[Type text] [Type text] [Type text]

  

   

2014

 

© Trade Science Inc.  

ISSN : 0974 - 7435 Volume 10 Issue 24

BioTechnology

An Indian JournalFULL PAPER

BTAIJ, 10(24), 2014 [15049-15059]

Novel MC-DTC control method for induction motor based on space vector modulation

Cai Bin-jun1,2, Nian Xiao-hong1,*

1School of Information Science and Engineering, Central South University, Changsha 410004, (P.R.CHINA)

2College of Electrical & Information Engineering, Hunan Institute of Engineering, Xiangtan 411101, (P.R.CHINA)

E-mail : xhnian@mail.csu.edu.cn

ABSTRACT This paper presents a novel MC-DTC method for fed induction motor based on spacevector modulation. The advantages of DTC method are combined with the advantages ofthe matrix converter based on space vector modulation technique. This proposed novelmethod provides a precious input power factor control capability beside the high controlperformances. Furthermore, Conventional principles of DTC and MC were described. Thecombination of the DTC and MC were given in details. The simulation and experimentresearch were carried out to identify the new method effectiveness. The results ofinduction motor control at steady state are shown to improve the low-speed performanceand strong adaptability of this novel control strategy. KEYWORDS Matrix converter; DTC; Induction motor; Space vector modulation; Low-speedperformance.

15050 Novel MC-DTC control method for induction motor based on space vector modulation BTAIJ, 10(24) 2014

INTRODUCTION

In two recent decades, due to the need to increase the quality and the efficiency of the power supply and usage, three phase matrix converter becomes a modern energy converter. The research of matrix converter has been carried on with many theoretical[1],[2],[3].These achievements along with the emergence of bidirectional switch make it possible to apply the matrix converter to practical applications. Various methods to control the matrix converter have been proposed [4],[5], being the scalar modulation and the indirect space vector modulation widely used. It fulfills all requirements of the conventionally used rectifier/dc link/ inverter structures. Some advantages of the matrix converter can be seen as following: the use of a compact voltage source, providing sinusoidal voltage with varying amplitude and frequency besides the sinusoidal input current and unity input power factor at power supply side. Matrix converter has a simple topology and a compact design due to the lack of dc-link capacitor for energy storage.

Since the DTC method has been proposed in the middle of 1980’s, DTC method becomes one of the high performance control strategies for AC machine to provide a very fast torque and flux control[6][7]. There are no requirements for coordinate transformation, no requirements for PWM generation and current regulators. It is widely known to produce a quick and fast response in AC drives by selecting the proper voltage space vector according to the switching status of inverter which is determined by the error signal of reference flux linkage and torque with their estimated values and the position of the estimated stator flux. Some research is being done to adapt DTC to new converters and also to reduce the torque ripple, which is one of its main drawbacks. DTC is the direct control of torque and flux of a drive by the selection, through a look-up table, of the inverter voltage space vectors. The main advantage of DTC is its structure, no coordinate transformations and no PWM generation are needed. However, torque and flux modulus values and the sector of the flux are needed. Not only it is a very simple and robust signal processing scheme but also a very quick and precise torque control response is achieved.

In this paper, a novel MC-DTC method for fed induction motor based on space vector modulation is proposed. The advantages of DTC method are combined with the advantages of the matrix converter based on space vector modulation technique.The appropriate switching configurations of the matrix converter for each constant time are presented in an opportune switching table[8],[9],[10]. The table is only entered by the imaginary voltage vector, which is generated from the DTC method for voltage source inverter, and the position of input voltage vector which can be measured exactly. Simulation and experiment at the high-speed and low-speed are carried out to prove the good performances of the novel method.

CONVENTIONAL DIRECT TORQUE CONTROL

Mathematical mode of induction machine

The mathematical mode of induction machine is shown in Figure 1.

Figure 1 : The mathematiacal mode of induction motor

According to Figure 1 flux-linkage equations of induction machines in the stator stationary reference frame as follows.

( )s s s sv R i dtα α αψ = −∫ (1)

( )s s s sv R i dtβ β βψ = −∫ (2)

Where sαψ and are theα -axis and β -axis component of sψ

r respectively; svα and svβ are the α -axis and β -axis

component of svr

respectively; siα and siβ are theα -axis and β -axis component of sir

. The electromagnetic torque can be expressed using the following equation.

3 3( ) ( )2 2e p s s p s s s sT n i n i iα β β αψ ψ ψ= × = −

rr (3)

Where eT is electromagnetic torque and pn is the number of rotor pole pairs.

sβψ

BTAIJ, 10(24) 2014 Cai Bin-jun and Nian Xiao-hong 15051

Principle of DTC The basic principle in conventional DTC for induction motors is to directly select stator voltage vectors by means of

a hysteresis stator flux and torque control. As it is shown in Figure 2.

*sψ

sψψE

ψH

TeH

*eT

eT

TeE

SASBSC

eT

sψ AIBI

rω

Figure 2 : The diagram block of basic DTC

From Figure 2 can obtain stator flux *sψ and torque *

eT references are compared with the corresponding estimated values. Both stator flux and torque errors, ψE and TeE , are processed by means of a hysteresis band comparators. In particular, stator flux is controlled by a two-level hysteresis comparator, whereas the torque is controlled by a three-level comparator. On the basis of the hysteresis comparators and stator flux sector a proper VSI voltage vector is selected by means of the switching table given in Table 1.

TABLE 1 : Basic DTC switching table

MC SPACE VECTOR MODULATION

Working principle of MC MC is an AC-AC converter, with mxn bidirectional switches, which connects an m-phase voltage source to an n-phase load. The three-phase, 3x3 switches, MC shown in Figure 3 is the most interesting. It connects a three phase voltage source to a three-phase load[11].

Figure 3 : The topology of matrix converter

In the MC shown in “Fig.3,” siv , { }, ,i A B C= are the source voltages, sii , { }, ,i A B C= are the source currents. jNv ,

{ }, ,j A B C= are the load voltages, ji , { }, ,j A B C= are the load currents, iv , { }, ,i A B C= are the MC input voltages and ii ,

{ }, ,i A B C= are the input currents. A switch, ijS , { }, ,i A B C= , { }, ,j a b c= can connect phase i of the input to phase j of

the load. With a suitable switching strategy, arbitrary voltages jNv at arbitrary frequency can be synthesized. Switches are characterized by the following equation.

01

ijij

ij ij

witch S is openS

witch S S close⎧⎪= ⎨⎪⎩

(4)

A mathematical model of MC can be derived from “Fig.3” as follows:

1

D

b

D

sio

z

d d d Wm

15052

The coDC and DC-AC

The cobidirectional sw

DC-AC conver 6 pow

space vector Us 60 degrees. T

output , ,A B C a The ou

Any ezero output vol

Where

MM TTd δ ==

NN mTTd δ ==

M ddd −−=10

Where ,M NT T modulation ind

No

onventional thrC. It is shown i

ontrol signal fowitches corresp

rter space vecwer switches of

1 6U U− and 2 zThey constituteand the input Dutput voltage a

expected outpultage 0U or 7U

e NM dd , and

( Jvm θ−060sin

Jvm θsin

N

is the switchindex of output vo

ovel MC-DTC con

⎢⎢

⎣

⎡

vvv

cN

bN

aN

(((⎢

⎢

⎣

⎡

iii

C

B

A

ree-phase to thin Figure 4.

Figure

for bidirectionapond to a 3x3 m

ctor modulatiof inverter with ero vector 0 ,Ue 6 uniform seDC , such as M

and the corresp

Figure 5 :

ut voltage spac. Suppose the

U

0d are the ratio

)J (8)

(9)

(10)

ng time of vecoltage. And

ntrol method for in

( )( )( )

( )( )( )⎢

⎢

⎣

⎡=

⎥⎥

⎦

⎤

tStStS

ttt

Ac

Ab

Aa

N

N

N

( )( )( )

( )( )( )⎢

⎢

⎣

⎡=

⎥⎥

⎦

⎤

tStStS

ttt

Ca

Ba

Aa

hree-phase matr

e 4 : The topol

al switches commatrix in each s

on 8 possible com

7,U . The phasegments. The th

101M = whichponding switchi

The output vo

e vector JU isangle between

dUdU MMJ +=

o cycles of MU

ctors MU and

nduction motor ba

( ) (( ) (( ) ( )tStS

tStStStS

CcBc

CbBb

CaBa

( ) ( )( ) ( )( ) ( )⎥

⎥tStStStStStS

CcCb

BcBb

AcAb

rix converter’s

logy of 3x3 m

me from the coswitching perio

mbinations showe angle betweehree digits in bh represents theing states are r

oltage vector a

formed by ad

n JU and MU

00UdUd NN +

NM U, and 0U r

MU respectiv

ased on space vect

)))

( )( )( )⎥⎥

⎦

⎤

⎢⎢

⎣

⎡⋅⎥⎥

⎦

⎤

tvtvtv

C

B

A (5)

) ( )( )( )⎥⎥

⎦

⎤

⎢⎢

⎣

⎡⋅⎥⎥

⎦

⎤

tititi

c

b

a (6)

modulation pr

matrix convert

ontrol circuit aod.

wn in Figure 3en one effectivebrackets exprese switching of trepresented in F

and switching

djacent two basis Jθ .

(7)

respectively. A

vely. Tδ is the

tor modulation

rocess consists

ter

and drive circu

3 are correspone voltage spacess the linking sthe switches SFigure 5.

g states

sic output volt

And

switching per

BTAIJ,

s of two proces

uits. The ratio c

nding to effectie vector and adstate between t

pCnBpA SSS ,, .

tage vectors U

riod of PWM.

J, 10(24) 2014

ses of AC-

cycles of 9

ive voltage djacent one three-phase

NM UU , and

vm is the

B

Wm

a

A

tor

M

pTiner

N

Dd

BTAIJ, 10(24) 201

Where omU an1=cm ,φ is the

When adjacent basic v

AC-DC conver The sp

opology is reprectification, th

MC space vect Three-

possible switchTable 2: group nput phase, gr

each SCs, the crepresented in F

Novel DTC-M The no

DTC for matridesired imagina

14

nd imU are the e input power fthe rotating sp

voltage space v

rter space vecpace vector mpresented in th

he DC voltage

tor modulatio-phase matrix hing configura

I (±1, ±2, …,roup II (0a, 0b, corresponding Figure 6.

MC based on SVovel DTC-MCix converter[12ary non-zero V

vm

amplitude of ofactor angle. pace vector JUvectors constitu

ctor modulatiomodulation proc

he left dotted is.

dU

n converter mo

ations (SCs), o ±9) consists o0c) consists ofoutput line-to

Figure 6 : (a

T

VM method will a2],[13]. Accor

VSI voltage vec

Cai Bin-ju

( ) (32 21omU=

output and inpu

J locates in a uting this segm

on cess of AC-DC

line frame of

1.5 cod c imm U=

odule includes nly 21 SCs caof the SCs whif the SCs whic-neutral voltag

a) The output v

TABLE 2 : MC

apply the directding to the inctor, the two no

un and Nian Xiao-

( )φcoscimmU

ut voltage, cm

segment, the lment and one ze

C is completelf “Fig.4”. The

osϕ

nine bidirectian be used to ich have two och have all outge vector and i

voltage; (b) T

C active and ze

t SVM techniqnput voltage linon-zero voltage

-hong

(11)

c is the input c

ocal average oero voltage spa

ly similar to the correspondin

(12)

ional switchesimplement the

output phases ctput phases coinput line curr

he intput line

ero vectors

que to overcomne to neutral ve vectors will b

current modula

of output voltagace vector.

he modulation ng formulas ar

as shown in e DTC algorithconnected to thnnected to a c

rent vector hav

current

me the disadvanvector sector lbe selected.

ation index, ge

ge can be form

process of Dre similar as w

Figure 6. Thhm for MC ashe same one o

common inputve the fixed di

ntages of the colocation, to co

15053

enerally set

med by two

DC-AC. Its well. After

ere are 27 s shown in of the other phase. For rections as

onventional ombine the

15054 Novel MC-DTC control method for induction motor based on space vector modulation BTAIJ, 10(24) 2014

The criteria utilized to implement the switching patterns for the matrix converter can be explained referring to the following example.

We can assume the imaginary VSI voltage vector is 1V , and the input voltage lint-to-neutral vector [14],[15] is located in sector 1 as shown in Figure 7.

Figure 7 : Block diagram of novel DTC-MC based on SVM

From Table 2, in order to generate a voltage vector in the same direction of 1V , there are 6 possible SCs (±1, ±2, ±3). According to the input voltage vector location, there are only 3 SCs having the voltage vectors as same direction to 1V : +1, -2 and -3. To synthesize the input current vector to be in phase with the input voltage vector located in sector 1, two SCs finally selected are +1 and -3. The switching table based on these criteria is shown in Table 3.

TABLE 3 : MC-DTC switching table using SVM

As shown in Figure 7, the output voltage of matrix converter for each SCs is calculated.

aA aB aC a a

S bA bB bC b b

cA cB cC c c

S S S V VV S S S V T V

S S S V V

⎡ ⎤ ⎡ ⎤ ⎡ ⎤⎢ ⎥ ⎢ ⎥ ⎢ ⎥= =⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎣ ⎦ ⎣ ⎦ ⎣ ⎦

(13)

Where the switching function ijS is 1 when the switch joining input line i to output line J is ON and i is 0 otherwise,

, ,i a b c= and = , ,J A B C . Matrix T represents the status of each switching configuration in Table 2. The input voltage vector of induction motor in the stationary reference frame for each sampling period.

1 2

1 112 2 23 3 30

2 2

asd

s x y bsq

C

Vv

v t T t T Vv

V

⎡ ⎤ ⎡ ⎤− −⎢ ⎥⎡ ⎤ ⎢ ⎥⎢ ⎥ ⎡ ⎤= = +⎢ ⎥ ⎣ ⎦ ⎢ ⎥⎢ ⎥⎣ ⎦ ⎢ ⎥⎣ ⎦⎢ ⎥⎣ ⎦

(14)

From Figure 6, the duty ratios of the two non-zero Scs. are calculated as follows.

1 2

21

1 2

sin( 6 ) sin( 6 )1

x y VSI

yx

i i

t V t V Vt Vt V

t tπ α π α

+ =

=− +

+ =

（15）

B

T

pc

thc

T

r

w

BTAIJ, 10(24) 201

(

(

1

2

sinco

sinco

t

t

π

π

=

=

The in

The m

The simulation In ord

period of 50μs.cage induction

=2.2nP kw ,U

The w

he matrix outpcurrent is calcu

The simulation At the

reference 0.6w Figure

waveform.

14

)( )

)( )

6os

6os

i

i

i

i

π αα

π αα

−

+

nduction motor

motor estimated

n model of MCder to verify th

The system simotor:

=380nU v , =4sR

whole system haput voltages anulated on the ba

n result of MCe high speed o

wb . The simulate 9 shows a go

(16

r stator flux can

d torque can be

SIMU

C-DTC based e behavior of timulation mod

.35Ω , =0.43rR Ω

as been simulad the input curasis of the matr

Figure 8 : T

C-DTC based operation, indution result show

ood performanc

Figu

Cai Bin-ju

6)

n be obtained f

(s s s sv R iϕ = −∫)

e obtained.

(32

psd sq

nT iϕ= −) )

ULATION OF

on SVM the proposed s

del is shown in

Ω , =2sL mH , rL

ated using the Srrents respectivrix input curren

The simulatio

on SVM uction motor iw in Figure 9-Fce in terms of

ure 9 : Flux at

un and Nian Xiao-

from the calcul

)s dt

)sq sdiϕ− )

MC-DTC BA

scheme, some Figure 8. The

=2r mH , =69mL

Simulink packvely, thus assunt.

n model MC-D

is running at Figure 10. stator flux. As

100rpm , 25Nm

-hong

lated input volt

(17)

(18)

ASED ON SVM

simulation hasmachine utiliz

9.31mH , 0.0J =

kage. Equation uming ideal sw

DTC based on

speed 1000rpm

s it can be seen

for MC-DTC

tage and the m

M

s been carried zed for simulati

2089kg m⋅ , =2P

(5) and Equatiwitching device

n SVM

m , rated load

n in Figure 9, s

C

easured stator

out assuming aions is a three-

.

ion (6) are usees. The mains f

torque 25Nm

stator flux show

15055

currents.

a sampling -phase 3kw

d to obtain filtered line

m and flux

ws a circle

1

Fs

r

cc

7fc

15056

FigureFurthermore, thsystem with reg

At thereference 0.6 w

Figurechange as a stecurrent wavefo

The n7.5kw MC. Thefour-step commcontrolled hyste

No

e 10 shows stathe stator curregard to the impe low speed op

wb . The simula

e 12 shows theep command frrms are almost

novel method he MC-DTC algmutation proceeresis brake pr

ovel MC-DTC con

Figure 10 : C

tor flux and elents have sinuplementation ofperation, induc

ation result show

Figur

Figure 13 :

e dynamic perfrom 25Nm+ tot sinusoidal im

EXPE

had been testegorithms were ess, required wrovides the load

 

 

ntrol method for in

Current flux a

ectromagnetic usoidal wavefof the novel MCction motor isw in Figure 11

re 11 : Flux at

Torque, curre

formances of tho 25Nm− . The

mmediately righ

ERIMENT OF

d. The parameimplemented

when bidirectiod torque. A blo

nduction motor ba

and torque at

torque of the

orms. This figC-DTC methods running at a- Figure 12.

t 100rpm , 25Nm

ent and flux at

he novel contre electromagneht after the step

F MC-DTC BA

eters are samein a TMS320Lonal switches ock diagram of

ased on space vect

1000rpm for M

induction motogure emphasized.

very low spee

for MC-DTC

t 100rpm for M

rol method at thetic torque shop command.

ASED ON SVM

e as simulationLF2407 DSP aare used, was

f the system set

tor modulation

MC-DTC

or fully followes the good p

ed 100rpm , loa

C

MC-DTC

he rotor speedows a very goo

M

n parameters. Tachieving a sams implementedtup is shown in

BTAIJ,

wing the referenperformance of

ad torque 20Nm

d 100rpm and lod response and

The motor wample period of d in a FPGA. n Figure 13.

J, 10(24) 2014

nce values. f the drive

m and flux

load torque d the stator

as fed by a f 50μs. The A current-

B

r

sfhbth

FD

r

BTAIJ, 10(24) 201

At thereference 0.6w

It can speed respectivflux. Furthermohysteresis bandbands in the prhe torque towa

Figure 15 : DuDTC

At thereference 0.6w

14

e high speed owb . The experim

be seen from Fvely. As regardore, in the casds in the proporoposed methodards its referen

uring a load to

e low speed opewb 0. The exper

Figure 13 :

operation, indument result sho

Figure 14 :

Figure 14-Figud the ripple, these of torque, thosed method ard can be seen ace value.

orque step com

eration, inductiriment result sh

Cai Bin-ju

The system s

uction motor iow in Figure 14

The result of

ure 15, electrome proposed novhe type of the re identical to tas security lim

mmand from

ion motor is ruhow in Figure 1

Wb

/sβ

ψ

un and Nian Xiao-

setup of MC-

is running at4-Figure 15.

flux at 1000rpm

magnetic torqu

vel method cleaapplied vectorthe torque hys

mits above whic

25Nm+ to 25N−

unning at a very16-Figure 17.

Wb/sαψ

-hong

-DTC based on

speed 1000rp

m , 25Nm for MC

ue, stator currearly improves rs is also showteresis bands ich the large ve

Nm the torque

y low speed 10

n SVM

m , rated load

C-DTC

ent, and stator the performan

wn. At the samin the classicalectors are used

e, current and

00rpm , load to

d torque 0.2Nm

flux performance of both the

me time, the inl method. Thusin order to qu

flux at 1000rp

orque 25Nm an

15057

m and flux

nce at high torque and

nner torque s, the outer

uickly force

m for MC-

nd flux

1

FD

H

s

sclosm

R

[

[

[

15058

Figure 17 : DuDTC

FigureHowever, this i

In ordseveral tests we

This psuccessfully cocontrols the indow and high sp

strategy shows method and it c

ProjecResearch Fund

1] Guo yougtheroy and

2] Ding wei2002,24(2

3] Lan zhiyojournal of

No

uring a load t

e 16-Figure 17 is not seen as a

der to study theere carried out

paper presents ombined with tduction motor peed range arethe better inpu

can make the fl

ct U1134108, 6of Hunan Prov

gui, Zhu jianlid applications,i, Zhu jianlin, 2):185-187. ong, Zhu jianlif xiangtan univ

ovel MC-DTC con

Figure

torque step co

show that the a constraint sine dependency oat low-speed. T

a new DTC-SVthe SVM methrequirements i

e shown to valiut current harmlux and torque

61075065, 613vincial educati

in，“New mo,2006,23(4):54“Matrix conv

in, “Simulationversity, 2010,27

ntrol method for in

(Wb)

sΒ

ψ

16 : The flux

ommand from

stator flux,torqce a re-design

of the torque riThe standard d

CO

VM method fohod on matrixis suggested. Tidate the effec

monic spectrumsmall and stab

ACKNO

321003, 51177on department

RE

dulaiton strate2-546. erter and its r

n reearch of th7(3):110-115.

nduction motor ba

(Wb)sαψ

at 100rpm , 25N

m 25Nm+ to 2−

que in the low of the MC inp

ipple with respdeviation of the

ONCLUSION

or matrix conv

x converter. A The simulation tiveness of the

m and low-speeble. It has adva

OWLEDGMEN

7040 supportedt.

FERENCES

egy on raising

research situati

hree phase to t

ased on space vect

Nm for MC-DT

5Nm the torqu

frequency rangut filter would

pect to both thee torque was ca

verter. The advnew switchingand experimen

e new control sed performanceantages and goo

NTS

d by NSFC/Pr

g voltage trans

ion,” natural s

three phase spa

tor modulation

TC

ue,current and

ge when compd be enough to oe torque referenalculated to me

vantages of theg table for thent results on thscheme. Furthee as compared od future, it is w

roject 13B014

sfer ratio for m

science journa

arse matrix con

BTAIJ,

d flux at 100rpm

ared with the hovercome this nce and the moeasure the torqu

e DTC method e DTC-SVM whe induction mermore, the novto the conventworth further s

Supported by

matrix convert

al of xiangtan

nverter,” natur

J, 10(24) 2014

m for MC-

high speed. drawback. otor speed, ue ripple.

have been which fully motor at the

vel control tional DTC studying.

y Scientific

ter,”control

university,

ral science

BTAIJ, 10(24) 2014 Cai Bin-jun and Nian Xiao-hong 15059

[4] CASDEI, D; SERRA, G; TANI, A; ZARRI, L. “Matrix Converter Modulation Strategies: A New General Approach on Space-Vector Representation of the Switch State”, IEEE Trans. On Industrial Electronics, Vol. 49,No. 2, April 2012.

[5] CASDEI, D; SERRA, G; TANI, A. “The Use of Matrix Converters in Direct Torque Control of Induction Machines”, IEEE Trans. On Industrial Electronics, Vol. 48, No. 6, December 2010.

[6] Romeo Ortega, Nikita Barabanov & Gerardo Escobar Valderrama, “Direct Torque Control of Induction Motors: Stability Analysis and Performance Improvement”, IEEE transactions on automatic control, Vol. 46, No. 8, August 2011.

[7] TAKAHASHI, I; NOGUSHI, T. “A New Quick-Response and High-Efficiency Control Strategy of an Induction Motor”, IEEE Trans. Industry Applications, Vol. 1A-22, pp 820-827, October 2006.

[8] DEPENBROCK, M. “Direct Self-Control (DSC) of Inverter-Fed Induction Machine”, IEEE Trans. on Power Electronics, Vol. 3, No 4, pp 420-429, October 2008.

[9] LEE K B, BLAABGERG F, “improved direct torque control for sensorless matrix converter drives with constant switching frequency and torque ripple reduction”, international journal of control, automation and system, 2012,4(1):113-123.

[10] C. Lascu, I. Boldea, and F. Blaabjerg, “Very-low-speed variable-structure control of sensorless induction machine drives without signal injection,” IEEE Trans. Ind. Appl., vol. 41, no. 2, pp.591-598, Mar./Apr. 2010.

[11] C. Lascu, I. Boldea, and F. Blaabjerg, “Direct torque control of sensorless induction motor drives: a sliding-mode approach,” IEEE Trans. Ind. Appl., vol. 40, no.2, pp.582-590, Mar/Apr. 2011.

[12] Z. Xu and M. F. Rahman, “Direct torque and flux regulation of an IPM synchronous motor drive using variable structure control approach,” IEEE Trans. Power Electron., vol.22, no. 6, pp.2487-2498, Nov. 2007.

[13] V. I. Utkin, “Sliding mode control design principles and applications to electric drives,” IEEE Trans. Ind.Electron., vol.40, no. 1, pp.23-36, Feb. 2009.

[14] C. Lascu and A. M. Traynadlowski, “Combining the principles of sliding mode, direct torque control, and space vector modulation in a high-performance sensorless AC drive,” IEEE Trans. Ind. Appl., vol.40, no. 1, pp. 170-177, Jan./Feb. 2011.

[15] A. Naassani, E. Monmasson, and J. P. Louis, “Synthesis of direct torque and rotor flux control algorithms by means of sliding-mode theory,” IEEE Trans. Ind.Electron., vol.52, no.3, pp. 785-799, June. 2011.