Preprints, 1st IFAC Conference on Modelling, Identification and

Preprints,
Preprints, 1st
Control of 1st IFAC Conference
IFAC
NonlinearConference
Systems onon Modelling,
Modelling, Identification
AvailableIdentification and
and
online at www.sciencedirect.com
Preprints,
Control
Control of 1st
of IFAC Conference
Nonlinear
Nonlinear Systems
Systems on Modelling, Identification and
June 24-26, 2015. Saint Petersburg, Russia
Control
June of Nonlinear
June 24-26,
24-26, 2015. Systems
2015. Saint
Saint Petersburg, Russia
Petersburg, Russia
June 24-26, 2015. Saint Petersburg, Russia
ScienceDirect
IFAC-PapersOnLine 48-11 (2015) 065–071

Comparative Study on Direct Torque Control of Interior Permanent Magnet

Comparative Study on Direct Torque Control of Interior Permanent Magnet
Comparative StudySynchronous
on Direct Torque
MotorControl of Interior
for Electric VehiclePermanent Magnet
Synchronous Motor for Electric Vehicle
Synchronous Motor for Electric Vehicle
Weijie Lin*. Dongliang Liu**. Qiuxuan Wu***. Qiang Lu****. Lili Cui*****. Jian Wang******
Weijie Lin*.
Weijie Lin*. Dongliang
Dongliang Liu**. Liu**. Qiuxuan
Qiuxuan Wu***.Wu***. Qiang
Qiang Lu****.
Lu****. Lili Lili Cui*****.
Cui*****. Jian Jian Wang******
Wang******
Weijie Lin*. Dongliang Liu**. Qiuxuan Wu***. Qiang Lu****. Lili Cui*****. Jian Wang******
*Hangzhou Dianzi University, Hangzhou, China PR(Tel: 86-571-86878535;
*Hangzhou Dianzi
*Hangzhou Dianzi University,
University, Hangzhou,
Hangzhou, China
China PR(Tel:
PR(Tel: 86-571-86878535;
86-571-86878535;
*Hangzhou Dianzi University, e-mail:Hangzhou,
linwj@hdu.edu.cn)
China PR(Tel: 86-571-86878535;
e-mail: linwj@hdu.edu.cn)
e-mail: linwj@hdu.edu.cn)
* Wolong Electric e-mail: Group Co. Ltd, Shangyu, China PR
linwj@hdu.edu.cn)
* Wolong
Wolong Electric
Electric Group Co. Ltd,
Ltd, Shangyu,
Shangyu, China China PR PR
** Hangzhou **Dianzi Wolong University, Group
Electric Group
Co.
Hangzhou, China
Co. Ltd, PR (e-mail:
Shangyu, liudl@hdu.edu.cn)
Chinaliudl@hdu.edu.cn)
PR
** Hangzhou
** Hangzhou Dianzi Dianzi University,
University, Hangzhou,
Hangzhou, China
China PR (e-mail:
PR (e-mail: liudl@hdu.edu.cn)
***
** Hangzhou Dianzi
Hangzhou Dianzi University,
University, Hangzhou,
Hangzhou, China
China PR PR (e-mail:
(e-mail: liudl@hdu.edu.cn)
wuqx@hdu.edu.cn)
*** Hangzhou
*** Hangzhou
Hangzhou Dianzi Dianzi University,
Dianzi University, Hangzhou,
University, Hangzhou, China
Hangzhou, China
China PR PR (e-mail:
PR (e-mail: wuqx@hdu.edu.cn)
(e-mail: lvqiang@hdu.edu.cn)
wuqx@hdu.edu.cn)
****
*** Hangzhou
Hangzhou Dianzi Dianzi University,
University, Hangzhou,
Hangzhou, China
China PR PR (e-mail:
(e-mail: lvqiang@hdu.edu.cn)
wuqx@hdu.edu.cn)
****
**** Hangzhou Dianzi University, Hangzhou, China PR (e-mail: lvqiang@hdu.edu.cn)
*****
**** Hangzhou
Hangzhou Dianzi
Dianzi University,
University, Hangzhou,
Hangzhou, China
China PR PR (e-mail:
(e-mail: cll@hdu.edu.cn)
lvqiang@hdu.edu.cn)
*****
***** Hangzhou
Hangzhou Dianzi
Dianzi University,
University, Hangzhou,
Hangzhou, China
China PR (e-mail: cll@hdu.edu.cn)
*********** Hangzhou
Hangzhou Dianzi
DianziUniversity,
University, Hangzhou,
Hangzhou, China
China PRPRPR
(e-mail:
(e-mail:
(e-mail:
cll@hdu.edu.cn)
wangjian@hdu.edu.cn)
cll@hdu.edu.cn)
****** Hangzhou Dianzi University, Hangzhou, China PR
****** Hangzhou Dianzi University, Hangzhou, China PR (e-mail: wangjian@hdu.edu.cn) (e-mail: wangjian@hdu.edu.cn)
****** Hangzhou Dianzi University, Hangzhou, China PR (e-mail: wangjian@hdu.edu.cn)
Abstract: Comparative studies on several direct torque control (DTC) strategies of interior permanent
Abstract: Comparative
Abstract: Comparative studies studies on on several
several direct
direct torque
torque control
control (DTC)
(DTC) strategies
strategies of of interior
interior permanent
permanent
magnet synchronous
Abstract: Comparative motor (IPMSM)
studies for electric
on several vehicles
direct torque (EVs) are discussed
control in details, namely basic
magnet
magnet synchronous
synchronous motor
motor (IPMSM)
(IPMSM) for electric
for electric vehicles
vehicles (EVs)(DTC)
(EVs) are strategies
are discussed
discussed in of
in interior
details,
details, permanent
namely
namely basic
basic
DTC,
magnet DTC combined
synchronous with space vector modulation (DTC-SVM), and deadbeat DTC (DB-DTC). These
DTC, DTC
DTC, DTC combinedmotor
combined (IPMSM)
with space
with space vector
vector formodulation
electric vehicles
modulation (EVs) are
(DTC-SVM), anddiscussed
deadbeat in DTC details, namelyThese
(DB-DTC). basic
DTC
DTC, strategies
DTC are reviewed,
combined with space meanwhile
vector dynamics(DTC-SVM),
modulation and steady-state
(DTC-SVM),
and
and
deadbeat
performance
deadbeat
DTC
DTC
(DB-DTC).
are analyzedThese
(DB-DTC). and
These
DTC strategies
DTC strategies are are reviewed,
reviewed, meanwhile
meanwhile dynamicsdynamics and and steady-state
steady-state performance
performance are are analyzed
analyzed and and
compared.
DTC
compared. Simulations
strategies
Simulations of
are reviewed, a 20kW
of aa 20kW IPMSM
meanwhile
20kW IPMSMdynamicsfor EVs
for EVs are carried
andcarried
EVs are
are out
steady-state
carried out forfor comparison
forperformance
comparison are studies
studies including:
analyzed and
including:
compared. Simulations of IPMSM for out comparison studies including:
ripple of
compared.
ripple of torque
torque and stator
Simulations
and stator
of aflux,
20kW
flux, sensitivity
IPMSM
sensitivity tofor
to machine's
EVs
machine'sare parameter,
carried
parameter, out computational
for comparison
computational complexity,
studies
complexity, and total
total
including:
and
ripple of torque and stator flux, sensitivity to machine's parameter, computational complexity, and total
harmonic distortion
ripple of torque and ofof stator
stator current.
flux, The results
sensitivity results can be
to machine's be used
used as as guidance
guidance for application
application of IPMSM
IPMSM to
harmonic
harmonic distortion
distortion of stator
stator current.
current. The
The results can
can be parameter,
used as computational
guidance for
for complexity,
application of
of and total
IPMSM to
to
EVs
EVs andand
harmonic others.
distortion of stator current. The results can be used as guidance for application of IPMSM to
and others.
others.
EVs
EVs
© andIFAC
2015, others.(International Federation
Keywords: Direct
Keywords: Direct torque control,
torque control, flux,ofinterior
flux,
Automatic
interior Control) magnet
permanent
permanent
Hosting synchronous
by Elsevier Ltd.
magnet synchronous
synchronous motor,
All rights
motor,
motor, electric
electric
reserved.
vehicle,
vehicle,
Keywords:
deadbeat, Direct torque
observer, control, flux,
simulation. interior permanent magnet electric vehicle,
Keywords: Direct
deadbeat, observer,
deadbeat, torque
observer, simulation. control,
simulation. flux, interior permanent magnet synchronous motor, electric vehicle,
deadbeat, observer, simulation.
dynamics and human comfort. Therefore, the torque open-
1. INTRODUCTION dynamics and
dynamics and human
human comfort.
comfort. Therefore,
Therefore, the the torque
torque open-
open-
1. INTRODUCTION
1. INTRODUCTION loop control
dynamics and ofhuman
CVC is not extremely
comfort. Therefore, suitable
the for IPMSM
torque open-
loop control
loop control of CVC is not extremely
of CVC is not extremely suitable for IPMSMsuitable for IPMSM
With the increasing1.demand INTRODUCTION
of living comfort, automobile control loop
control
of EVs.
control
of EVs. of CVC is not extremely suitable for IPMSM
With the
With the increasing
increasing demanddemand of of living
living comfort, automobile control of EVs.
comfort, automobile
gets
With more and more
the increasing utilized
demand for
of for
living individual
comfort, and and public control of EVs.
automobile
gets more
gets more and more
and more utilized
utilized for individual
individual and public Direct torque control (DTC) has closed-loop control both of
public
transportation
gets more and because
moreof of its convenience. But, accompanying
and public Direct Direct torque
torque control
control (DTC)
(DTC) has has closed-loop
closed-loop controlcontrol both
both ofof
transportation
transportation because
because ofutilized for individual
its convenience.
its convenience. But, accompanying
But, accompanying torque
Direct and stator
torque flux,(DTC)
control which has wasclosed-loop
firstly proposed controlforboth
IM by of
vast fuel
transportationconsumption
because ofandand environmental
its environmental
convenience. But, pollution become
accompanying torque
torque and
and stator
stator flux,
flux, which
which was
was firstly
firstly proposed
proposed for
for IM
IM by
by
vast fuel
vast fuel consumption
consumption and environmental pollution
pollution become Takahashi
become torque and and
stator Noguchi
flux, (1986)
which was and Depenbrock
firstly proposed (1988).
for IM At
by
more
vast serious. Automobile companies focus on become Takahashi and
electric Takahashi and Noguchi
Noguchi (1986) (1986) and and Depenbrock
Depenbrock (1988). (1988). At At
morefuel
more consumption
serious. Automobileand environmental
companies focus pollution
focus on electric present, DTC
electric Takahashi has become(1986)a powerful and widely used control
vehiclesserious.
more (EVs), Automobile
serious. hybrid electric
Automobile
companies
vehiclesfocus
companies (HEVs),on
on present, DTC
plug-in present,
electric DTC andhas Noguchi
has become aa powerful
become powerfuland andDepenbrock
and widely used
widely (1988).
used control
controlAt
vehicles (EVs),
vehicles
hybrid
(EVs), hybrid
electric
hybrid electric
vehicles
electric vehicles
(PHEVs),
vehicles (HEVs),
(HEVs), plug-in
and fuel-cell
plug-in present,
vehicles. strategy
strategy of
of
DTC
of AC
AC machines.
has become
AC machines.
machines. The
The
a dq
powerful
The dq axis
axis
and vector
widely
axis vector
decoupling
used
vector decoupling control
decoupling of
of
of
vehicleselectric
hybrid
hybrid (EVs), vehicles
electric hybrid electric
vehicles (PHEVs),
(PHEVs), vehicles (HEVs),vehicles.
and fuel-cell
and fuel-cell plug-in strategy
vehicles.
Among
hybrid these, electric
electric vehicles vehicles
(PHEVs), get andthe most
fuel-cellattraction
vehicles. field oriented
in strategy
field of ACcontrol
oriented machines.
control
(FOC)
(FOC)The is dqreplaced
is axis vector
replaced
by two
by two
hysteresis
decoupling
hysteresisof
Among these,
Among these, electric
electric vehicles
vehicles get get thethe most
most attraction
attraction in in field oriented
controllers of controlwhich
DTC, (FOC) is replaced
meets very wellby with
two thehysteresis
on-off
recent years
Amongyears these, for its only convenient and rechargeable battery field oriented controlwhich(FOC) is replaced by with
two thehysteresis
recent
recent years forelectric
only vehicles
its only convenient getand therechargeable
and most attraction
rechargeable in controllers
battery controllers of DTC,
ofof
DTC, which meets
meets very well
very well with the on-off
on-off
recent supplyfor
power years for
its
and
its simple
only
convenient
driver control
convenient and rechargeable
battery
structure. Interior
battery operation
controllers
operation ofof power
DTC,
power
transistors
which
transistorsmeets of inverter.
of very
inverter.well DTC
with
DTC
of
the
of
IPMSM
on-off
IPMSM
power supply
power supply and and simple
simple driver
driver control
control structure.
structure. Interior
Interior operation
was firstlyof power transistors
presented by Zhong (1997).of inverter.DTC DTC has someof IPMSM
virtues
permanent
power supply magnet
and synchronous
simple driver motor
control (IPMSM)
structure. hasInterior
some operation
was firstly of power
presented transistors
by Zhong of inverter.
(1997). DTC DTC
has of
some IPMSM
virtues
permanent
permanent magnet
magnet synchronous motor (IPMSM) has some wasboth
firstlycontrol
presented by Zhong (1997). and DTCdriver
has some virtues
merits
permanentof high powersynchronous
magnet density, highmotor (IPMSM)
efficiency, good has some of
reliability, wasboth
firstlycontrol
presented strategy
by Zhongframework
(1997). and DTCdriver
has some materials.
virtues
merits of
merits of high
high powersynchronous
power density, highmotor
density, high (IPMSM)
efficiency,
efficiency, good has
good some of
reliability,
reliability, of both
Comparing
control
with
strategy
strategy
FOC, DTC
framework
framework
does not
and
require
driver
any
materials.
materials.
coordinate
low torque
merits of high ripple
power and wide range of speed regulation, which of both control strategy framework and driver materials.
low torque
low torque ripple
ripple anddensity,
and wide range
wide high of
range efficiency,
of good reliability,
speed regulation,
speed regulation, Comparing with
which Comparing with FOC,FOC, DTC DTC does does notnot require
require any any coordinate
coordinate
make
low it much
torque fitter
ripple and for electric
wide range vehicles
of speed driving than which
regulation, other transformation
which Comparing
transformation withandand
FOC,
and
space
space
vector
DTCvectordoes not
vector
modulation.
require any
modulation.
Furthermore,
coordinate
Furthermore,
make
make it much
it electric fitter
much fitter for electric
for electric vehicles driving than other transformation space modulation. Furthermore,
general
make it much machines
fitter for such vehicles
electric as induction
vehicles
drivingmotor
driving
than (IM),
than
other DTC also does not require rotator position sensor which is
other transformation
DTC alsoalso does
does not and space
not require vector
require rotator modulation.
rotator position
position sensor Furthermore,
sensor whichwhich is is
general electric
general electric machines
machines such such as as induction
induction motor motor (IM),(IM), DTC essential for FOC. DTC isrotator natively sensorless leading to
brushless
general DC motor (BLDC), andas switch reluctance motor DTC alsofordoes not require position sensorleading
which to is
brushlesselectric
brushless DC motor
DC machines
motor (BLDC),
(BLDC), suchand induction
and switch
switch motor motor
reluctance
reluctance (IM), essential
motor essential
simplified
for FOC.
FOC. DTC is
DTC
implementation
is natively
natively
and lower
sensorless
sensorless
cost. DTC
leading to
has
(SR) as presented
brushless DC by Chan
motor (BLDC),(1996).and switch reluctance motor essential
simplified for FOC. DTC
implementation is natively
and lowersensorless
cost. leading
DTC to
has
(SR) as
(SR) as presented
presented by by Chan
Chan (1996).
(1996). simplified
comparable implementation
steady and dynamic and torquelower performance
cost. DTC with has
(SR) as presented by Chan of (1996). simplified
comparable implementation
steady and and dynamic
dynamicand torquelower performance
torque cost. DTC with
performance has
with
To promote performance IPMSM, current vector control comparable FOC. Additionally, steady DTCdynamic
has low torque sensitivity to parameters
To promote
To promote performance
performance of of IPMSM,
IPMSM, current current vector
vector control
control FOC.comparable
FOC. steady
Additionally,
Additionally, and
DTC has
DTC hasBut, low
lowthe sensitivity
sensitivityperformance
to with
parameters
to parameters
(CVC)
To promoteis theperformance
most widely used
of used
IPMSM, approach
currenttotovector
regulate the vibration
control of electric machine. disadvantages of basic
(CVC)
(CVC) is the
is the most
most widely
widely used approach
approach to regulate
regulate the FOC.
the vibration
vibrationAdditionally,
of electric
of DTC hasBut,
electric machine.
machine. lowthe
But, sensitivity
the disadvantages
disadvantagesto parameters
of basic
of basic
torque
(CVC) of IPMSM
is the most as as presented by Macminn (1991) and DTC is also obvious: torque and flux ripple, deteriorated
torque of
torque of IPMSM
IPMSM aswidely used by
presented
presented approach
by Macminn
Macminn to regulate
(1991) and
(1991) the vibration
and DTC
DTC is
is of electric
also
also obvious:
obvious: machine.
torque
torque But,
and
and the disadvantages
flux
flux ripple,
ripple, of basic
deteriorated
deteriorated
Morimoto
torque of IPMSM (1994). When using CVC,
as presented stator current,
by Macminn rather
(1991)ratherand performance atobvious:
low speed, and andvariable switching frequency
Morimoto
Morimoto (1994). When When using CVC,
CVC, stator current,
current, rather DTC is also at
performance at low
low speed,
speed,torque
and variable
variableflux switching
ripple, deteriorated
switching frequency
than torque(1994).
Morimoto and stator
(1994). When
using
flux,
using is CVC, stator
the closed-loop
stator controlled
current, rather
performance
of inverter. and frequency
than torque
than torque and and stator
stator flux,
flux, is is the
the closed-loop
closed-loop controlled
controlled of performance
inverter. at low speed, and variable switching frequency
of inverter.
variable.
than torque In CVC of
and stator IPMSM, voltage
flux, isvoltage space
the closed-loop vectors of
controlledthe
variable.
variable. In CVC
In CVC of IPMSM,
of IPMSM, voltage space vectors
space vectors the of
of the Forinverter.
inverter
variable. are
In the
CVC onlyof input
IPMSM, for voltage
closed-loop space control
vectors of ofstator the defects of basic DTC, much works have been made
inverter are
inverter are the
the only
only input
input forfor closed-loop
closed-loop controlcontrol of of of the For
stator
stator For the
over
the defects
defects of
thedefects
past few
of basic
basic DTC,
decades.
DTC, much much works
DTCmuch combined
works havehave been
with
been made
space
made
vector
current.
inverter The
are stator
the only current
input dynamics
for will
closed-loop affect
controltorque
of and
stator For
over the
the past few of basic
decades. DTC,DTC combinedworks have
with been
space made
vector
current. The
current. The stator
stator current
current dynamics
dynamics will will affect
affect torque
torque and and over the past (DTC-SVM),
modulation few decades. DTC combined
separately with space
proposed by T. vectorG.
stator
current.fluxTheperformance.
stator For dynamics
current EVs, IPMSM will driver
affect will
torquereceive
and over the
modulation past few decades.
(DTC-SVM), DTC combined
separately with
proposed space
by vector
T. G.
stator flux
stator flux performance.
performance. For For EVs,
EVs, IPMSM
IPMSM driverdriver will
will receive
receive Habetler
modulation (1992)(DTC-SVM),
for IM and separately
Zhang (2004)proposed
for by T.is G.
IPMSM, to
torque command from car controller with different operating modulation (DTC-SVM), separately proposed by T.
stator
torque flux
torque The performance.
command
command from
from For
car EVs,
controller
car controller IPMSM
with
with driver
different
different will receive
operating
operating Habetler (1992)
Habetlerconstant
achieve (1992) for for IM and Zhang
IM andfrequency
switching Zhang (2004)(2004)
of
for IPMSM,
for IPMSM,
inverter as well is G.
is
as
to
to
to
mode.
torque command torque, precisely
from car controlled
controller withor not, will
different affect
operatingcar Habetler
achieve (1992)
constant for IM
switching and Zhang
frequency (2004)
of for
inverter IPMSM,
as well is
as to
to
mode. The torque, precisely controlled or
mode. The torque, precisely controlled or not, will affect carnot, will affect car achieve constant switching frequency of inverter as well as to
mode. The torque, precisely controlled or not, will affect car achieve constant switching frequency of inverter as well as to
2405-8963 © 2015, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.
Copyright
Peer review©under
IFAC responsibility
2015 69 Control.
of International Federation of Automatic
Copyright
Copyright © IFAC
© IFAC 2015
2015
10.1016/j.ifacol.2015.09.161 69
69
Copyright © IFAC 2015 69
MICNON 2015
June 24-26, 2015. Saint Petersburg, Russia Weijie Lin et al. / IFAC-PapersOnLine 48-11 (2015) 065–071
66

obtain the desired torque and stator flux with little ripple by decided by stator flux vector. By applying appropriate
synthesizing an appropriate voltage space vector through voltage space vector, basic DTC can control the magnitude
SVM, which is more accurate than that of basic DTC to and angle of stator flux to obtain desired torque. The
compensate the error of desired and actual stator flux. framework of basic DTC is shown in Fig.1.
DTC with deadbeat control, namely DB-DTC, is proposed
for IM by Lee (2002) and Kenny (2003). Theoretically, DB-
DTC can reach the desired torque in one control period with
synthesized voltage space vector. The dynamics and steady-
state error of DB-DTC are better than basic DTC and DTC-
SVM even at low switching frequency, which is necessary
for high power machines used for EVs. DB-DTC for IPMSM
proposed by Lee (2011) contains a discrete time flux observer
for accurate torque and stator flux control. Moreover, a dq
rotating reference frame current observer is designed to
eliminate the sampling delay caused by digital control
implementation.
Fig. 1. Framework of basic DTC.
In this paper, basic DTC, DTC-SVM and DB-DTC will be
comparatively evaluated through simulations with various IPMSM stator winding currents are measured by hall current
criteria. The comparison results will be used as guidance for sensors, and its voltages are calculated by inverter switch
application of IPMSM to industry and electric vehicles. state. The actual stator flux and torque are calculated by flux
and torque observer. The actual stator flux and torque are
2. IPMSM MODEL compared with the reference values in two separate hysteresis
controllers. The hysteresis controllers are two-level
DTC is realized on base of IPMSM model. Expressions of
comparator whose outputs are labelled with  and  . If the
IPMSM in rotating dq reference frame are listed as follows
actual value is less than down bounds of the reference value,
d λd the comparator outputs true (also 1) meaning that the selected

ud = Rsid + dt − ωeλq voltage space vector should increase the actual value. On the
(1) other side, if the actual value is larger than up bounds of the

 u = R i + d λq + ω λ
reference value, the comparator outputs false(also 0) meaning
that the selected voltage space vector should decrease the
 q s q
dt
e d
actual value. According to the outputs of two hysteresis
= λd Ld id + λPM controllers and the stator flux position, an optimal voltage
 (2) space vector will be selected and applied to stator windings to
 λq = Lq iq minimize the error of stator flux and torque in each control
period. The selection of optimal voltage space vector is
where udq and idq are stator voltage and current; Rs is
referred to Table 1. In this table,  is the section of stator
stator resistance; Ld and Lq are d-axis and q-axis stator flux position.
inductance, respectively; dq is stator flux; e is electrical Table 1. Switch table of basic DTC
rotor angular velocity; PM is permanent magnet flux. The
 
torque of IPMSM is 
1 2 3 4 5 6
=Te 1.5 p ( λd iq − λq id ) (3) 1 u6 u2 u3 u1 u5 u4
1
where p is the number of rotator pole pairs. 0 u5 u4 u6 u2 u3 u1
1 u2 u3 u1 u5 u4 u6
3. DTC STRATEGY 0
0 u1 u5 u4 u6 u2 u3
3.1 Basic DTC
The parameters of IPMSM are listed in Table 2. The
For AC machines, the torque is proportional to vector product performance of basic DTC is shown in Fig.2. The bound of
  torque hysteresis control is 5N.m, and that of stator flux is
 s and rotator flux  r . For IPMSM, rotator
of stator flux 0.05Wb. Fig.2 demonstrates torque and stator flux are limited
flux is induced by permanent magnet, then r  PM , within bounds of the two separate hysteresis controllers. The
which is almost a constant. The angle velocity of rotator ripple of torque and stator flux is controllable. The dynamic
varies little when voltage space vector affects on stator response with step-up torque reference is about 0.15ms.
windings during one sample period, so the torque is only

70
MICNON 2015
June 24-26, 2015. Saint Petersburg, Russia Weijie Lin et al. / IFAC-PapersOnLine 48-11 (2015) 065–071 67

3.2 DTC-SVM
Torque(N.m)

60
For basic DTC, one of the only eight voltage space vectors
50 (six non-zero and two zero voltage space vectors ) can be
40 selected and applied to stator windings to minimize the error
of reference and actual of torque and stator flux. But, the
magnitude of voltage space vector is constant, and it will
Stator Flux(Wb)

activate within one whole control period continuously.

0.105 Therefore, the ripple of torque and stator flux are inevitable.
0.1 In order to recede the ripple of torque and flux, a more
accurate voltage space vector should be applied to stator
0.095
windings. Based on space vector synthesis, an arbitrary
desired voltage space vector can be synthesized by two
adjacent non-zero and one zero voltage space vectors. The
100
50 block diagram of DTC-SVM is shown in Fig. 3.
Ia(A)

0
-50
-100
0 0.1 0.2 0.3 0.4 0.5
Time(s)

(a)
Torque(N.m)

45
40
35
Stator Flux(Wb)

0.105
Fig. 3. Block diagram of DTC-SVM.
0.1
0.095 Reference flux vector calculator (RFVC) and space vector
0.2 0.201 0.202 0.203 0.204 0.205 modulator (SVM) substitute for the two hysteresis controllers
Time(s) of basic DTC. Flux and torque observer is just same as that of
basic DTC. Because of the torque's non-linearity to stator
(b) flux angle position, a PI regulator is used for torque variation
to stator flux angle position variation. The reference flux
vector calculator is just shown in Fig.4.
65
60
Torque(N.m)

55
50
45
0.15ms
40
35
Fig. 4. Block diagram of reference flux vector calculator.
0.2997 0.2998 0.2999 0.3 0.3001 0.3002 0.3003
Time(s) The performance of DTC-SVM is shown in Fig.5. The ripple
of torque is about 1.7N.m and the stator flux is nearly
(c) reference value with little error. The steady-state performance
of torque and stator flux is much better than that of basic
Fig. 2. Dynamic and steady performance of basic DTC.
DTC. The dynamic response is about 0.25ms, much longer
Mechanical angle velocity: 100rad/s; torque reference value:
than basic DTC, which is mainly caused by PI regulator
original 40N.m, step up to 60N.m at 0.3s; stator flux
existing in RFVC.
reference value: 0.1Wb. (a) Performance of IPMSM; (b)
Amplification; (c) Dynamics of torque.

71
MICNON 2015
June 24-26, 2015. Saint Petersburg, Russia Weijie Lin et al. / IFAC-PapersOnLine 48-11 (2015) 065–071
68

The torque of IPMSM is expressed as (3). Then, the

corresponding differential equation is
Torque(N.m)

60
dTe d di d di
50  1.5 p ( d iq  d q  q id  q d ) (4)
40 dt dt dt dt dt
The differential frame stator currents and stator flux of d-q
reference frame are substituted into (4) from (1) and (2). Thus,
Stator Flux(Wb)

the differential torque equation can be rewritten as (5),

0.1
0.0999 dTe
0.0998
dt
  Ld  Lq   L  Lq d  PM Lq
100  1.5 p ud q    uq d
50   Ld Lq  Ld Lq
Ia(A)

0 (5)
-50 e
-100 
Ld Lq
 Lq  Ld d  q   LqdPM 
2 2

0 0.1 0.2 0.3 0.4 0.5

Time(s)
Rsq 
(a)

L2d L2q
 L  L 
2
q
2
d d  L2qPM 

and the discrete time form of (5) is
Torque(N.m)

42
41
40
Te (k  1)  Te (k )
39 Ts
38
  Ld  Lq 
 1.5 p ud (k )q (k ) 
Stator Flux(Wb)


0.1001   d q 
 L L
0.1
0.0999  Ld  Lq d (k )  PM Lq
 uq ( k )
0.2 0.202 0.204 0.206 0.208 0.21 Ld Lq (6)
Time(s)
e ( k )
(b)

Ld Lq
 Lq  Ld d (k )2  q (k )2 
 Lqd (k )PM 
60
Rsq (k ) 
  L  L  (k )  L  
2 2 2
Torque(N.m)

55 q d d q PM
L2d L2q 
50
0.25ms Ts in (6) is sampling period. The torque variation in (6) can
45
be expressed as an variable Te by
40
Te (k )  Te (k  1)  Te (k ) (7)
0.2997 0.2998 0.2999 0.3 0.3001 0.3002 0.3003
The relationship of d-q axis stator voltage can be developed
Time(s)
from (6) including Te ( k ) as the desired torque change

Fig. 5. Dynamic and steady performance of DTC-SVM. uq (k )Ts  Mud (k )Ts  B (8)
Mechanical angle velocity: 100rad/s; torque reference value:
where
original 40N.m, step up to 60N.m at 0.3s; stator flux
reference value: 0.1Wb. (a) Performance of IPMSM; (b)
Amplification; (c) Dynamics of torque.  Lq  Ld q (k )
M
 Ld  Lq d (k )  LqPM
3.3 DB-DTC

72
MICNON 2015
June 24-26, 2015. Saint Petersburg, Russia Weijie Lin et al. / IFAC-PapersOnLine 48-11 (2015) 065–071 69

Ld Lq
B
 Ld  Lq d (k )  LqPM

Torque(N.m)
60

 2T (k )  T 50

 3 p  L L  Lq  Ld d (k )  q (k ) 
 e e s 2 2
40
 d q

Stator Flux(Wb)
 Lqd (k )PM  0.101
0.1
RsTsq (k ) 

L2d L2q
 L2q  L2d d (k )  L2qPM  0.099
0.098

Using (8), stator voltage vectors will be calculated to achieve
the desired torque variation over the next sample time, which 100
50
implements deadbeat torque control. But, the stator flux will

Ia(A)
0
not be controlled simultaneously with the torque and vary in -50
random. In order to achieve deadbeat flux control, the d-q -100
axis stator flux vectors should be taken into account and 0 0.1 0.2 0.3 0.4 0.5
imported into torque control. Then, the voltage model of Time(s)
stator flux is introduced as
(a)
dq   udq  Rs idq  dt (9) Torque(N.m)

When the stator resistance is small enough to be negligible, 42

41
the discrete time of stator flux vector is approximately 40
39
dq (k  1)  dq (k )  udq (k )Ts (10)
38
In (10), dq (k  1) is the desired stator flux, that is,
Stator Flux(Wb)

0.101

s* (k )  d (k  1) 2  q (k  1) 2 (11) 0.1

The block diagram of DB-DTC is shown in Fig.6. The flux 0.099

and torque estimation with current observer is different from 0.2 0.202 0.204 0.206 0.208 0.21
that of basic DTC and DTC-SVM. It contains current Time(s)
predictive model composed of PI regulators and current
filters presented by Lee (2011). The DB-DTC controller is (b)
equations with (7), (8), (10) and (11).

60

55
Torque(N.m)

50
0.85ms
45

40

0.2998 0.3 0.3002 0.3004 0.3006 0.3008 0.301 0.3012

Time(s)

(c)
Fig. 6. Block diagram of DB-DTC.
Fig. 7. Dynamic and steady performance of DB-DTC.
The performance of DB-DTC is shown in Fig. 7. The torque Mechanical angle velocity: 100rad/s; torque reference value:
ripple is less than 1N.m, and also less than that of DTC-SVM. original 40N.m, step up to 60N.m at 0.3s; stator flux
But, the stator flux ripple is nearly 1%, larger than that of reference value: 0.1Wb. (a) Performance of IPMSM; (b)
DTC-SVM. The dynamic response is about 0.85ms, much Amplification; (c) Dynamics of torque.
longer than basic DTC and DTV-SVM, mainly caused by PI
regulator existing in stator flux and torque estimation.

73
MICNON 2015
June 24-26, 2015. Saint Petersburg, Russia Weijie Lin et al. / IFAC-PapersOnLine 48-11 (2015) 065–071
70

4. Performance Evaluation

Torque(N.m)
45
4.1 Parameter Sensitivity 40
35
In various applications, IPMSM's parameters may vary with
ambient temperature. The machine's parameters mainly are

Stator Flux(Wb)
stator resistance, d-q axis inductance and rotator permanent 0.105
magnet flux. The parameter sensitivity of control strategies 0.1
will determinate the performance of IPMSM. Here, the 0.095
performance of DTC strategies with variation of machine's
0.2 0.201 0.202 0.203 0.204 0.205
parameters is evaluated through some simulations. Fig. 8 Time(s)
shows the torque ripple and flux ripple of the three different
DTC strategies with the value of stator resistance exceeding (a)
the actual value by 20% and q-axis inductance by less 20%.

Torque(N.m)
The simulation results demonstrate that, when the q-axis 42
41
inductance and stator resistance vary, the steady performance 40
of stator flux and torque are not much deteriorated. The 39
torque ripple gets increased a little while parameters vary. 38
Therefore, the three DTC strategies are non-sensitive to this

Stator Flux(Wb)
machine's parameters variation.
0.1001
0.1
0.0999
4.2 Computational Complexity
0.2 0.202 0.204 0.206 0.208 0.21
For these three DTC strategies, the computational complexity Time(s)
is linear to the core algorithm. The flux and torque observer
of basic DTC and DTC-SVM are identical. This observer (b)
only calculates actual values of stator flux and its angle
position and torque by the measured stator currents and stator
Torque(N.m)

42
voltages deduced by switch state of inverter. But, DB-DTC's 41
40
flux and torque estimation contains current observer with PI 39
and filters, who will consume more computational time. 38
Besides of flux and torque observer, basic DTC needs
Stator Flux(Wb)

minimum time for computation who has only hysteresis 0.101

comparator and voltage selection. On the other hand, DB-
0.1
DTC has the maximum time for computation because of its
computation complexity of reference d-q axis stator voltages. 0.099
DB-DTC have the same space vector modulation as DTC- 0.2 0.202 0.204 0.206 0.208 0.21

SVM for constant stator flux. Time(s)

(c)
4.3 Stator Current THD
Fig. 8. Performance with variation of machine's parameters.
For electric vehicle, power losses of motor and driver affect (a) basic DTC; (b) DTC-SVM; (c) DB-DTC.
the efficiency of car electrical system, and furthermore the
travel distance. The frequency spectrum of stator currents Table 2. Parameters of IPMSM
affects iron losses of motor. For AC machine, the total
harmonic distortion (THD) is widely used for frequency Rated power PN (kW) 20
spectrum performance. When steady, the stator currents and
their THD are shown in Fig.9. The simulation is set at Rated DC voltage U dc (V) 336
mechanical angle velocity of 100rad/s constantly and the Rated Torque Te (N.m) 64
desired torque of 40N.m for different DTC strategies.
Stator Resistance Rs (  ) 0.02
From Fig.9, it can be seen that basic DTC has the highest
THD, while DTC-SVM and DB-DTC have almost equivalent Inductance of d axis Ld (mH) 0.336
lower THD. Correspondingly as formulations before, basic
DTC has the highest torque ripple and flux ripple. The
Inductance of q axis Lq (mH) 0.5
highest stator current THD of basic DTC is primarily caused Permanent Magnet Flux of
by the torque ripple and flux ripple, which could be improved 0.1
Rotator PM (Wb)
by decreasing the hysteresis comparator bounds and
enhancing sampling and switching frequency of inverter. Pole pairs p 4

74
MICNON 2015
June 24-26, 2015. Saint Petersburg, Russia Weijie Lin et al. / IFAC-PapersOnLine 48-11 (2015) 065–071 71

ACKNOWLEDGEMENT
80 This work is financially supported by National Nature
40 Science Foundation of China under Grant 51277048, Nature
Ia(A)

0
-40
Science Foundation of Zhejiang Province under Grant
-80 LZ13E070002 and LQ13F030014, Science and Technology
0.2 0.25 Project of Zhejiang Province under Grant 2013R40030.
Time(s)
REFERENCES
Chan, C.C. and Chau K.T. (1996). An advanced permanent
magnet motor drive system for battery-powered electric
vehicles. IEEE Transactions on Vehicular Technology,
45(1), 180-188.
Depenbrock, M. (1988). Direct self-control (DSC) of
inverter-fed induction machine. IEEE Transactions on
Power Electronics, 3(4), 420-429.
(a) Habetler, T.G., Profumo, F., Pastorelli, M., and Tolbert, L.M.
(1992). Direct torque control of induction machines
using space vector modulation. IEEE Transactions on
80 Industrial Applications, 28(5), 1045-1053.
40 Kenny, B.H., and Lorenz, R.D. (2003). Stator and rotor flux
Ia(A)

0
based deadbeat direct torque control of an induction
-40
-80 machine. IEEE Transactions on Industrial Applications,
0.2 0.25 39(4), 1093-1101.
Time(s) Lee, J.H., Kim, C.G., and Youn, M.J. (2002). A dead-beat
type digital controller for the direct torque control of an
induction motor. IEEE Transactions on Power
Electronics, 17(5), 739-746.
Lee, J.S., Choi, C.H., Seok, J.K., and Lorenz, R.D. (2011).
Deadbeat-Direct Torque and Flux Control of Interior
Permanent Magnet Synchronous Machines With
Discrete Time Stator Current and Stator Flux Linkage
Observer. IEEE Transactions on Industrial Applications,
47(4), 1749-1758.
(b) Macminn, S.R., and Jahns, T.M. (1991). Control techniques
for improved high speed performance of interior PM
80
synchronous motor drives. IEEE Transactions on
40 Industrial Applications,27(5), 997-1004.
Ia(A)

0 Morimoto, S., Sanada, M., and Takeda, Y. (1994). Wide-

-40 speed operation of interior permanent magnet
-80 synchronous motors with high-performance current
0.2 0.25 regulator. IEEE Transactions on Industrial
Time(s) Applications,30(4), 920-926.
Tang, L.X., Zhong, L., Rahman, M.F., and Hu, Y.W. (2004).
A novel direct torque controlled interior permanent
magnet synchronous machine drive with low ripple in
flux and torque and fixed switching frequency. IEEE
Transactions on Power Electronics, 19(2), 346-354.
Takahashi, I., and Noguchi, T. (1986). A new quick-response
and high-efficiency control strategy of an induction
motor. IEEE Transactions on Industrial Applications,
22(5), 820-827.
(c) Zhong, L., Rahman, M.F., Hu, Y.W., and Lim, K.W. (1997).
Analysis of direct torque control in permanent magnet
Fig. 9. THD for stator current.(a)Basic DTC: Fundamental synchronous motor drives. IEEE Transactions on Power
(60Hz) = 67.37, THD= 16.63%; (b)DTC-SVM: Fundamental Electronics, 12(3), 528-536.
(60Hz) = 67.77, THD= 7.59%; (c)DB-DTC: Fundamental
(60Hz) = 67.71, THD= 7.57%.

75