18 - Congestion Management

ECE 573 – Power System Operations and
Control
18. Congestion Management
George Gross
Department of Electrical and Computer Engineering
University of Illinois at Urbana–Champaign
ECE 573 © 2007 – 2017 George Gross, University of Illinois at Urbana-Champaign, All Rights Reserved. 1
OUTLINE
q  Review of the centralized electricity market (CEM )
q  Transmission – unconstrained and constrained
markets
q  Market performance and congestion impact

metrics
q  Congestion and local market power

q  Integration of bilateral transactions with the CEM
q  Analytic framework for congestion analysis
BASIC PREMISE
q  Congestion is a major obstacle to the functioning
of vibrant, competitive electricity markets
q  A widely–used congestion management scheme
is based on the locational marginal prices (LMPs)
q  LMPs are purely short–term signals
q  The many lingering problems, such as

BASIC PREMISE
m  transmission services for all CEM players and
all bilateral transaction parties
m  determination of realistic price signals
m  effective use of the financial transmission
rights (FTR)
need to be addressed
q  The construction of a comprehensive analytic
framework is a key requirement

RTO CONGESTION COSTS
2,000
PJM
1,500
NYISO
CAISO
million $
1,000
500
Sources: NYISO, PJM, CAISO.

PJM CONGESTION COSTS
3.5
3.0
2.5
$/MWh
2.0
1.5
1.0
0.5
0.0
2000 2001 2002 2003 2004 2005 2006
Source: PJM State of the Market Reports
THE MARKET TIME FRAME
q  We define one hour as the smallest indecomposable

unit of time and focus on a specified hour h
q  We discuss the market decisions for that specified
hour that are made at various times
h – year h – month h – day h h + month
time
axis
year– month– day– specified settlement
ahead ahead ahead hour time
market decisions
THE CENTRALIZED ELECTRICITY
MARKET (CEM )
q  We discuss the structure of the forward day-ahead
market; we examine the centralized electricity market,
sometimes called the pool market, in detail to bring
a clear understanding of its features
q  In fact, the day–ahead market or DAM is a
collection of 24 separate commodity markets, one
for each hour of the next day

THE CENTRALIZED ELECTRICITY
MARKET (CEM)
q  Our focus is on the market corresponding to the
specified hour h
h – year h – month h – day h h + month

time
axis
day–ahead CEM
q  We suppress h in our notation in our discussion

CEM STRUCTURE
seller 1 ... seller i ... seller N s
MWh MWh MWh

$ $ $
CEM operator
MWh $ $ MWh $
MWh
buyer 1 ... buyer j ... buyer N b
CEM PARTICIPANTS
q  The CEM operator is responsible for this market
and deploys auction mechanisms to determine the
prices/quantities bought and sold in each hour
q  Sellers are generation entities and brokers/
marketers
q  Buyers are consumers, brokers/marketers,
distribution entities and generation entities

CEM
q  CEM sellers and buyers submit sealed offers and
bids, respectively; each offer/bid specifies the
price and quantity at which the players are willing
to sell and buy the electric commodity – the MWh
q  The CEM operator determines the successful offers
and bids and the market clearing price obtained by
the maximization of the social surplus
q  The auction results determine the unit commitment
and the dispatch of the physical units
SELLER s AT NODE i OFFER AND
IGO COSTS
offer curve β is ( p is ) the costs to the

$/MWh IGO of the seller
s sale of p is
MW at node i
are the area
under the curve
p is
C i ( p is ) = ∫ β is (ξ ) d ξ
0
p is, min p s
i
p is, max MWh/h
BUYER b AT NODE j BID AND
BENEFITS
p bj
B j ( p bj ) = ∫ β bj (ξ )dξ
0
$/MWh
bid curve β bj ( p bj )
the benefits of the
buyer at node j
are given by the
area under the
curve
b
0 p j MWh/h
THE SOCIAL WELFARE
The social welfare is defined as the total

benefits of the buyers minus the total costs
of the sellers:
Nb Ns
b s
S ! p
∑ j j
B ( ) − p
∑ i i
C ( )
j=1 i=1
social welfare total benefits total costs
MAXIMIZATION OF THE SOCIAL
WELFARE
q  The objective in markets is to maximize the social
welfare, so as to determine the maximum net
benefits for society as a whole
q  When we do not consider the transmission cons-
traints, the CEM operator solves the optimization
problem to determine the successful offers/bids
b s
( p ) − ∑C ( p )
N N
∑B
b s
max S = j j i i
j =1 i =1
s.t. b s
N N
∑p ∑p
b s
j
= i supply – demand balance
j =1 i =1
MAXIMIZATION OF THE SOCIAL
WELFARE
$/MWh demand
curve
market
clearing mar
equi ket
price social welfare libri
poin um
t
supply curve
market clearing MWh/h

quantity
MARKET CLEARING PRICE
q  The market clearing price ρ * (system marginal price)
is the change in the social welfare for a unit
change in the market clearing quantity
q  Each seller receives ρ * from the CEM operator for
each MWh sold
q  Each buyer pays ρ * to the CEM for each MWh
bought
q  The market clearing price is different from the
offer/bid price of nearly every player
A TWO – BUS SYSTEM EXAMPLE
s1 ~ ~ s3 ~ s2
bus 1 f max
ℓ
= 800 MW bus 2
lossless line
b1 b2 b3
OFFERS AND BIDS
$/MWh $/MWh
6 10
s1 : 5 b1 :
5
300 700 MWh/h 400 600 MWh/h
$/MWh $/MWh
8 12
b2 :
s2 : 4
200 500 MWh/h 500 MWh/h
$/MWh $/MWh
9
s3 : b3 : 7
3
700 MWh/h MWh/h
600 800
SOCIAL WELFARE MAXIMIZATION
IGNORING TRANSMISSION
$/MWh
500 @ 12 200 @ 7
400 @ 10
600 @ 9
S = 9,300
ρ* = 7 300 @ 8
400 @ 6 200 @ 5
300 @ 5
200 @ 4
700 @ 3
MWh/h
1,600
MARKET OUTCOMES WITHOUT
TRANSMISSION CONSIDERATION
s1 s2 s3
7 $/MWh 700 MWh
700 MWh 200 MWh 7 $/MWh
7 $/MWh
CEM operator
500 MWh 7 $/MWh

400 MWh 7 $/MWh
7 $/MWh 600 MWh
b1 b2 b3
total sales = 1,600 MW S = 9,300

UNCONSTRAINED SYSTEM
SETTLEMENT
quantity price revenues payments

participant
[MWh] [$/MWh] [$] [$]
s1 700 7.0 4,900 -
s2 200 7.0 1,400 -
s3 700 7.0 4,900 -
b1 400 7.0 - 2,800
b2 500 7.0 - 3,500
b3 700 7.0 - 4,900
total 1,600 7.0 11,200 11,200
A THREE – BUS SYSTEM EXAMPLE
s1 ~ lossless system ~ s2
bus 1 j 0.1 bus 2
b1 CEM b2
j 0.1 operator j 0.1
bus 3
max
f 13 = 200 MW
b3 ~ s3
THREE – BUS SYSTEM: OFFERS AND
BIDS
$/MWh $/MWh
10 70
b1 : 50
s1 : 5
300 600 MWh/h 200 300 MWh/h

$/MWh $/MWh
20 80
b 2 : 60
s 2 : 10
200 600 MWh/h 200 400 MWh/h

$/MWh $/MWh
s 3 : 29 300
b3 :
20
200 450 MWh/h MWh/h
800
THREE – BUS SYSTEM: TRANSMISSION
UNCONSTRAINED EQUILIBRIUM
$/MWh
800 @ 300
200 @ 80
200 @ 70
S = 265, 600 200 @ 60
100 @ 50
λ * = 29
400 @ 20
200 @ 10 100 @ 29
300 @ 5 200 @ 20
MWh/h
300 @ 10 1,500
UNCONSTRAINED SYSTEM
REVENUES AND PAYMENTS
quantity price revenues payments

participant
[MWh] [$/MWh] [$] [$]
s1 600 29 17,400 -
s2 600 29 17,400 -
s3 300 29 8,700 -
b1 300 29 - 8,700
b2 400 29 - 11,600
b3 800 29 - 23,200
total 1,500 29 43,500 43,500
BASIC MARKET PERFORMANCE
MEASURES
q  The social welfare is a measure of the performance
of the market as a whole but it fails to provide
insights into the performance of each individual
player
q  We define two key components of social welfare
m  producer surplus
m  consumer surplus
PRODUCER SURPLUS
q  For a seller s at node i, the individual producer surplus
measures the difference between the revenues
received for the sale at the market clearing price and
those that would be received at the offer price
S is = ρ * ⋅ p s − C i p s
i
( ) i
i = 1, ... , N s
revenues
costs
under bid under offer
q  The total producer surplus sis
N
S s
= ∑S s
i
i=1
CONSUMER SURPLUS
q  For each buyer b at node j, the individual consumer
surplus measures the difference between the
payments for the commodity at the bid prices of
the buyer and those at the market clearing price
S b
j
= B j ( p bj ) − ρ * ⋅( p bj ) j = 1, ... , N b
actual payments
*
benefits under ρ
q  The total consumer surplus is

b
N
S b
= ∑S b
j
j=1
THE TWO – BUS SYSTEM EXAMPLE
$/MWh
S s= S s
1
+S s
2
+S s
3
= 4, 400
500 @ 12
400 @ 10 200 @ 7
600 @ 9
300 @ 8
ρ* = 7 s
S s
S s S
400 @ 6 1 200 @ 5
3 2
300 @ 5
200 @ 4
700 @ 3
MWh/h
1,600
$/MWh
b b b b
500 @ 12 S =S 1
+S 2
+S 3
= 4, 900
400 @ 10 200 @ 7
S b 600 @ 9
2 b
S 1 S b
* 3 300 @ 8
ρ =7
400 @ 6 200 @ 5
300 @ 5
200 @ 4
700 @ 3
MWh/h
1,600
$/MWh
S = S s +S b
= 9, 300
500 @ 12
400 @ 10 200 @ 7
600 @ 9
300 @ 8
ρ* = 7
400 @ 6 200 @ 5
300 @ 5
200 @ 4
700 @ 3
MWh/h
1600
UNCONSTRAINED EQUILIBRIUM
b
$/MWh S 2
800 @ 300 b
S 1
200 @ 80
b
S 3 200 @ 70
200 @ 60
100 @ 50
λ * = 29 s
S s S 2
400 @ 20
100 @ 29
s
1
200 @ 10
S 3
300 @ 5 200 @ 20
MWh/h
S s 300 @ 10 s 1,500
1 S 2
UNCONSTRAINED DISPATCH
s1 ~ 600 MW ~ s2 600 MW
bus 1 bus 2
b1 CEM b2
300 MW 400 MW
lossless system
bus 3
f 13max = 200 MW
b3 ~ s3 λ *
= 29
800 MW 300 MW
A TRANSMISSION CONSTRAINT
s1 s3 s2
~ ~ ~
700 MW 700 MW 200 MW
bus 1 f max
ℓ
= 800 MW bus 2
actual flow: 1,000 MW

b1 b2 b3
400 MW 500 MW 700 MW
IMPACTS OF THE TRANSMISSION
NETWORK IN THE TWO – BUS SYSTEM
To eliminate the line overload, we reduce the sale
from s 1 by 100 MW and instead use the higher –
priced energy from s 2
s1 s3 s2
~ ~ ~
700 – 200 200 + 100
700 MW = 300 MW
= 500 MW
bus 1 f ℓmax = 800 MW bus 2
b1 1,000 – 200 = 800 MW b 2 b3
!###"### $
400 MW actual flow 500 MW 700 – 100
= 600 MW
TWO – BUS EXAMPLE: DISCONNECTED
MARKETS AT BUSES 1 AND 2
bus 1 bus 2
$/MWh $/MWh
500 @ 12
400 @ 10 12
600 @ 9
200 @ 7
200 @ 5 300 @ 8
400 @ 6
300 @ 5 200 @ 4
3 700 @ 3
MWh/h MWh/h
TWO – BUS EXAMPLE: INTERCONNECTED
MARKETS AT BUSES 1 AND 2
f max
ℓ
= 800 MW
bus 1 bus 2
TWO – BUS EXAMPLE:
UNCONSTRAINED MARKET AT BUS 1
$/MWh
400 @ 10
200 @ 5 400 @ 6
300 @ 5
3
700 @ 3
MWh/h
TWO – BUS EXAMPLE: BUS – 2 – DEMAND
– MODIFIED MARKET AT BUS 1
$/MWh modified demand
includes bus 2
500 @ 12 demand
400 @ 10
600 @ 9
200 @ 7
400 @ 6 200 @ 5
300 @ 5
3
700 @ 3
MWh/h
TWO – BUS EXAMPLE: CONSTRAINED
MARKET AT BUS 1
$/MWh
500 @ 12
400 @ 10 line
300 @ 9
constraint
ρ *1 = 6 impact
200 @ 5
300 @ 5
3 400 @ 6
700 @ 3
MWh/h
1,200
TWO – BUS EXAMPLE:
UNCONSTRAINED MARKET AT BUS 2
$/MWh
500 @ 12
12
600 @ 9
300 @ 8 200 @ 7
200 @ 4
MWh/h
TWO – BUS EXAMPLE: BUS – 1 – SUPPLY
– MODIFIED MARKET AT BUS 2
$/MWh
500 @ 12
12
200 @ 7
600 @ 9
300 @ 8
400 @ 6
300 @ 5
200 @ 4 modified supply
300 @ 3
includes bus 1 supply
MWh/h
TWO – BUS EXAMPLE: CONSTRAINED
MARKET AT BUS 2
$/MWh
500 @ 12
12
200 @ 7
600 @ 9
ρ *2 = 8 300 @ 8
200 @ 6
300 @ 5 line constraint
200 @ 4
300 @ 3 impact
MWh/h
1,100
TRANSMISSION – CONSTRAINED
SETTLEMENT
quantity price revenues payments change w.r.t. the
participants
[MWh] [$/MWh] [$] [$] unconstrained case [$]
s1 500 6.0 3,000 - – 1,900
s2 300 8.0 2,400 - 1,000
s3 700 6.0 4,200 - – 700
b1 400 6.0 - 2,400 – 400
b2 500 8.0 - 4,000 500
b3 600 8.0 - 4,800 100
total 1,500 9,600 11,200 –
difference is the congestion rents = 1,600

s 1 ~ 600 MW ~ s 2 600 MW
bus 1 bus 2
b1 IGO b2
300 MW 400 MW
lossless system
bus 3
max
f = 200 MW b3
λ* = 29
13
~ s3
800 MW 300 MW
THREE – BUS SYSTEM: PTDFs
1 MW 1/3 MW 1/3 MW 1 MW
1 2 1 2
3 3
1 MW 1 MW
s 1 ~ 600 MW ~ s 2 600 MW
bus 1 IGO bus 2
b1 2 1 b2
300 MW f13 = × 300 + × 200
3 3 400 MW
violation of = 266.67
the line
constraint bus 3
since b3 λ *
= 29
~ s3
f 13 > f max
13 800 MW 300 MW
q  The transmission unconstrained dispatch is infea-
max
sible because the line flow f 13 violates limit f 13
q  The net injections at buses 1 and 2 need to be
modified to drive the network to feasibility
q  The only choice of buyer b 3 is to bid sufficiently
high to induce seller s 3 to provide supply to meet
his load
UNDER TRANSMISSION CONSTRAINTS
Nb Ns
max S = ∑ j j ∑ i i)
B ( p b
) − C ( p s
j =1 i =1
s.t. set of
g n ( p 1s, ! , p sN s ; p 1b, ! , p bN b ) = 0 ∀ node n power flow
equations
real
f ℓ ( p 1s, " , p sN s ; p 1b, " , p bN b ) ≤ f max
∀ line ℓ power
ℓ
line flow
limits
CONSTRAINED MARKET CLEARING
PRICES
q  The equality constraints represent the balance of
power flow at each node n
q  The supply/demand equilibrium at each node n
may lead to different nodal market clearing prices
q  Each seller or buyer faces the market clearing
price at the node where he is located

CONGESTED LINE AND SYSTEM
q  We call a transmission line ℓ congested if the real
power line flow violates the line limit, i.e., the
corresponding inequality constraint becomes
binding or active :
fℓ ( p , " , p N s ; p , " , p N b ) = f
s
1
s b
1
b max
ℓ
q  We call the transmission system congested if there
are one or more congested lines in the network

CONGESTION
q  Power system reliability/security considerations
require secure operations not only under base case
conditions but also under the set of postulated
contingency cases
q  Congestion occurs if one or more limit violations
occur under either the base case or in any of the
postulated contingency cases
q  The consideration of transmission constraints
requires the representation of the base case and
all the postulated contingency cases
q  The transmission unconstrained dispatch is infea-
max
f
sible because the line flow f13 violates limit 13
q  The net injections at buses 1 and 2 have to be
modified to drive the network to feasibility
q  The only choice of buyer b 3 is to bid sufficiently
high to induce seller s 3 to provide the supply
needed to meet his load

THREE – BUS SYSTEM: ENSURING
TRANSMISSION FEASIBILITY
(300 – x) MW 1 2 (200 – 2x) MW
(500 – 3x) MW
2 1 max
×(300 − x ) + ×(200 − 2 x ) = 200 MW f 13
3 3
THREE – BUS SYSTEM: REDISPATCH
OF SUPPLY
q  x is the amount of redispatch due to the impacts
max
of the f 13 constraint on seller s 1
q  2x is the amount of redispatch due to the impacts
max
of the f 13 constraint on seller s 2
q  Redispatch calculation obtains
2 1
× (300 – x ) + × (200 – 2 x ) = 200 MW
3 3
so that
x = 50 MW
THREE – BUS SYSTEM: REDISPATCH
OF SUPPLY
q  Then, the IGO reduces the output of seller s 1 by 50
MW and that of seller s 2 by 100 MW
q  Since there is a willingness to pay by the buyer b 3,
the IGO increases the output of seller s 3 by 150 MW
q  The constrained dispatch changes the output of
each seller and may also impact the load supplied
to buyer b 3
CONSTRAINED DISPATCH
s 1 ~ 550 MW 500 MW ~ s2
bus 1 bus 2
b1 IGO b2
300 MW 400 MW
lossless system
bus 3
max
f 13 = 200 MW b3 ~ s3
800 MW 450 MW
THREE – BUS SYSTEM: LMPs
s 1 ~ 550 + 1 MW 500 + 1 MW ~ s 2
bus 1 λ *1 = 10 λ *2 = 20 bus 2
b1 IGO b2
300 + 1 MW 400 + 1 MW
lossless system
bus 3
b3 ~ s3
800 MW 450 MW
s 1 ~ 550 +Δ P1 MW 500 +Δ P2 MW ~ s 2
bus 1 λ *1 = 10 λ *2 = 20 bus 2
b1 IGO b2
300 MW 400 MW
lossless system
Δ P1 + Δ P2 = 1
bus 3 2 1
λ *3 = ? ⋅ Δ P1 + ⋅ Δ P2 = 0
b3 3 3
~ s3
800 + 1 MW 450 MW Δ P1 = − 1 Δ P2 = 2
s 1 ~ 550 MW 500 MW ~ s2
bus 1 λ *1 = 10 λ *2 = 20 bus 2
b1 IGO b2
300 MW 400 MW
lossless system
bus 3 λ *3 = 2•20+( – 1)•10

b3 ~ s3 = 30
800 MW 450 MW
s 1 ~ 550 MW 500 MW ~ s2
bus 1 λ *1 = 10 λ *2 = 20 bus 2
b1 IGO b2
300 MW 400 MW
lossless system
max
bus 3 λ *3 = 30
f 13 = 200 MW b3 ~ s3
800 MW 450 MW
SOCIAL WELFARE AND SURPLUSES
$/MWh
consumer surplus
producer surplus
MWh/h
EVALUATION OF METRICS
$/MWh
consumer
surplus
ρb
congestion
rents
ρs
market dead –
efficiency weight
loss loss
producer MWh/h
surplus
IMPACTS OF CONGESTION
q  Congestion in the system leads to a change from
the single market equilibrium point to different
nodal equilibrium points, possibly one at each node
q  Change in the preferred schedule for the required
generation–demand balance may lead to possible
curtailment in production or in consumption or in
both
q  The individual player surpluses change from the
unconstrained market values to those of the
separate markets at each bus under the
constrained conditions
OTHER LONGER – TERM
CONGESTION IMPACTS
q  Increase in costs arising from the delayed
connection of new generation resources
q  Reduction in overall system reliability/security
q  Pollution due to the commitment/dispatch of older
and less efficient plants that must be operated
only for reliability/security purposes

CONGESTION MEASURES
q  The impacts of congestion may be measured in
terms of the power/energy that needs to be
redispatched and/or the financial costs in $ on the
various players
q  Measures of congestion impacts in $
m  redispatch costs
m  congestion rents
m  market efficiency loss

CONGESTION RENTS
q  In the transmission – constrained case, we have
different prices at the system nodes, so the
players may face different market clearing prices,
depending on their respective locations
q  The social welfare in this case is given by
Nb Ns
ˆ ˆ s ˆ
S =S +S + b
∑ρ b
⋅ p − ∑ ρi ⋅ p s
j j i
j=1 i=1
congestion rents κ
CONGESTION RENTS
q  In the constrained case, the congestion rents are
part of the social welfare
b s
S =S +S +κ
c c c
q  The congestion rents are sometimes called the

merchandising surplus and correspond to the
difference between the amounts paid by buyers

and the amounts received by sellers; the
congestion rents are collected by the IGO

MARKET EFFICIENCY LOSS
q  Congestion may reduce the social welfare of the
market due to the physical constraint impacts
q  This reduction is called market efficiency loss and is
defined by
market constrained
efficiency social welfare
loss
E =− S ( c
−S
u )
q  In economics, the market efficiency loss is also
known as deadweight loss
TWO-BUS SYSTEM: CONSTRAINED
CASE – BUS 1
$/MWh
consumer
500 @ 12
surplus
400 @ 10
300 @ 9
400 @ 6
ρ *1 = 6
300 @ 5 200 @ 5
700 @ 3 producer
surplus
MWh/h
CASE – BUS 2
$/MWh
consumer
500 @ 12 surplus
200 @ 7
600 @ 9
*
ρ =8
2 300 @ 8
200 @ 6
300 @ 5
200 @ 4
300 @ 3 producer
surplus
MWh/h
CASE
seller surplus ($) buyer surplus ($)
s1 300 b1 1,600
s2 800 b2 2,000
s3 2,100 b3 6000
total 3,200 total 4,200
congestion rents ($) 1,600
social welfare ($) 9,000
TWO – BUS SYSTEM: MARKET
EFFICIENCY LOSS
q  For the unconstrained case we have
S = 9, 300
u
q  For the constrained case we have
S = 9, 000
c
q  The market efficiency loss is, therefore,
E = − S ( c
−S
u ) = 300
THREE – BUS SYSTEM: MARKET
EFFICIENCY LOSS
q  For the unconstrained case we have
S = 265, 600
u
q  For the constrained case we have
S = 263,750
c
q  The market efficiency loss is
E =− S ( c
−S
u ) = 1, 850
THREE – BUS SYSTEM: CONSTRAINED
CASE
seller surplus ($) buyer surplus ($)
s1 1,500 b1 16,000
s2 2,000 b2 20,000
s3 2,250 b3 216,000
total 5,750 total 252,000
congestion rents ($) 6,000
social welfare ($) 263,750
REDISPATCH COSTS
q  The blocks, whose prices are higher than the

*
market clearing price ρ in the unconstrained case,
are sometimes called the out – of – merit blocks
q  The redispatch costs correspond to the additional
costs incurred due to the dispatch of these out – of
– merit blocks in order to avoid violations of
transmission constraints
IMPACTS OF THE TRANSMISSION
NETWORK IN THE TWO – BUS SYSTEM
To eliminate the line overload, we reduce the
sale from s 1 by 200 MW and instead use the
higher-priced energy from s 2
s1 s3 s2
~ ~ ~
700 - 200 200 + 100
700 MW
= 500 MW = 300 MW
bus 1 f max
ℓ
= 800 MW bus 2
b1 - 200 = 800 MW b b
1,000
!###"### $
2 3
400 MW actual flow 500 MW 700 – 100

= 600 MW
TWO – BUS SYSTEM: REDISPATCH
COSTS
$/MWh
300 @ 8
ρ* = 7
block must
be
300 @ 5 unloaded
200 @ 4 to prevent
400 @ 6
700 @ 3 congestion
– 200
MWh/h
1600
TWO – BUS SYSTEM: REDISPATCH
COSTS
$/MWh
redispatch costs = ( 8 – 6 ) ×100 = 200
300 @ 8
*
block must
ρ =7 be
dispatched
to avoid
violation of
300 @ 5 the line flow
200 @ 4 400 @ 6 limit
700 @ 3 – 200 100
MWh/h
FACTORS CAUSING SOCIAL
WELFARE REDUCTION
q  The redispatch of higher – priced blocks to
replace the output of the lower – priced generation
that could be used when the constraints are not
considered
q  The decrease in market efficiency: the deadweight
loss in efficiency due to the fact that the output of
the displaced units cannot be used and sold
q  The decrease in the producer surplus of some
sellers
FACTORS CAUSING SOCIAL
WELFARE REDUCTION
q  The decrease in the consumer surplus of some
buyers
q  The need for additional ancillary services
provided by sellers charging the higher prices
q  The creation of situations that may lead to the
exercise of market power

LOCATIONAL MARGINAL PRICES
q  The costs to supply an additional MWh at different
nodes in a constrained system may be different
for each node
q  The locational marginal price or LMP µ n is
defined as the cost to supply an additional MWh
at node n
q  If there is no congestion, the LMPs are the same
at all the system nodes and therefore, each LMP
is equal to the marginal clearing price since none
of the transmission constraints is binding
LOCATIONAL MARGINAL PRICE
q  A buyer at node n pays µ n $/MWh for electricity
purchased
q  A seller at node m is paid µ m $/MWh for electricity
sold
q  The difference in the LMPs at a node pair includes
implicitly the charges for congestion; the IGO
collects these charges
q  Since the market outcomes are based on the
optimization results under transmission
constraints, the transmission capability is
allocated optimally among the buyers and sellers
LMPs FOR THE TWO – BUS SYSTEM
EXAMPLE
s1 s3 s2
~ ~ ~
[500 +1] MW 300 MW
700 MW
@6 @8
bus 1 800 MW bus 2

800 MW
!"#
b1 b2 b3
actual flow
[400 +1] MW 500 MW 600 MW
µ1 = 6 $ / MWh
LMPs FOR THE TWO – BUS SYSTEM
EXAMPLE
s1 s3 s2
~ ~ ~
500 MW [300 +1 ] MW
700 MW
@6 @8
bus 1 800 MW bus 2

800 MW
b1 b2 b3
400 MW 500 MW [600 +1] MW
µ2 = 8 $ / MWh
MARKET OUTCOMES INCORPORATING
TRANSMISSION CONSTRAINTS
s1 s2 s3
6 $/MWh 700 MWh
500 MWh 300 MWh 8 $/MWh
6 $/MWh
IGO
400 MWh 8 $/MWh

500 MWh 8 $/MWh
6 $/MWh 600 MWh
b1 b2 b3
s 1 ~ 550 MW 500 MW ~ s2
bus 1 λ *1 = 10 λ *2 = 20 bus 2
b1 IGO b2
300 MW 400 MW
lossless system
max
bus 3 λ *3 = 30
f 13 = 200 MW b3 ~ s3
800 MW 450 MW
SENSITIVITY STUDY IN THE TWO –
BUS SYSTEM
q  We investigate the impacts of changing the offer
of seller s 2 for his second block by varying the
offer price from 8 to 9.2 $/MWh; the other offers/
bids remain unchanged
q  We evaluate the resulting surpluses for the
various values of the offers submitted

BUS SYSTEM
$/MWh
500 @ 12
200 @ 7
600 @ 9
* * 9
ρρρρ ====8.6
*
2 2* 8.4
8.2
8
300 @
300 @
300
98.6
@ 8.2
8.4
22 300 @ 8
200 @ 6
300 @ 5
200 @ 4
300 @ 3
MWh/h
BUS SYSTEM
$/MWh
500 @ 12
200 @ 7
600 @ 9
ρ =9
*
2 300 @ 9
200 @ 6
300 @ 5
200 @ 4
300 @ 3
MWh/h
BUS SYSTEM
limiting price
3000 S
$ S
Series1
2
2500
B
2000 S
Series2
2
1500
B
S
Series3
3
1000
congestion
500 rents
Series4
0 market
8 8.2 8.4 8.6 8.8 9 9.2 efficiency
Series5
$/MWh loss
SENSITIVITY STUDY IN THE
THREE – BUS SYSTEM
q  We investigate the impacts of changing the offer
of seller s 3 for his second block by varying the
offer price from 29 to 330 $/MWh; all the other
offers/bids remain unchanged
q  We evaluate the resulting surpluses for the
various values of the offers submitted

THREE – BUS SYSTEM: λ * AND S s
3 3
60 350
50 300
40 250
$/MWh
200
10 3 $
30
150
20 interval where
λ *3 increasing price 100
10 does not impact the
S s 50
3 market outcomes
0 0
30 70 110 150 190 230 270 310
offer for the last block of seller s3 in $/MWh
THREE – BUS SYSTEM: PRODUCER
SURPLUS
90
80
s
70 S 2
60
10 3 $
50
40 s
S 3
30
20 S s
1
10
0 30 70 110 150 190 230 270 310

THREE – BUS SYSTEM: CONSUMER
SURPLUS
250
200 b
S 3
150
10 3 $
100
b b
S S
50 2 1
0
30 70 110 150 190 230 270 310
THREE – BUS SYSTEM: MARKET
PERFORMANCE MEASURES
100
90
80 congestion rents
70
60
market
10 3 $
50
efficiency losses
40
30
20
10
0
30 70 110 150 190 230 270 310
LOCAL MARKET POWER
q  Market power is the ability of a firm to profitably
raise the price of a product
q  The exercise of market power may be carried out
by:
m  the physical withholding of units
m  the financial withholding of units
q  Transmission constraints may create locational
market power since they may set up area markets
with limited importing capability
SIMULATION STUDIES
q  A seller changes his offer prices by varying the
offer price for the last block offered
q  We study the resulting variations of the producer
surplus, consumer surplus, congestion rents and
market efficiency loss
q  The simulations are performed on different
systems of various sizes

THE SEVEN – BUS SYSTEM EXAMPLE
offers bids
constrained line
s1 200 @ 5 600 @ 10 b1 100 @ 80 100 @ 50
b2 200 @ 100 -
~ s3 200 @ 40 100 @ 60
1 2 s4 200 @ 10 300 @ 15 b3 800 @ 500 -
s5 100 @ 20 300 @ 40 b4 200 @ 140 200 @ 120

~
4 s7 200 @ 30 200 @ 40 b5 100 @ 80 200 @ 50
~ b6 200 @ 120 200 @ 110

IGO
5 b7 100 @ 90 100 @ 50
3 6 7
~ ~
SEVEN – BUS SYSTEM: PRODUCER
SURPLUS
100
impacts of
changes in the
80 flow directions
S 3S
60
10 3 $
S 4S
40 S
S7
20
S1 S S 5S
0
0 100 200 300 400 500 600
SEVEN – BUS SYSTEM: CONSUMER
SURPLUS
350
300
250
10 3 $
200
150 b
S 3
100
50
0

SEVEN – BUS SYSTEM SENSITIVITY
300
congestion rents
250
200 market
10 3 $
150 efficiency
losses
100
50

THE IEEE 57 – BUS SYSTEM
~ 16
2
12
~ 1
17
~
3 ~
4
15
14
18
5
45
44
46 13
~
6 19 11
38
47
20
48 49 50
21
39 57
42 43 51
22 37
10
26
23
40 56 41
36
27 24
35
28 25 9
~
7 34
29 30
31
~ 32
8
52 53 54 33 55
THE IEEE 57 – BUS SYSTEM:
CONGESTION RENTS
10000
!
!
!
⌣!⌢
8000
6000
$
4000
2000
0
0 0.5 1 1.5 2 2.5 3
offer of seller s2 in $/MWh

THE IEEE 57 – BUS SYSTEM:
MARKET EFFICIENCY LOSS
7000
6000
5000
4000
$
3000
2000
1000
0
0 0.5 1 1.5 2 2.5 3
offer of seller s2 in $/MWh
SIMULATION RESULTS
q  Congestion situations produce, typically,
changes in the consumer, producer and social
surpluses, the additional congestion rents compon-
ent of the social welfare, and the market efficiency
loss with respect to the unconstrained case
q  Congestion creates situations which are
conducive to the exercise of market power

SIMULATION RESULTS
q  Under price – responsive demand, when a
particular seller increases his offer prices, the
impacts of congestion on the individual players
and the entire market are bounded due to the
asymptotic nature of the outcomes
q  We observe the existence of free – riders in the
market on both the supply – and demand – sides

SIMULATION RESULTS
q  There are also players that are negatively
impacted by the exercise of market power
q  The simulations underline the critical role of the
network topology and the relative location of the
market players in determining who are the losers
and the gainers as a result in such a market
power exercise attempt

BILATERAL TRANSACTIONS
q  We next incorporate the representation of
bilateral trades
q  A bilateral trade involves the delivery of a
specified quantity of energy from the seller to the
buyer; the delivery requires the provision of
transmission service by the IGO

h - year h - month h - day h h + month

time
axis
bilateral trade and transmission
scheduling
q  The bilateral transactions are negotiated/
scheduled at any time prior to their consumption

TRANSACTION DEFINITION
q  A bilateral trade is called a transaction and is
denoted by the triplet

ω w ! { m w , nw , t w }
injection withdrawal quantity

node node
q  We denote the set of all the transactions as
W ! {ω 1 , ω 2 ,!, ωW }
UNDER TRANSACTIONS
q  We incorporate the basic transactions into the
transmission-constrained social welfare
optimization problem statement
Nb Ns
max S = ∑ B j ( pB ) − ∑ C i ( pS )
j i
j=1 i=1
s.t.
g! n ( p 1s ,", pNs s ; p b1 ,", p Nb b ; ω 1 ," ω W ) = 0 ∀ node n
f!ℓ ( p 1s ,#, pNs s ; p b1 ,#, pNb b ; ω 1 ,# ω W ) ≤ f ℓmax ∀ line ℓ

UNDER TRANSACTIONS
q  Since ω 1 , ! , ω W do not appear in the objective
function, we may treat them as parameters and
rewrite the optimization problem with the
transactions explicitly represented

r
f
q  We define ℓ as the remaining real power flow limit
of line ℓ with all the basic transactions in the pool
taken into account

UNDER TRANSACTIONS
q  The formulation became
Nb Ns
max S = ∑ B j ( pB ) − ∑ C i ( pS )
j i
j=1 i=1
s.t.
g n ( p 1s ,!, pNs b ; p b1 ,!, pNb b ) = 0 ∀ node n
f ℓ ( p1s ,!, pNs s ; p1b ,!, pNb s ) ≤ f ℓr ∀ line ℓ

INCLUSION OF TRANSACTIONS IN
THE TWO – BUS EXAMPLE
100 MW ω1 = {1,2,100 MW } 100 MW
bus 1 bus 2
100 MW
50 MW ω 2 = {2,1,50 MW } 50 MW
bus 1 bus 2
50 MW
r
EVALUATION OF f IN THE ℓ
TWO – BUS EXAMPLE
f ℓmax = 800 MW
bus 1 bus 2
100 MW
50 MW
f ℓr = 800 - 100 +50 =750 MW

WITH TRANSACTIONS
s1 s3 s2
~ ~ ~
500 - 50 300 + 50
700 MW = 350 MW
= 450 MW
f ℓr =750 MW
bus 1 bus 2
800 – 50 = 750 MW
b1 b2 b3
400 MW 500 MW 600 MW
MARKET OUTCOMES
INCORPORATING TRANSACTIONS
s1 s2 s3
6 $/MWh 700 MWh
450 MWh 350 MWh 8 $/MWh
6 $/MWh
IGO
400 MWh 8 $/MWh

500 MWh 8 $/MWh
6 $/MWh 400 MWh
b1 b2 b3
S = 8, 950
IMPACTS OF TRANSACTIONS
q  f ℓr may be greater than, less than or equal to f ℓmax
q  The presence of transactions may increase,
decrease or maintain unchanged the active power
flow limit of a specified line; consequently,
transactions may relieve, exacerbate or keep
unchanged that line’s congestion

CONGESTION CHARGES
q  Whenever the presence of a transaction
exacerbates the congestion in a line, congestion
charges are imposed; these must be paid to avoid
the curtailment of the transaction
q  A commonly used scheme is to charge each MW
of a transaction the LMP difference between the
withdrawal and the injection nodes

CONGESTION CHARGES
q  For the transaction ω = {m, n, t } , the congestion
charges are
µn − µm t ( )
q  These charges are collected by the IGO at
settlement time
h - year h - month h - day h h + month

time
axis
settlement time
CONGESTION CHARGES IN THE TWO
– BUS SYSTEM EXAMPLE
q  The line is congested in the base case and
therefore LMPs are not equal
q  The congestion charges are
m  ω 1 = {1,2,100 MW } : IGO charges

$ 200 from ω 1
(µ 2 )
− µ 1 t 1 = (8 - 6) ×100 = $ 200
IGO pays
m  ω 2 = {2,1,50 MW } :
$ 100 to ω 2
(µ 1
− µ 2 ) t2 = (6 - 8) i 50 = − $ 100
h-year
seller
MWh
$ bilateral
transactions h-month
buyer
bilateral transactions are

day-ahead h-day
electricity trading contracts
market h
negotiated between a seller and
a buyer and are external to the h+month
day – ahead markets in the CEM time
ECE 573 © 2007 – 2017 George Gross, University of Illinois at Urbana-Champaign, All Rights Reserved. axis 125
TRANSMISSION SCHEDULING
CEM seller
bilateral
seller
IGO
bilateral
buyer
CEM buyer
transmission constraints
A NEW SCHEDULING TOOL
q  We develop a comprehensive formulation for the
transmission service provision that we call the
transmission allocation problem (TAP)
q  In the formulation, we
m  explicitly represent the contributions of each
bilateral transaction to the social welfare
m  consider the provision of transmission
services to all the customers on a consistent
basis in conformance with the willingness to
pay of each customer
TAP FORMULATION
q  The IGO’s TAP is essentially solved as a
scheduling problem for transmission service
provision with the objective to maximize total social
welfare
q  The formulation takes into account the side – by –
side operation of the CEM and the accommodation
of bilateral transactions
TRANSMISSION REQUESTS OF THE
seller at
ω ! m ,n ,t
w
{ w w w
} node m w
IGO tw
α w ( t w ) ≥ 0 for buyer at
node n w
t w ∈ ⎡⎣ 0, t w ⎤⎦
bilateral
willingness to transaction w
pay
FORMULATION OF THE TAP
q  We modify the social welfare expression to include
the contributions of the bilateral transactions
pool buyers’ pool sellers’

benefits costs
N W
# β b pb − β s p s % + α w t w
S = ∑$ n n n n & ∑ ( ) ( ) ( )
n=0 w =1
social welfare in conventional models

q  We explicitly represent the transaction impacts
in the constraints of the TAP
THE TAP STATEMENT
N W
max S = ∑ #$ β bn p bn − β ns p ns %& + ∑ α w t w
( ) ( ) ( )
p 0s , p 0 , p , p , t
b s b
n=0 w=1
s.t. p 0s − p 0b + pW0 = b T0 θ ↔ µ0
p s − p b + pW = B θ ↔ µ
max
Bd A θ ≤ f
s s
impacts of the 0≤ p ≤ p n n
n = 0,1, 2, ... , N
bilateral
0 ≤ pnb ≤ pnb n = 0,1, 2, ... , N
transactions
0 ≤ tw ≤ t w w = 1, 2, ... ,W
THE TAP SOLUTION
pool seller transmission

s*
at node i p i services for
w* seller at
t MW transaction node n
µ*i pis* from m to n
IGO
µ*j p s*j buyer at
(µ *
n −µ *
m )t w*
node m
p b*j
pool buyer
at node j congestion
charges
TAP : KEY ADVANTAGES
q  Transmission services are allocated in accordance
with each transaction’s willingness to pay: the
limited transfer capabilities are used so as to
provide services to entities – pool or bilateral

players – who value them most
q  The solution provides realistic short – term
economic signals: with the explicit representation
of the bilateral transactions, the LMPs determined

TAP : KEY ADVANTAGES
by the TAP solution reflect more realistically the
appropriate short-term price signals
q  The TAP solution improves the efficiency of the
market outcomes : the TAP results increase the
social welfare with the added bilateral transaction
contributions and entail smaller market efficiency
losses than those of conventional models

COMPARISON OF THE LMPs
COMPUTED FROM TAP AND TSP
IEEE 14 – bus system
45 TAP LMP TSP LMP
40
LMP ($/MWh)
35
30
25
20
15
10
5
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14
node n
LMP DEVIATION AS A FUNCTION OF
TRANSACTION VOLUME
1.4 µ n* − µ n*
TAP TSP
εµ =* bus 35
n
µ n*
1.2 TAP
bus 7
1
bus 54
0.8
εµ *
n
0.6 bus 12
bus 3
0.4 bus 28
bus 43
0.2
0
0 10 20 30 40 50 60 70
percentage of total load supplied by transactions
SOCIAL WELFARE DEVIATION AS A
FUNCTION OF TRANSACTION VOLUME
* *
S −S
TAP TSP
εS * = *
S
TAP
εS *
0 10 20 30 40 50 60
percentage of total load supplied by transactions
THE TAP
transmission
bilateral constraints energy sale
transmission quantities
requests for pool
customers
sellers’
offers
TAP LMPs
transmission
buyers’ services
bids for bilateral
transactions
UNCERTAINTY IN THE CONGESTION
CHARGES FROM TRANSACTIONS
h-year
congestion
charges
LMP seller h-month

difference
IGO
buyer day-ahead h-day
market
h
bilateral
transaction
h+month
time
axis
THE LMP DIFFERENCE VARIES
WITH TIME
BRANDONSH LMP BRUNSWICK LMP LMP differences
120
100
80
60
40
20
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
-20
data source: www.pjm.com

FINANCIAL TRANSMISSION RIGHTS
q  Financial transmission rights (FTR) are financial
instruments that entitle the holder to be reimbursed
for the congestion charges assessed in the DAM –
the day–ahead market
q  FTR provide price certainty to transmission
service customers in the electricity markets

THE FTR PAYOFF χ
χ ! µ n* − µ m* γ s
( )
χ FTR holder
receives from
γ IGO
* *
µ −µn m
FTR holder
pays to IGO
THE FTR MARKETS
h-year
FTR issuance
h-month
FTR secondary markets

h-day
h
FTR settlement h+month
time
ECE 573 © 2007 – 2017 George Gross, University of Illinois at Urbana-Champaign, All Rights Reserved. axis 143
FTR ISSUANCE MODEL
simultaneous
IGO feasibility
test
Γ1 =
{m1 , n1 , γ 1 , ρ1 }
ΓK γ K ρK
γ 1 ρ1 Γk
γ k ρk
FTR buyer 1 ... FTR buyer k ... FTR buyer K
THE SIMULTANEOUS FEASIBILITY
TEST (SFT)
FTR Γ1 ... Γk ... ΓK
fictitious
ω! 1 ω! k ! {mk , nk , γ k } ω! K
transactions
w" 1 w" k
PTDFs { }ϕℓ { } ϕ ℓ { } "
ϕ ℓwK
K
w" k
≤ f ℓmax yes
∑γ k ϕ ℓ feasible
k=1
∀ ℓ∈ L
no
infeasible
BASIC REQUIREMENTS FOR THE
FRAMEWORK
q  Simplicity : applications to the electricity market
analysis requires the models in the framework to
be simple and easy to understand
q  Comprehensiveness : close interactions among
the various elements of the problem requires
detailed representation of all of the three aspects
q  Flexibility : the lack of a standard market design
and the evolution of the electricity markets
requires the framework to be able to accommodate
different market designs
CONSTRUCTION OF THE FRAMEWORK
financial market layer
commodity market layer
physical network layer
CONSTRUCTION OF THE
FRAMEWORK
seller 1 ... seller i ... seller N s
MWh MWh MWh

$ $ $
IGO
MWh $ $ MWh $
MWh
buyer 1 ... buyer j ... buyer N b
CONSTRUCTION OF THE FRAMEWORK
CONSTRUCTION OF THE
FRAMEWORK
THE INFORMATION FLOWS

SFT
LMPs Ω
result
system Ω̂
states { p s , pb , t }
COMPUTATIONAL ISSUES
q  To provide computational simplicity, we adopt the
DC power flow assumptions and use the
distribution factors in the implementation
q  Such approximation may result in errors on the
outcomes of various markets
m  these factors are linear approximation of the
actual sensitivities
m  the factors calculated for the base case are
often used to evaluate impacts in changed
system conditions
ILLUSTRATIVE APPLICATIONS OF
THE FRAMEWORK
q  Assurance of the IGO revenue adequacy
q  Design of FTR markets with enhanced liquidity
m  new trading rules based on payoff parity
m  reconfiguration scheme based on FTR

valuation/pricing
q  Extension of the framework to analyze investment

issues: substitutability of transmission and
generation
ASSURANCE OF IGO REVENUE
ADEQUACY
pool
customers A = merchandising surplus
of the pool market
FTR
IGO holders
C = FTR
payoffs
bilateral B = congestion
transactions charges
IGO REVENUE ADEQUACY THEOREM
q  Given that
m  the set of existing FTR satisfy the SFT
m  the PTDFs used in the SFT equal the PTDFs
used in the TAP
q  The inequality
A+ B ≥ C
holds for the specified hour

.
REFERENCES
q  M. Liu and G. Gross, “Framework for the Design

and Analysis of Congestion Revenue Rights,”
IEEE Transactions on Power Systems, vol. 19:1,
pp. 243 – 251, February 2004.
q  M. Liu and G. Gross, “Congestion Rents and FTR
Evaluations in Mixed Pool-Bilateral Systems,”
International Journal of Electrical Power and
Energy Systems, vol. 30, issue 8, pp. 447 – 454,
October 2008.

18 - Congestion Management

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

18 - Congestion Management

Uploaded by

Copyright:

Available Formats

ECE 573 – Power System Operations and

18. Congestion Management

q Market performance and congestion impact

q Congestion and local market power

q Congestion is a major obstacle to the functioning

of vibrant, competitive electricity markets

q A widely–used congestion management scheme

is based on the locational marginal prices (LMPs)

q LMPs are purely short–term signals

q The many lingering problems, such as

all bilateral transaction parties

m determination of realistic price signals

m effective use of the financial transmission

q The construction of a comprehensive analytic

framework is a key requirement

Sources: NYISO, PJM, CAISO.

q We define one hour as the smallest indecomposable

market; we examine the centralized electricity market,

sometimes called the pool market, in detail to bring

a clear understanding of its features

q In fact, the day–ahead market or DAM is a

collection of 24 separate commodity markets, one

for each hour of the next day

h – year h – month h – day h h + month

q We suppress h in our notation in our discussion

seller 1 ... seller i ... seller N s

MWh MWh MWh

buyer 1 ... buyer j ... buyer N b

and deploys auction mechanisms to determine the

prices/quantities bought and sold in each hour

q Sellers are generation entities and brokers/

q Buyers are consumers, brokers/marketers,

distribution entities and generation entities

offer curve β is ( p is ) the costs to the

The social welfare is defined as the total

social welfare total benefits total costs

market clearing MWh/h

500 MWh 7 $/MWh

total sales = 1,600 MW S = 9,300

quantity price revenues payments

bus 1 j 0.1 bus 2

300 600 MWh/h 200 300 MWh/h

200 600 MWh/h 200 400 MWh/h

quantity price revenues payments

of the market as a whole but it fails to provide

insights into the performance of each individual

q We define two key components of social welfare

q The total consumer surplus is

actual flow: 1,000 MW

difference is the congestion rents = 1,600

bus 1 IGO bus 2

q The net injections at buses 1 and 2 need to be

modified to drive the network to feasibility

q The only choice of buyer b 3 is to bid sufficiently

high to induce seller s 3 to provide supply to meet

power flow at each node n

q The supply/demand equilibrium at each node n

may lead to different nodal market clearing prices

q Each seller or buyer faces the market clearing

price at the node where he is located

power line flow violates the line limit, i.e., the

corresponding inequality constraint becomes

q We call the transmission system congested if there

q  Market performance and congestion impact

q  Congestion and local market power

q  Congestion is a major obstacle to the functioning

q  A widely–used congestion management scheme

q  LMPs are purely short–term signals

q  The many lingering problems, such as

m  determination of realistic price signals

m  effective use of the financial transmission

q  The construction of a comprehensive analytic

q  We define one hour as the smallest indecomposable

q  In fact, the day–ahead market or DAM is a

q  We suppress h in our notation in our discussion

q  Sellers are generation entities and brokers/

q  Buyers are consumers, brokers/marketers,

q  We define two key components of social welfare

q  The total consumer surplus is

q  The net injections at buses 1 and 2 need to be

q  The only choice of buyer b 3 is to bid sufficiently

q  The supply/demand equilibrium at each node n

q  Each seller or buyer faces the market clearing

q  We call the transmission system congested if there

q  The net injections at buses 1 and 2 have to be

q  The only choice of buyer b 3 is to bid sufficiently

q  Since there is a willingness to pay by the buyer b 3,

q  The constrained dispatch changes the output of

q  Reduction in overall system reliability/security

q  Pollution due to the commitment/dispatch of older

m  market efficiency loss

q  The congestion rents are sometimes called the

q  For the constrained case we have

q  The market efficiency loss is, therefore,

q  For the constrained case we have

q  The market efficiency loss is

q  The blocks, whose prices are higher than the

q  The redispatch costs correspond to the additional

q  The need for additional ancillary services

q  The creation of situations that may lead to the

q  We evaluate the resulting surpluses for the

q  We evaluate the resulting surpluses for the

q  A seller changes his offer prices by varying the

q  We study the resulting variations of the producer

q  The simulations are performed on different