
right-only premium to the rights-and-obligations price initially assessed.

It should be also mentioned that options could be developed outside of a right-and-

obligation only co-ordinated auctioning mechanism, by power marketers taking over
the firmness risk.

April 2001 8/22

 D. Zone-to-zone versus Zone-to-hub

In the zone-to-zone model, the product sold is the right to implement a bilateral
transfer between the zone of production and the zone of delivery without further
liability for congestion charges. In the zone-to-hub model, the two parties of the
transaction buy independently the rights needed, both referring to a normative
transfer between the zone and a global hub (see the following figure).

~ B ~ B
A
A
hub

Zone to zone Zone to Hub

Zone to zone commodities are more stable for TSOs but less flexible for users.
Zone-to-hub commodities gets the opposite qualities.

For instance in a zone to hub model a user who bought an input right in zone A, can
use it for any cross-border trade referring to a generation in A : its counterparty can
change, say between auction clearing and run time, providing that its counterparty
has also bought on its own the cross-border rights in the off-take zone. This type of
commodity makes things easier for trade, specially on energy markets where
counterparties are not clearly specified.

On the other hand in a zone-to-zone model, a user who has bought an A to B

transfer right, would have more difficulty if its expected counterparty in B changes :
unless he can find another counterparty in zone B, he must sell back its A-B right
and buy another one, or buy the complementation right from B to the zone of its
new counterparty.

It is feasible to implement both models in coordinated auctioning. They will lead to

the same allocations in terms of rights-and-obligations. When dealing with rights-
without-obligations, some differences may appear due to the fact that input /off-take
leads to higher risks for TSOs than bilateral transfer rights.

E. Time horizon and reselling of transmission rights

As proposed in earlier papers, multiple auctions can be organised for different time
horizons (yearly, monthly, weekly and daily), thus allowing long term capacity
reservation needed for long term bilateral contracts, while retaining the flexibility
required for operational planning and last minute adjustments.

The method presented here can be implemented for these different time frames.

As far as transmission rights reselling are concerned, it is also just a choice to be

made at the stage of implementation (that is either within one time horizon frame, or
from – for instance – a yearly to monthly auction).

April 2001 9/22

 F. Integration of Co-ordinated Auctions with Bilateral Contracts and
Energy markets

All the developed principles can be used for bilateral trade, but also for trade related
to any energy markets, under some synchronisation of the processes

Cross border bilateral contracts, which associate generation on one side and supply
on the other, fit directly with the above mechanism.

In the same time energy markets, which deal with unilateral bids, may develop also
cross border exchanges as a result of their clearing process. Therefore they turn a
part of the bids they deal with into inter-zonal requests. These inter-zonal requests
could also be taken into account by the above mechanism which will ensure that
bilateral contracts and organised energy markets are treated equally regarding
congestion management.

Periodic auctions with yearly, monthly, weekly and daily terms can be used to
reserve capacity, and thus hedge bilateral contracts as well as short term energy
markets from the delivery congestion risk.

G. Regional implementation on a part of European network

To be implemented, the proposed method requires a high level of co-operation and

co-ordination among TSOs; and of course this co-operation is more easily feasible
on a «regional» basis (say few neighbouring control areas between which there are
electricity exchanges that induce the majority of the flows through the bottlenecks).
This implementation is feasible providing that the missing interconnected area is
represented by some type of equivalent.

The knowledge of the overall PTDF matrix factors, would allow an analysis of the
coupling between areas and thus would help the choice of the limit for the best
regional implementations.

April 2001 10/22

V. CONCLUSION

The co-ordinated auctioning mechanism presented here is likely to ensure the feasibility of
complex bilateral cross-border trade. It does not introduce additive costs at every border and
as such avoids any inefficient pancaking effect.

It favours a clear distinction between trade and physical flows, and shows how the market
participants can be relieved from handling the complexity of the electricity physical laws,
while having transparent information on the model used to approach at best the complex
physical reality.

The main idea is that a simple representation of the meshed network effects (through so-
called «PTDF factors») can take into account the main physical interactions. Thus, a
standard mathematical formulation makes it possible to select the bids that represent the
highest value for the market.

The method presented here is an extension of bilateral explicit auction mechanisms; in that
respect, it inherits the properties of explicit auctioning. The paper then introduces a
discussion about the compatibility between different implementation possibilities for this
extension of explicit auctioning and the different forms of markets : bilateral contracts, PXs,
etc. The conclusion is that there is no incompatibility with any chosen form, or the
combination of several: a high level of interoperability may then be achieved.

Among the properties of the proposed method, one is of special importance and should be
recalled here: except some special cases of implementation, bidders will pay only if their bid
is participating to an active constraint, which means if the bid contributes to saturate at least
one bottleneck. Otherwise the capacity is free of congestion charge.

In reference to the November Florence Forum Guidelines on Congestion Management, this

method seems to be a valuable alternative to market splitting:

[Guideline 2]: “The congestion management method(s) implemented should deal with
short-run congestion in an economically efficient manner whilst simultaneously
providing signals or incentives for efficient network and generation investment in the
right locations”. The method provides a competition among bidders based on sound
mathematical models, which guarantees the efficiency of the solution through proven
optimality.

[Guideline 3]: “In order to minimise the negative impact of congestion on trade, the
current network should be used at the maximum capacity that complies with the
safety standards of secure network operation”. As graphically shown in appendix 2,
one of the key point is that security constraints are explicitly taken into account : this
ensures that the maximum capacity within these constraints is proposed to the
bidders.

[Guideline 4]: “The TSOs should provide non-discriminatory and transparent

standards, which describe which congestion management methods they will apply
under which circumstances… ”. The TSOs would publish all critical data (security
limits) and parameters (the load flow factors)

April 2001 11/22

 [Guideline 5]: “Discrimination between the different types of cross-border
transactions, whether they are physical bilateral contracts or bids into foreign
organised markets, should be kept to a minimum when designing the rules of specific
methods for congestion management …”. The method can cope with bilateral and
other kinds of trade.

[Guidelines 6 & 7]: “Price signals that result from congestion management methods
should be directional”, “Every effort should be made to net the capacity requirements
of any power flows in opposite direction over the congested tie line…”. As shown in
appendix 1, the method exhibits marginal congestion prices and assumes netting.

[Guideline 8]: “Any unused capacity must become available to other agents (the use
or lose it principle)…”. The co-ordinated auctioning method it is compatible with the
use-it-or-lose-it principle.

To be implemented, the proposed method requires a high level of co-operation and co-
ordination among TSOs; of course this co-operation is more easily feasible on «regional»
basis (say few neighbouring control areas between which there are electricity exchanges
that induce the majority of the flows through the bottlenecks). The first implementations
should then be managed on such a «regional» basis, with further possible extension if the
situation asks for.

In that respect, the joined auction mechanism recently set up by four TSOs to manage the
import capacity of the Netherlands is a step forward towards improved co-ordination. This
area is thus a possible candidate for a first implementation of the co-ordinated auctioning
method, that takes into account network interactions.

Though requiring further work before implementation, the proposed system seems very
promising to improve the European Power System interoperability.

April 2001 12/22

APPENDIX 1 : Four node theoretical example

How the Coordinated Auction mechanism practically works

This example presents an application of coordinated auctions to a small system of four

nodes and four interconnecting lines. Each line having the same impedance, the PTDF
matrix and Bottleneck Capacities limits are shown on the following table, where C is taken as
hub, and represented in figure 1 diagram :

Interconnections max =
A 100 MW
Inputs A-B A-C B-C C-D max =
in A + 0.33 + 0.67 + 0.33 0 150 MW
max =
in B - 0.33 + 0.33 + 0.67 0
50 MW
in D 0 0 0 -1 C D
max =
« BC limits » 50 100 150 150 150 MW
B

Figure 1

In the example, full netting of opposite flows is adopted hence the PTDFs are only published
for one direction of transaction – those for the opposite direction automatically assume equal
and opposite values. The market participants submit their bids reflecting the value of the
transmission rights in relation to their transactions. Then, the clearing mechanism aimed at
maximising the value of allocated transmission capacity is applied. Finally the results are
given, showing the bids accepted and those rejected, as well as the interconnectors that
become constrained.

First set of bids :

Let us consider a first set of bids. The bids are supposed to be bilateral ones asking for a
transfer, that is to say a combined input and off-take :

Market Transfer Bid Bid Allocated

Partici requestd quantity price quantity A 100
pant (MW) (E/MW) (MW)
M1 A-C 50 3 50
0
M2 A-C 50 2.5 50 0
M3 B-C 50 2 50 C D
M4 B-C 50 1.5 50
B 100
M5 A-D 0

Figure 2

April 2001 13/22

In the coordinated clearing, all the bids are agreed, as their combination does not lead to
overpass the « BC limits » , see figure 1. The above diagram on the right side shows the
resulting flows after clearing. The A-C line reaches its limit.

Second set of bids :

Now let us add a fifth market participant M5 , which requests an A-D transfer at a high bid
price :

Market
Partici
Transfer Bid Bid
requestd quantity price
Cost to induce Allocated Marg.
+ or - 1 MW quantity Price
A 100
pant (MW) (E/MW) flow on A-C (MW) (E/MW)
M1 A-C 50 3 4.50 50 2,50
10
M2 A-C 50 2,5 3.75 40 2,50 0
M3 B-C 50 2 6.00 50 1,25 C D
M4 B-C 50 1,5 4.50 50 1,25
M5 A-D 10 3 4.50 10 2,50 B 100

Figure 3

After clearing, the only constrained line is AC. The requested quantity of the five bids cannot
be allocated. Due to the PTDF factors, the bids which proposes the highest values on A-C
flow are M3 (2 / 0.33 = 6 Euros/MW), then M1, M4 and M5 (4.5 Euros/MW) and finally M2
(3.75 Euros/MW). Therefore this last M2 bid is the first one to be reduced and the 100 MW
limit on A-C results in a limitation at 40 MW of M2 allocated quantity. Another way of
understanding this result is to start from the figure 2 and to consider that the 10 MWs now
requested by M5 wins against the same number of those requested by M2, as both have the
same influence on A-C flow and as M5 bids at a higher price.
M2 bid is the marginal bid and so it sets the marginal price on A-C at 2.5 Euros/MW. M2 bid
price applies also to M1 and M5 which have the same influence (= PTDF) on A-C flow.
M3 and M4 bids has and influence on A-C flow which is half of M2 ones, so the price they
are charged is 1.25 Euros/MW.

April 2001 14/22

Third set of bids :
Let us now consider a slight move in the bid price of M4, who is lowered down to 1
Euro/MW. The clearing changes because the value of M4 bid on A-C flow (which stays the
unique constrained line) is now 3 Euros/MW ( = 1 / 0.33) while the price proposed by M2 for
the same MW of A-C flow stays at 3.75 Euros/MW (= 2.5 / 0.67).

Market Transfer Bid Bid Cost to induce Allocated Marg.

Partici requestd quantity price
pant
+ or - 1 MW
(MW) (E/MW) flow on A-C
quantity
(MW)
Price
(E/MW)
A 100
M1 A-C 50 3 4.50 50 2,0 10
M2 A-C 50 2,5 3.75 50 2,0 10
M3 B-C 50 2 6.00 50 1,0
C D
M4 B-C 50 1 3.00 30 1,0
M5 A-D 10 3 4.50 10 2,0
B 90

Figure 4

Therefore M4 is now the first bid to be reduced and the 100 MW limit on AC flows results in
a limitation at 30 MW of M4 allocation. Another way of understanding this result is to
consider that, compared with the second set of bids, M2 wins back on M4 the 10 MW
allocation it has just lost against M5. M4 allocated quantity is reduced by 20 MW as 20 MW
of M4 are necessary to balance 10 MW of M2 due to the ratio between their respective
PTDFs on A-C flow.
M4 bid is the marginal bid and so it sets the marginal price on A-C at 1.0 Euros/MW. M4 bid
price applies also to M3 which has the same influence (= PTDF) on A-C flow.
M1, M2 and M5 bids has an influence on A-C flow which is twice the one of M4, so the price
they are charged is 2 Euros/MW.

Marginal pricing
With such a marginal pricing method, the above sequence illustrates that the price charged
is always under or equal to the bid price (equal for the marginals bids). It shows also that the
price requested to each participant is directly related to the influence of their allocation on
the constrained lines flows.

No congestion , no payment
It would be easy to show that a participant whose bid does not influence any constrained line
would have nothing to pay. For instance, in the above third set, an M6 bid requesting a C-D
transfer of less than 140 MW would participate to no congestion and thus would have
nothing to pay.

April 2001 15/22

Counterflows are rewarded whatever price they bid
As full netting of opposite flow has been taken as hypothesis, a bid inducing a counterflow is
rewarded at the marginal price of the constrained flow it alleviates. In the above third set, so
does an M7 bid requesting a B-A transfer.

Market Transfer Bid Bid Cost to induce Allocated Marg.

Partici
pant
requestd quantity price + or - 1 MW
(MW) (E/MW) flow on A-C
quantity
(MW)
Price
(E/MW)
A 100
M1 A-C 50 3 4.50 50 2,0 10
M2 A-C 50 2.5 3.75 50 2,0 8
M3 B-C 50 2 6.00 50 1,0 C D
M4 B-C 50 1 3.00 48 1,0
M5 A-D 10 3 4.50 10 2,0
B 108
M7 B-A 18 0.5 non relevant 18 -1,0

Figure 5

Let us assume an M7 bid of 18 MW. Each 3 MW of this M7 bid creates a 1 MW alleviation of

the constraint on A-C, which allows consequently 3 MW more of M4 to be allocated. This
phenomena leads to the above clearing and load flow.
M7 will therefore receive 1 Euro / MW for its bid, whatever positive bid price it has made
under the strict condition that he buys a combined right-and-obligation commodity.
Note that over 20 MW requested by M7 on B-A transfer, the system would face a new
situation. As all other bids would be fulfilled, the A-C congestion would disappear, its flow
getting under 100 MW, then nobody would then be charged at all and consequently no
counterflow reward would be given to M7.

April 2001 16/22

APPENDIX 2 : Three zone example
Comparison between NTCs based allocation and Coordinated Auctions
Let’s consider three zones A, B and C.

B C

C being taken as hub, we can write the PTDF matrix and the Bottleneck Capacity limits (BC)
as follows :
Tie lines PTDF towards inputs
Inputs A-C C-A B-C C-B A-B B-A
in A +72% -72% +28% -28% +28% -28%
in B +36% -36% +64% -64% -36% +36%
BC 2000 600 1000 500 1500 1000
The NTC calculation from A to C can be shown on the following figure, where (for instance)
the base case consists in the combination of a large exchange from A to C with a small one
from A to B (A-B transfer is figured as an A-C transfer combined with a B-C transfer). NTC
(A-C) is reached by levelling up the A-C transfer to the value where a constraint is found (the
reliability margin is not shown on the figure). With the above hypothesis, one can see3 that it
is possible to draw the line representative of AC border limit (see figure), defined by the
relation : 0,72 * transfer(A-C) + 0,36 * transfer (B-C) = 2000
B-C transfer

2000/0.72 = 2778 A-C

border
limit A-C
transfer
NTC(A-C)
Base case

Figure 1 : NTC (A-C) and AC border limit line

3
If, starting from the point where we have just reached NTC(A-C), we substract 1 MW from the A-C transfer and
then add to the B-C transfer 1 / 0.36 * 0.72 = 2 MW, the PTDF matrix told us that the flow on the A-C tie lines
comes back to its constraint limit.

April 2001 17/22

Following a similar construction, one can draw the limits in the six possible transfers, as
shown on figure 2.

B-C transfer

B-C
B-A bor-
border limit
der limit

NTC(B-A)

NTC(B-C)

A-C
border
limit A-C
C-A
transfer
border
limit NTC(C-A) NTC(A-C)
Base case

NTC(A-B)

NTC(B-C)
A-B
C-B
border limit
border limit

Fig 2 : Polygon of constraints

Note that, depending on the base case chosen, the same constraint line may be reached in
two different NTCs assessment, while some other contraints lines are not. In above
example, NTC (A-B) and NTC (B-C) are both limited by the A-B border limit, while C-B
border limit does not play any role in this NTC assessment. With a different base case, the
effect of limiting constraint lines would change.
This simplified representation may also show why contract path methods, whose allocation
limits are based on fixed values of NTCs, are poorly designed for using the full range of
physical possibilities. On figure 3, the full combination of the three NTCs previously found is
represented by an hexagon. The green horizontally-lined part is within the constraint limits,
the red bold vertically-lined part is outside, which means such positions cannot be allocated
with firmness on TSOs unless using redispatch means to take it back into the unconstrained
area.

April 2001 18/22

 B-C transfer

B-C
B-A bor-
border limit
der limit

NTC(B-C)

A-C
border
limit A-C
C-A
transfer
border
limit
Base case

NTC(A-B)
NTC(A-C)

A-B
C-B
border limit
border limit

Fig 3 : Combination of three NTCs

The availability of these redispatch means is quite delicate to set up as it should involve the
three zones (cross-border coordinated redispatching) , and thus the firmness on TSOs is not
easy to guaranty.
The part of the unconstrained area which can be allocated by such separated combination of
fixed transfers such as NTCs, with firmness on TSOs and controlled risks, is more or less
the vertically-lined green area shown on figure 4.

B-C transfer

B-C
B-A bor-
border limit
der limit

A-C
border
limit A-C
C-A
transfer
border
limit

A-B
C-B
border limit
border limit

A-B
transfer

Fig 4 : Possible risk-free allocation with combination of fixed transfers and firmness on TSOs

April 2001 19/22

The maximum transfers allowed in fig.4 are significantly under the theoretical NTCs
assessed in figure 2. The unconstrained area which is leftover by this fixed-transfer-based
allocation appears in white inside the constraint polygon. It is due to the fact that
counterflows created by commercial transfers are not taken in account to alleviate physical
constraints in this mode of allocation.
Moreover, these figures shows the strong dependency of NTCs assessment on the choice of
the Base Case transfers. This choice being somewhat empirical, the values of NTCs
assessed upon are thus questionable in an open trade market.
Finally, the setting up of the maximum transfers proposed in figure 4 would need a lot of
coordination between the three TSOs, maybe as much as Coordinated auctioning, for a
poorest result.

As for Coordinated Auctions, they do not depend on NTC values although it has been shown
how the matrix factors and BC values can be related with current NTC assessment. Once
the above constraint polygon is built, Coordinated Auctions combines any set of bids in order
to stay within its limits. If the result is on a constraint line, that means that this congestion is
reached and that some payment will be required from the participants whose bids are
responsible for reaching the congestion. If the result is out of the constraint lines, that means
that no congestion is reached and thus no payment is required from the corresponding
participants.
In case of rights-and-obligations, it is even possible to allocate transfers which would have
led to constraint if taken alone, but which are balanced by counterflows due to the other bids
(see fig 5).
B-C transfer

B-C
B-A bor-
border limit
der limit

A-C
border
limit A-C
C-A
A-B allocation transfer
border
limit A-C allocation
C-B
allocation
Limit
reached

A-B
C-B
border limit
border limit

Fig 5 : Allocation of rights-and-obligations with coordinated auctions

In the case represented on figure 5, A-B and A-C allocated transfers plays a direct role in the
reaching of the limit and thus will pay for such an allocation, while C-B allocated transfer has
a counter-flow effect and thus will receive money out of the Coordinated auction mechanism,
providing that marginal pricing variant is applied (see Appendix 1).

April 2001 20/22

APPENDIX 3 : Mathematical Formulation
Let us define a mathematical formulation of the problem :
The network is represented as a set of N nodes, each of them standing for a control area. Let us call by
capital letter these nodes : A, B, C, ... . They can be associated in oriented doublets [X,Y].
The neighbouring nodes are linked by I interconnections, let us call these interconnections by Greek
letters : α, β, γ, ...
Each bid consists in an identification, an origin, a destination, a requested quantity and a bid price.
Let us call XY1, XY2, XY3, ... the bids whose origin is in area X, and whose destination is in area Y.
Their bid price and requested quantity are written π (X,Y,1), ... and Qreq (X,Y,1), ... Let us consider
that they are systematically classified such as π (X,Y,1) ≥ π (X,Y,2) ≥ ....≥ π (X,Y,i) ≥ ...≥04
The clearing process results in a quantity allocated to each bid. We will write this quantity Qall
(X,Y,i). These quantities can be summed up all over a doublet [X,Y] to get the quantity allocated to
the doublet Qall(X,Y) = Σi {Qall(X,Y,i)}.
The «participation» matrix described in above chapter III is called R. Each element of R, written
r(X,Y,ε), represents the proportion of an XY exchange which flows through the ε interconnection.
This coefficient is often called in literature « participation factor » or «Power Transfer Distribution
Factor» .The matrix R = [r(X,Y,ε)] has N*(N-1) lines (each doublet is considered in both direction)
and I columns.
For each interconnection ε, a Bottleneck Capacity BC(ε) is evaluated by the neighbouring TSOs, this
value being an image of the maximum flow through the interconnection compatible with security
rules.
The optimisation problem can be written in two strictly equivalent ways :
Maximise the objective function ΣXYi {π(X,Y,i)*Qall(X,Y,i)} or
Minimise the objective function2 ΣXYi {π(X,Y,i)*[Qreq(X,Y,i) - Qall(X,Y,i)]}
under the constraints : Qall(X,Y,i) ≤ Qreq(X,Y,i), whatever X,Y,i
Qall ≤ AC (AC is an optional transfer limit, called auctioned capacity)
RT*Qall ≤ BC,
Qall(X,Y,i) ≥ 03, whatever X,Y,i
Note that if we call RT(α)*Qall = f(α) the «flow» on interconnection α, other constraints can be added
without compromising the generality of the problem posed, such as :
f(α) + f(β) + f(γ) ≤ F if the flows has to be limited on the sum of the three considered
interconnections. 4
What could be the price paid by each participant after the clearing process ? The idea developed here
is threefold:
• to make a participant pay for a bid only if this bid is participating to an active capacity constraint ;

4
No negative bids are considered at this level of our analysis.
2
This form of the criterion represents the «burden» of the network
3
Except possible reselling capacity acquired in previous longer-term auction rounds where Qreq could be
counted negatively in the objective function and the capacity constraints.
4
It is possible to enhance the complexity of constraints taken into account (as long as linearity is maintained, this does not
add special difficulties). For instance, suppose that in the example C wishes to limit the import flow from A+B (because of
interior network constraints within C) : then a linear constraint (sum of the flows limited by this bound) could be added.

April 2001 21/22

• to charge the same payment to each participant bidding in the same (X,Y) pair of area. This price
will be written π(X,Y) (i.e. marginal pricing principle) ;
• to never overpass the bid price of any participant when using π(X,Y) value.
The solution given by the Theory of Optimisation is to use a dual variable πBC associated to each BC.
For each interconnection ε, this dual variable equals the variation of the objective function associated
to a 1 MW move of BC.
πBC(ε) equals to zero if the interconnection is not saturated - that is to say if the capacity BC(ε)
exceeds the demand in the interconnection after clearing - it is positive if the interconnection is
saturated5.
From these set of πBCs, the marginal price π(X,Y) can be assessed by the formula:
π(X,Y) = Σε {r(X,Y,ε)* πBC(ε)}
r(X,Y, ε) are the coefficients of the participation matrix above defined.
Each bid participating to the constraint may then be charged this «marginal» price. It can be
mathematically shown that π(X,Y) is systematically lower that all the bid price put on (X,Y) pair of
area as soon as they get some capacity allocated.
The total auction revenue is then :
AR = ΣXY {π(X,Y)*Qall(X,Y)}
The summation being carried out only on the borders where congestion has actually occurred
( π(X,Y)=0 if the capacity is higher than the transmission demand).
About the revenues allocation to TSOs, here again the optimisation theory can be helpful : an idea
here could be to use the dual function in order to allocate the revenues, each TSO keeping half of the
revenue for each doublet.
First variant : allocation of auction revenues to TSOs according to the influence of the bids originated
from (or destined to) their area on the saturated interconnections :

Revenue of X = ½*[ΣY {π(X,Y)*Qall(X,Y)}

Second variant : allocation of auction revenue to TSOs according to the location of the saturated
interconnections :

Revenue of X = ½*[AR * ΣX{πBC(ε)} / Σε{πBC(ε)}]

AR = Total Auction Revenue
It is important to note that the "pay-as-bid" option, if chosen in the mechanism, would lead to
similar developments.

5
In some situations, the dual variable is discontinuous that is to say that the value when moving AC upwards
differs from the value when moving downwards. In such a case we will consider the maximum of these two
values in the following calculations.

April 2001 22/22

 Appendix VI

ETSO - Co-ordinated Use of Power Exchange for Congestion

Management

63
Co-ordinated use of Power Exchanges for
Congestion Management

Final Report
April 2001

1/12
 Table of contents

1. INTRODUCTION 3

2. MAIN PRINCIPLES FOR CONGESTION MANAGEMENT 3

3. THE NORDIC POWEREXCHANGE AND MARKET SPLITTING 4

4. REQUIREMENTS FORIMPLEMENTATION OF MARKET SPLITTING IN

EUROPE 8

5. OPTIONS FOR CO-ORDINATED USE OF POWER EXCHANGES FOR

CONGESTION MANAGEMENT IN EUROPE 9

6. CONCLUSIONS 12

2/12
1. INTRODUCTION

In the Conclusions from the Sixth meeting of the European Electricity Regulatory Forum,
Florence, 9.-10. November 2000 it is stated that appropriate mechanisms for cross-border
tarification and congestion management have to be in place to promote the development of
cross-border trade.

It is further stated that market based congestion management methods which

- ensure optimal utilisation of transmission capacity
- give appropriate price signals to the market parties and the TSOs involved
- are non-transaction based
are preferred.

According to the Florence Forum conclusions, market splitting meets these requirements, but is
considered to be too difficult to implement in the short-term, since it requires the existence of
exchanges or power pool based arrangements on both side of the interconnection.

The purpose of this paper is to discuss the obstacles and options for implementing a ”market
splitting” model in Continental Europe. First the Nordic market splitting model is described, then
the main obstacles for implementing a similar model in Continental Europe are outlined and
finally the options for a co-ordinated use of power exchanges for congestion management in the
area are described.

2. MAIN PRINCIPLES FOR CONGESTION MANAGEMENT

Although congestion management methods for open electricity markets exist in many variations,
the most commonly considered schemes can be categorised as:
- Explicit auctions, where only the transmission product (MW) between two areas is traded
- Implicit auctions, where both the energy (MWh) and the corresponding transmission product
(MW) between bidding/price areas are traded simultaneously and are coupled
- Counter trade/re-dispatch

In this context market splitting is a variant of implicit auctions, whereas counter trade/re-dispatch
is a decongestion method used by the TSOs to alleviate congestion for the benefit of market
players.

The main characteristic of implicit auctions is that transmission capacity and energy are coupled
and traded simultaneously, ensuring that transmission capacity is allocated according to energy
trading requirements.

3/12
3. THE NORDIC POWER EXCHANGE AND MARKET SPLITTING

Nord Pool is a power exchange including Denmark, Norway, Sweden, and Finland. These four
countries have liberalised their power markets and through the joint exchange, they have
established a joint power market.

Nord Pool has two market places: Elspot and Eltermin. Elspot is the spot market where physical
kWh are traded. Eltermin is a commercial centre in which price securing contracts are traded.

3.1 Elspot - the Spot Market

Participants who want to buy or sell kWh via Nord Pool’s Elspot, must send their bids and/or
offers to Nord Pool by 12 o’clock the day before delivery.

At Nord Pool, the bids and the offers for each hour the following day are put together to form one
total demand curve and one total supply curve. The so-called system price can be read at the
point where the two curves intersect one another. The system price is the price that would be
obtained in the whole Nord Pool area if there were no congestion.

Nord Pool determines a system price for each hour of the following day. This means that the
price on Elspot may fluctuate from hour to hour but the price is fixed for one hour at a time.

3.1.1 Bidding areas and price areas

Obviously, it may happen that, in one area, an amount of power, which is simply too small, has
been offered at the system price. This may happen if constraints in the grid makes it impossible
to transport enough power into this area. If so, the area becomes a so-called high price area, in
which the price is higher than the system price.

Correspondingly, it may occur that, in one area, too much power has been offered at the system
price, and the grid cannot transport the amount of power necessary out of the area. If this
happens, the area becomes a low price area: an area in which the price is lower than the system
price.

Every bid and offer is related to a given bidding area. Bidding areas turn into separate price
areas if congestion between them occurs as a result of the bids and offers.

4/12
The Nord Pool-area is divided into 6-8 geographically limited bidding areas. Norway can
internally be divided into several bidding areas, whereas Finland, Sweden, Jutland/Funen and
Zealand each make up one bidding area. The structure of the Nordic electricity system allows a
rather permanent definition of bidding areas, and it is always possible to calculate available
transmission capacity between them without significant interdependencies.

3.1.2 Constraints in the grid

All the transmission capacity between bidding areas is managed by Nord Pools spot market.

Nord Pool uses the capacity for conducting power out of low price areas and into high price
areas. Thereby, the price in high price areas is reduced and the price in low price areas is
raised. The system secures that for every hour of operation, all the capacity of the constraints is
utilised in accordance with current price signals.

This price discovery process is illustrated in the following figures.

Low price area High price area

Price/MWh Price/MWh
Offers
Bids Offers

PHigh - 0 cap.

PLow - 0 cap.
Bids

MWh/h MWh/h

Figure 1 Area prices without exchange between high and low price areas

Figure 1 illustrates how prices would have been set in the high and the low price areas without
any exchange between the areas. Figure 2 illustrates the change in area prices due to an
exchange from the low price area into the high price area according to the available
transmission capacity between the two areas.

5/12
 Low price area High price area

Price/ MWh Price/ MWh

Offers
Bids
Offers

PHigh - 0 cap.

P High

P Low

P Low - 0 cap.

Bids

MWh/h MWh/h
Capacity between
price areas

Figure 2 Area prices with exchange between low and high price areas

The exchange raises the price in the low price area from "P Low-0 cap. " to "PLow", and reduces the
price in the high price area from "PHigh-0 cap. " to "PHigh".

The next figure is included to illustrate how much transmission capacity is needed to avoid
congestion and thereby attain equal prices (=system price) in the two areas.

Low price area High price area

Price/ MWh Price/ MWh

Offers
Bids
Offers

PHigh - 0 cap.

P High
Systemprice - P Sys
P Low

P Low - 0 cap.

Bids

MWh/h MWh/h
Capacity between
price areas

Needed capacity to
attain equal prices

Figure 3 Needed transmission capacity to attain equal

prices (system price) in the two areas.

6/12
If the illustrated "needed capacity" is available, the exchange of electricity is sufficient to attain
equal prices in the two bidding areas and there will only be one price area.

With separate price areas sellers in the "exporting" low price area are paid the "low price" while
buyers in the importing high price area pay the "high price". The difference can be considered as
a capacity fee paid to the TSOs via the power exchange.

Market players cannot trade physical energy directly between bidding areas. This is impossible
because all the capacity of the constraint is managed by the power exchange.

3.2 Eltermin

Eltermin is a commercial centre where price-securing contracts are traded. In effect, both parties
involved in such contracts take out mutual insurance.

A contract is settled by comparing the average system price for the week concerned with the
hedge price in the contract. The difference in price is multiplied by the volume in the contract,
and this amount of money is transferred between the parties. A price-securing contract is
therefore not only a mutual insurance; it is also a mutual obligation.

The two parties involved in the futures contract do not know each other’s identity if the contract
has been made via Nord Pool’s market Eltermin. All settling of accounts takes place via Nord
Pool. Furthermore, Nord Pool guarantees the settling of accounts; Nord Pool enters the contract
if one of the parties cannot fulfil its obligations.

It is important to note that kWh are not exchanged between the parties of a price-securing
contract. Only money is exchanged.

The parties can submit bids and offers to Elspot with unspecified prices if they wish to trade the
energy physically. They do not need to worry about the price, because it is hedged in the futures
contract.

Since market players cannot trade physical energy directly between bidding areas, they have to
make a financial contract to be guaranteed a certain price. The two participants trade the
physical energy with the spot exchange, and afterwards they settle with each other in
accordance with the financial contract.

Participants within the same bidding area may also deal in physical kWh with each other. The
exchange has no monopoly within bidding areas.

7/12
4. REQUIREMENTS FOR IMPLEMENTATION OF MARKET SPLITTING
IN EUROPE

The Nordic implementation of the market splitting model is based on the following principles:
- all physical trade between congested bidding areas has to go via the power exchange (with
financial arrangements which have to be agreed upon by the involved parties)
- there is only one power exchange handling physical trade between bidding areas
- bidding areas are defined as rather permanent geographically limited areas (bottlenecks are
always at the same locations)
- all major permanent constraints are located at interconnections between bidding areas
- there is a low interdependency between individual net transfer capacities, so that they can
be calculated beforehand
- constraints within bidding areas are either handled as constraints on the interconnections
between price areas or handled by TSOs through re-dispatching.

It is evident that these conditions can not all be met immediately in the continental part of the
internal electricity market.

There are physical, structural and market obstacles:

- The highly meshed structure of the Continental Europe network complicates the definition of
bidding areas and available transmission capacity between these in advance. The physical
flow pattern and consequently the appearance of congestion changes considerably with both
demand and generation.
- The net transmission capacities across neighbouring areas are strongly interdependent and
cannot be calculated separately from one another. Thus the conventional “Market Splitting”
algorithm for the determination of the “maximum power” exchange between areas cannot be
applied.
- The electricity markets in Continental Europe are at present far from being compatible
enough to implement a common market splitting system
- Market participants should be asked for their agreement on the impact of market splitting on
bilateral trade between congested areas.

8/12
5. OPTIONS FOR CO-ORDINATED USE OF POWER EXCHANGES
FOR CONGESTION MANAGEMENT IN EUROPE

Assuming that market actors are ready to accept that direct bilateral trade between congested
areas may be affected financially by the congestion management scheme, it is then necessary
to find practicable solutions to overcome both the physical and the structural obstacles
mentioned above before considering the implementation of a market splitting model in
Continental Europe.

Market splitting does not solve the problem with varying and strongly interdependent constraints
in the transmission network in Continental Europe. As explained above, market splitting
presupposes well-defined bidding areas and available transmission capacities between these.

The simplest possible design is obtained by allowing only one power exchange to handle all
physical trade between bidding areas and thereby exclude all bilateral trade between these as in
the Nordic model. Nevertheless, this design does not seem to be acceptable by all European
market actors.

A more complex design would be required to allow:

- several power exchanges to handle physical trade between bidding areas and
- bilateral trading between bidding areas as well.

In such a design,
- strong co-ordination between power exchanges and bilateral trade and
- strong co-ordination between several power exchanges
must be developed to implement market splitting.

This is the minimum design to operate without capacity interdependencies between areas. The
case where interdependencies exist is addressed in paragraph 5.4

In the following it is assumed that there will be at least one power exchange operating a spot
market in a given bidding area.

5.1 Definition of bidding areas and available transmission capacity

A prerequisite for market splitting is a rather permanent definition of a number of geographically

limited bidding areas. The criteria for defining these areas is the location of the physical
constraints in the transmission network, that is: all major physical constraints are located at the
interconnections between bidding areas.

9/12
In the market splitting model the market players only experience constraints between bidding
areas. Constraints occurring within bidding areas should ideally be handled via re-dispatch by
the TSO, but can also be taken into account when setting the limits for the interconnections.
Market splitting as well as most kinds of explicit auctioning (except co-ordinated auctioning) has
to be based on a set of calculated values for available transmission capacity.

The definition of bidding areas and calculation of available capacity on interconnections should
be based on load-flow calculations which locate the major constraints in the transmission
network.

5.2 Co-ordinated use of several Power Exchanges

With more than one power exchange handling physical trade between bidding areas, it would be
necessary to ensure co-ordination between the implicit capacity auctions, by integrating the net
balances for each bidding/price area.

When congestion management is based on market splitting, the allocation of transmission

capacity must be based on the same marginal value for the difference between area prices at all
power exchanges, i.e. the capacity fee for a certain congestion between two price areas must be
the same at all power exchanges. This argument excludes ex-ante allocation of capacity
between the power exchanges.

The traded products and their prices in a certain area do not necessarily have to be the same at
all power exchanges. Products could for instance slightly differ with respect to time interval, risk
and fee, even if some requirements must be met in terms of power exchange gate closure
synchronisation, etc.

The allocation of transmission capacity between power exchanges could be based on

summarised bids for capacity per power exchange for each interconnection (capacity fee =
difference in area price).

A high degree of co-ordination is needed between Power Exchanges and TSOs.

5.3 Concurrent operation of Co-ordinated use of Power Exchanges and

Bilateral Trade

A system with both power exchange trade and bilateral trade between price areas calls for a
method to distribute the available transmission capacity between these two types of trading. In
this paper all spot trade on power exchanges is assumed to take place day ahead.

A possible method is an a-priori allocation of slices of transmission capacity to explicit capacity

auctions and power exchanges respectively.

10/12
The distribution of available capacity between power exchanges and explicit auctions has to be
decided before these are carried out. It could be decided to offer a slice corresponding to X% of
the available capacity in explicit auctions and allocate the rest to power exchanges. To ensure a
high utilisation of capacity the longer term explicit capacity auctions (for instance yearly and
monthly) have to be combined with implementation of the "use it or lose it" principle ahead of the
daily allocations. Unused capacity bought in long term explicit auctions would be entered into the
daily spot market.

Theoretically it is possible to sell capacity obtained in the longer-term capacity auctions on the
power exchange by entering a capacity-offer (price/MW) into the spot market. This would
however partly violate the "use it or lose it"-principle for long term capacity reservations.

An alternative to capacity slicing is to let power exchanges buy capacity in the explicit capacity
auctions on equal terms with all other market players, and then offer this capacity in the implicit
auctions.

With market splitting a capacity fee is obtained due to the difference between area prices in case
of congestion. This capacity fee is at first collected by the power exchange. If a slice of capacity
is put at free disposal for the power exchanges, without charge, the capacity fee should be paid
to the TSOs. If on the other hand the power exchanges beforehand have to buy capacity at their
own risk in explicit capacity auctions the capacity fee obtained by the later market splitting could
be considered a recovery of these prepaid costs.

Net revenues from congestion management should in all cases be allocated to grid investment.

In market splitting the allocation of capacity to trades in the power exchange is done implicitly on
the basis of the bids and the resulting difference between area prices. A way to include bilateral
trades in this capacity allocation is to allow bids for capacity alone (price/MW) in the spot
exchange along with bids for energy (price/MWh). By letting these bids enter into the implicit
auction and compete for capacity with the power exchange energy trades, it is possible to
allocate capacity to all day ahead trades on equal terms. Capacity is allocated to bilateral trades
with higher bids for capacity than the resulting difference in area prices.

5.4 Improvements to take into account network interdependencies

ETSO has also produced a document describing the co-ordinated auctioning allocation method.
The concepts used in this method could also be used to take into account network constraints
for the optimal management by TSOs of physical exchanges of power between different areas
managed by different Power Exchanges. Nevertheless, this sophisticated evolution of market
splitting has still to be worked out.

11/12
6. CONCLUSIONS

Market splitting is a very interesting principle for congestion management, but it has severe
requirements that have to be addressed before considering implementation outside Nordel.

In a system where the location of potential constraints is rather permanent and the capacity of
these can be calculated ex-ante, and where bilateral trade is not permitted, market splitting in
the Nordic form can ensure an optimal market based utilisation of congested lines.

In Continental Europe there are a number of physical, structural and market obstacles to market
splitting in its simple form:
- It is a highly meshed network where both the location and the capacity of congested lines
changes considerably with demand and generation.
- The Net Transmission Capacities across neighbouring constraints are strongly
interdependent and cannot be calculated independently.
- Market participants should be asked for their agreement on the impact of market splitting on
bilateral trade between congested areas.
- The electricity markets in Continental Europe are at present not compatible enough to
implement a common market splitting system.

An advanced and rather complicated form for market splitting based on the allocation principles
described in the ETSO paper "Co-ordinated Auctioning of Transmission Capacity in Meshed
Networks" could be a potential candidate for a future congestion management method in
Continental Europe, but this sophisticated evolution of market splitting has still to be worked out.
For the time being co-ordinated use of power exchanges for congestion management is not
considered to be a method that can be implemented in Continental Europe.

12/12