Design of an Interior Permanent Magnet Machine with Concentrated Windings for Field Weakening Applications By Lester Chong A thesis submitted to THE UNIVERSITY OF NEW SOUTH WALES in partial fulfilment of the requirements for the degree of Doctor of Philosophy (Electrical Engineering) August, 2011
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Design of an Interior Permanent Magnet Machine with Concentrated
Windings for Field WeakeningApplications
By
Lester Chong
A thesis submitted to
THE UNIVERSITY OF NEW SOUTH WALES
in partial fulfilment of the
requirements for the degree of
Doctor of Philosophy
(Electrical Engineering)
August, 2011
i
ACKNOWLEDGEMENTS
I am extremely grateful to my two supervisors Professor Faz Rahman and Dr Rukmi Dutta for
all their time, guidance and invaluable advice given to me over the duration of my PhD. I
thankfully acknowledge the inputs from my examiners and Associate Professor John Fletcher. I
would also like to thank the Australian Government and the University of New South Wales for
sponsoring my studies.
I am very grateful to the wonderful staff and at School of Electrical Engineering and
Telecommunications, especially Dr Baburaj Karnayil, Richard Tuck and Gamini Liyadipitiya. I
am also grateful for the advice given by Subash and Seetha from the school of mechanical
engineering. I would like to thank Dr Howard Lovatt and Colin Bilson at CSIRO for their help
and advice.
A special thank you to my family, especially my mother Jenny Ong who has selflessly worked
so hard to raise and support us; my fiancé Janice and the Liao family for their support from the
start to the final stages of my PhD. Last but not least I would like to thank my wonderful friends
and colleagues for making my time in Sydney so special and unforgettable.
ii
ABSTRACT
This thesis presents the design of an interior permanent magnet (IPM) machine with
concentrated windings (CW) for field-weakening applications. The initial phase of this
work involved a feasibility study and comparison with the CW surface permanent
magnet machine. Subsequently a CW-IPM machine was designed and constructed with
the aim of achieving a wide constant power speed range (CPSR). Lastly, based on the
constructed design, scalability and efficiency studies were performed.
The work done in this thesis has led to the successful construction of a prototype
machine achieving a very wide 7.2:1 CPSR. At the time of writing, there is no available
literature that clams such a wide CPSR in a concentrated wound permanent magnet
machine. Distributed winding machines capable of achieving such a CPSR have
complex rotors and hence manufacturing issues. The proposed design was subjected to
the same size constraint as two previously constructed 550W distributed winding IPM
machines. With this constraint, the advantage of shorter end winding length was
exploited and the effective length of the machine was increased. This resulted in a
significant increase in output power to 800W throughout the CPSR. A detailed study on
losses performed in this work showed that despite the increased harmonic content
generated by CW, frequency related losses can be minimized through design methods,
and over an 80% efficiency can be achieved throughout the CPSR. Mechanical stress
analysis of the rotor indicated that iron bridges were required between poles to prevent
excessive stress inflicted on the inter-pole link sections during high speed operation.
Based on confidence gained from the experimental verification of the CW-IPM machine
design, a scalability study was performed and designs up to 30kW were proposed. A
iii
study on efficiency optimization was also carried out and the prototype machine was
redesigned to produce up to 93% efficiency.
The work done in this thesis has setup a strong basis for future work on CW-IPM
machines for automotive traction drives and has also proven that this machine type is
suitable for high performance industrial applications requiring high efficiency over a
very wide CPSR.
iv
TABLE OF CONTENTS
ACKNOWLEDGEMENTS .................................................................................. i
ABSTRACT ............................................................................................................ ii
TABLE OF CONTENTS ...................................................................................... iv
LIST OF FIGURES ............................................................................................... x
LIST OF TABLES ................................................................................................. xix
NOMENCLATURE .............................................................................................. xx
2.2.5 Galerkin’s Method for Deriving Finite Element Equations ...................... 39
2.2.6 Solving Finite Element Equations with Newton Raphson Method ............ 41
2.2.7 Process of a Time Stepping Finite Element Model ...................................... 43
v
2.3 FINITE ELEMENT METHOD FOR DETERMINING MACHINE PARAMETERS AND PERFORMANCE CHARACTERISTICS .................................................................. 45
2.3.1 Construction of Geometry and Assignment of Mesh ................................. 45
2.3.2 Defining Material Properties ..................................................................... 47
2.3.3 Coupling of Electrical Circuits .................................................................. 48
3.4 COMPARING THE IPM AND SPM MACHINES WITH CONCENTRATED WINDINGS ..................................................................................................... 75
3.4.1 Airgap Flux Produced by the Magnets ...................................................... 76
3.4.2 Constant Power Speed Range .................................................................... 79
5.2 MMF HARMONICS AND LOSSES IN MACHINES WITH CONCENTRATED WINDINGS ................................................................................................... 121
5.3 CORE LOSS .................................................................................................. 127
5.3.1 Comparison of Steel Grades ................................................................... 129
5.3.2 Core Loss of the Final Design ................................................................ 131
5.4 MAGNET LOSS ............................................................................................. 132
5.4.1 Comparison SPM and IPM Magnet Losses ............................................ 134
5.4.2 Effects of Magnet Segmentation .............................................................. 136
5.4.3 Magnet Loss of the Final Design ............................................................ 139
5.5 STATOR WINDING LOSS .............................................................................. 140
5.7 FIELD-WEAKENING PERFORMANCE WITH THE INCLUSION OF LOSSES FROM THE FINITE ELEMENT MODEL ............................................................................. 147
APPENDIX A ...................................................................................................... 243
A. AC STANDSTILL TEST APPLIED TO THE FINITE ELEMENT MODEL OF THE SEGMENTED IPM MACHINE .......................................................................... 243
A.1 Results of AC Standstill Test Implemented on the Segmented IPM Machine 243
APPENDIX B ....................................................................................................... 245
B. SALIENCY RATIO OPTIMISATION .................................................................. 245
B.1 Optimisation of Saliency Ratio by Variation of Rotor Magnet Shape ........... 245
ix
APPENDIX C ...................................................................................................... 249
C. INDUCTANCE WAVEFORMS AND SALIENCY RATIO FOR VARIOUS SLOT/POLE COMBINATION AND FOR DOUBLE-LAYER WINDINGS .................................. 249
C.1 Inductance Waveform and Saliency Ratio – Comparison of Various Slot/pole Combinations ............................................................................................. 249
C.2 Inductance Waveform and Saliency Ratio – Comparison with Double-Layer Windings .................................................................................................... 251
APPENDIX D ...................................................................................................... 253
D. THERMAL MODEL ......................................................................................... 253
D.1 Thermal Model Approximating Temperature at Various Parts of the Machine ...................................................................................................... 253
APPENDIX E ....................................................................................................... 255
E. FINAL MACHINE DRAWINGS ........................................................................ 255
E.1 ABB Casing used (with Original Induction Motor) ..................................... 255E.2 Stator of the CW-IPM Prototype ................................................................. 256E.3 Rotor of the CW-IPM Prototype .................................................................. 257E.4 Shaft of the CW-IPM Prototype ................................................................... 259E.5 Key (shaft) of the CW-IPM Prototype ......................................................... 260E.6 End-plates of the CW-IPM Prototype .......................................................... 261
APPENDIX F ....................................................................................................... 262
F. EXPERIMENTAL SETUP ................................................................................. 262
APPENDIX G ...................................................................................................... 271
G. PUBLICATION LIST ....................................................................................... 271
x
LIST OF FIGURES
Fig. 1.1 Classification of AC machine types used for traction applications 2
Fig. 1.2 Various IPM motor geometries 4
Fig. 1.3 Various stator winding layouts 6
Fig. 1.4 Typical PMSM drive system block diagram 7
Fig. 1.5 Ideal field-weakening characteristics of a drive system 8
Fig. 2.1 Various methods to solve Maxwell equations and predict machine performance
28
Fig. 2.2 Typical finite elements 35
Fig. 2.3 Two dimensional triangular element 36
Fig. 2.4 Dirichelet boundaries 38
Fig. 2.5 Machine with quarter cyclic-symmetry 39
Fig. 2.6 Newton Raphson method 43
Fig. 2.7 Flow chart for the time-stepping finite element process 44
Fig. 2.8 Prototype machine geometry – showing regions 45
Fig. 2.9 DW-IPM showing the repletion of phase coils every quarter 46
Fig. 2.10 Mesh structure of the CW-IPM machine 47
Fig. 2.11 B-H Curve for hard and soft magnetic materials 48
Fig. 2.12 Representation of core material characteristics in terms of an analytic solution
49
Fig.2.13 Representation of permanent magnet material demagnetisation characteristics
49
Fig. 2.14 Three-phase star conneted circuit with current excitation 50
Fig. 2.15 Meshing of a three-layer airgap 52
Fig. 3.1 14-pole IPM machine with various slot and pole combinations 56
Fig. 3.2 EMF phasor diagram 59
xi
Fig. 3.3 3-phase EMF waveforms and corresponding frequency spectrum over 1 electrical cycle for 14-pole IPM machinewith different slot and pole combinations
61
Fig. 3.4 Cogging torque waveforms for various slot and pole combination
63
Fig. 3.5 Flux waveform and corresponding frequency spectrum of a 14-pole, double-layer, DW machine model
66
Fig. 3.6 Flux waveform due to armature reaction for single- and double-layer CW
67
Fig. 3.7 14-pole DW IPM with flux being channeled to the d- and q-axis
68
Fig. 3.8 14-pole, 18-slot, CW-IPM flux plot showing no obvious d or q-axis flux paths
69
Fig. 3.9 Inductance waveform measured from UNSW Segmented IPM machine
71
Fig. 3.10 Inductance waveform of an 18-slot, 14-pole CW-IPM machine
71
Fig. 3.11 d- and q-axis inductance comparison 72
Fig. 3.12 Estimated reduction in end winding length 73
Fig. 3.13 Advanced winding methods to achieve a high saliency ratio 74
Fig. 3.14 Performance comparison of different magnet shapes 75
Fig. 3.15 Flux density plot of the v-shaped IPM model showing saturation regions
77
Fig. 3.16 Airgap flux produced by the different rotor configurations 78
Fig. 3.17 Peak power and CPSR comparison between three rotor types 79
Fig. 4.1 Rotor types used for increasing saliency ratio 83
Fig. 4.2 Contour plot showing the variation of magnet remanent fluxdensity versus rated current
Fig. 4.6 Resistance and copper loss per phase as temperature increases 95
Fig. 4.7 ABB casing with inserted 18-slot stator core 96
Fig. 4.8 End winding length comparison between UNSW DW stator with windings done on a plastic model
97
Fig. 4.9 Hand winding methods 98
Fig. 4.10 Airgap length variation with output power and CPSR 99
Fig. 4.11 Key parameters defining the stator geometry 100
Fig. 4.12 Flux density plots showing peak flux density in the yoke and tooth
102
Fig. 4.13 Plastic stator made with three different slot opening widths 103
Fig. 4.14 IPM rotors showing inter-pole link sections 104
Fig. 4.15 V-angle variation 105
Fig. 4.16 Normalised power versus speed characteristics with variation of v-angle
105
Fig. 4.17 Various types of SPM rotors 106
Fig. 4.18 IPM rotor with buried single-piece/pole magnets 107
Fig. 4.19 Modelled solenoid-magnet model 108
Fig. 4.20 Sections for centrifugal force calculation 109
Fig. 4.21 Model showing outward normal pressure on each section of the rotor
109
Fig. 4.22 Mechanical stress analysis of rotor 110
Fig. 4.23 Final 18-slot, double-layer winding layout 112
Fig. 4.24 Three-phase induced line to neutral back EMF voltage from the FE model (at 50Hz)
114
Fig. 4.25 Induced line to line voltage versus speed 114
Fig. 4.26 Comparison of EMF waveform between the CW-IPM and DW-IPM
115
Fig. 4.27 Comparison of EMF waveform between the CW-IPM and DW-IPM – frequency spectrum
115
xiii
Fig. 4.28 Cogging torque of final design comapred to an equivaent integral-slot DW model
116
Fig. 4.29 Self- and mutual-inductance waveform of final model with 3Arms current excitation
117
Fig. 4.30 Torque ripple of final model at base speed 118
Fig. 4.31 CPSR of final design with base frequency of 50Hz 118
Fig. 5.1 Comparison of flux distribution in a CW and DW machine 121
Fig. 5.2 Ref. to fig. 3.5 122
Fig. 5.3 Ref. to fig. 3.6 122
Fig. 5.4 Contribution of eddy current and hysteresis loss 124
Fig. 5.5 Ref. to fig. 4.5 128
Fig. 5.6 Annular steel model to determine eddy current and hysteresis loss constants
128
Fig. 5.7 Core loss comparison at 50 and 500Hz with different steel grades
130
Fig. 5.8 Core loss with chosen steel grade at various frequencies 131
Fig. 5.9 Extrapolated values of measured hysteresis and eddy current loss for sintered neodymium magnets at 50Hz
133
Fig. 5.10 3D model of a single-pole and single-phase excitation – v-shaped IPM
134
Fig. 5.11 3D model of a single-pole and single-phase excitation – SPM 134
Fig. 5.13 Comparison of magnet losses between IPM and SPM machine 135
Fig. 5.14 Variation of magnet eddy current loss with number of magnet segments due to slot and carrier harmonics – Comparison between IPM, inset and SPM rotors
137
Fig. 5.15 Circulating eddy currents in a non-segmented a) SPM magnet pole, b) V-shaped IPM magnet pole
137
Fig. 5.16 Circulating eddy currents in a segmented magnet poles 138
Fig. 5.17 Magnet losses in an SPM machine with variation of segment number
138
xiv
Fig. 5.18 Magnet losses in an IPM machine with variation of segment number
139
Fig. 5.19 Magnet losses versus frequency in the final design withsintered NdFeB magnets
3-phase CW machines with double-layer windings require slots in multiples of 6 in
order for the phase coils to be equally distributed. Thus 6, 12 and 18 slots are chosen
and the same 14-pole rotor will be used.
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
61
Here the back EMF from the four designs shown earlier in fig. 3.1 is compared.
(a) 2Spp DW
(b) 1/7Spp CW
(c) 2/7Spp CW
(d) 3/7Spp CW
Fig. 3.3 3-phase EMF waveforms and corresponding frequency spectrum over 1 electrical cycle for 14-pole IPM machine with different slot and pole combinations
0
10
20
30
0 5 10 15
Volt
0
10
20
30
0 5 10 15
Volt
0
10
20
30
0 5 10 15
Volt
0
10
20
30
0 5 10 15
Volt
28.4V
21V
27.6V
14.5V
( )Time (s) Frequency components (n-order)
Time (s) Frequency components (n-order)
Time (s)0 5 10 15Frequency components (n-order)
Time (s)0 5 10 15
Frequency components (n-order)
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
62
By comparing the back EMF waveforms from fig. 3.3, it can be seen that all, except the
6-slot model, produced near-perfectly sinusoidal (line to neutral) back EMF waveforms.
The 84-slot, DW-IPM machine, (with a calculated winding factor of 0.933), generated
an EMF waveform with the highest magnitude of 28.4Vrms. From the values shown in
table 3.1, the 12-slot, 14-pole model also has a winding factor of 0.933. Therefore it
should produce the same back EMF magnitude. However, due to the rotor magnet pole
pitch being smaller than the stator pole pitch, the flux from the opposing magnet pole
cancels the total amount of flux that is linked to a phase coil, resulting in a lower than
expected winding factor. The same condition was seen in the 6-slot, 14-pole model.
This indicates that having a larger number of slots than poles are required.
Based on back EMF obtained by the DW-IPM machine, the 18-slot design, which has a
winding factor of 0.902 (from table 3.1), the calculated back EMF should by 27.5Vrms.
From the 18-slot FE model, it is shown that the back EMF magnitude achieved
(27.6Vrms) agrees with the calculated value.
3.2.2 Cogging Torque
Cogging torque is a result of the variation of stator inductances due to the interaction of
rotor and stator poles. It contributes to vibrations and noise, leading to severe
restrictions in the machine performance. The cogging torque equation was stated in
earlier in (3.3).
There are several methods by which cogging torque can be reduced, some of which are
rotor flux barrier shaping [122], stator chamfer angle shaping [125] and shaping of the
rotor outer radius [61]. These methods, are however, heavily based on trial and error
techniques to achieve the optimal geometry. Another common method to reduce
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
63
cogging torque in machines with integral slot windings is by skewing. However, in
fractional-slot windings skewing would lead to a significant decrease in induced EMF,
as well as increase the complexity and cost of constructing the machine. In actuality, the
use of fractional-slot windings itself aids in the reduction of cogging torque as it
eliminates periodicity between slots and poles.
To quantify the amount of reduction with different slot and pole combinations, the
lowest common multiple (LCM) method is used, which is basically the LCM of the
number of slots and poles. A higher LCM would yield a lower peak value in the
cogging torque waveform but result in higher frequency fluctuation. The LCM of
various combinations is stated in table 3.1. Fig. 3.4. gives an example of cogging torque
waveforms for three different Spp values. The 1, 2/7 and 3/7 Spp combinations have
LCMs of 42, 84 and 126 respectively. Results verify the reduction in cogging torque
magnitude as the LCM increases.
Fig. 3.4 Cogging torque waveforms for various slot and pole combination
From the FE results shown in this section, it can be seen that the choice of slot and pole
configurations directly affect both the winding factor and the cogging torque
magnitudes. In the abovementioned comparison, the 18-slot, 14-pole model seems to be
-1.3%
-0.8%
-0.3%
0.3%
0.8%
1.3%
0 0.2 0.4 0.6 0.8 1
Torq
ue (%
Rat
ed)
Normalized Time (P.U.)
42-slots, 14-polesmachine (1 Spp)
12-slots, 14-polesmachine (2/7 Spp)
18-slots, 14-polesmechine (3/7 Spp)
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
64
the most appropriate choice as it produces a high winding factor (0.902) and very low
cogging torque magnitude (as it has a high LCM of 126). Thus this combination will be
used as the basis for the prototype design in this thesis.
Other suitable candidates for this study would be the 12-slot, 10-pole model as well as
the 18-slot, 16-pole model. The 18-slot, 14-pole model was chosen over the 12-slot, 10-
pole model as a lower cogging torque is desired. (The LCM produced by the 12-slot,
10-pole model – the LCM was 60 as opposed to the 126 in the chosen configuration).
The 18-slot, 14-pole model was chosen over the 18-slot, 16-pole model because a
higher base speed was required. (An approximate base speed with 16-poles would be
375RPM as opposed to 429RPM in the 14-pole model).
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
65
3.3 PERFORMANCE IN COMPARISON TO DISTRIBUTED WINDING IPMMACHINE
The growing popularity of CW is due to factors such as:
Shorter end-windings due to non-overlapping coils.
Less amount of copper used.
Simplified and possible automation of manufacturing process [105].
Higher slot-fill factor [104, 105].
Lower stator copper loss.
Higher tolerance to phase faults [20].
Additional leakage inductance increases CPSR in SPM machines [130].
There are, however, also advantageous factors which still make DW a preferred choice
over CW. Some of these are:
DW generates less MMF harmonics leading to lower core and magnet losses.
Higher saliency ratio leading to higher reluctance torque.
Unity winding factor can be achieved resulting in larger electro-dynamic torque.
High performance control techniques including sensorless control are readily
available.
In this section, the key differences between CW and DW will be discussed.
3.3.1 Airgap Flux Harmonics
The popularity of DW is due to its ability to produce MMF waveforms which are close
to sinusoidal. On the contrary, CW produces MMF waveforms which are rich in
harmonics. The MMF produced by stator coils can be expressed follows:
= (3.7)
where,
= Flux density from the stator poles
= Airgap surface area
= Airgap reluctance
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
66
If the airgap length and overall slot opening widths are kept constant, the MMF is
directly proportional to the airgap flux density produced by the stator current. The
airgap flux density of a typical 14-pole, double-layer, DW machine with magnets
removed is shown in fig. 3.5:
Fig. 3.5 Flux waveform and corresponding frequency spectrum of a 14-pole, double-layer, DW machine model
The only term that contributes to electro-dynamic torque production is the fundamental
component. The other terms contribute to increased core and magnet eddy current losses,
as well as additional leakage inductance. A detailed mathematical explanation of how
the additional harmonic components lead to increased frequency-related losses will be
shown in chapter 5.
Due to CW being non-sinusoidal in nature, it is expected that significant harmonic
components will be present in the airgap flux waveform. Airgap flux waveforms in
single- and double-layer, 18-slot, 14-pole CW models with magnets removed are shown
in fig. 3.6a and 3.6b respectively:
-0.25
0
0.25
0 100 200
Tesla
0
0.1
0.2
0.3
0 25 50
Tesla
Fundamental component (torque-producing term)
00 00Airgap circumference (mm)
25Frequency components (n-order)
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
67
(a) 14-pole, 18-slot, single-layer CW machine model
(b) 14-pole, 18-slot, double-layer CW machine model
Fig. 3.6 Flux waveform due to armature reaction for single- and double-layer CW
Comparatively, the double-layer model produces much lower MMF harmonic
components than the single-layer model. Due to the nature of the machine designed for
this work, (where operating frequency can exceed several times the base frequency –
making the machine more susceptible to frequency related losses), the double-layer CW
was selected as the base layout in this work.
-1
-0.5
0
0.5
1
0 100 200
Tesla
0
0.1
0.2
0.3
0 25 50
Tesla
-0.5
-0.25
0
0.25
0.5
0 100 200
Tesla
0
0.1
0.2
0.3
0 25 50
Tesla
Fundamental component (torque-producingterm)
Fundamental component (torque-producing term)
00 00Airgap circumference (mm) Frequency components (n-order)
Airgap circumference (mm) Frequency components (n-order)
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
68
3.3.2 Saliency Ratio and Constant Power Capability
The two main factors contributing to the field-weakening performance of IPM machines
are saliency ratio and characteristic current. Saliency ratio ( ) given by (3.8),
contributes to the additional reluctance torque as shown in (3.1), which is additive to the
total IPM machine torque.
= (3.8)
where,
= d-axis inductance
= q-axis inductance
In an IPM machine < , resulting in > 1.
Measuring the saliency in integral-slot windings is less time consuming as the rotor can
be positioned according to the d- and q-axis flux paths (fig. 3.7) and the d- and q-axis
inductances can be measured accordingly. Fig. 3.7a shows the flux path across the pole
axis, while 3.7b shows the flux path over the inter-pole axis.
(a) Flux being channeled to the d-axis (pole axis)
(b) Flux being channeled to the q-axis (inter-pole axis)
Fig. 3.7 14-pole DW IPM with flux being channeled to the d- and q-axis
In fractional-slot CW, the d- and q-axis flux paths are not obvious due to the
aperiodicity between slots and poles as shown in fig. 3.8.
q-axis
d-axis
q-axis
d-axis
Chapter 3: Investigation of the Concentrated Winding IPM Machine for Wide Field-weakening Applications
69
Fig. 3.8 14-pole, 18-slot, CW-IPM flux plot showing no obvious d or q-axis flux paths
To measure d- and q-axis inductances in CW-IPM machines, the AC standstill test is
deemed most suitable [184]. The AC standstill test was implemented in FE analysis and
its accuracy was verified against the UNSW 4-pole segmented DW-IPM machine
(shown in Appendix A). With the FE results agreeing with the range of measured values,
this method was then used to measure the saliency ratio of the CW-IPM machine. To
account for saturation, measurements were taken at different current values. It should
also be noted that the effect of cross coupling was not studied as it would not
significantly affect the results.
With this method, inductances in the d and q-axes are determined from the self-
Stator and Rotor lamination thickness 0.35mmSlot-opening width 1.2mm
Number of slots 18slots / double-layerWinding type Double-layer concentratedPacking (slot-fill) factor 41.5%Number of turns per coil (around each tooth) 115turnsDiameter of each conductor strandConductor insulating material max. temp.Slot insulating material max. temp.Cooling arrangements Natural convection (Fin cooled)Area per half slot 86.3mm2
Number of poles 14polesNo. of magnet pieces 28pieces (2 X 14poles)Magnet dimensions 13.5mm X 2mm X 79mm Desired remnant flux density (Grade) 1.04 – 1.09T (Raremag - N24EH)Magnet temperature rating 200°CMagnet coating Ni-Cu-NiCore material Non-oriented silicon steelSaturation mag. of laminations 1.68T @5000A/mPredicted core loss at 50Hz/1.5T 2.60W/KgYield strength of laminations 2
Resistivity of lamination -cm
Rated voltage (RMS) 240V/phaseRated current (RMS) 2.55A/phaseMaximum rated torque at base speed 18NmMaximum operating speed 6857rpm d-axis inductance (at rated current) 81.16mHq-axis inductance (at rated current) 84.45mHStator resistance at ambient temperatureMagnet flux linkage 0.48Wb
Chapter 4: Design of an IPM Machine with Concentrated Windings for Field Weakening Applications
114
4.7.1 Back EMF from the Finite Element Model
The predicted back EMF from the final FE model is shown below:
Fig. 4.24 Three-phase induced line to neutral back EMF voltage from the FE model (at 50Hz)
It can be observed that the modelled phase back EMF is near-perfectly sinusoidal, all
three phases are balanced and 120º apart. The induced voltage per phase when the
machine is ran at 428.6RPM (50Hz) is 91.6Vrms. The modelled line to line induced
voltage at the same speed is sinusoidal, and has a value of 158.6Vrms. The linear
relationship between induced line to line voltages versus rotor speed is shown in fig.
4.25:
Fig. 4.25 Induced line to line voltage versus speed
-150
-100
-50
0
50
100
150
0 0.005 0.01 0.015 0.02
Va(L-N)
Vb(L-N)
Vc(L-N)
0200400600800
1000120014001600
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Indu
ced
Volta
ge (V
)
Time (s)
Indu
ced
Vol
tage
(Vrm
s)
Speed (RPM)
Chapter 4: Design of an IPM Machine with Concentrated Windings for Field Weakening Applications
115
To determine the winding factor, the final design was compared to an equivalent
integral-slot, double-layer, DW machine with the same rotor. This comparison is shown
in fig. 4.26 below. The corresponding harmonic spectrums of these waveforms are
shown in fig. 4.27.
Fig. 4.26 Comparison of EMF waveform between the CW-IPM and DW-IPM
Fig. 4.27 Comparison of EMF waveform between the CW-IPM and DW-IPM – frequency spectrum
It can be seen that both these winding types produce equally sinusoidal back EMF
waveforms. However, due to a higher winding factor, the DW-IPM machine produced a
higher fundamental term, (233.2Vrms), compared to the CW-IPM machine,
-250-200-150-100
-500
50100150200250
0 0.005 0.01 0.015 0.02
CW-IPM DW-IPM
0
50
100
150
200
250
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
CW-IPM DW-IPM
4 5 6 11 12 13
Indu
ced
Vol
tage
(V)
Time (s)
Indu
ced
Volta
ge (V
)
Harmonic number
224.3Vrms 233.2Vrms
Near-zero harmonics
Chapter 4: Design of an IPM Machine with Concentrated Windings for Field Weakening Applications
116
(224.3Vrms). The DW-IPM machine has a distribution factor of 0.9659, and a pitch
factor of 0.9695; this results in a winding factor of 0.933. Based on these values,
fundamental magnitude achieved for a unity winding factor should be 249.9Vrms and
the expected magnitude of the 18-slot, 14-pole model should be (249.9Vrms x 0.902 =
225.45Vrms). With a value of 224.3Vrms achieved, the result agrees with the tabulated
values in table 3.1.
4.7.2 Cogging Torque from the Finite Element Model
The cogging torque magnitude of the final design, (3/7Spp), compared to an integral-
slot (2Spp) DW model, is shown in fig. 4.28. It is shown that the final design produces
2.5 times lower peak to peak cogging torque magnitude compared to the DW model.
This is due to the elimination of periodicity of slots and poles, which lowers the
magnitude of the cogging torque but increases its fluctuating frequency.
Fig. 4.28 Cogging torque of final design comapred to an equivaent integral-slot DW model
4.7.3 Inductance and Saliency Ratio from the Finite Element Model
Since, the calculation of inductances by aligning the winding axes to the d- and q-axis
of the rotor is not accurate in CW machines, d- and q-axis inductances are calculated by
applying AC standstill test conditions to the FE model. The self- and mutual inductance
Chapter 4: Design of an IPM Machine with Concentrated Windings for Field Weakening Applications
117
waveforms of the final design from the FE model are shown in Fig. 4.29 (for single-
phase, 50Hz, 3Arms current excitation).
Fig. 4.29 Self- and mutual-inductance waveform of final model with 3Arms current excitationLd, Lq and saliency ratio obtained from the DC and second harmonic terms of these
waveforms are as follows.
= ( ) 2 += 81.16mH
= ( ) + 2 += 84.45mH= = 1.04
Chapter 4: Design of an IPM Machine with Concentrated Windings for Field Weakening Applications
118
4.7.4 Torque Performance from the Finite Element Model
In order to account for small fluctuations, steps/electrical cycle had to be small (200
steps per cycle). The characteristic current of the machine was found to be 2.55Arms (at
which the widest CPSR could be achieved). With the peak voltage set to 240V(1-l), the
rated torque at a base speed of 428.6rpm is 15.14Nm (equivalent to 679W of power).
The torque ripple at this speed is 2% of the torque produced as shown in fig. 4.30.
Fig. 4.30 Torque ripple of final model at base speed
With the abovementioned rated values, a > 10:1 CPSR can be achieved. The peak
power achieved was 880W (at 300Hz).
Fig. 4.31 CPSR of final design with base frequency of 50Hz
The power versus speed characteristic shown in fig. 4.31 is for the ideal case where
losses are ignored. Due to frequency-related losses, the output power and CPSR may be
affected. The performance with the inclusion of losses will be discussed in chapter 5.
14.714.814.9
1515.115.215.315.415.5
0 0.01 0.02 0.03 0.04
0
200
400
600
800
1000
0 500 1000 1500 2000 2500 3000 3500 4000 4500
Pow
er (W
)
Speed (RPM)
Torq
ue (N
m)
Time (s)
>10:1 CPSR
0.31Nm(p-p)
Chapter 4: Design of an IPM Machine with Concentrated Windings for Field Weakening Applications
119
4.8 CONCLUSION
This chapter showed the optimisation process of the 14-pole, 18-slot, double-layer CW-
IPM machine to achieve a wide CPSR and desired torque performance. The key steps in
this design were:
Positioning of phase windings to achieve highest winding factor
Material considerations
Rotor magnet design and structural considerations
Optimising the CPSR with by satisfying characteristic current equilibrium
conditions
Further extend the speed range by airgap length and rotor geometry variations
FE analysis showed that the final design can achieve over 10:1 (lossless) CPSR with the
capability of producing over 880W of peak power for a voltage limit of 240V(l-l). By
comparing FE models of the final CW-IPM machine with the integral-slot DW-IPM
machine, it was shown that the expected winding factor (0.902) and significantly lower
cogging torque can be achieved. It was also shown that high inductance values were
achieved with Ld being similar to Lq – resulting in a saliency ratio is almost unity (1.04).
In this chapter, the design process only considered the ideal scenario where both
electrical and mechanical losses were omitted. Losses will be considered separately in
the chapter 5. The test results from the constructed design will be shown in chapter 7,
where the FE results will be verified.
120
CHAPTER 5ANALYSIS OF LOSSES IN AN IPM MACHINE WITH CONCENTRATED WINDINGS FOR FIELD WEAKENING APPLICATIONS
5.1 INTRODUCTION
A key aim in almost all high performance PM machine design is to minimise losses. In
machines used for field weakening applications, frequency related losses in particular
have to be carefully considered.
Losses in PM machines are separated into two main areas – electrical losses and
mechanical losses.
Electrical Losses:
Core losses – Eddy current and hysteresis losses
Permanent magnet losses
Copper loss
Mechanical losses:
Bearing losses
Windage losses
Chapter 5 will highlight the causes of increased frequency-related losses in an IPM
machine with CW in comparison to DW. The effects of varying material type and
geometry will be investigated. The abovementioned electrical and mechanical losses
will be studied and quantified in terms of the final FE design presented in chapter 4.
Losses would be found at various frequencies and used to provide a more realistic field
weakening performance of the machine.
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
121
5.2 MMF HARMONICS AND LOSSES IN MACHINES WITH
CONCENTRATED WINDINGS
Compared to DW, the MMF waveform produced by CW contains additional MMF
harmonics and sub-harmonics not rotating in sync with the synchronous frequency. Fig.
5.1 show the magnetic flux distribution in a CW and DW machine. (The flux
distribution shown is by armature excitation only).
(a) Flux distribution of 14-IPM rotor with a fractional-slot CW stator
(b) Flux distribution of 14-IPM rotor with an integral-slot DW stator
Fig. 5.1 Comparison of flux distribution in a CW and DW machine
MMF ( ) produced by the stator coils is basically expressed as flux density produced by
the stator (Bs) linked across an airgap of surface area (Arot) with reluctance ( ).= (5.1)
If the machine geometry, including airgap length and slot opening width, is kept
constant, the only parameter that will affect MMF across the airgap is the flux density
produced by the armature reaction. The airgap flux waveforms generated by the DW
model (fig. 5.2), has a fundamental component at 7 times the base frequency, which is
the torque producing term in this 14-pole machine model. On the other hand, CW
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
122
produces a dominant fundamental in addition to several other significant harmonic and
sub-harmonic components. Fig. 5.3, gives a comparison of MMF waveforms produced
by single-layer and double-layer CW.
Fig. 5.2 Airgap flux waveform and harmonic spectrum produced from a DW model
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
123
The total loss in the core depends on two main terms hysteresis loss and eddy current
loss as shown in (5.2). = + (5.2)
where, = Eddy current loss constant = Hysteresis loss constant = Maximum flux density = Material dependent Steinmetz loss constant (in the range of 1.6 to 2.0) While hysteresis loss depend largely on material and flux density, eddy current loss is
proportional to the frequency squared, making it much more susceptible to leakage
harmonic components which are several times higher than the operating frequency
[208]. In the field-weakening region, an additional CPSR factor is further multiplied to
the fundamental and leakage components. For example: in operation at 10 times the
base frequency of 50Hz, a leakage harmonic component n=42 would be fluctuating at
3000Hz.
To better visualise the proportion of losses, it is common to divide the total loss by
frequency to keep hysteresis term constant; with the assigned as 2, (5.2) can be
simplified [192, 197, 209].
= + (5.3)
A graphical representation of this equation is shown in fig. 5.4.
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
124
Fig. 5.4 Contribution of eddy current and hysteresis loss
As the hysteresis loss is largely material dependent and little can be done to minimise it,
emphasis will be placed on eddy current loss, in which the choice of lamination
thickness can help in minimising losses.
For a general concept on how the increase in airgap harmonics affects eddy current loss,
Faraday and Maxwell equations can be used.
Equating the electric field ( ) in a closed path c along the surface to the induced voltage( ) and flux linkage we get:
= = = (5.4)
Equating current density ( ) to the electric field, we have: = (5.5)
Say a fluctuating magnetic flux density (B) acts on a thin piece of material with
conductivity thickness width (w). Eddy current (Peddy) induced in this
conductor can be expressed as follows:
esis lossFrequency (Hz)
t and hf1
Contribution of eddy current loss
Contribution of hysteresis loss
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
125
= 1 (5.6)
Assuming that flux density is uniform across the surface, flux linkage will be:
= (5.7)
The induced voltage in (5.4) can be expressed as:
= = (5.8)
If is reduced to infinitely thin sections (x), the electric field along the closed path in
(5.4) can be described by:
= 2 = (5.9)
From (5.5) the induced current density at the surface of the material is given by:
= = (5.10)
The eddy current loss shown in (5.6) in terms of the induced current density can be
described as follows:
= 1 1 (5.11)
= 12 (5.12)
For integral-slot DW, a sinusoidal airgap flux density (fig. 5.2) is assumed, then (5.12)
can be solved to obtain:
= 12 (5.13)
= 6 (1 + 2 ) (5.14)
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
126
where the terms in the square brackets are usually absorbed in the eddy current constant
(Ke).
In CW, the function would not only depend on the fundamental but other leakage
harmonic components as well. The eddy current loss for CW can thus be generally
expressed in (5.15) where Bn is the nth harmonic component of the waveform as shown
fig. 5.3.
= 12 1 ( ) (5.15)
From (5.15) it can be seen that the most effective way to reduce eddy current loss is by
minimising flux density harmonics. Although the introduction of additional harmonic
components is unavoidable with CW, double-layer CW windings substantially reduce
these components. For this particular reason, double-layer was chosen over single-layer
CW windings in the final design.
In the PM machine, eddy current loss occurs mainly in the core and magnets. As
mentioned in the chapter 4, non-oriented silicon steel is chosen as the core material due
to its price and performance characteristics. Within the range of silicon steel grades,
conductivity may vary up to 35% [198, 210], resulting in different core losses. The
thickness of the material also affects core loss. This chapter will study the effects of
core loss as the steel grade/thickness is varied.
For the magnet material, there is a huge variation of conductivity between the two
different methods by which NdFeB magnets are made. Sintered magnets have a typical
conductivity of 625X103 -m)-1 whereas bonded magnets have a conductivity of
7.14X103 -m)-1. Thus, in areas such as in an SPM machine used for FW applications
where magnet losses are high, bonded magnets are commonly used at the expense of
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
127
lower magnet remanent flux densities. In cases where high torque density and low
magnet losses are required, sintered-segmented magnets can be used [140, 211].
The comparison of SPM and IPM magnet losses, as well as the effects of magnet
segmentation and variation of magnet type will be quantified this chapter.
5.3 CORE LOSS
In machines used for field-weakening applications, frequency-related losses have to be
carefully considered and minimised due to constant operation at high frequencies.
Furthermore the increase in leakage harmonic terms as a result of applying CW makes
the machine more susceptible to increased core and rotor losses.
One method of separating the hysteresis and eddy current losses is derived from the
relationship mentioned by Yeadon [212]; the book states that in typical steel
laminations (grades M19 through to M45), hysteresis loss at 60Hz make up
approximately 67% of the total core loss and eddy current loss makes up the other 33%.
The material chosen (35RM300) has core loss 3.25W/kg at 60Hz/1.5T (shown in fig
5.5), which is similar to grade M19. Thus, this approximation is valid for the breakdown
of hysteresis and eddy current loss at 60Hz and the total core loss can be expressed as:= + (5.16)
where, = 13 (5.17)
= 23 (5.18)
This estimation gives us an eddy current loss of 1.08W/kg and hysteresis loss of
2.17W/kg. In order to determine core losses due to harmonics components, various
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
128
operating frequencies as well as saturation levels, the hysteresis ( ) and eddy current
loss constants ( ) must first be found.
Fig. 5.5 Core loss curve of 35RM300 provided by Sankey showing core loss at 60Hz/1.5T
To ensure consistency between FE models in Magsoft-Flux2D, a narrow annular-core
specimen consisting of thin laminations is created (as shown in fig. 5.6):
Fig. 5.6 Annular steel model to determine eddy current and hysteresis loss constants
The surface area of the core was chosen such that the flux distribution throughout the
core is at a relatively constant flux density – in this case 1.5T. Frequency of excitation
Constant flux density throughout surface area (3.3% difference margin)
Is (60Hz)
Annular laminated-core structure
Coils with number of turns producing 1.5T flux density in core
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
129
was set to be 60Hz. The volume of the core was chosen to be equivalent to 100g of
silicon steel.
From this model, the values of and was found by trial and error, to achieve eddy
current loss of 1.08W/kg and hysteresis loss of 2.17W/kg. The values were 0.29 and
151 respectively. These loss constants can then be used to obtain the core loss in the
machine model at various excitation frequencies. ( and is found in this way for
each steel grade).
5.3.1 Comparison of Steel Grades
Here the chosen steel grade (35RM300) will be compared with two other steel grades
with different thicknesses- 50JN350 and 65JN800 [198, 213]. Key features of the
specific grades are shown in table 5.1.
Table 5.1Properties of compared core grades
Material Type
Lamination Thickness
Saturation Mag. (B50)
Stacking Factor
Total Core Loss @60Hz/1.5T
35RM300 0.35mm 1.68T 95% 3.25W/kg 151 0.29
50JNE350 0.50mm 1.68T 96% 4.45W/kg 263 0.39
65JNE800 0.65mm 1.72T 97% 10.15W/kg 495 0.89
With and found using the annular model, the losses at 50 and 500Hz are shown in
fig. 5.7. Stacking factor in the above table was stated as a reference (being proportional
to output torque) – it should be noted that in the loss calculations, stacking factor was
not included as differences would not be very significant.
Fig. 5.13a Magnet losses in an IPM machine with sintered magnets
Fig. 5.13b Magnet losses in an SPM machine with sintered magnets
Fig. 5.13c Magnet losses in an SPM machine with bonded magnets
0.02 0.03 0.05
0.56
0.34
0.90
0
0.2
0.4
0.6
0.8
1
Eddy currentloss @50Hz
Hysteresisloss @50Hz
Total magnetloss @50Hz
Eddy currentloss @500Hz
Hysteresisloss @500Hz
Total magnetloss @500Hz
Pow
er lo
ss (W
)
2.1 4.1 6.3
67.3
37.9
105.2
0
20
40
60
80
100
120
Eddy currentloss @50Hz
Hysteresisloss @50Hz
Total magnetloss @50Hz
Eddy currentloss @500Hz
Hysteresisloss @500Hz
Total magnetloss @500Hz
Pow
er lo
ss (W
)
0.0 1.1 1.1 0.8
11.1 11.9
0
2
4
6
8
10
12
14
Eddy currentloss @50Hz
Hysteresis loss@50Hz
Total magnetloss @50Hz
Eddy currentloss @500Hz
Hysteresis loss@500Hz
Total magnetloss @500Hz
Pow
er lo
ss (W
)
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
136
Because the magnets are being exposed to the airgap flux containing high leakage
harmonic content, (and with no protecting sleeves around the magnets), total magnet
losses in the SPM models are very much higher than that of the IPM model. In [216,
208], the measurement of magnet losses in an SPM machine show that at frequencies
lower than 1kHz, hysteresis loss is higher compared to eddy current loss. Results in
table 5.2 is consistent with the measured losses, in the sense that hysteresis loss is
higher compared to Eddy current loss at 50Hz but lower at 500Hz.
It should also be mentioned that the inverse is true for rotor core losses (that is: the rotor
core loss in the SPM machine is negligible as compared to the IPM rotor core loss due
to the magnets damping most of the varying harmonic terms from armature reaction).
Fig. 5.13 also shows that the total magnet losses with SPM bonded magnets are about
10 times lower than that of the SPM model with sintered magnets.
While the use of bonded magnets is highly favourable in SPM machines, in IPM
machines the use of bonded magnets at the expense of a reduction in torque density is
clearly not, since sintered PM losses make up less than 0.1% of the power produced.
5.4.1 Effects of Magnet Segmentation
Despite the very low magnet loss obtained in the IPM model, it would be useful to study
the effects of magnet segmentation in the SPM and IPM models. The work done in this
section takes reference to well-established work in various literature, some of which are
[140, 214, 215, 218, 219].
In work done by Yamazaki et al. in [140], magnet losses due to carrier and slot
harmonics in different rotor types were compared. With the variation of the number of
segments, SPM machines show a gradual decrease in manget eddy current loss as the
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
137
number of segments is increased, whereas IPM machines exhibit peaks (shown in fig.
5.14).
Fig. 5.14 Variation of magnet eddy current loss with number of magnet segments due to slotand carrier harmonics – Comparison between IPM, inset and SPM rotors [140]
In this section, the effects of segmenting magnets in the IPM and SPM models are
compared. The same models shown in fig. 5.10 and 5.12 are used. Fig. 5.15 shows the
circulating eddy currents in the single SPM magnet pole piece, as well as in the IPM
magnet pole consisting of two magnet pole pieces.
(a) SPM magnet pole (b) V-shaped IPM magnet pole
Fig. 5.15 Circulating eddy currents in a non-segmented poles
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
138
Fig. 5.16 shows the circulating eddy currents same magnet poles when the magnets are
segmented.
(a) SPM magnet pole (b) V-shaped IPM magnet pole
Fig. 5.16 Circulating eddy currents in a segmented magnet poles
The total magnet losses in all 14-poles obtained at 50 and 500 Hz for the SPM and IPM
machine models are shown in fig. 5.17 and fig. 5.18 respectively.
Fig. 5.17 Magnet losses in an SPM machine with variation of segment number
0
20
40
60
80
100
120
No Segments 3 Segments 6 Segments 9 Segments 12 Segments
SPM @50Hz SPM @ 500Hz
Tota
l mag
net l
oss (
W)
100%
44.4% 41.8% 39.7% 39.6%
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
139
Fig. 5.18 Magnet losses in an IPM machine with variation of segment number
Fig. 5.17 and 5.18 show that, while there is a gradual decrease in PM losses for the SPM
model, the decrease in PM losses for the IPM model was not constant, exhibiting a peak
with 9 segments. These results comply with the results obtained by Yamazaki shown in
fig. 5.14. It can be seen that segmentation has greater effect on losses in the SPM
machine as compared to the IPM machine, where more segments are required to achieve
the same percentage of losses.
With a difference of a mere 49.3% (at 500Hz) with 12 magnet segments compared to a
single pole-piece, the cost of increased complexity and duration of manufacturing the
IPM rotor is not worthwhile for this design.
5.4.2 Magnet Losses in the Final Design
In the final sintered non-segmented magnets were used. The total magnet losses for the
final model us shown in fig. 5.19:
0
0.2
0.4
0.6
0.8
1
No Segments 3 Segments 6 Segments 9 Segments 12 Segments
IPM @50Hz IPM @ 500Hz
Tota
l mag
net l
oss (
W)
100%
70.9%59.2%
78.8%
50.7%
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
140
Fig. 5.19 Magnet losses versus frequency in the final design with sintered NdFeB magnets
The above figure shows that the total magnet loss for the final design is very low
(making of for less than 1.5% of total output power of the machine). The breakdown of
losses show that hysteresis loss increases linearly, and is the dominant loss in the
magnet for frequencies <200Hz, whereas the eddy current loss increases exponentially
with frequency due to its frequency squared dependence, and begins to dominate at
higher frequencies. Compared to total core loss in the rotor, magnet losses in the IPM is
very small. Majority of the losses occurs in the rotor iron.
5.5 STATOR WINDING LOSS
Stator winding loss – also known as I2R, copper or joule loss – occurs when the
armature windings are excited by an external source. Of the total loss in PM machines,
the largest portion is usually due to I2R loss [220]. I2R is not frequency dependent and is
constant throughout the speed range, so the CPSR is not affected by this loss. This is
due to the copper conductors being thin enough to have 100% skin-depth throughout the
operating region. I2R loss is described in the following formula:
= 2 (5.20)
where, = Number of turns per coil
= Conductor resistivity (1.68X10-8 for copper)
= Cross-sectional area of wire
00.20.40.60.8
11.21.41.6
0 50 100 200 300 400 500 600
Eddy currentloss
HysteresislossPo
wer
loss
(W)
00 300 40Frequency (Hz)
Peddy = 34%Phys = 66%
Peddy = 51%Phys = 49%
Peddy = 59%Phys = 41%
Peddy = 64%Phys = 36%
Chapter 5: Analysis of Losses in an IPM Machine With Concentrated Windings for Field Weakening Applications
141
A common standard for wire sizes is the American wire gauge (AWG) standard. Some
wire sizes for typical machines in the range of 1kW are shown in table 5.3 below [221,
222]. Values indicated are for a 25°C ambient temperature and for frequencies below
In the d- and q-axis reference frame, the voltage and currents are represented by space
vectors as shown in fig. 6.2. Control of the amplitude and phase of these vectors
determines the performance of the machine, thus the term ‘vector control’. As vector
control affects the spatial orientation of the rotor and stator fields, it is also commonly
referred to as field-oriented control (FOC) [174, 225, 226]. FOC is a well-known and
simple algorithm which can be easily implemented on a fixed point digital signal
processor.
Phase A axis ( r ) (Stator ref. frame) ((Direct axis
(Rotor ref. fame)
Phase C axis (Stator ref. frame)
Phase B axis (Stator ref. frame)
Quadrature axis (Rotor ref. frame)
r
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
151
Fig. 6.2 Space vector dq-axis phasor diagram
With vector control, current is decoupled into torque- (iq) and flux- (id) producing
components. This results in transient response characteristics similar to those of a
separately excited DC machine [227].
The aim of this work is to run the machine over a wide CPSR, as well as to achieve
error-free speed responses under steady state condition. To do so, the vector control
method is employed. A well-known field-weakening method proposed by Morimoto in
[43] was successfully implemented in available DW-IPM machines, however these
methods resulted in a slight ‘over-suppression’ of the permanent magnet flux in the
CW-IPM machine.
This chapter will state the vector control method, controller architecture and inversion
technique used to produce the final three-phase inputs to the prototype CW-IPM
machine. The id trajectories during field-weakening operation calculated by equations proposed by Morimoto will be compared with trajectories obtained by repetitive testings.
d-axis (flux axis)
q-axis (EMF axis)
Iq Ef IqR jIqXs
jIdXs IdR
V I Id
(fm
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
152
6.2 CONTROL METHODOLOGY
6.2.1 Basic Equations describing the PM Machine
The 3-phase voltage equation used for describing a PM machine involves stator currents,
flux linkages and reactance values expressed as follows [227-229]:
= + = + + ( ) (6.1)
= + = + + ( ) (6.2)
= + = + + ( ) (6.3)
where, , , = Phase voltages, , = Phase currents, , = Total flux linkage in each phase windings
= Synchronous inductance( , , ) = Flux linkage in each phase windings due to rotor field
= Stator winding resistance per phase
Park’s transformation which transforms the three phase quantities to two axis quantities
can be expressed by matrix A:
= 2323 4323 4312 12 12
(6.4)
Applying Park’s transformation matrix to the three phase voltage, flux and current gives
us the following voltage equations in the dq reference frame.
= + (6.5)
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
153
= + + + (6.6)
= + = 0 for the case where 3-phases are balanced (6.7)
and expressed in matrix form is given by:
= + + 0 (6.8)
where, , = d and q axes phase inductances
, = d and q axes phase currents
= Speed in electrical radians per second
= Peak permanent magnet flux linkage
For an IPM machine, the well-known torque equation derived in [39, 230], in the dq
reference frame is given by:
= 32 + (6.9)
It should be noted that saturation casues variation in and temperature causes
variation in . However within the operating range of the CW-IPM machine in this
thesis, the rotor and stator steel are operated below the saturation region; hence these
issues would not be studied here.
The parameters of the modelled CW-IPM machine required for torque calculation are:
= ( )/ = 0.41 Wb= 84.45 mH= 81.16 mH
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
154
Here torque will be split into two terms – the alignment term and the saliency term as
shown in fig. 6.3, where the variation of torque versus Id will be shown. Iq will not be
shown but will follow the relationship = .
Fig. 6.3 Calculated torque comprising of alignment and reluctance torque from (6.9)
From fig. 6.3 it can be seen that in the CW-IPM model, the contribution of the
alignment torque makes up most of the total torque produced by the machine. Since the
reluctance term is almost negligible, the torque equation may be simplified to that of an
SPM machine:
= 32 (6.10)
6.2.2 Variation of Current Phase Angle
The current phase angle ( ) is the angle between the current phasor and the back EMF
axis as shown previously in Fig. 6.2. Variation of this angle affects the values of the d-
and q-axis current, which in turn affects the flux and torque produced by the machine.
For control of the prototype CW-IPM machine, the current angle will be varied to
achieve maximum torque per unit current (MTPC) control, where the current angle is
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
155
in-phase or almost in-phase with the back EMF phasor (fig. 6.4a), and field-weakening
(FW) control, where the current angle is made to lead the back EMF phasor (fig. 6.4b).
(a) Current phase angle under maximum torque per unit current operation
(b) Current phase angle under field-weakening operation
Fig. 6.4 Current phasor under MTPC and Field-weakening operation
In 3-phase quantities, the current angle is simply a time delay added into the current
equations as follows: = (2 + ) (6.11)
= 2 23 + (6.12)
time
V Ef I
Ef IqR jIqXs V I
pm
Ef V I
time
Iq Ef IqR jIqXs
jIdXs IdR V I
Id pm
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
156
= 2 + 23 + (6.13)
With variation of this current angle, the machine can be controlled to operate with
optimal efficiency, maximised torque and a wider CPSR above base speed under the
constraints of current and voltage limits.
For the CW-IPM machine model, the variations of EMF, current and voltage with time
at base speed, (MTPC operation), as well as maximum speed, (FW operation), is shown
in fig. 6.5a and 6.5b respectively.
(a) Respective waveforms at base speed under MTPC operation
(b) Respective waveforms at maximum speed under field-weakening operation
Fig. 6.5 Back EMF, induced current and induced voltage waveforms under MTPC and maximum field-weakening operation
= 79
I Ef V
I Ef V
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
157
Fig 6.5a shows the current in phase with the emf waveform. While in fig. 6.5b current
leads. This leading current waveform suppresses the back emf induced in the coils.
This suppression also causes distortion in the voltage waveform as shown in the figure.
6.2.3 Current and Voltage Limits
The maximum torque produced by the machine depends on the imposed armature
current limits, the maximum speed is limited by the maximum output voltage of the
inverter. In order for the CW-IPM machine to achieve a wide constant power speed
range and optimal torque density, the operating limits of the drive should be determined
by the circle diagram [31, 230]. The circle diagram consists of current-limiting circles
and voltage-limiting ellipses as shown in fig 6.6. Fig. 6.6a and 6.6b show limiting
values and trajectories for the IPM machine and SPM machines respectively.
(a) Circle diagram for an IPM machine
(b) Circle diagram for a SPM machine
Fig. 6.6 Circle diagram for IPM and SPM machine showing current and voltage limits of the system
Current limiting circlesMTPC trajectory ybase
12Voltage limiting ellipses
MTField-weakening trajectory
, 0Is(max)
Current limiting circles CMTPC trajectory base1
2Voltage limiting ellipses
Field-weakening trajectory
, 0 Is(max)
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
158
Since the saliency ratio of the CW-IPM machine is so low, the id and iq current
trajectories would be similar to that of the SPM machine in fig 6.5b.
In the circle diagrams, the outer most limit of the current-limiting circle is expressed
simply as:
( ) = + (6.14)
and the voltage-limiting ellipse at various speeds is given by:
( ) = + (6.15)
where under ideal and steady state operations and derived earlier in (6.5) and
(6.6) is given by: = (6.16)= + (6.17)
Thus, in terms of current values, the voltage limiting ellipse is given by:
( ) = + + (6.18)
The centre of the voltage limiting ellipses lies at the point:
, 0 (6.19)
From standstill to base speed, the machine operates within the current limit along the
MTPC trajectory. When base speed is reached, the operating point follows anti-
clockwise along the current limiting circles. This results in an increase in - at the
expense of . An increase in - creates a larger opposing flux to the rotor pole axis,
which temporarily ‘weakens the field’ of the magnets. Field-weakening of the magnets
helps to limit the amount of flux that is linked to the stator windings, thus limiting the
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
159
back EMF generated. This helps to maintain voltage under operating limits as speed
continues to increase past base speed.
The control limits on the field-weakening range of the machine depend on the
characteristic current. Practical IPM machines can be classified in two categories:
Type 1: ( ) <Type 2: ( ) >In terms of field-weakening control, type 1 machines have a finite field-weakening
range, whereas type 2 have an infinite range. This is due to the centre of the voltage
limiting ellipse lying outside the current limiting circle for type 1 machines and inside
for type 2 as shown in 6.7a and 6.7b respectively. With the centre of the voltage limiting
ellipse lying outside of the current limiting circle, there exists a maximum speed beyond
which both the current limits and voltage limits can no longer be satisfied. Whereas if
the centre of the voltage limiting ellipse lies inside the current limits, both these limits
would always be satisfied with the application of the voltage-limited maximum output
trajectory, thus theoretically resulting in an infinite speed range.
(a) Type 1 machine with centre of voltage limiting ellipse lying outside of the current limiting circle
Field-weakening trajectory max
base MTPC trajectory
, 0
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
160
(b) Type 2 machine with centre of voltage limiting ellipse lying inside of the current limiting circle
Fig 6.7 Classification of machine type by characteristic current
The CW IPM machine falls under the category of a type 1 machine as the rated current
is lower than the characteristic current. Thus the voltage-limited maximum output
trajectory cannot be applied to this machine. For this work the machine would operate
only with the two trajectories- the MTPC trajectory and the field-weakening trajectory.
The id and iq current trajectory plotted for the final CW-IPM FE model is shown in fig.
6.8.
Fig. 6.8 id and iq field-weakening current trajectory for the CW-IPM model
max Voltage-limited trajectory
Field-weakening trajectory MTPC trajectory base
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
161
Fig. 6.8 shows the very large requirement of negative d-axis current for initial field-
weakening up to 200Hz. The requirement of additional negative d-axis current
decreases exponentially as speed increases. In the FE model, the limit on this field-
weakening trajectory as speed increases is where the armature field can no longer
maintain constant power, (higher than power produced at base speed), at a specific
voltage limit.
6.2.4 Maximum Torque Per unit Current and Field-Weakening Trajectories
Here two trajectories will be compared: firstly, the trajectory calculated by widely used
Morimoto’s equations in [43]; secondly, values obtained when the machine is regarded
as an SPM machine during MTPC operation and subsequently by repeated testing in the
field-weakening region.
Under MTPC operation, calculated by [43] is shown in (6.20), and the remaining
to produce torque at full load is shown in (6.21).
= 4 16 + 2 (6.20)
( ) = (6.21)
When regarded as an SPM machine, the maximum torque point will intersect with the
current limit circle along the q-axis. Thus will be equal to zero and at full load, will
be equal to the supply current. = 0 (6.22)
( ) = (6.23)
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
162
For both these trajectories, the speed at the maximum torque point is determined by
( ) as well as the voltage limit of the inverter ( ). = ( )+ ( ) (6.24)
In the field-weakening region, cross-coupling effects will be omitted, and will be
regarded as an independent quantity of . While is largely determined by the speed
controller to produce sufficient torque to achieve the desired speed, will be used to
supress flux produced by the magnets and maintain constant power.
From [43], for field-weakening is calculated by (6.25).
= + 1 (6.25)
Through repetitive testings, the values of required to weaken the magnet field are
lower as compared to (6.25). The trajectories comparing these two methods applied to
the CW-IPM FE model are shown in fig. 6.9.
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
163
Fig. 6.9 Comparison of d-axis current trajectories by Morimoto’s equations and through repetitive testings
From the fig. 6.9 it is shown that the CW-IPM model requires a lower amount of d-axis
current to maintain constant voltage/power, as speed is increased. Chapter 7 will show
that a similar characteristic is seen in the constructed prototype machine.
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
164
6.3 CONTROLLER ARCHITECTURE
The input to the controller is a speed reference signal ref), and the outputs of the
controller are the desired three-phase voltages and currents with desired magnitudes and
phase displacements. To obtain the desired magnitudes and phase displacements, a 3-
phase fully-controlled inverter is used. The controller block diagram is shown below:
Fig. 6.10 Vector control system block diagram
This system requires a position sensor, which is mounted to the shaft of the machine.
The position feedback is used to convert 3-phase quantities to dq-axis quantities; it is
also differentiated with respect to time to provide a speed feedback. Simple high-gain PI
controllers are implemented to achieve the desired speed and dq-axis current dynamics,
as well as to remove any tracking or following error. The compensated errors must be
first converted back to three-phase references to be fed into the inverter.
Speed controller Current ref. Generatoriq current controllerid current controller
dq -1
dq
Inverter
refCTref
ag
iq(ref)
id(ref)iq
id
Vref(a,b,c)
ia,b,c
mCW IPM
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
165
6.3.1 Three-phase Inversion Technique
For traction applications, the only available energy source is usually the battery
producing DC voltages. This DC supply is then inverted to desired 3-phase supply
currents and voltages with desired magnitudes and phase displacements. Fig. 6.11 gives
an example of the inverter architecture for the drive system used in this work. (A more
detailed schematic, as well as its connections to the controller board can be seen in
appendix F).
Fig. 6.11 Rectifier – Inverter for producing three-phase outputs to the machine
In the experimental set up, a 3-phase AC source is rectified using an uncontrolled full-
bridge rectifier to produce a desired DC bus input voltage to the inverter. This DC bus
voltage will then be converted to desired AC signals by modulating techniques. Two
common modulating techniques include the sinusoidal pulse width modulation (SPWM)
scheme, and the space vector modulation (SVM) scheme. Here, SVM is preferred due
Controlleria,b,c
m
S1
S2
S3
S4
S5
S6VDC
+
-Switching signals S1 – S6
CW IPMUncontrolled full-bridge rectifierThree-phase input from source
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
166
to its lower hardware requirements and a higher modulation ratio of 0.907 compared to
0.785 in the SPWM scheme.
In the SVM scheme, the desired three phase quantities Van, Vbn and Vcn are sampled in
time at a specified sampling frequency (fs) and are represented in terms of space vectors
as shown in fig. 6.12.
There are eight possible switching states (V0 to V7), six of which are termed switching
vectors (V1 to V6) and the other two (V0 and V7) are termed zero vectors.
Fig 6.12 Switching vectors of the space vector modulation method
The SVM method uses a look-up table (table 6.1) to determine the switching states.
This look-up table contains a set of switching rules which enable the voltage vector Vx to
rotate continuously with smooth transition from one sector to the next.
V1 (1,0,0)
V2 (1,1,0)V3 (0,1,0)
V4 (0,1,1)
V5 (0,0,1) V6 (1,0,1)
Sector 1Sector 2
Sector 3
Sector 4Sector 5
Sector 6v1
v2 SeVx
V0 (0,0,0) V7 (1,1,1)
Vref(max)
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
167
Table 6.1Space vector modulation look-up table
State S1 S2 S3 S4 S5 S6 Van Vbn Vcn Space Vector
0 OFF ON OFF ON OFF ON 0 0 0 V0 = 0,0,0
1 ON OFF OFF ON OFF ON V1 = 1,0,0
2 ON OFF ON OFF OFF ON V2 = 1,1,0
3 OFF ON ON OFF OFF ON V3 = 0,1,0
4 OFF ON ON OFF ON OFF V4 = 0,1,1
5 OFF ON OFF ON ON OFF V5 = 0,0,1
6 ON OFF OFF ON ON OFF V6 = 1,0,1
7 ON OFF ON OFF ON OFF 0 0 0 V7 = 1,1,1
Taking sector 1 as an example, the relationship between Vx and the two vectors V1 and
V2 is given by: (3 ) = 3 (6.26)
= 3 (6.27)
In terms of v1 and v2,
= 23 (3 ) (6.28)
= 23 (6.29)
Vectors v1 and v2, which are subsidiary vectors of vectors V1 and V2 respectively,
determines the position of the rotating vector Vx. The desired magnitude of the voltage
Vx is controlled by activating vectors V1 and V2 for durations t1 and t2 respectively over
half a period Ts/2. The expression of Vx in terms of activations times is given as:
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
168
= + = / + / + ( ) / (6.30)
where, = / (6.31)
= / (6.32)
= / (1 ) (6.33)
The switching pattern for sector 1 is shown in fig. 6.13:
Fig. 6.13 Switching pattern for sector 1
This process is repeated for the other five sectors. In this way, the magnitude, frequency
and phase of the voltage can be varied according to the desired output voltage signal as
specified by the controller.
V0 V1 V2 V7 V7 V2 V1 V0
t0/2 t1 t2 t0/2
Ts/2 Ts = 1/fs
Phase APhase BPhase C
Chapter 6: Vector Control of the IPM Machine with Concentrated Windings
169
6.4 CONCLUSION
This chapter has illustrated the control methodology of the CW-IPM machine. It
showed the voltage and current limits that the drive is subjected to by use of the circle
diagrams. Due to the centre of the voltage limiting ellipses lying outside of the current
limiting circle, the voltage limited trajectory is not applicable for the CW-IPM model.
The model was subjected to two different current trajectories: one being the widely-
used trajectory calculated by Morimoto’s equations and the other through repetitive
testings. It was shown that the actual current required to weakening the magnet field
and maintain constant voltage is lower than that calculated by available equations.
Lastly this chapter also showed the general controller architecture and the SVM
technique used to generate the three phase input quantities.
170
CHAPTER 7CONSTRUCTION AND PERFORMANCE ANALYSIS OF THE CONCENTRATED WINDING IPM MACHINE PROTOTYPE
7.1 INTRODUCTION
The work done in previous chapters provided a study of the CW-IPM machine for use
in field-weakening applications. From this study, a final 14-pole, 18-slot model was
designed and optimised using FE analysis. The final FE model was built to verify
studies and predicted performance characteristics.
This chapter will first give an overview of the construction process of the CW-IPM
prototype (which was the final FE design shown towards the end of chapter 4).
Problems encountered and lessons learnt during the construction process will be stated
in order to facilitate quicker and less problematic manufacturing for future designs.
Subsequently, the open circuit parameters of the prototype machine will be measured,
and compared with results achieved by the FE model. Control techniques shown in
chapter 6 will be used to run the machine in constant torque and field-weakening
regions. A steady state analysis will be performed and the torque performance of the
CW-IPM machine in both operating regions will be shown. Transient characteristics of
the machine in the MTPC region will also be briefly covered. Lastly, the performance of
the CW-IPM machine will be compared with two other similar-sized DW-IPM
machines.
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
171
7.2 CONSTRUCTION PROCESS
The construction process of the CW-IPM machine can be broken down into three steps:
the rotor and stator assembly; stator winding; and finally the total machine assembly.
The total manufacturing time of the final CW-IPM prototype was twelve months. This
section illustrates the abovementioned three steps of constructing the motor, as well as
the manufacturing timeline of the construction process.
7.2.1 Manufacturing Duration
The flow-chart in fig. 7.1 shows the entire construction process from the date when
designs were submitted to the manufacturers to the date the experimental setup was
completed. The reason for describing this process is to give future PhD
students/researchers a clearer picture of what has to be done if a prototype is to be
constructed. Key delays in the manufacturing process will be stated and suggestions to
speed up the process will be made.
Key delays for the manufacturing process and suggestions to avoid them include:
Delay: Redesigning of unmanufacturable portions of the initial model.
- Suggestion: Regular meetings with manufacturer during the design stage
Delay: Duration taken for the desired steel grade to be shipped.
- Suggestion: Choose and order desired core material before the design
Delay: Manufacturing errors in cutting of laminations.
- Suggestion: Not applicable.
Delay: Windings not being wound according to a specified layout.
- Suggestion: Commercial winders are not always adaptable to new winding
types. Coils can be fitted by the winders but phase connections can be done in
our labs.
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
172
Fig. 7.1 Duration breakdown and steps of construction process
Meeting with CSIRO:Discussion of design
Correction of unmanufacturable portions
Choice of alternate steel grade
Ordering of magnets from China (RareMag)
Correction of slot opening size, magnet slot size
Modelling performance with available steel grade
Odering of steel (Sankey) and lazer cutting (LazerXperts)
Rotor assembly (CSIRO):Stacking laminations; insertion of magnets; dynamic balancing
Stator assembly (CSIRO):Stacking laminations; heat shrinking into casing
Removal of original stator and rotor from ABB casing
Winding of stator (Atom Electrical)
Assembly of machine and experimental setup (UNSW)
Start of manufacturing process
Month 1 to Month 3
Month 4 to Month 8
Month 9 to Month 11
Month 12
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
173
7.2.2 Rotor Assembly
As mentioned in chapter 4, the material used for both the rotor and stator laminations
was 0.35mm thick (35RM300). A laser-cut rotor lamination of the prototype is shown in
fig. 7.2. The 79mm rotor stack was assembled by pressing one lamination at a time
down a shaft; tapered ends of the shaft simplify this process.
On completion of the stack, magnets are then inserted. Inserting the magnets is a fairly
easy task, as all the magnet pieces get drawn into the rotor stack. The completed rotor
stack with magnets inserted is shown in fig. 7.3.
Fig. 7.2 Laser-cut rotor lamination
It should also be noted that axially segmented magnets, (for the purpose of eddy current
loss reduction), can also be easily inserted piece by piece into the rotor. The segments
making up a magnet pole would not be repelled out of the slot as initially expected.
Magnet slots4mm d holes for bolts to holding end-plates and laminations together
6mm key hole
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
174
Fig. 7.3 Complete rotor stack with magnets inserted
Lastly the end plates are bolted on, bearings mounted and the rotor is dynamically
balanced. Dynamic balancing is done by drilling holes into the endplates to achieve
rotational weight balance.
Fig. 7.4 Completed dynamically-balanced rotor
Shaft
Magnets
Endplate
Bolts
Bearing mounted
Holes drilled in endplate to achieve dynamic balance
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
175
7.2.3 Stator Core Assembly and Stator Windings
The same rotor steel grade (35RM300) is used in the stator. The completed 80mm stator
stack is shown in fig. 7.5. Six equally spaced, w-shaped groves were cut into the outer
periphery of the stator for welding and alignment purposes. It should be noted that these
groves should be situated in the inter-slot section, so as not to hinder flux paths in the
stator yoke section.
Fig. 7.5 Completed stator stack
The stator was wound with the horizontal-fill method, as discussed previously in
chapter 4. The windings done by Atom Electrical are of acceptable quality, with 42%
slot-fill factor achieved. The initially achieved 14.5mm end winding length with the
plastic stator could not be achieved in the winding of the prototype stator; the actual end
winding length measured 18.5mm (shown in fig. 7.6).
Groves situated in the inter-slot sections
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
176
Fig. 7.6 Measurement of end winding length in UNSW CW-IPM stator
7.2.4 Overall Machine Assembly
Due to the high remanent flux density of the magnets, inserting the rotor into the stator
required the use of an aluminium sleeve. This method is tedious and causes damage to
both the stator and rotor laminations. An alternative method of assembling the machine
is to first remove the magnets from the rotor; insert the rotor; then re-insert the magnets.
For this machine, the time taken to assemble it using the latter method was slightly less
compared to the use of an aluminium sleeve. Fig. 7.7 shows the completed CW-IPM
machine assembly in comparison to the DW S-IPM machine.
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
177
Fig. 7.7 Comparison of UNSW CW-IPM machine assembly (right) and DW S-IPM machine (left)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
178
7.3 OPEN CIRCUIT PARAMETERS
In chapter 4, it was shown that with an 18-slot, 14-pole FE model, a high winding factor
and sinusoidal back EMF could be achieved. Due to the elimination of slot and pole
periodicity achieved by fractional-slot distribution, very low cogging torque can also be
achieved.
7.3.1 Back EMF
For back EMF measurements, the CW-IPM machine terminals were left open-circuited
and the machine was ran at various speeds by a prime mover (1kW Kollmorgen
machine). The measured line to line- and line to neutral- induced voltages were 154Vrms
and 89Vrms respectively. These values were very similar to the modelled values, albeit
slightly lower (157Vrms and 92Vrms). The near-sinusoidal shape of the measured
waveforms also followed very closely to the modelled ones as shown in fig. 7.8.
Fig. 7.8 Measured line to line and line to neutral back EMF waveforms compared against modelled values
-250-200-150-100
-500
50100150200250
0.005 0.01 0.015 0.02 0.025
ModelledVab @50Hz
MeasuredVab @50Hz
ModelledVan @50Hz
MeasuredVan @50Hz
Indu
ced
Vol
tage
(V)
Time (s)
VL-N(Measured) = 89Vrms
VL-N(Modelled) = 92Vrms
VL-L(Modelled) = 157Vrms
VL-L(Measured) = 154Vrms
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
179
Fig. 7.9 shows the linear increase in back EMF and speed.
Fig. 7.9 Measured line to line back EMF versus speed compared against modelled values
7.3.2 Cogging Torque
Cogging torque in the CW-IPM machine is expected to have a high frequency carrier
signal with a low frequency modulating signal. The experimental setup shown in fig.
7.10 consists of a 0.6m beam, balanced on both sides, weights, and a position sensor. By
gradually adding weights at each position, the amount of torque required to rotate the
machine at various angles can be calculated. The cogging torque measurements made in
the clockwise direction make up the positive half of each fluctuation, and vice versa.
Fig. 7.10 Cogging torque measurement setup
0100200300400500600700800
0 500 1000 1500 2000
ModelledVab Vs.Speed
MeasuredVab Vs.Speed
Speed (rpm)
Indu
ced
Vol
tage
(Vrm
s)
0.6m Beam (Balanced on both sides)
Position Sensor(Connected to DS1104 Board)
Weights (1g to 500g)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
180
Fig 7.11a (measurements from 3°-10°) and 7.11b (measurements from 12°-18.5°)
compares the cogging torque points measured over 1 cogging torque period (20
mechanical degrees), with the results obtained from the FE model.
(a) Measurement range – 3 to 10 degrees
(b) Measurement range – 12 to 18.5 degrees
Fig. 7.11 Measured cogging torque points compared against results obtained from FE model
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
181
From fig. 7.11, it is shown that the measured points agrees with values from the FE
model, with slight deviations in the rotor angle. The peak magnitude of the measured
points were higher (approximately 2x higher) than the peaks of the waveform obtained
by the FE model. Considering that the cogging torque values are extremely small (<
±0.018Nm(p-p)), effects on bearing striction and/or rotor eccentricity/balance (which
were not considered in the FE model), would contribute significantly to the magnitude
of measured torque points. Due to the high frequency fluctuations created by fractional
slot distribution, measured cogging torque points had to be taken in 0.2 degree intervals.
This contributed to the errors in the rotor positions where torque was measured.
From the measured cogging torque points, a curve achieved from FE analysis was fitted.
This curve termed as ‘expected cogging torque waveform’ (shown in fig. 7.12), would
be used in place of the actual measured cogging torque waveform in the subsequent
sections as a comparison to the DW-IPM models.
Fig. 7.12 Curve fitted cogging torque waveform
It can be seen that the modulation of the cogging torque waveform is proportional to the
width of the slot (i.e. 2 cycles per slot), similar to that of regular integral slot machines.
Envelope for peak cogging torque values
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
182
The only difference is the high frequency ripples, which are caused by the aperiodic
flux linkage in other slots and phases (a characteristic of fractional slot windings),
which has been reported in several recent papers as well.
7.3.3 Inductance and Saliency Ratio
The inductance waveform was measured using the AC standstill test method with
single-phase excitation, (stated in chapter 3). Static measurements were performed in
two degree increments throughout one electrical revolution, with the position being read
from a mounted encoder. The variation of voltage in self and mutual-phases was then
used to calculate the self- and mutual-inductance values at each position. The self and
mutual-inductance values (measured at 3A) from the finite element model, as shown in
chapter 4 are re-illustrated here as a comparison to the measured values. Fig. 7.13 shows
the inductance values from the FE model and fig. 7.14 shows the values measured from
the prototype.
Fig. 7.13 Modelled self- and mutual-inductance waveform from the FE model with 3Armscurrent excitation
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
183
Fig. 7.14 Measured self and mutual-inductance waveform from the prototype with 3Arms current excitation
From fig. 7.13 and 7.14, it is seen that the average and peak to peak magnitude of the
self-inductance obtained in both the FE model and prototype machine are relatively
similar. While the mean value mutual-inductances are also relatively similar, the
variations in the prototype machine are much higher compared to the FE model. This
larger variation in mutual–inductance leads to a higher saliency ratio of 1.12, compared
to an almost negligible saliency ratio of 1.04 in the FE model. Ld and Lq for the FE
model are 81.16mH and 84.45mH, whereas Ld and Lq for the actual machine are
82.52mH and 92.48mH respectively. The most probable reason for this difference is,
that 2D FE model was not able to account for some of the flux leakage occurring toward
the end of the windings. We can also conclude here that end-turn winding of the q-axis
inductance is more than the d-axis inductance. This is in fact advantageous for the
machine because it can be a tool to optimize the saliency ratio.
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
184
With different excitations, saturation of the core also causes the inductance to change
slightly. Fig. 7.15 compares the measured and modelled dq-axis inductance values
between 1Arms to 3Arms.
Fig. 7.15 Variation of dq-axis inductances with current
Fig. 7.15 shows that overall inductance values fall as the current increases. For the
measured inductance, the increase in current has a greater effect on the q-axis
inductance compared to the d-axis inductance, thus leading to a slight decrease in the
saliency ratio with higher currents.
0
20
40
60
80
100
1 1.5 2 2.5 3
Indu
ctan
ce (m
H)
Current (A)
d-axis inductance(modelled)
q-axis inductance(modelled)
d-axis inductance(measured)
q-axis inductance(measured)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
185
7.4 STEADY STATE ANALYSIS
The steady state analysis of the machine includes the torque/power versus speed curves
as well as the id and iq trajectory required to achieve the curves; the voltage trajectory
from zero to maximum speed; and efficiency over the CPSR. Steady state waveforms of
the input currents and voltages as well as id and iq values will also be shown.
7.4.1 Torque and Power Characteristics
In Chapter 4, it is shown that a 10:1 CPSR can be achieved for the FE model with an
input line to line voltage of 240Vrms and a base speed of 429rpm. Due to loading and
equipment constraints – torque transducer, gearbox and loading generator are rated for a
shaft torque of approximately 10Nm – as well as the desire to achieve a higher
efficiency, the voltage limit on the constructed CW-IPM prototype was increased:
Input line to line voltage: 340Vrms
Rated current: 2.2Arms
Base speed with rated voltage: 573rpm
The main components of the experimental setup are shown in fig. 7.16. (the full setup is
shown in appendix F)
Fig. 7.16 Experimental setup to measure back EMF and torque/power versus speed performance
CW-IPM motor(connected to inverter)
Torque transducer
Loading generatorWith external resistor bank (not shown)
Gearbox(4:1)
Encoder
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
186
The three-phase supply to the CW-IPM machine is from an inverter (diagram shown in
chapter 6, fig 6.7). The three-phase space vector modulation (SVM) generated supply
voltage and currents were derived based on controller outputs. The controller
implemented in C code, ran together with dSPACE control desk which provided the
graphical user interface on a Windows-based PC. Input and output signals of the
controller were handled by a DS1104 ADC/DAC control board.
From zero to base speed, the machine ran under MTPC conditions. For the CW-IPM
machine, it is assumed that under MTPC operation, id is zero and iq is equal to the
supply current is. This is due to the fact that the saliency ratio is close to unity. After
base speed, when the rated voltage is reached, id is then made to oppose the flux
produced by the rotor magnets. This is done to maintain constant voltage by suppressing
the induced back EMF as speed increases. The required id and iq current trajectory to
maintain constant voltage with increasing speed is shown in fig. 7.17:
Fig. 7.17 Measured dq-axis current points under field weakening operation
Maximum FW point
MTPC point
FW Region
Unstable region
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
187
The voltage measured throughout the speed range from operation in the MTPC region
to the maximum achievable speed in the field weakening region is shown in fig. 7.18:
Fig. 7.18 Measured line to line voltage versus speed
From the above figure, it can be seen that a constant voltage can be maintained after
base speed when the FW trajectory in fig. 7.17 is followed. Here the field-weakening
range/CPSR is taken from base speed (the point where the rated voltage is reached) to
the point where power drops below the value achieved at base speed. The torque and
power versus speed characteristics based on the current and voltage points from the
abovementioned figures are shown in fig. 7.19:
Fig. 7.19 Measured torque versus speed characteristics of the CW-IPM machine prototype
050
100150200250300350400
0 1000 2000 3000 4000 5000
Line to line voltage versus speed
Inpu
t vol
tage
(VL-
L)
2000 3000Speed (rpm)
MTPC Region
FW Region
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
188
With a voltage limit of 340VL-L, greater than 7:1 CPSR (7.2:1 max) can be achieved.
Output power at base speed is 762W; over 800W of power is achieved just slightly
above base speed all the way to the 7:1 FW point. The maximum power of 905W was
achieved at 1290rpm. Over 80.8% efficiency was achieved from base speed till the 6.2:1
FW point. The efficiency of the machine versus speed is plotted in fig. 7.20:
Fig. 7.20 Measured efficiency versus speed
Within the 6.2:1 range (indicated in fig. 7.20) the efficiency varied from 80.8% to 83%.
Measured efficiency before and after this range dropped rapidly to 61.4% at the lowest
measured speed of 191rpm, and to 65.1% at the highest achieved speed of 4889rpm,
(which was the maximum speed where constant voltage can be maintained, 8.5:1 point).
Therefore, the optimal operating range in terms of efficiency would be between 573rpm
(base speed) and 3562rpm, resulting in a 6.2:1 CPSR. Comparison between the
experimental results and the predicted efficiency (by calculations and FE analysis)
showed result were very close (with 0.2 – 3% error). This justifies the loss predictions
shown in chapter 5.
0102030405060708090
100
0 500 1000 1500 2000 2500 3000 3500 4000
Measured Efficiency vs. SpeedPredicted Efficiency vs. Speed
> 80% Efficiency over a 6.2:1 CPSR
Effic
ienc
y (%
)
Speed (rpm)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
189
7.4.2 Steady State Voltage and Current Characteristics
This section illustrates the steady state speed, currents and voltages in the dq-axis
reference frame, as well as the three phase current and voltage input to the machine.
These readings were taken at base and maximum FW speed. The speed waveforms
under full load conditions are shown in fig. 7.21. The signals in red (#1:2) are the
desired controller references, and the signals in green (#1:1) are the following/actual
output signals.
(a) Base Speed (b) Maximum Speed
Fig. 7.21 Steady-state speed waveforms
The steady state speed output follows the reference with zero error at base and
maximum field-weakening speed. The following speed signal at maximum speed is
constant, but at base speed the signal has low magnitude perturbations. At these two
speeds current and voltage signals in their corresponding reference frames are shown in
fig 7.22.
Time (s)
(RPM
)
Time (s)
(RPM
)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
190
(a) Base Speed (b) Maximum Speed
Fig. 7.22 Steady state id current waveforms
Reference id values are user inputs to the program and are therefore constant. The
following id signals contain high frequency fluctuations of substantial magnitude. These
fluctuations, which are also present in the in the iq actual signal, lead to vibrations and
additional acoustic noise in the very low speed and near the maximum speed of the
machine.
(a) Base Speed (b) Maximum SpeedFig. 7.23 Steady state iq current waveforms
The slight fluctuations in the reference iq current waveform were more pronounced at
higher speeds. The actual iq current waveforms follow the reference waveforms as long
as the required voltage does not exceed limited values. After the maximum speed of
Time (s)
(A)
Time (s)
(A)
Time (s)
(A)
Time (s)
(A)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
191
4889rpm, iq becomes uncontrollable and develops following errors, thus the CW-IPM
drive is limited to this speed.
The corresponding current waveforms and voltage pulses in the abc reference frame for
steady state operation is shown in fig. 7.24 below: (Switching frequency used here is
10kHz).
(a) Current and voltage waveforms at base speed of 60rad/s
(b) Signals at maximum FW speed of 426rad/s
Fig. 7.24 Steady state current and voltage inputs to the machine in abc reference frame
Time (s)
Curr
ent (
A)
Volta
ge (V
)Vo
ltage
(V)
Time (s)
Curr
ent (
A)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
192
7.5 TRANSIENT RESPONSE UNDER MTPC OPERATION
In this work, the steady state performances would be sufficient to verify the CPSR,
power density and efficiency of the prototype CW-IPM machine. Thus, only the basic
dynamic responses in the MTPC region (standstill to base speed) will be shown. Control
strategies to improve the dynamic performance of the CW-IPM machine are currently
being studied.
7.5.1 Transient Voltage and Current Characteristics
Fig. 7.26 shows the speed step response of the CW-IPM drive:
(a) No load (b) Full load
Fig. 7.25 Speed step from standstill to base speed
(a) No load (b) Full load
Fig. 7.26 id current waveforms with speed step from standstill to base speed
Time (s) Time (s)
(RPM
)
(RPM
)
Time (s) Time (s)
(RPM
)
(RPM
)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
193
The id current waveforms show large negative spikes during the speed transition,
getting more pronounced with a higher load, despite the reference being zero.
(a) No load (b) Full load
Fig. 7.27 iq current waveforms with speed step from standstill to base speed
The actual iq current waveforms follow the reference well, however, high frequency
fluctuations are present in the reference signal.
The corresponding current waveform and voltage pulses for the speed step at no load
and full load in the abc reference frame is shown in fig. 7.28 below:
(a) No load
Vol
tage
(V) C
urrent (A)
Time (s)
Time (s) Time (s)
(RPM
)
(RPM
)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
194
(b) Full loadFig. 7.28 Current and voltage inputs to the machine in abc reference frame with step change in
speed
7.5.2 Torque Transients
The output torque of the machine is determined from a torque transducer mounted
between the drive and the loading machine. Fig. 7.29 shows the torque transient at full
load when the machine accelerates from standstill to base speed. Fig. 7.30 shows the
corresponding torque ripple when steady state is reached.
Fig. 7.29 Torque transient when CW-IPM machine accelerates from standstill to base speed at full load
0
2
4
6
8
10
12
14
0 0.5 1 1.5 2
Current (A
)Vol
tage
(V)
Time (s)
Torq
ue (N
m)
Time (s)
Section of torque ripple shown in fig. 7.31
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
195
Fig. 7.30 Measured torque ripple at steady state
The output torque ripple was 2.1Nm(peak to peak), which amounts to 16.5% of the total
torque produced at base speed. The magnitude of torque ripple is higher than predicted.
This could be due three main reasons:
i) The noisy gearbox (in terms of vibrations) at the load – resulting in relative
fluctuations at the torque transducer.
ii) The second reason being the controller design. More detailed parameter
identification and control strategies which are currently being implemented
fall beyond the scope of this thesis. This strategies, implemented on the
CW-IPM machine will be shown in future publications.
iii) Lastly, increased torque ripple could be caused by the load machine and
misalignment in coupling.
0
2
4
6
8
10
12
14
1.8 1.85 1.9 1.95 2
Torque ripple magnitude = 2.1Nm(peak to peak) To
rque
(Nm
)
Time (s)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
196
7.6 PERFORMANCE COMPARED TO DISTRIBUTED WINDING IPM
MACHINES
The main aim of this thesis was to design and build a CW-IPM machine to achieve
higher torque/power density, wider CPSR, lower cogging torque, and of equivalent
efficiency as two other available DW-IPM machines of equal size [13]. As a
comparison, two other previously constructed, equally sized DW-IPM machines will be
used. All three machines were designed to fit in the same 550W ABB casing. The two
DW machines have 4-poles, while the CW machine has 14-poles. This is due to the
required 14-pole, 18-slot layout to achieve low cogging torque and an appropriate back
EMF waveform, as mentioned in earlier chapters. Since pole numbers are not equal, a
comparison of output torque would not be appropriate. Most other quantities such as
CPSR, output power, cogging torque and efficiency can still be compared.
The first DW IPM machine (shown in fig. 7.31a) has regular single-pieced, flat shaped
poles. It uses sintered magnets with Br = 1.05T. This machine will be named IPM-I.
The second DW-IPM machine, (shown in fig. 7.31b), has segmented magnets. It uses
bonded magnets with Br = 0.78T. This machine will be named S-IPM.
It should be pointed out that the IPM-I was designed based on prior knowledge and
experience. No optimisation strategy was implemented. The S-IPM machine was
designed and optimised in a similar fashion to the CW-IPM machine- with FE analysis
as shown in this work. Additional details on its optimisation can be seen in [174].
The final CW-IPM machine model is shown in fig. 7.31c. It uses a similar magnet grade
as IPM-I with Br = 1.04T.
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
197
(a) Single piece per pole IPM with DW (DW IPM-I)
(b) Segmented IPM machine with DW (DW S-IPM)
(c) Constructed CW-IPM prototype (CW-IPM)
Fig. 7.31 Comparison of three UNSW IPM machines
DW S-IPM
DW IPM-I
CW-IPM
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
198
7.6.1 Power and Torque versus Frequency Comparison
As the CW-IPM machine has 14 poles and the two other DW-IPM machines have only
4 poles each, frequency is used as the basis of comparison instead of mechanical speed.
Fig. 7.32 shows the power versus frequency comparison between the three IPM
machines. It should be noted that while the comparison against the DW IPM-I would
not be fair due to the fact that the machine was not optimised for field weakening
performance nor for high power density. It is included to show that despite a much
lower saliency ratio, magnet volume and magnet energy density, the design is still the
most crucial in achieving good field weakening performance.
Fig. 7.32 CPSR comparison between the three UNSW IPM machines
From the results obtained, it is shown that the CW-IPM machine not only achieved a
much wider CPSR – > 6.2:1, compared to 4:1 in the S-IPM machine and almost no field
weakening capability in IPM-I – it also achieved a 56% power increase over the
constant power region as compared to the two other DW-IPM machines.
In terms of peak torque under maximum torque per unit current (MTPC) operation, the
CW-IPM machine achieved a peak shaft torque of 12.7Nm, compared to 2.25 Nm and
0100200300400500600700800900
1000
0 50 100 150 200 250 300 350 400 450
Pow
er (W
)
Frequency (Hz)
> 6.2:1 CPSR
4:1 CPSR
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
199
2.3Nm in the S-IPM and IPM-I respectively. For a fair comparison of torque between
the 14-pole CW-IPM machine and the 4-pole DW-IPM machines, torque is normalised.
The torque magnitude of 1p.u. is assumed in the MTPC region.
Fig. 7.33 Normalised output torque comparison between the three UNSW IPM machines
The torque versus frequency waveforms clearly show that IPM-I has almost no field
weakening capability, with torque rapidly falling at the beginning of the field
weakening region. On the other hand, it is shown that both the S-IPM and CW-IPM
have excellent field weakening capability.
7.6.2 Cogging Torque Comparison
Comparing the two DW machines the S-IPM machine uses magnets with a lower
remanent flux density of 0.78T as compared to IPM-I, which uses magnets with remanet
flux density of 1.05T. Thus the cogging torque is naturally higher in the latter. However,
despite the use of high remanent flux density magnets in the CW-IPM machine (1.04T),
the cogging torque is substantially lower compared to the DW machines due to the
elimination of periodicity of slots and poles by using fraction-slot distribution.
0
0.2
0.4
0.6
0.8
1
0 50 100 150 200 250 300 350 400 450
Torq
ue (P
.U)
Frequency (Hz)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
200
Cogging torque comparisons between the three IPM machines is shown in fig. 7.34:
Fig. 7.34 Cogging torque comparison between the three UNSW IPM machines
From the comparison of the cogging torque values, it is shown that the CW-IPM
machine produced lower cogging torque magnitude, compared to the two other DW-
IPM machines. As expected IPM-I produced the largest peak cogging torque magnitude.
As higher generated torque will make the effects of cogging torque less significant,
another important comparison is, the amount cogging torque produced as a percentage
of the total torque generated in the MTPA region. This comparison is shown in fig.7.35
as follows:
Fig. 7.35 Cogging torque as a percentage of output torque at base speed – comparison between the three UNSW IPM machines
0.038Nm(p-p)0.126Nm(p-p)
0.502Nm(p-p)
0.3%(p-p)8.1% (p-p)
24.2%(p-p)
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
201
7.6.3 Efficiency Comparison
Data on efficiency versus speed for IPM-I is not readily available, although it is known
to be somewhere in the 60-65% range at rated speed, so efficiency was compared only
between the CW-IPM and SEG-IPM as shown in fig. 7.36:
Fig. 7.36 Efficiency comparison between the CW-IPM and S-IPM machine up to 200Hz
The SEG-IPM machine produced 84 to 85% efficiency from base speed up to the
measured 200Hz. In comparison the CW-IPM machine produced 80.8 to 83% efficiency
from base speed up to 420Hz. Copper loss, which is 2.4 times lower in the S-IPM
machine as compared to the CW-IPM machine, was the main reason for the lower
efficiency of the CW-IPM machine. Besides copper loss, other losses were lower in the
CW-IPM machine, despite the increase in MMF harmonics created by CW.
0102030405060708090
100
0 50 100 150 200
CW-IPM SEG-IPM Poly. (CW-IPM)Effic
ienc
y (%
)
Frequency (Hz)
Base speed
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
202
7.6.4 Magnet Volume Comparison
Due to increasing magnet prices, the amount of magnet used for present-day machine
designs is subjected to constraints. Thus it is of interest to briefly compare the amount
of magnet used in the three machines. Fig. 7.37 compares the volume of magnet per kW
in each machine.
Fig. 7.37 Comparison of Magnet volume per kW the three UNSW IPM machines
This factor can be further improved if the efficiency of the machine is optimised. It
should be noted that, despite similar amount of magnet material used in the CW-IPM
and SEG_IPM, a magnet grade with higher energy density was chosen to compensate
for the loss in reluctance torque. This was done at the expense of higher rotor core
saturation and higher magnet losses.
Chapter 7: Construction and Performance Analysis of the Concentrated Winding IPM Machine Prototype
203
7.7 CONCLUSION
This chapter firstly illustrated the manufacturing process of the prototype CW-IPM
machine. Problems were identified and manufacturing delays were stated as a guide to
improve the manufacturing process of future machines.
The measured performance characteristics of the constructed CW-IPM prototype were
shown and used to verify the performance of the FE model. It was shown that the
measured results from the prototype agreed with the FE results with a high degree of
accuracy. The machine parameters and performance characteristics from the FE model
were compared to two other equally sized DW-IPM machines. These results showed
that the CW-IPM outperformed the two other DW-IPM models in terms of power
density (up to 56% higher), CPSR (greater than 55%) and cogging torque performance
(almost negligible compared to the other two DW machines when compared as a
percentage of total torque). Despite the high copper loss, an 80.8% efficiency was
achieved throughout a 6.2:1 CPSR.
This chapter has verified studies and the design of the CW-IPM machine in this work.
The successful design of the first CW-IPM prototype in our labs has provided a strong
basis and confidence in the FE models.
204
CHAPTER 8EFFICIENCY OPTIMISATION AND SCALABILITY OF THE CONCENTRATED WINDING IPM MACHINE
8.1 INTRODUCTION
In previous chapters, the design and experimental verification of the CW-IPM machine
have been shown. The close correlation of the experimental and FE results gave
confidence for further designs and optimisation. On the basis of the original 800W FE
model, the scalability and efficiency optimization of the CW-IPM machine will be
studied in this chapter.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
205
8.2 EFFICIENCY OPTIMISATION
In chapter 7, it was shown that the CW-IPM machine operated at over 80% efficiency
throughout the CPSR. It was also shown, in chapter 5, that the majority of losses (over
90% of losses) in the CW-IPM machine were due to copper loss. With the main focus
on minimising copper loss, winding modifications and two geometric changes to the
original design are proposed to increase the efficiency of the machine.
8.2.1 Winding Modification
Two hand-winding methods were introduced in chapter 4 – the vertical-fill method and
the horizontal-fill method. The latter method was used in the prototype to reduce the
time and cost required to wind the machine. Here, for the purpose of optimising
efficiency, the vertical slot-fill method is used despite the increase in time and cost of
winding the machine. A 45% slot-fill factor was personally achieved in the prototype
stator. Thus a 45% slot-fill factor will be assumed in this optimisation study. The
difference in winding span and end winding length between these two methods is shown
in fig. 8.1 below:
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
206
(a) Horizontal slot-fill factor
(b) Vertical slot-fill factorFig. 8.1 Axial length comparison between two different slot-fill methods
With the vertical winding method, the overall length per turn is reduced significantly
(from 276mm per turn to 219mm per turn).
8.2.2 Proposed Designs for Efficiency Optimisation
The basic structure of the original design and the volume of the machine are preserved
in this modifications. For convenience, the first modified design will be called CW-IPM
R. For the CW IPM-R, the outer diameter and stack length of the machine is preserved
(fig 8.2a). The rotor diameter of the machine is made smaller to increase the space in the
stator for larger slots, permitting the use of larger conductors, thus lowering copper loss.
The second design will be called CW-IPM S. For the CW-IPM S, the outer diameter is
increased but the stack length is reduced by half in order to keep the volume constant
117mm axial winding length (measured)
21mm average winding span
95mm axial winding length (expected)
14.5mm average winding span
Horizontal slot-fill
Vertical slot-fill
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
207
(fig. 8.2b). The performance constraints are to preserve the same 6:1 CPSR and power
(a) (b)
Fig. 8.2 Designs used for efficiency optimisation indicating outer dimensions
Table 8.1 below, gives key parameters of the two optimised designs. Fig. 8.3a and 8.3b
give performance characteristics for CW-IPM R and CW-IPM S respectively.
(a)80mm
184m
m
130m
m Volume =
1063mm2
(b)40mm
Volume =
1063mm2
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
208
Table 8.1Key Specifications of the Two Optimised CW-IPM Machine Designs
CW-IPM R CW-IPM S
Stator outer diameter 130mm 183.85mm
Rotor outer diameter 36mm 45.7
Airgap length 1mm 1.41mm
Slot opening width 1.2mm 1.2mm
Stack length 80mm 40mm
Rated current 2.2Arms/ph 2.2Arms/ph
Current density 4.25x106A/m2 5.37x106A/m2
Conductor size AWG 20 AWG 21
No. of turns per coil 163turns 127turns
Stator resistance
Mag. remanent flux 1.13T 1.13T
Slot-fill factor 45% 45%
(a) CW-IPM R
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
209
(b) CW-IPM S
Fig. 8.3 Efficiency and power versus speed performance of the two efficiency optimised model
The results firstly show a significant increase in efficiency, thus increasing in output
power for both optimised designs compared to the original design (where 80%
efficiency was achieved). Comparing the two optimised designs, the CW-IPM S
achieves a higher efficiency of 93%, while the CW-IPM R achieves an efficiency of
91% throughout the CPSR. Despite a higher stator resistance, the CW-IPM S achieved a
higher efficiency, due to its capability of producing higher maximum output power. The
maximum output power in the CW-IPM R was limited by the stator and rotor outer
radius.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
210
8.3 SCALABILITY OF THE CONCENTRATED WOUND IPM MACHINE
This scalability study will be divided into two steps: Firstly, the machine parameters of
the original 800W prototype will be varied to study their effects on performance. From
this, a set of general design rules will be derived to ensure that the material properties
are fully utilised and that the machine is operating with the desired field-weakening
capability. Secondly, two up-scaled versions of the 800W prototype will be created. By
implementing the design rules, it will be shown that the predicted power density and
efficiency of the scaled version can be increased significantly. FE analysis will be used
as the basis of this optimisation.
8.3.1 Airgap Length Variation
CW results in increased harmonic and sub-harmonic content in the MMF waveform,
causing localised saturation, thus, affecting the field-weakening performance of the
machine, (as explained in chapter 4). The application of double-layer stator windings as
well as a larger airgap helps lower the effects of these harmonics. From the original
800W, CW-IPM design, the effects of varying airgap length on machine performance
are investigated.
Fig. 8.4 shows the effects on the CPSR and input power, (at base speed), when airgap
length is varied from 0.6 to 1.6mm.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
211
Fig. 8.4. Airgap length versus CPSR and input power
Fig. 8.5 shows the effects on total core loss, (measured at 500Hz), and overall efficiency
with the same variations in airgap length.
Fig. 8.5 Airgap length versus efficiency and core loss
From fig. 8.4, it can be seen that there is an obvious trade-off between input power and
the maximum achievable CPSR of the machine. While power falls almost linearly as
airgap length is increased, the CPSR increases exponentially with airgap length.
Fig. 8.5 shows that having a wider airgap length is beneficial in decreasing overall core
losses. However as airgap length increases, input power naturally decreases as well.
This makes the copper loss more prominent, thus reducing efficiency.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
212
8.3.2 Magnet Strength and Armature Current Variation
In order to achieve optimal field weakening range, both the saliency ratio and
characteristic current values need to be considered. It was explained in chapter 3 that
saliency ratio optimisation was not effective in the CW-IPM machine. Thus,
optimisation of CPSR would be based solely on optimising the characteristic current
condition.
An increase in magnet strength results in an increase in characteristic current, (reiterated
in (8.1)). Therefore when magnets of higher remanent flux density are chosen, the
machine has to be designed with a higher rated current.
= = (8.1)
The geometry of the machine was not altered; hence Ld is kept the same (81.16mH) as
before. Fig. 8.6 shows the variation of magnet remanent flux density and the current
required to satisfy the condition (8.1) and achieve a > 6:1 CPSR. Fig. 8.6 also shows the
lossless power produced at each point. The same core material specifications used for
the prototype CW-IPM machine 35RM300 (with a saturation magnetization of 1.68T at
5000A/m) is used here.
Fig. 8.6 Magnet remanent flux density versus input current and input power
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
213
It can be seen from fig. 8.6 that the current required to achieve equilibrium conditions
and the input power of the machine increase linearly with magnet remanent flux density,
until the core begins to saturate, (after Is=3.3A/Br=1.32T). Fig. 8.7 shows the flux
density plot for the machine under unsaturated, (with 2.3Arms of excitation current), and
saturated conditions, (with 4.1Arms of excitation current).In order to fully exploit the
material properties of the core, the machine should be made to operate at this
equilibrium point (Is=3.3A/Br=1.32T – equilibrium is different or each design).
(a) Unsaturated conditions (b) Saturated conditionsFig. 8.7. Flux density plot of CW-IPM machine under saturated and unsaturated conditions
As magnet remanent flux and stator current are increased, the core loss increases almost
proportionately, while the copper loss increases with the current squared. It is of interest
to know the extent to which losses and efficiency are affected as the power density of
the machine increases with magnet remanent flux density. Fig. 8.8 shows the total
losses and overall efficiency of the machine as magnet remanet flux and current are
increased.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
214
Fig. 8.8 Magnet remanent flux density versus total machine losses and efficiency
This section shows the effects of varying parameters – in particular, airgap length,
magnet remnant flux density and rated current – of the CW-IPM prototype machine
without significant alterations to the machine geometry. The study done here provides
designers with a set of rules to design the CW-IPM for field weakening operation.
These steps are listed as follows:
i. Decide on the desired split-ratio (the ratio of stator inner diameter to outer diameter).
A larger split-ratio would result in higher torque densities but decreased space for
stator slots, hence higher copper loss.
ii. Decide on the airgap length of the machine. A smaller airgap would result in higher
torque density and higher efficiency but a narrower CPSR.
iii. Decide on the required slot area, keeping the core saturation limits in mind. From the
FE model, ensure that the maximum flux density in the tooth and yoke are similar.
iv. Vary slot-opening width to fit desired conductor sizes. Essentially, slot-opening
widths should be kept as small as possible to achieve maximum flux linkage across
the airgap.
v. Determine the optimal equilibrium point of the machine (the point after which current
increases non-linearly with magnet remanent flux density to maintain desired CPSR).
Operating after this point yields a non-linear decrease in efficiency.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
215
8.3.3 Effects of Scaling the Machine Size
The purpose of this section is to study the effects of scaling up the CW-IPM machine
size, as well as to observe the effects of applying the abovementioned design rules to
larger machines. The aim is to achieve over 6:1 CPSR, over 80% efficiency throughout
the CPSR, as well as to fully utilise the steel properties, while operating at the optimal
equilibrium point.
Two designs are created based on the proven 800W model. In the first model, only the
machine outer diameter is increased by a factor of two; the stack length remains the
same as the 800W machine (80mm). In the second model, the outer diameter is three
times that of the 800W machine and the stack length is also increased by two (160mm)
as shown in fig. 8.9.
Fig. 8.9. Comparison between the three machine sizes
The output torque ( ) on machine sizing can be determined by (8.2), where the
output torque/power of the machine increases linearly with the effective stack length
( ) of the machine, and with the square of the machine outer diameter ( ).
Size of the 800W prototype
1st model
2nd model
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
216
= (8.2)
Assuming the torque constant is unaltered and the entire machine is scaled up
proportionately, then the output power for the first and second scaled models should be:
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
218
Fig. 8.10 Input power, output power and efficiency versus speed characteristics of the 5kW design
Fig. 8.11 Input power, output power and efficiency versus speed characteristics of the 30kW design
The results show that the scaled designs were both able to achieve over 8:1 CPSR. The
5kW model achieved over 92.2% efficiency and 30kW model achieved over 95.8%
efficiency throughout the CPSR.
In the optimised designs, the core material operates with higher flux densities, thus core
loss would naturally increase. Despite the increase in core loss, efficiency is increased
due to the significant reduction of copper loss (which contributes to the largest portion
of losses in the machine) with a larger machine outer diameter. Fig. 8.12 shows the loss
breakdown of the three machine models:
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
219
Fig. 8.12. Losses in the three machine sizes as a percentage of total loss
From scaling the size of the machine, the following key points can be noted:
While increasing the machine outer diameter results in an exponential increase
in power, increasing the stack length results only in a linear increase in power, in
compliance to (8.2). Additionally, increasing the machine outer diameter
increases the efficiency of the machine.
Changing the conductor size to achieve the desired rated voltage has no effect on
copper loss and output power, as long as ampere-turns and the slot-fill factor are
unchanged.
Chapter 8: Efficiency Optimisation and Scalability of the Concentrated Winding IPM Machine
220
8.4 CONCLUSION
This chapter has shown that the original 800W FE model can be effectively optimised
to provide high efficiency (over 93% throughout the CPSR), by the reduction of copper
loss. The comparison of two optimized models with the same volume, shows that
having a larger outer diameter and shorter stack length was effective in increasing both
efficiency and torque density of the machine.
The scalability of the machine was also studied in this chapter. Through variation of
machine parameters, a general set of design rules was illustrated to fully utilise material
properties as well as to achieve desired CPSR and power density. By applying these
design rules on two scaled models, power density in the 3.2kW and 14.4kW model can
be optimised to produce 5kW and 30kW respectively. Efficiency of over 92.2% and
95.8% were achieved by each model respectively.
221
CHAPTER 9CONCLUSION AND SUGGESTIONS FOR FUTURE WORK
9.1 CONCLUSION
The work done in this thesis has proven that the fractional-slot CW-IPM machine is a
suitable candidate for field-weakening applications. The constructed double-layer CW-
IPM machine with 14 v-shaped poles and 18-slots showed that a very wide 7.2:1 CPSR
could be achieved. Efficiency of over 80% was achieved over a 6.2:1 CPSR. The CW-
IPM also achieved a higher torque density, and much lower cogging torque compared to
two other equally sized DW-IPM machines.
In chapter 3, open-circuit characteristics of the CW-IPM machine were studied. Several
slot and pole combinations were investigated, and it was shown that the 18-slot, 14-pole
model achieved the highest induced back EMF magnitude, as compared to other 14-pole,
fractional-slot combinations. The calculated winding factor of 0.902 was confirmed by
comparison with an equivalent integral-slot DW machine model. This combination of
slots and poles also resulted in a near-perfectly sinusoidal EMF waveform, and very low
cogging torque magnitude. The comparison of two CW-IPM machines with different
magnet geometries, (rectangular and v-shaped), with the CW-SPM machine, showed
that the machine with rectangular, single-piece/pole magnets was inferior to the CW-
SPM machine in terms of torque density, and field-weakening capability. The CW-IPM
machine with v-shaped magnets, on the other hand, had improved field weakening
capability, but with slightly lower peak torque density as compared to the CW-SPM
machine, due to saturation effects.
Chapter 9: Conclusion and Suggestions for Future Work
222
In chapter 4, the CPSR of the 14-pole, 18-slot, v-shaped IPM design was optimised by
v-angle and airgap variations. It was shown that by satisfying the characteristic current
equilibrium conditions by variation of specific machine, the optimal CPSR could be
achieved. From optimisation done in this chapter, the final design was specified.
Parameters and performance characteristics of the final model were determined by FE
analysis.
In chapter 5, a detailed study of losses in the CW-IPM machine was done. It showed
how the increase in harmonics resulting from CW affected frequency related losses.
Electromagnetic losses, (consisting of core, magnet and I2R losses), and mechanical
losses, (consisting of bearing and windage losses), were studied and quantified. This
study highlighted the importance of choosing thin silicon steel laminations for the rotor
and stator core material. It also indicated that magnet losses in the CW-IPM were very
low, and that loss reduction by magnet segmentation was not as effective as magnet
segmentation in the equivalent CW-SPM machine. It was shown that the calculated
mechanical losses were low even at maximum operating speed of the machine. Chapter
5 also showed that copper loss made up for the majority of loss (> 90%) in the machine;
if the efficiency of the machine were to be improved, the main focus should be on
minimising copper loss.
Chapter 6 presented the control methodology of the CW-IPM machine. It showed the
voltage and current limits that the drive is subjected to by use of the circle diagrams. It
was shown that the widely-used trajectories calculated by Morimoto’s equations led to
the ‘over-weakening’ of the magnet fields. Thus a manually obtained current
trajectory was used in the control of the prototype CW-IPM machine.
Chapter 9: Conclusion and Suggestions for Future Work
223
Chapter 7 firstly illustrated the manufacturing process of the prototype CW-IPM
machine. Problems identified and manufacturing delays were stated as a guide to
improve the manufacturing process of future machines. The measured performance
characteristics of the constructed CW-IPM prototype was shown and used to verify the
performance of the FE model. It was shown that the measured results from the
prototype agreed with the FE results with a high degree of accuracy. The machine
parameters and performance characteristics from the FE model were compared to two
other equally sized DW-IPM machines. This comparison showed that the CW-IPM
outperformed the two other DW-IPM models in terms of power density, (up to 56%
higher), CPSR, (7.2:1 as opposed to a maximum of 4:1 achieved by the DW S-IPM), as
well as cogging torque performance, (0.3%(p-p) as compared 8.1%(p-p) and 24.2%(p-p)
achieved by the other two machines – as a percentage of total torque). Efficiency was
slightly lower in the CW-IPM machine, (80.8 to 83% over a 6.2: CPSR), as compared
with the DW S-IPM machine, (84 to 85% over a 4:1 CPSR).
With the successful design of the CW-IPM, achieving desired performance
characteristics, as well as the confidence gained from the close correlation of FE and
measured results, an efficiency optimization and scalability study was performed.
Chapter 8 showed that the original 800W FE model could be effectively optimised to
provide high efficiency, (up to 93%). Two optimised models with the same volume
were compared. It was shown that the model with larger outer diameter and shorter
stack length achieved a higher efficiency, as compared to one with the same dimensions
as the prototype machine, but with a smaller rotor. Chapter 8 also studied the scalability
of the machine. Through variation of machine parameters, a general set of design rules
were illustrated. This rules aid in designing the CW-IPM machine to achieve the desired
Chapter 9: Conclusion and Suggestions for Future Work
224
CPSR, and power density, as well as to fully utilise material properties. Two scaled up
models of the CW-IPM machine were created. Based on theoretical calculations, the
presumed output power of the two scaled models were 3.2kW and 14.4kW. However,
by application of the abovementioned design rules, the optimised models each produced
5kW and 30kW respectively.
The successful design, construction of the first CW prototype in our labs has provided a
strong basis, and confidence, in the field-weakening capability of the CW-IPM machine.
It has also created opportunities for future work to be done in this area, such as
performance optimisation, control implementation, as well as to determine the
suitability of the CW-IPM for various industrial applications.
Chapter 9: Conclusion and Suggestions for Future Work
225
9.2 SUGGESTION FOR FUTURE WORK
Due to the lack of a suitable loading generator, (capable of producing required torque
over the entire speed range), sensing equipment (torque transducer), as well as a time
constrain, the torque and field-weakening capability of the machine could not tested to
its maximum limits. With further testing using appropriate equipment, increased power
and a wider CPSR can be achieved. This work will soon be carried by a future student,
when the equipment becomes available. From simulations, it is noted that the saturation
in the steel is low (typically lower than 1.6T). This indicates that the tooth width and
yoke length can be made smaller to provide space for increased winding size- hence
lower copper loss.
In this thesis, it was shown that commonly used vector control techniques could not be
applied to the CW-IPM machine, thus a manually obtained id and iq trajectory was used
to operate the machine at optimal power throughout the speed range. With this method,
dynamic performance of this machine could not be fully tested. Therefore, a proper
control technique has to be implemented to achieve desired dynamic performance.
Subsequently, sensorless control can also be implemented. Above base speed (60rads),
common DTC control methods should work. However, due to the low saliency ratio of
1.12, achieved by the CW-IPM machine, the closed-loop observer and high-frequency
signal injection methods might not work. If a higher saliency ratio is required, the
machine can be redesigned based on rotor magnet geometry variation techniques
illustrated in appendix B, as well as the investigation of end-winding effects,
(contributing the q-axis inductance), can be done to further increase the saliency ratio.
Chapter 9: Conclusion and Suggestions for Future Work
226
Additional focus can be placed upon trying to improve the torque/power density of the
CW-IPM machine. Also, as the cost of rare earth magnets escalating, it would be
beneficial to design machines with less magnet material (NdFeB in particular).
Lastly, with a high number of poles, as well as the capability of achieving high
efficiencies, CW-IPM may by suitable for low speed applications, such as in a wind
generator.
227
REFERENCES[1] H. W. Beaty and J. L. Kirtley, Electric motor handbook, 1998.[2] J. M. Miller, Propulsion Systems for Hybrid Vehicles. United Kingdom: The
Institution of Engineering and Technology, 2004.[3] D. Hans[man, Brushless Permanent Magnet Motor Design, 2nd ed.: The
Writers' Collective, 2003.[4] Z. Q. Zhu and D. Howe, "Halbach permanent magnet machines and applications:
a review," Electric Power Applications, IEE Proceedings -, vol. 148, pp. 299-308, 2001.
[5] A. M. El-Refaie, "High Speed Operation of Permanent Magnet Machines," PhD Thesis PhD, University of Wisconsin-Madsion, 2005.
[6] T. Yamakawa, S. Wakao, K. Kondo, and T. Yoneyama, "A new flux weakening operation of interior permanent magnet synchronous motors for railway vehicle traction," in European Conference on Power Electronics and Applications 2005, pp. 1-6.
[7] N. Bianchi, S. Bolognani, and B. Ruzojcic, "Design of a 1000 HP Permanent Magnet synchronous motor for ship propulsion," in 13th European Conference on Power Electronics and Applications, EPE, 2009, pp. 1-8.
[8] T. M. Jahns and R. C. Van Nocker, "High-performance EHA controls using an interior permanent magnet motor," Aerospace and Electronic Systems, IEEE Transactions on, vol. 26, pp. 534-542, 1990.
[9] S. Suzuki, T. Katane, H. Saotome, and O. Saito, "Electric power-generating system using magnetic coupling for deeply implanted medical electronic devices," Magnetics, IEEE Transactions on, vol. 38, pp. 3006-3008, 2002.
[10] J. Zhang, Z. Chen, and M. Cheng, "Design and comparison of a novel stator interior permanent magnet generator for direct-drive wind turbines," Renewable Power Generation, IET, vol. 1, pp. 203-210, 2007.
[11] P. Kyoung-Jin, J. Dae-Sung, C. Hong-Soon, B. Jae-Nam, and L. Ju, "A study on the reducing of cogging torque of IPMSM," in International Conference on Electrical Machines and Systems, ICEMS, 2008, pp. 3137-3141.
[12] W. L. Soong and N. Ertugrul, "Field-weakening performance of interior permanent-magnet motors," Industry Applications, IEEE Transactions on, vol. 38, pp. 1251-1258, 2002.
[13] R. Dutta and M. F. Rahman, "Design and Analysis of an Interior Permanent Magnet (IPM) Machine With Very Wide Constant Power Operation Range," Energy Conversion, IEEE Transaction on, vol. 23, pp. 25-33, 2008.
[14] Toyota, "Toyota hybrid system - THS II," Toyota Motor Corporation, Public Affairs Division, Tokyo, Japan2003.
[15] L. Chong, R. Dutta, and M. F. Rahman, "Electromagnetic Losses in a 1kW Concentric Wound IPM Machine for Field Weakening Applications," Journal of Applied Superconductivity and Electromagnetism, vol. 1, pp. 95-100, 2010.
[16] J. Cros and P. Viarouge, "Synthesis of high performance PM motors with concentrated windings," Energy Conversion, IEEE Transaction on, vol. 17, pp. 248-253, 2002.
References
228
[17] F. Magnussen and C. Sadarangani, "Winding factors and Joule losses of permanent magnet machines with concentrated windings," in Electric Machines and Drives Conference, IEMDC'03. IEEE International, 2003, pp. 333-339 vol.1.
[18] A. M. El-Refaie and T. M. Jahns, "Impact of Winding Layer Number and Magnet Type on Synchronous Surface PM Machines Designed for Wide Constant-Power Speed Range Operation," Energy Conversion, IEEE Transactions on, vol. 23, pp. 53-60, 2008.
[19] D. Ishak, Z. Q. Zhu, and D. Howe, "Permanent-magnet brushless machines with unequal tooth widths and similar slot and pole numbers," Industry Applications, IEEE Transactions on, vol. 41, pp. 584-590, 2005.
[20] N. Bianchi, S. Bolognani, M. D. Pre, and G. A. G. G. Grezzani, "Design considerations for fractional-slot winding configurations of synchronous machines," Industry Applications, IEEE Transactions on, vol. 42, pp. 997-1006, 2006.
[21] F. Magnussen and H. Lendenmann, "Parasitic effects in PM machines with concentrated windings," in Industry Applications Conference, Fourtieth IAS Annual Meeting. , 2005, pp. 1044-1049 Vol. 2.
[22] J. Wang, Z. P. Xia, D. Howe, and S. A. Long, "Comparative Study of 3-Phase Permanent Magnet Brushless Machines with Concentrated, Distributed and Modular Windings," in The 3rd IET International Conference on Power Electronics, Machines and Drives, 2006, pp. 489-493.
[23] R. Wrobel, P. H. Mellor, N. McNeill, and D. A. Staton, "Thermal Performance of an Open-Slot Modular-Wound Machine With External Rotor," Energy Conversion, IEEE Transactions on, vol. 25, pp. 403-411, 2010.
[24] C. R. Steen, "Direct axis aiding permanent magnets for a laminated synchronous motor rotor," US Patent 4139790, 1979.
[25] V. B. Honsinger, "The Fields and Parameters of Interior Type AC Permanent Magnet Machines," Power Apparatus and Systems, IEEE Transactions on, vol. PAS-101, pp. 867-876, 1982.
[26] M. Rahman, T. Little, and G. Slemon, "Analytical models for interior-type permanent magnet synchronous motors," Magnetics, IEEE Transactions on, vol. 21, pp. 1741-1743, 1985.
[27] A. Consoli and G. Renna, "Interior type permanent magnet synchronous motor analysis by equivalent circuits," Energy Conversion, IEEE Transactions on, vol. 4, pp. 681-689, 1989.
[28] B. J. Chalmers, S. A. Hamed, and G. D. Baines, "Parameters and performance of a high-field permanent-magnet synchronous motor for variable-frequency operation," Electric Power Applications, IEE Proceedings B, vol. 132, pp. 117-124, 1985.
[29] R. Schifer and T. A. Lipo, "Core loss in buried magnet permanent magnet synchronous motors," Energy Conversion, IEEE Transactions on, vol. 4, pp. 279-284, 1989.
[30] T. M. Jahns, G. B. Kliman, and T. W. Neumann, "Interior Permanent-Magnet Synchronous Motors for Adjustable-Speed Drives," Industry Applications, IEEE Transactions on, vol. IA-22, pp. 738-747, 1986.
References
229
[31] T. M. Jahns, "Flux-Weakening Regime Operation of an Interior Permanent-Magnet Synchronous Motor Drive," Industry Applications, IEEE Transactions on, vol. IA-23, pp. 681-689, 1987.
[32] W. L. Soong and T. J. E. Miller, "Field-weakening performance of brushless synchronous AC motor drives," Electric Power Applications, IEE Proceedings -,vol. 141, pp. 331-340, 1994.
[33] W. L. Soong, "Design and Modelling of Axially-Laminated Interior Permanent Magnet Motor Drives for Field Weakening Applications," PhD, Deparmanent of Electronics and Electrical Engineering, The University of Glasgow, 1993.
[34] W. L. Soong, D. A. Staton, and T. J. E. Miller, "Design of a new axially-laminated interior permanent magnet motor," Industry Applications, IEEE Transactions on, vol. 31, pp. 358-367, 1995.
[35] Y. Honda, T. Higaki, S. Morimoto, and Y. Takeda, "Rotor design optimisation of a multi-layer interior permanent-magnet synchronous motor," Electric Power Applications, IEE Proceedings -, vol. 145, pp. 119-124, 1998.
[36] Y. Honda, T. Nakamura, T. Higaki, and Y. A. Takeda, "Motor design considerations and test results of an interior permanent magnet synchronous motor for electric vehicles," in Industry Applications Conference, Thirty-Second IAS Annual Meeting, IAS '97., Conference Record of the IEEE, 1997, pp. 75-82vol.1.
[37] L. Jolly, M. A. Jabbar, and Q. Liu, "Optimization of the constant power speed range of a saturated permanent-magnet synchronous motor," IndustryApplications, IEEE Transactions on, vol. 42, pp. 1024-1030, 2006.
[38] A. Fratta, A. Vagati, and F. Villata, "On the evolution of AC machines for spindle drive applications," Industry Applications, IEEE Transactions on, vol. 28, pp. 1081-1086, 1992.
[39] S. Morimoto, Y. Takeda, and T. Hirasa, "Current phase control methods for permanent magnet synchronous motors," Power Electronics, IEEE Transactions on, vol. 5, pp. 133-139, 1990.
[40] Z. Q. Zhu, Y. S. Chen, and D. Howe, "Iron loss in permanent-magnet brushless AC machines under maximum torque per ampere and flux weakening control," Magnetics, IEEE Transactions on, vol. 38, pp. 3285-3287, 2002.
[41] K. Kyung-Tae, K. Kwang-Suk, H. Sang-Moon, K. Tae-Jong, and J. Yoong-Ho, "Comparison of magnetic forces for IPM and SPM motor with rotor eccentricity," Magnetics, IEEE Transactions on, vol. 37, pp. 3448-3451, 2001.
[42] L. Jung Ho, K. Jung Chul, and H. Dong Seek, "Effect analysis of magnet on Ld and Lq inductance of permanent magnet assisted synchronous reluctance motor using finite element method," Magnetics, IEEE Transactions on, vol. 35, pp. 1199-1202, 1999.
[43] S. Morimoto, M. Sanada, and Y. Takeda, "Wide-speed operation of interior permanent magnet synchronous motors with high-performance current regulator," Industry Applications, IEEE Transactions on, vol. 30, pp. 920-926, 1994.
[44] M. N. Uddin, T. S. Radwan, G. H. George, and M. A. Rahman, "Performance of current controllers for VSI-fed IPMSM drive," Industry Applications, IEEE Transactions on, vol. 36, pp. 1531-1538, 2000.
[45] M. F. Rahman, L. Zhong, and L. Khiang Wee, "A direct torque-controlled interior permanent magnet synchronous motor drive incorporating field
References
230
weakening," Industry Applications, IEEE Transactions on, vol. 34, pp. 1246-1253, 1998.
[46] S. Ogasawara and H. Akagi, "Implementation and position control performance of a position-sensorless IPM motor drive system based on magnetic saliency," Industry Applications, IEEE Transactions on, vol. 34, pp. 806-812, 1998.
[47] S. Ogasawara and H. Akagi, "An approach to real-time position estimation at zero and low speed for a PM motor based on saliency," Industry Applications, IEEE Transactions on, vol. 34, pp. 163-168, 1998.
[48] M. J. Corley and R. D. Lorenz, "Rotor position and velocity estimation for a salient-pole permanent magnet synchronous machine at standstill and high speeds," Industry Applications, IEEE Transactions on, vol. 34, pp. 784-789,1998.
[49] M. E. Haque, Z. Limin, and M. F. Rahman, "A sensorless initial rotor position estimation scheme for a direct torque controlled interior permanent magnet synchronous motor drive," Power Electronics, IEEE Transactions on, vol. 18, pp. 1376-1383, 2003.
[50] P. Sergeant, F. De Belie, and J. Melkebeek, "Effect of Rotor Geometry and Magnetic Saturation in Sensorless Control of PM Synchronous Machines," Magnetics, IEEE Transactions on, vol. 45, pp. 1756-1759, 2009.
[51] K. Gyu-Hong, H. Jung-Pyo, K. Gyu-Tak, and P. Jung-Woo, "Improved parameter modeling of interior permanent magnet synchronous motor based on finite element analysis," Magnetics, IEEE Transactions on, vol. 36, pp. 1867-1870, 2000.
[52] N. Bianchi and S. Bolognani, "Influence of Rotor Geometry of an IPM Motor on Sensorless Control Feasibility," Industry Applications, IEEE Transactions on, vol. 43, pp. 87-96, 2007.
[53] N. Bianchi, S. Bolognani, J. Ji-Hoon, and S. Seung-Ki, "Comparison of PM Motor Structures and Sensorless Control Techniques for Zero-Speed Rotor Position Detection," Power Electronics, IEEE Transactions on, vol. 22, pp. 2466-2475, 2007.
[54] N. Bianchi, S. Bolognani, and M. Zigliotto, "Design Hints of an IPM Synchronous Motor for an Effective Position Sensorless Control," in IEEE 36th Power Electronics Specialists Conference, PESC, 2005, pp. 1560-1566.
[55] W. Shanshan, D. D. Reigosa, Y. Shibukawa, M. A. Leetmaa, R. D. Lorenz, and L. Yongdong, "Interior Permanent-Magnet Synchronous Motor Design for Improving Self-Sensing Performance at Very Low Speed," Industry Applications, IEEE Transactions on, vol. 45, pp. 1939-1946, 2009.
[56] E. R. Braga Filho, A. M. N. Lima, and T. S. Araujo, "Reducing cogging torque in interior permanent magnet machines without skewing," Magnetics, IEEE Transactions on, vol. 34, pp. 3652-3655, 1998.
[57] Y. Fujishima, S. Wakao, M. Kondo, and N. Terauchi, "An optimal design of interior permanent magnet synchronous motor for the next generation commuter train," Applied Superconductivity, IEEE Transactions on, vol. 14, pp. 1902-1905,2004.
[58] Y. Kano and N. Matsui, "A Design Approach for Direct-Drive Permanent-Magnet Motors," Industry Applications, IEEE Transactions on, vol. 44, pp. 543-554, 2008.
References
231
[59] K. Yamazaki, "Torque and efficiency calculation of an interior permanent magnet motor considering harmonic iron losses of both the stator and rotor," Magnetics, IEEE Transactions on, vol. 39, pp. 1460-1463, 2003.
[60] S. Dong-Joon, C. Dong-Hyeok, C. Jang-Sung, J. Hyun-Kyo, and C. Tae-Kyoung, "Efficiency optimization of interior permanent magnet synchronous motor using genetic algorithms," Magnetics, IEEE Transactions on, vol. 33, pp. 1880-1883,1997.
[61] D.-H. Kim, I.-H. Park, J.-H. Lee, and C.-E. Kim, "Optimal Shape Design of Iron Core to Reduce Cogging Torque of IPM Motor," Magnetics, IEEE Transactions on, vol. 39, pp. 1456-1459, 2003.
[62] K. Ki-Chan, K. Dae-Hyun, H. Jung-Pyo, and L. Ju, "A Study on the Characteristics Due to Pole-Arc to Pole-Pitch Ratio and Saliency to Improve Torque Performance of IPMSM," Magnetics, IEEE Transactions on, vol. 43, pp. 2516-2518, 2007.
[63] L. Parsa and H. Lei, "Interior Permanent Magnet Motors With Reduced Torque Pulsation," Industrial Electronics, IEEE Transactions on, vol. 55, pp. 602-609,2008.
[64] A. Kioumarsi, M. Moallem, and B. Fahimi, "Mitigation of Torque Ripple in Interior Permanent Magnet Motors by Optimal Shape Design," Magnetics, IEEE Transactions on, vol. 42, pp. 3706-3711, 2006.
[65] K. Gyu-Hong, S. Young-Dae, K. Gyu-Tak, and H. Jin, "A Novel Cogging Torque Reduction Method for Interior-Type Permanent-Magnet Motor," Industry Applications, IEEE Transactions on, vol. 45, pp. 161-167, 2009.
[66] S.-H. Han, T. M. Jahns, W. L. Soong, M. K. Guven, and M. S. Illindala, "Torque Ripple Reduction in Interior Permanent Magnet Synchronous Machines Using Stators With Odd Number of Slots Per Pole Pair," Energy Conversion, IEEETransactions on, vol. 25, pp. 118-127.
[67] M. Sanada, K. Hiramoto, S. Morimoto, and Y. Takeda, "Torque ripple improvement for synchronous reluctance motor using an asymmetric flux barrier arrangement," Industry Applications, IEEE Transactions on, vol. 40, pp. 1076-1082, 2004.
[68] F. Liang, K. Sung-Il, O. K. Soon, and H. Jung-Pyo, "Novel Double-Barrier Rotor Designs in Interior-PM<newline/> Motor for Reducing Torque Pulsation," Magnetics, IEEE Transactions on, vol. 46, pp. 2183-2186.
[69] S.-I. Kim, J.-Y. Lee, Y.-K. Kim, J.-P. Hong, Y. Hur, and Y.-H. Jung, "Optimization for reduction of torque ripple in interior permanent magnet motor by using the Taguchi method," Magnetics, IEEE Transactions on, vol. 41, pp. 1796-1799, 2005.
[70] Y. Kawase, T. Ota, and H. Fukunaga, "3-D eddy current analysis in permanent magnet of interior permanent magnet motors," Magnetics, IEEE Transactions on, vol. 36, pp. 1863-1866, 2000.
[71] V. Zivotic-Kukolj, W. L. Soong, and N. Ertugrul, "Iron Loss Reduction in an Interior PM Automotive Alternator," Industry Applications, IEEE Transactions on, vol. 42, pp. 1478-1486, 2006.
[72] D. M. Ionel, M. Popescu, M. I. McGilp, T. J. E. Miller, S. J. Dellinger, and R. J. Heideman, "Computation of Core Losses in Electrical Machines Using Improved Models for Laminated Steel," Industry Applications, IEEE Transactions on, vol. 43, pp. 1554-1564, 2007.
References
232
[73] K. Yamazaki and Y. Seto, "Iron loss analysis of interior permanent-magnet synchronous motors-variation of main loss factors due to driving condition," Industry Applications, IEEE Transactions on, vol. 42, pp. 1045-1052, 2006.
[74] A. Wang, H. Li, and C.-T. Liu, "On the Material and Temperature Impacts of Interior Permanent Magnet Machine for Electric Vehicle Applications," Magnetics, IEEE Transactions on, vol. 44, pp. 4329-4332, 2008.
[75] K. Yamazaki and Y. Kanou, "Rotor Loss Analysis of Interior Permanent Magnet Motors Using Combination of 2-D and 3-D Finite Element Method," Magnetics, IEEE Transactions on, vol. 45, pp. 1772-1775, 2009.
[76] T. Okitsu, D. Matsuhashi, and K. Muramatsu, "Method for Evaluating the Eddy Current Loss of a Permanent Magnet in a PM Motor Driven by an Inverter Power Supply Using Coupled 2-D and 3-D Finite Element Analyses," Magnetics, IEEE Transactions on, vol. 45, pp. 4574-4577, 2009.
[77] J.-H. Seo, S.-Y. Kwak, S.-Y. Jung, C.-G. Lee, T.-K. Chung, and H.-K. Jung, "A Research on Iron Loss of IPMSM With a Fractional Number of Slot Per Pole," Magnetics, IEEE Transactions on, vol. 45, pp. 1824-1827, 2009.
[78] K. Yamazaki and H. Ishigami, "Rotor-Shape Optimization of Interior-Permanent-Magnet Motors to Reduce Harmonic Iron Losses," Industrial Electronics, IEEE Transactions on, vol. 57, pp. 61-69.
[79] K. Yamazaki and A. Abe, "Loss Investigation of Interior Permanent-Magnet Motors Considering Carrier Harmonics and Magnet Eddy Currents," Industry Applications, IEEE Transactions on, vol. 45, pp. 659-665, 2009.
[80] H. Seok-Hee, W. L. Soong, T. M. Jahns, M. K. Guven, and M. S. Illindala, "Reducing Harmonic Eddy-Current Losses in the Stator Teeth of Interior Permanent Magnet Synchronous Machines During Flux Weakening," Energy Conversion, IEEE Transactions on, vol. 25, pp. 441-449.
[81] K. J. Tseng and S. B. Wee, "Analysis of flux distribution and core losses in interior permanent magnet motor," Energy Conversion, IEEE Transactions on, vol. 14, pp. 969-975, 1999.
[82] B. Stumberger, A. Hamler, and B. Hribernik, "Analysis of iron loss in interior permanent magnet synchronous motor over a wide-speed range of constant output power operation," Magnetics, IEEE Transactions on, vol. 36, pp. 1846-1849, 2000.
[83] L. Ma, M. Sanada, S. Morimoto, and Y. Takeda, "Iron loss prediction considering the rotational field and flux density harmonics in IPMSM and SynRM," Electric Power Applications, IEE Proceedings -, vol. 150, pp. 747-751, 2003.
[84] M. Barcaro, N. Bianchi, and F. Magnussen, "Rotor Flux-Barrier Geometry Design to Reduce Stator Iron Losses in Synchronous IPM Motors Under FW Operations," Industry Applications, IEEE Transactions on, vol. 46, pp. 1950-1958, 2010.
[85] F. Fernandez-Bernal, A. Garcia-Cerrada, and R. Faure, "Determination of parameters in interior permanent-magnet synchronous motors with iron losses without torque measurement," Industry Applications, IEEE Transactions on, vol. 37, pp. 1265-1272, 2001.
[86] J.-Y. Lee, S.-H. Lee, G.-H. Lee, J.-P. Hong, and J. Hur, "Determination of parameters considering magnetic nonlinearity in an interior permanent magnet
References
233
synchronous motor," Magnetics, IEEE Transactions on, vol. 42, pp. 1303-1306, 2006.
[87] Z. Li, S. Z. Jiang, Z. Q. Zhu, and C. C. Chan, "Analytical Modeling of Open-Circuit Air-Gap Field Distributions in Multisegment and Multilayer Interior Permanent-Magnet Machines," Magnetics, IEEE Transactions on, vol. 45, pp. 3121-3130, 2009.
[88] E. C. Lovelace, T. M. Jahns, T. A. Keim, and J. H. Lang, "Mechanical design considerations for conventionally laminated, high-speed, interior PM synchronous machine rotors," Industry Applications, IEEE Transactions on, vol. 40, pp. 806-812, 2004.
[89] G.-B. Hwang, S.-M. Hwang, W.-B. Jung, B.-S. Kang, I. C. Hwang, and C.-U. Kim, "Analysis of magnetic forces in magnetically saturated permanent magnet motors by considering mechanical and magnetic coupling effects," Journal of Applied Physics, vol. 91, pp. 6976-6978, 2002.
[90] A. M. El-Refaie, N. C. Harris, T. M. Jahns, and K. M. Rahman, "Thermal analysis of multibarrier interior PM synchronous Machine using lumped parameter model," Energy Conversion, IEEE Transactions on, vol. 19, pp. 303-309, 2004.
[91] B. A. Welchko, T. M. Jahns, and S. Hiti, "IPM synchronous machine drive response to a single-phase open circuit fault," Power Electronics, IEEE Transactions on, vol. 17, pp. 764-771, 2002.
[92] N. Bianchi, S. Bolognani, M. Zigliotto, and M. Zordan, "Innovative remedial strategies for inverter faults in IPM synchronous motor drives," Energy Conversion, IEEE Transactions on, vol. 18, pp. 306-314, 2003.
[93] J. Cros, P. Viarouge, and C. Gelinas, "Design of PM brushless motors using iron-resin composites for automotive applications," in Industry Applications Conference, 1998. Thirty-Third IAS Annual Meeting. The 1998 IEEE, 1998, pp. 5-11 vol.1.
[94] M. Nakano and H. Kometani, "A study on eddy-current losses in rotors of surface permanent magnet synchronous machines," in Industry Applications Conference, 2004. 39th IAS Annual Meeting. Conference Record of the 2004 IEEE, 2004, pp. 1696-1702 vol.3.
[95] P. B. Reddy, T. M. Jahns, and A. M. El-Refaie, "Impact of Winding Layer Number and Slot/Pole Combination on AC Armature Losses of Synchronous Surface PM Machines Designed for Wide Constant-Power Speed Range Operation," in Industry Applications Society Annual Meeting, 2008. IAS '08. IEEE, 2008, pp. 1-8.
[96] H. Jussila, P. Salminen, and J. Pyrhonen, "Losses of a Permanent Magnet Synchronous Motor with Concentrated Windings," in Power Electronics, Machines and Drives, 2006. The 3rd IET International Conference on, 2006, pp. 207-211.
[97] H. Jussila, P. Salminen, M. Niemela, and J. Pyrhonen, "Guidelines for Designing Concentrated Winding Fractional Slot Permanent Magnet Machines," in Power Engineering, Energy and Electrical Drives, 2007. POWERENG 2007. International Conference on, 2007, pp. 191-194.
[98] A. M. El-Refaie and T. M. Jahns, "Scalability of surface PM Machines with concentrated windings designed to achieve wide speed ranges of constant-power
References
234
operation," Energy Conversion, IEEE Transaction on, vol. 21, pp. 362-369, 2006.
[99] Y. Xu and Y. Sun, "Fractional-slot low speed large torque permanent magnet brushless motors," in 4th IEEE Conference on Industrial Electronics and Applications, ICIEA . , 2009, pp. 3565-3569.
[100] P. Salminen, T. Jokinen, and J. Pyrhonen, "Pull-out torque of fractional-slot PM-motors with concentrated winding," Electric Power Applications, IEE Proceedings -, vol. 152, pp. 1440-1444, 2005.
[101] P. Salminen, M. Niemela, J. Pyrhonen, and J. A. M. J. Mantere, "High-torque low-torque-ripple fractional-slot PM-motors," in Electric Machines and Drives, 2005 IEEE International Conference on, 2005, pp. 144-148.
[102] D. Gerada, A. Mebarki, and C. Gerada, "Optimal design of a high speed concentrated wound PMSM," in International Conference on Electrical Machines and Systems, ICEMS, 2009, pp. 1-6.
[103] H. Murakami, H. Kataoka, Y. Honda, S. Morimoto, and Y. Takeda, "Highly efficient brushless motor design for an air-conditioner of the next generation 42 V vehicle," in Industry Applications Conference, 2001. Thirty-Sixth IAS Annual Meeting. Conference Record of the 2001 IEEE, 2001, pp. 461-466 vol.1.
[104] A. G. Jack, B. C. Mecrow, P. G. Dickinson, D. Stephenson, J. S. Burdess, N. Fawcett, and J. T. Evans, "Permanent-magnet machines with powdered iron cores and prepressed windings," Industry Applications, IEEE Transactions on, vol. 36, pp. 1077-1084, 2000.
[105] H. Akita, Y. Nakahara, N. Miyake, and T. Oikawa, "New core structure and manufacturing method for high efficiency of permanent magnet motors," in Industry Applications Conference, 2003. 38th IAS Annual Meeting. Conference Record of the, 2003, pp. 367-372 vol.1.
[106] H. Akita, Y. Nakahara, N. Miyake, and T. Oikawa, "A new core," IEEE Industry Applications Magazine, vol. 11, pp. 38-43, 2005.
[107] W. Wang, M. E. Carrier, P. G. Michaels, J. S. Rose, and J. J. Jurkowski, "End cap for interconnecting winding coils of a segmented stator to reduce phase-on-phase conditions and associated methods," US Patent 7116023, 2006.
[108] S. Wainio, J. Ahlvik, M. Poyhonen, and I. T. F. I. Ikonen, "Segmented stator and the winding of it," US Patent 20040150287, 2004.
[109] K. A. Sheeran, P. Rassoolkhani, and P. G. Michaels, "Segmented Stator with Improved Handling and Winding Characteristics," US Patent 20080129142, 2008.
[110] Y. Mitsui and K. Fujita, "Method of Manufacturing Laminated Core," US Patent 20100192357, 2010.
[111] G. E. Horst and K. I. Hoemann, "Reduced coil segmented stator," US Patent 7122933, 2006.
[112] Y. Asano, Y. Honda, H. Murakami, Y. Takeda, and S. Morimoto, "Novel noise improvement technique for a PMSM with concentrated winding," in Power Conversion Conference, 2002. PCC Osaka 2002. Proceedings of the, 2002, pp. 460-465 vol.2.
[113] F. Magnussen, P. Thelin, and C. Sadarangani, "Performance evaluation of permanent magnet synchronous machines with concentrated and distributed windings including the effect of field-weakening," in Power Electronics,
References
235
Machines and Drives, 2004. (PEMD 2004). Second International Conference on (Conf. Publ. No. 498), 2004, pp. 679-685 Vol.2.
[114] S. O. Kwon, K. Sung-Il, Z. Peng, and H. Jung-Pyo, "Performance comparison of IPMSM with distributed and concentrated windings," in Industry Applications Conference, 41st IAS Annual Meeting. Conference Record of the IEEE, 2006, pp. 1984-1988.
[115] D. Ishak, Z. Q. Zhu, and D. Howe, "Comparison of PM brushless motors, having either all teeth or alternate teeth wound," Energy Conversion, IEEE Transactions on, vol. 21, pp. 95-103, 2006.
[116] A. M. El-Refaie, Z. Q. Zhu, T. M. Jahns, and D. Howe, "Winding Inductances of Fractional Slot Surface-Mounted Permanent Magnet Brushless Machines," in Industry Applications Society Annual Meeting, 2008. IAS '08. IEEE, 2008, pp. 1-8.
[117] M. T. Abolhassani and H. A. Toliyat, "Fault tolerant permanent magnet motor drives for electric vehicles," in Electric Machines and Drives Conference, 2009. IEMDC '09. IEEE International, 2009, pp. 1146-1152.
[118] M. R. Shah, A. M. El-Refaie, and K. Sivasubramaniam, "Analysis of turn-to-turn faults in surface PM machines with multi-layer fractional-slot concentrated windings," in Electrical Machines, 2008. ICEM 2008. 18th International Conference on, 2008, pp. 1-4.
[119] P. M. Lindh, H. K. Jussila, M. Niemela, A. Parviainen, and J. Pyrhonen, "Comparison of Concentrated Winding Permanent Magnet Motors With Embedded and Surface-Mounted Rotor Magnets," Magnetics, IEEE Transactions on, vol. 45, pp. 2085-2089, 2009.
[120] P. Salminen, J. Pyrhonen, H. Jussila, and M. Niemela, "Concentrated Wound Permanent Magnet Machines with Different Rotor Designs," in Power Engineering, Energy and Electrical Drives, 2007. POWERENG 2007. International Conference on, 2007, pp. 514-517.
[121] S. H. Han, T. M. Jahns, and Z. Q. Zhu, "Analysis of Rotor Core Eddy-Current Losses in Interior Permanent Magnet Synchronous Machines," in Industry Applications Society Annual Meeting, IAS '08. IEEE, 2008, pp. 1-8.
[122] Y. Kawaguchi, T. Sato, I. Miki, and M. Nakamura, "A reduction method of cogging torque for IPMSM," in Electrical Machines and Systems, 2005. ICEMS 2005. Proceedings of the Eighth International Conference on, 2005, pp. 248-250Vol. 1.
[123] Y. Kano, T. Terahai, T. Kosaka, N. Matsui, and T. Nakanishi, "A new flux-barrier design of torque ripple reduction in saliency-based sensorless drive IPM motors for general industrial applications," in Energy Conversion Congress and Exposition, 2009. ECCE 2009. IEEE, 2009, pp. 1939-1945.
[124] Y. Kano, T. Kosaka, N. Matsui, and T. Nakanishi, "A new technique of torque ripple reduction in saliency-based sensorless drive IPM motors for general industrial applications," in 13th European Conference on Power Electronics and Applications, EPE '09. , 2009, pp. 1-10.
[125] S.-I. Kim, J.-H. Bhan, J.-P. Hong, and K.-C. Lim, "Optimization Technique for Improving Torque Performance of Concentrated Winding Interior PM Synchronous Motor with Wide Speed Range," in Industry Applications Conference, 41st IAS Annual Meeting. Conference Record of the 2006 IEEE,2006, pp. 1933-1940.
References
236
[126] K.-j. Lee, K. Kim, S. Kim, J.-S. Ahn, S. Lim, and J. Lee, "Optimal magnet shape to improve torque characteristics of interior permanent magnet synchronous motor," Journal of Applied Physics, vol. 97, pp. 10Q505-10Q505-3, 2005.
[127] S. Y. Lim and J. Lee, "A design for improved performance of interior permanent magnet synchronous motor for hybrid electric vehicle," Journal of Applied Physics, vol. 99, pp. 08R308-08R308-3, 2006.
[128] K.-C. Kim, K. Kim, H. J. Kim, and J. Lee, "Demagnetization Analysis of Permanent Magnets According to Rotor Types of Interior Permanent Magnet Synchronous Motor," Magnetics, IEEE Transactions on, vol. 45, pp. 2799-2802, 2009.
[129] A. M. El-Refaie and T. M. Jahns, "Comparison of synchronous PM machine types for wide constant-power speed range operation," in Industry Applications Conference, Fourtieth IAS Annual Meeting, 2005, pp. 1015-1022 Vol. 2.
[130] A. M. El-Refaie and T. M. Jahns, "Optimal flux weakening in surface PM machines using fractional-slot concentrated windings," Industry Applications, IEEE Transactions on, vol. 41, pp. 790-800, 2005.
[131] A. M. El-Refaie, T. M. Jahns, P. J. McCleer, and J. W. McKeever, "Experimental verification of optimal flux weakening in surface PM Machines using concentrated windings," Industry Applications, IEEE Transactions on, vol. 42, pp. 443-453, 2006.
[132] W. L. Soong, P. B. Reddy, A. M. El-Refaie, T. M. Jahns, and N. Ertugrul, "Surface PM Machine Parameter Selection for Wide Field-Weakening Applications," in Industry Applications Conference, 2007. 42nd IAS Annual Meeting. Conference Record of the 2007 IEEE, 2007, pp. 882-889.
[133] A. R. Munoz, F. Liang, and M. W. Degner, "Evaluation of Interior PM and Surface PM Synchronous machines with distributed and concentrated windings," in 34th Annual Conference of IEEE Industrial Electronics, IECON. , 2008, pp. 1189-1193.
[134] C. Deak, A. Binder, B. Funieru, and M. Mirzaei, "Extended field weakening and overloading of high-torque density permanent magnet motors," in Energy Conversion Congress and Exposition, 2009. ECCE 2009. IEEE, 2009, pp. 2347-2353.
[135] P. H. Mellor, R. Wrobel, and D. Holliday, "A computationally efficient iron loss model for brushless AC machines that caters for rated flux and field weakened operation," in IEEE International Electric Machines and Drives Conference, IEMDC '09. , 2009, pp. 490-494.
[136] F. Meier and J. Soulard, "Analysis of Flux Measurements on a PMSM With Non-Overlapping Concentrated Windings," in IEEE Industry Applications Society Annual Meeting, IAS '08., 2008, pp. 1-8.
[137] R. Nuscheler, "Two-dimensional analytical model for eddy-current loss calculation in the magnets and solid rotor yokes of permanent magnet synchronous machines," in 18th International Conference on Electrical Machines, ICEM. , 2008, pp. 1-6.
[138] O. Bottauscio, G. Pellegrino, P. Guglielmi, M. Chiampi, and A. Vagati, "Rotor loss estimation in permanent magnet machines with concentrated windings," Magnetics, IEEE Transactions on, vol. 41, pp. 3913-3915, 2005.
References
237
[139] M. Nakano, H. Kometani, and M. Kawamura, "A study on eddy-current losses in rotors of surface permanent-magnet synchronous machines," Industry Applications, IEEE Transactions on, vol. 42, pp. 429-435, 2006.
[140] K. Yamazaki, M. Shina, Y. Kanou, M. Miwa, and J. Hagiwara, "Effect of Eddy Current Loss Reduction by Segmentation of Magnets in Synchronous Motors: Difference Between Interior and Surface Types," Magnetics, IEEE Transactions on, vol. 45, pp. 4756-4759, 2009.
[141] K. Yamazaki, Y. Fukushima, and M. Sato, "Loss Analysis of Permanent Magnet Motors With Concentrated Windings - Variation of Magnet Eddy Current Loss Due to Stator and Rotor Shapes," Industry Applications, IEEE Transactions on, vol. 45, pp. 1334 - 1342 2009.
[142] H. Polinder, M. J. Hoeijmakers, and M. Scuotto, "Eddy-current losses in the solid back-iron of permanent magnet machines with concentrated fractional pitch windings," in The 3rd IET International Conference on Power Electronics Machines and Drives. PEMD 2006. , 2006, pp. 479-483.
[143] J. Wang, Z. P. Xia, D. Howe, and S. A. Long, "Vibration Characteristics of Modular Permanent Magnet Brushless AC Machines," in IEEE Industry Applications Conference, 41st IAS Annual Meeting. , 2006, pp. 1501-1506.
[144] S.-H. Lee and J.-P. Hong, "A Study on the Acoustic Noise Reduction of Interior Permanent Magnet Motor with Concentrated Winding," in Industry Applications Society Annual Meeting, 2008. IAS '08. IEEE, 2008, pp. 1-5.
[145] S.-H. Lee, J.-P. Hong, S.-M. Hwang, W.-T. Lee, J.-Y. Lee, and Y.-K. Kim, "Optimal Design for Noise Reduction in Interior Permanent-Magnet Motor," Industry Applications, IEEE Transactions on, vol. 45, pp. 1954-1960, 2009.
[146] S. Araki, T. Kondou, and A. Yamagiwa, "Analyses of electromagnetic forces of concentrated winding permanent magnet brushless motors with rotor eccentricity," in International Conference on Electrical Machines and Systems, ICEMS, 2009, pp. 1-4.
[147] F. Xu and T. Li, "A research on the radial unbalanced force in the unsymmetrical brushless PM motors," in Industrial Electronics and Applications, 2008. ICIEA 2008. 3rd IEEE Conference on, 2008, pp. 1695-1698.
[148] W. Jingai, Z. Dongqi, and A. Satake, "The torque research on concentrated winding brushless PM motor," in TENCON '02. Proceedings. 2002 IEEE Region 10 Conference on Computers, Communications, Control and Power Engineering, 2002, pp. 2026-2029 vol.3.
[149] R. Qu and T. A. Lipo, "General closed-form analytical expressions of air-gap inductances for surface-mounted permanent magnet and induction machines," in Electric Machines and Drives Conference, 2003. IEMDC'03. IEEE International,2003, pp. 443-448 vol.1.
[150] I. Abdennadher, R. Kessentini, and A. Masmoudi, "Analytical derivation and FEA validation of the inductances of CWPMM," in 6th International Multi-Conference on Systems, Signals and Devices, SSD, 2009, pp. 1-7.
[151] J. K. Tangudu, T. M. Jahns, A. El-Refaie, and Z. Q. Zhu, "Lumped parameter magnetic circuit model for fractional-slot concentrated-winding interior permanent magnet machines," in Energy Conversion Congress and Exposition, 2009. ECCE 2009. IEEE, 2009, pp. 2423-2430.
[152] J. K. Tangudu, T. M. Jahns, A. M. El-Refaie, and Z. Q. Zhu, "Segregation of torque components in fractional-slot concentrated-winding interior PM machines
References
238
using frozen permeability," in Energy Conversion Congress and Exposition, 2009. ECCE 2009. IEEE, 2009, pp. 3814-3821.
[153] F. Meier and J. Soulard, "dq theory applied to a permanent magnet synchronous machine with concentrated windings," in 4th IET Conference on Power Electronics, Machines and Drives. PEMD. , 2008, pp. 194-198.
[154] Y. Duan, R. G. Harley, and T. G. Habetler, "Method for multi-objective optimized designs of Surface Mount Permanent Magnet motors with concentrated or distributed stator windings," in Electric Machines and Drives Conference, 2009. IEMDC '09. IEEE International, 2009, pp. 323-328.
[155] D. Reigosa, K. Akatsu, N. Limsuwan, Y. Shibukawa, and R. D. Lorenz, "Self-sensing comparison of fractional slot pitch winding vs. distributed winding for FW- and FI-IPMSMs based on carrier signal injection at very low speed," in Energy Conversion Congress and Exposition, 2009. ECCE 2009. IEEE, 2009, pp. 3806-3813.
[156] A. Eilenberger, M. Schroedl, and J. Heissenberger, "Comparison of outer rotor PMSM with single- and double-layer windings at same machine geometry with respect to the sensorless control capability," in 13th European Conference on Power Electronics and Applications, EPE, 2009, pp. 1-7.
[157] A. Eilenberger, M. Schroedl, F. Demmelmayr, and M. Troyer, "Short-circuit-proofed outer rotor PMSM with a wide field weakening range for high efficiency traction applications," in Industrial Electronics, 2009. IECON '09. 35th Annual Conference of IEEE, 2009, pp. 1294-1297.
[158] N. Imai, S. Morimoto, M. Sanada, and Y. Takeda, "Influence of Rotor Configuration on Sensorless Control for Permanent-Magnet Synchronous Motors," Industry Applications, IEEE Transactions on, vol. 44, pp. 93-100, 2008.
[159] K. Kojima, M. Hasegawa, T. Ohnuma, M. Tomita, S. Doki, S. Okuma, and K. Matsui, "Position sensorless control by reduced-order extended-flux observer applicable to IPMSM with concentrated windings," in Power Electronics Electrical Drives Automation and Motion (SPEEDAM), 2010 International Symposium on, pp. 1018-1023.
[160] Y. Wang, W. Xuhui, S. Xue, T. Fan, and Z. Lili, "Analysis and design of high power factor interior permanent magnet motor with concentrated windings for undersea vehicle propulsion," in IEEE Vehicle Power and Propulsion Conference, VPPC 2008, pp. 1-6.
[161] D. M. Ionel, "High-efficiency variable-speed electric motor drive technologies for energy savings in the US residential sector," in 12th International Conference on Optimization of Electrical and Electronic Equipment (OPTIM) 2010, pp. 1403-1414.
[162] M. V. Cistelecan and M. Popescu, "Study of the Number of Slots/Pole Combinations for Low Speed Permanent Magnet Synchronous Generators," in Electric Machines & Drives Conference, 2007. IEMDC '07. IEEE International,2007, pp. 1616-1620.
[163] E. V. Kazmin, E. A. Lomonova, and J. J. H. Paulides, "Brushless traction PM machines using commercial drive technology, Part II: Comparative study of the motor configurations," in International Conference on Electrical Machines and Systems, ICEMS., 2008, pp. 3772-3780.
References
239
[164] L. Alberti, M. Barcaro, M. D. Pre, A. Faggion, L. Sgarbossa, N. Bianchi, and S. Bolognani, "IPM Machine Drive Design and Tests for an Integrated Starter-Alternator Application," Industry Applications, IEEE Transactions on, vol. 46, pp. 993-1001, 2010.
[165] J. Germishuizen and M. Kamper, "Design and Performance Characteristics of IPM Machines with Single Layer Non-Overlapping Concentrated Windings," in Industry Applications Conference, 2007. 42nd IAS Annual Meeting. Conference Record of the 2007 IEEE, 2007, pp. 141-147.
[166] J. J. Germishuizen and M. J. Kamper, "IPM Traction Machine With Single Layer Non-Overlapping Concentrated Windings," Industry Applications, IEEE Transactions on, vol. 45, pp. 1387-1394, 2009.
[167] A. M. El-Refaie, "Fractional-Slot Concentrated-Windings Synchronous Permanent Magnet Machines: Opportunities and Challenges," Industrial Electronics, IEEE Transactions on, vol. 57, pp. 107-121, 2010.
[168] R. D. Cook, D. S. Malkus, M. E. Plesha, and R. J. Witt, Concepts and Applications of Finite Element Analysis, 4th ed.: New York, NY : Wiley, 2002.
[169] S. V. Ahamed and E. A. Erdelyi, "Nonlinear Theory of Salient Pole Machines," Power Apparatus and Systems, IEEE Transactions on, vol. PAS-85, pp. 61-70, 1966.
[170] E. A. Erdelyi, S. V. Ahamed, and R. D. Burtness, "Flux Distribution in Saturated DC Machines at No-Load," Power Apparatus and Systems, IEEE Transactions on, vol. 84, pp. 375-381, 1965.
[171] P. Silvester and M. V. K. Chari, "Finite Element Solution of Saturable Magnetic Field Problems," Power Apparatus and Systems, IEEE Transactions on, vol. PAS-89, pp. 1642-1651, 1970.
[172] M. V. K. Chari and P. Silvester, "Finite-Element Analysis of Magnetically Saturated D-C Machines," Power Apparatus and Systems, IEEE Transactions on, vol. PAS-90, pp. 2362-2372, 1971.
[173] J. P. A. Bastos and N. Sadowski, Electromagnetic Modelling By Finite Element Methods. New York, NY: Marcel Dekker, Inc., 2003.
[174] R. Dutta, "A Segmented Interior Permanent Magnet Synchronous Machine with Wide Field-weakening Range," PhD Thesis, School of Electrical Engineering and Telecommunications, University of New South Wales, Australia, 2007.
[175] D. Jing, "Numerical Analysis of a Permanent Magnet Synchronous Machine Using Numerical Techniques," PhD Thesis, Department of Electrical and Computer Engineering, National University of Singapore, Singapore, 2004.
[176] D. Zhong, "Finite Element Analysis of Synchronous Machines," PhD Thesis, Department of Electrical Engineering, The Pennsylvania State University, 2010.
[177] J. Jin, The Finite Element Method in Electromagnetics. New York, NY: John Wiley & Sons, 2002.
[178] N. Bianchi, Electrical Machine Analysis Using Finite Elements. Boca Raton, FL: Taylor and Francis Group, LLC, 2005.
[179] P. P. Silvester and R. L. Ferrari, Finite Elements for Electrical Engineers:Cambridge University Press, 1996.
[180] M. V. K. Chari and P. Silvester, "Analysis of Turboalternator Magnetic Fields by Finite Elements," Power Apparatus and Systems, IEEE Transactions on, vol. PAS-90, pp. 454-464, 1971.
References
240
[181] E. P. Furlani, Permanent Magnet and Electromechanical Devices. Rochester, NY: Academic Press, 2001.
[183] D. C. Hanselman, "Minimum torque ripple, maximum efficiency excitation of brushless permanent magnet motors," Industrial Electronics, IEEE Transactions on, vol. 41, pp. 292-300, 1994.
[184] A. Chiba, F. Nakamura, T. Fukao, and M. A. Rahman, "Inductances of cageless reluctance-synchronous machines having nonsinusoidal space distributions," Industry Applications, IEEE Transactions on, vol. 27, pp. 44-51, 1991.
[185] A. M. El-Refaie, T. M. Jahns, and D. W. Novotny, "Analysis of surface permanent magnet machines with fractional-slot concentrated windings," Energy Conversion, IEEE Transaction on, vol. 21, pp. 34-43, 2006.
[186] D. A. Staton, T. J. E. Miller, and S. E. Wood, "Maximising the saliency ratio of the synchronous reluctance motor," Electric Power Applications, IEE Proceedings., vol. 140, pp. 249-259, 1993.
[187] L. Chong and M. F. Rahman, "Saliency ratio optimization in an IPM machine with fractional-slot concentrated windings," in International Conference on Electrical Machines and Systems, ICEMS. , 2008, pp. 2921-2926.
[188] L. Chong and M. F. Rahman, "Saliency ratio derivation and optimisation for an interior permanent magnet machine with concentrated windings using finite-element analysis," Electric Power Applications, IET, vol. 4, pp. 249-258, 2010.
[189] T. J. E. Miller, Brushless permanent-magnet and reluctance motor drives. New York: Oxford University Press, 1989.
[190] W. L. Soong, S. Han, and T. M. Jahns, "Design of interior PM machines for field-weakening applications," in Electrical Machines and Systems, 2007. ICEMS. International Conference on, 2007, pp. 654-664.
[191] E. P. Furlani, Permanent Magnet and Electromechanical Devices - Materials, Analysis and Applications: Academic Press, 2001.
[192] A. Goldman, Modern Ferrite Technology, Second ed.: Springer, 2006.[193] (24/03/2001). Magnetic Component Engineering: Neodymium Iron Boron vs.
Samarium Cobalt- A Comparison. Available: http://www.mceproducts.com/knowledge-base/article/article-dtl.asp?id=32
[194] "Magnet Sales and Manufacturing Inc. - High Performance Permanent Magnets 7," ed. California, 1995.
[196] Magnetic Properties of Sintered NdFeB Magnets [Online]. Available: http://www.raremagnet.com/en/product/product1_1.asp
[197] "Advaced MotorTech, LLC: Short Course on the Design of Interior Permanent Magnet and Brushless DC Machines- Taking Theory to Practice," ed, May 2010.
[201] L. Chong, R. Dutta, and M. F. Rahman, "Field Weakening Performance of a Concentrated Wound PM Machine with Rotor and Magnet Geometry Variation " in Power Engineering Society General Meeting, PES, Minneapolis, USA, 2010.
[202] N. Bianchi, S. Bolognani, and P. Frare, "Design criteria of high efficiency SPM synchronous motors," in Electric Machines and Drives Conference, IEMDC'03. IEEE International, 2003, pp. 1042-1048 vol.2.
[203] A. Borisavljevic, H. Polinder, and J. A. Ferreira, "On the Speed Limits of Permanent-Magnet Machines," Industrial Electronics, IEEE Transactions on, vol. 57, pp. 220-227, 2010.
[204] T. Kikuchi and T. Kenjo. (1997, 22/03/2011). A Unique Desk-top Electrical Machinery Laboratory for the Mechatronics Age. Available: http://ewh.ieee.org/soc/es/Nov1997/09/INDEX.HTM
[205] N. Bianchi, S. Bolognani, and F. Luise, "Potentials and limits of high speed PM motors," in Industry Applications Conference, 2003. 38th IAS Annual Meeting. Conference Record of the, 2003, pp. 1918-1925 vol.3.
[206] C. Han-Wook, J. Seok-Myeong, and C. Sang-Kyu, "A Design Approach to Reduce Rotor Losses in High-Speed Permanent Magnet Machine for Turbo-Compressor," Magnetics, IEEE Transactions on, vol. 42, pp. 3521-3523, 2006.
[207] A. Beiser, Schaum's outline of theory and problems of physical science:McGraw-Hill, 1988.
[208] K. Shinichi, T. Norio, and K. Takeharu, "Measurement and analysis of AC loss of NdFeB sintered magnet," Electrical Engineering in Japan, vol. 154, pp. 8-15,2006.
[211] A. M. El-Refaie and T. M. Jahns, "Impact of Winding Layer Number and Magnet Type on Synchronous Surface PM Machines Designed for Wide Constant-Power Speed Range Operation," in Industry Applications Conference, 41st IAS Annual Meeting., 2006, pp. 1486-1493.
[212] W. H. Yeadon and A. W. Yeadon, Handbook of Small Electric Motors:McGraw-Hill, 2001.
[213] "Sankey - 35RM300 Nonoriented Silicon Steel Datasheet," ed, 2010.[214] H. Jussila, P. Salminen, A. Parviainen, J. Nerg, and J. Pyrhonen, "Concentrated
winding axial flux permanent magnet motor with plastic bonded magnets and sintered segmented magnets," in 18th International Conference on Electrical Machines, ICEM, 2008, pp. 1-5.
[215] N. Takahashi, H. Shinagawa, D. Miyagi, Y. Doi, and K. Miyata, "Factors affecting eddy current losses of segmented Nd-Fe-B sintered magnets without insulation in large PM motors," in IEEE International Electric Machines and Drives Conference, IEMDC. , 2009, pp. 24-29.
[216] A. Fukuma, S. Kanazawa, D. Miyagi, and N. Takahashi, "Investigation of AC loss of permanent magnet of SPM motor considering hysteresis and eddy-current losses," Magnetics, IEEE Transactions on, vol. 41, pp. 1964-1967, 2005.
[217] K. Akatsu, K. Narita, Y. Sakashita, and T. Yamada, "Characteristics comparison between SPMSM and IPMSM under high flux density condition by both
experimental and analysis results," in Electrical Machines and Systems, 2008. ICEMS 2008. International Conference on, 2008, pp. 2848-2853.
[218] D. Ishak, Z. Q. Zhu, and D. Howe, "Eddy-current loss in the rotor magnets of permanent-magnet brushless machines having a fractional number of slots per pole," Magnetics, IEEE Transactions on, vol. 41, pp. 2462-2469, 2005.
[219] P. Sergeant and A. Van den Bossche, "Segmentation of Magnets to Reduce Losses in Permanent-Magnet Synchronous Machines," Magnetics, IEEE Transactions on, vol. 44, pp. 4409-4412, 2008.
[220] Juha Pyrhonen, Tapani Jokien, and V. Hrabovcova, Design of RotatingElectrical Machines: John Wiely & Sons, Ltd., 2008.
[222] Wikipedia. (2011). American Wire Guage. Available: http://en.wikipedia.org/wiki/American_wire_gauge
[223] A. Emadi, Energy-Efficient Electric Motors, Third ed. New York: Marcel Dekker, Inc., 2005.
[224] R. H. Park, "Two-reaction theory of synchronous machines generalized method of analysis-part I," American Institute of Electrical Engineers, Transactions of the, vol. 48, pp. 716-727, 1929.
[225] D. W. Novotny and T. A. Lipo, Vector control and dynamics of AC drives. New York: Oxford University Press, 1996.
[226] M. F. Rahman, Electrical Machines and Drives Lecture Notes - Lecture 6: CSI of Synchronous Motor Drive. Australia: University of New South Wales, 2007.
[227] P. Vas, Vector Control of AC Machines. New York: Oxford University Press, 1990.
[228] P. C. Krause, O. Wasynczuk, and S. D. Sudhoff, Analysis of Electric Machinery.New York: IEEE Press, 1995.
[229] M. F. Rahman, Electrical Machines and Drives Lecture Notes - Lecture 9: Dynamic Model of the Three-Phase Synchronous Motor and its Vector Control.Australia: The University of New South Wales, 2007.
[230] S. Morimoto, Y. Takeda, T. Hirasa, and K. Taniguchi, "Expansion of operating limits for permanent magnet motor by current vector control considering inverter capacity," Industry Applications, IEEE Transactions on, vol. 26, pp. 866-871, 1990.
Fig. D.1 Approximate model used in thermal analysis
The model as shown in fig D.1 was set to operate with 250Hz excitation frequency with
an ambient temperature of 30°C. It was subjected to the worse case scenario where the
machine was completely enclosed in the case (without fins) and made to run
continuously at full load current over a timeframe of 20 mins. The steady state
temperatures at various parts of the machine are given in table D.2:
Table D.2Estimated temperature at various parts of the machine
Housing 94°C
Rotor yoke and rotor surface 107°C
Magnet 107°C
Stator surface and stator tooth 103°C
Stator yoke 100°C
Winding 121°C
Housing 94°C
Rotor yoke
Housing
Rotor surface
Magnet
Stator tooth
Winding
Stator surfaceStator yoke
255
APPENDIX E
FINAL MACHINE DRAWINGS
E.1 ABB Casing used (with Original Induction Motor)
Appendix E
256
E.2 Stator of the CW-IPM Prototype
Appendix E
257
E.3 Rotor of the CW-IPM Prototype
Appendix E
258
Appendix E
259
E.4 Shaft of the CW-IPM Prototype
Appendix E
260
E.5 Key (shaft) of the CW-IPM Prototype
Appendix E
261
E.6 End-plates of the CW-IPM Prototype
262
APPENDIX F
EXPERIMENTAL SETUP
F.1 The Experimental Setup
The experimental setup of for testing the performance of the CW-IPM prototype,
(shown in fig. F.1), consists of the following:
(i) CW-IPM machine
(ii) Kollmorgen PM machine (loading machine)
(iii) -
(iv) Windows based PC
(v) DS1104 Controller board
(vi) 3-phase IGBT inverter
(vii) 10Nm Torque transducer (HBM – T20WN)
(viii) Position Sensor (Heidenhain ROD 426 – 5000 pulses per revolution)
(ix) Power analyser
(x) 415V Voltage regulator
Fig. F.1 Complete experimental Setup
(i)
(viii)
(vi) (x) (vi)
(iii)
(ix)(ii)
(vii)
Appendix F
263
In this setup, the inverter was supplied by 3-phase, 415V mains through the voltage
regulator. The DS1104 controller board receives feedback signals from the inverter,
(supply current from two phases), as well as from the position sensor. These signals,
together with the applied control algorithm and SVM produces the desired current
references to the inverter, (through the controller board). The controller algorithm is
written in C-code, and is applied in real-time via d-space control desk on a windows
based PC, (this interface is shown in fig. F.2). 3-phase supply is fed into the CW-IPM
machine through a power analyser, (which measures the input quantities to the motor, as
well as the power factor). The CW-IPM machine is loaded by the Kollmorgen machine,
connected to a load bank. At low speeds, the Kollmorgen machine is connected to 4:1
gearbox to achieve higher loading torque. Output shaft torque from the CW-IPM
machine is measured by the 10Nm HBM torque transducer.
Fig. F.2 d-space control desk, real-time graphical user interface
Appendix F
264
F.2 Control Algorithm
The key portion of the control algorithm, written C, and ran with d-space control desk is
shown below:
/***************************************************************************************************** Rotor Field Oriented Control of the CW-IPM (DS1104) *****************************************************************************************************/#include <c:\dSPACE\DS1104\RTLib\brtenv.h>#include <ctrl.h> /* general header for motor control */#include <vlimtfun.c>#include <svmodc.c>
/********************************************************************************************//************************** Determination of Reference and Max id ***************************//********************************************************************************************/
/* initialize control parameters */varinit();/* periodic event with timer0 */ds1104_start_isr_timer0(timer0_period, isr_timer0);/* Background tasks */while(1){
RTLIB_BACKGROUND_SERVICE(); /* ControlDesk service */}
}
/*********************************************** End ********************************************************/
F.3 3-phase IGBT Inverter
The 3-phse IGBT inverter which supplies power to the CW-IPM machine is shown fig.
F.3. Details of its circuit diagrams are shown in fig. F.4 to fig. F.5.
Fig. F.3 3-phse IGBT inverter (casing off) [174]
Appendix F
268
Fig. F.4 IGBT inverter schematic
Appendix F
269
Fig. F.5 Connections between the IGBT inverter and control boards
Appendix F
270
F.4 Kollmorgen PM Machine Specifications
Specifications of the Kollmorgen machine, (used as the loading machine), are as listed
in table F.1.
Table F.1Kollmorgen PM Machine specifications
Model AKM33H
Rated Current 5.63Arms
Rated Torque 2.88Nm
Rated Voltage 320VDC
Rated Speed 5500RPM
Rated Power 1.31kW
Stator Resistance (l-l)
Appendix G
271
APPENDIX GPUBLICATION LIST
Patents:
[1] M. F. Rahman, Rukmi Dutta, Lester Chong “Interior permanent magnet machine”, Provisional patent no. : 2011903320
Journal Publications:
[2] Lester Chong, Rukmi Dutta, M. F. Rahman, “Application of Concentrated Windings in Interior Permanent Magnet Machine”, Australian Journal of Electrical & Electronics Engineering (AJEEE) 2008. ISSN: 1448837X
[3] Lester Chong, M. F. Rahman, “Saliency Ratio Derivation and Optimization for an IPM Machine with Concentrated Windings Using Finite Element Analysis ”, Institution of Engineering and Technology (IET) 2009. ISSN: 1751-8660
[4] Lester Chong, Rukmi Dutta, M. F. Rahman, “Electromagnetic Losses in a 1kW Concentric Wound IPM Machine for Field Weakening Applications”, Journal of Applied Superconductivity and Electromagnetics (JASEM), 2010. ISSN 1836-7151
Conference Publications:
[5] Lester Chong, Rukmi Dutta, M. F. Rahman, “Application of Concentrated Windings in Interior Permanent Magnet Machine”, Australasian Universities Power Engineering Conference (AUPEC) 2007, Australia. ISBN: 978-0-646-49488-3
[6] Lester Chong, Rukmi Dutta, M. F. Rahman, “Open Circuit Analysis of Concentrated Winding in Interior Permanent Magnet Machines with Fractional Slot Distribution”, 4th IET International Conference on Power Electronics, Machines and Drives (PEMD) 2008, UK. ISBN: 978-0-86341-900-3
[7] Lester Chong, M. F. Rahman, “Comparison of d- and q-axis Inductances in an IPM machine with Integral-slot Distributed and Fractional-slot Concentrated Windings”, 18th International Conference on Electrical Machines (ICEM) 2008, Portugal. ISBN: 978-1-4244-1735-3
[8] Lester Chong, M. F. Rahman, “Saliency Ratio Optimization in an IPM Machine with Fractional-slot Concentrated Windings” 11th International Conference on Electrical Machines and Systems (ICEMS) 2008, China. ISBN: 978-1-4244-3826-6
[9] Lester Chong, Rukmi Dutta, M. F. Rahman, “Parameter Analysis of an IPM Machine with Fractional-slot Concentrated Windings, Part I: Open-circuit Analysis”, Australasian Universities Power Engineering Conference (AUPEC) 2008, Australia. ISBN: 978-0-7334-2715-2
[10] Lester Chong, Rukmi Dutta, M. F. Rahman, “Parameter Analysis of an IPM Machine with Fractional-slot Concentrated Windings, Part II: Including Armature-reaction”, Australasian Universities Power Engineering Conference (AUPEC) 2008, Australia. ISBN: 978-0-7334-2715-2
[11] Lester Chong, Rukmi Dutta, M. F. Rahman, “Design of IPM machine with Concentrated Windings for Vehicular Applications”, European Power Engineering Conference, 2009, Spain. ISBN: 978-1-4244-4432-8
Appendix G
272
[12] Lester Chong, Rukmi Dutta, M. F. Rahman, “Design and Mechanical Consideration of an IPM Machine with Concentrated Windings”, Australasian Universities Power Engineering Conference (AUPEC) 2009, Australia. ISBN: 978-1-4244-5153-1
[13] Lester Chong, Rukmi Dutta and M. F. Rahman, "Design of an interior permanent magnet machine with concentrated winding for field weakening applications," in Proc. of IEEE Int. Electric Machines & Drives Conf. (IEMDC), 2009, pp. 1985-1992.
[14] Lester Chong, Rukmi Dutta, M. F. Rahman, “Design and Thermal Considerations of an Interior Permanent Magnet Machine with Concentrated Windings”, International Conference on Electrical Machines and Systems (ICEMS) 2009, Japan. ISBN: 978-1-4244-5177-7
[15] Lester Chong, Rukmi Dutta, M. F. Rahman, “Field Weakening Performance of a Concentrated Wound PM Machine with Rotor and Magnet Geometry Variation”, Power and Engineering Society General meeting (PES) 2010, USA. ISBN: 978-1-4244-6549-1
[16] Lester Chong, Rukmi Dutta, M. F. Rahman, “Design of a Highly Efficient 1kW IPM Machine with a Very Wide Constant Power Speed Range”, International Power Electronics Conference -ECCE-Asia (IPEC) 2010, Japan. ISBN: 978-1-4244-5394-8
[17] Lester Chong, Rukmi Dutta, M. F. Rahman, “A Comparative Study of Rotor Losses in an IPM with Single and Double Layer Concentrated Windings”, International Conference on Electrical Machines and Systems (ICEMS) 2010, Korea. ISBN: 978-1-4244-7720-3
[18] Lester Chong, Rukmi Dutta, Howard Lovatt, Nguyen Quang Dai, M. F. Rahman, “Comparison of Concentrated and Distributed Windings in an IPM Machine for Field Weakening Applications”, Australasian Universities Power Engineering Conference (AUPEC) 2010, Australia. ISBN: 978-1-4244-8379-2
[19] Lester Chong, Rukmi Dutta, M. F. Rahman, Howard Lovatt “Experimental verification of Rotor Losses in a Concentrated Wound IPM Machine with V-Shaped magnets”, International Electrical Machines and Drives Conference (IEMDC) 2011, Canada
[20] Rukmi Dutta, Lester Chong, M. F. Rahman, “Analysis of CPSR in Motoring and Generating Modes of an IPM Motor”, International Electrical Machines and Drives Conference (IEMDC) 2011, Canada
[21] Lester Chong, Rukmi Dutta, M. F. Rahman, Howard Lovatt “Open Circuit Analysis of an IPM Machine with Concentrated Windings Including Experimental Verification”, International Conference on Electrical Machines and Systems (ICEMS) 2011, China
[22] Lester Chong, Rukmi Dutta, D. Xiao, M. F. Rahman “Performance Comparison between Concentrated and Distributed Wound IPM Machines used for Field Weakening Applications”, Aegean Conference on Electric Machines and Power Electronics(ACEMP) 2011, Turkey