Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 118 (2015) 1127 – 1136

International Conference on Sustainable Design, Engineering and Construction

Evaluation for Voltage Stability Indices in Power System Using

Artificial Neural Network
H.H. Goha,*, Q.S. Chuaa, S.W. Lee a, B.C. Koka, K.C. Goha, K.T.K. Teo b
a
Department of Electrical Engineering, Faculty of Electrical and Electronic Engineering,
Universiti Tun Hussein Onn Malaysia, 86400 Parit Raja, Batu Pahat, Johor, Malaysia .
b
Modelling, Simulation and Computing Laboratory, Level 3, Block C, S chool of Engineering and Information Technology,
Universiti Malaysia Sabah, Jalan UMS, 88400, Kota Kinabalu, Sabah, Malaysia

Abstract

At present, the evaluation of voltage stability assessment experiences sizeable anxiety in the safe operation of power systems.
This is due to the complications of a strain power system. With the snowballing of power demand by the consumers and also the
restricted amount of power sources, therefore, the system has to perform at its maximum proficiency. Consequently, the
noteworthy to discover the maximum ability boundary prior to voltage collapse should be undertaken. A preliminary warning can
be perceived to evade the interruption of power system’s capacity. This paper considered the implementation of real-time system
monitoring methods that able to provide a timely warning in the power system before the voltage collapse occurred. Numerous
types of line voltage stability indices (LVSI) are differentiated in this paper to resolve their effectuality to determine the weakest
lines for the power system. The line voltage stability indices are assessed using the IEEE 9-Bus and IEEE 14-Bus Systems to
validate their practicability. Besides that, this paper also introduced the implementation of real-time voltage stability monitoring
by using Artificial Neural Network (ANN). Results demonstrated that the calculated indices and the estimated indices by us ing
ANN are practically relevant in predicting the manifestation of voltage collapse in the system. Therefore, essential actions can be
taken by the operators in order to dodge voltage collapse incident from arising.
©©2015
2015TheThe Authors.
Authors. Published
Published by Elsevier
by Elsevier Ltd.
Ltd. This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of organizing committee of the International Conference on Sustainable Design, Engineering
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of organizing committee of the International Conference on Sustainable Design, Engineering and
and Construction
Construction 2015 2015.

Keywords: Artificial Neural Network (ANN); Line Voltage Stability Indices (LVSI); Maximum Loadability; Voltage Stability Analysis;

* Corresponding author. Tel.: +6016-741-6824.

E-mail address: hhgoh@uthm.edu.my (H.H. Goh)

1877-7058 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of organizing committee of the International Conference on Sustainable Design, Engineering and Construction 2015
doi:10.1016/j.proeng.2015.08.454
1128 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136

1. Introduction

Present-day, the power systems need to adapt to the new situation since the actual scene no longer exists as it used
to be. Due to the climate change throughout the world, it is expected to lead the electricity consuming demand to
operate closely to the numbers of generated electricity [1, 2]. Besides that, aggressive business conditions have
enforced electric utilit ies to fully make use of accessible resources. Moreover, the current power systems are
extremely loaded as compared to the past because of the arising demand, maximu m economic advantages and the
effectiveness of utilizing the available transmission capacity [3-5].
After examining the sequence of incidents that caused the major blackout in the year 2003, the reasons of the
blackouts were due to a shortage of reliable real-t ime data [6, 7]. The established decentralized way of operating
systems by Transmission System Operators (TSOs) where each TSO take cares of its own control area and limited
informat ion to exchange, resulted in insufficient and delay response towards contingencies. Therefore, a real-t ime
security assessment and control are needed to maintain the system security [7]. The significance of real-t ime data is
to allow the operators to carry out important and practically p reventive action to avoid cascading or else will lead the
system to incorrect or delayed correction actions and thus will give a chance of instability occurrence.
Vo ltage stability assessment and control are not considered as any new issue [8], but they have now attained
special attentions to maintain the stability of the transmission networks in order to avoid recurrence of major
blackouts as experienced by the particular countries. The power system can be classified in the voltage stability
region if it can maintain steady acceptable voltages at all buses in the system under normal operating conditions and
after being subjected to a disturbance [9, 10]. In order to be reliable, the power system must be stable at most of the
time. The research works on voltage stability can be break down into various approaches, but the estimat ion on the
power system’s distance towards voltage collapse can be very handy to the operators before they take any remedial
actions [11, 12]. The details on the distance towards voltage instability can be obtained by using Voltage Stability
Indices (VSI) [13].
Vo ltage stability analysis is still widely being imp lemented in the industries by calculating the P -V and Q-V
curves at selected load buses [14]. Co mmon ly these curves are created by a large nu mber of load flows using
conventional methods and models. However, these methods are time consuming and do not provide sufficient
practical information towards the stability problems [15].
Various causes are recited in [16] as the commencement of the power systems outages. Most of all, power
systems outages can be categorized into two types; unpreventable event and preventable event. Some of the p ower
systems outages can be classified as the unpreventable events from the system operator. During the unpreventable
event took place; the system operator cannot control the damaged level happening in the power system.
In the mean while, in several cases, the power system outages can be prevented with the utilizat ion of a sufficient
system protection and situational awareness. If the power system is not equipped with the suitable p rotection for the
system, then the power system is prone to critical operatio nal situations and leads to instabilit ies. Hence, voltage
instability is one of the significant problems in causing the power outages.
Basically, voltage stability indices can be classified into Jacobian matrix – based voltage stability ind ices and
system variable – based voltage stability indices. In this paper, the motive for focusing on the system variable –
based voltage stability indices is because it requires fewer amounts of co mputing time. Besides that, it can precisely
verify the weak bus or lines in the power systems. Therefore, the scope has been narrowed into focusing on system
variable based voltage stability indices.
In addition, two test-power system cases are utilized in this paper, wh ich is IEEE 9-Bus test case and IEEE 14-
Bus test case. IEEE 9-Bus test case represents a portion of the Western System Coordinating Council (WSCC) 3 -
Machines 9-Bus system. Fundamentally, this IEEE 9-Bus test case consists of three generators, nine buses and three
loads. The IEEE 14-Bus test case actually represents a part of the A merican Electric Power System wh ich is situated
in the Midwestern US. Fundamentally, this system consists of two generators, three units of synchronous
condensers, 14 buses and 11 loads.
The rest of this paper was organized as follows. A background study on the overview of definit ions of voltage
stability and voltage stability indices will be discussed in section 2. The implementation of the voltage stability
indices and the details of the simulat ion cases will be talked about in section 3. Results and discussion will be
comprised in section 4. Finally, this paper is concluded in section 5.
 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136 1129

2. Background of study

2.1. Definitions of voltage stability

Vo ltage stability can be interpreted as the potential of the power system to sustain steady voltages at all buses in
the system after being vulnerable to a disturbance from a given initial operating condition [9]. Besides that, voltage
stability is the consequence on the ability of the power system to maintain o r restore the equilibriu m between the
load demand and load supply [9, 17]. A system is treated as voltage instable if at least one bus in the system
experiencing voltage magnitude decreases once the reactive power injection is increasing [18]. In addition, voltage
stability can also be considered as load stability. If the power system lacks of the capability to transfer an infinite
amount of electrical power to the loads, hence voltage instability will be present. The main reason for contributing to
voltage instability is the inability of the power system to meet the requirements for reactive power in the extremely
stressed system keeping the desired voltages [19]. In order to restore the increasing demand of loads in the systems
will cause further voltage decrement [17]. Vo ltage collapse is the event when acco mpanied by voltage instability
leads to a blackout or abnormally low voltages in a significant part of the power system [9]
Vo ltage instability and collapse effectively extent a t ime range fro m seconds to one hour. The time length of a
distraction in a power sys tem, originating a possibility of a voltage instability problem, can be categorized into short
term and long term. The examp le of short-term or transient voltage instability can be found main ly caused by rotor
angle variance or loss of synchronism. In the meanwh ile, long term voltage instability problems main ly occurred in
heavily loaded systems where the electrical distance is huge between generator and the load.
Mainly, in order to analyze the voltage stability, it is often convenient to categorize the problem into small-
disturbance and large-disturbance voltage stability. Small-disturbance voltage stability refers to the system’s
capability to continue steady voltages when disposed to small perturbations such as incremental changes in system
load [9]. Generally, the analysis for small-disturbances is done in steady-state stability analysis. Steady-state
stability analysis is useful in obtaining the qualitative overview of the system such as how stress the system is and
the system’s stability to the point of instability. Aside fro m, the large-disturbance voltage stability refers to the
system’s ability to maintain stable voltages followed by large d isturbances such as system fau lts, disappearance of
generation and loss of line. Large-d isturbance voltage stability can be analyzed by using non -linear time-do main
simulation in short term time frame and load flow analysis in the long term time frame [19].

2.2. Voltage Stability Indices

Vo ltage stability indices are very applicab le in retrieving the voltage stability of the power system. Vo ltage
stability indices are the scalar magnitudes that are being imp lemented to observe the changes of the parameters in
the system. Besides that, the indices are also used to quantify the distance of the particular operating point with the
point of voltage collapse [20]. These indices will be very handy to the operators before they started to imp lement the
prevention actions [12].
According to [21, 22], the authors mentioned that voltage stability indices particularly could be subdivided into
two parts, which are Jacobian matrix based voltage stability indices and system va riables based voltage stability
indices. Jacobian matrix based voltage stability ind ices are able to calculate the voltage collapse point or maximu m
load ability of the system and discover the voltage stability marg in. However, these indices required h igh
computational time and for this part icular reason, the Jacobian matrix based voltage stability indices are not
appropriate for online assessment. In the meanwhile, system variables based voltage stability indices required less
computational time. The reasons are due to the system variab le based voltage stability indices that used the elements
of the admittance matrix and some system variables such as bus voltages or power flow through the lines. With the
benefit of less computational time, system variab les based voltage stability indices are suitable to be imp lemented on
the online assessment and monitoring purposes. At the same time, system variables based voltage stability indices
cannot efficiently estimate the margin because their responsibilit ies are mo re to determine the critical lines and
buses.
1130 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136

The differences between Jacobian matrix based voltage stability indices and system variables based voltage
stability indices is provided in Table 1. The d ifferentiation is more likely based on the two aspects which were being
defined in [9]. The two aspects are proximity towards voltage collapse – (How close is the system to voltage
instability?) and mechanism of voltage instability – (How and why does instability occur?).

T able 1. Differentiation between Jacobian matrix based voltage stability indices and system variables based voltage stability indices.
Jacobian matrix based voltage stability indice s Syste m variable s base d voltage stability indice s
Require more amount of computing time Require less amount of computing time
Suitable for offline monitoring purpose Suitable for online monitoring purpose
Discover voltage stability margin Discover weak buses and lines
(Proximity towards voltage collapse) (Mechanism of voltage instability)

3. Methodologies

3.1. Power System Test Cases

Two different types of test cases are being utilized in this paper. They are IEEE 9-Bus test system and IEEE 14-
Bus test system. The main intention of increasing the number of buses for each test case is to determine the
feasibility of line voltage stability indices.
At first, IEEE 9-Bus test system network was used for the simulat ion. This IEEE 9-Bus system represents a
portion of the Western System Coordinating Council (WSCC) 3-Machines 9-Bus System. Basically, this 9-Bus
system contains of three generators, nine buses and three loads. The wiring d iagram for IEEE 9-Bus system is
illustrated in Fig. 1(a). The IEEE 14-Bus test case represents a portion of the American Electric Power System
which is located in the Midwestern US as of February, 1962. Basically, this system consists of five generators, 14
buses and 11 loads. The wiring diagram for IEEE 14-Bus system is illustrated in Fig. 1(b).

a b

Fig. 1. (a) IEEE 9-Bus test network; (b) IEEE 14-Bus test network.

3.2. Voltage stability indices formulation

Basically in this research, six different types of line voltage stability indices are being implemented. They are
Lmn index, LQP index, FVSI index, VCPI(Po wer), VCPI(Loss) and LCPI index. Main ly, the line stability indices
are formulated based on the power transmission concept in a single line as shown in Fig. 2.
 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136 1131

Fig. 2. 2-Bus system.

Where:
VS and VR are the sending end and receiving end voltages respectively
G S and G R are the phase angle at the sending and receiving buses
Z is the line impedance
R is the line resistance
X is the line reactance
T is the line impedance angle
PR is the active power at the receiving end
QR is the reactive power at the receiving end

3.2.1. Line stability index (Lmn)

This index was derived by [23] by using an overall system stability index based on the concept of power flow
through a single line as shown in Fig. 3. This index imp lemented a technique to reduce the system into a single line
equivalent network. The Lmn index is provided in Equation 1.

4 XQR
Lmn (1)
>VS sin(T  G )@
2

3.2.2. Line stability factor (LQP)

This line stability factor (LQP) is derived in [24] by the authors using the same notion in section 3.2.1. Hence,
LQP index is calculated as provided in Equation 2.

ª§ X ·§ X ·º
LQP = 4 «¨ 2 ¸¨ 2 PS 2 + QR ¸ » (2)
«¬© VS ¹© VS ¹ »¼

3.2.3. Fast voltage stability index (FVSI)

A novel fast voltage stability index (FVSI) was proposed by the authors in [25]. Th is index is being simplified
fro m a pre-developed voltage stability index referred to a line fro m the voltage quadratic equation at the sending end
of a representation of a two bus system. The formu lated index was tes ted on the IEEE 30-Bus reliability test system
in order to verify the performance of the proposed indicator. Hence, the equation for FVSI index can be defined in
Equation 3.

4Z 2 QR
FVSI = (3)
VS 2 X
1132 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136

3.2.4. Voltage collapse proximity indicators (VCPI)

The voltage collapse proximity indicators (VCPI) is proposed in [26] are based on the maximu m power
transferred through a line in the power network. The VCPI(Power) in Equation 4 and VCPI(Loss) in Equation 5 are
used to find the stability point for each line connection between two bus bars in an interconnected network. As long
as the indices remain less than one, then the system is considered to be stable.

PR QR
VCPI ( Power ) = or (4)
PR (max) QR (max)

Ploss Qloss
VCPI ( Loss) = or (5)
Ploss (max) Qloss (max)

3.2.5. Line collapse proximity indicator (VCPI)

The authors in [27] have proposed LCPI index based on the derivation of A BCD parameters in transmission line.
The forte of this index is by taking into consideration the line charg ing reactance during the derivation of equations.
The significant of line charg ing reactance may support voltage stability. The relat ionship between the parameters
can be explained in Equation 6.

4 A cos D  PR B cos E  QR B sin E 

LCPI = (6)
VS cos G 
2

Where:
A = 1  Z *Y / 2  { A‘D
B = Z { B‘E
C = Y * 1  Z *Y / 4
D=A

3.3. Feed forward back propagation neural network (FFBPNN)

A two-layer feed forward neural network with error back-propagation learning is one of the most commonly
implemented neural networks [28]. Neural networks are used to model co mplex relat ionships between inputs and
outputs or to find patterns in data. In this paper, ANN is used to find the patterns and equation fro m the data. It is a
structure (network) co mposed of a nu mber of interconn ected units. In this paper, the input data was as many as 100
data in which 30 data were used as testing data. The algorithm for this study is illustrated in Fig. 3.

Fig. 3. Simulation process of FFBPNN method.

 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136 1133

4. Results and discussion

In this paper, there are six line voltage stability indices and each of the indices consists of different number of
inputs but with one general output. The details of the inputs for the indices are summarised in the Table 2.

T able 2. T he details of the number of inputs for different line voltage stability indices.
Inde x Numbe r of inputs
Lmn 6
FVSI 4
LQP 4
VCPI (Power) 5
VCPI (Loss) 5
LCPI 9

4.1. Prediction for the most critical line in IEEE 9-Bus test system

The results indicated that line 9 – 6 is the most crit ical line at 465 seconds during the real-t ime simu lation.
Therefore, the prediction is mainly focused on the most critical line by using the artificial neural network model.

Fig. 4. T he comparison between the actual and the predicted line voltage stability indices based on line 9 – 6 at 465 seconds in IEEE 9-Bus test
system.

The blue bar in Fig. 4 rep resents the actual output for the line voltage stability indices while the red bar indicates
the predicted line voltage stability indices by using artificial neural network. It can be observed from the figure, the
trend of the pred icted output values is almost identical with the original calculated values. As it can be seen, the
errors found were not significant and there is only a minor difference between both data values in the figure which is
acceptable. Overall, the calcu lated and the predicted values are similar and this showed that artificial neural network
is also sufficient to be used for voltage stability monitoring purpose.
The details fo r the line stability indices based on line 9 – 6 at 465 seconds are presented in Tab le 3. The first
column of the table shows the type of the index. Besides that, column two showed the actual output values for the
indices based on the calculation. Consequently, column three showed the prediction output values from the art ificial
neural network model. Moreover, colu mn four and five showed the network error and error percentage
subsequently. In general, the predict ion outputs shown in Table 3 are relevant with the actual output and this once
again indicates artificial neural network is sufficient to be used for voltage stability asse ssment purpose.
1134 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136

T able 3. Details for line voltage stability indices based on line 9 – 6 at 465 seconds in IEEE 9-Bus test system.
Error Pe rce ntage
Inde x Actual O utput Pre diction O utput Ne twork Error
(%)
Lmn 0.994159 0.994160 -0.000001 0.000028
FVSI 0.922800 0.922799 0.000002 0.000166
LQ P 0.980285 0.980284 0.000001 0.000056
VCPI (Powe r) 1.010020 1.010000 0.000020 0.001970
VCPI (Loss) 0.674455 0.674453 0.000002 0.000319
LCPI 1.011878 1.011878 0.000000 0.000017

4.2. Prediction for the most critical line in IEEE 14-Bus test system

The line voltage stability results indicated that line 5 – 6 is the most critical line at 55 seconds. Therefore, the
prediction is based on the most critical line by applying the artificial neural network model.

Fig. 5. T he comparison between the actual and the predicted line voltage stability indices based on line 5 – 6 at 55 seconds in IEEE 14-Bus test
system.

The blue bar illustrated in Fig . 5 represents the actual output for the line voltage stability ind ices while the red bar
indicates the predicted line voltage stability indices by using artificial neural network. It can be observed from Fig.
5, the trend of the predicted output values is almost similar with the original calculated values. As it can be seen, the
errors found were not significant and there is only a minor difference between both data values in the figure which is
acceptable. Overall, the calcu lated and the predicted values are similar and this showed that artificial neural network
is also sufficient to be used for voltage stability monitoring purpose.

T able 4. Details for line voltage stability indices based on line 5 – 6 at 55 seconds in IEEE 14-Bus test system.
Error Pe rce ntage
Inde x Actual O utput Pre diction O utput Ne twork Error
(%)
Lmn 1.036085 1.036085 0.000000 0.000000
FVSI 0.991151 0.992118 -0.000967 0.097611
LQ P 1.081265 1.080954 0.000311 0.028734
VCPI (Powe r) 1.074979 1.074977 0.000002 0.000195
VCPI (Loss) 0.681614 0.681430 0.000184 0.027010
LCPI 1.081265 1.080954 0.000311 0.028734
 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136 1135

The summaries for the line stability indices based on line 5 – 6 at 55 seconds are presented in Table 4. The first
column of the table shows the type of the index. Besides that, column two showed the actual output values for the
indices based on the calculation. Consequently, column three showed the prediction output values from the art ificial
neural network model. Moreover, co lu mn four and five showed the network error and error percentage respectively.
In general, the prediction outputs shown in Table 4 are similar with the actual output and this once again indicates
artificial neural network is sufficient to be used for voltage stability assessment purpose.

5. Conclusion

In the nut shell, the application of art ificial neural network h ad been executed in this paper as well. The main
purpose for conducting this was to predict the most critical line in the power system based on the line voltage
stability indices assessment. In the meantime, this artificial intelligent technique is sufficient to be imp lemented in
the voltage stability monitoring purpose. Feed forward back propagation artificial neural network was used to
predict the line voltage stability indices for the most critical line in three different types of power system test
networks. The results showed that the implemented ANN model is indicative in forecasting the occurrence of system
collapse and hence suitable prevention action can be taken beforehand to avoid voltage collapse.

Acknowledgements

The authors would like to thank the Ministry of Science, Technology and In novation, Malaysia (MOSTI), and the
Office for Research, Innovation, Commercialization, Consultancy Management (ORICC), Universiti Tun Hussein
Onn Malaysia (UTHM) for financially supporting this research under the Science Fund grant No.S023.

References

[1] Parkpoom, S., G. Harrison, and J. Bialek. Climate change impacts on electricity demand. in Universities Power Engineering Conference,
2004. UPEC 2004. 39th International. 2004. IEEE.
[2] Belzer, D.B., M.J. Scott, and R.D. Sands, Climate change impacts on US commercial building energy consumption: an analysis using sample
survey data. Energy Sources, 1996. 18(2): p. 177-201.
[3] Wu, Y.-K., A novel algorithm for AT C calculations and applications in deregulated electricity markets. International Journal of Electrical
Power & Energy Systems, 2007. 29(10): p. 810-821.
[4] Rahi, O., et al., Power System Voltage Stability Assessment through Artificial Neural Network. Procedia Engineering, 2012. 30: p. 53-60.
[5] Savulescu, S.C., Real-time stability assessment in modern power system control centers. Vol. 42. 2009: John Wiley & Sons.
[6] Pourbeik, P., P.S. Kundur, and C.W. Taylor, The anatomy of a power grid blackout - Root causes and dynamics of recent major blackouts.
Power and Energy Magazine, IEEE, 2006. 4(5): p. 22-29.
[7] Bialek, J.W. Why has it happened again? Comparison between the UCTE blackout in 2006 and the blackouts of 2003. in Power T ech, 2007
IEEE Lausanne. 2007.
[8] Ajjarapu, V., Computational techniques for voltage stability assessment and control. 2007: Springer.
[9] Kundur, P., N.J. Balu, and M.G. Lauby, Power system stability and control. Vol. 7. 1994: McGraw-hill New York.
[10] Azmy, A.M. and I. Erlich. Impact of distributed generation on the stability of electrical power system. in Power Engineering Society General
Meeting, 2005. IEEE. 2005. IEEE.
[11] Goh, H., et al., Power stability monitoring based on voltage instability prediction approach through wide area system. Americ an Journal of
Applied Sciences, 2014. 11(5): p. 717-731.
[12] Goh, H., et al., Comparative study of different Kalman filter implementations in power system stability. American Journal of Applied
Sciences, 2014. 11(8): p. 1379-1390.
[13]Wen, J., et al., Optimal coordinated voltage control for power system voltage stability. Power Systems, IEEE Transactions on, 2004. 19(2): p.
1115-1122.
[14] Ajjarapu, V. and C. Christy, The continuation power flow: a tool for steady state voltage stability analysis. Power Systems, IEEE
T ransactions on, 1992. 7(1): p. 416-423.
[15] Balamourougan, V., T. Sidhu, and M. Sachdev. Technique for online prediction of voltage collapse. in Generation, Transmission and
Distribution, IEE Proceedings-. 2004. IET.
[16] McLinn, J., Major Power Outages in the US, and around the World. IEEE Reliability Society, 2009.
[17] Kundur, P., et al., Definition and classification of power system stability IEEE/CIGRE joint task force on stability terms an d definitions.
Power Systems, IEEE Transactions on, 2004. 19(3): p. 1387-1401.
[18] Kundur, P., J. Paserba, and S. Vitet. Overview on definition and classification of power system stability. in Quality and Security of Electric
Power Delivery Systems, 2003. CIGRE/PES 2003. CIGRE/IEEE PES International Symposium. 2003. IEEE.
1136 H.H. Goh et al. / Procedia Engineering 118 (2015) 1127 – 1136

[19] Chakrabarti, A. and S. Halder, Power System Analysis: Operation And Control 3Rd Ed. 2010: PHI Learning Pvt. Ltd.
[20] Lof, P.-A., G. Andersson, and D. Hill, Voltage stability indices for stressed power systems. Power Systems, IEEE Transactions on, 1993.
8(1): p. 326-335.
[21] Karbalaei, F., H. Soleymani, and S. Afsharnia. A comparison of voltage collapse proximity indicators. in IPEC, 2010 Conference
Proceedings. 2010. IEEE.
[22] Gong, Y., N. Schulz, and A. Guzman. Synchrophasor-based real-time voltage stability index. in Power Systems Conference and Exp osition,
2006. PSCE'06. 2006 IEEE PES. 2006. IEEE.
[23] Moghavvemi, M. and F. Omar, Technique for contingency monitoring and voltage collapse prediction. IEE Proceedings-Generation,
T ransmission and Distribution, 1998. 145(6): p. 634-640.
[24] Mohamed, A., G. Jasmon, and S. Yusoff, A static voltage collapse indicator using line stability factors. Journal of industrial technology,
1989. 7(1): p. 73-85.
[25] Musirin, I. and T .A. Rahman. Novel fast voltage stability index (FVSI) for voltage stability analysis in power transmission system. in
Research and Development, 2002. SCOReD 2002. Student Conference on. 2002. IEEE.
[26] Moghavvemi, M. and O. Faruque, Real-time contingency evaluation and ranking technique. Generation, Transmission and Distribution, IEE
Proceedings-, 1998. 145(5): p. 517-524.
[27] Tiwari, R., K.R. Niazi, and V. Gupta, Line collapse proximity index for prediction of voltage collapse in power systems. Inte rnational
Journal of Electrical Power & Energy Systems, 2012. 41(1): p. 105 -111.
[28] Fu, Y., Hybrid network reduction and artificial neural network based approaches in fast voltage stability analysis. 2000, The Hon g Kong
Polytechnic University.