applied

sciences
Article
Active Disturbance Compensation Based Robust
Control for Speed Regulation System of Permanent
Magnet Synchronous Motor
Yuxiang Ma and Yunhua Li *
School of Automation Science and Electric Engineering, Beihang University, XueYuan Road No.37, HaiDian
District, Beijing 100191, China; mayuxiang1218@[Link]
* Correspondence: yhli@[Link]; Tel.: +86-10-8233-9038




Received: 20 December 2019; Accepted: 15 January 2020; Published: 19 January 2020


Abstract: This paper deals with the robust control method for permanent magnet synchronous motor
(PMSM) speed-regulation system based on active disturbance compensation. Different from the
classical PMSM disturbance compensation scheme, a novel disturbance feed-forward compensation
based on extended state observer (ESO) is designed for speed loop and q-axis current loop of PMSM.
The disturbances of current loop include unmodeled dynamics of back electromotive force and
parameters variations of stator are considered as lumped disturbance to compensate actively. In this
way, the dynamic response of q-axis current loop can be improved to guarantee the anti-disturbance
ability. A composite controller using sliding mode control and ESO is designed as speed loop controller,
and an ESO-based proportional-integral controller is designed for q-axis current loop. Moreover,
a transition process of reference signal is introduced to replace the step reference signal, which reduces
the initial error and increases the range of feedback gain to improve system robustness. Finally,
simulations and experiments are given to demonstrate the effectiveness of the proposed strategy.

Keywords: feedforward-based control; active compensation; extended state observer; permanent

magnet synchronous motor; q-axis current loop

1. Introduction
As an important power of mechanical system, permanent magnet synchronous motor (PMSM) is
widely used because of the compact structure, high power density, and low noise. Especially in the
field where volume is strictly limited, such as robots, aerospace, high-performance service system,
PMSM is widely accepted. As we know, proportional-integral (PI) control is already used in PMSM
system because of its simple implementation. However, PMSM is a nonlinear, time-varying system and
it is difficult to achieve satisfactory control performance by using PI control. Therefore, it is necessary
to use nonlinear control algorithm in PMSM system [1].
In recent years, many complex control algorithms have been implemented because of the
development of micro controller, e.g., robust control [2–4], fuzzy control [5–7], adaptive control [8–10],
sliding mode control (SMC) [11–13], backsteeping control [14–16], etc. These advanced algorithms
can effectively improve control performance. Among these methods, SMC attracts more attention
because of its robustness. SMC is a discontinuous control method that can force the system move
along the sliding surface. Moreover, sliding motion can be designed and it is not affected by parameter
change and external disturbance. In [17], a SMC speed-regulation controller consisting of a tracking
controller and two synchronous controllers is designed for multi-motor system. In [18], a nonlinear
speed-control algorithm based on sliding mode reaching law (SMRL) is designed for PMSM system
and this algorithm can dynamically adapt to the operation statues. But SMC is a feedback-based

Appl. Sci. 2020, 10, 709; doi:10.3390/app10020709 [Link]/journal/applsci

Appl. Sci. 2020, 10, 709 2 of 13

control, which only works when there is an error. So it cannot react promptly to disturbances such
as load change, unmodeled dynamics, and parameters variations. As an effective method to reject
disturbance, feed-forward compensation is introduced in control system.
Feedforward-based control is a method to compensate disturbance, which can effectively reduce the
impact of disturbance. The premise of compensation is to estimate disturbance, but most disturbances
cannot be measured directly, so the disturbance observer is proposed. Disturbance observer (DOB)
is proposed by Ohnishi [19] in 1987, which is an active estimation method for external disturbance
and has been applied in robotic [20], aerial vehicle control [21], PMSM [22], etc. In [23], a composite
sliding mode controller based on DOB is proposed for PMSM, the disturbance is compensated by
DOB can take a smaller switching gain without sacrificing disturbance rejection. In [24], a disturbance
observer-based sliding mode controller is designed for PMSM control system, a DOB is designed to
reconstruct the external disturbance and counteract the influence of external disturbance to reduce the
velocity resonance caused by disturbance. In [25], an augmented disturbance observer is proposed
by embedding uncertainty/disturbance in the DOB, which can accurately estimate disturbances of
different channels.
Another disturbance observation technique has been widely accepted is extended state observer
(ESO) which is proposed by Han [26]. ESO regards lumped disturbance, which include external
disturbance and internal disturbance, as a new state variable to reconstruct a new extended state
equation, and then estimate the state variable of the extended state equation. ESO does not require
a model of disturbance and does not need to measure disturbances directly, so ESO is a practical
disturbance observer. Disturbance feed-forward compensation based on ESO or active disturbance
rejection control (ADRC) [27] has been widely applied in PMSM systems, power converter, robotic
systems, etc., in [28], a second-order model of PMSM system is introduced to design a speed regulation
controller; moreover, an active feed-forward compensation is designed to reduce chattering of the
sliding mode speed controller and improve the robustness. In [29], a linear extended state observer
(LESO) is used to achieve high precision position estimation. The LESO is embedded into current
controller to realize robustness in the full-speed range. In [30], an ADRC controller using ESO is
proposed for elevator traction machine. The nonlinear disturbance during elevator startup is estimated,
which improves the dynamic response without using weight sensor.
As we know, current loop also have disturbances, such as unmodeled dynamics of back
electromotive force (EMF), current fluctuation, stator resistance change. Also active disturbance
compensation is widely used in PMSM systems, but, almost all disturbance compensation is for speed
loop and position loop, few papers focuses on current loop. With the development of hardware, it
has been reported that disturbance compensation technology is used in current loop. In [31], ESO is
introduced to estimate the disturbance of q-axis current loop, and the disturbance is compensated by
feed-forward. Simulation and experiment show that feed-forward compensation of current loop can
effectively improve the anti-disturbance performance. In [32,33], a current loop controller based on
sliding mode control is proposed, and simulation results show that the new current controller is more
accurate then classical PI controller.
In this paper, by introducing sliding mode control and feed-forward compensation, a composite
controller is designed for not only speed loop but also q-axis current loop. The lumped disturbances of
speed loop, which include parameters variations and load change, are estimated by ESO. ESO-based
feed-forward compensation can actively reject disturbances. Moreover, the dynamics of q-axis
current loop are also considered in this paper, the lumped disturbances of q-axis current loop, which
include unmodeled dynamics of back electromotive force and resistance variations, are estimated
by ESO for active compensation. Here, a composite control law using SMC and ESO is designed
for speed loop. An ESO-based proportional-integral control law is designed for q-axis current loop.
A tracking-differentiator is used as transition process. The rest of this paper is organized as follows:
in Section 2, mathematical model of PMSM and problem formation of PMSM drive system are
given. In Section 3, a composite control law using SMC and ESO for speed loop and an ESO-based
Appl. Sci. 2020, 10, x 3 of 13
Appl. Sci. 2020, 10, 709 3 of 13
signal is introduced to replace step reference signal. In Section 4, simulation and experiment results
are presented. Finally, conclusions are shown in Section 5.
proportional-integral for q-axis current loop are designed. Moreover, a transition process of reference
signal is introduced
2. Problem Formationto replace step reference signal. In Section 4, simulation and experiment results
are presented. Finally, conclusions are shown in Section 5.
To simplify the mode of the PMSM, let us first assume: the three-phase windings of the stator
are evenly distributed, the saturation effect of the core is ignored, and air gap magnetic field is
2. Problem Formation
sinusoidal. The governing equations of PMSM under d-q reference frame can be described as follows:
To simplify the mode of the PMSM, let us first assume: the three-phase windings of the stator are
evenly distributed, the saturation effect  Rofs the core is ignored,  and airudgap  magnetic field is sinusoidal.
pω
 − L d-qnreference 0  beLdescribed

The governing equations of PMSM under frame can as follows:
 id   i   d 
d

 n pψ f     uq  
d
 . i  =  − Rs
 id  q   −nLpdω n−p ω −0  iq + 
Rs ud
(1)
 .      Lq Lq   id   Lq Ludq 
ψ
 ω=  −np ω − Rs − n p f   ω    Lq 
 iq  
 .    n pLψq f Lq 
B  iq  +  TL (1)
np ψ f   −  TL 

ω  0 0 − m
− J J 
B m
 ω  J− J
  
 
JJ

where RRss is stator resistance

where resistance (Ω), (Ω), ψ
ψff is rotor flux
flux linkage
linkage (Wb),
(Wb), TTLL is load torque (N/m).
(N/m). J is moment of
inertia (Kg·m ), ω is angular velocity(rad/s),
inertia (Kg·m 2
2 velocity(rad/s),BBmm is viscous
viscous friction
friction coefficient,
coefficient,uudd and
and uuqq are d and q axis
stator voltage
stator voltage (V),(V), LLdd and Lqq are d and q axis induction (H), idd and and iqq are
are dd and
and qq axis
axis stator
stator current
current (A),
npp is the number of pole pairs.
current vector
The current vector is is decomposed
decomposed into into flux-producing
flux-producing component
component iidd and and torque-producing
torque-producing
component iiqq by coordinate transformation,
component transformation, therefore
thereforeiidd and iqq can bebe regulated
regulated separately.
separately. Here make
idd == 0,
0, so
so that
that the
the electromagnetic
electromagnetic torque torque of the motor can be controlled only by iqq. Finally, the PMSM
speed-regulationproblem
speed-regulation problem is simplified
is simplified to iq to iq regulation.
regulation. The general
The general structure structure
of PMSMofspeed-regulation
PMSM speed-
regulation
system system
is shown is shown
in Figure in Figure
1. The [Link] The PMSM system
includes includes
a speed loop anda speed loop and
two current two
loops. current
Here a PI
loops. Here
controller is aadopted
PI controller is adopted
in d-axis current in d-axis
loop. current
In this loop.
paper, weIn this paper,
mainly designwe mainly design
controller controller
and disturbance
and disturbance
compensator compensator
for speed loop and foriqspeed
loop. loop and iq loop.

Figure [Link]
Theblock
block diagram
diagram of permanent
of permanent magnet
magnet synchronous
synchronous motormotor
(PMSM)(PMSM) speed-regulation
speed-regulation system.
system.
3. Controller Design
3. Controller Design
In practical industrial application, PMSM is affected by disturbance, which includes external
disturbance and internal disturbance.
In practical industrial application,Despite
PMSMcontroller is by
is affected robust, but thesewhich
disturbance, nonlinear disturbances
includes external
still have an impact on system, moreover, feedback-based control method cannot react promptly
disturbance and internal disturbance. Despite controller is robust, but these nonlinear disturbances
to
stilldisturbances.
have an impact Foronspeed loop,
system, a sliding
moreover, mode speed regulation
feedback-based controller
control method cannotis react
designed, and to
promptly a
feed-forward compensator based on ESO is designed for not only speed loop but q-axis current
disturbances. For speed loop, a sliding mode speed regulation controller is designed, and a feed- loop.
forward compensator based on ESO is designed for not only speed loop but q-axis current loop.
3.1. Speed Loop Controller Design
The process of the system reaching the sliding surface from any initial state is called reaching
motion, the sliding mode accessibility condition can only ensure that the system state can reach sliding
surface, but there is no limitation on the motion trajectory. The reaching law can make the system
Appl. Sci. 2020, 10, 709 4 of 13

state reach sliding surface according to the designed trajectory and improve the dynamic quality of
reaching motion.
The differential equation of mechanical angular speed is shown as:

. np ψ f TL Bm ω
ω= iq − − (2)
J J J

Define the state variable is defined as:

x = ω∗ − ω (3)

where ω* is reference speed. Because the first-order model of PMSM is adopted, so iq is approximately
equal iq * . Conventionally, the response time of current loop is faster than speed loop, if control period
of speed loop is more than 10 times of current loop, the approximate error can be neglected.
The state equation can be rewritten as:
.
x = ax + bi∗q + d (4)

np ψ f T
where a = − BJm , b = − J , d = JL + BJm ω∗ .
There are two main disturbances in PMSM operation: external disturbance which includes load
change and reference signal change, and inertial disturbance which includes parameters change and
unmolded dynamics.
As we know the disturbance of the system mainly consists of the following three parts:
(1) disturbance caused by load change, (2) disturbance caused by speed change, (3) disturbance
caused by reference speed change.
However, it would be difficult to observe their disturbances separately, so we define the set
of all disturbances as lumped disturbance R, and the same effect can be achieved by compensating
the lumped disturbance R. Indeed, we expended R as a new state variable x2 ; x defined as x1 , and
establishes a new second-order model.
 .
∗
 x1 = x. 2 + biq


 . (5)
 x2 = R


Bm TL
R= ω+ (6)
J J
T
where JL is the disturbance caused by load change, BJm ω is the viscous friction disturbance caused by
speed change.
The ESO is designed as:
( .
z1 = z2 + α1 (x1 − z1 ) + bi∗q
. (7)
z2 = α2 (x1 − z1 )
where α1 , α2 are constant gain, z1 , z2 are the estimated values of speed error and lumped disturbance,
respectively. The high-gain error feedback in ESO makes the observer dynamic much faster than
system; ESO is equivalent to the fast-changing subsystem in the system, which can ensure the fast
convergence of the error and high estimation accuracy. But the high gain observer will produce a peak
when initial set value differs greatly from system actual value. For example, for a step reference signal,
the initial value of error e = ω* at t = 0 s. Large error can cause a large initial impact on system to cause
a peak. For peak caused by difference initial value, a transition process is designed.

t 3

100( t j ) , 0 ≤ t ≤ t j

β=

(8)
100, t ≥ tj

Appl. Sci. 2020, 10, 709 5 of 13

where t is time, tj is transition time, β is a gain. In the initial stage t → 0, β → 0, which mitigate peak
caused by different initial values, with t → tj , β; can smoothly transit to the normal gain value. The
ESO can be redesigned as:
( .
z1 = z2 + α1 β(x1 − z1 ) + bi∗q
. (9)
z2 = α2 β(x1 − z1 )
In order to design the sliding mode control law for Equation (8), here, we define an integral sliding
surface as:
Zt
s = x1 + c x1 dt (10)
0

where c is the integral gain (c > 0). The reaching law is chosen as:
.
s = −εsign(s) − ks (11)

where ε, k are positive real numbers, sign() is symbolic function. Substituting the estimated lumped
disturbance z2 into Equation (5) yields the derivative of s as:
. .
s = x1 + cx1
(12)
= z2 + bi∗q + cx1 = −εsgn(s) − ks

and an ESO-based sliding mode controller can be designed as:

−cx1 − εsgn(s) − ks − z2
i∗q = (13)
b
Defining a Lyapunov function is as:
1 2
V= s (14)
2
To differential V yields:
. .
V = ss = s(R + b∗q + cx1 )
= s(R − cx1 − εsign(s) − ks − z2 + cx1 ) (15)
= s(R − z2 ) − ε|s| − ks2

here we quote a conclusion in [34], if the gain of ESO (11) is large enough, then the observer error
(R − z2 ) = αα12 w0 → 0 , w0 is the derivative of lumped disturbance.
From Equation (17)
.
V = −ε|s| − ks2 ≤ 0 (16)

When the sliding mode motion occurs on sliding surface s = 0, we can derive
. .
s = 0 → s = x1 + cx1 = 0 (17)

If c > 0, the existence condition of sliding mode and exponential convergence of x1 can
be guaranteed.

3.2. Design of q-Axis Current Controller

As the inner loop of a cascade control structure, the disturbance of q-axis current loop cannot
be ignored. Conventionally, compensating the disturbance of q-axis current loop can improve the
robustness of the whole cascade structure.
Appl. Sci. 2020, 10, x 6 of 13

3.2. Design of q-Axis Current Controller

As 2020,
Appl. Sci. the inner
10, 709loop
of a cascade control structure, the disturbance of q-axis current loop cannot
6 ofbe
13
ignored. Conventionally, compensating the disturbance of q-axis current loop can improve the
robustness of the whole cascade structure.
The differential equation of q-axis current loop is shown as:
The differential equation of q-axis current loop is shown as:

q =ω
uq u= ωe Le dLiddid++((R + pLqq))iiqq ++ωω
Rss + ψ
eψe f f
(18)
(18)

where pp is
where is differential operator,ωωee is
differentialoperator, is synchronous
synchronousspeed
speed(rad/s).
(rad/s).
From Equation (19), we can obtain:
From Equation (19), we can obtain:
ψ uq
. i = −RRss i − n ω i − ψff n ω + u q (19)
iq q= − Lqiqq− nppωidd − Lq npp ω +Lq (19)
Lq Lq Lq
We define the lumped disturbance of q-axis current loop as:
We define the lumped disturbance of q-axis current loop as:
R ψf
Rq = − Rss iq − n pω id − ψ f n pω (20)
Rq = − Lq iq − np ωid − Lq np ω (20)
Lq Lq
Here, Rq is expended as a new state variable. The state-space form is as follows:
Here, Rq is expended as a new state variable. The state-space form is as follows:
 . x = R + 1 u
 1q q 1 q
 x1q = Rq +LLqq uq


.
(21)
(21)
 x. 2q 
 x2 q== RRqq



The ESO for q-axis current

current loop
loop is is designed
designed as: as:

 ) +1 L1uq uq q
 .
 z1qz1=

 z2q + α 1q β
q = z 2 q + α1q
β((xx11qq −−zz11q
q)+
Lq (22)
 z2q = α2q β(x1q − z1q )
.

 (22)
 z = α β ( x − z )
 2q 2q 1q 1q
Finally, the controller can be designed as:
Finally, the controller can be designed as: Z
∗
ki (*i∗q − iq )dt − z2q
q ) + ki  (iq − iq )dt − z2 q
uq =
uq k=p k(ipq(i−q* −
iq i) + (23)
(23)

q-axis current
The block diagram of ESO-based q-axis current loop
loop is
is shown
shown in
in Figure
Figure 2.
2.

Figure 2. The block diagram of q-axis current controller based on extended state observer
observer (ESO).
(ESO).

Remark 1. Some scholars may doubt the effectiveness of ESO because the quantitative stability analysis has
Remark 1. Some
not been given. Inscholars
[34], themay
errordoubt the effectiveness
analysis of ESO
of second-order ESObecause the quantitative
under constant stability
disturbance analysis
is given, and has
the
not been given. In [34], the error analysis of second-order ESO under constant disturbance is given,
conclusion is that the observation error of ESO can be made small enough by selecting appropriate parameters. and the
conclusion is thatisthe
But the premise observationand
steady-state error of ESOdisturbance,
constant can be made and
small
theenough by selecting
quantitative appropriate
analysis parameters.
is also difficult and
But the premise is steady-state and constant disturbance, and the quantitative analysis is also
challenging. Actually, ESO has been applied in many papers and practical system, and many cases show that difficult and
challenging. Actually,
ESO is an effective ESO hasmethod.
estimation been applied in many
Therefore, ESOpapers and in
is applied practical system,
this paper, andand
ourmany cases show
simulations and
that ESO is an
experiments effective
prove estimation of
the effectiveness method.
ESO. Therefore, ESO is applied in this paper, and our simulations and
experiments prove the effectiveness of ESO.
Appl. Sci. 2020, 10, x 7 of 13

3.3. Transition Peocess Based on TD

Appl. Sci. 2020, 10, 709 7 of 13
We hope the output speed ω of system to track the step reference signal ω* as fast as possible, so
a large error proportional gain is generally used. However, increasing the error proportional gain
will Transition
3.3. make thePeocess
closed-loop
Based onsystem
TD become an under-damped and overshoot occurs, which is a
contradiction between rapidity and overshoot. In fact, any system has inertia, the output speed ω can
the ω signal ωthe
* as fast as possible, so a
only We hopefrom
change the output speed
zero, but of system
reference to track
signal ω* isthe stepsignal.
a step reference Therefore, initial value of error
large
e = ω*error
at t =proportional gainproportional
0 s. Large error is generally [Link] can However,
cause aincreasing
large initial theimpact
error proportional
on the system,gain will
which
make
make the
the closed-loop
system easysystem become
overshoot. an initial
If the under-damped
error can and overshoot
be reduced, occurs,
a large which
error is a contradiction
proportional can be
between rapidity and overshoot. In fact, any system has inertia,
used to speed up the transition process. Therefore, a suitable transition process is arranged the output speed ω can only change
according
from
to thezero, but the
reference reference
signal signal
ω*, and ω* is a step
the output ω tracks signal. theTherefore,
transition the initialtovalue
process reachofthe
error e = ωtarget.
control
* at t = 0
In
s. Large
this error
paper proportional gain can (TD)
a tracking-differentiator causefroma large ADRC initial is impact
used ason thethe system, which
transition make the system
process.
easy overshoot. If the initial error can be reduced, a large error proportional can be used to speed up
 H1 ( k +
the transition process. Therefore, a 1) = H 1 ( k transition
suitable ) + hH 2 ( k ) process is arranged according to the reference
 (24)
2 ( k ), ωthe
*
signal ω , and the output ωtracks
* H 2 ( k +the
1) =transition
H 2 ( k ) + h ( process
fst ( H 1 ( k ),
toHreach , r,control
h )) target. In this paper a
tracking-differentiator (TD) from ADRC is used as the transition process.
d = rh
H1 (k +1d) =
(
H1 (k) + hH2 (k)
 0 = hd (24)
H2 (k + 1) = H2 (k) + h( f st(H1 (k), H2 (k), ω∗ , r, h))
 y = H1 + hH 2
 
 d= = d 2 + 8r y
a0 rh




d0 = hd ( a0 − d )


H+2 + sign( y ), , y > d 0


  (25)

 a = q
y = H hH 2 2

1

 


2 + y8r y

 a0 = Hd +

, , y ≤ d0


  

2
h( a −d )
(25)

0
>

H + sign ( y ) , y d
 
2 0
 
a= 2
  rsign a ),y , a > d

H2 (+

 


 y ≤ d0


h
fst = − 

a

, (a), a|a|> >
 (
  rrsign dd



 fst = −  d r a ,

|a| > d


d
where hh is
where is the
the integration
integration time
time step,
step, rr is
is the
the speed
speed factor
factor that
that determines
determines thethe tracking
tracking speed,
speed, sign()
sign() is
is
symbolic function, H is the transition process of reference signal, H is the differential of transition
symbolic function, H1 is the transition process of reference signal, H2 is the differential of transition
1 2

process.
process.
Finally, the
Finally, the feedforward-based
feedforward-based PMSM PMSM speedspeed regulation
regulationsystem
systemisisshown
shownin inFigure
Figure3.3.

Figure 3. Diagram of feedforward-based PMSM system.

4. Simulation and
4. Simulation and Experiment
Experiment Results
Results
Simulations
Simulations and
and experiments
experiments are are done
done to
to verify
verify the
the effectiveness
effectiveness of
of the
the proposed
proposed method.
method. In this
In this
section, ESO-based method only for speed loop is named single-ESO method, ESO-based
section, ESO-based method only for speed loop is named single-ESO method, ESO-based method for method for
speed loop and
speed loop q-axis is
and q-axis is named
named double-ESO
double-ESO method.
[Link]
parametersofofPMSM
PMSMare follows:RR
areasasfollows: s is
s is 4
4Ω,Ω, L and
Ld and
d L are 0.0006 H, J is 0.002 Kg·m 2 , B is 0.00054, n is 4.
q 0.0006 H, J is 0.002 Kg·m , Bm ism0.00054, np isp4.
Lq are 2

4.1. Simulation Results

The simulations are simulated by Simulink of MATLAB 2016b. Solver is ode3, fundamental
sample time is 0.00001 s, reference speed ω* = 1000 rpm, the load torque change from 0.1 N/m to
 Appl. Sci. 2020, 10, x 8 of 13
Appl. Sci. 2020, 10, x 8 of 13
4.1. Simulation Results
Appl. Sci. 2020, 10, 709 8 of 13
4.1. Simulation Results
The simulations are simulated by Simulink of MATLAB 2016b. Solver is ode3, fundamental
sample Thetime is 0.00001are
simulations s, reference
simulated speed ω* = 1000 of
by Simulink rpm, the load2016b.
MATLAB torqueSolver
changeis from
ode3,0.1 N/m to 0.3
fundamental
0.3 N/m
N/m
sample T
atatTtime=iss,
= 1.5 1.5 thes,proportional
0.00001 thes, proportional
referencegain andgain
speed ω* = and
integral
1000gainintegral
rpm,forthe gain
current for current
loop
load torque arechange loop
100 and are
1000,
from 100
N/m andto 0.31000,
0.1respectively.
αN/m at T =α1.5
1 = 10,000,
respectively. α
1 2== 50,000,
s, the
10,000, αα1q ==3000,
proportional
2 α 2q =αand
gain
50,000, 1q =
100,000, ε
integral
3000, = α
0.5.
gain
2q = for current
100,000, ε =
loop are
0.5. 100 and 1000, respectively.
αThe The
1 = 10,000, estimated
α2 = 50,000,
estimated resulted
α1q = of
resulted ofESO
3000,ESO =for
α2q for speed εloop
100,000,
speed = [Link]
loop andq-axis
q-axiscurrent
current loop areare
loop shownshown in Figure 4, 4,
in Figure
respectively.
The It can
estimated be seen
resulted from
of Figure
ESO for 4 that
speed the
loopestimated
and value
q-axis of ESO
current
respectively. It can be seen from Figure 4 that the estimated value of ESO is similar to the theoretical is
loop similar
are to
shown the theoretical
in Figure 4,
value
value of
[Link] disturbance,
It can be which
of the disturbance, which
seen from indicates
Figurethat
indicates 4 thatthat ESO
thecan
ESO can
estimated accurately
value
accurately estimate
of ESO is
estimate the
similar
the disturbance of
to the theoretical
disturbance the
of the system.
system.
value ofThis provides a basis forindicates
compensating the can
disturbances.
accuratelyThe speedthe curves of single-ESO
This provides a basis for compensating the disturbances. The speed curves of single-ESOofmethod
the disturbance, which that ESO estimate disturbance the
method
system. and Thisdouble-ESO
provides a basismethod forare shown in Figure
compensating 5, respectively.
the disturbances. The It speed
can be curves
seen that of the system
single-ESO
and double-ESO method are shown in Figure 5, respectively. It can be seen that the system using
using
methoddouble-ESO
and double-ESO method methodhas are
batter shown disturbance
in Figure 5, rejection ability.
respectively. In be
It can Figure 6, disturbance
seen that the system
double-ESO
compensation
method has batter
improves
disturbance
the dynamic response
rejection ability.
of q-axis rejection
In Figure
current loop.
6, disturbance
In Figure 7, a transition
compensation
process
using double-ESO method has batter disturbance ability. In Figure 6, disturbance
improves
is the dynamic
obtained
compensation by improves
TD, the response
system of
will
the dynamic q-axis
track current ofloop.
q-axisIn
this transition
response Figure
signal
current 7, aIn
instead
loop. transition
of 7, aprocess
step signal.
Figure is obtained
Therefore,
transition the by
process
TD,transition
the system
is obtained will
process track
by TD,extends this transition
the range
the system signal
of system
will track instead
parameters
this transition of step signal.
and improves
signal Therefore,
instead ofthe the
robustness.
step transition
signal. Therefore, the process
extends the range
transition process of extends
system parameters
the range of systemand improves
parameters the and
robustness.
improves the robustness.

Figure 4. Comparison of the actual disturbance with ESO estimates: (a) actual disturbance R,
estimated
Figure
Figure 4. disturbance
4. Comparison of zthe
Comparison ofactual
2 of speed loop,disturbance
the actual (b) actual
disturbance withdisturbance
ESO Rq,
with estimates:
ESO estimated
estimates: (a)disturbance
(a) actual actual z2q of
R, q-axis
disturbance
disturbance R,
estimated
loop.
estimatedz2disturbance
disturbance z2 of (b)
of speed loop, speed loop,disturbance
actual (b) actual disturbance Rq, estimated
Rq , estimated disturbancedisturbance z2q ofloop.
z2q of q-axis q-axis
loop.

Figure
Figure 5. Speed
5. Speed curvesunder
curves undersignal-ESO
signal-ESO and
and double-ESO:
double-ESO:(a)(a)response of of
response speed under
speed signal-ESO,
under signal-ESO,
Appl. Sci.
(b) 2020,
(b) 10,5.x of
response
Figure
response of speed
Speed under
curves
speed double-ESO.
under
under signal-ESO and double-ESO: (a) response of speed under signal-ESO,
double-ESO. 9 of 13
(b) response of speed under double-ESO.

Figure 6. iiqq and

Figure 6. and iiqq reference
reference with
with current
current loop
loop compensation
compensation and
and without
without current
current loop
loop compensation:
compensation:
(a) curve of iq and iq reference without q-axis current compensation, (b) curve of iq and iq reference
(a) curve of iq and iq reference without q-axis current compensation, (b) curve of i q and iq with
reference with
q-axis current compensation.
q-axis current compensation.
 Figure 6. iq and iq reference with current loop compensation and without current loop compensation:
Appl. (a)
Sci. curve of 709
2020, 10, iq and iq reference without q-axis current compensation, (b) curve of iq and iq reference with9 of 13
q-axis current compensation.

Figure
Figure 7. Curves
Curves of
of transition
transition process
process and
and speed
speed curves
curves using transition process: (a)
(a) speed
speed curve
curve and
acceleration
acceleration curve of transition process,
process, (b)
(b) speed
speed curve under cc == 15, (c) speed under c =
curve under = 30,
30, (d)
(d) speed
speed
Appl. Sci. 2020, 10, x 10 of 13
under c ==50.
under 50.

mode controller,Results
4.2. Experiment
Experiment and disturbance compensation based on ESO is designed for speed loop and q-axis
4.2. Results
current loop.
AA can
It test be
test device
device isthat
seenis established
established
PI has thetoworst
to verifyperformance
verify the proposed
the proposedandmethod,
method, and
cannotand is shown
is
handle shown in Figure
in Figure
disturbance 8. The
8. The
very test
[Link]
In
deice
deice consists
consists of
of controller,
controller, motor
motor driver,
driver, PMSM.
PMSM. The
The controller
controller is
is
Figures 10 and 11, speed sliding mode controller and speed loop disturbance compensation STM32F103,
STM32F103, FOC,
FOC, and
and SVPWM
SVPWM
are implemented
are implemented
technology by C-program,
by C-program,
can significantly the clock
the
improve clock frequency of
frequency
the anti-disturbanceof controller
controller
ability ofisis
the72system.
72 MHz, the
MHz, the sampling
sampling
In Figures time13,
12 time
and of
of
speed
speed loop
currentloop is 10 ms, the
loopis disturbance
10 ms, the PWM PWM frequency
frequency
technology is 16
canisspeed KHz,
16 KHz, current
current
up the sensing
sensing
dynamic by isolated
by isolated
response current
of thecurrent sensors
systemsensors (ICS),
(ICS),
and shorten
an
an encoder
encoder of
of 1000
1000 lines
lines is
is used.
used. The
The max
max speed
speed of
of motor
motor is
is 3000
3000 rpm
rpm and
and the
the
the adjustment time. Moreover, it can make iq tracking iq reference faster and more stable. A detail rated
rated current
current is
is 3.13
3.13 A.
A.
The
The serial port
serial port
comparison communicating
communicating
is shown baud rate between the STM32F103 and computer
in Table [Link] rate between the STM32F103 and computer is 115,200. is 115,200.
In order to verify the effectiveness of the proposed method, the proposed method is compared
with other controllers. The speed curve under PI speed controller is shown in Figure 9, the
proportional gain and integral gain of speed PI controller are 200 A/rad/s and 50 A/rad, respectively.
The proportional gain and integral gain of current loop PI controller are 572 V/A and 121 V/A s,
respectively. The parameters of sliding mode speed controller are as follows: c = 21, ε= 50, k = 0.0075.
The parameters of ESO are as follows: α1 = 3000, α2 = 3000. The speed curve under speed loop sliding
mode controller is shown in Figure 10, the current loop controllers are PI. The speed curve under
SMC with speed loop compensation is shown in Figure 11. In Figure 12, the speed controller is sliding

Experimental test device.

Figure 8. Experimental

In order to verify the effectiveness of the proposed method, the proposed method is compared with
other controllers. The speed curve under PI speed controller is shown in Figure 9, the proportional gain
and integral gain of speed PI controller are 200 A/rad/s and 50 A/rad, respectively. The proportional gain
and integral gain of current loop PI controller are 572 V/A and 121 V/A s, respectively. The parameters
of sliding mode speed controller are as follows: c = 21, ε= 50, k = 0.0075. The parameters of ESO
are as follows: α1 = 3000, α2 = 3000. The speed curve under speed loop sliding mode controller is

Figure 9. Speed curve under PI controller: (a) the speed curve, (b) speed curve when load disturbance.
Appl. Sci. 2020, 10, 709 10 of 13

shown in Figure 10, the current loop controllers are PI. The speed curve under SMC with speed loop
compensation is shown in Figure 11. In Figure 12, the speed controller is sliding mode controller,
and disturbance compensation based on ESO
Figure
Figure is designed
[Link]
Experimental for
test
test speed loop and q-axis current loop.
device.
device.

Figure
Figure9.
Figure [Link]
9. Speedcurve
Speed curveunder
curve underPI
under PIcontroller:
PI controller:(a)
controller: (a)the
(a) thespeed
the speedcurve,
speed curve,(b)
curve, (b)speed
(b) speedcurve
speed curvewhen
curve whenload
when loaddisturbance.
load disturbance.
disturbance.

Figure
Figure10.
[Link]
Speed curve
curve under
undersliding
under slidingmode
sliding modecontrol
mode control(SMC):
control (SMC):(a)
(SMC): (a)the
(a) thespeed
the speedcurve,
speed curve,(b)
curve, (b)speed
speedcurve
speed curvewhen
when
Appl.
Appl.
loadSci. 2020,
2020, 10,
10, xx
[Link].
load disturbance. 11
11 of
of 13
13

Figure
Figure 11.11.
Figure 11. Speed curve
Speedcurve
Speed under
curveunder SMC
underSMC with
SMCwith speed loop
with speed loop compensation:
loop compensation: (a)
compensation: (a)the
(a) thespeed
the speed curve,
curve,
speed (b)
curve, speed
(b)(b)
speed
speed
curve
curve
curve when
when
when load
load
load disturbance.
disturbance.
disturbance.

Figure
Figure
Figure 12.12.
12. Speed curve
Speedcurve
Speed under
curveunder SMC
underSMC with
SMCwith speed
speed loop
with speed loop compensation
compensationand
compensation andcurrent
and current
currentloop
loop compensation:
compensation:
loop compensation:
(a)
(a) the
the speed
speed curve,
curve, (b)
(b) speed
speed curve
curve when
when load
load disturbance.
disturbance.
(a) the speed curve, (b) speed curve when load disturbance.
 curve when load disturbance.

Appl. Sci. 2020, 10, 709 11 of 13

It can be seen that PI has the worst performance and cannot handle disturbance very well.
In Figures 10 and 11, speed sliding mode controller and speed loop disturbance compensation
technology can significantly improve the anti-disturbance ability of the system. In Figures 12 and 13,
current loop disturbance technology can speed up the dynamic response of the system and shorten
the adjustment [Link]
Figure 12. Speed Moreover, it canwith
under SMC make iq tracking
speed iq reference
loop compensation faster
and andloop
current more stable. A detail
compensation:
comparison is shown
(a) the speed curve,in(b)
Table 1. curve when load disturbance.
speed

Figure 13. iqq and

andiiqqreference
referencewith
withcurrent
currentloop
loopcompensation
compensation and
and without
without current
current loop
loop compensation:
compensation:
(a) curve
(a) curve of
of iiqq and iqq reference
reference without q-axis current
current compensation,
compensation,(b)
(b)curve
curveof
ofiiqq and iqq reference with
current compensation.
q-axis current compensation.

Table 1. Comparisons between control methods.

Table 1. Comparisons between control methods.
Speed During Disturbance Steady-State
Control Method Overshoot Setting
Setting Time Speed During Disturbance Steady-State
Control Method Overshoot Decrease Increase Error
Time Decrease Increase Error
Speed: PI
Speed: PI
Current: PI
26% 0.3s −240 rpm 336 rpm ±18 rpm
26% 0.3s −240 rpm 336 rpm ±18 rpm
Current: PI
Speed: SMC
0% 1.27s −246 rpm 48 rpm ±12 rpm
Current: PI
Speed: SMC
Speed: SMC + ESO 0%
0% 1.27s
1.06s −246 rpm
−112 rpm 1848 rpm
rpm ±12±12
rpmrpm
Current: PIPI
Current:
Speed: SMC + ESO
Speed: SMC +
Current: PI + ESO
0% 0.531s −28 rpm 18 rpm ±10 rpm
ESO 0% 1.06s −112 rpm 18 rpm ±12 rpm
Current: PI
5. Conclusions
Speed: SMC +
In this ESO
paper, a robust control
0% method 0.531s
based on feed-forward
−28 rpm compensation
18 rpm has been proposed
±10 rpm for
PMSM speed-regulation
Current: PI + ESO system. A feed-forward based on ESO is introduced in speed loop and q-axis
current loop to improve the dynamic response and robustness. By means of the disturbance estimation
for feed-forward compensation, the compensator can generate a compensation signal to reduce the
interference of lumped disturbance, which include external disturbance and internal disturbance,
in the system. In particular, compensation for current loop disturbances significantly improves the
performance and robustness of the system. A transition process of reference signal is introduced,
which can reduce the initial error and increase the range of error feedback gain to improve system
robustness. Both simulation and experimental results have been presented to verify the feasibility of
the proposed method.

Author Contributions: Conceptualization: Y.M., methodology: Y.M. and Y.L.; formal analysis: Y.L. and Y.M.;
writing—original draft preparation: Y.M.; writing—review and editing: Y.L.; supervision: Y.L. All authors have
read and agreed to the published version of the manuscript.
Funding: This work was partly supported by National Natural Science Foundation of China (Grant no. U1910211).
Conflicts of Interest: The authors declare no conflict of interest.
Appl. Sci. 2020, 10, 709 12 of 13

References
1. Grcar, B.; Cafuta, P.; Znidaric, M.; Gausch, F. Nonlinear control of synchronous servo drive. IEEE Trans.
Control Syst. Technol. 1996, 4, 177–184. [CrossRef]
2. Zhou, A.M.; Wu, Y.; Zhong, L.; Chen, Q.; Guo, J. Adaptive robust control for high-speed permanent magnet
synchronous motor systems. In Proceedings of the 2017 3rd IEEE International Conference on Control
Science and Systems Engineering (ICCSSE), Beijing, China, 17–19 August 2017; pp. 346–350.
3. Xu, Y.P.; Hou, Y.L.; Li, Z.H. Robust predictive speed control for SPMSM drives based on extended state
observers. J. Power Electron. 2019, 19, 497–508.
4. Ye, J.G.; Yang, J.H.; Xie, D.S.; Huang, B.Z. Strong robust and optimal chaos control for permanent magnet
linear synchronous motor. IEEE Access 2019, 7, 57907–57916. [CrossRef]
5. Choi, H.H.; Jung, J.W. Takagi-sugeno fuzzy speed controller design for a permanent magnet synchronous
motor. Mechatronics 2011, 21, 1317–1328. [CrossRef]
6. Hou, L.M.; Li, W.H.; Shen, H.S.; Li, T.C. Fuzzy sliding mode control for systems with matched and mismatched
uncertainties/disturbances based on ENDOB. IEEE Access 2019, 7, 666–673. [CrossRef]
7. Khorashadizadeh, S.; Sadeghijaleh, M. Adaptive fuzzy tracking control of robot manipulators actuated by
permanent magnet synchronous motors. Comput. Electr. Eng. 2018, 72, 100–111. [CrossRef]
8. Liu, L.; Cartes, D.A. Synchronisation based adaptive parameter identification for permanent magnet
synchronous motors. IET Control Theory Appl. 2007, 1, 1015–1022. [CrossRef]
9. Luo, Y.; Chen, Y.; Ahn, H.S.; Pi, Y. Dynamic high order periodic adaptive learning compensator for cogging
effect in permanent magnet synchronous motor servo system. IET Control Theory Appl. 2011, 5, 669–680.
[CrossRef]
10. Li, S.H.; Liu, Z.G. Adaptive speed control for permanent-magnet synchronous motors system with variations
of load inertia. IEEE Trans. Ind. Electron. 2009, 5, 3050–3059.
11. Barambones, O.; Garrido, A.J.; Maseda, F.J. Integral sliding-mode controller for induction motor based on
field-oriented control theory. IET Control Theory Appl. 2007, 1, 786–794. [CrossRef]
12. Huang, Y.; Sung, C. Implementation of sliding mode controller for linear synchronous motors based on
direct thrust control theory. IET Control Theory Appl. 2010, 4, 326–338. [CrossRef]
13. Wang, X.; Reitz, M.; Yaz, E.E. Field oriented sliding mode control of surface-mounted permanent magnet
ac motors: Theory and applications to electrified vehicles. IEEE Trans. Veh. Technol. 2018, 67, 10343–10356.
[CrossRef]
14. Linares-Flores, J.; García-Rodríguez, C.; Sira-Ramírez, H.; Ramírez-Cárdenas, O.D. Robust backstepping
tracking controller for low-speed PMSM positioning system: Design, analysis, and implementation.
IEEE Trans. Ind. Inform. 2015, 11, 1130–1141. [CrossRef]
15. Wu, S.F.; Zhang, J.W. A terminal sliding mode observer based robust backstepping sensorless speed control
for interior permanent magnet synchronous motor. Int. J. Control Autom. Syst. 2018, 16, 2743–2753. [CrossRef]
16. Ning, B.; Cheng, S.M.; Qin, Y. Direct torque control of PMSM using sliding mode backstepping control with
extended state observer. J. Vib. Control 2018, 24, 694–707. [CrossRef]
17. Zhao, D.Z.; Li, C.W.; Ren, J. Speed synchronization of multiple induction motors with adjacent cross-coupling
control. IET Control Theory Appl. 2010, 4, 119–128. [CrossRef]
18. Zhang, X.; Sun, L. Nonlinear speed control for PMSM using sliding mode control and disturbance
compensation techniques. IEEE Trans. Power Electron. 2013, 28, 1358–1368. [CrossRef]
19. Ohnishi, K. A new servo method in mechatronics. Trans. Jpn. Soc. Electr. Eng. 1987, 107, 83–86.
20. Wei, X.; Guo, Z.; Liu, X. Composite hierarchical anti-disturbance control for nonlinear robotic systems with
robust nonlinear disturbance observer. In Proceedings of the 32nd Chinese Control Conference, Xi’an, China,
26–28 July 2013; pp. 2694–2699.
21. He, X.; Kou, G.; Calaf, M. In-ground-effect modeling and nonlinear-disturbance observer for multirotor
unmammed aerial vehicle control. J. Dyn. Syst. Meas. Control-Trans. ASME 2019, 141, 071013. [CrossRef]
22. Zhao, W.X.; Jiao, S.; Chen, Q.; Xu, D.Z. Senseless control of a linear permanent-magnet motor based on an
improved disturbance observer. IEEE Trans. Ind. Electron. 2018, 65, 9291–9300. [CrossRef]
23. Li, S.; Zhou, M.; Yu, X. Design and implementation of terminal sliding mode control method for PMSM
speed regulation system. IEEE Trans. Ind. Inform. 2013, 7, 1879–1891. [CrossRef]
Appl. Sci. 2020, 10, 709 13 of 13

24. Yan, J.; Wang, H.; Huang, S. Load disturbance observer-based complementary sliding mode control for
PMSM of the mine traction electric locomotive. Int. J. Fuzzy Syst. 2019, 21, 1051–1058. [CrossRef]
25. Yan, Y.D.; Yang, J.; Sun, Z.X.; Zhang, C.L.; Li, S.H. Robust speed regulation for PMSM servo system with
multiple sources of disturbances via an augmented disturbance observer. IEEE-ASME Trans. Mechatron.
2018, 23, 769–780. [CrossRef]
26. Han, J.Q. The extended state observer of a class of uncertain systems. Control Decis 1995, 10, 85–88. (In
Chinese)
27. Han, J.Q. From PID to active disturbance rejection control. IEEE Trans. Ind. Electron. 2009, 56, 900–906.
[CrossRef]
28. Li, S.; Zong, K.; Liu, H. A composite speed controller based on a second-order model of PMSM system. Trans.
Inst. Meas. Control 2011, 33, 522–541.
29. Du, B.; Wu, S.; Han, S.; Cui, S. Application of linear active disturbance rejection controller for senseless
control of internal permanent-magnet synchronous motor. IEEE Trans. Ind. Electron. 2016, 63, 3019–3027.
[CrossRef]
30. Wang, G.L.; Wang, B.W.; Li, C.; Xu, D.G. Weight-transducerless control strategy based on active disturbance
rejection theory for gearless elevator drives. IET Electr. Power Appl. 2016, 11, 289–299. [CrossRef]
31. Li, S.; Xia, C.; Zhou, X. Disturbance rejection control method for permanent magnet synchronous motor
speed-regulation system. Mechatronics 2012, 22, 706–714. [CrossRef]
32. Zwerger, T.; Mercorelli, P. Combining an internal SMC with an external MTPA control loop for an interior
PMSM. In Proceedings of the 2018 23rd International Conference on Methods and Models in Automation
and Robotics, Miedzyzdroje, Poland, 27–30 August 2018; art no. 8485900. pp. 674–679.
33. Zwerger, T.; Mercorelli, P. Combining SMC and MTPA using an EKF to estimate parameters and states of
an interior PMSM. In Proceedings of the 2019 20th International Carpathian Control Conference (ICCC),
Krakow-Wieliczka, Poland, 26–29 May 2019; pp. 1–6.
34. Han, J.Q. Error analysis of the second order ESO. J. Syst. Sci. Math. Sci. 1999, 19, 465–471. (In Chinese)

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license ([Link]