Vol. 3, No. 2 | July - December 2020

Design of Low-Cost Synchronous Reluctance Motor with a

Surrogate-Assisted Optimization Technique

Syed Abid Ali Shah Bukhari1*, Imtiaz Ali Leghari1, Junaid Iqbal Bhatti1,

Saleem Raza1, Mohsin Ali Koondhar1, Irfan Ali Channa2, Muhammad Usman

Keerio1

Abstract:

This paper presents the performance of centrifugal pump coupled with synchronous

reluctance motor of 24kW. The 4-pole 36 slots induction motor stator has been used for the

prototype. The reduction of torque ripple and power losses is the main aim of research. Different

arrangement of rotor flux barrier shapes has been tested through finite element analysis method.

The rotor optimization is done by varying different parameters are used such as, barrier edge

angle, width of flux carriers, flux barriers and shaft diameter. In order to improve the motor

performance, the method of particle swarm optimization has been used by generating samples

with surrogate assisted optimization technique. Distinct flux barrier shapes and designs have

been checked through simulation. For the prototype development, final V-shaped best model is

chosen. The proposed rotor shows the excellent performance in terms of reduced torque ripple

(27%) and improved average torque (154Nm) in experimental results. The suggested design also

has good thermal and mechanical performances with the capability to use in various industrial

applications.

Keywords: torque ripple, particle swarm optimization, surrogate based technique, flux barrier,

machine design, synchronous reluctance motor, direct drive.

1. Introduction

The construction of transverse-laminated
rotor structure of synchronous reluctance
motor (SynRM) is attracting the attention by
the different users, because of its simple
construction, cost, robustness and
manufacturing process [1],[2]. Typically, a
pumping system is mostly used to couple an
induction motor of 1500 rpm with a
centrifugal pump but, this system has a lot of
disadvantages such as heavy and bulky motor
drive system with low energy efficient. In this

1 Quaid-e-Awam University of Engineering, Science and Technology, Pakistan
2 Department of Automation, Beijing University of Chemical Technology, Beijing, China
Corresponding Author: abidshah@quest.edu.pk

regard, for high-torque applications,
permanent magnet synchronous machines
(PMSM) are dominated, but the most
important issue is the recycling of these rare-
earth metal permanent magnets. Additionally,
the cost of magnetic material collectively with
demand and supply chain issues is actually
preventing these machines from assuming its
rightful place as a motor of choice near future.
So as the price rate of rare-earth magnets is
growing, at the same time their market
stability is declining [3]. Thus, a universal
satisfactory solution has not been adopted for

mailto:abidshah@quest.edu.pk

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

application of closed coupled centrifugal
pumps. After an improvement in electrical
drives in 20th century due to enhanced
efficiency and fast dynamic response of the
drives, the reluctance motors represents a
much possible alternatives [4],[5].

The reluctance motors have a unique rotor
structure which is arranged from laminated
steel and do not use permanent magnet. There
is salient rotor and works on the basis of
reluctance torque. There are two types of
reluctance machine the number one is
switched reluctance motor (SRM) and second
is synchronous reluctance motor (SynRM). If
the stator is of round configuration and fed
with AC supply with different phases the
machine is called synchronous reluctance
machines. If the stator has salient poles
structure, the machine is called switched
reluctance machine. The SRM has high
torque ripple and separate DC input source is
required to excite the winding. Whereas
SynRM utilize sinsoidally distributed
windings as in induction or synchronous
machines and requires easily available
universal sinusoidal supply, so that’s why
author has decided to take SynRM for the
research.

Many industrial applications require low
torque ripple, therefore lots of studies have
been conducted and to enhance the
performance of the machine, like in
[6],[7],[8],[9], the torque ripple is
investigated and multiple solutions are tested.
Further to enhance the torque density of the
machine, the rotor design is investigated in
[10],[3],[11] in which the saliency ratio is
examined which denotes the ratio between the
d-axis and q- axis.

For SynRM rotor, the type of axially
laminated anisotropy (ALA) is mostly used in
a high-speed operation as proved in many
research investigations [12-15]. Due to rigid
structure and improved saliency ratio [16-18].
However, because of very thicker laminations
it produces unwanted flux oscillations with
very high iron losses. Therefore, TLA type
rotor is most preferable because of simple
construction and low losses [19-22]. Hence,
due to above advantages of SynRM
simulation model design it has been decided
to use TLA type rotor. It is aimed to select the

number and shape of flux barriers to achieve
the low power losses along with low torque
ripples. The paper is distributed in three
different sections. The section I describes the
background of research.

(a)

(b)

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

(c)

Fig 1: Four flux barrier rotor’s different

designs of SynRM (a) 4 V-shaped flux barrier

rotor (b) 4 square shaped rotor (c) 4 straight

flux barriers with slight angle at the edges.

The detailed rotor design with three and
four flux barriers and its advantages have
been described through simulation in section
II. Also, in order to minimize torque ripple
and to enhance efficiency, the analysis on flux
barriers and flux carriers carried out by
varying the width, and shaft diameter while
using Infolytica Magnet software. In section
III the development of rotor from
manufacturing point of view and its process is
mentioned. In order to validate the proposed
rotor model, a test bench is set up by
manufacturing the designed rotor and the
prototype development steps are conducted
with the experimental results and discussion
with no-load test, low-slip test and Over speed
test operation and experimental results of the
proposed machine have been described in
section IV. In section V, the brief analysis of
results has been presented. Finally, section VI
describes the research methodologies along
with application and future trends.

(a)

(b)

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

(c)

(d)

Fig 2: Three and four flux barrier rotor

designs for SynRM (a) 4 square shaped flux

barrier rotor (b) 3 V- shaped rotor (c) 4 square

shaped flux barriers with slight angle at the

end and 3 square shaped flux barriers.

2. Working prniciple of SynRM

In SynRM, to achieve the maximum
output, the saliency ratio (Ld-Lq) should be
maximized. In this rotor design 0.6 to 0.7
percent saliency ratio has been used.
However, the torque is generated through
reluctance variation at different rotor
positions. The motor torque is directly
proportional to the transformation of the
magnetizing inductance (d-q axis) which
impacts the coordinate system of rotor
reference frame. In machine design, the main
challenge is to maintain thermal constraints
because of stator and rotor windings [23].
Therefore, it is essentially needed to reduce
the involvement of winding material and
magnetic mass. In [24] a low speed domestic
application with direct-drive SynRM having
six rotor geometries has been presented. Most
of the SynRM employs advanced
transversally laminated rotor structure. The
rotor cage design is taken out as the machine
can be easily started synchronously through
inverter control system. Therefore d-q
equations of SynRM [25] are established as
follows.

𝑉𝑑𝑠 = 𝑟𝑠 𝑖𝑑𝑠┿
𝑑

𝑑𝑡
𝐿𝑑𝑠 𝑖𝑑𝑠 – 𝜔𝑟 𝐿𝑑𝑠 𝑖𝑞𝑠 (1)

𝑉𝑞𝑠 = 𝑟𝑠 𝑖𝑞𝑠┿
𝑑

𝑑𝑡
𝐿𝑞𝑠 𝑖𝑞𝑠 – 𝜔𝑟 𝐿𝑑𝑠 𝑖𝑑𝑠(2)

The Lds and Lqs represents the quadrature
and direct axis inductances. The (ωr) is speed,
whereas each phase of stator resistance is
denoted by (𝑟𝑠 ). The electromagnetic torque
in terms of d-q variables, is identical to that of
a SynRM, namely

𝑇𝑒 =
3

𝑃

2
(𝐿𝑑𝑠– 𝐿𝑞𝑠) 𝑖𝑑𝑠 𝑖𝑞𝑠 (3)

Whereas number of poles are denoted by P.

In order to model the asynchronous machine,

above equations are used.

3. design and simulation of SynRM

In order to choose the best rotor design,
seven different geometries of three and four
flux barrier have been examined and finite
element analysis carried out. All the test was
conducted at 1500RPM as Fig. 1 shows the
different flux barrier designs of the rotor. (a)

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

shows the four flux barriers with V-shaped
designs in this design the flux barrier edge
angle is kept 49 degree and the torque ripple
is 27% which is good but machine efficiency
is very low and the total average torque
produces by the machine is 134Nm and one
additional factor which is the flux density is
higher up to 2.29. In design (b) the more angle
is provided to sharp edges but the the
efficiency reduces to 37% and torque ripple is
still higher to 70% and overall average torque
is also reducing to 93Nm with flux density
2.2T. In design 132Nm torque is produced
with 2.38T flux density and the torque ripple
is 42%. In Fig. 2 (a) circular shape with four
flux barrier design are tested which gives the
efficiency 47% with 118Nm torque and 2.18T
whereas, torque ripple is slightly higher 44%.
In (b) three V-shaped flux barrier design are
tested which produce the highest torque
158Nm with 70% and low torque ripple 30%
but flux design is higher 2.7T. The design (c)
gives 134Nm torque and 2.5T density and
torque ripple is higher upto 46% with 63%
efficiency. In (d) three square shaped flux
barrier are used which produces 134Nm
torque and 47% torque ripple with 2.17T flux
density.

4. Development of Synchronous
reluctance Motor

Keeping above data analysis in terms of
torque, efficiency, torque ripple and flux
density in section III, the author decide to take
three V-shaped design as a main investigation
point due to its higher high efficiency and
torque production. Further design variables
flux barrier and carrier width with edges and
shaft diameter are modified accordingly.
Consequently, a 3-flux barrier’s rotor of
24kW SynRM with motor has been simulated
with the specifications shown in table, I and
II.

Fig 3: Design rotor dimensions of SynRM

Table I: 24kW SynRM specifications.

Parameter Units Value Symbol

Number of phases - 3 m

total flux barriers - 3 -

speed rpm 1500 nN

Number of turns turns 12 Np

Number of stator

poles

- 4 Ns

frequency Hz 50 f

Rated current Amps 26 I

angle of edges o 4 deg

Rated power kW 24 P

Saliency ratio 0.7

Phase Voltage V 380 Vdc

Stator diameter mm 310 ds

Air-gap mm 1 g

Machine length mm 200 l

Average torque Nm 154 Tav

diameter mm 188 dr

Shaft diameter mm 50 dsh

Table II. SynRM rotor design Specifications

Item Value Item Value

Rotor

Materi

M350-

50Amp

carrier

width

12.06,

9.92,9.97,10

.21

barrie

width

9.02,

8.97,

8.81

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

Overal

carrier

width

42.17 Overa

barrie

width

26.822

Rotor

188 mm Flux

barrie

r edge

angle

3 and 4

deg

length of first barrier (starting at outer

diameter)

Lower 35.9mm Upper 36mm

length of second barrier

Lower 36mm Upper 36mm

length of third flux barrier

Lower 35.63mm Upper 35.91m

Note: All the dimensions are depicted in

rotor’s half or 4th quadrant in Fig. 3.

At first four and five barriers with square
shaped ware simulated. However, their
performance was not reasonable and machine
was generating 2.7T magnetic flux density
and the produced torque was also low in
between (100 to 120Nm). Then
configurations of 3 flux barrier were
simulated with V-shaped flux barrier design
has been proposed, as depicted in fig. 4(a)-
and two-dimensional mesh view is shown in
Fig. 4(b). The 2.1 Tesla of flux density,
current waveform and produced torque is
shown in Fig. 5 (a to c) respectively.

(a)

(b)

Fig 4: (a) Model of two dimensional SynRM

(b) Triangular view of meshes

At 500 milliseconds transient simulation

with velocity-driven settings carried out, thus
machine can produce effective starting torque
180 Nm and 154.30Nm average torque with
90% efficiency. The portion of the shaft kept
hollow to avoid additional magnetic losses
therefore its losses are very negligible, so
that’s why has author has not considered in
this work. the copper, hysteresis and eddy
current losses are 2.358watts and the stator
hysteresis and eddy current losses are
379watts,

The 1000/65 Newcore nonlinear
lamination material selected for the stator and
for rotor lamination M350-50A material used
in the machine.

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

(b)

(a)

(c)

Fig. 5 (a) 2.1 (Tesla) flux density (b) Flux linkage (c) generated torque of the SynRM

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

Fig 6: The 4 pole 36 slots double layer stator

5. Experimental Validation

Subsequently the optimization of rotor
design scheme the machine performance was
examined and the 3-flux barrier rotor design
was chosen for further refinement and
Surrogate based optimization technique was
applied with four main variables of rotor
design which are Flux barrier width, flux
carrier width, shaft diameter and barrier edge
angle. Total 20 Latin hypercube samples are
generated to check the toraue, efficiency,
torque ripples and magnetic flux density. And
finally, best design is selected for the
prototype development.

Fig 7: Prototype development stages of the proposed rotor (a) early manufacturing from the

factory (b) flux barrier filling with magic and PVC (c) assembled rotor with shaft (d) final

assembled rotor after machining

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

In order to improve the mechanical
integrity of the rotor manufacturing process,
the polyvinylchloride material was inserted
inside of the flux barrier and it took almost 2
days to dry the rotor. The complete rotor
manufacturing stages from manufacturing in
the factory to assembled rotor in the
machining process are shown in Fig. 7. The
stator for the proposed motor used is identical
of standard Cummins BCI-184F machine is
shown in Fig. 8, whereas the complete
specifications are described in [26, 27]. There
are short-pitched 2/3 winding arrangement
with double layer star connection. Each slot
area is 144mm2 for individual layer. The
experimental test rig of the whole was built
and tested as depicted in Fig. 8.

Fig 8: The Standard stator from a Cummins

BCI-184F machine

A-NO-LOAD TEST

An experimental set-up was conducted in
order to measure the rated voltage, current,
power and other parameters at no-load and is
shown is Fig. 9. The no load speed test is
conducted at 13.3⁰C ambient temperature by
coupling the shaft of proposed machine with
55kW permanent magnet synchronous
machine. The simulation results achieved
through the FEM analysis are validated by
experimental measurements on the test bench.
The designed SynRM was tested under no-
load condition, at fifteen different driven
speed and ten different frequencies. The
frequency varied at steps: 5, 10, 15, 20, 25,
30, 33, 35, 38, 40, 42, 45 and 50. Thus the
machine speed has been changing according

to synchronous speed: 150, 299, 452, 600,
752, 900, 995, 1052, 1140, 1200, 1260, 1314,
1350, 1433, 1500 RPM.

Table VII. No-load experimental data of

speed, current, voltage, power, power factor,

harmonics and vibration from 0-50Hz

frequency.

Fig 9: Experimental test rig

A. The SynRM

B. Siemens Drive

C. 55kW load motor

D. ABB Drive

E. Temperature meter

F. Oscilloscope Tektronix TDS 2024B

G. Torque meter JN-338

H. Fluke power meter

I. Mutimeter

J. Speed meter

K. Torque transducer

L. Siemens circuit breaker CDM10-100/3300

Delixi

M. Motor side coupling

N. N. Load side coupling

43.8, 45, 47.8, 50 Hz frequency. Table III
presents the rms value of current and voltage
with power, power factor, total harmonics
distortion and vibration reading at various
speed.

Table IV shows the temperature readings
at for 150 to 1500 rpm driven speed
respectively and Fig. 10. depicts the
temperature graph at 5 to 50Hz frequency. As
it can be noted that the from no-load test
results the number of harmonics and vibration
is decreasing as the machine is getting higher

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

speed mainly due to fact that as the machine’s
performance becomes more stable. The
reduction of harmonics and vibration content

is therefore and important achievement in the
results.

Table III. No-load Test Temperature experimental data of speed, current, voltage, power,

power factor and harmonics from 0-50Hz frequency.

B-LOW-SLIP TEST

In order to verify the analysis in the paper,
the low slip test is performed. In the low slip
test different performance of synchronous
reluctance motor carried out, the speed is
varied from 150 to 1500rpm by Siemens VFD
and the load motor 55kW is rotated anti
clockwise at different speed and torque
settings and all the parameters are noted such
as current, voltage, power, power factor and

harmonics. The test machine is derived by
Siemens-Micromaster-440 drive and the
55kW coupled machine is derived by ABB
ACS800 drive.

The temperature readings along with
current, voltage, power, power factor and
harmonics are shown in table IV and V,
whereas the harmonics reading and parameter
setting of load drive motor is shown in table
VI.

Hz RPM I V kW P.f Harmonics Vibration

m/s2

5 150 17.8 155 0.28 0.08 254 102.1

10 299 18.3 163.5 0.29 0.09 141.8 99.9

15 452 18.2 172.9 0.36 0.10 165% 98

20 600 18 182 0.34 0.09 116% 99.6

25 752 18 190 0.36 0.10 16.7/0.4/5.3 98.5

30 900 18.1 196.2 0.36 0.10 9.2/6.7/0.3 100

33 995 19.7 199 0.38 0.11 13.6/2.4/14.7 97.2

35 1052 18.2 202 0.40 0.11 8.5/7.5/0.4 97.7

38 1140 18.4 208 0.42 0.11 14.5/12.4/7.1 97.9

40 1200 18.4 213 0.45 0.12 14.3/11.2/ 96

42 1260 18.8 217.4 0.52 0.13 8.26% 91.5

43.80 1314 21.7 221 0.53 0.13 8.3/13.7 91

45 1350 18.4 229 0.52 0.13 6.7/4./0.8 92

47.8 1433 18 236 0.55 .13 8.3/13.7 93

50 1500 18.4 249 0.56 0.14 6.4/5.1/0.6 93

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

Fig 10: Temperature graph as different input frequency and speed

Table IV. Low-Slip Test Temperature readings of the test machine from 0-50 frequency and 0

to 1500RPM rated speed

Frequency T1 T2 T3 T4 T5 RPM

5Hz 29.3 30 31.9 32 27.6 150

17Hz 32.3 34.1 37.2 30 31.4 513

25Hz 32.3 34 36.6 31.3 31.8 750

Not Torque applied by ABB at 50Hz

25Hz 32.8 34.5 36.8 33 32.7 750

30Hz 32.8 34.2 36.7 33.2 33.2 899

Now Torque Applied on the machine

30Hz 32.7 34.1 36.7 33.4 33.6 900

35Hz 33 34.5 37.1 33.6 34.1 1050

Now Torque Applied on the machine

35Hz 33.5 34.9 37.5 34.3 34.6 1050

40Hz 34.1 35.2 37.9 34.8 35.7 1200

Now Torque Applied on the machine

40Hz 34.2 35.2 37.9 34.4 35.9 1200

45Hz 35.4 36.2 39.5 36.2 36.8 1356

Now Torque Applied on the machine

45Hz 36.1 37 40.2 37.1 37.8 1350

50Hz 36.6 37.7 41.1 37.6 38.3 1500

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

Table V. Low-slip test

Frequency RPM Current Voltage Power P.f

5Hz 150 18 139 0.22 0.09

17Hz 513 18.2 163 0.36 0.12

25Hz 750 17.8 181 41 0.14

Now Torque applied by ABB at 50Hz

25Hz 750 18 181 0.51 0.16

30Hz 899 18 192 0.54 0.17

Now Torque Applied on the machine

30Hz 900 18.2 191 0.68/0.76 0.21

35Hz 1050 19 312 0.85 0.224

Now Torque Applied on the machine

35Hz 1050 18.8 200 0.94 0.25

40Hz 1200 18.9 211 1.10 0.26

Now Torque Applied on the machine

Same all 1200 18.9 211 1.10 0.27

45Hz 1356 20.7 227 1.17 0.28

Now Torque Applied on the machine

45Hz 1350 20.7 226 1.76 0.37

50Hz 1500 23.8 239 2.29 0.41

Table VI. Harmonics produces by the SynRM and parameter settings of load motor.

Harmonics Torque

ABB drive reference

settings Hz/Rpm
ABB Drive current

512% 0.2 0.1Hz/

44.7% 0.42 0.6/-9.8 2.4

0.5 0.6/-9.8 2.9

Not Torque applied by ABB at 50Hz

14.5/4.9/0.4 0.8/-11.7/-52% 45/6.3

8.3/3.0/0.5 0.9 0.8/-11.7 45/6.3

Now Torque Applied on the machine

8.5/3.3/0.2 1.4 1/68% 68/10.1

7.5/4.8/0.3 1.7 1.1/-77% 67/11.4

Now Torque Applied on the machine

7.4/4.2/0.2 1.7 1.2/-16rpm/-84% 72/15

12/8.2/2 2.3 1.2/-16rpm/-84% 72/15

Now Torque Applied on the machine

12/8.1/0.3 2.3 1.4/-19/99 86/19

7.2/6.3/0.2 2.9 1.4/19.6 84/11.8

Now Torque Applied on the machine

7.2/6.2/0.3 3.4 1.7/-23.8/117.74T 102.84/12

5.9/6.1/0.3 3.9 1.7/-24.7rpm/-113T 99.74/18

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

C-OVERSPEED TEST

In order to verify the analysis in the paper,
the over speed and test is performed. Firstly,
the machine was speedup to normal speed at
1500RM and then gradually speed was

increased while varying the frequency from
51-55Hz. It was observed the machine was
running good with thermally stable. The
experimental results of overspeed test are
shown at Table VII and VIII.

Table VII. Low-slip test experimental data of speed, current, voltage, power, power factor and

harmonics from 0-50Hz frequency.

Speed

Rpm

Frequency Current Voltage Power P.f Harmonics

1530 51Hz 1530 18.9 245 0.44 0.15

1560 52Hz 1560 22.5 248 0.63 0.124

1590 53Hz 1590 17.8 249 0.0.59 0.13

1620 54Hz 1620 18.1 251.9 0.65 0.14

1635 54.5Hz 1635 17.3 250 0.64 0.14

1650 55Hz 1650 17.2 251 0.64 0.15

Table VIII. Over speed Test Temperature readings of the test machine from 0-50 frequency

and 0 to 1500RPM rated speed.

Speed Rpm Frequency T1 T2 T3 T4 T5

1530 51Hz 29.5 29.6 32.1 23.4 31

1560 52Hz 29.8 30 32.4 27.4 31.7

1590 53Hz 30.2 30.2 32.7 29.2 31.9

1620 54Hz 30.3 30.3 33 31.7 32.2

1635 54.5Hz 30.6 30.8 33 31.3 22.6

1650 55Hz 30.9 30.8 33.1 32.3 32.8

6. Analysis of Experiment Results

The main aim of the experimental
validation is to confirm the simulation model
at no-load, low slip and over speed test.
Addition to this, temperature at different
speed of each method is determined. In the
over speed test as per Siemens drive default
settings machine is started at 5Hz frequency
which runs at 150RPM at ambient
temperature 16.2 degree centigrade. There are
5 thermocouple temperature sensors have
been located at different positions in the test
machine.

(a)

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

(b)

(c)

(d)

Fig.11 Experimental results (a) The voltage

wave form; (b) power and power factor

reading; (c) total harmonics distortion; (d)

torque wave form at different input frequency

and speed

The three sensors are used for winding
and their location is 120 mechanical degree
apart and two sensors are located at stator
outer body. The results of over speed test of
all the parameters and temperature parameters
are shown in Table VII and VIII respectively.
The reading T1, T2 and T3 shows the winding
temperature whereas T4 and T5 is the stator
outer body temperature. The speed is
controlled from the input supply frequency
from Siemens drive. It can be observed from
the data reading that the T3 has the highest
reading 33.1⁰C at 15Hz frequency with
1650RPM whereas remaining all the
temperature reading is below the maximum
which reflects that the machine the thermally
stable. The graph of bar chart also depicts the
temperature reading under various frequency
changes as shown in Fig 10. The no-load
experimental data of speed, current, voltage,
power, power factor, harmonics and vibration
are shown in Table. III and their
corresponding photos of 3-phase current
wave form, amperes, voltage and frequency
values with total harmonics distortion and
generated torque wave form are shown in Fig.
11 (a-d). It is observed that initially the
machine has high vibration at low speed but
as the speed is increased the machine is also
stable which is very good in terms of
mechanical point of view at can be seen from
the data at 50Hz frequency and 1500RPM the
vibration is 93 m/s2 which is much better than
at the time of starting.

7. Conclusion

This paper has presented the simulation
and experimental studies on synchronous
reluctance machine with a symmetrical rotor
design for centrifugal pumps. A technique
based on surrogate with particle swarm
optimization has been developed to aid in the
machine. Distinct shapes of the rotor are
analyzed through simulation and modified the
flux barrier shape, shaft diameter and angle of
edges to reduce the torque ripple and losses of
the machine. The experimental results
validate the numerical designs for the
proposed machine. A permanent magnet
material could also be inserted in rotor flux

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

barriers to improve flux density. As this
would be beneficial for high torque machines.
The proposed rotor A symmetrical rotor can
improve effective thermal and mechanical
performance of the machine and can be
beneficial for some applications which are
uni-directional in machine operations, as this
work is targeted.

References

[1] S. Yammine, C. Henaux, M. Fadel, S.
Desharnais, and L. Calégari,
Synchronous reluctance machine flux
barrier design based on the flux line
patterns in a solid rotor," "in Electrical
Machines (ICEM), 2014 International
Conference on, 2014, pp. 297-302.

[2] P. S. Ghahfarokhi, A. Belahcen, A.
Kallaste, T. Vaimann, L. Gerokov, and
A. Rassolkin, "Thermal Analysis of a
SynRM Using a Thermal Network and a
Hybrid Model," in 2018 XIII
International Conference on Electrical
Machines (ICEM), 2018, pp. 2682-2688.

[3] E. Howard, M. J. Kamper, and S. Gerber,
"Asymmetric flux barrier and skew
design optimization of reluctance
synchronous machines," IEEE
Transactions on Industry Applications,
vol. 51, pp. 3751-3760, 2015.

[4] S. Taghavi and P. Pillay, "A sizing
methodology of the synchronous
reluctance motor for traction
applications," IEEE Journal of Emerging
and Selected Topics in Power
Electronics, vol. 2, pp. 329-340, 2014.

[5] A. Boglietti, A. Cavagnino, M. Pastorelli,
and A. Vagati, "Experimental
comparison of induction and
synchronous reluctance motors
performance," in Industry Applications
Conference, 2005. Fourtieth IAS Annual
Meeting. Conference Record of the 2005,
2005, pp. 474-479.

[6] R.-R. Moghaddam, F. Magnussen, and C.
Sadarangani, "Novel rotor design
optimization of synchronous reluctance
machine for low torque ripple," in 2012
XXth International Conference on
Electrical Machines, 2012, pp. 720-724.

[7] N. Bianchi, S. Bolognani, D. Bon, and M.
Dai PrÉ, "Torque harmonic
compensation in a synchronous
reluctance motor," IEEE Transactions on
energy conversion, vol. 23, pp. 466-473,
2008.

[8] M. Sanada, K. Hiramoto, S. Morimoto,
and Y. Takeda, "Torque ripple
improvement for synchronous reluctance
motor using an asymmetric flux barrier

arrangement," IEEE Transactions on
Industry Applications, vol. 40, pp. 1076-
1082, 2004.

[9] M. Gamba, G. Pellegrino, and A. Vagati,
"Design of non conventional
Synchronous Reluctance machine,"
2018.

[10] D. Staton, T. Miller, and S. Wood,
"Maximising the saliency ratio of the
synchronous reluctance motor," in IEE
Proceedings B (Electric Power
Applications), 1993, pp. 249-259.

[11] M. Kamper and A. Volsdhenk, "Effect of
rotor dimensions and cross magnetisation
on Ld and Lq inductances of reluctance
synchronous machine with cageless flux
barrier rotor," IEE Proceedings-Electric
Power Applications, vol. 141, pp. 213-
220, 1994.

[12] J. Ikaheimo, J. Kolehmainen, T.
Kansakangas, V. Kivela, and R. R.
Moghaddam, "Synchronous high-speed
reluctance machine with novel rotor
construction," IEEE Transactions on
Industrial Electronics, vol. 61, pp. 2969-
2975, 2014.

[13] M. E. H. Zaim, "High-speed solid rotor
synchronous reluctance machine design
and optimization," IEEE Transactions on
Magnetics, vol. 45, pp. 1796-1799, 2009.

[14] H. Hofmann and S. R. Sanders, "High-
speed synchronous reluctance machine
with minimized rotor losses," IEEE
Transactions on Industry Applications,
vol. 36, pp. 531-539, 2000.

[15] Y. Gessese and A. Binder, "Axially
slitted, high-speed solid-rotor induction
motor technology with copper end-
rings," in Electrical Machines and
Systems, 2009. ICEMS 2009.
International Conference on, 2009, pp. 1-
6.

[16] I. Boldea, Z. Fu, and S. Nasar,
"Performance evaluation of axially-
laminated anisotropic (ALA) rotor
reluctance synchronous motors," IEEE
transactions on industry applications,
vol. 30, pp. 977-985, 1994.

[17] D. Platt, "Reluctance motor with strong
rotor anisotropy," in Industry
Applications Society Annual Meeting,
1990., Conference Record of the 1990
IEEE, 1990, pp. 224-229.

[18] D. Platt, "Reluctance motor with strong
rotor anisotropy," IEEE transactions on
industry applications, vol. 28, pp. 652-
658, 1992.

[19] M. J. Kamper, F. Van der Merwe, and S.
Williamson, "Direct finite element
design optimisation of the cageless
reluctance synchronous machine," IEEE

Syed Abid Ali Shah (et al.) Design of Low-Cost Synchronous Reluctance Motor with a Surrogate-Assisted Optimization

Technique (pp. 24 – 39)

Transactions on Energy Conversion, vol.
11, pp. 547-555, 1996.

[20] S. Taghavi and P. Pillay, "A
mechanically robust rotor with
transverse-laminations for a synchronous
reluctance machine for traction
applications," in Energy Conversion
Congress and Exposition (ECCE), 2014
IEEE, 2014, pp. 5131-5137.

[21] S. M. Taghavi and P. Pillay, "A
mechanically robust rotor with transverse
laminations for a wide-speed-range
synchronous reluctance traction motor,"
IEEE Transactions on Industry
Applications, vol. 51, pp. 4404-4414,
2015.

[22] J. Kostko, "Polyphase reaction
synchronous motors," Journal of the
American Institute of Electrical
Engineers, vol. 42, pp. 1162-1168, 1923.

[23] A. A. S. Bukhari, S. Ali, S. H. Shaikh, T.
A. Soomro, Z. Lin, and W. Cao, "Design
of a high speed 18/12 switched reluctance
motor drive with an asymmetrical bridge
converter for electric vehicles," in
Computing, Mathematics and
Engineering Technologies (iCoMET),

2018 International Conference on, 2018,
pp. 1-6.

[24] M. A. Raj and A. Kavitha, "Effect of
Rotor Geometry on Peak and Average
Torque of External Rotor Synchronous
Reluctance Motor (Ex-R SynRM) in
comparison with Switched Reluctance
Motor for Low Speed Direct Drive
Domestic Application," IEEE Trans. Ind.
Appl, vol. 53, 2015.

[25] T. Matsuo and T. A. Lipo, "Rotor design
optimization of synchronous reluctance
machine," IEEE Transactions on Energy
Conversion, vol. 9, pp. 359-365, 1994.

[26] N. Yang, W. Cao, Z. Liu, Z. Tan, Y.
Zhang, S. Yu, et al., "Novel
asymmetrical rotor design for easy
assembly and repair of rotor windings in
synchronous generators," in Magnetics
Conference (INTERMAG), 2015 IEEE,
2015, pp. 1-1.

[27] T. Yuan, N. Yang, W. Zhang, W. Cao, N.
Xing, Z. Tan, et al., "Improved
Synchronous Machine Rotor Design for
the Easy Assembly of Excitation Coils
Based on Surrogate Optimization,"
Energies, vol. 11, p. 1311, 2018.