Energy and Power Engineering, 2012, 4, 274-282

http://dx.doi.org/10.4236/epe.2012.44037 Published Online July 2012 (http://www.SciRP.org/journal/epe)

Evaluation of Operating Domains in

Power Systems
Mohamed A. El-Kady, Yasir A. Alturki, Mamdooh S. Al-Saud
SEC Chair in Power System Reliability and Security, King Saud University, Riyadh, KSA
Email: [email protected]

Received April 29, 2012; revised May 31, 2012; accepted June 17, 2012

ABSTRACT
This paper presents the results of a recent major industry-supported study with the aim to provide power system opera-
tors with more meaningful and effective means to quickly identify feasible operating boundaries as well as more flexi-
bility to select alternate operating scenarios. In this regard, the paper outlines the main theoretical basis and computa-
tional framework for the development of innovative computerized schemes capable of identifying and processing vari-
ous system integrity domains. The novel framework allows system operators to determine—in a fast and reliable man-
ner—the most favorable operating scenarios which maintain system security, reliability and operating performance
quality. For demonstration purposes, and without loss of generality, an emphasis is given to the dynamic system secu-
rity problem where the Transient Energy Function (TEF) method is used to define quantitative measures of the level
(degree) of system security for a given operating scenario. Nonetheless, the framework presented is applicable quite as
well to other system performance functions and criteria that may be considered. A demonstrative application is pre-
sented for a 9-bus benchmark system, widely used in the literature. In addition, a practical application is also presented
for the Saudi Electricity Company (SEC) power system where the operating security domain was evaluated in the oper-
ating parameter space spanned by two major interface flows in the system.

Keywords: Power Systems; Dynamic Security; Operating Parameters; Performance Optimization

1. Introduction reliability thresholds, voltage profile and VAR guidelines,

supply-demand balance requirements, etc. The complex-
Power system operators often make operating decisions ity of the overall problem and the multi-discipline nature
based on two basic objectives, namely: 1) ensure system of the research involved have resulted in division of the
integrity (security, reliability and quality); and 2) mini- research into two main areas, namely power system per-
mize operating costs. In real life situations, these two ob- formance quality and operational economy. The objec-
jectives are in conflict with each other and the power sys- tive in the operational economy area is to determine the
tem operator has, in many cases, to analyze and check—in optimum schedule of utility generating units that mini-
a trial and error manner—operating scenarios, which are mizes the total operation cost subject to constraints and
thought to be most economical, against the system secu- limitations on equipment and system control devices. On
rity, reliability and quality requirements and, conse- the other hand, the objective of system performance as-
quently, reject those operating scenarios which do not sessment is to ensure the system ability to withstand
meet the system integrity constraints [1,2]. In the case some unforeseen, but probable, disturbances with the
where dynamic system security is at stake, the automated minimum disruption of service or reduction of service
identification of the system dynamic security boundaries quality [8-10]. In recent years, the loading of transmis-
would be of great help to system operators as it would sion network and the amount of power transfer between
provide them with a practical way of identifying the dy- interconnected systems has increased to the point where
namic operating security modes of the system in a fast, power system security and reliability constraints started
accurate and reliable manner [3-7]. to influence the generation commitment and loading de-
Due to the critical importance of electric energy and cisions. This is because in real power systems, any re-
the rising cost of its production, power utilities around distribution of generator power output to minimize fuel
the world are compelled to minimize production cost and other operational costs would also influence the sys-
while, in the mean time, operating within acceptable sys- tem static and dynamic behavior when a contingency
tem integrity limits such as operating security constraints, occurs [11-13].

Copyright © 2012 SciRes. EPE

 M. A. EL-KADY ET AL. 275

In a recent major industry-supported study, extensive tricity Company (SEC) power system is also presented in
research and development work was conducted with the the paper.
aim to provide power system operators with more mean-
ingful and effective means to quickly identify feasible 2. Theoretical Formulation
operating boundaries as well as more flexibility to select
alternate operating scenarios. In this regard, this paper 2.1. System Integrity Domains
outlines the main theoretical basis and computational Consider Figure 1 which provides schematic representa-
framework for the development of innovative computer- tion to illustrate the concept of the system integrity do-
ized schemes capable of identifying and processing mains, where the level of system integrity (security, reli-
various system integrity domains and, therefore, allowing ability or quality) associated with different operating
system operators to determine—in a fast and reliable scenarios is measured in terms of how far the location of
manner—the most favorable operating scenarios which the operating point is from the system integrity bounda-
maintain system security, reliability and operating per- ries. The concepts and principles outlined in this section
formance quality. The methodology adopted in this work are applicable to both single and multi (interacting) sys-
includes the development of advanced, highly efficient tem integrity domains of either one or more operating
computerized algorithms for fast identification of the functions.
system integrity domains, including feasible dynamic In the general formulation of the problem on hand, the
operating security and reliability modes of power system. field of real numbers is denoted by  while the field of
One of the salient outcomes of this work is the develop- complex numbers is denoted by C. The vector space over
ment of a novel framework for identification and repre- , of n-tuples (z1, z2, ···, zn), zi   is denoted by n. We
sentation of the system integrity domains as well as classify the problem variables into two groups, namely
evaluation of the security, reliability and performance the dependent (state) variables xi  n, which are grouped
quality levels associated with different operating scenar- into the n-dimensional column vector x, and the inde-
ios. The system integrity constraints are in fact domains pendent (control) decision variables uk  m, which are
(boundaries) that surround all potentially acceptable op- grouped into the m-dimensional column vector u.
erating modes (scenarios) of the power system. In other We note that the states xi are those variables which are
words, these domains form the feasible operating region— of interest to the problem but can only be observed (for
in the parameter space spanned by various operating example, transmission line flows), while the controls uk
variables—within which the system can safely be oper- are those variables which can be adjusted (manipulated)
ated. The identification of the system feasible operating in practice (for example, plant output powers, trans-
domains can be performed using various available nu- former tap-settings, controlled bus voltages, var additions,
merical techniques including artificial intelligence (arti- etc.). The state and control variables are related through
ficial neural networks, genetic algorithms, fuzzy logic, as set of n equality constraints h (x, u) = 0 representing,
etc.), pattern recognition, non-linear optimization-based for example, the network flow equations.
methods for centering, tolerance and tuning, etc. For a
given operating scenario, the associated system integrity Operating Scenario 1(u)
level is measured by the position of the operating point  Secure w.r.t. Integrity Domain 1(u)
 Unreliable w.r.t. Integrity Domain 2(u)
with respect to the feasible operating domain boundary,
measured either as a simple “distance” (for example, the Operating Scenario 2(u)
Integrity
Euclidean norm) in the decision parameter space or using  Insecure w.r.t. Integrity Domain 1(u)
(Security)
 Reliable w.r.t. Integrity Domain 2(u)
an operating performance merit function (e.g., security Domain
1(u)
margin).
While the concepts and principles presented are gen-
Operating Decision Variable u2

eral, the work of this paper is confined to the interpreta- Integrity

tion of the system integrity as maintaining acceptable (Reliability)
levels of system security, reliability and operating per- Domain
2 (u)
formance quality. For demonstration purposes, an em-
phasis is given to the dynamic system security problem
where the Transient Energy Function (TEF) method is Operating Scenario 3(u)
used to define quantitative measures of the level (degree)  Secure w.r.t. Integrity Domain 1(u)
of system security for a given operating scenario [14-16].  Reliable w.r.t. Integrity Domain 2(u)
Nonetheless, the framework presented is applicable quite Operating Decision Variable u1
as well to other system performance functions that may
be considered. A practical application to the Saudi Elec- Figure 1. System integrity domains.

Copyright © 2012 SciRes. EPE

276 M. A. EL-KADY ET AL.

2.2. Operating Scenarios and Integrity Indices solely by the control variables. The reduced gradients can
be calculated using the method of Lagrange multipliers,
A set of potential operating scenarios s (u), s =1, 2, ···,
in which the partial derivatives of fℓ (x, u) with respect to
NS, where NS denotes number of scenarios, are defined to
both x and u are denoted by f x = [fℓ/x] and f u =
represent different groupings of the control variables
[fℓ/u], respectively, while the partial derivatives of the
associated with particular operating decisions that cause
equality constraints h (x, u) with respect to both x and u
the power system to reside in a particular mode of opera- are denoted by Hx = [hT/x]T and Hu = [hT/u]T, re-
tion. In other words, s (u) defines a particular setting spectively. The Lagrange multipliers are obtained by
{uks  m} of the system operating modes resulting from solving the set of linear equations H xT ℓ = f x and are
certain operator decisions. then used to calculate the reduced gradients as dfℓ/du =
Now, the problem on hand is formulated in terms of [ f u – H uT ℓ]. It is important to note that the reduced
one (or more) System Integrity Index fℓ (x, u) which gradients dfℓ/du represent a powerful means for measur-
measures the distance in the control parameter space, in ing the sensitivity of the system integrity index (for ex-
terms of a particular norm (for example, Euclidean norm), ample, security level) with respect to various operating
between certain system operating scenario {uks  m} decision variables. In other words, they provide invalu-
and the system integrity domain ℓ (u), ℓ = 1, 2, ···, NL, able information on how the system integrity is impacted
where NL denotes the number of integrity domains con- by various operating decisions made by the system op-
sidered in the problem. In some study cases, several sys- erators.
tem integrity domains are considered to either represent
multi-criteria such as system security, reliability, etc. or
3. Dynamic Security Domains
denote different potential contingencies and/or operating
schemes that are likely to occur during actual system 3.1. Background
operation. The system integrity domain ℓ (u) is the fea-
The importance of efficient and secure operation of the
sible space spanned by the problem variables within
power grid has always been acknowledged by the electric
which all operating scenarios are considered feasible
power utilities. In this regard, power system operation
(acceptable). Therefore, a security domain ℓ (u) is de-
aims, in principle, at maintaining reliable and secure sup-
fined, in general, by a number of inequality constraints gℓ
ply of electricity while minimizing the total cost of op-
(x, u)  0, which may include—as special case—simple
eration. In theory, there are two main objectives that
upper and lower bounds on the state and control variables
could be considered, namely the maximization of system
xL  x  xU & uL  u  uU. In the present work, the sys-
security and the minimization of total operating cost of
tem security index fℓ (x, u) will be defined as the mini- supplying energy. In practice, however, the security re-
mum of the distance functional norms between a feasible quirements are included as constraints rather than for-
operating scenario and all relevant inequality constraints mulating the problem as a security maximization man-
defining the integrity domain ℓ (u). In other words, date. The system security constraints are in fact bounda-
fℓ (x, u) = Minimum {over j} of  w j  g j  g j  , where ries that surround all possible operating modes (scenarios)
g j is the value of the j-th inequality constraint evaluated of the power system. In other words, these boundaries
at the operating scenario s (u), g j is the value of the form the feasible operating domain—in the parameter
boundary (threshold) value of the j-th inequality con- space spanned by various operating variables—within
straint at ℓ (u) and wj is a weighting factor associated which the system can safely be operated.
the jth element of fℓ (x, u). Obviously large values of For a given operating scenario, the associated security
fℓ (x, u) would indicate more secure operating modes level is measured by the “distance” (for example, the
Euclidean norm) of the operating point from the security
while small values of fℓ (x, u) would indicate less secure
domain boundary. Alternatively, the security level could
modes.
also be measured in terms of the energy margin value
(see the TEF formulation in the next sub-section) at the
2.3. Reduced Gradients
operating point.
In order to deal with the presence of the state variables in Security constraints could either be “static” or “dy-
the problem formulation in an efficient and proper namic”. The term “static security” means that all con-
mathematical manner, the concept of reduced gradients straints reflect steady-state quantities such as steady-state
(total derivatives) could be employed. The reduced gra- bus voltage violations and steady-state transmission line
dients of the system integrity index fℓ (x, u) with respect overloading. In conventional approaches, the dynamic
to the control variables u are denoted by dfℓ/du and, security constraints are either neglected or checked sub-
therefore, represent the sensitivity of the system integrity sequently and independent of the operating cost minimi-
index fℓ (x, u) with respect to u in the sub-space spanned zation scheme. In real power system operations, however,

Copyright © 2012 SciRes. EPE

 M. A. EL-KADY ET AL. 277

any re-distribution of generator powers to minimize fuel generators bus voltage angle, i is the generator speed,
costs (economic dispatch) would also influence the sys- Mi is the moment of inertia and
tem dynamic behavior (stability) when a contingency NG

occurs (for example, a fault which is cleared by a trans- MT   Mi (2)

i 1
mission line outage). Modeling complexity as well as the
multi-discipline nature of the research required have tra- The bus angles as well as generator internal angles i
ditionally prevented the inclusion of the dynamic system and rotor speeds expressed in the COI frame of reference,
security constraints in the overall optimization procedure, are given by
which would include both static and dynamic security i  i   0 ,  i   i   0 , i   i  0 (3)
constraints. While the static security constraints ensure
that the system integrity is maintained during the steady- Also
state operation, the dynamic security constraints ensure NG

that the system would maintain its dynamic robustness  M ii  0 (4)
i 1
during contingency situations. In addition to static and
dynamic security constraints, other boundaries are de- where the over ~ denotes quantities in the synchronous
fined, for example, in terms of equipment physical con- frame of reference.
straints. Converting loads to constant shunt admittances and
This section focuses on the use of the transient energy transforming rotor angles and speed to the COI reference,
function (TEF) method in order to define quantitative the swing equation of the NG generators can be written
measures of the level (degree) of system security for a in the following compact form
given operating scenario. Mi
f  V ,  ,    M i i  Pmi  PGi  PCOI  0 (5)
MT
3.2. Transient Energy Function (TEF)
where Pmi and PGi are the mechanical power input
Until recently, transient stability analysis has been per-
and generation power output respectively, V is the bus
formed by power utilities using exclusively numerical
voltage magnitude and
integration techniques to calculate the synchronous gen-
NG
erator response to a given disturbance. Transient Energy
Function (TEF) method provides an alternative approach
PCOI   Pmi  PGi
i 1
  (6)
to the conventional transient stability analysis. In contrast
to the time-domain approach, TEF method determines The equilibrium points of the system are the points
the system stability directly based on energy functions. where the right hand side of Equation (5) is equal to 0.
The TEF method not only avoids the time-consuming Among such equilibrium points, the Stable Equilibrium
step-by-step time-domain solutions, but it also provides a Point (SEP) and the controlling Unstable Equilibrium
quantitative measure of the degree of system stability Point (UEP) are of interest for the purpose of the tran-
known as the Energy Margin (EM). Positive values of sient stability analysis. The only difference between the
EM indicate a stable system while negative values indi- determination of the SEP and the UEP is the initial con-
cate unstable system. dition provided to the solution algorithm. For the SEP,
The EM and its sensitivities with system operating pa- the condition at fault clearing is used while, for the UEP,
rameters make the TEF method very attractive when sta- the so-called “ray point” is normally used unless for
bility limits must be calculated quickly. Although the stressed systems in which more robust techniques are
needed to solve for the UEP.
direct methods have a long development history, it was
Having solved for the SEP and the UEP, the transient
until recently thought to be impractical for large-scale
energy function V (, ω) is expressed as
power systems analysis. However, recent developments
have made the direct method quite suitable for stability NG NG NG i

analysis of large-scale power systems.

i 1 i 1
 
V  ,    0.5  M i i   Pmi i  is    PGi i (7)
2

i 1  s
The TEF can be formulated directly using the Center i

of Inertia (COI) frame of reference, which has the ad-

in which the three RHS terms represent the kinetic en-
vantage of removing the energy change associated with
ergy, position energy and magnetic and dissipation en-
the motion of the system COI. The COI is defined as
ergy of the system, respectively.
1 NG
1 NG The stability assessment is done by comparing two
0 
MT
 i M i and 0  M  i M i (1) values of the transient energy V. These are value of V
i 1 T i 1
computed at the end of disturbance (e.g., at fault clearing)
where NG is the number of generator busses, i is the Vcl and the critical value Vcr which is the potential energy

Copyright © 2012 SciRes. EPE

278 M. A. EL-KADY ET AL.

at the controlling UEP, for the particular disturbance un-

der investigation. Substituting for Vcr and Vcl in (7) and
using the concept of kinetic energy correction, the energy
margin can be obtained as
NG NG


EM  0.5  M eq eq    Pmi  iu   icl 
2

i 1 i 1
NG
(8)
iu
   cl PGi di
i
i 1

in which ω and  cl are calculated using either the

cl

step-by-step method or directly assuming constant accel-

eration. The dissipation energy term can be evaluated
only if the system trajectory is known. The fairly accu-
rate approximation, assuming the so-called linear angle
path, is used in the present analysis.

3.3. Problem Variables and Indices

The specific formulation for the identification of dy-
namic security domains involves the following:
Figure 2. WSCC 3-machine, 9-bus system (all impedances
1) The set of control (decision) variables involve, in
are in pu on a 100 MVA base) [17].
general, plant power outputs, interface-flows, controlled
bus voltages, under-load transformer tap settings, in-
jected VAR from controlled reactive power devices, and  pPd   1.0 pu 
u    
other decision variables on interest to power system op-  Vg 2
   1.025 pu 
erators.
2) The set of potential operating scenarios s (u), s = 1, One system integrity domain 1 (u) is considered in
2, ···, NS, involve various groupings of the control vari- this application defining a security criterion in terms of
ables associated with particular operating decisions under both the transient energy margin (EM) and the minimum
different identified contingency cases. incidental load bus voltage in the system (Vmin) in addi-
3) A system integrity (security) index fℓ (x, u), which tion to practical upper and lower bounds on the control
associated with a given system integrity (security) do- (decision) variables (u). In other words, the system in-
main ℓ (u), is used to measure the degree of system tegrity (security) domain 1 (u) is, in this application, a
security in terms of compliance (or violation) of a set of system security domain defined by four inequality con-
inequality constraints representing security boundaries, straints, namely
on of which is a minimum threshold on the energy mar- g11  x , u   0   EM  EM   0,
gin value.
EM  0  EM  0
4. Illustrative Application g12  x, u   0  Vmin  Vmin
   0,
As an illustrative application, consider the popular West-  =0.95  Vmin  0.95
Vmin
ern System Coordinated Council (WSCC) 3-machine,
g13  x, u   0   pPd  pPd   0,
9-bus system shown in Figure 2 [17], which is widely
used in the literature [18]. The detailed system and ma- pPd  0.8  pPd  0.8
chine data is listed in [17,18]. The system has been
and
simulated with a classical model for the generators and
one disturbance is considered where a three-phase fault g14  x , u   0   Vg 2  Vg2   0,
occurs near bus 7 at the end of line 5 - 7. The fault is Vg  0.95  Vg  Vg
cleared by opening line 5 - 7. In this demonstrative ap- 2 2 2

plication scenario, two control variables are considered where EM , Vmin  , pPd and Vg represent, respec-
2
representing, respectively, the overall system loading tively, the security limit (threshold) values associated
(per-unit) level (pPd) and the generator voltage at Bus #2 with the transient energy margin (EM), minimum bus
(Vg2). voltage (Vmin), system loading level (pPd) and voltage at
The base values for these two control variables are generator bus #2 ( Vg ). Therefore, the system integrity
2
[18] (security) index f1 (x, u) in this application is measured

Copyright © 2012 SciRes. EPE

 M. A. EL-KADY ET AL. 279

for any operating scenario s (u) as: Table 1. System security index at three operating scenarios
of the 9-bus system.

f1  x , u   Minimum g 1  EM  ,10  g 2  Vmin
 ,
Operating Scenario
g 1  pP , Vg2  g 1  Quantity
1 (u) 2 (u) 3 (u)
or u1 = pPd 0.9 1.0 1.1


f1  x, u   Minimum g  0 ,10  g  0.95 ,

11

11 u2 = Vg
2
1.0 1.05 1.05

g11  0.8 , 1.1  g11  EM 0.2285 0.0633 −0.2117

Vmin 0.9952 1.0050 0.9959

Observe that weighting factors of 10, 5 and 5 are as-
sumed for the second, third and forth elements of f1 (x, u), g 0

11 0.2285 0.0633 −0.2117
respectively, to account for different measurement units
10  g11  0.95 0.452 0.550 0.459
and/or relative importance (severity/concern) of each
component from the operating perspective. Figure 3 de- 5  g  0.8

11 0.5 1.0 1.5
picts the operating security domain of the sample power
5  1.1  g11 0.5 0.25 0.25
system in the parameter space spanned by the two control
parameters (pPd and Vg2). For demonstration purpose, f1 (x, u) 0.2285 0.0633 −0.2117
three operating scenarios (1 (u), 2 (u) and 3 (u)), as
shown in the figure, are examined in this application,
representing three operating points in the control pa- are secure while the third operating scenario (3 (u)) is
rameter space spanned by pPd and Vg , namely insecure with a negative energy margin of −0.2117.
2
The two secure operating scenarios have, however,

1  u   pPd  0.9 pu & Vg  1.0 pu  different levels of security as determined by the system
security index f1 (x, u). While the operating scenario 1
2


 2  u   pPd  1.0 pu & Vg  1.05 pu 2
 (u) enjoys a security index value of 0.2285, the operating
scenario 2 (u) has a security index value of only 0.0633
3  u    pPd  1.1 pu & Vg2  1.05 pu
(relatively closer to the security boundary).

Table 1 shows the values of the inequality constraints 5. Application to Saudi Power System
g1 (x, u) = {g11 (x, u), g12 (x, u), g13 (x, u)} defining the 5.1. System Model
system integrity (security) domain 1 (u) for the three
operating scenarios  (u) = {1 (u), 2 (u), 3 (u)} as In this section, a practical application is presented using
well as the resulting values of the system integrity (secu- the actual Saudi electricity system grid. The power sys-
rity) index f1 (x, u). It is clear from the results obtained tem used in the application is a dynamically reduced
that the first two operating scenarios (1 (u) and 2 (u)) model of the full-interconnected Saudi Electricity Com-
pany (SEC) power grid, which consists of two main re-
C5-Graph {9-B A&F Book} <pd> vs <v2> gions, namely the SEC-C (Central Region) and SEC-E
5
(Eastern Region). The two SEC systems are intercon-
nected through two 380 kV and one 230 kV double- cir-
1 cuit lines.
2(u) 3(u) In the original (unreduced) system model, the inter-
5
connected SEC bulk electricity system comprises 150
VM-B#2

generator buses, 637 load buses, a total of 1168 trans-

1(u)
mission lines and transformers. In order to prepare a
1 emv=0
vm =0.95 meaningful system model, which is suitable for the pre-
sent security assessment study, a coherency-based re-
duced network model derived from the original base-case
« I MO- P A »

5
is used, which comprises 119 buses (19 generators, 100
loads), 334 lines and 122 transformers.
9
0.8 0.9 1 1.1 1.2 This system model will be referred to as the 19-Gen-
PU-Load erator model. The nineteen generators are distributed as
Figure 3. Operating security domain of 9-bus system in two 11 in the SEC-C area and 8 in the SEC-E area as shown
operating parameter space. in Figure 4.

Copyright © 2012 SciRes. EPE

280 M. A. EL-KADY ET AL.

6 SEC CENTRAL 83 81 SEC EAST

PP8X
12
112 110
111
41 2 4 1 47 82 44
87 85 84 48
GAZLAN 45 46
40
8 7 15 14 5 23 BERI 51 103
JSWCC
34
NORTH AREA 49
QPP2 77
PP8B 61 62
38
75 74 63 89 92 68 67
PP8A 119
78 39 37 90 91
PP5
80 78
9
RIYADH AREA 50 93
79 114
86 107 106 105 104 94
113 19 73
115
PP9
DAMMAM
116 56
108 DAMAMAM AREA
88 26 30
98 97 96 95 99
SHEDGUM
35 109
21 71
100 29 28
43 PP4
66
QPP3 65 59
42 76 64
20 57 60
16
36
PP7B PP7A 54
10 55 70 69
17
QURAIAH
QASSIM AREA 117 102 101
27 18 58
13
LAYLA 53
33 32 31 24
118 25 11 22
72 SOUTH AREA
3 AL-KAHARJ AREA
52
FARAS

Figure 4. SEC 19-generators system model.

5.2. System Security Assessment the Operating security domain of SEC interconnected
system in the two operating parameters’ space. Four op-
In this practical application, two control variables are
erating scenarios, 1 (u), 2 (u), 3 (u) and 4 (u), are
considered representing, respectively, the East-to-Central
examined in the present application as shown in Figure 5,
interface flow level (Flow-E) and the Central-to-Qassim
representing four operating points in the control parame-
interface flow level (Flow-Q), both measured in MW.
ter space spanned, namely
The base values for these two control variables are
1  u   Flow-E  400 MW & Flow-Q  500 MW
 Flow-E  1100 MW 
u  
 Flow-Q   800 MW   2  u   Flow-E  1000 MW & Flow-Q  100 MW
One system integrity domain 1 (u) is considered in  3  u   Flow-E  1000 MW & Flow-Q  700 MW
this application defining a security criterion in terms of
both the transient energy margin (EM) and the minimum  4  u   Flow-E  700 MW & Flow-Q  1100 MW
incidental load bus voltage in the system (Vmin) in addi-
tion to lower bounds on the control (decision) variables The first two operating scenarios (1 (u) and 2 (u))
(u). In the present application, a three-phase fault close to are secure as they fall within the security domain de-
a major power plant (PP8A at bus #74 of Figure 4) is picted in Figure 5. The energy margin values associated
considered resulting in the complete outage of this power with these operating scenarios are 0.61 pu and 1.59 pu,
plant. The upper bounds on control variables (maximum respectively, which are both above the boundary limit of
interface flows) represent inactive constrains and would 0. On the other hand, the minimum occurring load volt-
not impact on the security domain of this application. In age associated with these operating scenarios are 1.01 pu
other words, the system integrity (security) domain 1 (u) and 0.93 pu, respectively, which are also above the
is, in this application, a system security domain defined boundary limit of 0.9 pu.
by four inequality constraints, namely EM ≥ 0 and Vmin ≥ The third and forth operating scenarios, 3 (u) and 4
0.95, Flow-E ≥ 0 and Flow-Q ≥ 0. (u) are insecure as they fall outside the operating security
While the stability-limit (zero value) is considered in domain of Figure 5. The operating scenario 3 (u), al-
this application to indicate the security domain, a slightly though enjoying a positive EM value of 0.62 pu, which is
higher value is normally considered in practice by power almost the same as the first scenario 3 (u), is insecure
utilities as an “operating security limit”. Figure 5 depicts because the minimum load voltage is only 0.81 pu which

Copyright © 2012 SciRes. EPE

 M. A. EL-KADY ET AL. 281

C5-Graph {SEC 119/19 R} <FE> vs <FQ> The Qassim region depends on the Central region for
4(u) emv=0 supplying a portion of its demand. Therefore, electricity
1200 vm =0.9
flows in normal cases from the Central region to Qassim
1000
region in the west. In other words, the Central region
3(u) sees Qassim as an added load to be supplied. Higher
Qassim demand levels would degrade both system stabil-
Flow-Q {MW}

800
ity and minimum occurring voltage magnitude. For an
600
East-to-Central flow level of 800 MW, an increase in the
Central-to-Qassim from 460 MW to 1150 MW would be
400
sufficient to bring the energy margin down from 1 to 0.

« I M O- P A »
On the other hand, for an East-to-Central flow level of
1(u) 800 MW, an increase in the Central-to-Qassim from 150
200
2(u) MW to 680 MW would be sufficient to bring the mini-
200 400 600 800 1000 1200 1400 1600 1800
Flow-E {MW}
mum voltage in the system down from 0.95 to 0.85 pu.

Figure 5. Operating security domain of SEC inter-con-

nected system in two operating parameter space.
6. Conclusions
This paper presents the results of a recent major industry-
is below the boundary value of 0.9 pu. The fourth oper- supported study which has addressed one of the impor-
ating scenario 4 (u), on the other hand violates both tant issues currently of concern to power system opera-
energy margin and minimum voltage limits where the tion and management, namely the identification of oper-
energy margin is 0.93 and the minimum load voltage is ating security domains within which the system can op-
0.78. erate safely in order to maintain reliable and secure sup-
ply of electricity to the consumers. The paper has out-
5.3. Discussion of Results lined the main theoretical basis and computational frame-
The electricity flow from the Eastern region (which has work for the development of innovative computerized
most of the generation resources) to the Central region in schemes capable of identifying and processing various
the interconnected Saudi power system is essential to system integrity domains and, therefore, allowing system
cover the relatively large demand in the Central region operators to determine—in a fast and reliable manner—
which suffers from generation shortage in relation to its the most favorable operating scenarios which maintain
required load. More East-to-Central flow would also re- system security, reliability and operating performance
lief some power plants in the Central region from having quality.
to operate at full capacity during high-demand coupled For demonstration purposes, and without loss of gen-
with adverse weather conditions during which increased erality, an emphasis was given in the paper to the dy-
possibility exists for major contingencies close to the namic system security problem where the Transient En-
massive load centers. In other words, avoiding extreme ergy Function (TEF) method is used to define quantita-
high output from power plants in the Central region (by tive measures of the level (degree) of system security for
increasing the support from Easter region) would im- a given operating scenario. Nonetheless, the framework
prove the stability of the system, where more East-to- presented is applicable quite as well to other system per-
Central flow would improve the energy margin for a formance functions and criteria that may be considered.
given Qassim demand value. For a Central-to-Qassim The applications presented for the Saudi electricity sys-
flow level of 800 MW, a drop in the East-to-Central from tem have revealed several important findings, which
1100 MW to 750 MW would be sufficient to bring the demonstrate the powerful features and usefulness of the
energy margin down from 1 to 0. On the other hand, be- present methodology. In this regard, the East-to-Central
cause of the long East-Central tie-lines as well as the flow has contradicting impacts on the overall system
relatively long “electrical” distances, which electricity security. While more East-to-Central flow would im-
has to go from Eastern region to reach the major load prove the energy margin (for a given Qassim demand
centers in the Central region, the incidental minimum level) due to relieving power plants in the Central region
load voltage would suffer as the East-to-Central flow from having to operate at full capacity, it would however
increases. For example, for a Central-to-Qassim flow cause degradation to the minimum occurring voltage
level of 400 MW, an increase in the East-to-Central from because of the relatively long distances, which electricity
600 MW to 1200 MW would be sufficient to bring the has to go from Eastern region to reach the major load
minimum occurring voltage in the system down from centers in the Central region. On the other hand, the Cen-
0.95 to 0.85 pu. tral region sees Qassim as an added load to be supplied

Copyright © 2012 SciRes. EPE

282 M. A. EL-KADY ET AL.

and, therefore, higher Qassim demand levels would de- 559-564. doi:10.1109/ICIINFS.2007.4579239
grade both system stability and minimum occurring vol- [9] M. A. El-Kady, B. A. Alaskar, A. M. Shaalan and B. M.
tage magnitude. Al-Shammri, “Composite Reliability and Quality As-
sessment of Interconnected Power Systems,” Interna-
tional Journal for Computation and Mathematic in Elec-
7. Acknowledgements trical and Electronic Engineering, Vol. 26, No. 1, 2007,
This work was supported by Saudi Electricity Company. Paper SSD05-PES-12.
[10] K. Tomsovic, D. E. Bakken, V. Venkatasubramanian and
A. Bose, “Designing the Next Generation of Real-Time
REFERENCES Control, Communication, and Computations for Large
[1] D. P. Popovic, D. M. Dobrijevic, N. A. Mijuskovic and D. Power Systems,” Proceeding of the IEEE, Vol. 93, No. 5,
J. Vlaisavljevic, “An Advanced Methodology for Steady- 2005, pp. 965-979. doi:10.1109/JPROC.2005.847249
State Security Assessment of Power Systems,” ETEP, [11] M. A. El-Kady, M. S. El-Sobki and N. K. Sinha, “Reli-
Vol. 1, No. 4, 2001, pp. 227-233. ability Evaluation for Optimally Operated Large Electric
doi:10.1002/etep.4450110403 Power Systems,” IEEE Transactions on Reliability, Vol.
[2] M. S. Owayedh and M. A. El-Kady, “Integration of R-35, No. 1, 1986, pp. 41-47.
Economy and Quality Requirements: Practical Applica- doi:10.1109/TR.1986.4335340
tion to Power System Operation,” Proceeding of the 7th [12] Z. Benguo, W. Dapeng, L. Lei, X. S. Niu, H. Z. Zang and
Saudi Engineering Conference, Riyadh, 26-28 November Y. S. Zhao, “Research of Incentive Revelation Mecha-
2007, Paper # SEC-013. nism in Power System Optimal Security Control,” Pro-
[3] J. Aghaei, H. A. Shayanfar and N. Amjady, “Joint Market ceeding of the 2006 International Conference on Power
Clearing in a Stochastic Framework Considering Power System Technology, Chongqing, 22-26 October 2006, 5 p.
System Security,” Applied Energy, Vol. 86, No. 9, 2009, [13] M. A. El-Kady, B. D. Bell, V. F. Carvalho, R. C. Burchett,
pp. 1675-1682. doi:10.1016/j.apenergy.2009.01.021 H. H. Happ and D. R. Vierath, “Assessment of Real-Time
[4] M. A. El-Kady, “Direct Analysis of Transient Stability Optimal Voltage Control,” IEEE Transactions on Power
for Large Power Systems,” Final Report, No. EL-4980, Systems, Vol. PWRS-1, 1986, pp. 98-107.
EPRI Project No. 2206-1, Iowa State University, Ontario doi:10.1109/TPWRS.1986.4334912
Hydro, 1986. [14] T. Athay, R. Podmore and S. Virmani, “A Practical Me-
[5] J. Srivani and K. S. Swarup, “Power System Static Secu- thod for Direct Analysis of Transient Energy Stability,”
rity Assessment and Evaluation Using External System IEEE Transactions on Power Apparatus and Systems,
Equivalents,” Electrical Power and Energy Systems, Vol. Vol. 98, No. 2, 1979, pp. 573-584.
30, 2008, pp. 83-92. doi:10.1016/j.ijepes.2007.06.008 doi:10.1109/TPAS.1979.319407
[6] Y. Zeng, P. Zhang, M. Wang, H. Jia, Y. Yu and S. T. Lee, [15] M. A. El-Kady, C. K. Tang, V. F. Carvalho, A. A. Fouad
“Development of a New Tool for Dynamic Security As- and V. Vittal, “Dynamic Security Assessment Utilizing
sessment Using Dynamic Security Region,” Proceeding the Transient Energy Function Method,” IEEE Transac-
of the 2006 International Conference on Power System tions on Power Apparatus and Systems, Vol. PWRS-1,
Technology, Chongqing, 22-26 October 2006, 5 p. No. 3, 1986, pp. 284-291.
doi:10.1109/ICPST.2006.321756 [16] M. A. El-Kady and R. W. D. Ganton, “Fast Decoupled
[7] M. A. El-Kady and E. A. Al-Ammar, “Framework for Transient Stability,” Canadian Electrical Engineering
Identification of Power System Operating Security Re- Journal, Vol. 14, No. 2, 1989, pp. 63-66.
gions,” Proceedings of 3rd International Conference on [17] R. Patel, T. S. Bhatti and D. P. Kothari, “MATLAB/
Network and System Security, Gold Coast, 19-21 October Simulink-Based Transient Stability Analysis of a Multi-
2009. doi:10.1109/NSS.2009.16 machine Power System,” International Journal of Elec-
[8] U. D. Annakkage and B. Jayasekara, “Ensuring Power trical Engineering Education, Vol. 39, No. 4, 2002, pp.
System Security against Contingencies,” Proceeding of 320-336.
the 2nd International Conference on Industrial and In- [18] P. M. Anderson and A. A. Fouad, “Power System Control
formation Systems, Sri Lanka, 8-11 August 2007, pp. and Stability,” Iowa State University Press, Ames, 1977.

Copyright © 2012 SciRes. EPE