2404 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO.

5, SEPTEMBER 2015

Transient Stability and Voltage Regulation

in Multimachine Power Systems Vis-à-Vis
STATCOM and Battery Energy Storage
Adirak Kanchanaharuthai, Member, IEEE, Vira Chankong, Senior Member, IEEE, and Kenneth A.
Loparo, Fellow, IEEE

Abstract—This paper examines the application of STATCOM unavoidably occur in the system, including the variability in-
and battery energy storage to enhance the transient stability of troduced by distributed renewable energy resources (DRERs),
large-scale multimachine power systems with synchronous and including the integration of wind power into conventional power
doubly-fed induction generators (DFIGs). A passivity-based con-
trol design method [interconnection and damping assignment grids, e.g., [1] and [2]. As the diversity of the generation mix in-
passivity-based control (IDA-PBC)] is developed for multimachine creases and larger and larger amounts of power are being obtained
power systems and its performance is evaluated on a two-area from renewable generating sources, differences in the dynamic
system consisting of two synchronous generators (SGs) and two behavior between conventional and alternative generators can
DFIG along with STATCOM/battery energy storage system. play an increasingly important role in the stability and security
The main contributions of this paper are threefold: 1) use of a
STATCOM and battery energy storage system to enhance transient
of the power system when faced with both small- or large-signal
stability and provide voltage regulation with SG and DFIG; 2) disturbances.
demonstrating the application of nonlinear control theory (specif- This paper investigates the application of energy storage for
ically the IDA-PBC methodology) for the design of a stabilizing improving the transient stability and voltage regulation in a
feedback controller in large-scale power systems to improve tran- large-scale power system with conventional and nonconven-
sient system performance; and 3) developing a methodology that
tional generators through the design of a high-performance sta-
can use the additional degrees of freedom in large-scale power
systems in order to further improve system performance, in par- bilizing control system that is capable of keeping the closed-loop
ticular the transient stability margin [measured through critical system transiently stable under a diversity of multimachine
clearing time (CCT)] and the dynamic transient performance of power system operating conditions. Transient stability as studied
the system. In order to achieve power angle stability along with the in this paper addresses the impact of large disturbances such as
simultaneous regulation of frequency and voltage, the performance faults, loss of generating units, large sudden changes in loads,
of the proposed control scheme after the occurrence of large dis-
turbances is evaluated and compared with a conventional power etc., on the ability of the system to converge to a stable post-fault
system stabilizer and a feedback linearizing controller. equilibrium after the fault is cleared from the system.
Challenges with integrating renewable generation into a
Index Terms—Battery energy storage systems (BESS), inter-
connection and damping assignment passivity-based control power system are related to intermittent power availability as
(IDA-PBC), multimachine power systems, static synchronous com- well as differences in their dynamic response characteristics as
pensator (STATCOM), transient stability, wind power systems. compared with conventional synchronous machines. Energy
storage has been discussed as an important technology for
addressing the intermittency of renewable resources as well as
I. INTRODUCTION providing a means for load leveling. However, energy storage
can also play an important role in the real-time control of

W ITH the rapid increase of the size and complexity of

power systems, power system stability and performance
are of increasing importance in the operation of power systems.
real and reactive power, thereby enhancing both small- and
large-signal stability and security of the power system. Of par-
ticular interest is the use of energy storage not only to improve
Recently, power system operation is faced with the difficult
transient stability and damping of power oscillations but also
task of maintaining stability when a diversity of disturbances
to increase power transfer capability of power systems.
Energy storage technologies [3] have been considered for
Manuscript received January 26, 2014; revised May 19, 2014; accepted multimachine power systems to deal with the intermittency of
September 09, 2014. Date of publication October 08, 2014; date of current
version July 17, 2015. This work was supported in part by C.S. Draper Labo-
alternative energy sources [4] and for improving power quality,
ratory, Cambridge, MA, USA, through University Research and Development especially frequency and power angle stability [5]–[7]. For sta-
(URAD) grant. Paper no. TPWRS-00109-2014. bility enhancement of wind energy conversion systems, battery
A. Kanchanaharuthai is with the Department of Electrical Engineering, Col-
lege of Engineering, Rangsit University, Patum-Thani 12000, Thailand.
energy storage [8], [9] has been used to improve frequency sta-
V. Chankong and K. A. Loparo are with the Department of Electrical En- bility through the regulation of active power levels.
gineering and Computer Science, Case School of Engineering, Case Western With continuing developments in power electronic technolo-
Reserve University, Cleveland, OH 44106 USA.
Color versions of one or more of the figures in this paper are available online
gies, flexible ac transmission system (FACTS) devices [10], [11]
at http://ieeexplore.ieee.org. are being used to further enhance power transfer capability of
Digital Object Identifier 10.1109/TPWRS.2014.2359659 ac transmission networks and to improve the controllability of

0885-8950 © 2014 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.
KANCHANAHARUTHAI et al.: TRANSIENT STABILITY AND VOLTAGE REGULATION IN MULTIMACHINE POWER SYSTEMS VIS-à-VIS STATCOM 2405

power flow and voltage thus augmenting power system stability.

Among the FACTS devices, the Static Synchronous Compen-
sator (STATCOM) is of particular interest in this study because
this device is able to improve the transfer capability of a power
system by improving voltage regulation and stability and can
significantly provide smooth and rapid reactive power compen-
sation for voltage support and for enhancing both damping of
power oscillations and transient stability.
In general, STATCOM and (battery) energy storage systems
have been used independently to improve power system opera-
tions, and integrating these devices provides an opportunity to
improve overall small-signal and transient stability of the power
system. Considerable research has addressed the integration of
Fig. 1. STATCOM/battery model. (a) Schematic diagram. (b) Equivalent cir-
either STATCOM or battery energy storage into the power net- cuit (controllable voltage source) (c) Equivalent circuit (controllable current
work, with less attention devoted to the integration of STATCOM source).
and battery energy storage as a system, and the development of
dedicated control systems that simultaneously manage bidirec-
tional real and reactive power flow for multi-machine power electronic voltage-source converter and will be used as a regu-
system applications. Ali [8] and Muyeen et al. [9] have indicated lating device in the ac transmission network by delivering and
that the integration of STATCOM and battery along with other absorbing both active and reactive power simultaneously. The
types of energy storage is more effective than the STATCOM system can support electricity networks by improving power
alone in wind power systems. Chakraborty et al. [13] have pro- factor and voltage regulation and helping to damp electro-
vided a review and application regarding the practical realization mechanical oscillations and enhance transient stability of the
of STATCOM/battery integration, especially in connection with first swing dynamics that result from severe fault conditions.
intermittent renewable energy sources. Baran et al. [14] have A basic schematic diagram of a STATCOM and energy
shown that the combined STATCOM and battery can be used to storage system is shown in Fig. 1(a). The STATCOM/battery
smooth intermittent wind farm power and compensate reactive includes a voltage-source converter, a transformer or reactor,
power simultaneously through simulation studies. and a dc-link with the energy storage device, for example, with
This paper continues this line of investigation and examines battery energy storage systems that are considered in this paper.
the application of an integrated STATCOM and battery system The STATCOM/battery can be modeled as either a con-
using an advanced nonlinear controller to enhance the transient trollable voltage source ( ) as shown in
stability of a power system that includes both conventional SG Fig. 1(b) or a shunt-controllable current source , as shown
and DFIG as found in wind energy conversion systems. The in Fig. 1(c), where the STATCOM is used to regulate the ter-
IDA-PBC control design methodology is used to achieve power minal voltage by controlling the reactive power injected
angle stability and to provide both frequency and voltage reg- or absorbed from the power system while energy stored in the
ulation. To evaluate the effectiveness of the proposed approach battery is used to maintain active power output, regulate system
for improving transient stability, simulation studies are carried frequency, and damp power oscillations, hence increasing the
out on a classical four-machine benchmark system, and perfor- operational reliability and security of the power system. For
mance results of the proposed system are compared to conven-
instance, as a controllable current source, the converter is able
tional control system implementations.
to inject or absorb a current into the power
The remainder of this paper is organized as follows. System
network via the transformer. Both active and reactive power
models are briefly described in Section II. IDA-PBC method-
are able to instantaneously be exchanged with the network
ology for our controller design is presented in Section III. Simu-
by controlling the amplitude and angle of the injected current
lation results are given in Section IV, and conclusions are given
based on the capacity of the battery energy storage device.
in Section V.
The combined nonlinear STATCOM and battery state equa-
II. SYSTEM MODEL tions for the equivalent circuit model, particularly the control-
lable shunt current source that is used to describe the transient
Consider the power network that includes -conventional performance of the STATCOM and battery, are as follows.
machines (SG) and -wind power systems [doubly-fed induc- In coordinates, the dynamic model of a STATCOM and
tion generators (DFIGs)] with -STATCOM/battery units and battery [12] are as follows:
load that is modeled as constant impedances. Dynamic models
of this multimachine power system can be divided into three
main groups: namely, synchronous generators (SGs), DFIGs,
and STATCOM/battery systems as follows.
(1)
A. STATCOM/Battery Models
Here, the model of an integrated STATCOM and battery en-
ergy storage system is developed. This system relies on a power
2406 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO. 5, SEPTEMBER 2015

where and are the injected or absorbed per unit battery where
and STATCOM currents, respectively, is the per unit voltage denotes the and axis rotor voltage input of the
across the capacitor , and model the STATCOM/battery th DFIG, denotes the internal voltage behind transient
transformer losses, denotes the per unit battery voltage, reactances and is regarded as a polar variable converted
represents the per unit system side (ac) bus voltage, and from rectangular variables, the and axis voltages behind
. , and model the battery and the switching the transient reactances and , respectively, with the
losses, respectively, and and represent the PWM modulation transformation [16] . Similarly,
gain and firing angle, respectively, and are the control variables denotes a polar variable converted by the transformation [16]
that need to be chosen. is a constant voltage. Refer to [12] . is the stator terminal voltage. is
for more details. We also define and the rotor circuit constant.
as control variables. Based on our assumption that is Remark 1: Although this DFIG model is analogous to a one-
constant, the STATCOM/battery dynamic model becomes axis model of a synchronous machine, there are substantial dif-
ferences [19], namely: 1) is analogous to a voltage behind a
transient reactance in a synchronous generator despite the fact
that it is not generated from an external excitation current; 2)
(2)
the angle is similar to the angle of the rotor flux magnitude
with respect to the synchronously rotating reference frame not
a stroboscopic angle of rotation of the shaft; and 3) the angle
where , , and are shown in Section III. dynamics in the first equation of (4) contains two extra terms
as compared with the angle equation of a synchronous machine
B. SG and DFIG Modeling because of the variable speed operation of the DFIG. Replacing
the second and third terms of the angle equation with
For the -machine conventional power system the nonlinear results in
dynamic model of the th synchronous generator [15] can be
written as follows:

(3) (5)

where is the rotor slip of the th DFIG. When combining

(5) and the third equation of (4), the DFIG rotor voltage inputs
where and can be directly computed. The representation of
the DFIG model has a distinct advantage because the form of
the model is similar to the SG and we can directly apply existing
nonlinear control design techniques like feedback linearization,
backstepping, etc.
is the power angle of the th generator in radians, is the For both conventional (SG) and wind power systems (DFIG)
rotor speed of the th generator in rad/s, , is the including STATCOM/battery, the output electrical power of the
internal voltage in pu., is the -axis reactor in pu. is th synchronous generator and the th DFIG can be computed,
the -axis transient reactor of the th generator in pu. is the and, after some lengthy but straightforward calculations [18],
voltage of the field circuit of the th generator, the control input, they are written as follows:
in pu. is the inertia coefficient of the th generator in seconds,
is the damping constant, in pu.; is the field circuit time
constant, in pu.; is the mechanical power in pu., and is
the -axis current in pu. denote the real and imaginary
parts of the th element of the system admittance matrix.
For the -machine wind power system, the nonlinear dy- (6)
namic equations [16]–[18] of the th DFIG can be written as

where , ,
, and

(4)
KANCHANAHARUTHAI et al.: TRANSIENT STABILITY AND VOLTAGE REGULATION IN MULTIMACHINE POWER SYSTEMS VIS-à-VIS STATCOM 2407

III. IDA-PBC METHODOLOGY a left inverse of , i.e., , respectively. After

Interconnection and damping assignment-a formulation of multiplying (9) from the left by , we obtain a partial differ-
passivity-based control (PBC), introduced by Ortega et al. [20], ential equation (PDE) and an algebraic equation, respectively, as
is a general design method for high-performance nonlinear con- follows:
trol systems that can be described by a port-Hamiltonian model. (11)
This method not only assigns suitable dynamics to the closed-
(12)
loop system, but, as it is a Hamiltonian formulation, it is also
capable of providing a control design that achieves stabilization From the PDE (11) above, and are free to be
by rending the system passive with respect to a desired storage chosen by the designer given that is skew-symmetric
function and the injection of a suitable level of damping. and is positive semidefinite; may be totally or
Here, we present a brief recapitulation of the the IDA-PBC partially fixed if we can ensure that the Lyapunov stability
method applied to the control of a power system with conditions are satisfied: namely: 1) and 2)
STATCOM/battery. The interested reader is referred to the . The functions , and need
survey and tutorial paper [20] for more details, and, in par- to be determined to satisfy (11) with the Hamiltonian having
ticular, [21] and [22] for applications to transient stability of an isolated minimum at the desired equilibrium point
power systems with synchronous generators alone. so that the equilibrium is stabile, with the Lyapunov stability
Consider a nonlinear system that is affine in the control input conditions also satisfied. In addition, is in the largest in-
with dynamics given by the following set of differential variant set under the closed-loop dynamics (8) contained in
equations: . An estimate of the domain of
attraction of the closed-loop system is then given by the largest
(7)
bounded level set .
with the state variables and the control input As a result, the control law can be directly com-
. and are smooth vector functions in the puted from (12) as follows:
appropriate dimensions. The idea of IDA-PBC is to make the
closed-loop system with a stabilizing (static) feedback control
(13)
an explicit port-Hamiltonian system of the form
The key step in this design method is the solution of the
(8) PDE that guarantees the stability of the closed-loop system.
This technique relies on the concept of exact model matching
where the matrices ) and
of the closed-loop system with desired behavior determined by
, which denote the desired closed-loop interconnection struc-
the pre-specified interconnection structure and dissipation ma-
ture and dissipation, respectively, are determined by the de-
trices.
signer, a Hamiltonian function is the desired
In order to solve the matching equation above, there are dif-
total storage function for the closed-loop system that satisfies
ferent approaches to solve the PDE (11) as follows.
. is the gradient
• Nonparameterized IDA: and are selected to
of .
accomplish the desired structure of the closed-loop system;
In order to make (7) equal to (8), a solution of the so-called
subsequently, all assignable energy functions compatible
matching equation needs to be determined from
with that structure are characterized. This characterization
(9) is provided in terms of a solution of the PDE (9). In addi-
tion, from the family of solutions, we choose the one with
with the matrices and as design variables. equilibrium .
The equilibrium point (can be transferred to the origin) of the • Algebraic IDA: when the desired energy function to be as-
closed-loop system is stable if the desired Hamiltonian signed is selected a priori, then PDE (9) becomes an al-
is positive definite. Consequently, the time derivative of gebraic equation in and . Eventually, the con-
along closed-loop trajectories becomes troller is designed using (13).
Refer to the overview of these and other approaches [20] along
(10) with different applications [23], [24] for further details.
In this multimachine power system application, the Algebraic
Therefore, serves as a Lyapunov function for the closed- IDA approach is used to convert the PDE to an algebraic equa-
loop system and the equilibrium point (the origin) is stable. tion and this equation is used to find the controller that is ca-
In (10), asymptotic stability can be guaranteed provided that pable of achieving the desired closed-loop system performance
is strictly positive definite. In order to solve the matching requirements, namely: 1) the equilibrium point is asymptoti-
equation in (9), we split (9) into two parts: a fully actuated cally stable and 2) power angle stability along with voltage and
part and an unactuated part. Let be , frequency regulation are simultaneously achieved.
where denotes a full-rank left annihilator of , i.e., Implementation of the IDA-PBC design methodology to
, while denotes a practical power system requires modeling a multimachine
2408 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO. 5, SEPTEMBER 2015

system that consists of -SG and -DFIG and includes the where are positive. One consequence of (15) is that
STATCOM/battery system. the two Lyapunov stability conditions above are satisfied and
Consider the dynamic models of -SG, -DFIG and a the closed loop is stabilized by the control law. The proposed
STATCOM/battery system written as an affine nonlinear system controller consists of an excitation controller , DFIG rotor
as follows: voltage input along with STATCOM/battery control
input , generalizing the results obtained in [21] which
only contains the control law for stabilizing a SG excitation
system. We can also rewrite the closed-loop dynamics (7) with
(14) in a port-Hamiltonian representation as shown in Proposi-
tion 1 below.
Proposition 1: The closed-loop system of the multi-machine
power system ( -SG and -DFIG) including a STATCOM/bat-
tery unit can achieve the expected performance requirements
1)–2) with the following state feedback control law:
(14)
(17)

where

.. .. .. (18)
. . .

with , ,
and ,
and

..
.

For this case, and

.
The IDA-PBC design using Algebraic IDA will be used to
find a nonlinear controller for the transient stabilization of the
multi-machine power system with STATCOM/battery equation ..
(14) at a desired equilibrium . The first problem is to select .
an appropriate closed-loop Hamiltonian storage (energy) func-
tion . Note that, in prior work [21], a Lyapunov function
candidate based on synchronous generator dynamics alone can ..
.
be used for the transient stabilization problem. In this paper, the
combined dynamics of SG, DFIG, and STATCOM/battery will
be used in the desired closed-loop energy function. Following
an idea given in [21], we now show that the matrix structure (19)
( and ) can be determined using the following en-
ergy function: where are given in Appendix A.
Further, the closed-loop system is asymptotically stable at a de-
sired equilibrium point

(15) with the Lyapunov and energy functions (15). A domain of

attraction is given by the largest bounded invariant level set
(16) that contains the desired equilib-
rium point . Furthermore, the desired interconnection matrix
KANCHANAHARUTHAI et al.: TRANSIENT STABILITY AND VOLTAGE REGULATION IN MULTIMACHINE POWER SYSTEMS VIS-à-VIS STATCOM 2409

and the damping matrix of the closed-loop dy- ( and ) of the closed-loop system. In particular, in
namics (8) are selected as shown in (20) at the bottom of the (20), the matrices and are selected in order to
page, where . have coupling between the multi-machine power system and the
Proof: The proof and design procedure are given in the STATCOM/battery dynamics.
Appendix A. Remark 4: The IDA-PBC methodology has a universal sta-
Remark 2: The implementation of the proposed controller bilization property [20], [25]. This means that it is able to gen-
requires the following three steps. erate asymptotically stabilizing controllers for nonlinear sys-
Step 1) Initialization: tems of the form (7). Apart from solving the PDE, another dif-
• Find the steady-state values using load flow anal- ficulty in applying the IDA-PBC method is selecting the de-
ysis for: sired closed-loop interconnection structure and dissipation ma-
• synchronous generators: trices ( and ) when the energy function is
; selected a priori. Suitable selection of these matrices is essen-
• DFIG: tial for the success of the IDA-PBC design, and there is cur-
, and then compute and rently no general method for selecting them. In this paper, we
from the above-mentioned transformation. first choose the energy function based on physical considera-
• STATCOM/battery unit: tions, see [21], [24] for example, because it is simple and also
Step 2) Computations: satisfies the above mentioned Lyapunov stability conditions.
• Define the energy function as in (15) and Then, we choose the of and to obtain addi-
an interconnection structure , in partic- tional damping through a combination of mechanical subsys-
ular the elements, and the dissipation matrix tems, i.e., and electrical subsystems,
as in (20). i.e., in (20) where the are free parameters to be
• Compute using chosen as part of the design methodology.
(23)–(26). For the sake of brevity, during the simulation results, to
Step 3) Controller equations: evaluate the capability of the proposed controller presented in
• Implement (17)–(19). Section IV, there are two different types of contingencies used:
Remark 3: In general, SGs, DFIGs, or both exhibit poor me- short circuits and load variations. These types of disturbances
chanical damping, and, when there is no coupling between the belong to classes of very severe contingencies and may result
electrical damping and the mechanical damping, this results in in instability of power systems where only conventional con-
significant power oscillations in the system under fault condi- trollers are employed.
tions. In order to avoid and mitigate any damage to the sys-
tems, these oscillations have to be damped effectively. Thus, IV. SIMULATION RESULTS
additional damping should be included through a coupling be- Here, the proposed control methodology is implemented on a
tween the mechanical subsystem (SGs and DFIGs) and the elec- multimachine power system with a STATCOM/battery system,
trical subsystem (STATCOM/battery) which can be assigned and the closed-loop performance of the system is evaluated
by selecting suitable interconnection and damping structures using computer simulation studies under various transient

.. .. .. .. .. ..
. . . . . .

(20)
2410 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO. 5, SEPTEMBER 2015

Fig. 2. Simple two-area system with STATCOM/battery.

conditions. The complete system dynamics is obtained solving The control performance of the power systems with
the differential equation (14) in the MATLAB environment. STATCOM/battery for short circuit fault and changes in load
The time domain simulations are carried out to investigate the are given in the following subsections.
damping performance of the designed controllers, as given
in (17), in the system under study. The performance of the A. Three-Phase Fault Test
proposed controller (IDA-PBC with STATCOM/battery) under A three-phase fault on the transmission network is the most
a variety of test scenarios is compared with the following severe disturbance to a power system. In this paper, a symmet-
existing controllers. rical three-phase fault occurs at bus 9 and the following fault
• FBLC: A feedback linearizing (FBL) controller [15] con- sequence is taken into account to evaluate the performance of
sisting excitation controller ( ), DFIG rotor voltage input the proposed controller: 1) fault occurs at 1 s and 2) fault
and STATCOM/battery control input . is cleared at 1.1 s. In general, when a three-phase fault oc-
• IDA-PBC-alone: An IDA-PBC controller alone curs, the dynamics during the postfault period can become un-
( ) without STATCOM/battery control input. stable because insufficient damping is provided from the exci-
• PSS: A linear conventional lead-lag power system sta- tation systems of each generator. This instability needs to be
bilizer for conventional generators ( ) [26] resolved and PSSs are widely used in conventional power sys-
and a power system stabilizer for wind generators tems to provide some additional damping to the system. The
( ) [27]. STATCOM/battery is integrated into a power system that in-
In order to gain insight into the performance improvements cludes both conventional and wind power generators (DFIGs),
that can be achieved in multi-machine power systems using the this combination offers additional degrees of freedom to add
proposed control design approach, we study a classical four- damping to the power system beyond that of the PSS alone,
machine power system [26] as shown in Fig. 2. We recognize if managed appropriately, and assist with mitigating the insta-
that even though these results are preliminary, they do suggest bility problem. The relative rotor angle responses of genera-
the improvements that can be achieved using the proposed ap- tors 2–4 with respect to generator 1 are shown in Fig. 3, in
proach. In particular, we evaluate the effectiveness of the pro- which it can be seen that the oscillations in the relative rotor
posed IDA-PBC control design methodology on transient sta- angles are sluggishly damped by the the PSS. The postfault
bility enhancement when a STATCOM/battery is installed at the system responses are quite oscillatory and eventually settle to
midpoint of the transmission line between area 1 and area 2. the pre-fault values after the transient fault is cleared. The addi-
This system is modified by replacing two conventional gener- tion of STATCOM/battery and advanced (IDA-PBC and FBL)
ator and with two wind farms, in particular controllers stabilize the system without sustained oscillations by
and in area 2 for the test system used in this paper. For providing additional damping to the system beyond that of the
controller design, synchronous generators and of area PSS and the IDA-PBC-alone.
1 are conventional plants or aggregates of plants along with ag- To evaluate improvements in transient stability of a highly
gregated wind generators (DFIGs) in area 2 [28]. The param- nonlinear power system, simulation studies using the proposed
eters and the normal operating condition along with the data (IDA-PBC) controller are compared with an existing nonlinear
used for the synchronous generators, transformers, and lines (FBL-feedback linearizing) controller and IDA-PBC controller
of the four-machine system can be found in [26]. Data for the alone. In this test, we assume that the two types of generators
wind generators (DFIGs) and STATCOM/battery are given in supply constant power to the system given that during the
the Appendix. fault period the input mechanical power is constant; thus,
KANCHANAHARUTHAI et al.: TRANSIENT STABILITY AND VOLTAGE REGULATION IN MULTIMACHINE POWER SYSTEMS VIS-à-VIS STATCOM 2411

TABLE I
CCT OF 2-SG AND 2-DFIG CASES

Fig. 3. Controller performance in case of three-phase fault-relative rotor angles

, , and (deg.) (solid line: IDA-PBC with STATCOM/
battery; dashed line: FBLC; dotted line: IDA-PBC-alone; dashed–dotted line:
PSS). and the conventional PSS, the performance of the IDA-PBC
controller with STATCOM/battery is much better when a
three-phase short circuit fault occurs at Bus 9. Because the
proposed control scheme is based on a nonlinear model of the
power system, it is capable of improving the transient stability
margin (measured through CCT) following this major distur-
bance. Comparison of the transient stability margin for the two
area system among the IDA-PBC control with STATCOM/bat-
tery, the FBLC and IDA-PBC-alone, and the conventional
PSS is shown in Table I. Table I presents the critical clearing
times (CCT) in percentage for the two area power system that
includes 2-SG and 2-DFIG with STATCOM/battery along with
the following configurations: IDA-PBC with STATCOM/bat-
tery, FBLC and IDA-PBC-alone compared to PSS. From
Table I, it is obvious that when the fault takes place in buses 6,
Fig. 4. Controller performance in case of three-phase fault-transmission power 9, and 11, respectively, the proposed IDA-PBC control scheme
from area 1 to area 2 ( ) and voltage at bus 8 ( ) (solid line: IDA-PBC with STATCOM/battery significantly increases the transient
with STATCOM/battery; dashed line: FBLC; dotted line: IDA-PBC-alone;
dashed–dotted line: PSS).
stability margin (CCT) when compared with the FBLC and
IDA-PBC alone and PSS control schemes.

the rotor angles of the generators should settle to the pre-fault B. Load Change Test
steady-state operating condition after the fault is cleared. Load changes are considered as a type of transient distur-
Fig. 3 illustrates the relative rotor angles of generators 2–4 bance in this study. For example, a sudden increase (or decrease)
with respect to generator 1 when a three-phase fault occurs on in load demand at particular buses can induce transients in the
Bus 9. As shown, the proposed controller not only responds system, that if inadequately damped can lead to system insta-
faster than the FBL and IDA-PBC controllers alone to maintain bility. The performance of the two area power system is in-
the postfault steady-state conditions but also provides better vestigated for a transient 50% decrease of the load at Bus 7
transient response performance in terms of reduced overshoot during the time period 1.0–1.5 s. Consequently, such a
and faster reduction of oscillations when compared with the decrease in load results in changes to the values of and
PSS. in the system's bus admittance matrix, and this effects the net-
Another important requirement considered in this work is work configuration and the power transfer characteristics of the
the post-fault steady-state voltage and frequency regulation. power system. Temporary differences in the power balance be-
Fig. 4 indicates the terminal voltage at Bus 8 and power tween the mechanical power (assumed constant) and the elec-
transfer from area 1 to area 2, and illustrates that both voltage trical power of each machine can lead to acceleration (deceler-
and transferred power experience a brief transient period in ation) of rotor angles for the entire power system. For this test,
response to changes in the network configuration of the power the STATCOM/battery system is installed to provide some ad-
system but then settle back to their original steady-state (equi- ditional damping to the system to mitigate the acceleration (or
librium) points. In particular, the IDA-PBC controller with deceleration) that results after the occurrence of load changes.
STATCOM/battery, FBLC and IDA-PBC-alone exhibit better The results are illustrated in Figs. 5 and 6 that show the relative
transient response characteristics than the conventional PSS rotor angles, power exchanged between area 1 and area 2, and
controller. As compared with the FBLC and IDA-PBC-alone terminal voltage at Bus 8, respectively. These figures indicate
2412 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO. 5, SEPTEMBER 2015

Fig. 5. Controller performance in case of load changes-Relative rotor angles Fig. 7. Dominant eigenvalues of the two-area test system with each controller
, , and (deg.)(Solid: IDA-PBC with STATCOM/battery, (IDA-PBC with STATCOM/battery: ; FBLC: , IDA-PBC alone: ; PSS: ;
Dashed: FBLC, Dotted: IDA-PBC-alone, Dashdot: PSS). No control: ).

test system in order to gain insight into the performance

improvements, in particular, the damping of power system
oscillations. Fig. 7 show the dominant eigenvalue analysis
results at a normal operating point for the two-area test system
with the different control methods. It can be seen in Fig. 7 that,
for the open-loop system without any stabilizers, there is a
positive real eigenvalue ( ), leading to a dynamically
unstable system. It is therefore necessary to install the PSS
to improve the dynamic performance. The introduction of the
PSS on the two SG and the two DFIG can push all system
eigenvalues into the left half plane and simultaneously confine
them to a conic sector with a minimum damping ratio .
In particular, it is obvious that the system minimum damping
Fig. 6. Controller performance in case of load changes-Transmission power ratio ( )
from area 1 to area 2 ( ) and voltage at bus 8 ( ) (solid line: IDA-PBC from installing the PSS greatly improves the system response
with STATCOM/battery; dashed line: FBLC; dotted line: IDA-PBC-alone;
dashed–dotted line: PSS). as compared with the case without the stabilizers. Similar
improvements in damping ratios are also achieved using
IDA-PBC-alone ( ) and FBLC ( ),
that similar to the previous short-circuit fault tests, the post-fault respectively. This results in better damping of the power system
responses of the PSS are quite oscillatory but ultimately settle oscillations as compared with the PSS and with no stabilizers.
back to the postfault steady-state within a prolonged time pe- Moreover, if the IDA-PBC technique is combined with the
riod. On the other hand, with adequate damping provided by the STATCOM and battery (IDA-PBC with STATCOM/battery)
proposed IDA-PBC with STATCOM/battery, FBLC and IDA- the minimum damping ratio improves ( ) and this
PBC-alone controllers, their post-fault responses are hardly os- configuration provides the best damping characteristics of the
cillatory and quickly return to the postfault steady state. From those studied and greatly enhances the dynamic stability of the
these simulation results, based on the characteristics of the tran- power system. As observed from Fig. 7, as the damping ratio
sient dynamics after the disturbance, we conclude that the pro- increases the damping of power system oscillations improves.
posed IDA-PBC controller with STATCOM/battery performs As the time responses in Figs. 3–6 show, the system perfor-
better than the FBLC and IDA-PBC-alone and the conventional mance with the IDA-PBC and STATCOM/battery outperforms
PSS, and that the FBLC and IDA-PBC-alone outperform the the other design approaches and the power system oscillations
PSS. are suppressed much more rapidly.
Additionally, we can discuss the impact of the different From the simulation results provided, we can conclude that,
control methodologies on the damping of power oscillations by regardless of the steady-state operating point of the system, the
analyzing the dominant eigenvalues of the linerarized system. proposed IDA-PBC controller is capable of not only achieving
In this study, the eigenvalue analysis [26] is performed on the expected performance requirements 1) and 2) mentioned
the linearized system model of the test system without any previously, but it also improves the transient dynamic response
stabilizer as well as with PSS, IDA-PBC-alone, FBLC, and performance of the system after the occurrence of a tempo-
IDA-PBC with STATCOM/battery, respectively. Table II shows rary fault and transient changes in load. In comparison with
the system eigenvalues and minimum damping ratios of the the FBLC and IDA-PBC-alone and the conventional power
KANCHANAHARUTHAI et al.: TRANSIENT STABILITY AND VOLTAGE REGULATION IN MULTIMACHINE POWER SYSTEMS VIS-à-VIS STATCOM 2413

TABLE II V. CONCLUSION
EIGENVALUES ( ) AND MINIMUM DAMPING RATIOS ( ) OF EACH
CONTROLLER ( ) In this paper, the IDA-PBC nonlinear control design method
has been proposed to enhance the transient stability and fre-
quency and voltage regulation of multi-machine power systems
that include both SG and DFIG and a STATCOM/battery
energy storage device. Integrating the STATCOM/battery into
the power systems using the IDA-PBC control design method-
ology provides additional damping to the system and dynamic
simulations on a two-area multi-machine power system have
shown that the nonlinear IDA-PBC control design approach
is capable of providing improved system damping and better
transient stability performance. Moreover, the proposed control
scheme offers several advantages. It is well-known that conven-
tional control strategies such as PSS are based on linear design
concepts and other nonlinear control design methods such as
feedback linearization (FBLC) rely on finding a coordinate
transformation that transforms the original system to normal
form. In contrast, the proposed scheme is a nonlinear control
design methodology for nonlinear systems that does not rely
on linearization or feedback linearization, and offers increased
degrees of freedom in the design process through the selection
system stabilizer, the proposed IDA-PBC controller with of a suitable energy function and a desired interconnection and
STATCOM/battery not only effectively damps the oscillations damping structure. The dynamic simulation results have clearly
of relative rotor angles between area 1 and area 2, but also has indicated that power angle (transient) stability and voltage and
superior performance in maintaining the terminal bus voltage frequency regulation improvements are achievable for large
magnitudes close to their reference voltage values defined for (transient) disturbances. In particular, we observed significant
the normal operating conditions, and provides effective voltage increases in CCT (IDA-PBC with STATCOM/battery) for the
regulation to the desired pre-fault steady-state values after the two test studies of interest when compared to operation with a
occurrence of major transient disturbances. Additionally, the feedback linearizing control (FBLC), IDA-PBC-alone and the
IDA-PBC nonlinear controller with STATCOM/battery can combined conventional and alternative (wind) power system
maintain synchronization of generators along with effective controllers (PSS), thereby providing a significant improvement
damping of the power oscillations when compared to the in power system stability, robustness and dynamic security.
conventional control configurations. In closing, we recognize that the multi-machine power system
As indicated in the results above, the IDA-PBC control results presented in this paper are preliminary and that there is
strategy was effectively designed to improve transient stability considerable work that needs to be done to extend this approach
and voltage regulation following short-circuit and load change to large-scale power systems. However, we have clearly shown
conditions. The proposed IDA-PBC design methodology has that integrating a STATCOM/battery into power systems using a
numerous advantages: it is a systematic methodology that en- nonlinear control (e.g., IDA-PBC) design methodology has the
sures closed-loop stability and improved transient performance potential to improve power system performance. We are cur-
for multi-machine power systems with STATCOM/battery rently working to demonstrate these results on the laboratory
and facilitates the determination of controller parameters as scale power system discussed in [29], [30].
compared with conventional control design strategies that
include iterative tuning approaches. The IDA-PBC design APPENDIX A
strategy facilitates any additional damping to the system via IDA-PBC DESIGN PROCEDURE
the selection of appropriate matrices ( ), where
each matrix includes the coupling between the electrical Proof of Proposition 1:
damping and the mechanical damping, thereby mitigating and
Consider the desired interconnection and damping matrices,
suppressing heavy oscillations in the system. The proposed
which are to be found in (20) and the energy function (15). After
method also provides improved dynamic response as compared
some simple calculation, for each th row, ,
with well-known strategies such as FBLC, IDA-PBC-alone and
we obtain a PDE that becomes an algebraic equation of the form
linear control design. The direct controller design approach for
(nonlinear) multi-machine power systems is not limited by the
use of a linearized model of the system around a steady-state
operating point (PSS) or to the construction of a coordinate
transformation that transforms an affine nonlinear system into (21)
the so-called normal form (FBL) with the appropriate selection
of an output function [15]. where .
2414 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO. 5, SEPTEMBER 2015

Substituting (15) into the expressions above, recalling the

Leibniz integral formula and then evaluating at the equilibrium
point , we ..
have .

(22)

where and ..
.

Substituting these expressions back into (21) and then using (25)
some lengthy but straightforward calculations, we obtain
as follows:

(26)
..
.
where and
are arbitrary positive con-
stants. The control law can be computed by selecting the
free parameter as and using
the pre-fault equilibrium equations as given in (17). The
.. Lyapunov function satisfying Poincare's Lemma [20] can be
.
obtained by including all elements ( )
.. in a vector ( ), requiring that
. , directly integrating ,
and then substituting back into (15). Further, to ensure the
minimum condition ( and ) the
Hessian of evaluated at the desired equilibrium also needs
.. to be positive semidefinite ( ) as the SG-DFIG case
.
[18]. From the analysis above, it follows that, with selecting
in (20), the closed-loop dynamics consisting of
(14) with (17) matches the model (8). Consequently, the time
derivative of the energy function along the trajectories
of (8) satisfies the following equality:
(23)

..
.

(27)

Note that plays the role of Lyapunov function, which

ensures the Lyapunov stability at the point (
(24) ). Moreover, from
KANCHANAHARUTHAI et al.: TRANSIENT STABILITY AND VOLTAGE REGULATION IN MULTIMACHINE POWER SYSTEMS VIS-à-VIS STATCOM 2415

the dynamical description of (14) with (17), it follows that [15] Q. Lu, Y. Sun, and S. Mei, Nonlinear Control Systems and Power Sys-
as . This means that tems Dynamics. London, U.K.: Kluwer, 2001, pp. 223–234.
[16] K. Elkington, “Modelling and control of doubly fed induction genera-
from the equilibrium condition. Therefore, the result is con- tors in power Systems: Towards understanding the impact of large wind
cluded that the closed-loop system is asymptotically stable at parks on power system stability,” Ph.D. dissertation, Sch. Electr. Eng.,
the desired equilibrium . The domain of attraction can be KTH Royal Inst. of Tech, Stockholm, Sweden, 2009.
explicitly determined using the invariance principle, namely, [17] A. Kanchanaharuthai, V. Chankong, and K. A. Loparo, “Small-signal
stability enhancement of power systems with renewable distributed
and where . energy resources,” in Proc. 18th IFAC World Congr., Milano, Italy,
Aug.-Sep. 28–2, 2011, pp. 513–518.
[18] A. Kanchanaharuthai, “Small-signal stability, transient stability and
APPENDIX B voltage regulation enhancement of power systems with renewable
DFIG AND STATCOM/BATTERY PARAMETERS distributed energy resources,” Ph.D. dissertation, Dept. Electr. Eng.
Comput. Sci., Case Western Reserve Univ., Cleveland, OH, USA,
Doubly Fed Induction Generator Data: 2012.
[19] S. Ahmed-Zaid and M. Taleb, “Structural modeling of small and large
• Common generator data: 0.01 pu, 0.1 pu, induction machines using integral manifolds,” IEEE Trans. Energy
0.1 pu., 3 pu. Convers., vol. 6, no. 3, pp. 529–535, 1991.
[20] R. Ortega and E. Garcia-Canseco, “Interconnection and damping as-
• : 2.5 pu., , signment passivity-based control: A survey,” Eur. J. Control, vol. 10,
• : 2.0 pu., . no. 5, pp. 432–450, 2004.
[21] R. Ortega, M. Galaz, A. Astolfi, Y. Sun, and T. Shen, “Transient sta-
STATCOM/Battery: bilization of multimachine power systems with nontrivial transfer con-
ductances,” IEEE Trans. Autom. Control, vol. 50, no. 1, pp. 60–75, Jan.
• STATCOM: Rated at 100 MVAr, 0.001 pu. 2005.
[22] M. Galaz, R. Ortega, A. S. Bazanella, and A. M. Stankovic, “An
0.22 pu. 40 kV, .
energy-shaping approach to the design of excitation control of
• Battery: 0.01 pu., 800 V. synchronous generators,” Automatica, vol. 39, no. 1, pp. 111–119,
2003.
[23] F. Dorfler, J. K. Johnsen, and F. Allgower, “An introduction to
REFERENCES interconnection and damping assignment passivity-based control
[1] T. Ackermann, Wind Power in Power Systems. Hoboken, NJ, USA: in process engineering,” J. Process Control, vol. 19, no. 9, pp.
Wiley, 2005. 1413–1426, 2009.
[2] S. Heier, Grid Integration of Wind Energy Conversion Systems, 2nd [24] R. Ortega, F. Castanos, and A. Astolfi, “Control by interconnection
ed. Hoboken, NJ, USA: Wiley, 2006. and standard passivity-based control of Port-Hamiltonian systems,”
[3] P. Ribeiro, B. Johnson, M. Crow, A. Arsoy, and Y. Liu, “Energy storage IEEE Trans. Autom. Control, vol. 53, no. 1, pp. 2527–2542, Jan.
systems for advanced power applications,” Proc. IEEE, vol. 89, no. 12, 2008.
pp. 1744–1756, 2000. [25] R. Ortega, A. Van der Schaft, B. Maschke, and G. Escobar, “Intercon-
[4] S. M. Muyeen, H. M. Hasanien, and A. Al-Durra, “Transient stability nection and damping assignment passivity-based control of port-con-
enhancement of wind farm connected to a multi-machine power trolled Hamiltonian systems,” Automatica, vol. 38, no. 4, pp. 585–596,
system by using an adaptive ANN-controlled SMES,” Energ. Convers. 2002.
Manag., vol. 78, pp. 412–420, 2014. [26] P. Kundur, Power System Stability and Control. New York, NY,
[5] M.-H. Wang and H.-C. Chen, “Transient stability control of multi- USA: McGraw-Hill, 1994.
machine power systems using flywheel energy injection,” Proc. Inst. [27] F. M. Huhges, O. Anaya-Laro, N. Jenkins, and G. Strbac, “A power
Electr. Eng.—Gener. Transm. Distrib., vol. 152, pp. 589–596, 2005. system stabilizer for DFIG-based wind generation,” IEEE Trans.
[6] S. M. Sadeghzadeh, M. Ehsan, N. Hadj Said, and R. Feuillet, “Tran- Power Syst., vol. 21, no. 2, pp. 763–772, May 2006.
sient stability improvement of multi-machine power systems using [28] L. Fernandez, C. Garcia, J. Saenz, and F. Jurado, “Equivalent models of
on-line fuzzy control of SMES,” Control Eng. Pract., vol. 7, no. 4, pp. wind farms by using aggregated wind turbines and equivalent winds,”
531–536, 1999. Energ. Convers. Manage., vol. 50, no. 3, pp. 691–704, 2009.
[7] K.-I. Kawabe and A. Yokoyama, “Transient stability improvement of [29] D. H. Moore, H. M. Abdar, J. M. Murray, A. Chakraverty, and
multi-machine power system with large-capacity battery systems,” K. A. Loparo, “Grid of the future demonstration system,” in Proc.
IEEJ Trans. Power Energy, vol. 132, no. 1, pp. 47–56, 2012. IEEE EnergyTech Conf., Cleveland, OH, USA, May 29–31, 2012,
[8] M. H. Ali, Wind Energy Systems: Solutions for Power Quality and Sta- pp. 1–6.
bilization. Boca Raton, FL, USA: CRC, 2012. [30] H. M. Abdar, A. Chakraverty, D. H. Moore, J. M. Murray, and K. A.
[9] S. M. Muyeen, J. Tamuru, and T. Murata, Stability Augmentation Loparo, “Design and implementation a specific grid-tie inverter for an
of a Grid-Connected Wind Farm. London, U.K.: Springer-Verlag, agent-based microgrid,” in Proc. IEEE EnergyTech Conf., Cleveland,
2009. OH, USA, May 29–31, 2012, pp. 1–6.
[10] N. G. Hingorani and L. Gyugyi, Understanding FACTS: Concepts and
Technology of flexible AC Transmission Systems. Piscataway, NJ,
USA: IEEE, 1999.
[11] Y. H. Song and A. T. John, Flexible AC Transmission Systems
(FACTS), ser. Power and Energy 30. London, U.K.: IEE , 1999.
[12] Z. Yang, C. Shen, L. Zhang, M. L. Crow, and S. Atcitty, “Integration
of a STATCOM and battery energy storage,” IEEE Trans. Power Syst.,
Adirak Kanchanaharuthai (M'12) received the B.Eng degree in control engi-
vol. 16, no. 2, pp. 254–260, May 2001.
neering from King Mongkut's Institute of Technology Ladkrabang, Thailand, in
[13] A. Chakraborty, S. K. Musunuri, A. K. Srivastava, and A. K. Kond- 1995, the M.Eng. degree in electrical engineering from Chulalongkorn Univer-
abathini, “Integrating STATCOM and battery energy storage systems sity, Thailand, in 1997, and the Ph.D. degree in systems and control engineering
for power system transient stability: A review and application,” Ad- from Case Western Reserve University, Cleveland, OH, USA, in 2012 .
vances in Power Electron., vol. 2012, 2012. Currently, he is an Assistant Professor with the Department of Electrical En-
[14] M. E. Baran, S. Teleke, L. Anderson, A. Huang, S. Bhattacharya, and gineering, Rangsit University, Patum-Thani, Thailand. His main research inter-
S. Atcitty, “STATCOM with energy storage for smoothing intermittent ests include power system dynamics, stability and control, renewable energy
wind farm power,” in Proc. IEEE PES Gen. Meet., Jul. 20–14, 2008, systems, energy storage systems, and applications of nonlinear control theory
pp. 1–6. to power systems and flexible ac transmission systems (FACTS) devices.
2416 IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 30, NO. 5, SEPTEMBER 2015

Vira Chankong (LSM'14) received the Ph.D. de- Kenneth A. Loparo (F'99) received the Ph.D. de-
gree in systems and control engineering from Case gree in systems and control engineering from Case
Western Reserve University, Cleveland, OH, USA, Western Reserve University, Cleveland, OH, USA,
in 1977. in 1977.
He is an Associate Professor of electrical engi- He is the Nord Professor of Engineering and
neering and computer science engineering with Case Chair of the Department of Electrical Engineering
Western Reserve University, Cleveland, OH, USA. and Computer Science at Case Western Reserve
He is well-known for his work in systems thinking University. His research interests include stability
and systems methodology for complex problem and control of nonlinear systems with applications
solving, as well as multi-objective and large-scale to large-scale electric power systems, nonlinear
optimization. His current research interests are in filtering with applications to monitoring, fault
the areas of large-scale optimization, multiobjective optimization, logic-based detection, diagnosis and reconfigurable control.
discrete optimization, intelligent computing, and computational engineering Dr. Loparo has received numerous undergraduate and graduate teaching
and science with specific application to energy and biomedicine. He is the awards and has held numerous positions in the IEEE Control Systems Society.
senior author of Multiobjective Decision-Making: Theory and Methodology
(Elsevier-North Holland, 1983 and Dover, 2010. He has published several
papers in multiple objective optimization, large-scale optimization, systems
methodology, and applications of decision theory, optimization and information
technology to medical research and engineering design and operation problems.
His current research focuses on application of systems concepts and systems
tools to radiation treatment planning, signaling pathways and other systems
biology problems. His special interests are in the development of specialized
large-scale optimization algorithms for treatment planning of Gamma Knife
radiosurgery and Intensity Modulated Radiation Therapy, data mining, and
supply chain management.
Prof. Chankong is a senior member of IIE and a member of SIAM and IN-
FORMS.