J. Mod. Power Syst.

Clean Energy (2018) 6(3):473–481

https://doi.org/10.1007/s40565-017-0363-3

Observer based sliding mode frequency control for

multi-machine power systems with high renewable energy
Ayyarao S.L.V. TUMMALA1, Ravikiran INAPAKURTHI1,
P. V. RAMANARAO2

Abstract With the global consciousness of climate Keywords Renewable energy, Disturbance observer,
change, renewable energy systems are prioritized over the Sliding mode control, Disturbance compensation,
conventional energy systems. The deep injection of Frequency control
renewables into the power systems is creating several
challenges to the grid due to wide variations in their output
power depending on the time of the day, weather etc. Of 1 Introduction
these challenges, frequency change plays a vital role in
maintaining the power quality. This paper presents a novel Renewable energy systems have become the cog in the
sliding mode controller with non-linear disturbance wheel of power systems due to the outcry on global
observer to effectively mitigate the wide changes in the warming. Also, most of the emerging countries are trying
frequency. A sliding mode surface based on estimated to shift towards green energy systems by changing their
disturbance along with states is designed. A sliding mode policies towards carbon negativity. Among those renew-
control law is proposed to compensate disturbances ables, wind and solar plants are mushrooming due to their
including variations in renewables, load and parameters availability. But the major disadvantage of the wind and
under mismatched uncertainties. The proposed observer solar energy is their stochastic nature as they depend on the
based controller is tested for three area multi-machine weather conditions. This poses a serious threat to power
power system in MATLAB/Simulink. The simulated system stability. Increased penetration of wind energy
results proved to alleviate the frequency variations effec- leads to increased rate of variation of net load demand on
tively compared to the conventional controllers. the grid with reduced operational stability [1]. Large scale
integration of wind energy may lead to increase in inter-
area oscillations [2]. When these sources are synchronized
CrossCheck date: 2 November 2017
with thermal power generation stations, there are many
Received: 3 November 2016 / Accepted: 2 November 2017 / Pub- issues arising such as increased frequency fluctuations,
lished online: 9 January 2018 increased losses and reduced operating efficiency.
Ó The Author(s) 2018. This article is an open access publication
Automatic generation control (AGC) is used to balance
& Ayyarao S.L.V. TUMMALA the power and counteract the frequency variations. Power
[email protected]
system security and stability greatly depends on the
Ravikiran INAPAKURTHI effectiveness of AGCs in the thermal power plants to
[email protected]
counteract the power imbalance [3]. There are several
P. V. RAMANARAO traditional methods that address the load frequency control
[email protected]
such as integral, proportional-integral (PI) and propor-
1
Department of Electrical and Electronic Engineering, GMR tional-integral-differential (PID) controllers. The perfor-
Institute of Technology, Rajam, Andhra Pradesh, India mance of these controllers greatly depends on the
2
Department of Electrical and Electronic Engineering, controller parameter values [4]. The major issue with these
University College of Engineering and Technology, ANU, controllers is disturbance attenuation or compensation.
Guntur, India

123
474 A.S.L.V. TUMMALA et al.

Other issues are conventional controllers cannot handle 2 System configuration under study
parameter variations, load variations and unmodelled
dynamics. Frequency regulation by controlling load side The system under study is a three area multi-machine
demand is proposed in [5]. But this method is not viable for power system as shown in Fig. 1. Area-1 consists of ther-
large power systems with multiple load centers and gen- mal power generator synchronized with wind and solar
erating units. Takagi-Sugeno fuzzy model of power sys- power generators which are connected to load center as
tems is considered for the design of fuzzy based frequency shown in Fig. 2. Area-2 and Area-3 consists of thermal
control in [6]. Large disturbances in wind & solar energy, power generators connected to load centers. The intermit-
load and parameter variations has drastic effect on the tent nature of renewable energy sources leads to large
frequency. Thus there is a need to compensate these dis- frequency fluctuations in Area-1. The conventional PI or
turbances. The effect of disturbance can be minimized by PID controllers may not deal with this issue. The system
estimating and compensating the disturbance. Disturbance performance can be improved by compensating the dis-
affecting the system can be estimated using linear distur- turbance. Therefore the objective of this paper is to design
bance observer, extended state observer [7], nonlinear a frequency controller for Area-1 that can stabilize the
disturbance observer etc. For matched uncertainty, distur- overall system even under multiple disturbances.
bances can be estimated and are compensated directly The power system is a more complex and distributed
through control input [8]. When the case of mismatched system. It displays highly non-linear behavior due to con-
uncertainties arise, the direct compensation through control stantly changing dynamics in the systems. This becomes
input is not effective. more profound when parameter changes are also to be
Sliding mode control is one of the robust control tools considered. In conjunction with the load-demand, the
that can handle the various disturbances on the system. generation is also increasing rapidly. This paved a way for
Because of the advantages like insensitivity to parameter exhaustive integration of renewables into the system. This
variations, external disturbances, unmodelled dynamics leads to large disturbances owing to the stochastic nature of
and finite time convergence of sliding surface, the appli- the renewables. Conventionally, the operating point may be
cations of sliding mode control are not limited to speed preserved by forecasting the renewable energy. But, the
control of PMSM [9], sensorless speed control of induction validity of the intraday forecast itself is questionable due to
motor [10], missile control [11] etc. A normal sliding mode intermittent weather conditions.
control cannot effectively handle mismatched uncertainty. Generally, the power system is linearized at an operating
Sliding mode control is used for frequency control of point and the controller is designed so that the system oper-
thermal and hydro power plants [12]. The effect of ates at that point. If the system deviates far from the operating
parameter variations and the effect of renewables are not point, severe non-linearities are introduced and the conven-
discussed. Sliding mode frequency control for multi-ma- tional controller may lose its control over the system.
chine power systems with matched uncertainty and mis- In this paper, a linearized model of power system for ith
matched uncertainty was proposed in [13]. A disturbance area with thermal, wind and solar energy systems is
observer based sliding mode control (DO-SMC) for fre- considered.
quency regulation is proposed in [14, 15]. But this method
requires two functioning controllers. One controller is Area-1 Area-2
integral controller for handling steady state conditions and
the other is sliding mode control for compensating distur- Area-3

bances. In this paper a novel sliding mode surface, as a

Fig. 1 Hybrid energy system
function of states and estimated disturbance, is proposed
for frequency control. The disturbance in the system is
estimated using nonlinear disturbance observer. This pro- Area-1
posed controller serves dual function of controlling the
system under normal conditions as well as compensating Wind farm +
the disturbance.
The paper is organized as follows. Section 2 discusses Thermal power station +

the configuration of hybrid energy system under study.

Solar power + Load center
Section 3 outlines the procedure for nonlinear disturbance
observer design and the design of sliding mode controller.
Tie line power +
Disturbance observer and sliding mode controller are
designed for the proposed system in Section 4. Section 5
analyzes the simulated results for the proposed system. Fig. 2 Area-1 with renewable energy sources

123
Observer based sliding mode frequency control for multi-machine power systems with high 475

1 KPi KPi Assumption 2 The net disturbance on the system is

Df_i ðtÞ ¼  Dfi ðtÞ þ DPgi ðtÞ  DPdi ðtÞ
TPi TPi TPi assumed to be bounded i.e. d\dm , where dm is a scalar.
KPi X  
 Ksij Ddi ðtÞ  Ddj ðtÞ 3.1 Control structure
2pTPi j2N ð1Þ
j6¼i

þ KTPiPi DPwi ðtÞ þ KTPiPi DPsi ðtÞ þ w The proposed control structure is shown in Fig. 3. It
consists of a disturbance observer that estimates the
lumped disturbance on the system based on the states and
1 1
DP_ gi ðtÞ ¼  DPgi ðtÞ þ DXgi ðtÞ ð2Þ the control law. The estimated disturbance is given to the
TTi TTi sliding-mode controller which compensates the disturbance
1 1 1 and controls the states.
DX_ gi ðtÞ ¼  Dfi ðtÞ  DXgi ðtÞ  ui ðtÞ ð3Þ
Ri TGi TGi TGi
3.2 Disturbance observer design
where Dfi is the change in frequency; DPgi is the change in
generator power output; DPdi is the change in load demand;
Initially, a disturbance observer is designed to com-
DPwi is the change in wind power input; DPsi is the change
pensate the disturbances in the system. The structure of
in solar power input; KPi is power system gain; TPi is the
disturbance observer is shown in Fig. 4.
power system time constant; TTi is the turbine time con-
stant; DXgi is the change in governor valve position; TGi is b_ ¼  GB2 ðb þ GxÞ  GðAx þ B1 uÞ ð8Þ
the governor time constant; Ri is the speed regulation
d^ ¼ b þ Gx ð9Þ
coefficient; ui is the control input; Ddi is the rotor angle
deviation; Ksij is the interconnection gain; w denotes where d^ is the estimated disturbance; b is the virtual state;
unmodelled dynamics and change in parameter values. G is disturbance observer gain to be designed.
All the load variations, unmodelled dynamics, parameter Taking the derivative of (9)
variations and renewable energy changes are treated as _
lumped disturbance on the system. d^ ¼  GB2 ðd^  dÞ ð10Þ
Accordingly (1) is modified as: If GB2 [ 0, then
1 KPi
Df_i ðtÞ ¼  Dfi ðtÞ þ DPgi ðtÞ þ di ð4Þ lim ðd^  dÞ ¼ 0 ð11Þ
TPi TPi t!1

where di denotes the lumped disturbance on the system, is 3.3 Design of sliding mode controller
expressed as:
KPi KPi X   The design of sliding mode controller involves two
di ¼  DPdi ðtÞ  Ksij Ddi ðtÞ  Ddj ðtÞ steps. The first being the design of sliding surface and the
TPi 2pTPi j2N
j6¼i
KPi KPi
þ DPwi ðtÞ þ DPsi ðtÞ þ w ð5Þ u x
TPi TPi Plant

3 Disturbance observer based sliding mode

Sliding mode controller d̂ Disturbance observer
control

Let us consider a dynamic system in the form of

Fig. 3 Disturbance observer
x_ ¼ Ax þ B1 u þ B2 dðx; t; wÞ ð6Þ
y ¼ Cx ð7Þ
u x
Plant
where x 2 Rn is the state vector; u 2 Rm is the control x=Ax+B1u+B2d(x,t,w)
d
vector; dðx; t; wÞ is the net disturbance on the system; y is
output vector; B1 ¼ ½ 0 b T ; B2 ¼ ½ 1 0 T ; d̂
  ∫ β =−GB2(β +Gx)−G(Ax+B1u) d̂ =β +Gx
0 1
C ¼ ½ 1 0 T ; A ¼ .
a1 a2
Disturbance observer
Assumption 1 This system comes under the case of mis-
matched certainty. This is because B1 ¼
6 kB2 , where k is a scalar. Fig. 4 Sliding mode control

123
476 A.S.L.V. TUMMALA et al.

Sliding surface
x
B1 ¼ ½ 0 0 13:21 T
ρ =x2+kx1+dˆ
d̂
B2 ¼ ½ 1 0 0 T
Sliding mode controller
ρ
C ¼ ½1 0 0
x u=−b−1(a1x1+a2x2+kx2+kd+
ˆ γ sign(ρ ))
d̂
To plant Transforming the system represented in (15) using the
transformation matrix Tc such that the new state vector
 T
Fig. 5 Sliding mode controller n ¼ T c x, where T c ¼ C CA CA2 , n ¼ ½ n1 n2 n3 T .
The new modified system obtained is given below.
second is the design of control input as shown in Fig. 5. A
novel sliding surface based on the states and disturbance n_ 1 ¼ n2 þ d ð16Þ
estimation is formulated as:
n_ 2 ¼ n3  0:05d ð17Þ
q ¼ x2 þ kx1 þ d^ ð12Þ
n_ 3 ¼ 124:71n1  46:03n2  16:54n3  275:19u
where k [ 0 is the parameter to be designed. þ 0:002d ð18Þ
Sliding mode control input is designed as:
  The above system is represented as:
u ¼  b1 a1 x1 þ a2 x2 þ kx2 þ kd^ þ csignðqÞ ð13Þ
n_ ¼ AC n þ BC u þ DC d ð19Þ
Differentiating (12) leads to
where AC ¼ T c AT 1
c ; BC ¼ T c B1 ; DC ¼ T c B2 .
_
q_ ¼ a1 x1 þ a2 x2 þ bu þ kðx2 þ dÞ þ d^ ð14Þ There exists
2 3
0 1 0
AC ¼ 4 0 0 1 5
4 Controller design for proposed system
124:71 46:03 16:54
The above mentioned concept of control can be used for BC ¼ ½ 0 0 275:19 T
design of decentralized frequency regulator in multi-ma-
chine power systems. The system (see (2)–(4)) is repre- DC ¼ ½ 1 0:05 0:002 T
sented in dynamic state space form as: The objective of this paper is to design control input that

x_ ¼ Ax þ B1 u þ B2 d stabilizes the system operating frequency by counteracting
ð15Þ
y ¼ Cx the disturbances.
The lumped disturbance of the system is estimated using
T
where x ¼ ½ Dfi DPgi DXgi  nonlinear disturbance observer as given in [16]. The
2 3 dynamics of the disturbance observer for estimating the
1 KPi
 0 disturbance on the system are given below.
6 TPi TPi 7
6 1 1 7
6 7 b_ ¼  GDC ðb þ GnÞ  GðAC n þ BC uÞ ð20Þ
A¼6 0  7
6 TTi TTi 7
4 1 5

1
0  d^ ¼ b þ Gn ð21Þ
Ri TGi TGi
 T The matrix G is to be designed such that the eigenvalues
1 of GBC lie on the left half of s-plane.
B1 ¼ 0 0 
TGi To address the uncertainties including variations in wind
and solar energy generation, the disturbance observer based
B2 ¼ ½ 1 0 0 T
sliding surface is formulated as:
The parameters of the system under study are given in  
q ¼ k1 n1 þ k2 n2 þ d^ þ k3 n3 ð22Þ
Appendix A. As discussed earlier, the controller is
designed for Area-1 where cumulative disturbance is where k1, k2, k3 are constants to be selected such that the
maximum. The matrices for Area-1 are given below. roots of (23) lie on the left half of s-plane.
2 3
0:050 6 0 f ðsÞ ¼ k3 s2 þ k2 s þ k1 ð23Þ
A¼4 0 3:472 3:472 5
5:878 0 13:021 A control input is designed to stabilize the system
operating frequency.

123
Observer based sliding mode frequency control for multi-machine power systems with high 477

 
u ¼ 0:003 124:71n1  46:03n2  16:54n3 þ 0:002d^
     

þ 0:003 k31 k1 n2 þ d^ þ k2 n3  0:05d^ þ csignðqÞ

ð24Þ
1) Stability proof
Let us consider the Lyapunov function as:
1
V ¼ q2 ð25Þ
2
Taking the derivative of (25), there exists
V_ ¼ qq_ ¼ cjqj þ ðk1  0:05k2 þ 0:002k3 þ GDC Þ
 
 d^  d q   ½c  ðk1  0:05k2 þ 0:002k3
 
þGDC Þ dm  d^ jqj ð26Þ
  Fig. 6 Actual and estimated disturbance
If c [ ðk1  0:05k2 þ 0:002k3 þ GDC Þ dm  d^ , then
the stability of the sliding mode control can be
guaranteed.
disturbance observer is able to track the actual dis-
2) Remarks turbance in 0.4 s. Due to the variation in estimated and
actual disturbance along with the states, the sliding
a) In the absence of disturbances and parameter surface is also seen to be non-zero. The sliding surface
variations, the function of proposed sliding mode q reaches zero when the states and disturbance tracks
control is similar to conventional sliding mode the true value as shown in Fig. 7. The control input to
control. the system is shown in Fig. 8. In Fig. 9, the effec-
b) Even for large variation in actual and estimated tiveness of regulating the frequency with sliding mode
disturbance, from (10) and (11), the estimated control is compared with conventional PI controller. It
disturbance tracks the true value in finite time. can be observed that with PI controller, the change in
c) It can handle both steady state and transient frequency has an undershoot of - 0.25 Hz. But with
conditions. DO-SMC, the undershoot of change in frequency
reduced significantly to - 0.065 Hz. Also the settling
time in the proposed control is very small compared to
5 Simulation results the conventional PI controller.
From this we can infer that the performance of DO-
The proposed controller is tested on a three-area multi-
machine power system discussed in Section 2 in a simu-
lated environment using MATLAB/Simulink ver.2009a
with Ode4 solver and a sample time of 1 ms. The proposed
controller is applied for Area-1 and PI controllers are used
in the other two areas.
The following values are selected for the design of the
controller and non-linear disturbance observer: G ¼
½ 10 0 0:01 , k1 ¼ 575, k2 ¼ 48, k3 ¼ 1, c ¼ 100.

1) Case 1: effect of load disturbance

In this case, only load disturbance is considered
without wind and solar power input to the system.
Load is suddenly increased by 0.1 p.u. With nonlinear
disturbance observer, the net disturbance is estimated
and the true value of disturbance is tracked within a
short period though initial condition of the observer is
different from the actual and the same is shown in Fig. 7 Sliding surface for load disturbance
Fig. 6. From the figure it can be observed that the

123
478 A.S.L.V. TUMMALA et al.

Fig. 8 Control input Fig. 10 Change in frequency with GRC

4) Case 4: effect of wind variations

In this case, wind generation is modeled as a Gaussian
noise with 0.1 p.u. mean and variance of 10-4 to
represent the physical system. The wind input is
shown in Fig. 12. The continuous variation in wind
generation leads to large fluctuations in operating
frequency with conventional controller. The applied
nonlinear disturbance observer tracks even fast chang-
ing disturbances and is shown in Fig. 13. Since the
sliding surface q is a function of disturbance as well as
states, it changes continuously with the disturbance as
shown in Fig. 14. Frequency distortions are drastically
reduced with sliding mode controller when compared
to PI controller as shown in Fig. 15. The correspond-
ing control input with variable wind generation is
Fig. 9 Change in frequency shown in Fig. 16.
5) Case 5: effect of multiple disturbances
In this case, the system is inflicted with wind, solar and
SMC is superior in terms of peak overshoot, settling load disturbances leading to more severity. Similar to
time and disturbance rejection. wind, solar generation is also modeled as a Gaussian
2) Case 2: effect of generation rate constraint (GRC)
To represent practical power systems, GRC is included
in the simulation model [17, 18]. Figure 10 shows the
change in frequency for a 5% step load change with a
GRC value of 0.005 p.u.MW/s. However, due to GRC,
the settling time is increased.
3) Case 3: effect of governor dead band and time delay
Further, the performance of the proposed controller is
evaluated by inserting other nonlinearities like gover-
nor dead-band and actuator time delay along with
GRC in the model [17, 18]. The change in frequency
with GRC of 0.005 p.u.MW/s, governor dead band
(GDB) as 5% and delay time as 0.05 s is shown in
Fig. 11. The proposed controller is robust to above
said non-linearities.
Fig. 11 Change in frequency with GRC, GDB and time delay

123
Observer based sliding mode frequency control for multi-machine power systems with high 479

Fig. 12 Wind input

Fig. 15 Change in frequency for wind variations

Fig. 13 Actual and estimated disturbance for wind variations

Fig. 16 Control input for wind variations

Figs. 18, 19 and 20 respectively. The frequency

deviation with sliding mode control is slightly high
initially because of the deviation of estimated distur-
bance. When disturbance estimated tracks the true
value, the frequency deviation is drastically reduced.
This shows the superior performance of the proposed
sliding mode controller.

Fig. 14 Sliding surface for wind variations

noise with mean of 0.01 p.u. and variance of 10-6.

This solar model injects low amplitude, high fre-
quency disturbances into the system. Along with this
the load is suddenly reduced by 5% which will add to
the disturbance created by the renewable generation.
The disturbance observer based estimated lumped
disturbance is shown in Fig. 17. Sliding surface,
control input and change in frequency are shown in
Fig. 17 Actual and estimated disturbance for multiple disturbances

123
480 A.S.L.V. TUMMALA et al.

Fig. 21 Change in frequency with and without parameter variations

Fig. 18 Sliding surface for multiple disturbances

proposed controller for parameter variations, the

system is simulated under time varying parameters.
A time varying parameter in the function of cos(t) are
introduced in (4) and this is represented in (27).
 
1
Df_i ðtÞ ¼  þ cosðtÞ Dfi ðtÞ
TPi 
KPi
þ þ cosð1:5tÞ DPgi ðtÞ þ di ð27Þ
TPi
The simulation results show that frequency changes with
and without parameter variations are similar as shown in
Fig. 21. Thus frequency change with parameter variation is
negligible with the proposed sliding mode control.

Fig. 19 Control input for multiple disturbances

6 Conclusion

A non-linear observer is designed for estimating the

lumped disturbance on the system. A control law for
compensating the lumped disturbance based upon a novel
sliding surface for frequency regulation in power system is
proposed in this paper. The proposed system is tested on a
three area multi-machine power system under various
operating conditions. The sliding surface varies with the
estimated disturbance facilitating limited variation in fre-
quency. The results for all the cases, renewable distur-
bance, non-linearities (GRC, GDB and time delay), load
perturbations and parameter variations, are shown to be
superior compared to the conventional controllers. The
proposed control law can be extended to large scale power
Fig. 20 Change in frequency for multiple disturbances systems with FACTS devices.

Open Access This article is distributed under the terms of the

6) Case 6: effect of parameter variations Creative Commons Attribution 4.0 International License (http://
Most of the practical systems are subjected to param- creativecommons.org/licenses/by/4.0/), which permits unrestricted
eter variations and this in turn affects the overall use, distribution, and reproduction in any medium, provided you give
appropriate credit to the original author(s) and the source, provide a
performance of the system. Thus there is a need to link to the Creative Commons license, and indicate if changes were
compensate the parameter variations. To examine the made.

123
Observer based sliding mode frequency control for multi-machine power systems with high 481

Appendix A [11] Zhou J, Yang J (2015) Smooth sliding mode control for missile
interception with finite-time convergence. J Guid Control Dyn
38:1311–1318
The parameters of three area multi-machine power [12] Vrdoljak K, Perić N, Petrović I (2010) Sliding mode based load-
systems is given in Table A1 [13]. frequency control in power systems. Electr Power Syst Res
80(5):514–527
[13] Mi Y, Fu Y, Wang C et al (2013) Decentralized sliding mode
Table A1 Parameters of three area multi-machine power systems load frequency control for multi-area power systems. IEEE
Trans Power Syst 28(4):4301–4309
Area TPi KPi TTi TGi Ri KEi KBi KSij [14] Mi Y, Fu Y, Li D et al (2016) The sliding mode load frequency
control for hybrid power system based on disturbance observer.
1 20 120.0 0.30 0.080 2.4 10.0 0.41 0.550 Int J Electr Power Energy Syst 74:446–452
2 25 112.5 0.33 0.072 2.7 9.0 0.37 0.650 [15] Wang C, Mi Y, Fu Y et al (2016) Frequency control of an
3 20 115.0 0.35 0.070 2.5 7.1 0.40 0.545 isolated micro-grid using double sliding mode controllers and
disturbance observer. IEEE Trans Smart Grid. https://doi.org/10.
1109/TSG.2016.2571439
[16] Yang J, Li S, Yu X (2013) Sliding-mode control for systems
with mismatched uncertainties via a disturbance observer. IEEE
Trans Ind Electron 60(1):160–169
[17] Jiang L, Yao W, Wu QH et al (2012) Delay-dependent stability
References for load frequency control with constant and time-varying
delays. IEEE Trans Power Syst 27(2):932–941
[1] Yuan X (2013) Overview of problems in large-scale wind [18] Zhang CK, Jiang L, Wu QH et al (2013) Delay-dependent
integrations. J Mod Power Syst Clean Energy 1(1):22–25 robust load frequency control for time delay power systems.
[2] He P, Wen F, Ledwich G et al (2013) Small signal stability IEEE Trans Power Syst 28(3):2192–2201
analysis of power systems with high penetration of wind power.
J Mod Power Syst Clean Energy 1(3):241–248
[3] He P, Wen F, Ledwich G et al (2013) Effects of various power Ayyarao S.L.V. TUMMALA received B.Tech and M.Tech in
system stabilizers on improving power system dynamic per- electrical and electronics engineering from JNTU, Andhra Pradesh.
formance. Int J Electr Power Energy Syst 46(1):175–183 He is pursuing his Ph.D. in ANU, India. He is presently working as
[4] Gozde H, Taplamacioglu MC (2011) Automatic generation assistant professor in Department of Electrical and Electronics
control application with craziness based particle swarm opti- Engineering, GMR Institute of Technology, Rajam. His research
mization in a thermal power system. Int J Electr Power Energy interests are application of optimization techniques to power systems,
Syst 33(1):8–16 dynamic state estimation and control of power systems.
[5] Molina-Garcia A, Bouffard F, Kirschen DS (2011) Decentral-
ized demand-side contribution to primary frequency control. Ravikiran INAPAKURTHI received M.Tech in power electronics
IEEE Trans Power Syst 26(1):411–419 from National Institute of Technology, Calicut and received his
[6] Lee H, Park J, Joo Y (2006) Robust load-frequency control for B.Tech from JNT University, Hyderabad. He is presently working as
uncertain nonlinear power systems: a fuzzy logic approach. Inf an assistant professor in the Department of Electrical and Electronics
Sci 176(23):3520–3537 Engineering, GMR Institute of Technology. His research interests are
[7] Li S, Yang J, Chen WH et al (2012) Generalized extended state power electronic converters, smart grid, power system stabilizers and
observer based control for systems with mismatched uncer- control engineering.
tainties. IEEE Trans Ind Electron 59(12):4792–4802
[8] Shtessel Y, Edwards C, Fridman L et al (2014) Sliding mode P.V. Ramanarao is working as professor and head of department in
control and observation. Birkhäuser, New York Electrical and Electronics Engineering in University College of
[9] Zhang X, Sun L, Zhao K et al (2013) Nonlinear speed control Engineering, ANU, Guntur. He has guided several research scholars
for PMSM system using sliding-mode control and disturbance in diverse fields of power systems. His research interests include
compensation techniques. IEEE Trans Power Electron power system stabilizers, DFIG controller for wind power generation.
28(3):1358–1365
[10] Gennaro SD, Dominguez JR, Meza MA (2014) Sensorless high
order sliding mode control of induction motors with core loss.
IEEE Trans Ind Electron 61(6):2678–2689

123