You are on page 1of 12

International Journal of Electronic Engineering Research

ISSN 0975 - 6450 Volume 1 Number 3 (2009) pp. 233244


Research India Publications
http://www.ripublication.com/ijeer.htm

AI based Direct Torque Fuzzy


Control of AC Drives
1

Jagadish H. Pujar and 2S.F. Kodad

Research Scholar (JNTU, Hyderabad) & Faculty, Department of EEE, BVB


College of Engg. & Tech., Hubli, India
E-mail: jhpujar@bvb.edu
2
Professor & Head, Department of EEE, Auroras Engineering College,
Hyderabad, India
E-mail: kodadsf@rediffmail.com

Abstract
Fuzzy logic is recently getting increasing emphasis in soft control
applications. This paper presents an application of fuzzy logic to control the
speed of an AC drive system is proposed for space vector pulse width
modulated (SVPWM) voltage source inverter-fed induction motor drive. This
paper presents the results of an investigation into the suitability of a Direct
Torque Fuzzy Control (DTFC) method for an induction motor drive system.
Direct torque control is known to produce quick and robust response in AC
drive system. However during steady state, torque, flux and current ripple
occurs. An improvement of the AC electric dive speed regulation can be
obtained using a new DTFC method based on space vector modulation (SVM)
which reduces the torque and flux ripples. To validate the proposed method,
the simulation has been carried out using MATLAB-SIMULINK. Simulated
results presented in this paper shown to have low current distortion, low
torque ripple, and a fast torque and speed responses. The simulation results
verify the superiority of the proposed method to the conventional DTC
method.
Keywords: Fuzzy logic, Direct Torque Control (DTC), Direct Torque Fuzzy
Control (DTFC), Induction Motor (IM), Space Vector Modulation (SVM),
switching table.

234

Jagadish H. Pujar and S.F. Kodad

Introduction
Fuzzy logic is recently getting increasing emphasis in soft computing applications in
the recent days. The fuzzy logic control technique has been an active research topic in
automation and control theory since the work of Mamdani proposed in 1974 based on
the fuzzy sets theory of Zadeh proposed in 1965 to deal with the system control
problems which are not easy to be modeled [1]. Hence, the development of highperformance control strategies for AC motor drives resulted in a rapid evolution. One
of the most popular methods, known as field oriented control has been proposed by F.
Blaschke[2]. In the vector control the motor equations are transformed in a field
coordinate system that rotates in synchronism with the rotor flux vector and hence
FOC controls the induction motor in a same manner as separately excited DC motor.
The disadvantage of this control scheme is inclusion of the pulse encoder, indirect
torque control and also it is quite complex due to reference frame transformations. To
overcome these disadvantages, in the middle of 1980s, a new quick response
technique for the torque control of induction motors was proposed by Takahashi as
direct torque control (DTC) [3]. DTC provides very quick response with simple control
structure and hence, this technique is gaining popularity in industries [4]. Though DTC
has high dynamic performance, it has few drawbacks such as high ripple in torque,
flux, current and variation in switching frequency of the inverter. The effects of flux
and torque hysteresis band amplitudes in the induction motor drive performance have
been analyzed in [5]. To improve the performance of conventional DTC, modern
resources of artificial intelligence like neural networks, fuzzy logic and genetic
algorithms are implemented [6]. In the following, we will describe the application of
fuzzy logic in DTC [7] [8].
The main objective of this paper was to develop the digital simulation software for
direct torque fuzzy control of SVPWM voltage source inverter-fed induction motor
drive to reduce the ripples at all modulation indices and to improve the speed
performance, under transient and steady state uncertainties caused by variation in load
torque replacing conventional DTC by fuzzy based DTC, yields a new type of
induction motor control method called as DTFC. The MATLAB SIMULINK is used
to perform the simulation. The simulated results of this method are discussed and
compared with conventional DTC.

Principle of Conventional DTC


DTC principle is widely employed for induction motor drives with fast dynamics [6].
The main notion of the conventional DTC is the rate of change of torque is
proportional to the instantaneous slip between the stator flux and rotor flux under
constant stator flux linkage [5]. DTC has been widely recognized for its fast and robust
torque and flux control. The rotor flux linkage changes slowly compared to the stator
flux linkage, as the rotor time constant of a standard squirrel-cage induction machine is
very large. However, the rotor flux is almost unchanged during a short transient. Thus
rapid changes of the electromagnetic torque can be produced by rotating the stator flux
in the required direction, as directed by the torque command. On the other hand the
stator flux can instantaneously be accelerated or decelerated by applying proper stator

AI based Direct Torque Fuzzy Control of AC Drives

235

voltage phasors. Thus, the simultaneous and decoupled control of torque and flux is
achieved by direct adjustment of the stator voltage in response to the torque and flux
errors [9]. The DTC regularly applies the appropriate voltage vector in order to
maintain the torque and stator flux within two hysteresis bands which results bangbang
behavior and produces variation in witching frequency and significant ripple in flux,
torque and current [8].

DTFC with Voltage Source Inverter


The schematic of the basic functional blocks used to implement the proposed DTFC of
induction motor drive is shown in Fig.1. A voltage source inverter supplies the motor
and instantaneous values of the stator flux and torque are calculated from stator
variable by using a closed loop estimator [3]. Stator flux and torque can be controlled
directly and independently by properly selecting the inverter switching.

ref

DTFC

Tref

Tem

VS

S1
Switching S2
Table
S3

Flu x
sector

VDC

PWM
Inverter

S
V
M

Flu x and
Torque
Estimator
AC
MO TOR

Figure 1: Direct Torque Fuzzy Control scheme for AC motor drives.

A. Vector Model of Inverter Output Voltage


The SVPWM technique is used to approximate the voltage vector by employing the
one out of eight possible combination of vectors generated by 3- phase voltage
source inverter for AC motor drives is as shown in Fig.2.

236

Jagadish H. Pujar and S.F. Kodad

S1

S2

S3

A
B

VDC

N
C

S1

S2

S3

Inducton
Motor

Figure 2: SVPWM Iverter for AC motor drives.

In a 3-, voltage source inverter, the switching commands of each limb are
complementary. So, for each limb, logic state Si (where i=1 to 3) is ON (1) or OFF
(0) can be defined. As there are three independent limbs, there will be eight different
logic states, provides eight different voltages obtained applying the vector
transformation described as:
Vs =

2
4
j
j
2
VDC S1 + S2e 3 + S3e 3
3

(1)

Eight switching combinations can be taken according to the above expression (1).
The partitions of d-q plane in to two zero voltage vectors and six non-zero voltage
vectors are show in Fig.3.

Figure 3: Partition of the d-q planes in to six angular sectors.

B. Stator Flux and Torque Estimation


The components of the current (Id, Iq) and stator voltage (Vd, Vq) are obtained by the
application of the transformation [3] given by (2) to (5). The components of the stator

AI based Direct Torque Fuzzy Control of AC Drives

237

flux (d, q) given by (6) and (7) . The stator flux linkage per phase and the
electromagnetic torque estimated are given by (8) and (9) respectively.
2
(2)
Id =
IA
3

1
(I B I C )
Iq =
2
2
1

V DC S 1 (S 2 + S 3 )
3
2

Vd =

1
VDC (S 2 S 3 )
2

Vq =

d = (V d RS I d )dt
t

(3)
(4)
(5)
(6)

0
t

q = V q RS I q dt

(7)

s =

d2 + q2

(8)

Tem = p (d I q q I d )
(9)
The stator resistance RS can be assumed constant during a large number of
converter switching periods TS. The voltage vector applied to the induction motor
remains also constant over the time period TS. Therefore, resolving first equation of
system leads to:
t
(10)
S = (V S RS I S )dt

S (t ) = S (0 ) + V S TS

(11)
In equation (11), S(0) stands for the initial stator flux condition. This equation
shows that when the term RSIS can be neglected in high speed operating condition of
the extremity of stator flux vector VS. Also, the instantaneous flux speed is only
governed by voltage vector amplitude [3] given in (12).
d S
(12)
VS
dt

Therefore, by selecting adequate voltage vector one can increase or decrease the
stator flux amplitude and phase to obtain the required performances. The deviation
obtained at the end of the switching period TS can be approached by the first order
Taylor Series. By choosing a suitable sequence of the vectors of tension, one can force
the end of the vector flux to follow a desired trajectory. To function with a module of
practically constant flux S, it is enough to choose an almost circular trajectory for the
end of the vector flux [3] [5].
C. Switching Table Formation
The vectors Vi+1 or Vi-1 are selected to increase the amplitude of flux, and Vi+2 or Vi-2 to
decrease it when flux is in sector I. Which shows that the choice of the vector tension
depends on the sign of the error of flux is independent of its amplitude [3]. Obviously,
the exit of the corrector of flux must be a Boolean variable. One adds a band of
hysteresis around zero to avoid unwanted commutations when the error of flux is very

238

Jagadish H. Pujar and S.F. Kodad

small [3] [4]. Indeed, with this type of corrector in spite of its simplicity, one can easily
control and maintain the end of the vector flux in a circular ring form. The switching
table proposed by Takahashi [3] is as given in Table I.

Table I: Switching table for DTFC basis.


Sector
Flux Torque
T=1
F=1
T=0
T=-1
T=1
F=0
T=0
T=-1

II

III IV

VI

V2
V7
V6
V3
V0
V5

V3
V0
V1
V4
V7
V6

V4
V7
V2
V5
V0
V1

V6
V7
V4
V1
V0
V3

V1
V0
V5
V2
V7
V4

V5
V0
V3
V6
V7
V2

Principle of Direct Torque Fuzzy Control


The DTFC of induction motor drive is designed to have three fuzzy input variables and
one output control variable to achieve fuzzy logic based DTC of the induction machine
[7] [8]. Its functional block diagram is as shown in Fig.4. It requires three input
variables, the stator flux error, electro magnetic torque error and angle of stator flux.
The output is the voltage space vector. The proposed DTFC consists of fuzzification,
rule base, data base, decision making and defuzzification blocks as shown in Fig.4.
The input variables , T and are fuzzyfied using fuzzy functions over the
respective domains. The output of DTFC is also fuzzyfied using fuzzy singletons. The
all possible fuzzy rules are stored in fuzzy rule base. DTFC takes the decision for the
given input crisp variables by firing this rule base.

F
U
Z
Z
I
T F
I
C

A
T
I
O
N

RULE BASE
()
(T)
()

DECISION
MAKING
FUZZY
V OV/ P

DATA BASE

D
E
F
U
Z
Z
I VS
F
I
C
A
T
I
O
N

Figure 4: Proposed DTFC functional block diagram.

AI based Direct Torque Fuzzy Control of AC Drives

239

A. Flux Linkage Error Fuzzification


The flux linkage error is given by = ref - S. Where ref is the related value of
stator flux and S is actual stator flux. Three linguistic values, negative, zero and
positive denoted as N, Z and P respectively are used to fuzzify flux linkage error
domain [9] [10] as shown in Fig.5.

A
N

-0.05

0.05

Figure 5: Flux linkage error domain fuzzification.

B. Electromagnetic Torque Error Fuzzification


The torque error given by T = Tref - Tem. Where Tref is desired torque and Tem is actual
torque. Five linguistic values, negative large, negative small zero, positive small and
positive large denoted as NL,NS, ZE, PS and PL respectively are used to fuzzify
torque error domain [9] [10] as shown in Fig.6.

B
1

NL

-0.025

NS

ZE

PS

0.1

-0.1

PL

0.025 T

Figure 6: Electromagnetic torque error domain fuzzification.

C. Stator Flux Linkage Angle Fuzzification


The stator flux linkage angle is an angle between stator flux S and a reference axis
is defined as = tan 1 ( q d ) . Twelve linguistic values, 0 to 12 are used to fuzzify the
domain of stator flux linkage angle [9] [10] as shown in Fig.7.

240

Jagadish H. Pujar and S.F. Kodad


C

1 0

11

12

3
12

5 7 9 11 13 15 17 19 21 23
12 12 12 12 12 12 12 12 12 12

12

Figure 7: Stator flux linkage angle domain fuzzification.

D. Space Vector Voltage Fuzzification


The space vector voltage Vi ( i=1 to 6 ) is fuzzified using six singleton linguistic
values, as V1,V2,V3,V4,V5 and V6 [9] [10] as shown in Fig.8.

D
1

V1

V2

V3

V4

V5

V6

Figure 8: Space vector voltage domain fuzzification.

E. DTFC Rule Base Design


Each fuzzy rule can be described using the three input linguistic variables , T and
and one output linguistic variable VS in the DTFC rule base. The ith IF-THEN fuzzy
rule Ri of DTFC can be expressed as:
Ri: IF is Ai, T is Bi and is Ci THEN VS is Di
Where Ai, Bi, Ci and Di are fuzzy membership function values obtained from A,
B, C and D membership functions of inputs and output linguistic variables of
DTFC respectively for ith rule [9] [10].
In this paper, Mamdanis fuzzy rule base model is developed to perform the
function of proposed DTFC. Obviously the Mamdanis implication is used to infer the
output of DTFC rule base for any crisp in put values of , T and . In the
defuzzification stage, a crisp value of the DTFC output is computed by using the center
of gravity method.

Simulation and the Results


To verify the proposed scheme, a numerical simulation has been carried out by using
MATLAB/SIMULINK. In the performed simulation, certain stator flux and torque
references are compared to the values calculated in flux and torque estimator and the

AI based Direct Torque Fuzzy Control of AC Drives

241

corresponding errors are sending to the Direct Torque Fuzzy Controller of the
induction motor drive system. The outputs of the flux and torque comparators along
with stator flux linkage angle are used in order to determine the appropriate voltage
vector and stator flux space vector. Vector locations are shown in order to validate the
control strategies as discussed above. A digital simulation studies were made by using
the S-factions and SIMULINK software of MATLAB environment for the system
described in Fig.1. The flux and torque loops of the drive were also designed and
simulated respectively using fuzzy logic control techniques. The feedback control
algorithms were iterated until best simulation results were obtained. The proposed
DTFC is simulink model is developed by using Fuzzy Toolbox.
For the simulation, a four pole, 3 KW, 1500 rpm, three phase, 220/440 volts and 50
Hz induction motor AC drive system is developed using S-functions, Fuzzy Toolbox
and SIMULINK of the MATLAB. The rated parameters of induction motor are
Rs=4.85, Rr=3.805, Ls=0.275H, Lr=0.274H, Msr=0.258H, J=0.031kgm2 and
B=0.00039. The inverter dc bus voltage is 160 volts. The proposed DTFC AC drive
system dynamic responses were obtained by simulation of the proposed AC drive
system of Fig.1 at a sampling time period of 1000 s. The pulsating ripple in the actual
torque simulated response of DTFC shown in Fig.11 is very much reduced compare to
the actual torque simulated response of conventional DTC shown in Fig.10 for the
same corresponding estimated torque simulated response shown in Fig.9.
The Direct Torque Fuzzy Control of induction motor presents the advanced
performance to achieve tracking of the desired smooth circular trajectory of stator flux
locus shown in Fig.13. The DTFC speed response shown in Fig.12 rejects the load
disturbance rapidly with no overshoot and with a negligible steady state error
compared to the conventional DTC. The reason for superior performance of DTFC of
AC drive system is basically adaptive in nature and is able to realize and fire
corresponding stable control fuzzy rules for each input state of inputs , T and .

Figure 9: Estimated torque simulated response.

242

Jagadish H. Pujar and S.F. Kodad

Figure 10: Conventional DTC torque simulated response.

Figure 11: DTFC torque simulated response.

Figure 12: DTC and DTFC actual speed simulated responses.

AI based Direct Torque Fuzzy Control of AC Drives

243

Figure 13: DTFC stator flux locus simulated response.

Conclusions
The paper presents a new approach for speed control of 3- induction motor using
fuzzy logic technique. The paper develops a fuzzy logic based DTC methodology for
AC drive systems is intended for an efficient control of the torque and flux without
changing the motor parameters. Also the flux and torque can be directly controlled
with the inverter voltage vector using SVPWM methodology.
A complete direct torque fuzzy control of SVPWM voltage source inverter-fed
induction motor drive system has been described. The system was analyzed, designed
and performances were studied extensively by simulation to validate the theoretical
concept. The simulation results shows that the proposed DTFC method is superior to
conventional DTC in robustness and in tracking precision. It appears from the response
properties that it has a high performance in presence of load disturbances.
The control of speed by DTFC gives fast dynamic response with no overshoot and
negligible steady state error. So, it is concluded that DTFC can be applied for AC
drives and is reliable in a wide speed range. Especially in AC servo motor applications,
where high dynamic performance is demanded. DTFC has a great advantage over
other control methods due to its property of fast torque response. In order to increase
the performance, control period should be selected as short as possible. When the
sampling interval is selected smaller, it is possible to keep the bandwidth smaller and
to control the stator magnetic flux more accurately.
The simulation study clearly indicates the superior performance of DTFC over
conventional DTC, because it is inherently adaptive in nature. To make this DTFC AC
drive system fully adaptive, the hybridization of fuzzy with other soft computing
techniques like ANN and Rough Sets will be preferred.

References
[1]

Mamdani, E. H. Applications of Fuzzy Algorithms for Simple Dynamic


Plants, Proc. of the IEE 121, pp. 1585 1588, 1974.

244

Jagadish H. Pujar and S.F. Kodad

[2]

F. Blaschke The principle of field orientation as applied to the new


TRANSVECTOR closed loop control system for rotating field machines,
Siemens Review XXXIX, 1972, (5): 217220.
Takahashi I, Naguchi T. A new quick-response and high-efficiency control
strategy of an induction motor. IEEE Transactions on Industry Application
[ISSN 0093-9994], 1986, IA-22(5): 820-827.
P. Tiitinen, M. Surandra, The next generation motor control method, dtc
direct torque control, Proc on Power Electronics, 1996, 1: 3743. Drives
and Energy Systems for industrial growth.
D. Casadei, G. Grandi, G. Serra, A. Tani Effectes of flux and torque
hysteresis band amplitude in direct torque control of induction machines,
IEEE-IECON-94, 1994, 299304.
Hui-Hui Xiao, Shan Li, Pei-Lin Wan, Ming-Fu Zhao, Study on Fuzzy Direct
Torque Control System, Proceedings of the Fourth International Conference
on Machine Learning and Cybernetics, Beijing, 4-5 August 2002.
R.Toufouti S.Meziane, H. Benalla, Direct Torque Control for Induction
Motor Using Fuzzy Logic, Proc. ACSE Journal, Vol. 6, Issue 2, pp. 19 26,
Jun. 2006.
S. Benaicha, R. Nait-Said, F. Zidani, M-S. Nait-Said, B. Abdelhadi, A direct
torque fuzzy control of SVM inverter-fed Induction Motor drive, Proc. of the
International Journal of Artificial Intelligence and Soft Computing Vol. 1,
Nos. 2-4, pp. 259 270, 2009.
Hui-Hui Xia0, Shan Li, Pei-Lin Wan, Ming-Fu Zhao, Study on Fuzzy Direct
Torque Control System, Proceedings of the Fourth International Conference
on Machine Learning and Cybernetics, Beijing, pp. 4-5 August 2002.
Yang Xia and Oghanna, W Study on fuzzy control of induction machine with
direct torque control approach, Industrial Electronics, 1997. ISIE apos; 97,
Proceedings of the IEEE International Symposium on Volume 2, Issue, pp:
625 630, 7-11 Jul 1997.

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]