See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/252039587
Direct torque control of brushless DC motor drives with reduced starting
current using fuzzy logic controller
Article · August 2011
DOI: 10.1109/URKE.2011.6007863
CITATIONS
READS
9
2,077
3 authors, including:
Mehdi Shafiei
Mojtaba Bahrami Kouhshahi
Queensland University of Technology
Nexteer Automotive
28 PUBLICATIONS 182 CITATIONS
15 PUBLICATIONS 101 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Magnetically Geared Lead Screw View project
Australian Renewable Energy Agency (ARENA) and industry funded project: Increasing Visibility of Distribution Networks to Maximise PV Penetration Levels View project
All content following this page was uploaded by Mehdi Shafiei on 21 July 2015.
The user has requested enhancement of the downloaded file.
2011 International Conference on Uncertainty Reasoning and Knowledge Engineering
Direct Torque Control of Brushless DC Motor Drives with Reduced Starting
Current Using Fuzzy Logic Controller
N. Parhizkar, M. Shafiei, and M. Bahrami Kouhshahi
Department of Electrical Engineering, Science and Research Branch, Islamic Azad University, Fars, Iran
E-mail: parhizkar@ssriau.ac.ir, mehdi.shafiei.7206@gmail.com, mojtaba.bahrami@gmail.com
Simulation results for PI speed controller and FLC are
presented. Particle Swarm Optimization (PSO) has been used
to regulate PI parameters of speed controller to achieve
optimum torque and speed responses [3], [4]. Effectiveness
of proposed method can be deduced from comparison of
simulation results.
Abstract— This paper presents a direct torque control
technique for brushless DC motors with non-sinusoidal back
electromotive force. Direct torque control has some benefits
such as faster torque response and reduced torque ripple for
driving the brushless DC motors. In order to solve the
problems associated with conventional PI speed controller, a
speed control based on fuzzy logic controller is proposed to
reduce starting current, eliminate overshoot in the torque and
speed responses, simplify designing and eliminating complex
math formulas. To reduce rising time PD controller used with
fuzzy logic control. The effectiveness of proposed system has
been validated by simulation results.
Keywords-Brushless dc (BLDC) motor; DTC; fuzzy logic
controller; starting current
I.
BLDC MOTOR OPERATION PRINCIPLE
II.
INTRODUCTION
The Brushless DC (BLDC) motors are a type of electric
motors, which are AC synchronous and look quite similar to
the DC motors. BLDC motors have come to dominate many
applications such as home appliances, automobiles,
information technology equipment, industries, transportation,
aerospace, defense equipment, power tool, vision, sound
equipment, and laboratory medical equipment in wide range
of power. 0
Direct Torque Control (DTC) method proposed in [1], [2]
has been utilized to drive the BLDC motors. DTC has
features like fast torque response, simple and robust design.
These features have made it popular in industrial applications.
This control method operates in two phase conduction mode
which is simplified to just a torque controlled drive by
intentionally keeping the stator flux linkage amplitude
almost constant by eliminating the flux control in the
constant torque region. Due to sharp changes in commutation
region, amplitude of stator flux linkage cannot easily be
controlled.
In this paper PI speed controller is replaced by Fuzzy
Logic Controller (FLC). Although PI speed controller has a
simple structure and fast response but this controller cannot
completely eliminate overshoot in the torque and speed
responses. In addition, low starting current will be achieved
if FLC is used, as will be described in section IV. As regards
to amount of current that flows through the power switches
is the most important switching cost parameters, so reduction
of starting current leads to low cost inverter switches.
Furthermore, high starting current during start up leads to
magnetic saturation and it reduces torque constant due to the
nonlinearity of the magnetic circuit. In practice, by designing
a suitable control system during startup, efficiency of BLDC
motor is improved.
978-1-4244-9983-0/11/$26.00 ©2011 IEEE
129
BLDC motors are truly an inside-out DC Commutator
with the mechanical commutator replaced by an electronic
switching converter. The stator of BLDC motors is the coil,
and the rotor is the permanent magnet. The stator generates
the magnetic field to make the rotor rotating. Hall Effect
sensors detect the rotor position as the commutating signals.
Configuration of BLDC motor drive system, back EMF
pattern and reference current generation are shown in Fig. 1
and Fig. 2 [5]-[7].
S1
S3
S5
S4
S6
S2
Figure 1. Configuration of BLDC motor drive system
Figure 2. Back EMF pattern and reference current generation
The analysis of a BLDC motor is represented as the
following equations:
ªv a º
«v »
« b»
¬«v c ¼»
ªR
«0
«
¬« 0
0
R
0
ªL M
«« 0
¬« 0
0 º ªi a º
0 »» ««i b »»
R ¼» ¬«i c ¼»
0
L M
0
º
d
0 »»
dt
L M ¼»
0
ªi a º ªe a º
«i » «e »
« b» « b»
¬« i c ¼» ¬«ec ¼»
given for eliminating the flux control. First, since the
magnitude of the phase back-EMF is below 50% of the dclink voltage in the constant torque region there is no need to
control the flux amplitude. Second, with the two-phase
conduction mode, sudden sharp dips in the stator flux
linkage locus occur that complicate the control scheme. The
size of these sharp dips is unpredictable. Third, regardless of
the stator flux linkage amplitude, the phase currents tend to
match with the flat top portion of the corresponding
trapezoidal back-EMF to generate constant torque.
As it can be seen from Table I, the switching pattern of
the inverter can be determined according to the torque status
from the output of hysteresis. Therefore, in the DTC of a
in the voltage vector
BLDC motor drive the flux error Fst
selection look-up table is always selected as zero and only
the torque error Tst is used depending on the error level of the
actual torque from the reference torque. If the reference
torque is bigger than the actual torque, within the hysteresis
bandwidth, the torque error Tst is defined as "TI", otherwise it
is "TD", as shown in Table I. By means of this method
BLDC motor will be successfully driven.
(1)
where va, vb and vc are the phase voltages, ia, ib and ic are the
phase currents, ea, eb and ec are the phase back-EMF
waveforms, R is the phase resistance, L is the self inductance
of each phase and M is the mutual inductance between any
two phases.
So the electromagnetic torque can be obtained as:
T e (e a i a eb i b ec i c ) / Zr
(2)
where r is the mechanical speed of the rotor.
d
Zr (T e T L B Zr ) / J
(3)
dt
where B is the damping constant, J is the moment of inertia
of the drive and TL is mechanical torque.
The electrical speed e related to the mechanical speed
for a motor with P number of poles is:
Ze (P / 2)Zr
(4)
III.
IV.
DESIGN OF SPEED CONTROLLER BASED ON FUZZY
LOGIC CONTROL
The main idea of FLC is to use the control ability of
human beings which includes experience and intuition of
experts. It is one of useful control technique for uncertain
and ill-defined nonlinear systems. Control actions of a FLC
are described by some linguistic rules. This property makes
the control algorithm easy to understand. Heuristic FLC
incorporate experience or knowledge into rules. Fig. 4 shows
a typical FLC that consist of the input and the output scaling
factors, the fuzzification and defuzzification blocks, the rule
base, and the fuzzy interface engine [10], [11]. FLC has been
proved to be powerful and able to resolve many problems.
FLC is used to improve the dynamic response and reduce the
overshoot before the motor reaches the desired speed. The
inputs of FLC are speed error and the differential of speed
error. The output of controller is command torque. In order
to facilitate computation, triangular membership functions
are utilized in this study for both the inputs and output
membership functions. The Mamdani type inference is
adopted here because it is one of the most widely used
inference methods.
Table II shows the rule base chosen for the FLC, [12].
These five membership functions are labeled as: negative big
(NB), negative small (NS), zero (ZO), positive small (PS)
and positive big (PB). The control rules are built based on
the difference between actual speed and command speed.
Though FLC has many advantages that were mentioned
above, but it increases rising time, so the output of FLC is
added with output of PD controller. Therefore two
controllers applied for two reasons; first, PD controller’s
input is the difference between the actual speed and the
command speed, when the motor is at standstill state this
input has considerable value.
DIRECT TORQUE CONTROL OF BLDC MOTOR DRIVE
USING TWO – PHASE CONDUCTION MODE
DTC was originally developed for induction machine
drives, and directly controls the flux linkage and
electromagnetic torque, considering the electrical machine,
the power electronic inverter, and the control strategy at the
system level.
The key issue in the DTC of a BLDC motor drive in the
constant torque region is to estimate the electromagnetic
torque correctly. For surface-mounted BLDC motors the
back-EMF waveform is non-sinusoidal (trapezoidal),
therefore equation (5) which is given in the stationary
reference frame can be used for the electromagnetic torque
calculation [1], [2], [8], [9].
3P 1
ªeD i s D e E i s E º¼
Te
(5)
2 2 Ze ¬
where e
is the electrical rotor position and
e, e, is, is
are motor back-EMFs and stator currents, respectively.
Fig. 3 shows the basic functional blocks used to
implement the core of the DTC scheme. Four key blocks
interact to provide the primary control required. These are
the speed controller, the motor torque estimator, the torque
comparator, and the switching logic which selects the
inverter switching state. The torque comparator contain
hysteresis control block to compare the torque reference with
its corresponding actual value. The actual speed is compared
to the speed command to form the torque command from the
PI speed regulator.
Although in conventional DTC torque and flux are
considered in the overall control system but three reasons are
130
¦
¦
¦
3
2
5
6
T1
T3
T4
T6
T5
4
d
T2
dt
m
is = isa
is =
1
(isa + 2isb)
3
P
2
Te
3p 1
ªeˆ i eˆ i º
ˆe ¬ q sq d sd ¼
2 2Z
d m
dt
e
Figure 3. Overall block diagram of the DTC of a BLDC motor dive in the constant torque region and Speed Controller Based on Fuzzy Logic
TABLE I.
SWITCHING STATE SELECTOR FOR DTC, TORQUE INCREASE (TI), TORQUE DECREASE (TD) AND NO-CHANGE IN FLUX (F)
These two speed controllers are as follows; PI speed
controller and FLC.
TABLE III.
BLDC MOTOR PARAMETER USED FOR SIMULATION
Figure 4. Typical FLC scheme
TABLE II.
'Z
RULE BASE WITH FIVE MEMBERSHIP FUNCTION
'Z c
According to (6) output of PD controller has considerable
value too, so the speed rises rapidly from standstill state and
reaches the steady state. Second, until speed is low PD
controller is effective, with increased speed, more precise
controller should be applied to eliminate overshoot and
improve response.
d
y (t ) K p e (t ) K d
e (t )
(6)
dt
where Kp and Kd are the proportional and derivative gains.
e(t) and y(t) are the input and output of PD controller
respectively.
V.
SIMULATION RESULTS
In order to evaluate effectiveness of proposed method,
drive system shown in Fig. 3 has been simulated by
MATLAB/Simulink. Motor parameters used for the
simulation are given in Table III. Two different speed
controllers are applied on the motor and results are compared.
131
Because of the system nonlinear behavior, the PI
parameters can be regulated by intelligent algorithms.
Nowadays, PSO algorithm is one of the fast and accurate
methods in comparison with other intelligent algorithm, so in
this paper it is utilized to achieve optimal PI controller
parameters. Although PI controller decreases rise time but it
can’t eliminate the overshoot.
Fig. 5.a shows rotor speed by using PI controller.
Overshoot of the response is obviously distinguished with
the circle and zoomed region. As shown in Fig. 5.b by using
FLC the overshoot is completely removed.
Key advantage of using FLC over PI controller is
reducing starting current without causing any starting
problems.
Fig. 6 shows that this analysis is true for torque response.
It is clear from Fig. 7 that starting current by using FLC
has a value about half of the applying PI controller.
Therefore low price switches can be used, so overall cost of
inverter is reduced. Due to low starting current by using FLC,
although starting torque reduced compare with using PI
controller, as shown in Fig. 6, but it can’t lead to starting
problems.
controller has been verified through simulation and the
results were presented.
1500
20
1200
10
5
600
Iabc (A )
S peed (rpm )
15
900
300
0
0
-5
-10
0
0.2
0.4
0.6
0.8
Time (sec)
1
1.2
1.4
-15
(a)
-20
0
0.01
0.02
0.03
0.04
0.05
Time (sec)
0.06
0.07
0.08
0.09
0.06
0.07
0.08
0.09
(a)
1500
20
1200
10
5
Iabc (A )
S peed(rpm )
15
900
600
0
-5
300
-10
0
0
0.2
0.4
0.6
0.8
Time(sec)
1
1.2
1.4
-15
-20
(b)
Figure 5. Rotor speed a) By using PI controller b) By using FLC
E lec trom agnetic Torque (N.m )
11
10
0.02
0.03
0.04
0.05
Time (sec)
9
REFERENCES
S. B. Ozturk, W. C. Alexander, and H .A. Toliyat, "Direct torque
control of four-switch brushless dc motor with non sinusoidal backEMF," IEEE Trans Power Electronics, vol. 25, pp. 263-271, February
2010.
[2] S. B. Ozturk, and H .A. Toliyat, "Sensorless direct torque and indirect
flux control of brushless dc motor with non-sinusoidal back-EMF’’,
IEEE International Industrial Electronics, vol. 3, pp. 1373-1378
November. 2008.
[3] M. Clerc, Particle Swarm Optimization, ISTE, London, UK, 2006.
[4] Y.Shi and R.Eberhart, K.Muhammed Zakariah, "A modified particle
swarm optimizer," in Proc. 1998 IEEE Int. Conf. on Evolutionary
Computation, USA, pp.69-73.
[5] Grasblum P. "3-phase BLDC motor contents control with Hall
sensors using DSP56F80x," Motorola App Note AN1916/D, 2004.
[6] Toliyat HA, Gopalarathnam T., AC machines controlled as DC
machines (Brushless DC machines/electronics), In: Skvarenina TL,
editor. The power electronics handbook. New York: CRC Press; 2002.
[7] Ji Hua, Li Zhiyong, "Simulation of sensorless permanent magnetic
brushless dc motor control system," in Proc. 2008 IEEE Int. Conf. on
Automation and Logistics, Qingdao, China, pp. 2847-2851.
[8] Yong Liu, Zi Qiang Zhu, D. Howe, "Instantaneous torque estimation
in sensorless direct-torque-controlled brushless dc motors", IEEE
Trans. Industry Applications, vol. 42, no. 5, September/October 2006.
[9] Yong Liu, Zi Qiang Zhu, D. Howe, "Commutation-torque-ripple
minimization in dirrect-torque-controlles pm brushless dc drives,"
IEEE Trans Industry Application, vol. 43, no. 4, July/August 2007.
[10] M.ali Akcayol, Aydin Cetin, and Cetin Elmas, "An educational tool
for fuzzy logic-controlled BDCM," IEEE Trans Education, vol. 45,
no. 1, February 2002.
[11] R.Shanmugasundram, K.Muhammed Zakariah, and N.Yadaiah,
‘’Digital implementation of fuzzy logic controller for wide range
speed control of brushless dc motor," in Proc. 2009 IEEE Int. Conf. ,
pp 119 -124.
[12] Dan Liu, Changliang Xia, Maohua Zhang, Yingfa Wang "Control of
brushless dc motor using fuzzy set based immune feedback PID
controller’’, 2007 IEEE International Symposium on Industrial
Electronics, pp 1045 - 1049
8
7
[1]
6
5
4
3
2
1
0
0
0.2
0.4
0.6
0.8
Time (sec)
1
1.2
1.4
(a)
12
11
E lec trom agnetic T orque (N .m )
0.01
(b)
Figure 7. a) Starting current by using PI controller b) Starting Current by
using FLC
12
10
9
8
7
6
5
4
3
2
1
0
0
0.2
0.4
0.6
0.8
Time (sec)
(b)
Figure 6. a) Electromagnetic torque
Electromagnetic torque by using FLC
1
by
1.2
using
1.4
PI
controller
b)
VI. CONCLUSION
In this work Direct Torque Control based on Fuzzy Logic
Controller has demonstrated. DTC offers some advantages
such as simple algorithm, simplicity to implement, faster
torque response, reduced torque ripple and less sensitivity to
parameters variations, so the proposed system has utilized
DTC methods, in order to benefit from mentioned
advantages. In order to eliminate overshoot exists in speed
and torque responses, PI controller replaced by FLC. In
addition by using FLC, starting current reduced due to
reliability of this controller. The effectiveness of FLC
132
View publication stats
0