www.ietdl.

org
Published in IET Control Theory and Applications
Received on 2nd March 2014
Revised on 23rd August 2014
Accepted on 9th September 2014
doi: 10.1049/iet-cta.2014.0220

Special Issue on Sliding-Mode Based

Disturbance Estimation, Attenuation and
Fault Detection
ISSN 1751-8644

Extended state observer-based sliding mode control

for PWM-based DC–DC buck power converter
systems with mismatched disturbances
Junxiao Wang, Shihua Li, JunYang, Bin Wu, Qi Li
Key Laboratory of Measurement and Control of Complex Systems of Engineering, Ministry of Education, School of
Automation, Southeast University, Nanjing, Jiangsu 210096, People’s Republic of China
E-mail: [email protected]

Abstract: This study investigates an extended state observer (ESO)-based sliding mode control (SMC) approach for pulse-
width modulation-based DC–DC buck converter systems subject to mismatched disturbances. By designing a novel sliding-
mode manifold incorporated with a disturbance estimation technique, an ESO-based SMC method is designed to deal with
mismatched disturbances. A rigorous stability analysis is also presented. As compared with the nominal SMC method, the
proposed method obtains a better disturbance rejection ability even the disturbances do not satisfy the so-called matching
condition. Simulations and experimental comparison results are implemented to verify the effectiveness of the proposed control
method.

1 Introduction control [25] etc. These methods can improve the perfor-
mance of DC–DC switched converter systems in different
Modern electronic systems require high-quality, reliable and aspects.
efficient power supplies. Hence DC–DC power converters The higher and higher requirement on product precision
are widely used in industrial systems such as DC motor has motivated the work of various researchers to improve the
drives, computer systems, communication equipments and control performance by designing the controller directly on
automotive systems etc. [1–3]. It is well known that DC–DC the basis of the average model in the last decades. Recently,
switched mode power converters are variable-structure non- owing to the undesirable influences of external disturbances
linear and time-varying systems [4–6]. Firstly, the parameter and model parameter uncertainties, the researchers have
uncertainties of the system models are inevitable from been raised an increasing deal of interest in disturbance esti-
the modelling errors such as the magnetic characteristics mate methods. Disturbance observer-based control (DOBC)
of inductance are actually uncertain and non-linear espe- has been proved to be effective in reducing the effects of
cially in the presence of large magnetic flux density in unknown external disturbances and model uncertainties in
the ferromagnetic core in the circuit as studied in [7, 8]. two-link robotic manipulator systems by compensating it
Secondly, disturbance from variation of the load, change from the feedforward channel [21, 26]. The major advan-
of input voltage and electro-magnetic interference which is tage of the DOBC is that the disturbance reject ability
produced by the switching actions of semiconductor such of closed-loop system is improved without sacrificing its
as switch transistors, diode, variable frequency transformer nominal control performance [27–29]. It could completely
etc. [9–11]. Thus, to obtain a satisfactory control perfor- remove the non-vanishing disturbances from system as long
mance for these systems, a high-performance controller of as they can be estimated accurately [30]. Another kind of
DC–DC switched mode converter systems is required to disturbance estimate technique is extended state observer
have nice disturbance rejection ability, small steady-state (ESO) which regards the lumped disturbances as a new state
error, fast dynamical response and low overshoot. So non- of system. It can observe both system states and distur-
linear control method is regarded as a natural consideration. bances by simple computations, then the observed value of
Many kinds of non-linear control methods have been pro- disturbances can be employed to compensate in the feedfor-
posed on the basis of the average system model, such as ward channel to improve the performance of system [31].
passivity control [2], input/output feedback linearisation, Owing to such a promising feature, an ESO-based con-
backstepping control approach [3, 7], linear matrix inequal- trol frame also called active disturbance rejective control
ity (LMI)-based control [12], adaptive control [13], robust has been developed [32, 33]. This method has also been
control [14], sliding-mode control (SMC) [11, 13, 15–23], applied to permanent magnet synchronous motor (PMSM)
optimal control [8], predictive control [24], fuzzy logic servo systems [34–36].

IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586 579
doi: 10.1049/iet-cta.2014.0220 © The Institution of Engineering and Technology 2015
www.ietdl.org
However, most of the existing DOBC methods are only
insensitive to matched disturbances but sensitive to mis-
matched ones. In many practical systems, the disturbances
would not rigorously satisfy the so-called matching condi-
tion which implies that the disturbances act via the same
channel as the control input, such as MAGLEV suspension
vehicle, PMSM system, power electronics system and so
on [37]. The output voltage performance of DC–DC buck
system is also influenced by the mismatched uncertain load
in [20, 38]. The authors in [39] conclude that the SMC
input signal is actually load dependent for DC–DC system. a
If the controller wants to have good regulation performance
against the load changes, the instantaneous load value should
be fed back. However, this would require additional sensors
and cumbersome computations, which makes the controller
complicated, the aforementioned design of SMC method
does not take into account the load resistance disturbances,
so it is necessary to design a controller to deal with the load
resistance disturbances for DC–DC buck system.
Owing to the importance of compensating mismatched
disturbances in both theory and engineering applications,
serval researchers have been paid attention in solving such
a problem. In [40], offset caused by mismatched distur- b
bances is removed in the context of model predictive con-
trol by correcting the prediction error via a disturbance
observer. An equivalent input disturbance-based control
scheme is proposed for mismatched disturbance attenuation
in [41, 42]. By designing a specific disturbance compen-
sation gain, a new DOBC framework was proposed to
compensate mismatched disturbances in [43] for MAGLEV
suspension vehicle system. Sliding-mode controller based on
disturbance observer is designed to deal with mismatched
uncertainties/disturbances for various industrial systems in
[37, 44–48].
In this paper, aiming to improve the performance c
of the controlled DC–DC buck converter system with
mismatched/matched disturbances, an ESO-based SMC Fig. 1 Average model circuit of buck converter
method is developed to completely counteract the mis- a Circuit of buck converter
matched/matched disturbances for DC–DC system. By fully b Case when the switch is on
taking into account the estimation value of disturbances, c Case when the switch is off
a dynamic sliding-mode surface is firstly designed which
is insensitive to not only matched disturbances but also 2 Model description of DC–DC buck power
mismatched ones. As a result, the DC–DC buck converter converter system
system output voltage can be driven to the reference sig-
nal asymptotically by sliding motion along the sliding-mode A typical PWM-based DC–DC buck power converter struc-
surface. Note that the sliding-mode surface here is specific ture is shown in Fig. 1, where iL is the inductance current,
since an estimate of the mismatched disturbances based on υs the average output capacitor voltage, υr the reference out-
ESO is introduced. In addition, considering the typical dis- put voltage, L the circuit inductance, C the circuit capacitor,
advantages of having variable switching frequency operation R the circuit load resistance, E the voltage of the external
in hysteresis modulation-based controller, this paper adopts source and μ ∈ [0, 1] the duty ratio function taken as the
the fixed-frequency pulse-width modulation (PWM)-based control signal of PWM [2, 3]. In practice, the load resistance
controller which is easy to implement by power electronics R in the circuit may continually change subject to uncer-
engineers. Moreover, a better convergence performance and tain outer circumstances such as temperature etc. [39]. The
a stronger disturbance rejection ability against mismatched nominal value of R is assumed to be R0 .
resistance load disturbances are achieved with the proposed The converter dynamic model is deduced as follows.
control scheme, which has been shown by simulation and When the switch is on
experimental results.
The paper is organised as follows. Section 2 gives the sys- E = Li̇L + υs
tem model description. Details on the problem of nominal υs
C υ̇s = iL − (1)
sliding-mode controller design with mismatched/matched R
disturbances are discussed in the later section. The dis-
turbance estimation and the new sliding-mode controller When the switch is of
are designed in Section 4. In Section 5, numerical simu-
Li̇L = −υs
lation and experimental results will show the effectiveness
of the proposed controller. Then the conclusion ends the υs
C υ̇s = iL − (2)
paper. R

580 IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586
© The Institution of Engineering and Technology 2015 doi: 10.1049/iet-cta.2014.0220
 www.ietdl.org
The average system model is

Li̇L = μE − υs
υs
C υ̇s = iL − (3)
R
Let x1 = e = υs − υr , then the model can be rewritten as
follows
iL υs
ẋ1 = ė = − + d1 (t) (4)
C R0 C
where d1 (t) = −[υs /RC] + [υs /R0 C]. Let x2 = [iL /C] −
[υs /R0 C], we have
 
(μE − υs ) 1 iL υs Fig. 2 Control system structure of buck converter
ẋ2 = − − (5)
LC R0 C C RC
ESO-based SMC method, it is an essential reason why the
The average model could be simply deduced as nominal SMC design is only insensitive to matched distur-
bance but sensitive to mismatched disturbances. In addition,
ẋ1 = x2 + d1 (t) the integral sliding-mode surface can deal with mismatched
  disturbance, but this method always brings some adverse
(μE − υr ) x1 1 iL υs
ẋ2 = − − − (6) effects to the control systems, such as large overshoot and
LC LC R0 C C RC
long settling time [37]. The solution of directly using the
Denote u = [{μE − υr }/LC], d2 (t) = −[1/R0 C]([υs /R0 C] − information of ẋ1 by differentiation of x1 may bring large
[υs /RC]), the model is further reduced to noise and influence the closed-loop performance.

ẋ1 = x2 + d1 (t) 4 Novel sliding-mode controller design

x1 x2
ẋ2 = u − − + d2 (t) (7) In this paper, we focus on the mismatched disturbance rejec-
LC R0 C
tion problem of the DC–DC buck converter system (7). A
novel sliding controller is proposed by the following two
3 Problem of the nominal SMC method steps. First, an ESO is employed to estimate the mismatched
and matched disturbances, respectively. A novel sliding-
Assumption 1: For the above DC–DC system (7), sup- mode controller is then designed for DC–DC buck converter
pose that lumped disturbances d1 (t), d2 (t) are bounded, system based on the disturbance observation.
that is, there exist constants d ∗ > 0, satisfying d ∗ = supt>0 The control structure of DC–DC buck converter system
| kd1 (t) + d2 (t) |. is designed as Fig. 2, where the voltage and current infor-
The sliding-mode surface and control law of the nominal mation can be obtained from sensors, and the control output
SMC for DC–DC system (7) are designed as follows will generate PWM signal.
s = x2 + kx1 (8) 4.1 Extended state observer
x1 x2
u = −[kx2 + η sgn(s)] + + (9) The objective of this section is to design a disturbance
LC R0 C
LCu + υr estimator for observing the disturbances caused by load
μ(t) = (10) resistance changes. On the basis of DC–DC mathematical
E model (7), the mismatched disturbance estimation of d1 (t) is
Taking the derivative of (8), and combining (7)–(9) yields designed based on the ESO technique, which is designed as

ṡ = −η sgn(s) + kd1 (t) + d2 (t) (11) ż1 = z2 + x2 − β1 (z1 − x1 ) (13)

ż2 = −β2 (z1 − x1 ) (14)
It can be concluded that the states of DC–DC system will
reach the nominal sliding-mode surface s = 0 in finite time where the z1 = x̂1 , z2 = d̂1 (t) and β1 > 0, β2 > 0.
if the switching gain η of the controller is selected such that The matched disturbance estimation of d2 (t) is directly
η > d ∗ . Once the nominal sliding surface s = 0 is reached, derived from the d̂1 (t), given by
the sliding motion dynamics can be described as
1
ẋ1 = −kx1 + d1 (t) (12) d̂2 (t) = − d̂1 (t) (15)
R0 C
Remark 1: Equation (12) implies that if the d1 (t) = 0, the Assumption 2: For the above DC–DC system (7), suppose
system output can track the reference signal, but the sys- the lumped disturbances ḋ1 (t), ḋ2 (t) are bounded and satisfy
tem states cannot be driven to the desired equilibrium point limt→0 ḋ1 (t) = 0, limt→0 ḋ2 (t) = 0.
because of the existence of mismatched disturbances d1 (t)  =
0 although the control law can force the DC–DC system Lemma 1 [49]: Consider a non-linear system ẋ = F(x, w)
states to reach the sliding-mode surface in a finite time. This which is input-to-state stable (ISS). If the inputs satisfy
nominal SMC design is only for the comparison of proposed limt→0 w(t) = 0, then the states satisfy limt→0 x(t) = 0.

IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586 581
doi: 10.1049/iet-cta.2014.0220 © The Institution of Engineering and Technology 2015
www.ietdl.org
For the system of (7), (13)–(15), we give a stability anal- Table 1 Component values of DC–DC buck converter
ysis of ESO, choosing the following Lyapunov function
Descriptions Parameters Nominal values
β2 1
Ve = ex21 + ed21 (16) input voltage υin (E ) 20 (V)
2 2 desired output voltage υr 10 (V)
where β2 > 0, ed1 (t) = d1 (t) − d̂1 (t), ex1 (t) = x1 − z1 , ed2 (t) = inductance L 4.7 (mH)
d2 (t) − d̂2 (t). Taking the time derivative of ed1 , ex1 , it yields capacitance C 1000 (μF)
load resistance R 0 ()–220 ()
ėx1 = ed1 − β1 ex1 (17)
ėd1 = ḋ1 − β2 ex1 (18)
(7), (13)–(15) and (24), choosing the following Lyapunov
By taking the time derivative of Ve , we obtain function
1
V̇e = β2 ex1 ėx1 + ed1 ėd1 (19) V = sT s (25)
2
Taking derivative of V based on (23), we obtain that
Then, by using the above (17) and (18) it yields
˙
V̇e = β2 ex1 (ed1 − β1 ex1 ) + ed1 (ḋ1 − β2 ex1 ) (20) V̇ = sṡ = s(ẋ2 + k ẋ1 + d̂1 (t)) (26)

= −β1 β2 ex21 + β2 ex1 ed1 − β2 ed1 ex1 + ed1 ḋ1 (21) Substituting (24) into (23) yields
= −β1 β2 ex21 + ed1 ḋ1 (22) 
x1 x2
ẋ2 = u − − + d2 (t) = − k(x2 + d̂1 (t))
where β1 β2 > 0. Suppose the DC–DC system satisfies LC R0 C

Assumptions 1 and 2, then the error system of (17) and x1 x2 ˙
(18) is ISS, noting that limt→0 ḋ1 (t) = 0, from Lemma 1, − − + d̂2 (t) + d̂1 (t) + η sgn(s)
LC R0 C
then it concludes that limt→0 ex1 (t) = 0, limt→0 ed1 (t) = x1 x2
0, limt→0 ed2 (t) = 0. This implies that the error system states − − + d2 (t) (27)
of (17) and (18) will converge to the desired equilibrium LC R0 C
point asymptotically. Substituting (27) into (26) yields
Remark 2: For the system stability of (17) and (18), the dis-  
x1 x2 ˙
turbances and system states estimation errors ed1 (t), ed2 (t) V̇ = s − k(x2 + d̂1 (t)) − − + d̂2 (t) + d̂1 (t)
are bounded, there exist a constant l > 0 satisfying l = LC R0 C
supt>0 | ed1 (t) + ed2 (t) |. x1 x2
+ η sgn(s) ] − − + d2 (t)
LC R0 C

4.2 Controller design ˙
+ k(x2 + d1 (t)) + d̂1 (t) ≤| s | (−η + ked1 + ed2 )
A novel sliding-mode manifold for system (7) under mis-
matched disturbance is designed as   12
1 1 T
= 2 2 (−η + ked1 + ed2 ) s s
2
s = x2 (t) + kx1 (t) + d̂1 (t) (23) 1 1
= −2 2 (η − ked1 − ed2 )V 2 (28)
where k > 0.
The proposed ESO-based SMC law is designed as where ed1 = d1 − d̂1 , ed2 = d2 − d̂2 . It follows from (28) and
Assumption 2 that the system states are driven to sliding-
u = −[k(x2 + d̂1 (t)) + η sgn(s)] mode surface s = 0 in finite time when η > l.
x1 x2 ˙ From the above analysis, the system states could reach
+ + − d̂2 (t) − d̂1 (t) the sliding surface s = 0 from any initial condition in finite
LC R0 C
time. Let s = 0, the sliding motion is then described as
LCu + υr
μ(t) = (24)
E s = x2 + kx1 + d̂1 (t) = 0 (29)
where k > 0, η > l, then the output voltage of closed-loop
system can asymptotically track the desired voltage υr . Substituting (7) into (29) gives

Theorem 1: Considering the above system (7) with mis- s = ẋ1 + kx1 + d̂1 (t) − d1 (t) = 0 (30)
matched/matched disturbances, we choose a sliding-mode
controller of the form shown as (13)–(15), (24). Suppose which implies that
the DC–DC system satisfies Assumptions 1 and 1, the μ(t)
control law is designed as (24). If the switching gain is cho- ẋ1 = −kx1 + ed1 (31)
sen such that η > l and k > 0, then the closed-loop system
of (7), (13)–(15) and (24) is globally asymptotically stable. According to Lemma 1, we know that the system states can
converge to the origin along the sliding surface asymptot-
Proof: The closed-loop system stability analysis here con- ically under the proposed control law (13)–(15) and (24).
sists of the analysis for the DC–DC closed-loop system of This completes the proof. 

582 IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586
© The Institution of Engineering and Technology 2015 doi: 10.1049/iet-cta.2014.0220
 www.ietdl.org
Remark 3: The above proof implies that states of DC–DC
system can be driven to the desired equilibrium point and
the control law can force the DC–DC system states to reach
the sliding-mode surface in finite time. This is the main
reason why the proposed SMC is not only insensitive to
matched disturbances, but also insensitive to mismatched
disturbances.

Fig. 4 Response curves under SMC+ESO controller (24)

and controller (10) when the load resistance is increased
(simulation)
a Output voltage vs
b Current iL
c Duty ratio μ(t)

5 Simulation and experimental studies

Fig. 3 Response curves under SMC+ESO controller (24)
and controller (10) when the load resistance is decreased In what follows, the performances of the proposed
(simulation) SMC+ESO controller are illustrated via both numerical sim-
a Output voltage vs ulations and experiments. The desired output voltage is set
b Current iL to be υr = 10 V, and the parameters in this test are shown
c Duty ratio μ(t) in Table 1.

IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586 583
doi: 10.1049/iet-cta.2014.0220 © The Institution of Engineering and Technology 2015
www.ietdl.org

Fig. 5 Control system configuration of the DC–DC buck con-

verter

Fig. 6 Experimental test system setup of the DC–DC buck con-

verter

5.1 Numerical simulations

c
First, the DC–DC buck converter is numerically simulated
using the average model (7) with mismatched/matched dis-
turbances. In order to show the advantages of the novel
SMC method proposed in this paper to its nominal slid-
ing control counterpart, we will use simulations to compare
the performance between them for the average model (7). To
have a fair comparison, the control signal amplitudes of both
controllers are in the same level, proper gains are carefully
chosen for the controllers (24) and (10).
The parameters in SMC+ESO controller (24) are selected
as k = 10, η = 800, β1 = 400, β2 = 1600, while the param-
eters in (10) are selected as k = 8, η = 2500. In Figs. 3
and 4, it is observed that the two control input amplitudes are
almost in the same level. Figs. 3 and 4 show the closed-loop
performance comparisons under controllers (24) and (10). It d
is shown in Figs. 3 and 4 that the SMC+ESO controller can
Fig. 7 Response curves under SMC+ESO controller (24) and
produce a better convergence rate than that of the nominal
controller (10) when the load resistance is decreased from 100
SMC controller, the chattering of the inductance current is
to 60  (experiment)
smaller than that of the nominal SMC method.
To verify the disturbance rejection ability of the proposed a Output voltage vs
b Current iL
method, simulations are also carried out for SMC for com-
c Duty ratio μ(t) under controller (24)
parisons. Load variation of resistance is considered here, that d Duty ratio μ(t) under controller (10)
is, resistance changes from 100 to 60  at 7 s and from 60 to

584 IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586
© The Institution of Engineering and Technology 2015 doi: 10.1049/iet-cta.2014.0220
 www.ietdl.org
80  at 10 s in Fig. 3 It is observed from Figs. 4a–c and 4a–
c that a better disturbance reject ability of the closed-loop
system is obtained by SMC+ESO controller.

5.2 Experimental results

To verify the mismatched disturbance rejection ability of

the proposed SMC+ESO control method, experimental com-
parison results are presented between controllers (24) and
(10). The configuration and experimental test setup are
shown in Figs. 5 and 6, respectively. The control algo-
rithms in the experimentation are implemented using the
LabView system. The reason for selecting LabView as a
a real-time controller is that it is easy to implement with vari-
ous inputs and outputs, provides easy graphic user interface
modules such as user-made scopes, and comprises rich soft-
ware tools that can be applied in real time. A DAQ control
board from National Instrument is used as a real-time con-
troller. The PWM signal is generated by comparing control
input with the ramp waveform. The DC–DC converter is
operating at continuous conduction mode when selecting
the proper component values. In system (7), the parame-
ters in the SMC+ESO controller (24) are selected as k =
5, η = 2000, β1 = 80, β2 = 640, and the control parameters
for nominal SMC (10) are selected as k = 10, η = 2500.
The response curves in this case are shown in Figs. 7
and 8. The experimental results show that the closed-loop
performance comparison between controllers (24) and (10).
b The disturbance rejection abilities of the closed-loop per-
formance are shown in Figs. 7a, b, 8a and b. The solid
lines represent the response curves under controller (24),
and the dotted lines represent the response curves under
controller (10). It is clear that the SMC+ESO controller
can still ensure strong disturbance rejection abilities using
much smaller switching gains when the mismatched load
resistance changes from 100 to 60  at 10 s in Fig. 7 and
from 100 to 190  at 10 s in Fig. 8, respectively. This
advantage directly guarantees the chattering reduction of the
SMC+ESO method.

c
6 Conclusion
The mismatched/matched disturbance rejection control prob-
lem of the buck power DC–DC converter system has been
studied in this paper. By utilising the disturbance estimation
technique based on ESO, a novel sliding-mode controller has
been developed. The controller not only makes the states of
closed-loop system obtain better tracking performance, but
also provides a better disturbance rejection ability against
resistance load variation.

7 Acknowledments
This work is supported in part by the Graduate Innovation
d Program of Jiangsu Province under grant (KYLX_0138),
National Natural Science Foundation (NNSF) of China
Fig. 8 Response curves under SMC+ESO controller (24) and under grants (61203011), Natural Science Foundation of
controller (10) when the load resistance is increased from 100 Jiangsu Province under grant (BK2012327, BK20130018),
to 190 (experiment) High-level Talents Program in Six Industries of Jiangsu
a Output voltage vs Province (DZXX-30), PhD programmes Foundation of Min-
b Current iL istry of Education of China under grant (20120092120031)
c Duty ratio μ(t) under controller (24) and Thermal Process of Intelligent Control of Jiangsu
d Duty ratio μ(t) under controller (10) Province Key Laboratory (BM2013016).

IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586 585
doi: 10.1049/iet-cta.2014.0220 © The Institution of Engineering and Technology 2015
www.ietdl.org
8 References 25 Cetin, E., Omer, D., Huseyin, S.: ‘Adaptive fuzzy logic controller for
DC–DC converters’, Expert Syst. Appl., 2009, 36, (2), pp. 1540–1548
1 Sira-Ramirez, H., Rios-Bolívar, M.: ‘Sliding mode control of DC– 26 Tan, S.C., Lai, Y.M., Tse, C.K.: ‘Indirect sliding mode control of
DC power converters via extended linearization’, IEEE Trans. Circuits power converters via double integral sliding surface’, IEEE Trans.
Syst., Fundam. Theory Appl., 1994, 41, (10), pp. 652–660 Power Electron., 2008, 23, (2), pp. 600–611
2 Sira-Ramirez, H., Perez-Moreno, R., Ortega, R., Garcia-Esteban, M.: 27 Guo, L., Chen, W.H.: ‘Disturbance attenuation and rejection for sys-
‘Passivity based controllers for the stabilization of DC-to-DC power tems with nonlinearity via DOBC approach’, Int. J. Robust Nonlinear
converters’, Automatica, 1997, 33, (4), pp. 499–513 Control, 2005, 15, (3), pp. 109–125
3 Fadil, E., Giri, F.: ‘Backstepping based control of PWM DC–DC boost 28 Davila, J., Fridman, L., Levant, A.: ‘Second-order sliding-mode
power converters’. IEEE Int. Symp. on Industrial Electronics, 2007, observer for mechanical systems’, IEEE Trans. Autom. Control, 2005,
pp. 395–400 50, (11), pp. 1785–1789
4 Middlebrook, R., Cuk, S.: ‘A general unified approach to modelling 29 Shtessel, Y., Edwards, C., Fridman, L., Levant, A.: ‘Sliding mode
switching converter power stages’, Power Electron. Spec. Conf., 1976, control and observation’ (Springer, 2014)
1, pp. 18–34 30 Chen, W.H., Ballance, D.J., Gawthrop, P.J., Reilly, J.O.: ‘A nonlin-
5 Martin, A., Guillaume, G., Thierry, M.: ‘Implementation of a peak cur- ear disturbance observer for robotic manipulators’, IEEE Trans. Ind.
rent control algorithm within a field-programmable gate array’, IEEE Electron., 2000, 47, (4), pp. 932–938
Trans. Ind. Electron., 2007, 54, (1), pp. 406–418 31 Sun, B., Gao, Z.: ‘A DSP-based active disturbance rejection control
6 Pallab, M., Philip, K., Matthew, G., Matthew, F.: ‘Sensorless current design for a 1-kW H-bridge DC–DC power converter’, IEEE Trans.
mode control an observer-based technique for DC–DC converters’, Ind. Electron., 2007, 52, (5), pp. 1271–1277
Control Eng. Pract., 2001, 16, (4), pp. 522–526 32 Han, J.: ‘From PID to active disturbance rejection control’, IEEE
7 Fadil, E., Giri, F., Magueri, E., Chaoui, H.: ‘Control of DC–DC power Trans. Ind. Electron., 2009, 56, (3), pp. 900–906
converters in the presence of coil magnetic saturation’, Control Eng. 33 Xia, Y., Shi, P., Liu, G.P., Rees, D., Han, J.: ‘Active disturbance
Pract., 2009, 17, pp. 849–862 rejection control for uncertain multivariable systems with time-delay’,
8 Olalla, C., Queinnec, I., Leyva, R., Aroudi, A.E.: ‘Robust optimal IET Control Theory Appl., 2007, 1, (1), pp. 75–81
control of bilinear DC–DC converters’, Control Eng. Pract., 2011, 34 Li, S.H., Zong, K., Liu, H.X.: ‘A composite speed controller based
19, (7), pp. 688–699 on a second order model of permanent magnet synchronous motor
9 Nagrial, M.H., Hellany, A.: ‘EMI/EMC issues in switch mode power system’, Trans. Inst. Meas. Control, 2011, 33, (5), pp. 522–541
supplies(smps)’. Int. Conf. and Exhibition on Electromagnetic Com- 35 Li, S.H., Zhou, M.M., Yu, X.H.: ‘Design and implementation of termi-
patibility, EMC York, 1999, vol. 99, pp. 180–185 nal sliding mode control method for PMSM speed regulation system’,
10 Majid, A., Saleem, J., Bertilsson, K.: ‘EMI filter design for high fre- IEEE Trans. Ind. Inf., 2013, 9, (4), pp. 1879–1891
quency power converters’. 11th Int. Conf. Environment and Electrical 36 Liu, H.X., Li, S.H.: ‘Speed control for PMSM servo system using
Engineering, 2012, pp. 586–589 predictive functional control and extended state observer’, IEEE Trans.
11 Said, O., Guo, L.P.: ‘PWM-based adaptive sliding mode control for Ind. Electron., 2012, 59, (2), pp. 1171–1183
boost DC–DC converters’, IEEE Trans. Ind. Electron., 2013, 60, (8), 37 Yang, J., Li, S.H., Yu, X.H.: ‘Sliding mode control for systems with
pp. 3291–3294 mismatched uncertainties via a disturbance observer’, IEEE Trans. Ind.
12 Olalla, C., Leyva, R., ElAroudi, A., Queinnec, I.: ‘LMI robust control Electron., 2013, 60, (1), pp. 160–169
design for boost PWM converters’, IET Power Electron., 2010, 3, 38 Tan, S.C., Lai, Y.M., Cheung, M.K. et al.: ‘On the practical design of
pp. 75–85 a sliding mode voltage controlled buck converter’, IEEE Trans. Power
13 Tan, S.C., Lai, Y.M., Tse, C.K., Cheung, M.K.: ‘Adaptive feedfor- Electron., 2005, 20, (2), pp. 425–437
ward and feedback control schemes for sliding mode controlled power 39 Tan, S.C., Lai, Y.M., Tse, C.K.: ‘A unified approach to the design of
converters’, IEEE Trans. Power Electron., 2006, 21, (1), pp. 182–192 PWM-based sliding-mode voltage controllers for basic DC–DC con-
14 Zhang, C.L., Wang, J.X., Li, S.H., Wu, B., Qian, C.J.: ‘Robust control verters in continuous conduction mode’, IEEE Trans. Circuits Syst. I,
for PWM-based DC–DC buck power converters with uncertainty via Regul. Pap., 2006, 53, (8), pp. 1816–1827
sampled-data output feedback’, IEEE Trans. Power Electron., 2014, 40 Maeder, U., Morari, M.: ‘Offset free reference tracking with model
30, (1), pp. 504–515 predictive control’, Automatica, 2010, 46, (9), pp. 1469–1476
15 He, Y., Luo, F.L.: ‘Sliding-mode control for DC–DC converters with 41 She, J.H., Fang, M., Ohyama, Y. et al.: ‘Improving disturbance rejec-
constant switching frequency’, IEE Proc. Control Theory Appl., 1998, tion performance based on an equivalent-input-disturbance approach’,
145, (3), pp. 37–45 IEEE Trans. Ind. Electron., 2008, 55, (1), pp. 380–389
16 Jiao, Y., Luo, F.L., Zhu, M.: ‘Generalised modelling and sliding mode 42 She, J.H., Xin, X., Pan, Y.: ‘Equivalent input disturbance approach
control for n-cell cascade super-lift DC–DC converters’, IET Power analysis and application to disturbance rejection in dual-stage feed
Electron., 2011, 4, (5), pp. 532–540 drive control system’, IEEE/ASME Trans. Mechatron., 2011, 16, (2),
17 Tan, S.C., Lai, Y.M., Cheung, M.K.H., Tse, C.K.: ‘A pulse wave mod- pp. 330–340
ulation based sliding mode controller for buck converter’. IEEE Power 43 Li, S.H., Yang, J., Chen, W.H., Chen, X.S.: ‘Generalized extended
Electronics Specialists Conf. (PESC), 2004, pp. 3647–3653 state observer based control for systems with mismatched uncertain-
18 Cortesa, D., Alvarezb, J., Alvarezc, J.: ‘Output feedback and dynami- ties’, IEEE Trans. Ind. Electron., 2012, 59, (2), pp. 4792–4802
cal sliding-mode control of power converters’, Int. J. Electron., 2011, 44 Yang, J., Li, S.H., Su, J.Y., Yu, X.H.: ‘Continuous nonsingular termi-
98, (4), pp. 505–519 nal sliding mode control for systems with mismatched disturbances’,
19 Komurcugil, H.: ‘Non-singular terminal sliding mode control of DC– Automatica, 2013, 49, pp. 2287–2291
DC buck converters’, Control Eng. Pract., 2013, 21, pp. 321–332 45 Estrada, A., Fridman, L.: ‘Quasi-continuous HOSM control for sys-
20 Tan, S.C., Lai, Y.M., Tse, C.K.: ‘General design issues of sliding mode tems with unmatched perturbations’, Automatica, 2010, 46, (11),
controllers in DC–DC converters’, IEEE Trans. Ind. Electron., 2008, pp. 1916–1919
55, pp. 1160–1174 46 Loza, A.F.„ Bejarano, F.J., Fridman, L.: ‘Unmatched uncertainties
21 Li, S.H., Yang, J., Chen, W.H., Chen, X.S.: ‘Disturbance observer- compensation based on high-order sliding mode observation’, Int. J.
based control: methods and applications’ (CRC Press, 2014) Robust Nonlinear Control, 2013, 23, (7), pp. 754–764
22 Tan, S.C., Lai, Y.M., Tse, C.K.: ‘Sliding mode control of switching 47 Ferreira de Loza, A., Punta, E., Fridman, L. et al.: ‘Nested backward
power converters’ (CRC Press, 2011) compensation of unmatched perturbations via HOSM observation’, J.
23 Chiu, C.S., Shen, C.T.: ‘Finite-time control of DC–DC buck converters Franklin Inst., 2014, 351, (5), pp. 2397–2410
via integral terminal sliding mode’, Int. J. Electron., 2012, 99, (5), 48 Ginoya, D., Shendge, P.D., Phadke, S.B.: ‘Sliding mode control
pp. 643–655 for mismatched uncertain systems using an extended disturbance
24 Zhou, G., Xu, J., Jin, Y.: ‘Improved digital peak current predictive observer’, IEEE Trans. Ind. Electron., 2014, 61, (4), pp. 1983–1992
control for switching DC–DC converters’, IET Power Electron., 2011, 49 Khalil, H.K.: ‘Nonlinear systems’ (Prentice-Hall, Upper Saddle River,
4, (2), pp. 227–234 NJ, 1996, 2nd edn.)

586 IET Control Theory Appl., 2015, Vol. 9, Iss. 4, pp. 579–586
© The Institution of Engineering and Technology 2015 doi: 10.1049/iet-cta.2014.0220