Contingency Selection

Performance Index (PI)

Measure how much a particular outage might affect the PS Overload Performance Index :

PI

P !( P
l

flowl max l

)2n

If n is a large number
PI will be a small number if all flows are within limits PI will be a large number if one or more lines are overloaded

How to use this PI

Problem

Suppose n = 1


A table of PI values, one for each line can be calculated

Selection Procedure:

Order the PI table from largest value to least Lines corresponding to the top of the list are then the candidates for the short list pick Nc entries

Problems faced when n=1

PI does not snap from zero to infinity as the branch exceeds its limit It rises as a quadratic function A line just below its limit contributes to PI almost equal to one that is just over its limit PI may be large when many lines are loaded just below their limit Thus PIs ability to distinguish / detect bad cases is limited ordering PI values results in a list that is not at all representative of bad cases at the top

1P1Q Method


Way to perform an outage case selection Decoupled power flow is used Solution procedure is interrupted after one iteration - one P- calculation & one Q-V calculation PI can use large n value

1P1Q Contingency Selection Procedure

Advantage
a.

Sufficient information at the end of First iteration to give a reasonable PI Voltages can also be included in the PI

a.

PI with voltage magnitude

P PI ! ( P
l

flowl 2 n ) max l

( Ei (E

) 2m max

( Ei

(E

max
-

difference between voltage magnitude (solved at the end of 1P1Q procedure) & base case voltage Value set by utility engineers, indicating limit of a bus voltage from changing on one outage case

To complete security analysis



PI list is sorted so that largest PI appears at the top Security analysis can then start by executing full power flows with the case at the top of the list Then solve the second case and so on This continues until either
 

a fixed number of cases is solved or until a predetermined number of cases are solved which do not have any alarms

Concentric Relaxation


Another idea for security analysis Outage has only a limited geographical effect even though line is high-voltage line, heavily loaded

To realize any benefit from limited geographical effect of an outage

Power system

Affected part

Unaffected part

Power System Division



Layer zero: Buses at the end of the outaged line Layer 1: Buses that are one transmission line or transformer from layer zero Continue this until all the buses in the entire network are included

Choose some arbitrary number of layers All buses included in that layer and lower-numbered layers are solved as power flow with the outage in place Buses in higher-numbered layers are kept as constant voltage & phase angle (reference bus)

Layering of outage effects

Application
1) Solution of layers included becomes Final

Solution of that case and all overloads & voltage violations are determined from this power flow
2) Solution is used to form PI for that outage

Concentric Relaxation Procedure

Proposed by Zaborsky Trouble:  Requires more layers for circuits whose influence is felt further from the outage

Bounding


Brandwajn solves atleast one of the problems using concentric relaxation method using an adjustable region around the outage applied only to linear (DC) power flow Extended for AC network analysis

Analysis in Bounding Technique

Define Three Subsystems of PS:
N1 = subsystem immediately surrounding the outaged line N2 = External subsystem that we shall not solve in detail N3 = Set of boundry buses that separate N1 & N2

Layers used in bounding analysis

Fact behind Bounding method

We can make certain phase angle assumptions about spread across lines in N2, given injections in N1 & maximum phase angle appearing across any two buses in N3 (Refer appendix 11A line outage modeling
using injections)

Let

Transmission line in N2

with flow fopq

There is a max. amount that the flow on pq can shift

max (f pq

 smaller[( f pq

0 0 f pq ), ( f pq

 f pq )]

 f pq pUpperLimit

f pq p Lo erLimit

Translation into max. change in phase angle difference

f
pq

1 x pq
1 x pq

(U

Uq )

( f pq !

((U p  (U q )

( (U p  ( U q )

max

! (f

max pq x pq

From [2] theorem states that

(U p  (U q

(U i  (U j

I & j are any pair of buses in N3

i

largest smallest

in N3 in N3

| i j| provides an upper limit to max. change in angular spread across any circuit in N2
max (f pq x pq

(U i  (U j

Graphical interpretation of Bounding process

I fig. cannot go over limit II fig. could go over limit

max (f pq x pq represents the line point where circuit pq will go into overload

Any value right of dotted line represents an overload

Solid line

upper limit on

(f pq x pq

If solid line is left to the dotted line, then the circuit cannot go into overload If solid line is right to the dotted line, the circuit may be shifted in flow due to outage so as to violate a limit

All circuits in N2 are safe from overload if the value of ( U i  ( U j is less than the smallest value of max ( f pq x pq over all pairs pq, where pq corresponds to the buses at the ends of circuits in N2

If this condition fails, we have to expand N1, calculate a new (U i  (U j in N3 and rerun the test over the newly defined N2 region circuits When an N2 is found which passes the test, only region N1 need be studied

PROBLEM


Show how bounding technique works so that not all of the circuits in the system need be analyzed. Consider the six-bus system used previously with outage of transmission line 3-6

LINE DATA
From bus
1 1 1 2 2 2 2 3 3 4 5

To bus
2 4 5 3 4 5 6 5 6 5 6

R (pu)
0.10 0.05 0.08 0.05 0.05 0.10 0.07 0.12 0.02 0.20 0.10

X (pu)
0.20 0.20 0.30 0.25 0.10 0.30 0.20 0.26 0.10 0.40 0.30

BCAP (pu)
0.02 0.02 0.03 0.03 0.01 0.02 0.025 0.025 0.01 0.04 0.03

BUS DATA
Bus No.
1 2 3 4 5 6

Bus type
Swing Gen Gen Load Load Load

V
(pu) 1.05 1.05 1.07

Pgen

Pload

Qload
(pu MVAR)

(pu MW) (pu MW)

0.50 0.60 0.0 0.0 0.0

0.0 0.0 0.7 0.7 0.7

0.0 0.0 0.7 0.7 0.7

X matrix for six-bus system

Base case DC power flow

MW limits on transmission lines

Line 1-2 1-4 1-5 2-3 2-4 2-5 2-6 3-5 3-6 4-5 5-6 MW limit 30 50 40 20 40 20 30 20 60 20 20

100 MVA base

Line 1-2 1-4 1-5 2-3 2-4 2-5 2-6 3-5 3-6 4-5 5-6 MW limit
(pu)
f
Q pq

( pu )

max (f pq

x pq 0.20 0.20 0.30 0.25 0.10 0.30 0.20 0.26 0.40 0.30

max ( f pq x pq

0.30 0.50 0.40 0.20 0.40 0.20 0.30 0.20 0.60 0.20 0.20

0.253 0.416 0.331 0.018 0.325 0.162 0.248 0.169 0.449 0.041 0.003

0.047 0.084 0.069 0.182 0.075 0.038 0.052 0.031 0.159 0.197

0.0094 0.0168 0.0207 0.0455 0.0075 0.0114 0.0104 0.00806 0.0636 0.0591

Analysis


Analyze system as if N1 & N3 regions consist of only line 3-6 If bounding criteria is met, no other analysis need be done as it establish that no overloads exist anywhere in the system

If bounding criteria fails, expand the bounded region ie, it includes buses 2,3,5 & 6 ie, boundary of region N3 consists of buses 2 & 5 Check bounding criteria, if it satisfies, stop the analysis

H i , nm

( X in  X im ) x k ! x k  ( X nn  X mm  2 X nm )

H i , nm
- sensitivity factor

(U i ! Pnm

Pnm - original power flow over line nm (k) before it was dropped phase angle xk - reactance of line k X - element of X matrix of the system

H i , nm

X in x k ! x k  X nn

for m = ref. bus

H i , nm

 X im x k ! x k  mm

for n = ref. bus

( X 33  X 36 )x36 H 3,36 ! ! 0.12865 x36  ( X 33  X 66  2 X 36 )

H 6,36 !

x 36

( X 63  X 66 ) x 36 ! 0.11953  ( X 33  X 66  2 X 36 )

( U 3  ( U 6 ! 0 . 111437
Smallest value of

(U

 ( U q ! ( U 2  ( U 4 ! 0 . 0075

Criteria fails, Consider new region

Calculate

(U 2  (U 5 ! 0.003564
Bounding criteria is satisfied If d factors are used, all line flows are to be found out

Line outage distribution factor

line l=1 (1-2) l=2 (1-4) l=3 (1-5) l=4 (2-3) l=5 (2-4) l=6 (2-5) l=7 (2-6) l=8 (3-5) l=9 (3-6) l=10 (4-5) l=11 (5-6) K=1 (1-2) K=2 (1-4) 0.64 0.59 0.41 -0.1 -0.59 -0.19 -0.12 -0.12 0.01 0.01 0.11 0.36 -0.03 0.76 -0.06 -0.04 -0.04 0 -0.24 0.03 0.18 -0.17 0.33 0.21 0.2 -0.03 0.29 -0.18 0.16 0.22 0.51 -0.38 -0.62 0.13 0.12 0.23 0.15 0.14 -0.02 -0.39 -0.13 0.27 0.27 -0.03 0.24 -0.23 -0.17 0.64 0.14 0.36 0.6 0.15 -0.4 -0.02 0.42 -0.18 K=3 (1-5) 0.54 0.46 K=4 (2-3) -0.11 -0.03 0.15 K=5 (2-4) -0.5 0.61 -0.11 0.12 K=6 (2-5) -0.21 -0.06 0.27 0.23 0.3 K=7 (2-6) -0.12 -0.04 0.16 0.47 0.17 0.24 K=8 (3-5) -0.14 -0.04 0.18 -0.4 0.19 0.27 -0.20 K=9 (3-6) 0.01 0 -0.02 -0.53 -0.02 -0.03 0.58 0.47 K=10 (4-5) 0.01 -0.33 0.32 0.17 -0.67 0.31 0.20 0.19 -0.02 K=11 (5-6) 0.13 0.04 -0.17 0.13 -0.19 -0.26 0.44 0.42 0.56 -0.15

Power flow using d factors

Line 1-2 1-4 1-5 2-3 2-4 2-5 2-6 3-5 3-6 4-5 5-6 MW limit
(pu)
f
Q pq

3 f pq 6 out

( pu )

( pu )

0.30 0.50 0.40 0.20 0.40 0.20 0.30 0.20 0.60 0.20 0.20

0.253 0.416 0.331 0.018 0.325 0.162 0.248 0.169 0.449 0.041 0.003

0.257 0.416 0.322 -0.220 0.316 0.148 0.508 0.380 0.032 0.191

Overloads exist on lines 2-3, 2-6, 3-5 These lines are within bounded region N1 + N3

Reference
1.

Allen J.Wood and Bruce F.Woolenberg, Power generation, operation and control, John Wiley & sons Inc. Brandwajn, Efficient Bounding method for Linear Contingency Analysis, IEEE Transactions on Power Systems, Vol. 3, No. 1, February 1988, pp.38-43.

2.