POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

Control Strategy for Switched Reluctance Motor

With Rotary Encoder Based Rotor Position
Detection

Slamet RIYADI

Department of Electrical Engineering, Faculty of Engineering, Soegijapranata Catholic University,

Pawiyatan Luhur IV-1, Semarang, Indonesia

[email protected]

DOI: 10.15598/aeee.v16i3.@2545

Abstract. Development of electric drive tends to the is more interesting due to its simplicity, robustness,
use of Switched Reluctance Motor (SRM) for their ad- and low cost although it has complexities. A rotor po-
vantages and green technology issues. The SRM takes sition information is required by SRM drive to provide
significant place in development for its simplicity, ro- sequence pulses that are properly synchronized with
bust construction, and low cost. Sensorless method can the rotor position to achieve the required speed. In
be applied to drive SRM, it is less expansive but has a sensor-based SRM control, rotor position detectors
more complexity and limitation. On the other hand, are commonly implemented by using Hall Effect or any
although sensor-based rotor position detection needs a other sensors. For the outputs of such sensors are time
hardware assembled on the shaft, some advantages can interval sectors, so it is difficult to adjust the instant
be obtained. In this paper, a control strategy for SRM time to turn on and off the switches.
drive with rotary encoder based rotor position detection
is proposed, core of the strategy implements digital sig- In sensorless concept, no hardware is required to de-
nal controller. The problem associated with wide range tect the rotor position. Some problems arise in apply-
speed and standstill operation can be overcome by this ing this pattern so there are some limitations in such
strategy. This is also capable to vary the time to turn sensorless control at standstill, low and high speeds [3].
the switches on and off by software. The analysis was A sensorless method can use measurement of the mutu-
verified by simulations and experiments. ally induced voltage in an inactive phase [4], derivative
of the phase current [5], flux-current approach as the
integration of voltage and current [6], etc.
Keywords In sensor concepts, by using Giant Magneto Resistive
(GMR) sensor embedded in the air gap of the motor,
Demagnetizing, freewheeling, magnetizing, ro- the rotor position can be estimated [7]. Two SRM can
tary encoder position detection, stepping mode, be connected on the same shaft, one is operated as a
switched reluctance motor. motor and the other as a rotor position detector that
is called Switched Reluctance Position (SRP) sensor
[8]. SRM can also be operated to mimic an inductive
1. Introduction rotor position sensor by injecting high frequency sig-
nal. This motor will act as a position sensor due to
its unsaturated phase inductance [9]. A non-contact
Electric drives take significant role in industry applica- optical sensor can also be used to detect the relative
tions. They formerly use DC motors and AC induction position between rotor and stator at high speed [10].
motors. Due to the rapid progress of the power elec- Commonly, the direct rotor detections at high speed
tronics and digital control developments, BLDC and result in fluctuation in switching angles, analogue en-
SRM types of motors begin to take parts in electric coder can be used to synchronize the stator phase ex-
drive. Green technology issues have been triggering citation and the rotor position of SRM [11].
the developments of Electric Vehicles (EV) and Hy-
brid Electric Vehicles (HEV) [1] and [2]. Although the The performance of SRM is influenced by the ave-
torque generated by SRM is smaller than BLDC, SRM rage torque generated, the better performance can be

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 261
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

obtained under larger positive torque and smaller neg- energy (Wf ). For the power is the rate of change of
ative torque. The instant time when the switches are energy, then:
turned on and off will determine the motor perfor- ∂λ ∂Wm ∂Wf
mance. In this paper, a control strategy for SRM using i· = + . (3)
∂t ∂t ∂t
rotary encoder based rotor position detection is pro-
posed. The proposed strategy is capable to change the If the relationship between the mechanical power
points when the static switches are on and off by soft- (Pm ) and torque (T ) is defined as:
ware. This can provide proper excitation to the stator
winding so the higher average value of the torque can ∂Wm ∂θ
Pm = =T ·ω =T · , (4)
be generated. To verify the analysis, simulations and ∂t ∂t
experiments were carried out. then
∂λ ∂Wf
T =i· − . (5)
∂θ ∂θ
2. Switched Reluctance Motor By using energy-coenergy concept as depicted in
Fig. 2, the stored field energy (Wf ) and coenergy (Wc )
A Switched Reluctance Motor (SRM) has very simple can be written as the following:
construction with no permanent magnet on its rotor Z λ
and has simple phase winding on its stator. An equiv- W f = i(θ, λ)∂λ, (6)
0
alent circuit of SRM is shown in Fig. 1.
Z i
Wc = λ(θ, i)∂i. (7)
0
+ i � The area of the sum of the energy and coenergy can
R vR be declared by

� Wf + Wc = λ · i. (8)

vs d � �� , i � by differentiating Eq. (8), then

L e�
dt
�
-
flux linkage

Wf
Fig. 1: Block diagram for measuring traction parameters.
Wc
To simplify the analysis, effects caused by saturation,
fringing and leakage fluxs, and mutual inductance are i
neglected so we obtain [12]:
current
∂λ(θ, i)
v = R · i + e(θ, i) = R · i + , (1) Fig. 2: The curve of energy-coenergy concept.
∂t
where R, i, e, θ and λ are the phase resistance, phase ∂Wf + ∂Wc = λ∂i + i∂λ, (9)
current, back EMF, rotor position and linkage flux. To
have the power quantity, Eq. (1) must be multiplied by and
∂(−λ∂i + ∂Wc )
the current. T = . (10)
∂θ
∂λ(θ, i)
v · i = R · i2 + i · . (2) Assuming the current is constant,
∂t

d · 12 L(θ)i i


The left-hand side term of the equation indicates the ∂Wc
T = = , (11)
power supplied from the source while the first term of ∂θ ∂θ
right-hand side is the ohmic dissipation and the second and finally the torque equation can be calculated as
term of right-hand side is the sum of the mechanical 1 ∂L(θ)
power (Wm ) as the output and the stored magnetic T = i2 . (12)
2 ∂θ

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 262
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

3. Converter Topology
D1 D2 D3
A SRM with three phase stator winding and four pole Q4

rotor (6/4 SRM) requires three phase converter to gen- Vdc

La Lb Lc
erate sequential voltage waveforms to drive. Some con- D4
D4
verter topologies can be used as an electric drive. In
this paper, (n + 1) topology is used as a power circuit. Q2 Q3
Q1
For three phase application, this topology consists of
four static switches and four diodes (Fig. 3). Three
mode of operation can be implemented to operate this (a) With upper side switch on.
converter. Energizing the SRM phase stator winding
can be done by switching on the upper side switch (Q4)
and one of the lower side switches (magnetizing mode). D1 D2 D3
Q4
This will make the phase current increase, when the Vdc
slope of the inductance profile of the appropriate phase La Lb Lc
D4
stator is positive then the positive torque is developed D4

(Fig. 4(a)).
Q1 Q2 Q3

D1 D2 D3
Q4
Vdc (b) With lower side switch on.
La Lb Lc
D4 Fig. 5: The (n + 1) converter under freewheeling mode.

Q1 Q2 Q3

second mode is freewheeling that make the phase stator

winding short circuit then the current goes down slowly
Fig. 3: Topology of (n + 1) converter for 6/4 SRM.
(Fig. 5).

D1 D2 D3
Q4
Vdc 4. The Proposed Control
La Lb Lc
D4
Strategy

Q1 Q2 Q3 The proposed system consists of 6/4 SRM with rotary

encoder as the detector for rotor position, a converter
(a) The converter under magnetizing mode. which is configured as the (n + 1) type and a digital
control as depicted in Fig. 6. By using 2000 PPR (pulse
per rotation) rotary encoder and setting the unaligned
position of rotor to the phase-A winding as the starting
D1 D2 D3
Q4
pulse, the mapping of the inductance profile due to the
Vdc pulses of rotary encoder can be represented as in Fig. 7.
La Lb Lc
D4

Q1 Q2 Q3
2000 PPR
Rotary Encoder
DC Source (n+1) type
(b) The converter under demagnetizing mode. SRM RE
Converter
Fig. 4: Topology of (n + 1) converter for 6/4 SRM.

To turn off the current flowing in the active phase TLP 250 Digital Signal
winding, two modes of operation can be used. The first driver Controller

mode is called demagnetizing, this make the opposite

polarity of the source connected to the stator winding
then the current will go down fast (Fig. 4(b)). The Fig. 6: The proposed drive system.

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 263
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

1 rotation � min
2

unaligned-3

unaligned-4
unaligned-1

unaligned-2
inductance 45o

Lmax 0o

2000 pulses
Lmin
500 pulses gate pulse
Fig. 7: Mapping of the phase-A inductance profile due to the
pulses of the rotary encoder.

y x
For the SRM has four pole rotor, it means that every
unaligned position is displaced by 90 degree (mechani-
cal degree) and represents 500 pulses of rotary encoder. � on � off
If the rotor runs one rotation, there will be an induc-
tance profile of the phase stator that repeats four times. Fig. 9: Controlling the width of the pulse for the SRM drive.
It results in the 30 degree displacement angle between
two adjacent phase winding inductance profile (Fig. 8).
number of pulses generated by the rotary encoder as-
sociated with the maximum inductance profile can be
900 �m � � 3600 �e � derived as:
3
La Lb Lc Amax (n) =
6
500 + (n − 1)500, (15)

5
Bmax (n) = 500 + (n − 1)500, (16)
6
1
Cmax (n) = 500 + (n − 1)500, (17)
ia 6
where Amax , Bmax and Cmax are the number of the
pulses generated by the rotary encoder at maximum
ib inductance of the stator winding-A, stator winding-B
and stator winding-C. In the operation of the SRM,
ic changing θon and θoff can be used to improve the per-
formance of the motor due to Eq. (11).
Basically, to have high average value of the torque
120 0
�e � generated by the SRM, higher current should be given
in the phase winding when its inductance profile slope
Fig. 8: Displacement angle of each phase inductance profile and is positive. Unfortunately due to characteristic of in-
ideal phase current waveforms for excitation. ductive circuit, the current is delayed to flow. Ener-
gizing the phase winding at the earlier instant time is
very important to overcome the delay problem. The
If it is assumed that minimum inductance equals position of θon will determine the motor performance.
θmin , the voltage must be applied to energize the phase- When θon is too closer to 0◦ , there is a time interval
A winding at angle θon (turn on angle) and switched when the current flow under zero inductance profile
off at turn off angle θoff where slope. This means the SRM absorbs the power from
θmin the source with no torque generated. But if θon is near
0 ≤ θon ≤ , (13) the point(θmin /2), then a the inductance profile starts
2
to increase without adequate current value. The torque
θmin generated by the SRM under this condition may not
≤ θoff ≤ 45◦ (14)
2 be large enough. Different impacts arise on the place-
Using a rotary encoder as a rotor position detector in ment of θoff , when the switching off occurs at or greater
SRM makes the relationship between the inductance than 45◦ the current remain to flow under the negative
profile and the number of pulses more important. The slope of the inductance profile. This results in negative

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 264
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

START Read ADC

N
INT1 ON first stage

p = p+1

Initialize ADC
Initialize Timer (A-x)<p<(A-y)
N
(B-x)<p<(B-y)
N
(C-x)<p<(C-y)
N

Initialize INT1,INT2
Value x & y Y Y Y Q1 �OFF
Q2 �OFF
m == 0 Q3 �OFF
N N N
ADC>TMR ADC>TMR ADC>TMR Q4 �OFF

Y Y Y

Q1-Q4 ON Q1 �ON Q1 �OFF Q2 �ON Q2 �OFF Q3 �ON Q3 �OFF

Y Q4 �ON Q4 �OFF Q4 �ON Q4 �OFF Q4 �ON Q4 �OFF
Q2-Q3 OFF INT2 ON
delay
N Fig. 12: Flowchart of the control strategy for the second stage
with magnetizing and demagnetizing operation modes.
Q2-Q4 ON
Y
Q1-Q3 OFF INT2 ON
delay
θon and θoff to produce higher average torque. By us-
N
ing rotary encoder, the position of the SRM rotor can
Q3-Q4 ON
be mapped into the number of pulses generated. For
Y rotary encoder has some outputs that represent the
Q1-Q2 OFF INT2 ON
delay number pulses generated (pin A) and indicate pulses
N every revolution (pin Z). The Digital Signal Controller
N (DSC) can be implemented as the core of the control
strategy. We know that the pulses generated by the
second m == 1
p=0
m == 1 rotary encoder are free running pulses so tagging the
stage Y
pulse as the initial position reference is required. Due
to this problem, the control strategy to drive SRM is
Fig. 10: Flowchart of the control strategy for the first stage. divided into two stages. The first stage is addressed to
find the initial position reference by using sequence ex-
citation given to the SRM stator windings. The rotor
N
Read ADC INT1 ON first stage will start to move until the position reference obtained
Y
Y and then it automatically changes into the second stage
p = p+1 that is functionalized to drive the motor.
In the first stage, the SRM is driven like a stepping
N N N
motor for a moment until the pulse from pin Z of rotary
(A-x)<p<(A-y) (B-x)<p<(B-y) (C-x)<p<(C-y)
encoder that connected to INT2 is generated to set the
variable m that indicates that the origin pulse of pin
Y Y Y
A is started to count and is marked by the increment
Q1�ON
Q2�OFF
Q1�OFF
Q2�ON
Q1�OFF
Q2�OFF
value of p (a variable that counts the pulses in every
Q3�OFF Q3�OFF Q3�ON step in a revolution). To control the time when the
switches go to on and off states by software, parameters
N x and y in Fig. 11 and Fig. 12 vary.
ADC>TMR

Q4�ON Q4�OFF 5. Results and Discussion

To validate the above analysis, computer simulations
Fig. 11: Flowchart of the control strategy for the second stage
and experiments were carried out. The (n + 1) con-
with magnetizing and freewheeling operation modes.
verter was used in the works as the power circuit (de-
picted in Fig. 6). By using PSIM, simulation were per-
torque produced then the average value of the torque formed with parameters described in Tab. 1.
will be low.
The SRM speed, phase current, phase voltage and
In designing the control strategy, the above analy- inductance profile are investigated in the simulations
sis must be taken as the technical considerations. The to prove that the control strategy is capable to change
control strategy is also capable to set the position of the value of the θon and θoff . This can affect the SRM

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 265
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

Tab. 1: SRM parameters for simulation and experimental 1500

1200
80

works 1000
1000
60
800

(RPM)
Resistance 0.5 Ohm

(A)
600

Current
Max. Inductance 3 mH 40
500

speed
400
Min. inductance 1 mH
200
Stator / Rotor Pole 6/4 20 0

0
Min. Inductance Angle 20 degree
0
-200
-500
0.970 0.10.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.99 0.7 0.995 0.9
0.8 1
1
time
time(s)
(s)

performance especially the developed torque. Under (a) Phase current.

the same load connected to the SRM, the higher torque 1500
1200
4
causes higher value of the motor speed. Operating un- 1000

der θon = 0◦ and θoff = 45◦ causes that the current 1000
3
800

Inductance (mH)
speed (RPM)
flowing in the stator winding has high value at the time 600
2
500
when the inductance profile begins to rise (Fig. 13). 400

But the current is still high under negative slope of the 200
10

inductance profile so the average torque is not large 0

0
and finally results in low speed (Fig. 14). -200
-500
0.970 0.1 0.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.990.7 0.995 0.9
0.8 1
1
time
time(s)
(s)

Finally, a prototype was designed to verify the pro- (b) Inductance profile.
posed control strategy (Fig. 23). The 2000 PPR (pulse
per revolution) rotary encoder is used to detect the ro- 1500
1200
8

tor position. The core of the controller implements the 1000

6
1000
800
16 bit digital signal controller (dsPIC30F4012). The 4
Torque(RPM)
(Nm)

600
output pins A and Z of the rotary encoder are con- 2500
speed

400
nected to INT1 and INT2 pins of the dsPIC to inform 0
200
0
the pulses generated and the reset pulse every revolu- -2
0
tion. Parameters for this work are presented in Tab. 1. -4
-500
-200
0.97 0 0.10.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.99 0.7 0.995 0.9
0.8 1
1
Tests of SRM with converter operated in magnetizing time
time(s)
(s)
and freewheeling modes are depicted in Fig. 24 and (c) Developed torque.
Fig. 25.
1500
1200
60
Those experimental results show that the currents 1000
50
flowing in the stator winding tend to decrease slowly 1000
800
40
(RPM)

because the voltage applied on this winding is zero

(V)

600
30
Voltage

500
(freewheeling mode).
speed

20
400

10
200
Experimental work using on equivalent to 1 pulse 0
00
◦
number (θon = 0 ) and θoff equivalent to 400 pulse
-10
-500
-200
number (θoff = 360◦ ) rotate the motor on the speed 0.970 0.1 0.975 0.2 0.98
0.3 0.4 0.985
time
0.5
time(s)
(s)
0.6 0.990.7 0.995 0.9
0.8 1
1

4285 RPM as shown in Fig. 24. If the off enlarges to 500

(d) Phase voltage.
pulse number, the lower speed is obtained (2857 RPM)
as depicted in Fig. 25. Using the last parameter but the
Fig. 13: Simulation result of the SRM drive under magnetizing
converter operated under magnetizing and demagnetiz- and freewheeling modes (θon = 0◦ and θoff = 45◦ ).
ing modes results in higher speed (3864 RPM). Under
this condition, the current flowing in the stator wind- 1500

ing is drastically reduced because the voltage applied

1000
on the stator winding has negative polarity (Fig. 26).
speed (RPM)

Due to Eq. (4), under constant torque, the output 500

power is proportional to the rotor speed. The proper

θon and θoff will result in higher speed and higher out- 0

put power because the negative torque can be mini-

mized, thus the efficiency is improved. -500
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
time (s)
Adjusting the θon and θoff by software changes the
motor speed, this means that the proposed control Fig. 14: Simulation result of the SRM drive speed under mag-
method is effective in driving the SRM. It can be done netizing and freewheeling modes (θon = 0◦ and
θoff = 45◦ ).
by entering the value of x and y (Fig. 10) that represent
the value of θon and θoff .

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 266
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

1500
1200
70 1500
1200
60

1000
60 1000
50
1000 1000
800
50 800
(RPM)

(RPM)
(A)

(A)
600
40 40
600
Current

Current
500 500
speed

speed
30
400 400
30
20
200 200
0 0
20
10 0 0

0
-200
-500 10
-500
-200
0.970 0.10.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.99 0.7 0.995 0.9
0.8 1
1 3.970 0.13.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.99 0.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(a) Phase current. (a) Phase current.

1500
1200
4 1500
1200
4

1000 1000
1000
3 1000
3
800 800
Inductance (mH)

Inductance (mH)
speed (RPM)

speed (RPM)
600 600
2
500 2
500
400 400

200 200
10 10

0 0

0
-200
-500 0
-200
-500
0.970 0.1 0.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.990.7 0.995 0.9
0.8 1
1 3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(b) Inductance profile. (b) Inductance profile.

1200
1500
8 1500
1200
4

1000
6 1000
3
1000 1000
4800 800
2
Torque(RPM)

Torque(RPM)
(Nm)

(Nm)

2600 600
500 1500
speed

speed

0400 400
0
-2200 200
0 0
-4 0 -1
0

-6
-500
-200 -2
-500
-200
0.97 0 0.10.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.99 0.7 0.995 0.9
0.8 1
1 3.97 0 0.13.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.99 0.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(c) Developed torque. (c) Developed torque.

1200
1500
60 1500
1200
60

1000
50 1000
50
1000 1000
800
40 800
40
(RPM)

(RPM)
(V)

(V)

600
30 600
30
Voltage

Voltage

500 500
speed

speed

20
400 20
400

10
200 10
200
0 0
00 00

-10
-500
-200 -10
-500
-200
0.970 0.1 0.975 0.2 0.98
0.3 0.4 0.985
0.5 0.6 0.990.7 0.995 0.9
0.8 1
1 3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(d) Phase voltage. (d) Phase voltage.

Fig. 15: Simulation result of the SRM drive under magnetizing Fig. 17: Simulation result of the SRM drive under magnetizing
and freewheeling modes (θon = 5◦ and θoff = 45◦ ). and freewheeling modes (θon = 0◦ and θoff = 30◦ ).

1200
1500 3000
1200
1500

1000 2500
1000
1000 1000
800 2000800
speed (RPM)

speed (RPM)

600 1500600
500 500
400 1000400

200 500200
0 0
0 0 0

-500
-200 -500
-500
-200
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0 0.5
0.1 0.21 0.3 1.5 0.4 20.5 0.6
2.5 0.7 3 0.8 3.50.9 1
4
time (s) time
time(s)(s)

Fig. 16: Simulation result of the SRM drive speed under mag- Fig. 18: Simulation result of the SRM drive speed under mag-
netizing and freewheeling modes (θon = 5◦ and netizing and freewheeling modes (θon = 0◦ and
θoff = 45◦ ). θoff = 30◦ ).

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 267
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

1500
1200
50 1500
1200
40

1000 1000
40 30
1000 1000
800 800
(RPM)

(RPM)
(A)

(A)
30
600 20
600
Current

Current
500 500
speed

speed
400
20 400
10

200 200
0 0
10 0
0 0

0
-200
-500 -10
-500
-200
3.970 0.13.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.99 0.7 3.995 0.9
0.8 4
1 3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(a) Phase current. (a) Phase current.

1500
4 1200 1500
1200
4

1000 1000

3 1000 1000
3
800 800
(mH)

Inductance (mH)
speed (RPM)

speed (RPM)
600 600
Inductance

2 500 2
500
400 400

200 200
1 0 10

0 0

0 -200
-500 0
-200
-500
3.97 0 3.975
0.1 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.99 0.7 3.995
0.8 0.9 14 3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1
time (s)(s)
time time
time(s)
(s)

(b) Inductance profile. (b) Inductance profile.

1500
1200
2.5 1500
1200
2

1000
2 1000
1000 1000
1.5
1.5800 800
Torque(RPM)

Torque(RPM)
(Nm)

(Nm)

1600 600
500 1
500
speed

speed

0.5400 400

0200 200
0 0.5 0
-0.5 0 0
-1
-500
-200 0
3.97 0 -200
-500
0.13.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 0.83.995 0.9 14 3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(c) Developed torque. (c) Developed torque.

60
1500
1200 1500
1200
60
50
1000 1000
40
1000 1000
40
800 800
(RPM)

20
(RPM)
(V)

(V)

30
600 600
Voltage

Voltage

500 0
500
speed

speed

20
400 400
-20
10
200 200
0
-40 0
00 0
-60
-10
-500
-200 -200
-500
3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1 3.970 0.1 3.975 0.2 3.98
0.3 0.4 3.985
0.5 0.6 3.990.7 3.995 0.9
0.8 4
1
time
time(s)
(s) time
time(s)
(s)

(d) Phase voltage. (d) Phase voltage.

Fig. 19: Simulation result of the SRM drive under magnetizing Fig. 21: Simulation result of the SRM drive under magnetizing
and freewheeling modes (θon = 5◦ and θoff = 30◦ ). and freewheeling modes (θon = 5◦ and θoff = 30◦ ).

3000
1200
2500
1500 3000
1200
2500
1500

2500
1000 2500
1000
2000 2000
1000 1000
2000800 2000800
speed (RPM)

speed (RPM)

1500 1500
1500600 1500600
500 500
1000400 1000400
1000 1000
500200 500200
0 0
500 500
0 0 0 0

-500
-500
-200
0 -500
-500
-200
0
0 0 0.5
0.10.5 0.21 1 0.3 1.51.50.4 20.5
2 0.6
2.5
2.5 0.7 33 0.8 3.5
3.50.9 1
4 0 0 0.5
0.10.5 0.21 1 0.3 1.51.50.4 20.5
2 0.6
2.5
2.5 0.7 33 0.8 3.5
3.50.9 1
4
time
time(s)(s) time
time(s)(s)

Fig. 20: Simulation result of the SRM drive speed under mag- Fig. 22: Simulation result of the SRM drive speed under mag-
netizing and freewheeling modes (θon = 5◦ and netizing and freewheeling modes (θon = 5◦ and
θoff = 30◦ ). θoff = 30◦ ).

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 268
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

6. Conclusion
The control strategy to drive the SRM motor using ro-
tary encoder as the rotor position detection has been
presented in this paper. The strategy consists of two
stages, the first stage is used to search the initial pulse
number as the reference and the second stage is the
main control to drive the SRM. The proposed con-
trol strategy is capable to tune the angle of turning on
and turning off the switches by software so the torque
produced is better. Simulations and experiments were
Fig. 23: The prototype for experimental works.
carried out to verify the analysis and these results can
prove the effectiveness of the proposed control strategy.

Acknowledgment
This work was supported by The Ministry of Research,
Technology and Higher Education, Republic of Indone-
sia.

Fig. 24: Experimental result of the SRM drive under magne-

tizing and freewheeling modes (a) phase voltage (b) References
phase current (pulse number on: 0, pulse number off:
400).
[1] EMADI, A., Y. J. LEE and K. RAJASHEKARA.
Power Electronics and Motor Drives in Electric,
Hybrid Electric, and Plug-In Hybrid Electric Ve-
hicles. IEEE Transactions on Industrial Electron-
ics. 2008, vol. 55, iss. 6, pp. 2237–2245. ISSN 1000-
6644. DOI: 10.1109/TIE.2008.922768.

[2] SHAO, J. An Improved Microcontroller-Based

Sensorless Brushless DC (BLDC) Motor
Drive for Automotive Applications. IEEE
Transactions on Industry Applications. 2006,
vol. 42, iss. 5, pp. 1216–1221. ISSN 0093-9994.
Fig. 25: Experimental result of the SRM drive under magne-
tizing and freewheeling modes (a) phase current (b) DOI: 10.1109/TIA.2006.880888.
phase voltage (pulse number on: 0, pulse number off:
500). [3] EHSANI, M. and B. FAHIMI. Elimination of
Position Sensors in Switched Reluctance Mo-
tor Drives: State of the Art and Future
Trends. Transactions on Industrial Electronics.
2002, vol. 49, iss. 1, pp. 40–47. ISSN 0278-0046.
DOI: 10.1109/41.982246.

[4] HUSAIN, I. and M. EHSANI. Rotor Position

Sensing in Switched Reluctance Motor Drives
by Measuring Mutually Induced Voltages. IEEE
Transactions on Industry Applications. 1994,
vol. 30, iss. 3, pp. 665–672. ISSN 0093-9994.
DOI: 10.1109/28.293715.

[5] XU, Z. Sensorless Switched Reluctance Mo-

Fig. 26: Experimental result of the SRM drive under magne-
tor Driver System Control by Detecting the
tizing and freewheeling modes (a) phase current (b) Derivative of the Phase Current. In: The 2011
phase voltage 3846 RPM (pulse number θon : 0, pulse International Conference on Electric Informa-
number θoff : 500). tion and Control Engineering. Wuhan: IEEE,

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 269
POWER ENGINEERING AND ELECTRICAL ENGINEERING VOLUME: 16 | NUMBER: 3 | 2018 | SEPTEMBER

2011, pp. 1116–1119. ISBN 978-1-4244-8036-4. Machines Based on Non-contact Optical Sen-
DOI: 10.1109/ICEICE.2011.5777929. sors. In: The 2017 IEEE Electric Machines and
Drives Conference (IEMDC). Miami: IEEE,
[6] PANDA, D. and V. RAMANARANAYAN. 2017, pp. 1–6. ISBN 978-1-5090-4281-4.
Sensorless Control of Switched Reluctance Motor DOI: 10.1109/IEMDC.2017.8002289.
Drive with Self-Measured Flux-Linked Char-
acteristics. In: The IEEE 31st Annual Power [11] AHN, J. W., S. J. PARK and D. H. LEE.
Electronics Specialists Conference. Galway: IEEE, Novel Encoder for Switching Angle Control of
2000, pp. 1596–1574. ISBN 0-7803-5692-6. SRM. IEEE Transactions on Industrial Electron-
DOI: 10.1109/PESC.2000.880539. ics. 2006, vol. 53, no. 3, pp. 848–854. ISSN 0278-
0046. DOI: 10.1109/TIE.2006.874278.
[7] TRAORE, W. F. and R. A. MCCANN. Ro-
tor Position Estimation in a Switched Re- [12] DIRENZO, M. T. Application Report:
luctance Motor Using Embedded Magnetic SPRA420A. Switched Reluctance Motor Con-
Field Sensors. In: The 2010 IEEE Power and trol - Basic Operation and Example Using the
Energy Society General Meeting. Providence, TMS320F240. Texas Instruments, 2000.
RI: IEEE, 2010, pp. 1–11. ISBN 978-1-4244-6549-
1. DOI: 10.1109/PES.2010.5590141.
[8] CAI, J. and Z. DENG. Switched-Reluctance Po- About Authors
sition Sensor. IEEE Transactions on Magnetics.
2014, vol. 50, iss. 11, pp. 1–4. ISSN 1478-7148. Slamet RIYADI was born in Semarang, Indonesia.
DOI: 10.1109/TMAG.2014.2329957. He received his B.Sc. from Diponegoro University
[9] CAI, J., Z. LIU, H. JIA, X. ZHAO and F. XU. Semarang in 1991, M.Sc. from Bandung Institute
An Inductive Position Sensor with Switched Re- of Technology, Indonesia in 1997. In 2006, he re-
luctance Motor Structure. In: The 20th Interna- ceived Ph.D. degree in Electrical Engineering from
tional Conference on Electrical Machines and Sys- Bandung Institute of Technology with Ph.D. partial
tems (ICEMS). Sydney: IEEE, 2017, pp. 1–4. research done in ENSEEIHT-INP Toulouse France.
DOI: 10.1109/ICEMS.2017.8056495. His research interests include power converter, power
quality, PV-Grid Connected System and Electric
[10] GONG, C., T. HABETLER, J. RESTREPO, Drive. Currently he is with the Department of Elec-
and B. SODERHOLM. Direct Position Con- trical Engineering, Soegijapranata Catholic University
trol for Ultra-high Speed Switched Reluctance Semarang Indonesia.

c 2018 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 270