Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

IEEE TRANSACTIONS ON SUSTAINABLE ENERGY 1

Transient Stability Augmentation of

PV/DFIG/SG-Based Hybrid Power System by
Nonlinear Control-Based Variable Resistive FCL
Md. Kamal Hossain, Student Member, IEEE, and Mohd. Hasan Ali, Senior Member, IEEE

Abstract—This paper proposes three nonlinear controllers such PV Photovoltaic.

as fuzzy logic controller (FLC), static nonlinear controller (SNC), RSC Rotor side converter.
and adaptive-network-based fuzzy inference system (ANFIS)- SG Synchronous generator.
based variable resistive-type fault current limiter (VR-FCL) to
augment the transient stability of a large-scale hybrid power sys- SNC Static nonlinear controller.
tem consisting of a doubly fed induction generator (DFIG)-based TPD Total power deviation.
wind farm, a photovoltaic (PV) plant, and a synchronous genera- VR-FCL Variable resistive-type fault current limiter.
tor (SG). Appropriate resistance generation of the VR-FCL during VSC Voltage source converter.
a grid fault to provide better transient stability is the main con- VSWT Variable speed wind turbine.
tribution of the work. The effectiveness of the proposed control
methods in improving the transient stability of the hybrid power ΔVPCC PCC voltage deviation.
network is verified by applying both balanced and unbalanced ΔωSG Speed deviation of the SG.
faults in one of the double circuit transmission lines connected to 1LG Single-line-to-ground fault.
the system. Simulation results show that the proposed FLC-, SNC-, 2LG Double-line-to-ground fault.
or ANFIS-based VR-FCL are effective in improving the transient 3LG Three-line-to-ground fault.
stability of the studied hybrid system. Moreover, all the proposed
methods exhibit almost similar performance. Therefore, any of the
methods can be chosen for the transient stability enhancement of B. Subscripts
the hybrid power system. ACeff AC-side effective.
Index Terms—Adaptive-network-based fuzzy inference system d, q d- and q-axis.
(ANFIS), doubly fed induction generator (DFIG), duty ratio (d), g Grid.
fuzzy logic controller (FLC), photovoltaic (PV), static nonlinear mes, ref Measured and reference.
controller (SNC), synchronous generator (SG), transient stability, s, r Stator and rotor windings.
variable resistive-type fault current limiter (VR-FCL).
w Wind.

N OMENCLATURE I. I NTRODUCTION
A. Abbreviations
ANFIS Adaptive-network-based fuzzy inference system.
ANN Artificial neural network.
A MONG the alternative renewable energy sources, the
wind energy generating system (WEGS) today is an
established source of renewable energy, with an annual increase
CB Circuit breaker. rate of 20% and the total worldwide installation capacity of
D Duty cycle. 238 000 MW at the end of 2011 [1]. DFIG, which is exploited as
DFIG Doubly fed induction generator. a VSWT generator, is employed extensively as wind generator
FCL Fault current limiters. due to its several merits, such as higher efficiency and indepen-
FLC Fuzzy logic controller. dent control of active and reactive powers by exploiting power
FRT Fault ride-through. electronic interfaces for better grid connection [2]. On the other
GSC Grid side converter. hand, sunlight being ubiquitous and the emerging technologies
IGBT Insulated-gate-bipolar-transistor. of thin film and crystalline solar cells render the PV as a pop-
MF Membership function. ular renewable energy source. The solar energy will attain the
MPPT Maximum power point tracker. important position among the renewable energy sources within
PCC Point of common coupling. 2040, fulfilling almost 28% of all world energy demand [3].
p.u. Per unit. In order to increase network reliability, renewable energy
sources, like wind generator, PV, etc., are intergraded with
Manuscript received January 26, 2015; revised May 15, 2015 and June 25,
2015; accepted July 26, 2015. Paper no. TSTE-00048-2015. the existing and conventional SG-based power system [1], [2].
The authors are with the Department of Electrical and Computer However, the occurrence of the grid faults causes the stability
Engineering, University of Memphis, Memphis, TN 38152 USA (e-mail: problem of these grid-connected energy sources. Transient sta-
mhssain1@memphis.edu; mhali@memphis.edu).
bility is the property of a power system to regain its normal
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. operating condition following sudden and severe faults in the
Digital Object Identifier 10.1109/TSTE.2015.2463286 system [4]. During a transmission line fault, opening of the CB
1949-3029 © 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

2 IEEE TRANSACTIONS ON SUSTAINABLE ENERGY

is required to protect the healthy section of the power system. In this work, three nonlinear controller schemes, such as an
When the fault arc is deionized, CBs reclose again to maintain FLC, a SNC, and an ANFIS-based VR-FCL are proposed to
the normal operation. The transient stability study is extremely improve the transient stability of a hybrid power system con-
important for maintaining the continuity of the power flow and sisting of PV, wind, and SGs, and this is the main contribution
properly controlling the modern electrical power systems with of this work.
multiple renewable energy sources integrated to it. Transient Since renewable energy sources, such as PV and wind, pro-
moment that is the time interest for the transient stability is usu- duce variable output, variable resistance generation of the FCL
ally limited to 3–5 s, following the disturbance, although it may during the transient moment to attain prefault condition is of
extend to about 10 s for very large systems [4]. utmost important. In this work, the TPD and voltage deviation
The DFIG stator being directly connected to the grid (ΔVPCC ) at the PCC during the transient moment are used as
is extremely sensitive to the grid disturbances [5]–[8]. inputs to FLC, whereas, TPD is employed as the input to SNC.
Applications of superconducting magnetic energy storage Both the controllers generate duty ratio to provide switching
(SMES) and high-voltage direct current (HVDC) link for of the IGBT switch. For implementing the ANFIS controller,
enhancing the dynamic performance of grid-connected wind PCC voltage deviation (ΔVPCC ) and the speed deviation of the
and PV generating systems are reported in [1] and [9], respec- SG are employed as the input to generate the resistance value.
tively. Also, flexible ac transmission system (FACTS) devices, The ANFIS controller was trained by exploiting the generated
such as dynamic voltage restorer (DVR), static synchronous database of the FLC-based VR-FCL. The effectiveness of the
series compensator (SSSC), static var compensator (SVC), proposed controllers is verified by extensive simulations con-
and static synchronous compensator (STATCOM), etc., are ducted in MATLAB/Simulink environment under the studied
employed for enhancing the stability of the power system hybrid system subject to both balanced and unbalanced faults.
network including wind generators [2], [11], [13], [14]. The
impacts of large-scale penetration of PV power to the grid
have been reported in [13]. Applications of the braking resistor
II. C ONCEPT OF S TABILITY E NHANCEMENT BY VR-FCL
(BR) [14], superconducting fault current limiter (SFCL) [15],
SMES [16], and STATCOM [17] are available in the literature The basic concept of fuzzy logic, static, or ANFIS controller-
to augment the stability of the SG-based power system. based VR-FCL for enhancing the stability is to introduce a
Fault current limiters (FCLs) [18] are implemented in medium of active power evacuation properly during the grid
the power system networks for suppressing the short-circuit fault. During the network fault, fault current is fed from power
current, enhancement of transient stability, FRT capability sources to the faulty node due to huge voltage sag at that node
enhancement, power quality improvement, and transformer which causes very small active power and voltage generation at
inrush-current limitation [19]. In [18], a nonsuperconducting the rotating machines (DFIG and SG) and the PV plant. This
bridge-type FCL (BFCL) is proposed with the feature of con- leads the rotating machines of the system to lack of the equilib-
trolling the fault current magnitude. A modified BFCL is rium state and may incur the instability. This situation can be
employed in [20] to improve the FRT of grid-connected wind explained by the swing equation as follows [11]:
generator. To achieve better transient stability, an optimal resis-
tance of the FCL should be inserted, and this optimal resistance 2H d2 δ
= Pm − Pe (1)
is related to the prefault conditions [21]. ω dt2
The penetration of PV and wind power to an SG-based
power system adds to the nonlinearity of a system, where these where Pm is the input mechanical power, Pe is the output elec-
nonlinearities are incurred by intermittent behavior of solar trical power, δ is the rotor angle, and H is the inertia constant
intensity, stochastic variation of wind energy, the real power of the machine. From (1), it can be seen that the stability of
variation from SG-based power system, and the switching phe- the machine can be maintained by making the output electrical
nomena of the inverters and power converters [1]. Both FLC power equivalent to the mechanical power. Introduction of the
and artificial neural network (ANN) are powerful tools and resistor of the FCL during the fault causes power dissipation
have prevalent applications in embedded control systems and and also stator voltage is developed due to voltage drop across
information processing [14], [16]. Fuzzy logic is capable of the resistor of the WG and SG. Delivery of the electrical power
providing definite conclusions in a very simple way from vague is maintained by DFIG and SG, and the desired power balance
or ambiguous information. The ANFIS [22] is a powerful tool is achieved. For the PV, the occurrence of a fault at the grid
that can be obtained by the combination of ANN and FLC. makes the inverter incapable of delivering power generated by
The ANFIS combines the self-learning ability of the ANN with the PV source due to decrease in the grid voltage. This excess
the knowledge-based linguistic expression of the Fuzzy logic. power causes the dc-link voltage to go high due to the power
The ANFIS has been employed to control thyristor-controlled- imbalance between the grid side and PV side [26]. The VR-
switched capacitor (TCSC) to augment both the rotor angle FCL concept is employed to balance the active power at the
stability and system voltage profile in [23]. ANFIS-controlled both sides of the converters during the abnormal grid condition
SSSC damping controller has been exploited in enhancing the (fault). Dynamic inclusion of a resistor between fault point and
stability of the VSWT-based offshore wind farm and single the PV terminal can balance the active power. Again, the inser-
machine system in [24] and [25], respectively. Control signal tion of a resistor would increase the voltage at the connection
for the SVC has been generated by the ANFIS structure in [7] point of the PV and thereby prevents the dc-link voltage to go
to augment the stability of power system. high sharply.
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

HOSSAIN AND ALI: TRANSIENT STABILITY AUGMENTATION OF PV/DFIG/SG-BASED HYBRID POWER SYSTEM 3

Fig. 1. Hybrid power system model.

III. M ODEL S YSTEM

This work employed the power system model depicted in
Fig. 1 for transient stability analysis. It consists of one SG
(100 MW, SG)-based single machine infinite-bus system, to
which one PV farm of 50 MW and one DFIG-based wind
Fig. 2. Block diagram of the (a) GSC controller and (b) RSC controller of the
farm of 60 MW are integrated through a short transmission studied DFIG.
line. These energy sources are delivering power to the utility
grid through double circuit transmission lines. The equations
described in [4] are used to model the SG in this work. The SG
is equipped with automatic voltage regulator (AVR) and gover-
nor (GOV) control systems [27]. The SG parameters used for
the design consideration are available in [27].
Although a practical wind power station consists of many
generators, it is considered to contain a large equivalent aggre-
gated single generator. The wind generator is driven by an
equivalent aggregated variable-speed wind turbine (VSWT)
through an equivalent gearbox [2]. The DFIG stator winding Fig. 3. Block diagram of the VSC controller of the studied PV system.
is connected to the grid by 0.69/66 kV step-up transformer,
whereas the rotor winding is connected to the low voltage
as follows. Literature shows that coordinated operation of
side of the step-up transformer through the power electronic
SFCL and SMES is employed at the PCC to enhance the
converters (RSC and GSC) and a transformer. A dc link is
FRT and the power fluctuation minimization purpose of DFIG-
connected between the RSC and GSC. The GSC is used to
based wind farm [30]. Resistor-based SFCL [15], [31] and
maintain a constant dc-link voltage and regulate the stator volt-
high-temperature superconducting FCL [32] are placed at the
age. Control block diagram of the GSC and RSC are shown in
PCC point in order to enhance the transient stability of SG-
Fig. 2(a) and (b), respectively.
based single-machine-infinite-bus (SMIB) power system. The
The PV plant consists of a large number of PV modules con-
Bonneville Power Administration (BPA) installed a BR which
nected in series-parallel combination to attain the desired power
can evacuate 1400 MW of fixed power to enhance power system
level, and it is connected to the PCC by a boost converter and
stability [33] and the VR-FCL works with the same principle
an equivalent aggregated dc to ac inverter. In this work, 100
like BR. Dangjin power plant operating under utility company
PV modules are connected in series, and total 2500 branches
Korean Power Exchange (KPX) has multiple thermal power
are connected in parallel to form a PV plant of rated 50 MW.
generating units connected to a common bus (PCC) and SFCL
KC200GT PV module with 200 W of peak power is employed
is placed at the PCC to enhance the power system stability [34].
for this work [28]. The perturb and observe (P&O) MPPT [29]
Based on this background of placing the FCLs at the PCC, in
for the PV is implemented on the boost converter for extracting
this paper, we have placed the proposed VR-FCL at the PCC.
maximum PV power at all environment conditions. A three-
level, three-phase VSC is used for the conversion of dc power
to ac power for grid interfacing. Fig. 3 represents the block
diagram of the VSC. IV. VARIABLE R ESISTIVE -T YPE FAULT C URRENT
L IMITER
It is to note here that the nonlinear control-based VR-
FCL is placed at the PCC as shown in Fig. 1. The reason Fig. 4 shows the configuration of VR-FCL, which consists of
to choose locating the VR-FCL at the PCC is described in an isolation transformer connected in series with the power line,
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

4 IEEE TRANSACTIONS ON SUSTAINABLE ENERGY

Fig. 5. Control block diagram for generating the inputs (TPD and ΔPPCC ) of
Fig. 4. Configuration of VR-FCL. the FLC.

a three-phase diode rectifier bridge, a small dc-link inductor

(Ldc ) and a parallel connected resistor (R) to a semiconduc-
tor (IGBT) switch. Ldc works as the current snubber for the
semiconductor switch and also it contributes a negligible volt-
age drop during the normal operation. At normal operating
conditions, the semiconductor switch is ON, and the evacuat-
ing resistor is bypassed. The dc-side voltage equation of the
VR-FCL is given by [21]
6 3
Vdc = sin nA (2)
π π
Fig. 6. MF of the FLC.
where n and A are the transformation ratio and amplitude of
the line voltage, respectively. For a fixed value of R, if the where Vd and Vq are the d- and q-axis voltage component,
bypassed switch is opened during the fault, a dc current with a respectively. Fig. 5 demonstrates the calculation circuit to deter-
value of Idc = Vdc /R flows through it. The maximum allowable mine the TPD by filtering out the fluctuation component of the
current during the fault can be exploited to set the value of R. power due to the fault. When the fault is detected, the semicon-
Incorporation of the switching mechanism of the bypass switch ductor switch starts switching with a duty ratio generated by the
with the duty ratio can make the R dynamic and adaptable as FLC. According to the duty ratio generation, the value of Rdc
follows: can be controlled to evacuate the power generated by the energy
Rdc = (1 − D)R (3) sources properly. For the design consideration of VR-FCL, the
value of R is chosen as 1.60 p.u. Also, the value of Ldc and n
where D and Rdc are the duty ratio and the effective dc used for this work is 0.05H and 1, respectively.
resistance appeared at the dc side, respectively. The effective
resistance (RACeff ) appeared on the ac side can be computed
V. D ESIGN OF FLC FOR VR-FCL
by employing the concept of equal active powers on ac and dc
sides of the diode bridge as follows [21]: An FLC is based on the IF–THEN rules, and it is effective
 √ 2 6 2 when the mathematical modeling is difficult to implement, and
3
A/ 2 π sin π nA there is existence of nonlinearity or uncertainty [35]. The design
3 = . (4) of the proposed FLC is described below.
RACeff Rdc
This results in the following:
A. Fuzzification
π2
RACeff = Rdc . (5) For designing the FLC, the TPD from all the energy
18n2
sources and the voltage deviation at PCC are employed as
Besides limiting fault current, another significant feature of inputs, and the duty cycle (D) for switching the VR-FCL
the VR-FCL is to generate dynamic resistance. The suitable is used as the output. In this work, TPD can be defined
value of the resistance can be inserted to achieve better transient as the difference between the total power (PTotal ) of all
stability by controlling the duty ratio. TPD and ΔVPCC during the energy sources at transient state and that at the normal
the fault are exploited as the inputs to the FLC to generate the operating condition, i.e., TPD = (PTotal at transient state) −
corresponding duty ratio. For better transient stability, fast fault (PTotal at normal operating condition). For the small variation
detection is required and in this work, root-mean-square (rms) of the inputs, triangular MFs that generate better damping
method is employed for this purpose [26]. The PCC rms voltage characteristics than that of other MFs are adopted for both
is computed without additional transformation given by the inputs and output as shown in Fig. 6. For designing the
 FLC, the input variables (TPD and ΔVPCC ) were observed in
VPCC = Vd 2 + Vq 2 (6) different conditions, especially, before and after applying the
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

HOSSAIN AND ALI: TRANSIENT STABILITY AUGMENTATION OF PV/DFIG/SG-BASED HYBRID POWER SYSTEM 5

TABLE I
F UZZY RULES

most severe fault at the transmission line without applying any

protection scheme. The maximum PCC voltage deviation was
about 1.03 p.u. with the system subject to most severe fault (i.e.,
3LG fault). Also, the maximum TPD (TPDmax ) was found to be
2.1 p.u. under the same severe disturbance since the hybrid sys-
tem is capable of delivering maximum active power of 2.1 p.u.
to the grid. Therefore, the universe of discourse (UOD), which
is the range of values associated with fuzzy variables, are set
from 0 to 1.03 p.u. for ΔVPCC and from 0 to 2.1 p.u. for TPD
as shown in Fig. 5. Duty ratio for switching the IGBT switch is
varied from 0 to 1 that can vary the resistance of VR-FCL from Fig. 7. (a) ANFIS internal structure. (b) Block diagram of the ANFIS controller
implemented in this work.
maximum set value to the zero resistance value. The linguistic
variables denoted as NB, NS, ZO, PS, and PB stand for nega-
tive big, negative small, zero, positive small, and positive big,
respectively. For both of the inputs, the equation of the triangu- D. Defuzzification
lar MF employed to determine the grade of membership values In the last step, the output crispy values (duty cycle, D) are
(μAi (x)) is given by [36] determined by simple defuzzification method called center-of-
1 area, which is defined by [36]
μAi (x) = (b − 2|x − a|) (7)  
b
D= Wi d i / Wi (9)
where x is the input to the FLC, a represents the coordinate of
the point at which the grade of membership is 1 and b is the
where di represents the value D, which is expressed in terms of
width of MFs.
linguistic variable in the fuzzy rule table.

B. Fuzzy Rule Base

VI. D ESIGN OF THE ANFIS C ONTROLLER
It is the heart of an FLC, since the control scheme used to
conduct the closed-loop system by storing all possible combi- A. ANFIS Structure
nations of inputs and proper outputs for each of them. For the In this work, Sugeno-type ANFIS structure and learning
sake of demonstration, one of the rules extraction procedures algorithms are adopted [37]. Both ANN and FLC are inde-
is described here. Considering the model system of Fig. 1, if pendent of system modeling and they possess the common
ΔVPCC is PB or TPD lies in PS MF then the system is sub- ability to handle the uncertainties. ANFIS structure is orga-
jected to the severe disturbance and VR-FCL should come into nized into two parts, such as the antecedent and the conclusion
the action with a large resistance. According to (3), for gener- part. Similar to FL system, these two parts are related to each
ating large value of resistance, a small duty cycle is required. other by rules. For the ANFIS structure with the Sugeno-type
Therefore, if TPD is PB and ΔVPCC is PS, then the output is inference system, the rules are given by
NB. The proposed FLC for controlling VR-FCL employs 25
control rules as shown in Table I. If (x1 = Ai ) and (x2 = Bi ) then fi = pi x1 + qi x2 + ri
(10)
C. Fuzzy Inference
where x1 and x2 are the inputs to the ANFIS structure with x1
Mamdani’s method [36] has been used for the inference denotes ΔVPCC and x2 represents the absolute value of speed
mechanism while designing the proposed fuzzy controller. deviation of the SG in this work; Ai and Bi denote the fuzzy
According to the Mamdani’s method, each fuzzy rule with the sets; fi is the output within the fuzzy region specified by the
degree of conformity (Wi ) is given by [36] fuzzy rules; pi , qi , and ri denote the designed parameters that

Wi = μAi (x1 ) × μAi (x2 ) (8) were determined during the training process; and i denotes the
number of MFs of each input.
where (μAi (x1 )) and (μAi (x2 )) are the values of grade of mem- As it can be seen from Fig. 7(a), the elements of the ANFIS
berships and i is the rule number. structure are the same as that of the typical FLC except a layer
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

6 IEEE TRANSACTIONS ON SUSTAINABLE ENERGY

of hidden neurons perform the computation at each stage. In

Fig. 7(a), five layers are shown, where the first layer (Inputs)
with each neuron represents a linguistic variable and the out-
put is equal to the MF of this linguistic variable. In the second
layer (Input MF), the incoming signals are multiplied at each
node and the products are sent out that corresponds to the fir-
ing strength (wi ) of a rule [38]. Every node in the third layer
(Rules), determines the ratio (w̄i ) of the ith rule’s firing strength
to the sum of all rules firing strengths. In the fourth layer (out-
put MF), each node generates the output that is the product
of the relative firing strength of ith rule and the rule fi . The
final layer (output) determines the overall structure output as
the summation of the incoming signals from layer 4 [38].

B. Design of an ANFIS Controller for VR-FCL

Fig. 8. Control surface of the proposed ANFIS.
The PCC voltage deviation (ΔVPCC ) and the absolute value
of speed deviation of the SG (|ΔωSG |) have been utilized as the
input to generate the VR-FCL resistance value by the ANFIS
controller as shown in Fig. 7(b). Designing of the ANFIS
controller requires some basic steps such as: 1) data genera-
tion; 2) rule extraction and MFs; 3) training and testing; and
4) results [7].
1) Data generation: Training data are required to implement
the ANFIS controller. 2-D input vectors and the associ-
ated 1-D output vector were generated by sampling the
input variable ΔVPCC and |ΔωSG | uniformly and estimat-
ing the VR-FCL resistance value R for each sampled Fig. 9. Static nonlinear controller.
point.
2) Rule extraction and MFs: Initial rules are estimated after
generating the training data. The initial rules are opti- will require less memory as compared to the FLC. The surface
mized by the hybrid learning algorithm [7]. The algorithm plot of the input and output signals is shown in Fig. 8. It can be
employs the iterative process to learn the parameters seen from the Fig. 8 that the absolute value of SG rotor speed
of the previous MFs through back-propagation and the deviation varies from 0 to 4 × 10−4 p.u. and the PCC voltage
parameter optimization of the resulted equations are deviation ranges from 0 to 1.03 pu, whereas the resultant out-
accomplished by linear regression analysis [25]. put resistance RVR-FCL is varied from 0 to 0.70 pu. From the
3) Training and testing: Training procedure is conducted control surface, it can be seen that the large ΔVPCC and |ΔωSG |
until reaching the desired error level. lead to the insertion of higher value of resistance of VR-FCL.
4) Determination of the parameters by learning algorithm: Employing the proposed ANFIS with these input/output rela-
ANFIS employs a hybrid learning procedure to esti- tionships, the hybrid power system subject to different faults
mates the parameters of the Sugeno-type fuzzy inference can be effectively stabilized.
systems (FIS). It exploits the least square and the back-
propagation gradient descent methods for the training
purpose of the FIS MF parameters to imitate a given VII. D ESIGN OF THE SNC
system’s training data set [1]. In this work, an alternative SNC is also implemented [39]
ANFIS editor toolbox available in MATLAB/Simulink is for evaluating the performance of the FLC-based VR-FCL and
exploited to check the ANFIS controller. The MFs utilized are ANFIS-based VR-FCL. The simple governing equation of the
Bell-shaped with the number of epochs are 3000. Like [25], nonlinear controller is given by
the sample data for training the ANFIS were collected from
the generated database of the system subjected to severe distur- D/ = K × TPD2 . (11)
bance with the protection scheme of FLC-based VR-FCL. The
ANFIS parameters that are generated after the training are: the The controller parameter K associated with the controller
number of nodes = 35, the number of linear parameters = 9, can be fine-tuned to obtain optimum performance. The con-
the number of nonlinear parameters = 18, the number of train- troller parameter K is set at 0.14, which is found to be
ing data pairs = 2501, and the number of fuzzy rules = 9. As the suitable value for enhancing the stability margin. Fig. 9
it can be seen that the ANFIS controller optimizes the control represents the block diagram of the static controller.
rules and makes it only 9 as contrast to the 25 rules required Since the penetration of PV and wind power to an SG-based
in the case of fuzzy controller. Therefore, the ANFIS controller power system adds to the nonlinearity of a system, the intention
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

HOSSAIN AND ALI: TRANSIENT STABILITY AUGMENTATION OF PV/DFIG/SG-BASED HYBRID POWER SYSTEM 7

Fig. 10. Total power at PCC. Fig. 11. Effective dc resistance of VR-FCL.

was to investigate a simple nonlinear controller that can be Case 3) with FLC-based VR-FCL;
incorporated to generate variable resistance. The square of Case 4) with ANFIS-based VR-FCL.
the TPD is chosen as it represents a very simple nonlinear
controller. Initially, we have tried with the other nonlinear func- B. Transient Stability Improvement by VR-FCL During
tions such as the cubic and biquadratic functions. However, Balanced Fault
quadratic nonlinear function showed the better transient stabil- Fig. 11 shows the effective dc resistance of VR-FCL dur-
ity enhancing capability than that of other functions considering ing the three line-to-ground (3LG) fault at the grid side. The
the different fault scenario extending from most common (1LG) ANFIS- or FLC-based VR-FCL imposes smaller value of resis-
to most severe (3LG) fault. Although the quadratic function tance than that of the static nonlinear control-based VR-FCL.
showed better performance for this hybrid power system, other The resistor of the VR-FCL allows evacuation of the active
nonlinear functions might work well for other power system power of the power sources and also it causes voltage drop at
networks. the PCC, i.e., voltage boosting at the terminals of the power
As a simple SNC is implemented, the value of the controller sources.
constant K plays a significant role on the operation on the Fig. 12 demonstrates the comparative transient responses
nonlinear controller-based VR-FCL. It was observed that if the of the DFIG, when the hybrid power system subject to 3LG
value of K is beyond some ranges, then the performance of VR- fault. DFIG responses are plotted to compare the damping
FCL in terms of stability margin deteriorates rapidly. Within characteristics obtained by the VR-FCL combined with the
the certain constrain of duty cycle that is the duty cycle should three proposed nonlinear control schemes such as the SNC, the
remain in the range of 0–1, we carried out the same procedure. fuzzy controller, and the ANFIS controller. It is evident from
For different fault magnitudes, we used the same value of K the DFIG responses that the VR-FCL joined with any of the
and observed both the system responses and the stability index proposed nonlinear controller offers better transient stability
values of the system. However, beyond a certain limit, the sta- than without the protection scheme. Simulation results of the
bility of the hybrid system tends to be deteriorated due to over SG load angle, real power, rotor speed, and terminal voltage
or under compensation. are shown in Fig. 13(a)–(d), respectively. From the compar-
ative simulation responses of SG, it can be seen that ANFIS
controlled VR-FCL shows superior performance than that of
VIII. S IMULATION R ESULTS other controller-based VR-FCLs. The SG responses are worst
A. Simulation Scenario without any protection scheme.
The dc-link voltage of the PV inverter is shown in Fig. 14.
In this work, simulations have been carried out using the
Low-voltage fault appeared at the grid side leads to the imbal-
MATLAB/Simulink software and considering both balanced
ance in power supplied from the dc side to the ac side of the
and unbalanced temporary faults at F1 location as shown in
PV inverter which causes excessive voltage rise in dc link.
Fig. 1. In the case of PV farm, variable power is generated by
Also, overcurrent in the ac side of the PV inverter may damage
the variable irradiance profile, while the wind generator and SG
the power electronic converter [26]. Application of the VR-
are assumed to generate rated power, which results in a vari-
FCL prevents the PV dc-link voltage and ac-side current to go
able power delivery by the hybrid system to the grid as shown
high and hence protects the power electronic converter during
in Fig. 10. Therefore, the TPD is variable for the hybrid power
grid low-voltage disturbance. In this case also, ANFIS-based
system and VR-FCL needs to be dynamic to produce variable
VR-FCL shows little better performance than that of VR-FCL
resistance for the proper evacuation of the active power. The
joined with other controllers.
fault is considered to occur at 0.1 s, the breakers of the lines
are opened at 0.2 s (after 5 cycles) and reclosed at 1.2 s (after
50 cycles). A total simulation time of 10 s with 0.04-ms time C. Transient Stability Improvement by VR-FCL During
Unbalanced Fault
step is considered. Simulations are conducted for the following
four cases. Stability analysis regarding the unbalanced fault condi-
Case 1) no auxiliary controller; tion is investigated under the proposed protection schemes.
Case 2) with SNC-based VR-FCL; Fig. 15(a)–(c) shows the SG load angle, DFIG rotor speed,
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

8 IEEE TRANSACTIONS ON SUSTAINABLE ENERGY

Fig. 12. Comparative transient responses considering the studied system sub- Fig. 13. Comparative transient responses considering the studied system sub-
ject to 3LG fault. (a) Equivalent DFIG-based wind farm rotor speed. (b) DFIG ject to 3LG fault. (a) SG load angle. (b) SG terminal voltage. (c) SG real power.
dc-link voltage. (c) DFIG real power. (d) DFIG terminal voltage. (d) SG rotor speed.

and PV dc-link voltage profiles for the unsymmetrical double the power system for the unsymmetrical fault is also enhanced
line-to-ground (2LG) fault at F1 location of the studied power by the VR-FCL joined alternatively with three proposed non-
system. It can be seen from Fig. 15 that the transient stability of linear controller schemes.
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

HOSSAIN AND ALI: TRANSIENT STABILITY AUGMENTATION OF PV/DFIG/SG-BASED HYBRID POWER SYSTEM 9

TABLE II
VALUES OF I NDICES FOR P ERFORMANCE C OMPARISON D URING
3LG FAULT

Fig. 14. Comparative transient responses PV dc-link voltage considering the

studied system subject to 3LG fault. D. Index-Based Transient Stability Performance
For more clear perspective, transient stability of the hybrid
system is evaluated by exploiting several performance indices,
such as, dfigpow (p.u. · s), dfigvlt (p.u. · s), dfigspd (p.u. · s), sgang,
sgpow (p.u. · s), sgvlt (p.u. · s), sgspd (p.u. · s), and PVvlt (p.u. · s).
Lower values of the indices indicate improved system perfor-
mance. Mathematical representations of the performance index
of each machine’s one parameter can be defined as follows:
T
dfigspd (p.u. · s) = |Δωwg |dt (12)
0
T
PVvlt (p.u. · s) = |ΔVdc |dt (13)
0
T
sgang (deg · s) = |Δδ|dt (14)
0

where Δωwg , ΔVdc , and Δδ represent speed deviation of DFIG,

dc-link voltage deviation of PV generator, and the load angle
deviation of SG, respectively. Also, T is the total simulation
time of 10 s. Other performance indices of the machines can be
represented mathematically in the same way.
Table II represents the values of the indices for the 3LG
fault. It is evident from Table II that the system performance is
the worst without any protection scheme. However, a notable
improvement in stability performance can be observed with
ANFIS-based VR-FCL, FLC-based VR-FCL, and static nonlin-
ear control-based VR-FCL in the system. Also, the ANFIS- and
FLC-based VR-FCL performs better than the static nonlinear
static control-based VR-FCL in every aspect.

IX. ROLE OF THE P ROPOSED VR-FCL U NDER

U NSTABLE C ASE
Usually, the fault event is followed by the CB corrective
action. The breaker opens the faulted line before the sys-
tem loses synchronism [12]. However, in the absence of the
proposed protection schemes, long lasting electromechanical
Fig. 15. Comparative transient responses of the power system under 2LG fault oscillations are visible because of a wide variation in the phase
at F1 location. (a) Equivalent DFIG-based wind farm rotor speed. (b) PV dc-link angle between the PCC point and the infinite bus. It may cause
voltage. (c) SG load angle. severe adverse impact on the equipment connected to these
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

10 IEEE TRANSACTIONS ON SUSTAINABLE ENERGY

TABLE III
VALUES OF MCCT FOR P ERFORMANCE C OMPARISON D URING
P ROLONGED 3LG FAULT

lines as well as the nearby buses. Transient stability margin is

investigated under the proposed protection schemes. Therefore,
maximum critical clearing time (MCCT) under the system sub-
jected to severe 3LG faults are investigated, since these can be
utilized as the transient stability margin [11]. The MCCT is the
maximum allowed time at which a fault must be cleared to pre-
serve and maintain the stability of the whole system [40]. Lower
value of MCCT indicates less stable situation for power system
transient stability studies [40]. Table III shows the comparative
MCCT, and it can be seen that the MCCT is lowest in the case
of system with no controller. However, a significant increase in
MCCT can be observed if the hybrid power system is equipped
with the proposed controller-based protection schemes.
A case where the system goes to the unstable situation is also
investigated and the roles of the proposed controller schemes
are observed. The simulations have been conducted alterna-
tively with no controller schemes, as well as the proposed
protection schemes with the same operating conditions under
a 3LG fault and for prolonged fault duration of 0.15 s. It can
be seen from Fig. 16 that the hybrid system without any protec-
tion scheme tends to be unstable beyond the MCCT of 140 ms.
However, the SNC-based, FLC-based, or ANFIS-based VR-
FCL is capable of bringing back the stability of the hybrid
system. For clear perspective, Fig. 16(b) and (c) is plotted
with their enlarged version. Therefore, an improvement can be
observed that the proposed protection schemes can enhance the
stability margin and assist the system to retrieve the operation
after faults. From the graphical representation and the MCCT
analysis, it can be seen that ANFIS-based VR-FCL shows the
superior performance than that of the other controller-based
VR-FCLs for handling the unstable case.
Fig. 16. Comparative simulation results with the system under prolonged
(150 ms) 3LG fault at F1 location. (a) DFIG-based wind farm rotor speed.
X. VOLTAGE AND P OWER F LOW B ETWEEN H YBRID (b) PV dc-link voltage. (c) SG load angle.
S YSTEM AND G RID W ITH VR-FCL
In this work, the VR-FCL works as a series voltage booster.
In a whole, the proposed ANFIS-based, fuzzy logic controlled nonlinear controller-based VR-FCL maintains the PCC voltage
or SNC-based VR-FCL introduces a voltage booster that offers level to ±0.1 p.u. of nominal voltage.
series voltage compensation capability and provides a means of Also the PCC voltage is plotted in Fig. 17(a) with the E.ON
power evacuation to mitigate the power imbalance during the grid-code [41], and it can be seen that PCC voltage of the
grid fault. Fig. 17 shows the voltage and power profiles that are hybrid system complies with the grid code by staying above
inserted to the grid. In this case, all the power sources generate the E.ON grid-code line. However, without any auxiliary
the rated power of 2.1 pu. The proposed nonlinear controller- controller in the system, PCC voltage about to violate the
based series compensating devices (VR-FCL) can maintain the grid-code near 1.8 s. Therefore, hybrid power system con-
voltage stability after clearing the fault. Without any controller, nected to a weak power system can be controlled to recover
the PCC voltage oscillates beyond the level to ±0.1 p.u. of the terminal voltages of the renewable energy generator as grid
nominal voltage. However, from the voltage stability point of code suggests. Fig. 17(b) shows that the hybrid power system
view, the power system should maintain steady acceptable volt- capable of delivering an improved power profile to the grid
age at PCC after being subjected to fault. The PCC voltage with all type of VR-FCL. Without the VR-FCL protection
should be between 0.9 and 1.1 p.u. after clearing the fault. The scheme the inserted power to the grid is more oscillatory in
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

HOSSAIN AND ALI: TRANSIENT STABILITY AUGMENTATION OF PV/DFIG/SG-BASED HYBRID POWER SYSTEM 11

In our future work, we will look into other protection and

stabilizing devices for renewable energy sources integration to
the grid.

R EFERENCES
[1] L. Wang and M. S. N. Thi, “Stability enhancement of large-scale inte-
gration of wind, solar, and marine-current power generation fed to an
SG-based power system through an LCC-HVDC link,” IEEE Trans.
Sustain. Energy, vol. 5, no. 1, pp. 160–170, Jan. 2014.
[2] L. Wang and D. N. Truong, “Stability enhancement of DFIG-based off-
shore wind farm fed to a multi-machine system using a STATCOM,”
IEEE Trans. Power Syst., vol. 28, no. 3, pp. 2882–2889, Aug. 2013.
[3] M. A. G. de Brito, L. P. Sampaio, G. Luigi, G. A. e Melo and
C. A. Canesin, “Comparative analysis of MPPT techniques for PV appli-
cations,” in Proc. Int. Conf. Clean Elect. Power (ICCEP), pp. 99–104,
Jun. 2011.
[4] P. Kundur, Power System Stability and Control. New York, NY, USA:
McGraw-Hill, 1994.
[5] L. Wang and L.-Y. Chen, “Reduction of power fluctuations of a large-
scale grid-connected offshore wind farm using a variable frequency
transformer,” IEEE Trans. Sustain. Energy, vol. 2, no. 3, pp. 226–234,
Jul. 2011.
[6] K. E. Okedu, S. M. Muyeen, R. Takahashi, and J. Tamura, “Wind farms
fault ride through using DFIG with new protection scheme,” IEEE Trans.
Sustain. Energy, vol. 3, no. 2, pp. 242–254, Apr. 2012.
[7] L. Wang and D. Truong, “Stability enhancement of a power system with
a PMSG-based and a DFIG-based offshore wind farm using a SVC with
an adaptive-network-based fuzzy inference system,” IEEE Trans. Ind.
Electron., vol. 60, no. 7, pp. 2799–2807, Jul. 2013.
[8] S. Alarai et al., “Voltage booster schemes for fault ride-through enhance-
ment of variable speed wind turbines,” IEEE Trans. Sustain. Energy,
vol. 4, no. 4, pp. 1071–1081, Oct. 2013.
[9] S. T. Kim, B. K. Kang, S. H. Bae, and J. W. Park, “Application of SMES
and grid code compliance to wind/photovoltaic generation system,” IEEE
Trans. Appl. Supercond., vol. 23, no. 3, pp. 5000804–5000807, Jun. 2013.
Fig. 17. (a) PCC voltage. (b) Power flow between hybrid system and grid with [10] L. Wang and K. H. Wang, “Dynamic stability analysis of a DFIG-based
VR-FCL. offshore wind farm connected to a power grid through an HVDC link,”
IEEE Trans. Power Syst., vol. 26, no. 3, pp. 1501–1510, Aug. 2011.
[11] M. S. El Moursi, K. Goweily, J. L. Kirtley, and M. Abdel-Rahman,
nature. Therefore, nonlinear controller-based VR-FCL helps “Application of series voltage boosting schemes for enhanced fault ride-
through performance of fixed speed wind turbines,” IEEE Trans. Power
improving the overall power quality. Del., vol. 29, no. 1, pp. 61–71, Feb. 2014.
[12] R. G. Wandhare and V. Agarwal, “Novel stability enhancing control strat-
egy for centralized PV-Grid systems for smart grid applications,” IEEE
Trans. Smart Grid, vol. 5, no. 3, pp. 1389–1396, May 2014.
XI. C ONCLUSION [13] S. Eftekharnejad, V. Vittal, G. T. Heydt, B. Keel, and J. Loehr, “IMPACT
This paper proposes three nonlinear controllers to generate of increased penetration of photovoltaic generation on power systems,”
IEEE Trans. Power Syst., vol. 28, no. 2, pp. 893–901, May 2013.
the variable resistance for the VR-FCL to enhance the tran- [14] M. H. Ali, T. Murata, and J. Tamura, “The effect of temperature rise of
sient stability of a large-scale hybrid power system consisting of the fuzzy logic-controlled braking resistors on transient stability,” IEEE
DFIG-based wind generator, PV, and SG. Based on simulation Trans. Power Syst., vol. 19, no. 2, pp. 1085–1095, May 2004.
[15] B. C. Sung, D. K. Park, J.-W. Park, and T. K. Ko, “Study on a series
results, the following points are noteworthy. resistive SFCL to improve power system transient stability: Modeling,
1) The proposed ANFIS-, fuzzy logic-, and SNC-based VR- simulation, and experimental verification,” IEEE Trans. Ind. Electron.,
FCL are capable of generating the variable resistance and vol. 56, no. 7, pp. 2412–2419, Jul. 2009.
[16] M. H. Ali, T. Murata, and J. Tamura, “A fuzzy logic-controlled super-
ensure the transient stability augmentation of the hybrid conducting magnetic energy storage for transient stability augmentation,”
power system. IEEE Trans. Control Syst. Technol., vol. 15, no. 1, pp. 144–150, Jan. 2007.
2) During the grid-fault, the sharp rise of dc-link voltage of [17] N. C. Sahoo, B. K. Panigrahi, P. K. Dash, and G. Panda, “Multivariable
nonlinear control of STATCOM for synchronous generator stabilization,”
the PV system can be prevented by proposed protection Int. J. Electr. Power Energy Syst., vol. 26, no. 1, pp. 37–48, Jan. 2004.
schemes. [18] M. T. Hagh and M. Abapour, “Nonsuperconducting fault current limiter
3) The proposed protection methods are also capable of sup- with controlling the magnitudes of fault currents,” IEEE Trans. Power
Electron., vol. 24, no. 3, pp. 613–619, Mar. 2009.
pressing the dc-link overvoltage of the DFIG-based wind [19] M. T. Hagh, S. B. Naderi, and M. Jafari, “New resonance type fault cur-
generator during the appearance of grid-fault. rent limiter,” in Proc. IEEE Int. Conf. Power Energy (PECon’10), 2010,
4) Nonlinear controller-based VR-FCL significantly pp. 507–511.
[20] M. Firouzi and G. B. Gharehpetian, “Improving fault ride-through capa-
improves the transient stability margin. bility of fixed-speed wind turbine by using bridge-type fault current
5) The VR-FCL joined with the proposed controllers is limiter,” IEEE Trans. Energy Convers., vol. 28, no. 2, pp. 361–369, Jun.
capable of stabilizing the power system almost equally. 2013.
[21] S. B. Naderi, M. Jafari, and M. Tarafdar Hagh, “Controllable resistive
Therefore, any of the methods can be chosen for the type fault current limiter (CR-FCL) with frequency and pulse duty-cycle,”
transient stability enhancement of hybrid power system. Int. J. Elect. Power Energy Syst., vol. 61, pp. 11–19, Oct. 2014.
 Thisarticlehasbeenacceptedforinclusioninafutureissueofthisjournal.Contentisfinalaspresented,withtheexceptionofpagination.

12 IEEE TRANSACTIONS ON SUSTAINABLE ENERGY

[22] J. S. R. Jang, “ANFIS: Adaptive-network-based fuzzy inference system,” [38] D.-N. Truong and V.-T. Ngo, “Designed damping controller for SSSC to
IEEE Trans. Syst. Man Cybern., vol. 23, no. 3, pp. 665–685, May/Jun. improve stability of a hybrid offshore wind farms considering time delay,”
1993. Int. J. Elect. Power Energy Syst., vol. 65, pp. 425–431, Feb. 2015.
[23] S. R. Khuntia and S. Panda, “ANFIS approach for TCSC-based controller [39] M. H. Ali, T. Murata, and J. Tamura, “Transient stability enhancement
design for power system stability improvement,” in Proc. IEEE Commun. by fuzzy logic-controlled SMES considering coordination with optimal
Control Comput. Technol. (ICCCCT), 2010, pp. 149–154. reclosing of circuit breakers,” IEEE Trans. Power Syst., vol. 23, no. 2,
[24] D.-N. Truong and V.-T. Ngo, “Designed damping controller for SSSC to pp. 631–640, May 2008.
improve stability of a hybrid offshore wind farms considering time delay,” [40] N. C. Kar, T. Murata, and J. Tamura, “A new method to evaluate the
Int. J. Elect. Power Energy Syst., vol. 65, pp. 425–431, 2015. q-axis saturation characteristic of cylindrical-rotor synchronous,” IEEE
[25] S. R. Khuntia and S. Panda, “ANFIS approach for SSSC controller design Trans. Energy Convers., vol. 15, no. 3, pp. 269–276, Sep. 2000.
for the improvement of transient stability performance,” Math. Comput. [41] N. R. Ullah, T. Thiringer, and D. Karlsson, “Voltage and transient stability
Modell., vol. 57, no. 1–2, pp. 289–300, Jan. 2013. support by wind farms complying with the E.ON Netz grid code,” IEEE
[26] F. Yang, L. Yang, and X. Ma, “An advanced control strategy of PV sys- Trans. Power Syst., vol. 22, no. 4, pp. 1647–1656, Nov. 2007.
tem for low-voltage ride-through capability enhancement,” Sol. Energy,
vol. 109, pp. 24–35, Nov. 2014.
[27] M. H. Ali, Wind Energy Systems: Solutions for Power Quality and
Stabilization. Boca Raton, FL, USA: CRC Press, 2012. Md. Kamal Hossain (S’12) received the Bachelor’s
[28] M. G. Villalva, J. R. Gazoli, and E. R. Filho, “Comprehensive approach degree in electrical and electronic engineering from
to modeling and simulation of photovoltaic arrays,” IEEE Trans. Power Rajshahi University of Engineering and Technology
Electron., vol. 24, no. 5, pp. 1198–1208, May 2009. (RUET), Rajshahi, Bangladesh, in 2009, and the
[29] M. K. Hossain and M. H. Ali, “Overview on maximum power point track- M.Sc. degree in electrical and computer engineer-
ing (MPPT) techniques for photovoltaic power systems,” Int. Rev. Electr. ing from the University of Memphis, Memphis, TN,
Eng., vol. 8, no. 4, pp. 1363–1378, Aug. 2013. USA, in 2013, where he is currently pursuing the
[30] I. Ngamroo and T. Karaipoom, “Cooperative control of SFCL and SMES Ph.D. degree.
for enhancing fault ride through capability and smoothing power fluctua- He served as a Lecturer with the Electrical
tion of DFIG wind farm,” IEEE Trans. Appl. Supercond., vol. 24, no. 5, and Electronic Engineering Department, Stamford
pp. 1–4, Oct. 2014. University Bangladesh, Dhaka, Bangladesh from
[31] M. Tsuda, Y. Mitani, K. Tsuji, and K. Kakihana, “Application of resis- 2009 to 2011. His research interests include the areas of advanced power and
tor based superconducting fault current limiter to enhancement of power energy systems, smart-grid and micro-grid systems, integration of renewable
system transient stability,” IEEE Trans. Appl. Supercond., vol. 11, no. 1, energy sources to the conventional power grid, and FACTS.
pp. 2122–2125, Mar. 2001.
[32] S. M. Alaraifi and M. S. El Moursi, “Hybrid HTS-FCL configura-
tion with adaptive voltage compensation capability,” IEEE Trans. Appl.
Supercond., vol. 24, no. 6, p. 5602208, Dec. 2014.
[33] M. L. Shelton, P. F. Winkelman, W. A. Mittelstadt, and W. J. Bellerby, Mohd. Hasan Ali (SM’08) received the Ph.D. degree
“Bonneville power administration 1400-MW braking resistor,” IEEE in electrical and electronic engineering from Kitami
Trans. Power App. Syst., vol. 94, no. 2, pp. 602–611, Mar. 1975. Institute of Technology, Kitami, Japan, in 2004.
[34] S. Seo, S. J. Kim, Y. H. Moon, and B. Lee, “A hybrid superconduct- He is currently working as an Assistant Professor
ing fault current limiter for enhancing transient stability in Korean power with the Electrical and Computer Engineering
systems,” Phys. C Supercond. Appl., vol. 494, pp. 331–334, Nov. 2013. Department, University of Memphis, Memphis,
[35] M. Datta, T. Senjyu, A. Yona, T. Funabashi, and C. H. Kim, “A frequency- TN, USA. Prior to joining this university, he
control approach by photovoltaic generator in a PV-diesel hybrid power worked as a Faculty with the Department of
system,” IEEE Trans. Energy Convers., vol. 26, no. 2, pp. 559–571, Jun. Electrical Engineering, University of South Carolina,
2011. Columbia, SC, USA, until August 2011. His research
[36] M. H. Ali, M. Park, I. K. Yu, T. Murata, J. Tamura, and B. Wu, interests include advanced power systems, smart-grid
“Enhancement of transient stability by fuzzy logic-controlled SMES con- and micro-grid systems, renewable energy systems, energy storage systems,
sidering communication delay,” Int. J. Elect. Power Energy Syst., vol. 31, and FACTS. He has more than 140 publications including 1 book, 2 book chap-
no. 7–8, pp. 402–408, Sep. 2009. ters, 53 top-ranked journal papers, 65 peer-reviewed international conference
[37] M. Sugeno and G. T. Kang, “Structure identification of fuzzy model,” papers, and 20 national conference papers.
Fuzzy Sets Syst., vol. 28, pp. 15–33, Oct. 1988. Dr. Ali is the Chair of the PES of the IEEE Memphis Section.