Power System Operations and

Control (EPE 609)

Lecture Set (3)

Power System Security and
Contingency Analysis
Prof. Muwaffaq I. Alomoush
Department of Electrical Power Engineering
Faculty of Engineering Technology
Yarmouk University

e-mail: [email protected]

First Semester 2023/2024

1
Power system security
• Power security: The ability of the power system to
maintain the flow of electricity from the generators to the
loads (customers) under disturbed conditions; it refers to
the ability of the system to withstand particular equipment
outages without loss of service, where all power system
components are operating within their limits.

• The power system must be operated at all times such that

the system will not be left in dangerous conditions (in a
secure state): all system parameters are operating as
desired with all voltages within their specified limits, no
power lines overloaded, all loads on line and being
provided with power.
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 2
Power system security: Cont.

 N-1 criterion is a widely used term in security studies, which means that
the system will remain in a secure operating state if any single outage event
or failure occurs.

 System security involves practices designed to keep the system operating

when components fail. For examples:
1. a generating unit may have to be taken off line because of auxiliary
equipment failure.
2. a transmission line may be damaged by a storm and taken out by
automatic relaying.

 Because the specific times at which initiating events that cause components
to fail are unpredictable, power system equipment is designed to be operated
within certain limits. Most pieces of equipment are protected by automatic
devices that can cause equipment to be switched out of the system if these
limits are violated.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 3
Power system security: Cont.

 If any events occur on a system that leaves it operating with limits

violated, the event may be followed by a series of further actions
that switch other equipment out of service. If this process of
cascading failures continues, the entire system or large parts of it
may completely collapse. This is usually referred to as a system
blackout.

 If any events occur on a system that leaves it operating with limits

violated, the event may be followed by a series of further actions
that switch other equipment out of service. If this process of
cascading failures continues, the entire system or large parts of it
may completely collapse. This is usually referred to as a system
blackout.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 4
Power system security: Cont.

 An example of the type of event sequence that can

cause a blackout start with a single line being opened
due to an insulation failure, the remaining transmission
circuit will take up the flow that was flowing on the
now-opened line. If one of the remaining lines is too
heavily loaded it may open due to relay action, and
this cause more load on the remaining lines. This type
of process is often termed a cascading outage.

 Most power system are operated such that any single

initial failure event will not leave other components
heavily overloaded, specially to avoid cascading
failures.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 5
Functions of Power System Security

System security can be divided into three major

functions that are carried out in an operation control
center:

(1) System monitoring,

(2) Contingency analysis, and

(3) Security-constrained optimal power flow.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 6
System Monitoring

System monitoring provides the operator of the power system

with up–to-date (real-time, or on–time) information on the
conditions on the power system.
Critical quantities are measured from a unit that supplying a
group of loads and transmitted to a control location. Such
system of measurement and data transmission called telemetry
systems.
The computer can check incoming information against pre-
stored limits and alarm the operators in the event of an overload
or out of limit voltage.
The critical information that can be measured (telemetered) are
voltage, currents, power flows, frequency, outputs of generating
units, and transformer tap positions.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 7
System Monitoring: Cont.

Digital computers are usually installed in operation control

centers to gather the telemetered data, process them and
display them on display monitors, and the computer can check
incoming information against pre-stored limits and alarm the
operators in the event of an overload or out of limit voltage.
Such systems are usually combined with supervisor control
systems that allow operators to control circuit breakers and
disconnect switches and transformer taps remotely. Together
these systems are often referred to as Supervisory Control
And Data Acquisition (SCADA) system.
The SCADA system allows a few operators to monitor the
generation and high voltage transmission systems and to take
action to correct line over loads or out-of-limit voltages.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 8
Contingency Analysis (Detection of Network Problems)

 The results of contingency analysis allow system to

be operated defensively (modeling possible
systems troubles before they arise). Transmission
line outage and generation outages result in
changes in flows and voltages on transmission
network.
 Many modern operations computers are equipped
with contingency analysis programs. These
programs are based on a model of power system
and are used to study outage events and alarm the
operators to any potential over loads or out of limit
voltages

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 9
 the procedure used for contingency analysis.

10

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Security Constrained Optimal Power Flow

 The third major security function is security-

constrained optimal power flow (OPF).

 In this function, a contingency analysis is combined

with an optimal power flow which seeks to make
changes to the optimal dispatch of generation, as
well as other adjustment, so that when security
analysis is run, no contingencies result in violations.

 To show how this can be done, we shall divide the

power system into four operating states, as follows:

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 11
1. Optimal dispatch : This is the state that the power system is in
prior to any contingency. It is optimal with
respect to economic operation, but it may not
be secure.

2. Post Contingency : The state of the power system after contingency

has occurred. It is assumed here that this
condition has a security violation such as bus
voltages and line beyond its flow limit.

3. Secure Dispatch : The state of the system with no contingency

outages, but with corrections to the operating
parameters to account for security violations.

4. Secure Post Contingency : The state of the system when the contingency
applied to the base- operating parameter
condition–with corrections.
Let us illustrate the above states with an example. Suppose that a simple
power system consisting of two generators supplying a load through
double circuit line as shown below in Figure 2 (ignore losses).
12
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Optimal Dispatch
The optimum dispatch for the system has shown that
there are 𝟓𝟎𝟎𝐌𝐖 from unit 1 and 𝟕𝟎𝟎𝐌𝐖 from unit 2.
Each circuit of the double circuit line can carry a
maximum of 400𝑀𝑊.
Cheap Unit Expensive Unit
Limit: 500 MW Limit: 1000 MW

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 13
Post Contingency State
If one of the two circuit making up the transmission line
has been opened because of a failure, as shown:

There is an over load on the remaining circuit

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 14
Secure Dispatch
Now there is an overload on the remaining circuit: This
state is not acceptable; the condition is corrected by
lowering the output of the generating with unit 1 to 400 MW.
The secure dispatch is as shown
Cheap Unit Expensive Unit
Limit: 500 MW Limit: 1000 MW

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 15
Secure Post Contingency State.
Now if the same contingency analysis is done the post-
contingency condition is as shown
Cheap Unit Expensive Unit
Limit: 500 MW Limit: 1000 MW

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 16
 By adjusting the output of generating units 1 and 2 to
consider possibility of one line outage, we have prevented the
post contingency operating state from having an overload.
This is the essence of what is called “Security Corrections”.

 Programs which can make control adjustments to the base or

pre-contingency operation to prevent violations in the post-
contingency conditions are called “Security–Constrained
Optimal Power Flow” or SCOPF. These programs can take
into account many contingencies and calculate adjustment to
generate MW, generator voltages, transformer taps,
interchanges between interconnected networks, etc.

 Together, the functions of system monitoring, contingency

analysis, and corrective action analysis form comprise a very
complex set of tools that can aid in the secure operation of
power system.
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 17
 Enough generation must always be available to meet the
load demand.
 Generator units must be managed in such a way that
planned outages of units, as well as forced outages, should
not result in a shortage of generation.
 The installed generation capacity should be greater than the
maximum demand, and it has to meet specific security
criteria.
 If
a generator is forced out, the remaining generators on-line
must have enough excess capacity to make up for the loss.
This excess capacity is called spinning reserve.
A common criterion for operational security is this ability of
the system to withstand the outage of any one piece of
equipment. The difficulty with this criterion is that checking
the effects of an outage on the transmission system
requires significant amount of computation.
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 18
Off-Line and On-Line Security Analysis
 Contingency screening is determining which contingencies pose the
biggest hazards (i.e, bad effects on system), which is achieved by
performing a fast approximate power flow solution for each
contingency case, for example 1-iteration of the decoupled load flow,
or the dc-load flow solutions if active line flow is the only concern.
 In real time, as the system conditions change, the contingencies,
which cause insecure operation, may also change, and would be
different from those predicted by off-line simulation studies.
 The selection of contingencies to be studied by the on-line security
analysis program should be based on the current operating
conditions, and should be adaptive and not a fixed list based upon
off-line studies.
 AC load flow equations are used for the evaluation of the effect of
transmission and generation outages on voltage quality.
 System-performance index and the dc load flow model are used to
rank contingencies with respect to real power flows.
 Full ac load flows must still be carried out, but only on the harmful
cases, by running the load flow for each case starting at the top of the
list and stopping when the cases do not give problems. 19
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
 Ensuring transmission security in real time requires significant on-
line computation of contingency scenarios.
 The computer automatically studies hundreds of possible
contingencies that could happen on the power system and
determines how well the system can withstand them.
 This requires running hundreds of power flow solutions and then
checking for line loading or voltage violations to alert the operator,
and it must be done within a few minutes for the information to be
useful.
 This is quite a computational burden in terms of both the number of
power flow solutions and the data sifting (‫ )غربلة‬needed for checking
violations. Thus much of the development of static security
assessment tools in the last decades has concentrated on making
this computation more efficient.
 Instead of finding full power flow solutions for all hundreds of
contingencies, more approximate but fast solutions are obtained to
determine which contingencies pose the biggest hazards. This
calculation is known as contingency screening.
20
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
 Most of the time, for well-planned systems, single
contingencies should not cause any limit violations,
and the main purpose of the contingency screening
is to isolate the very few problem cases from the
hundreds of non-threatening contingencies.
 In addition to running fast approximate solutions, the
screening must evaluate these solutions by a prober
severity index to determine which contingencies are
the worst.
 The severity indices must reflect line overloads and
voltage violations such that the contingencies can be
ranked according to their severity.
 Once this is done, only the worst contingencies are
further studied with accurate power flow solutions
and the resulting overloads and under voltages are
reported to the operator as alerting messages.
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 21
Static and Dynamic Security Assessment

Static Security Assessment

 The static security assessment program is designed to alert
the operator if a particular contingency would cause the
system to violate operational limits.
 The operator, if so alerted, must then decide whether to take
preventive action right away so that this contingency does
not make a problem or to take no action at the present time
but be ready to take corrective action if the contingency does
occur.
 In most cases of overloading or under voltages, the operator
usually has several minutes to take corrective action and so
the latter course is most often taken. This approach saves the
operator from making expensive changes in the operating
condition and the contingency may never occur.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 22
Static and Dynamic Security Assessment: Cont.

 The static security assessment checks for limit

violations after outages but it assumes that the system
reaches steady-state after these outages occur.
 Since outages are usually the result of an accidental
short-circuit which causes the protective systems to
isolate the short-circuited elements, the power system
may experience significant deviations in the voltages
and power flows during this disturbance.
 If the disturbance is severe enough, these swings may
cause generators to become unstable, by losing
synchronism, in which case there would be widespread
outages instead of the single outage expected.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 23
Dynamic Security Assessment

 The dynamic security assessment identifies those short-

circuits or contingencies that causes instabilities.
 Again, for a properly planned system no contingencies
should make the system unstable if operated within its
limits. However, in real time operation the power system
does end up in conditions that were not anticipated when
the planning was done.
 Thus, it is important to check whether contingencies can
make the system unstable. The problem is that the
stability calculations are even more time consuming than
the power flow calculations and the on-line checking of
stability for hundreds of possible contingencies is a
daunting task.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 24
Dynamic Security Assessment: cont.

 In those rare cases where the dynamic security assessment

detects instabilities, the operator, once alerted, needs to take
preventive action. This is because once the contingency
occurs, the onset of instability is very rapid and there is no
possibility of the operator taking manual corrective action after
the fact.
 In some cases, the operator may be able to alarm special
protection devices to shed load or generation that ensures
stability.
 More commonly, the preventive action available to the operator
is modification of the generating pattern. Since this invariably
increases the cost of operation, researchers are trying to find
methods to quickly calculate the minimum changes required to
maintain stability for a particular contingency.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 25
Distribution (or Shift or Sensitivity) Factors

 The problem of studying thousands of possible outages

becomes very difficult to solve if it is desired to present
the results quickly.
 One of the easiest ways to provide a quick calculation of
possible overloads is to use linear sensitivity factors.
 Sensitivity factors give the approximate change in line
flows for changes in generation or line outages and are
derived from the DC load flow.
 Sensitivityfactors basically come down to two types:
(1) Generation Shift Factors, or Generator Sensitivity
Factors, or Generator Outage Distribution Factors.
(2) Line Outage Distribution Factors.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 26
 Generator Outage Distribution Factors (GODFs) 𝒂𝒊𝒋,𝒌
The 𝑎𝑖𝑗,𝑘 factor then represents the sensitivity of the flow on line 𝑖𝑗 to
change in generation at bus 𝑘. The generation shift factors (𝑎𝑖𝑗,𝑘 ) have
the following definition:
𝜟𝑷𝒊𝒋
𝒂𝒊𝒋,𝒌 = or 𝜟𝑷𝒊𝒋 = 𝒂𝒊𝒋,𝒌 × 𝜟𝑷𝒌
𝜟𝑷𝒌
where
𝑖𝑗 Line index (line whose flow is affected)
𝑘 Bus index (bus at which generation or load has changed)
𝛥𝑃𝑖𝑗 Change in power flow on line 𝑖𝑗 when a change in generation ,
𝛥𝑃𝑘 , occurs at bus 𝑘
The new power flow on each line in the network for a change in
generation , 𝛥𝑃𝑘 , at bus 𝑘 is given by:
𝒏𝒆𝒘 𝒐𝒍𝒅
෡ 𝒊𝒋 = 𝑷𝒐𝒊𝒋 + 𝒂𝒊𝒋,𝒌 𝜟𝑷𝒌 𝐨𝐫 𝑷 = 𝑷 + 𝒂𝒊𝒋,𝒌 𝜟𝑷𝒌 (𝟏) (𝐨)
𝑷𝒊𝒋 = 𝑷𝒊𝒋 + 𝜟𝑷𝒊𝒋 𝐨𝐫 𝑷 𝒊𝒋 𝒊𝒋

It is assumed in this definition that the change in generation, 𝛥𝑃𝑘 , is

exactly compensated by an opposite change in generation at the
reference bus, and that all other generators remain fixed.
27
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
 Suppose one wanted to study the outage of a large
generation unit and it was assumed that all the generation
lost would be made up by the reference generation.
 If the generator in question was generating 𝑃𝑘𝑜 MW and it
was lost, we would represent 𝛥𝑃𝑘 as:

𝜟𝑷𝒌 = −𝑷𝒐𝒌 ; 𝑷𝒐𝒌 ≡ 𝑷𝒐𝑮𝒌

 The new power flow on each line in the network could be
calculated using a pre-calculated set of “ 𝑎 ” factors as
follows:
෡ 𝒊𝒋 = 𝑷𝒐𝒊𝒋 + 𝒂𝒊𝒋,𝒌 𝜟𝑷𝒌 for all lines 𝒊𝒋
𝑷
where
𝑃෠𝑖𝑗 = flow on line 𝑖𝑗 after the generator on bus 𝑘 fails
𝑃𝑖𝑗𝑜 = flow before the failure

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 28
Where 𝑎𝑖𝑗,𝑘 is given by:
(𝑿𝒊𝒌 −𝑿𝒋𝒌 )
𝒂𝒊𝒋,𝒌 =
𝒙𝒊𝒋
The matrix 𝑿 is given by (assuming bus 1 is the reference bus):
0 0 ... 0
0
𝑿=
⋮ [−𝑩′ ]−1
0
where [𝑩] is the susceptance matrix ([𝒀] = [𝑮] + 𝑗 [𝑩]), and
[𝑩′] is [𝑩] but with the colmumn and the row of reference bus in
[𝑩] are being removed.

The “outage flow” 𝑃෠𝑖𝑗 on each line can be compared to its limit
and those exceeding their limit flagged for alarming. This would
tell the operations personnel that the loss of the generator on bus
𝑘 would result in an overload on line 𝑖𝑗. 29
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
The effect of simultaneous changes in several generation buses
The effect of simultaneous changes in several generation buses can be
calculated using superposition. Suppose, for example, that the loss of the
generator on bus 𝒌 were compensated by governor action on machines
throughout the interconnected system. One frequently used method
assumes that the remaining generators pick up in proportion to the
maximum MW rating. Thus, the proportion of generation pick up from unit
𝑚 (𝑚 ≠ 𝑘), when generator on bus 𝑘 was lost, would be:
𝑷𝒎𝒂𝒙
𝒎
𝜸 𝒎,𝒌 = 𝑵𝒈
σ 𝒏 𝑷𝒎𝒂𝒙
𝒏
𝒎≠𝒌
𝒏≠𝒌
𝑃𝑛𝑚𝑎𝑥 maximum MW rating for generator 𝑛
𝛾𝑚,𝑘 proportionality factor for pick up on generating unit m when unit 𝑘
fails.
Then, to test for flow on line 𝑖𝑗, under the assumption that all the generators
in the interconnection participate in making up the loss , use the following:
(0)
𝑃෠𝑖𝑗 = 𝑃𝑖𝑗𝑜 + 𝑎𝑖𝑗,𝑘 𝛥𝑃𝑘 − σ𝑛≠𝑘 𝑎𝑖𝑗,𝑘 𝛾𝑛,𝑘 𝛥𝑃𝑘 ; 𝛥𝑃𝑘 = −𝑃𝐺𝑘
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 30
Example:
For a system that has four generators at buses 1, 2, 3, and 4, if G3
is outaged (failed), then:
𝑃1𝑚𝑎𝑥
𝛾1,3 = 𝑚𝑎𝑥
𝑃1 + 𝑃2𝑚𝑎𝑥 + 𝑃4𝑚𝑎𝑥
𝑃2𝑚𝑎𝑥
𝛾2,3 = 𝑚𝑎𝑥
𝑃1 + 𝑃2𝑚𝑎𝑥 + 𝑃4𝑚𝑎𝑥
𝑃4𝑚𝑎𝑥
𝛾4,3 = 𝑚𝑎𝑥
𝑃1 + 𝑃2𝑚𝑎𝑥 + 𝑃4𝑚𝑎𝑥
The changes in power productions of other units (G1 , G2 , G4)
are:
(0)
𝛥𝑃1 = −𝛾1,3 𝛥𝑃3 , 𝛥𝑃2 = −𝛾2,3 𝛥𝑃3 , 𝛥𝑃4 = −𝛾4,3 𝛥𝑃3 where 𝛥𝑃3 = −𝑃𝐺𝑘
𝑃෠12 = 𝑃12
𝑜
+ 𝑎12,3 𝛥𝑃3 + 𝑎12,1 𝛥𝑃1 + 𝑎12,2 𝛥𝑃2 +𝑎12,4 𝛥𝑃4
𝑃෠12 = 𝑃12
𝑜
+ 𝑎12,3 𝛥𝑃3 − (𝑎12,1 𝛾1,3 𝛥𝑃3 + 𝑎12,2 𝛾2,3 𝛥𝑃3 +𝑎12,4 𝛾4,3 𝛥𝑃3 )
𝑃෠23 = 𝑃23
𝑜
+ 𝑎23,3 𝛥𝑃3 − (𝑎23,1 𝛾1,3 𝛥𝑃3 + 𝑎23,2 𝛾2,3 𝛥𝑃3 + 𝑎23,4 𝛾4,3 𝛥𝑃3 )
31
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Line Outage Distribution Factors (LODFs)

By pre-calculating the line outage distribution factors, a very fast

procedure can be set up to test all lines in the network for
overload for the outage of a particular line.
The line outage distribution factors are used in a similar manner,
only they are used to the testing for overloads when
transmission circuits are lost. By definition, the line outage
distribution factor the following meaning:
𝜟𝑷𝒊𝒋
𝒅𝒊𝒋,𝒎𝒏 = or 𝜟𝑷𝒊𝒋 = 𝒅𝒊𝒋,𝒎𝒏 × 𝑷𝒐𝒎𝒏
𝑷𝒐𝒎𝒏
𝑑𝑖𝑗,𝑚𝑛 line outage distribution factor when monitoring line 𝑖𝑗 after an
outage on line 𝑚𝑛
𝛥𝑃𝑖𝑗 change in MW flow on line 𝑖𝑗
𝑜
𝑃𝑚𝑛 original flow on line 𝑚𝑛 before it was outaged (open)

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 32
Line Outage Distribution Factors (LODFs): Cont.

If one knows the power on line 𝑖𝑗 and line 𝑚𝑛 before outage,

the flow on line 𝑖𝑗 with line 𝑚𝑛 out can be determined using
“𝑑” factors:
𝑛𝑒𝑤 𝑜𝑙𝑑
𝑜 or 𝑃 (1) = 𝑃 (𝑜) + 𝑑
𝑃𝑖𝑗 = 𝑃𝑖𝑗 + 𝛥𝑃𝑖𝑗 or 𝑃෠𝑖𝑗 = 𝑃𝑖𝑗𝑜 + 𝑑𝑖𝑗,𝑚𝑛 𝑃𝑚𝑛 𝑃
(𝑜)
𝑖𝑗 𝑖𝑗 𝑖𝑗,𝑚𝑛 𝑚𝑛

𝑃𝑖𝑗𝑜 , 𝑃𝑚𝑛
𝑜 pre-outage flows on lines 𝑖𝑗 and 𝑚𝑛, respectively
𝑃෠𝑖𝑗 flow on line 𝑖𝑗with line 𝑚𝑛 out
where 𝑑𝑖𝑗,𝑛𝑚 is calculated as follows:
[(𝑋𝑖𝑛 −𝑋𝑖𝑚)− (𝑋𝑗𝑛 −𝑋𝑗𝑚 )] 𝑥𝑛𝑚
𝑑𝑖𝑗,𝑛𝑚 =
𝑥𝑖𝑗 [𝑥𝑛𝑚 −(𝑋𝑛𝑛 +𝑋𝑚𝑚 −2𝑋𝑛𝑚 )]

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 33
Contingency Ranking
 In contingency analysis, we initially quantify the severity of each outage,
then rank the contingencies using a scalar performance index (𝑷𝑰) that
measures various types of system stress.
 Following the calculation of the PIs, contingencies can be tested directly
beginning with the most severe and proceeding down the list to the least
sever ones.
 The ideal case of contingency selection is reached, if all actual critical
contingencies are ranked at the top of the contingency list. Severity is
reflected in voltage level degradation, and circuit overloads.
 Contingency Selection is to select the few contingencies that actually
affect the security of the power system and then analyze only these
cases in more details.
 The few contingencies at the top of the list have the most effect on
system security, and so should be analyzed in more details.
 There are criteria to decide how many case of this list should be chosen
for more details.
 The main problems facing ranking algorithms is masking effect. This
effect leads to a misclassification of contingencies and thereby to an
incorrect judgment of the critical contingency cases.
34
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Contingency Ranking: Cont.

 The analysis of the effects of hundreds or even

thousands of outages on line flows and bus voltages is
required for the real-time security analysis.
 This increases the demand on the speed, accuracy and
acceptability of the solution methods.
 Ideally, a full ac power flow should be solved for each
contingency, followed by a check of limit violations.
 Unfortunately, it is not practical in real life to solve using
full ac power flow. The fact that the vast majority of these
outages do not cause major shifts or violations leads to
various approximate methods.
 The purpose is to shorten the original list of branch
contingencies by eliminating the vast majority of
harmless cases.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 35
Performance Indices: PIMW , 𝐏𝐈𝐕 , 𝐏𝐈𝐕𝐐

 The system performance index is a measure that can be used to

evaluate the relative severity of a contingency.
 System performance indices are not unique and take on different forms
depending on the parameters that are of most importance to the
engineer (operator).
 The most common form of system performance indices gives a
measure of the deviation from rated values of system variables such as
line flows, bus power injection, etc.
 The PI is a scalar function of the network variables such as voltage
magnitude, real and reactive power flows. It has essentially two
aspects of functions:
(1) Differentiation between the critical outages from the non-critical
ones, and
(2) Prediction of relative severity of critical outages.
 There are different relations suggested for the PI. The selection of a
certain PI relation is based on the values we are interested in.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 36
Real Power Performance Index: PIMW

Real Power Performance Index (PIMW) gives a measure of line MW

overloads. It is given by the following equation:
𝑵𝑳
𝑷𝓵
𝑷𝑰𝑴𝑾 = ෍ 𝑾𝓵 ( 𝟐 )
𝒍𝒊𝒎
𝓵=𝟏 𝑷 𝓵
𝓵 An index refers to line number
𝑵𝑳 Number of lines of the system under study
𝑷𝓵 MW flow in line 𝓵
𝑷𝒍𝒊𝒎
𝓵 MW capacity (limit) of line 𝓵
𝑾𝓵 Weighting factor of active power limit violation
 The index 𝑷𝑰𝑴𝑾 is an index for quantifying the extent of line overloads.
 The dc distribution factor method, which is significantly faster than any
other technique, but less accurate, can be used to prescreen
contingencies for the more accurate ac load flow analysis.
 The main drawback of this analysis is it does not provide voltage, and
reactive power predictions.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 37
Voltage Performance Index: 𝑷𝑰𝑽

The Voltage Performance Index (𝑃𝐼𝑉 ) gives a measure of bus

voltage violations. It is chosen to quantify system deficiency
due to out-of-limit bus voltages. It is given by the following
equation:
𝑵𝑩 𝟐𝒏
𝑽𝒊 − 𝑽𝒍𝒊𝒎
𝒊
𝑷𝑰𝑽 = ෍ 𝑾𝑽𝒊
𝑽𝒍𝒊𝒎
𝒊
𝒊=𝟏
𝑖 An index refers to bus number
𝑁𝐵 Number of buses of the system under study
𝑉𝑖 Voltage magnitude at bus 𝑖
𝑉𝑖𝑙𝑖𝑚 Voltage magnitude limit at bus 𝑖
𝑛 Exponent of penalty function.
𝑊𝑉𝑖 Weighting factor of voltage magnitude at bus 𝑖
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 38
Voltage Performance Index: Cont.

Another form of PIV is given in an incremental form as

shown in the following equation:
𝑵𝑩 𝒔𝒑 𝟐
𝑾 𝑽𝒊 |𝑽𝒊 | − |𝑽𝒊 |
𝑷𝑰𝑽 = ෍
𝟐𝒏 𝜟𝑽𝑳𝒊𝒎𝒊
𝒊=𝟏
𝑽𝒊 Voltage at bus i.
𝒔𝒑
𝑽𝒊 Rated voltage at bus i.
𝜟𝑽𝑳𝒊𝒎
𝒊 Voltage deviation limit.
𝑾𝑽𝒊 Weighting factor of voltage limit violation.

 PIV measures the severity of the out-of-limit bus voltages,

 For a set of contingencies, PIV provides a direct means of
comparing the relative severity of the different outages
on the system voltage profile.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 39
Voltage-Reactive Power Performance Index: PIVQ

 Since the bus voltage levels depend mainly on the

reactive power flows, and therefore on the reactive power
production of the generators (and reactive power
production units, e.g., SC).
 In order to reflect the reactive power capability
constraints of the generators in the contingency selection
for voltage analysis, a generalized voltage-reactive power
performance index is used.
 The voltagereactive power performance Index (PIVQ)
gives a measure of bus voltage and reactive power
violations.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 40
Voltage-Reactive Power Performance Index: Cont.

The voltage-reactive power performance Index (PIVQ) is given by the

following equation:
𝑵𝑩 𝟐𝒏 𝑵𝑮 𝟐𝒏
𝑾𝑽𝒊 |𝑽𝒊 | − |𝑽𝒔𝒑
𝒊 | 𝑾𝑸𝒊 𝑸𝒊
𝑷𝑰𝑽𝑸 =෍ +෍
𝟐𝒏 𝜟𝑽𝑳𝒊𝒎
𝒊 𝟐𝒏 𝑸𝒎𝒂𝒙
𝒊
𝒊=𝟏 𝒊=𝟏
𝑸𝒊 Reactive power generated at bus i.
𝑸𝑴𝒂𝒙
𝒊 Rated reactive power generated at bus i.
𝑵𝑮 Number of reactive power generators.
n Exponent of penalty function.
𝑾𝑸𝒊 Weighting factor of reactive power limit violation.

 The reactive power weighting factors are set to zero if the effect of
the reactive power is not included.
 It is essential to perform an ac-load flow if the computation of the
contingency performance indices 𝑷𝑰𝑽 , 𝑷𝑰𝑸 ( or 𝑷𝑰𝑽𝑸 ) are needed.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 41
Voltage-Reactive Power Performance Index: Cont.

The 𝑃𝐼𝑉𝑄 can also be expressed in an incremental form as shown

by the following equation.
𝑵𝑩 𝟐𝒏 𝑵𝑩 𝟐𝒏
𝑽𝒊 − 𝑽𝒍𝒊𝒎
𝒊 𝑸𝒊 − 𝑸𝒍𝒊𝒎
𝒊
𝑷𝑰𝑽𝑸 = ෍ 𝑾𝑽𝒊 + ෍ 𝑾𝑸𝒊
𝑽𝒍𝒊𝒎
𝒊 𝑸𝒍𝒊𝒎
𝒊
𝒊=𝟏 𝓵=𝟏

𝑖 An index refers to bus number

𝑁𝐵 Number of buses of the system under study
𝑉𝑖 Voltage magnitude at bus 𝑖
𝑉𝑖𝑙𝑖𝑚 Voltage magnitude limit at bus 𝑖
𝑄𝑖 Reactive power injection at bus 𝑖
𝑄𝑖𝑙𝑖𝑚 Reactive power limit at bus 𝑖
𝑊𝑉𝑖 Weighting factor of voltage magnitude at bus 𝑖
𝑊𝑄𝑖 Weighting factor of reactive power at bus 𝑖

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 42
Contingency Selection

 DC or AC load flow can be used to calculate the performance indices for

every outage.
 Once the ranking of contingencies is done according to their severity, the
more severe cases may be analyzed further for more details with more
accuracy.
 The selection procedure then involves ordering the 𝑃𝐼 Table from largest
value to least. Selection criteria may take different forms. A selection
criterion of the form “ study the top 𝑁 cases in the ranked list of outages”
is attractive because it reduces the execution time if the number N is
kept low but it does not analyze outage cases ranked below 𝑁.
 In the real time environment, it is hardly feasible to perform exhaustive
contingency testing to determine how well the system, in its present state
can withstand the contingencies.
 At present, a contingency list to be studied by the security analysis
program is created based on the operator experience and off-line
simulation studies.

43
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Contingency Selection: Cont.

 In real time, as the system conditions change, the

contingencies, which cause insecure operation, may also
change, and would be different from those predicted by off-line
simulation studies.
 Non-linear ac load flow equations are used for the evaluation of
the effect of transmission and generation outages on a voltage
quality system-performance index and the dc load flow model to
rank contingencies with respect to real power flows. This does
not explicitly indicate whether contingency is going to give bus
voltage or circuit overload problems. Rather, it indicates the
severity of each contingency relative to the others.
 Full ac load flows must still be carried out –but only on the
“Problem” cases. This is easily determined by simply running
the load flow for each case starting at the top of the list and
stopping when the cases do not give problems.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 44
 Value of PI

45

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Summary of "Distribution Factors": Example

For the system shown, all values are given in p.u. and
bus. (1) is the reference bus.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 46
Base-Case Solution (values before outages (contingencies))
−𝑗15 𝑗10 𝑗5
[𝒀] = 𝑗10 −𝑗14 𝑗4
𝑗5 𝑗4 −𝑗9
−15 10 5
𝑩 = 10 −14 4
5 4 −9
14 −4
−[𝑩′] = ,
−4 9

0 0 0
0.0818 0.0364
−[𝑩′]−𝟏 = , [𝑿] = 0 0.0818 0.0364 , 𝜹 = 𝑿 𝑷
0.0364 0.1273
0 0.0364 0.1273
𝛿1 0 0 0 𝑃𝐺1 − 𝑃𝐷1 0.0
𝛿2 = 0 0.0818 0.0364 3 = 0.1
𝛿3 0 0.0364 0.1273 −4 0.4

Base-Case Line Flows (before any outage takes place):

(0) 𝛿 −𝛿 (0) 𝛿 −𝛿 (0) 𝛿 −𝛿
𝑃12 = 𝑃12 = 1 2 = −1 𝑃13 = 𝑃13 = 1 3 = 2 𝑃23 = 𝑃23 = 2 3 = 2
𝑥12 𝑥13 𝑥23

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 47
For outage of line 1-2:

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 48
For outage of line 1-3:

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 49
For outage of line 2-3:

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 50
For outage of Generator 2:

𝛿1 0 0 0 𝑃𝐺1 − 𝑃𝐷1 0 0 0 𝑃𝐺1 − 𝑃𝐷1

𝛿2 = 0 0.0818 0.0364 𝑃𝐺2 − 𝑃𝐷2 = 0 0.0818 0.0364 0
𝛿3 0 0.0364 0.1273 𝑃𝐺3 − 𝑃𝐷3 0 0.0364 0.1273 −4
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 51
For outage of Generator 3:

𝛿1 0 0 0 𝑃𝐺1 − 𝑃𝐷1 0 0 0 𝑃𝐺1 − 𝑃𝐷1

𝛿2 = 0 0.0818 0.0364 𝑃𝐺2 − 𝑃𝐷2 = 0 0.0818 0.0364 3
𝛿3 0 0.0364 0.1273 𝑃𝐺3 − 𝑃𝐷3 0 0.0364 0.1273 −5
52
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Summary of Results

Single Outage (Contingency)

Line Flow Base Case Line 1-2 Line 1-3 Line 2-3 G2 G3
𝑃12 -1 0 1 -3 1.4545 -0.6364
𝑃13 2 1 0 4 2.5455 2.6364
𝑃23 2 3 4 0 1.4545 2.3636

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 53
Line Outage Distribution Factors (𝒅𝒊𝒋,𝒏𝒎 )

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
The line outage distribution factor (𝑑𝑖𝑗,𝑛𝑚 ) represents the sensitivity of the
flow on line 𝑖𝑗 to outage of the line 𝑚𝑛 the line outage, which has the
following definition:
𝜟𝑷𝒊𝒋 (𝒐)
𝒅𝒊𝒋,𝒎𝒏 = (𝒐) or 𝜟𝑷𝒊𝒋 = 𝒅𝒊𝒋,𝒎𝒏 × 𝑷𝒎𝒏
𝑷𝒎𝒏
[(𝑿𝒊𝒏 −𝑿𝒊𝒎)− (𝑿𝒋𝒏 −𝑿𝒋𝒎 )] 𝒙𝒏𝒎
𝒅𝒊𝒋,𝒏𝒎 =
𝒙𝒊𝒋 [𝒙𝒏𝒎 −(𝑿𝒏𝒏 +𝑿𝒎𝒎 −𝟐𝑿𝒏𝒎 )]
If one knows the power on line 𝑖𝑗 and line 𝑚𝑛, the flow on line 𝑖𝑗 with line
𝑚𝑛 out can be determined using “𝑑” factors as follows:
(𝒏𝒆𝒘) (𝒐𝒍𝒅) ෡ 𝒊𝒋 = 𝑷(𝟎) + 𝜟𝒑𝒊𝒋 or ෡ 𝒊𝒋 = 𝑷(𝟎) + 𝒅𝒊𝒋,𝒎𝒏 𝑷(𝒐)
𝑷𝒊𝒋 = 𝑷𝒊𝒋 + 𝜟𝒑𝒊𝒋 or 𝑷 𝒊𝒋 𝑷 𝒊𝒋 𝒎𝒏
𝑑𝑖𝑗,𝑚𝑛 : Line outage distribution factor when monitoring line 𝑖𝑗 after an
outage on line 𝑚𝑛
𝛥𝑃𝑖𝑗 : Change in MW flow on line 𝑖𝑗
𝑜
𝑃𝑚𝑛 : Original flow on line 𝑚𝑛 before it was outaged (open)
(𝑜)
𝑃𝑖𝑗 , 𝑃𝑚𝑛
(𝑜) : Pre-outage flows on lines 𝑖𝑗 and 𝑚𝑛, respectively
𝑃෠𝑖𝑗 : : Flow on line 𝑖𝑗 with the line 𝑚𝑛 out 54
Calculations of Distribution Factors and Line Flows after Outages

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
0 0 0
[𝑿] = 0 0.0818 0.0364
0 0.0364 0.1273
(𝑋 𝑖𝑘 −𝑋 𝑗𝑘 )
 Generator Distribution Factors: 𝑎𝑖𝑗 ,𝑘 =
𝑥 𝑖𝑗

Generator Outaged or Changed

Line G1 G2 G3
12 (𝑋11 − 𝑋21 ) (𝑋12 − 𝑋22 ) (𝑋13 − 𝑋23 )
𝑎12,1 = 𝑎12,2 = 𝑎12,3 =
𝑥12 𝑥12 𝑥12
13 (𝑋11 − 𝑋31 ) (𝑋12 − 𝑋32 ) (𝑋13 − 𝑋33 )
𝑎13,1 = 𝑎13,2 = 𝑎13,3 =
𝑥13 𝑥13 𝑥13
23 (𝑋21 − 𝑋31 ) (𝑋22 − 𝑋32 ) (𝑋23 − 𝑋33 )
𝑎23,1 = 𝑎23,2 = 𝑎23,3 =
𝑥23 𝑥23 𝑥23

Generator Outaged or Changed

Line G1 G2 G3
12 𝑎12,1 = 0 𝑎12,2 = −0.8182 𝑎12,3 = −0.3636
13 𝑎13,1 = 0 𝑎13,2 = −0.1818 𝑎13,3 = −0.6364
23 𝑎23,1 = 0 𝑎23,2 = −0.1818 𝑎23,3 = −0.3636
55
 [(𝑋 𝑖𝑛 −𝑋 𝑖𝑚 )−(𝑋 𝑗𝑛 −𝑋 𝑗𝑚 )] 𝑥 𝑛𝑚
 Line Outage Distribution Factors: 𝑑𝑖𝑗 ,𝑛𝑚 =
𝑥 𝑖𝑗 [𝑥 𝑛𝑚 −(𝑋𝑛𝑛 +𝑋𝑚𝑚 −2𝑋𝑛𝑚 )]

Line Outaged (𝒎𝒏)

Line 𝒊𝒋 12 13 23
12 𝑑12,12 =  [(𝑋11 − 𝑋13 ) − (𝑋21 − 𝑋23 )] 𝑥13 [(𝑋12 − 𝑋13 ) − (𝑋22 − 𝑋23 )] 𝑥23
𝑑12,13 = 𝑑12,23 =
𝑥12 [𝑥13 − (𝑋11 + 𝑋33 − 2𝑋13 )] 𝑥12 [𝑥23 − (𝑋22 + 𝑋33 − 2𝑋23 )]

13
[(𝑋11 − 𝑋12 ) − (𝑋31 − 𝑋32 )] 𝑥12 𝑑13,13 =  [(𝑋12 − 𝑋13 ) − (𝑋32 − 𝑋33 )] 𝑥23
𝑑13,12 = 𝑑13,23 =
𝑥13 [𝑥12 − (𝑋11 + 𝑋22 − 2𝑋12 )] 𝑥13 [𝑥23 − (𝑋22 + 𝑋33 − 2𝑋23 )]

23
[(𝑋21 − 𝑋22 ) − (𝑋31 − 𝑋32 )] 𝑥12 [(𝑋21 − 𝑋23 ) − (𝑋31 − 𝑋33 )] 𝑥13 𝑑23,23 = 
𝑑23,12 = 𝑑23,13 =
𝑥23 [𝑥12 − (𝑋11 + 𝑋22 − 2𝑋12 )] 𝑥23 [𝑥13 − (𝑋11 + 𝑋33 − 2𝑋13 )]

Line Outaged (𝒎𝒏)

Line 𝒊𝒋 12 13 23
12 𝑑12,12 =  𝑑12,13 = 1 𝑑12,23 = −1
13 𝑑13,12 = 1 𝑑13,13 =  𝑑13,23 = 1
23 𝑑23,12 = −1 𝑑23,13 = 1 𝑑23,23 = 
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 56
 Single Outage (Contingency)
Line Flow Base Case Line 1-2 Line 1-3 Line 2-3 G2 G3
𝑃12 -1 0 1 -3 1.4545 -0.6364
𝑃13 2 1 0 4 2.5455 2.6364
𝑃23 2 3 4 0 1.4545 2.3636
Flows after line outages:
Outage of line 1-2 𝑃෠12 = 0 (0) (𝑜)
𝑃෠13 = 𝑃13 + 𝑑13,12 𝑃12
(0) (𝑜)
𝑃෠23 = 𝑃23 + 𝑑23,12 𝑃12
= 2 + (1)(2) = 1 = 2 + (−1)(−1) = 3
Outage of line 1-3 (0) (𝑜)
𝑃෠12 = 𝑃12 + 𝑑12,13 𝑃13 𝑃෠13 = 0 (0) (𝑜)
𝑃෠23 = 𝑃23 + 𝑑23,13 𝑃13
= −1 + (1)(2) = 1 = 2 + (1)(2) = 4
Outage of line 2-3 (0) (𝑜)
𝑃෠12 = 𝑃12 + 𝑑12,23 𝑃23
(0) (𝑜)
𝑃෠13 = 𝑃13 + 𝑑13,23 𝑃23 𝑃෠23 = 0

= −1 + (−1)(2) = −3 = 2 + (1)(2) = 4

Flows after generator outages:

Outage of G1 Not permitted (because bus is the refernce bus)
Outage of G2 (0) (0) (0)
𝑃෠12 = 𝑃12 + 𝑎12,2 𝛥𝑃2 𝑃෠13 = 𝑃13 + 𝑎13,2 𝛥𝑃2 𝑃෠23 = 𝑃23 + 𝑎23,2 𝛥𝑃2
= −1 + (−0.8182)(−3) = 2 + (−0.1818)(−3) = 2 + (−0.1818)(−3)
= 1.4545 = 2.5455 = 1.4545
Outage of G3 (0) (0) (0)
𝑃෠12 = 𝑃12 + 𝑎12,3 𝛥𝑃3 𝑃෠13 = 𝑃13 + 𝑎13,3 𝛥𝑃3 𝑃෠23 = 𝑃23 + 𝑎23,3 𝛥𝑃3
= −1 + (−0.3636)(−1) = 2 + (−0.6364)(−1) = 2 + (−0.3636)(−1)
= −0.6364 = 2.6364 = 2.3636 57
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
Example on "Power System Security": (DC Analysis)
Contingency Analysis: Line and Generator Outages

For the system shown, all values are given in p.u. and bus
(1) is the reference bus. Let line limits are:
𝑚𝑎𝑥 𝑚𝑎𝑥 𝑚𝑎𝑥
𝑃12 = 𝑃12 = 𝑃23 = 2 𝑝. 𝑢.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 58
Base-Case Line Flows:
(0) 𝛿 1 −𝛿 2 (0) 𝛿 1 −𝛿 3 (0) 𝛿 2 −𝛿 3
𝑃12 = 𝑃12 = = −1 𝑃13 = 𝑃13 = =2 𝑃23 = 𝑃23 = =2
𝑥 12 𝑥 13 𝑥 23

Real Power Performance Index (PI) of Contingencies (Outages):

𝑁𝐿=3
𝑃ℓ 2
𝑃𝐼𝑀𝑊 = ෍ 𝑊ℓ ( 𝑚𝑎𝑥 2
) ; 𝐴𝑠𝑠𝑢𝑚𝑒 𝑊ℓ = 1
ℓ=1
𝑃ℓ

Single Outage (Contingency)

Line Flow Base Case Line 1-2 Line 1-3 Line 2-3 G2 G3
𝑃12 -1 0 1 -3 1.4545 -0.6364
𝑃13 2 1 0 4 2.5455 2.6364
𝑃23 2 3 4 0 1.4545 2.3636

Performance Index (PI) 2.50 4.25 6.25 2.6768 3.2454

Contingency Rank 5 2 1 4 3

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 59
Example: Contingency Analysis of Multiple outages
For the system shown, use DC Load Flow Analysis to find the new line
flows using sensitivity factors for the following two cases:
(a)PG3 is increased by 0.2 p.u. and line 1 − 2 is outaged.
(b) 𝑃𝐷3 is increased by 0.5 p.u. and line 1 − 3 is outaged.

60
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
 Solution using Traditional DC Load Flow Analysis
To compare our results later obtained from sensitivity analysis, let’s
find the new line flows from the traditional DC load flow analysis:
(a)Find new line flows if 𝑷𝑮𝟑 is increased by 0.2 p.u. and line 𝟏 − 𝟐 is
outaged:
𝑷𝑮𝟑 = 𝟏 + ∆𝑷𝑮𝟑 = 𝟏. 𝟐 , Line 𝟏 − 𝟐: Open

61
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
 −𝑗5 0 𝑗5 −5 0 5
4 −4
𝒀= 0 −𝑗4 𝑗4 , 𝑩= 0 −4 4 , −𝑩′ =
−4 9
𝑗5 𝑗4 −𝑗9 5 4 −9
9 4
′ −1 1 9 4 20 20 0.45 0.20
− 𝑩 = = 4 4 =
4 9 − −4 −4 4 4 0.20 0.20
20 20
0 0 0
𝑿= 0 0.45 0.20
0 0.20 0.20
𝛿1 𝑃1 0 0 0 𝑃𝐺1 − 𝑃𝐷1 0.00 𝑟𝑎𝑑
𝛿2 = 𝑿 𝑃2 = 0 0.45 0.20 3 = 0.59 𝑟𝑎𝑑
𝛿3 𝑃3 0 0.20 0.20 −3.8 −0.16 𝑟𝑎𝑑

(2) (2) 𝛿1 −𝛿3 (2) 𝛿2 −𝛿3

𝑃12 = 0.0 outaged , 𝑃13 = = 0.8 𝑝. 𝑢. , 𝑃23 = = 3 𝑝. 𝑢.
𝑥13 𝑥23
(𝟐)
Where 𝑷𝒊𝒋 indicates the real power flow in the line 𝑖𝑗 after the two changes.
(2) (2)
Note that we should obtain 𝑃13 = 0.8 𝑝. 𝑢. and 𝑃23 = 3 𝑝. 𝑢. using
sensitivity factors.
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 62
(b) Find new line flows if 𝑷𝑫𝟑 is increased by 0.5 p.u. and line 𝟏 − 𝟑 is
outaged:
𝑷𝑫𝟑 = 𝟓 + ∆𝑷𝑫𝟑 = 𝟓. 𝟓 , Line 𝟏 − 𝟑: Open

63
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
 −𝑗10 𝑗10 0 −10 10 0
−14 4
𝒀 = 𝑗10 −𝑗14 𝑗4 , 𝑩 = 10 −14 4 , 𝑩′ =
4 −4
0 𝑗4 −𝑗4 0 4 −4
14 −4
−𝑩′ =
−4 4
4 4
1 4 4 40 40 0.10 0.10
− 𝑩′ −1
= = 4 14 =
14 4 − −4 −4 4 14 0.10 0.35
40 40

0 0 0
𝑿= 0 0.10 0.10
0 0.10 0.35
𝛿1 0 0 0 𝑃𝐺1 − 𝑃𝐷1 0.000 𝑟𝑎𝑑
𝛿2 = 0 0.10 0.10 3 = −0.150 𝑟𝑎𝑑
𝛿3 0 0.10 0.35 −4.5 −1.275 𝑟𝑎𝑑

2 𝛿1 −𝛿2 2 (2) 𝛿2 −𝛿3

𝑃12 = = 1.5 𝑝. 𝑢. , 𝑃13 = 0.0 𝑜𝑢𝑡𝑔𝑎𝑔𝑒𝑑 , 𝑃23 = = 4.5 𝑝. 𝑢.
𝑥12 𝑥23

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 64
Solution Using Sensitivity Factors
(a) If 𝑷𝑮𝟑 is increased by 0.2 p.u. and line 𝟏 − 𝟐 is outaged.
If one wants to have the sensitivity factors between line
𝒏𝒎 and the generator at bus 𝟑 when line 𝟏 − 𝟐 was
opened:
this is calculated by first assuming that the change in
generation on bus 3 (∆𝑃3 = +0.2) has direct effect on the line
𝑛𝑚 and indirect effect through its influence on the power flow
on line 1 − 2 (the line which was opened), which in turn
influences line 𝑛𝑚 when line 1 − 2 is out.
Let
1
𝑃13 : 𝑃13 after the first change (𝑃𝐺3 is increased by 0.2 p. u: ∆𝑃3 = +0.2)
2
𝑃13 : 𝑃13 after the second change (outage of line 1 − 2) (i.e., 𝑃13 after
the two changes)
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 65
Solution Using Sensitivity Factors: cont.
If 𝑷𝑮𝟑 is increased by 0.2 p.u. and line 𝟏 − 𝟐 is outaged.

After the first change (𝑷𝑮𝟑 𝐢𝐬 𝐢𝐧𝐜𝐫𝐞𝐚𝐬𝐞𝐝 𝐛𝐲 𝟎. 𝟐 𝐩. 𝐮: ∆𝑷𝟑 = +𝟎. 𝟐) :

1 0 1 (0)
𝑃13 = 𝑃13 + 𝑎13,3 𝛥𝑃3 ; 𝑃12 = 𝑃12 + 𝑎12,3 𝛥𝑃3
2 1 1 2 1 1
𝑃13 = 𝑃13 + 𝑑13,12 𝛥𝑃12 ; 𝑃23 = 𝑃23 + 𝑑23,12 𝛥𝑃12

After the second change (outage of line 𝟏 − 𝟐) :

2 1 1 (0) 0
𝑃13 = 𝑃13 + 𝑑13,12 ∆𝑃12 = 𝑃13 + 𝑎13,3 𝛥𝑃3 + 𝑑13,12 (𝑃12 + 𝑎12,3 𝛥𝑃3 )
2 1 1 (0) 0
𝑃23 = 𝑃23 + 𝑑23,12 ∆𝑃12 = 𝑃23 +𝑎23,3 𝛥𝑃3 + 𝑑23,12 (𝑃23 + 𝑎23,3 𝛥𝑃3 )

In general; In general, for two changes (increase/decrease in real

power at bus 𝒌 and outage of line 𝒊𝒋, The line flow in the line 𝒏𝒎
after the two changes is given by:

(𝟐) (𝟎) 𝟎
𝑷𝒏𝒎 = 𝑷𝒏𝒎 + 𝒂𝒏𝒎,𝒌 𝜟𝑷𝒌 + 𝒅𝒏𝒎,𝒊𝒋 [ 𝑷𝒊𝒋 + 𝒂𝒊𝒋,𝒌 𝜟𝑷𝒌 ]
Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 66
 1
If ∆𝑃12 is the change in line flow after the first change

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
(Change in 𝑃3 : 𝛥𝑃3 = +0.2)
1 (0)
∆𝑃12 = 𝑃12 + 𝑎12,3 𝛥𝑃3
1 (0)
𝑃13 = 𝑃13 + 𝑎13,3 𝛥𝑃3
2 1 1
𝑃13 = 𝑃13 + 𝑑13,12 ∆𝑃12

2 (0) 0
𝑃13 = 𝑃13 + 𝑎13,3 𝛥𝑃3 + 𝑑13,12 [ 𝑃12 + 𝑎12,3 𝛥𝑃3 ]

2
𝑃13 = 2 + −0.6364 +0.2 + 1 −1 + −0.3636 +0.2
= 2 − 0.12728 − 1.07272 = 2 − 1.2 = 0.8
or use the general equation:
2 (0) 0
𝑃𝑛𝑚 = 𝑃𝑛𝑚 + 𝑎𝑛𝑚,𝑘 𝛥𝑃𝑘 + 𝑑𝑛𝑚,𝑖𝑗 [ 𝑃𝑖𝑗 + 𝑎𝑖𝑗,𝑘 𝛥𝑃𝑘 ]
2 (0) 0
𝑃13 = 𝑃13 + 𝑎13,3 𝛥𝑃3 + 𝑑13,12 [𝑃12 + 𝑎12,3 𝛥𝑃3 ]
= 2 −0.6364 +0.2 + 1 −1 + −0.3636 +0.2 = 0.8
67
 2
Same is done for 𝑃23 :

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024
1 (0)
∆𝑃12 = 𝑃12 + 𝑎12,3 𝛥𝑃3
1 (0)
𝑃23 = 𝑃23 + 𝑎23,3 𝛥𝑃3
2 1 1
𝑃23 = 𝑃෠23 + 𝑑23,12 ∆𝑃෠12

2 (0) 0
𝑃23 = 𝑃23 + 𝑎23,3 𝛥𝑃3 + 𝑑23,12 [ 𝑃12 + 𝑎12,3 𝛥𝑃3 ]
= 2 + −0.3636 +0.2 + 1 −1 + −0.3636 +0.2
= 2 − 0.07272 + 1.07272 = 2 + 1 = 3
or use the general equation:
(2) (0) 0
𝑃𝑛𝑚 = 𝑃𝑛𝑚 + 𝑎𝑛𝑚,𝑘 𝛥𝑃𝑘 + 𝑑𝑛𝑚,𝑖𝑗 [ 𝑃𝑖𝑗 + 𝑎𝑖𝑗,𝑘 𝛥𝑃𝑘 ]
(2) (0) 0
𝑃23 = 𝑃23 + 𝑎23,3 𝛥𝑃3 + 𝑑23,12 [𝑃12 + 𝑎23,3 𝛥𝑃3 ]
= 2 + −0.3636 +0.2 + 1 −1 + −0.3636 +0.2 =3
68
Summary of results:
If 𝑷𝑮𝟑 is increased by 0.2 p.u. and line 𝟏 − 𝟐 is outaged.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 69
(b)Find new line flows using sensitivity factors if 𝑷𝑫𝟑 is
increased by 0.5 p.u. and line 𝟏 − 𝟑 is outaged.

𝜟𝑷𝟑 = −𝟎. 𝟓 𝒑. 𝒖. (same as decrease in 𝑷𝑮𝟑 )

(2) (0) 0
𝑃𝑛𝑚 = 𝑃𝑛𝑚 + 𝑎𝑛𝑚,𝑘 𝛥𝑃𝑘 + 𝑑𝑛𝑚,𝑖𝑗 [ 𝑃𝑖𝑗 + 𝑎𝑖𝑗,𝑘 𝛥𝑃𝑘 ]

(2) (0) 0
𝑃12 = 𝑃12 + 𝑎12,3 𝛥𝑃3 + 𝑑12,13 [𝑃13 + 𝑎13,3 𝛥𝑃3 ]
= −1 + −0.3636 −0.5 + 1 2 + −0.6364 −0.5
= −1 + 0.1818 + 2.3182 = −1 + 2.5 = 1.5

(2) (0) 0
𝑃23 = 𝑃23 + 𝑎23,3 𝛥𝑃3 + 𝑑23,13 [ 𝑃13 + 𝑎13,3 𝛥𝑃3 ]
= 2 + −0.3636 −0.5 + 1 2 + −0.6364 −0.5
= 2 + 0.1818 + 2.3182 = 2 + 2.5 = 4.5

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 70
Summary of results:
If 𝑷𝑫𝟑 is increased by 0.5 p.u. and line 𝟏 − 𝟑 is outaged.

Prof. Muwaffaq I. Alomoush, Yarmouk University, Power Engineering Department, First Semester 2023/2024 71