587

FIELD ORIENTED CONTROL OF PERMANENT MAGNET

SYNCHRONOUS MOTOR (PMSM) USING FUZZY LOGIC
CONTROLLER
1
B.Adhavan Member IEEE, 2A. Kuppuswamy, 3G.Jayabaskaran and 4Dr.V.Jagannathan Senior Member IEEE.

only under transient conditions. A simple gain scheduled PI

Abstract— speed controller has been proposed in [1], where the
The Permanent Magnet Synchronous Motor controller gains are varied with the input error signal. This
(PMSM) has been widely used in the low to medium power controller suffers from the drawback that for its proper
system due to its characteristics of high efficiency, high performance, the limits of the controller gains and the rate at
torque to inertia ratio, high reliability and fast dynamic which they would change have to be appropriately chosen.
performance. With the advent of the vector control methods Fuzzy based gain scheduling of PI controller has been
,permanent magnet synchronous motor can be operated like proposed in, but the limits of the gains have to be determined
separately excited dc motor high performance application. by the user manually.
The complexity of PI controller tuning and high response This paper is organized as follows Section II provides the
time is overcome by Fuzzy controller. Which has less principle of vector control of PMSM. Section III gives the
response time and high accuracy without any mathematical proposed system. Section IV shows the design procedure of
calculation .This paper presents a simulation of speed control fuzzy logic controller. Section V gives the results and discussion
system on fuzzy logic approach for an indirect vector of MATLAB SIMULINK simulation waveforms. Section VI
controlled permanent magnet synchronous drive by applying gives the conclusion and merits of the proposed system.
space vector modulation. The design, analysis and simulation
of the proposed system is done using MATLAB version 8. II. PRINCIPLE OF VECTOR CONTROL:
[6]The invention of vector control in the beginning of
Index terms —. Fuzzy controller, PI controller, Permanent 1970’s and demonstrated in induction motor can be controlled
Magnet Synchronous Motor (PMSM) by separately excited DC motor, which increase the efficiency
of AC drives. Vector control is applicable to both
I. INTRODUCTION Asynchronous and synchronous motor drive. The fig.1shows
the block diagram of conventional vector control of PMSM.
[1-6] Permanent Magnet Synchronous Motors
(PMSM) have advantages like high efficiency, high power
factor, high power density and maintenance free operation,
and these motors are nowadays preferred in a variety of
applications. Vector controlled PMSM drive provides better
dynamic response and lesser torque ripples, and necessitates
only a constant switching frequency. The outer speed loop in
vector control greatly affects the system performance.
Proportional plus integral (PI) controllers are usually
preferred, but due to its fixed proportional gain and integral
time constant, the performance of the PI controllers are
affected by parameter variations, load disturbances and speed
variations. These problems can be overcome by the fuzzy
logic controllers, which do not require any mathematical
model and are based on the linguistic rules obtained from the
experience of the system operator. But the performance of the Fig.1 Conventional Vector control of PMSM
fuzzy controller as compared to the PI controller is superior The synchronous inductance Ls and the corresponding
armature flux Ψa (=Ls Is ) very small, that is, Ψs= Ψm= Ψf . For
maximum torque sensitivity with the stator current(i.e.,
B.Adhavan is working as Assistant professor in EEE (PG) department of Sri maximum efficiency), we can set ids=0 and Is = iqs , and the
Ramakrishna Engineering College, coimbatore-641022 India.
(Corresponding author to provide phone: +919994869720; e-mail: adhav14 stator resistance Rs is neglected for simplicity. This condition
@ yahoo.com). also gives a minimal inverter power rating. The developed
A.Kuppuswamy and G.Jayabaskaran is pursuing M.E. (Power Electronics torque expression can be easily derived and is shown in (1).
& Drives) in Sri Ramakrishna Engineering college, coimbatore-641022, India
3⎛ p⎞
Dr.V.Jagannathan is working as HOD, Department of Electrical and
Electronics Engineeringg, in Coimbatore Institute of Technology,
Te = ⎜ ⎟ψmf iqs (1)
Coimbatore-641014.
2⎝ 2 ⎠

978-1-4244-9477-4/11/$26.00 ©2011 IEEE

 588

[6]Where Ψf is the space vector magnitude (√2 Ψf ) and Ψs A.PRINCIPLE OF SPACE VECTOR PWM
cos φ = Ψs cos δ = Ψf . This indicates that the torque is
propotional to iqs and the power factor angle φ equals the Space Vector PWM (SVPWM) is a more sophisticated
torque angle δ. The stator command current iqs is derived from technique for generating a fundamental sine wave that
the speed control loop. Its polarity is positive for motoring provides a higher voltage to the motor and lower total
mode, but negative for regeneration mode. The rotating frame harmonic distortion when compared to sinusoidal PWM. It is
signals are converted to stator current commands with the also well-suited for use in vector control (Field orientation) of
help of unit vector signals (cos θe , sin θe ) as shown. The AC motors. The space vectors technique is nowadays
position control loop can be added easily, if desired. commonly known as space vector modulation (SVM).
The switching sector is shown in the figure 3 and the table 1
III. PROPOSED SYSTEM shows the switching vector for inverter.
[1] In traditional PI controller it suffers from overshoot and
undershoots of response, when some unknown nonlinearities
are present in system. The Fuzzy controller overcomes these
disadvantages.[2-6].The fuzzy logic controller(FLC) executes
the rule based taking the inputs and gives the output by
defuzzification , inputs are speed error (e) and change in
speed error, (ce) the output is torque limit (T*). Which is
equivalent to iqs*.The Fig.2. Shows the block diagram of the
proposed system.

Fig 3 Switching vectors and sectors

TABLE 1: SWITCHING VECTORS FOR INVERTER

Vector A+ B+ C+ A− B− C− VAB VBC VCA
V0 = OFF OFF OFF ON ON ON 0 0 0
{000}
V1 = ON OFF OFF OFF ON ON +Vdc 0 −Vdc
{100}
V2 = ON ON OFF OFF OFF ON 0 +Vdc −Vdc
{110}
V3 = OFF ON OFF ON OFF ON −Vdc +Vdc 0
{010}
V4 = OFF ON ON ON OFF OFF −Vdc 0 +Vdc
{011}
Fig.2 Block diagram of proposed system V5 = OFF OFF ON ON ON OFF 0 −Vdc +Vdc
{001}
The d,q projections of the stator currents are then compared V6 = ON OFF ON OFF ON OFF +Vdc −Vdc 0
to their reference values isq* and isd*=0 ( to get maximum {101}
torque, set to zero if there is no field weakening ) and V7 = ON ON ON OFF OFF OFF 0 0 0
{111}
corrected by means of PI current controllers. The outputs of
the current controller are passed through the inverse Park
To implement the space vector PWM, the voltage
transform and a new stator voltage vector is impressed to the
equations in the abc reference frame can be transformed into
motor using the space vector pulse width modulation
the stationary αβ reference frame that consists of the
(SVPWM) technique.
horizontal (α) and vertical (β) axes, as a result, six non-zero
In order to control the mechanical speed of the motor, In
vectors and two zero vectors are possible. Six nonzero
order to control the mechanical speed of the motor, an outer
vectors (V1 - V6) shape the axes of a hexagonal as depicted
loop is driving the reference current isq*[2]The PI regulator
in Fig 3 and feed electric power to the load or DC link
compares the speed set point with the measured mechanical
voltage is supplied to the load. The angle between any
speed of the rotor and produces the stator current quadrature
adjacent two non-zero vectors is 60 degrees. Meanwhile, two
axis reference, isq*. That is the speed controller outputs a
zero vectors (V0 and V7) are at the origin and apply zero
reference torque, which is proportional to the quadrature-axis
voltage to the load. The eight vectors are called the basic
stator current component isq*. The mechanical speed
space vectors and are denoted by V0, V1, V2, V3, V4, V5,
reference is denoted "ωref". The information about the phase
V6, and V7. The same transformation can be applied to the
current can be obtained by two current sensors and the
desired output voltage to get the desired reference voltage
information about the rotor position can be obtained by
vector Vref in the d-q plane. The objective of space vector
position sensor like incremental encoder.
PWM technique is to approximate the reference voltage
vector Vref using the eight switching patterns.
 589

B. PI CONTROLLER TUNING [9] INPUTS

The speed loop PI regulator compares the current iqref with
the actual current iq and produces the Vsqref. And the other
PI regulator compares the current isdref with the measured
current id and produces the Vsdref.
The tuning of PI Controller is done by Ziegler-Nichols
method.The table 2 shows the Ziegler–Nichols tuning chart
for PI controller.
TABLE 2: ZIEGLER–NICHOLS TUNING CHART
Ziegler–Nichols method
Control Fig.5 Speed error
Kp Ki Kd
Type
Ku / 1.2Kp /
PI -
2.2 Tu

IV. FUZZY LOGIC CONTROLLER

[2-3]The fuzzy logic can be considered as a
mathematical theory combining multi-valued logic,
probability theory, and artificial intelligence to simulate the
human approach in the solution of various problems by using
an approximate reasoning to relate different data sets and to
make decisions.
It has been reported that fuzzy controllers are more
robust to plant parameter changes than classical PI or PID Fig.6 change in speed error
controllers and have better noise rejection capabilities. The
proposed scheme exploits the simplicity of the Mamdani type OUTPUTS
fuzzy systems that are used in the design of the controller.
The fig.4 shows the block diagram of fuzzy logic controller.

Fig.4 Fuzzy logic controller

Fig.7 Torque Limit
A. Membership Functions
[4]The Fuzzy Logic Controller initially converts the crisp TABLE 3: RULES
error and change in error variables into fuzzy variables and
then are mapped into linguistic labels. Membership functions NE ZE PE
are associated with each label as shown in which consists of
two inputs and one output. NCE NT NT ZT
Inputs- Triangular membership function
ZCE NT ZT PT
1. Speed error : NE, ZE, and PE.
2. Change in speed error : NCE, ZCE, and PCE. PCE ZT PT PT
Output- Triangular membership function
1. Torque Limit : NT*, ZT*, and PT*.
[2] The linguistic labels are divided into three groups. They B. Knowledge Rule Base
are: N-Negative, Z-zero, P-positive. Each of the inputs and
the output contain membership functions with all these three [7][8] The mapping of the fuzzy inputs into the required
linguistics. This method of formulation of control algorithms output is derived with the help of a rule base as given in Table
allows implementing heuristic strategies. A straightforward 1.Each rule of the FLC is characterized with an IF part, called
source of deriving the linguistic control strategies are human the antecedent, and with a THEN part called the consequent.
experience and understanding, which essentially contain the The antecedent of a rule contains a set of conditions and the
model of the control system in an implicit form. The fig.5 consequent contains a conclusion. If the conditions of the
shows the speed error, fig.6 shows the change in speed error antecedents are satisfied, then the conclusions of the
and fig.7 shows the torque limit. consequent apply. Considering the first rule, it can be
 590

interpreted as: IF change in speed error is N and change is

speed is N, THEN the output will be N.
C. Defuzzification <Electromagnetic torque Te (N*m)>

100
[2]Generally the output obtained is fuzzy in nature
and has to be converted into a crisp value by using any
Defuzzification technique. 80

Torque(Nm)
V. RESULTS AND DISCUSSION 60
The comparative results of traditional PI controller and
Fuzzy logic controller was performed using MATLAB 40
version 8, results are shown Fig.8, 9, 10, 11,12,13,14
,15,16,17 and 18.
20

<Rotor speed wm (rad/s)>

70 0
0 1 2 3 4 5
Time(S)
60
Fig.11 Torque using PI
50
Rotor speed(rad/s)

<Electromagnetic torque Te (N*m)>

40 8

30 7

6
20
5
10 Torque(Nm)
4
0
0 1 2 3 4 5 3
Time(S)
Fig.8 Rotor speed using PI 2
<Rotor speed wm (rad/s)>

70 1

0
69.98 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time(s)
69.96
Fig.12 Dynamic Torque using PI
Rotot speed(rad/s)

69.94
<Rotor speed wm (rad/s)>

69.92 80

70
69.9
60
Rotor speed(rad/s)

69.88
50
69.86
0.3 0.35 0.4 0.45 0.5 40
Time(S)
30
Fig .9 Zoomed Rotor speed using PI-no load
<Rotor speed wm (rad/s)> 20
70
10
69.9
0
0 1 2 3 4 5
Time(s)
Rotot speed(rad/s)

69.8
Fig.13 Rotor speed using FLC
69.7

69.6

69.5

69.4
0.5 0.6 0.7 0.8 0.9 1
Time(S)
Fig .10 Zoomed Rotor speed using PI- load
 591

<Rotor speed wm (rad/s)> <Electromagnetic torque Te (N*m)>

70 8

69.98 7

6
69.96

Torque(Nm)
5
69.94
4
Rotor speed(rad/s)

69.92
3
69.9
2

69.88 1

69.86 0
0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0.3 0.35 0.4 0.45 0.5 Time(s)
Time(s)
Fig. 17 Dynamic Torque using FLC
Fig.14 Zoomed Rotor speed using FLC-no load

<Rotor speed wm (rad/s)>

70 40

69.98 30

69.96 20

Stator current Is (A)

69.94 10
Rotor speed(rad/s)

69.92 0

69.9 -10

69.88 -20

-30
69.86
0.5 0.6 0.7 0.8 0.9 1 -40
0 0.2 0.4 0.6 0.8 1
Time(s) Time(s)
Fig.15 Zoomed Rotor speed using FLC- load Fig.18 Stator Current using FLC

<Electromagnetic torque Te (N*m)> From the rotor speed waveforms of PMSM controlled by
50
traditional PI fig.12 and FLC fig.17 it is inferred that the
dynamic response time is reduced. From the zoomed speed
40 waveforms of fig.9 and fig.14 it is inferred that the steady
state error at no load has been reduced from 0.06 rad/s to
30 0.005rad/s and From the zoomed speed waveforms of fig.10
and fig.15 the steady state error at load it has been reduced
Torque(Nm)

20 from .6rad/s to .14rad/s The initial starting torque of the

machine is reduced so the initial stator current drawn by the
10
motor is also reduced from the fig.11 and fig.16.
VI.CONCLUSION
A fuzzy rule-base design of fuzzy logic speed control has
0 been studied for speed control of vector controlled PMSM
0 1 2 3 4 5
Time(s) Drive. A set of fuzzy decision rules are formulated based on
Fig. 16Torque using FLC the literature review of the controller’s design. In this paper,
comparative results for traditional PI controller and Fuzzy
logic controller for speed response during start-up under no
load, load disturbance and changes in command settings. The
Simulation study is realized in MATLAB environment. The
investigation of fuzzy controllers is carried out based on
several selected speed response. The work can be upgraded to
sensorless speed controller by using speed and postion
estimators.
 592

VII. APPENDIX
MOTOR PARAMETERS 1.B.Adhavan, Received
Type PMSM B.E.,(Electrical and Electronics
Rated speed 75(rad/sec) Engineering) degree from
Number of phases 3 Kumaraguru College of
Number of poles (P) 8 Technology, Coimbatore, Affiliated
Base current 8A to Bharathiyar University in the
Rated voltage 300 V year 2001, and received M.E.,
Stator resistance per phase( R) 0.9585 ohm (Power Electronics and Drives)
q-axis inductance(Lq) 0.00525 H degree from P.S.G College of
d-axis inductance (Ld ) 0.00525 H technology, Affiliated to Bharathiyar University in the year
Stator flux linkages per phase 2003, Currently working as Assistant Professor in EEE (PG)
due to rotor magnet ( A f ) 0. 1827V/ (rad/s) of Sri Ramakrishna Engineering College, Coimbatore-641
Moment of inertia (J) 0.0006329Kg/m^2 022. His area of interested includes Power Electronics and
Friction Factor (F) 0.0003035(N.m.s) Drives, Embedded Controls, Special Electrical Machines and
Controls.
REFERENCES 2. A.Kuppuswamy, Currently pursuing M.E. degree in Sri
[1] Amit Vilas Sant and K. R. Rajagopal PM Synchronous Ramakrishna engineering college, coimbatore, Affiliated
Motor Speed Control Using Hybrid Fuzzy-PI With Novel Anna University of technology, coimbatore, India. Currently
Switching functions IEEE transactions on magnetic, vol. 45, he is working on project regarding PMSM speed control
no. 10, October 2009 pp 4672-4675 using fuzzy logic control.
[2] Siti Noormiza Mat Isa, Zulkifilie Ibrahim, Fazlli Patkar 3. G.Jayabaskaran Currently pursuing M.E. degree in Sri
Comparative Study of Fuzzy Logic Speed Controller in Ramakrishna engineering college, coimbatore, Affiliated
vector Controlled PMSM Drive: Minimum Number of Fuzzy Anna University of technology, coimbatore, India. Currently
Rule-Base 2009 Conference on Innovative Technologies in he is working on project regarding PMSM.
Intelligent Systems and Industrial Applications (CITISIA
2009) Monash University, Sunway campus, Malaysia, 25th & 4.Dr.V.Jagannathan, Received B.E.,
26th July 2009. (Electrical and Electronics
[3] Bhim Singh, Senior Member, IEEE, B.P. Singh, Senior Engineering.) degree from Madras
Member, IEEE and Sanjeet Dwivedi, Dsp based University in the year of 1965. M.Sc
implementation of Hybrid Speed Controller for Vector (Power system Engineering) degree
Controlled Permanent Magnet Synchronous Motor Drive. from Madras University in the year
Emerging electric power system vol.8,no 2,pp1-22 2007 . of 1971. and Ph.D (Power Electronics and Drives) in IIT,
[4] M. Nasir Uddin, Senior Member, IEEE, M. A. Abido, Kharagpur in 1990.He has 46 years of teaching and Research
Member, IEEE, and M. A. Rahman, Fellow, IEEE Real-Time field. He has published Three book in Electrical Engineering.
Performance Evaluation of a Genetic-Algorithm-Based Fuzzy Currently he is working as HOD, Department of Electrical
Logic IEEE transactions on industry applications, VOL. 41, and Electronics Engg. In Coimbatore Institute of Technology,
NO. 1, January/February 2005Controller for IPM Motor Coimbatore-14. His area of interested includes Power
Drives Electronics and Drives, Embedded Controls, Special
[5] M. A. Rahman, M. Vilathgamuwa, M. N. Uddin, and K. Electrical Machines and Controls.
J. Tseng, ‘‘Nonlinear control of interior permanent magnet
synchronous motor,’’ IEEE Trans. Ind. Appl., vol. 30, no. 2,
pp. 408---416, Mar./Apr. 2003.
[6]Bimal K. Bose, "Modern Power Electronics and AC
drives," Pearson Education Asia, Low Price Edition (LPE)
2003.
[7] K. Erenay, I. Ciprut, L. Tezduyar, and Y. Istefanopulos,
‘‘Application of fuzzy algorithms to the speed control of
washing machines with brushless DC motors,’’ in Proc. Int.
Conf. Electric Machines, Istanbul, Turkey, 1998, pp. 1231---
1236.
[8] M. N. Uddin, M. A. Abido, and M. A. Rahman,
“Development and implementation of a hybrid intelligent
controller for interior permanent magnet synchronous motor
drive,” IEEE Trans. Ind. Appl. vol. 40, no. 1, pp. 68–76,
Jan./Feb. 2004.
[9] J. C.Basilio, S. R. Matos, “Design of PI and PID
controllers with transient performance specification,” IEEE
Transactions on Education, vol. 45, no. 4, pp. 364–370,
November 2002.