Multilayer winding arrangements in axial flux PM machines

Transcription

New Frontiers of Brushless PM
Motor Technology
Dan M. Ionel, Ph.D., IEEE Fellow
dan.ionel@uky.edu
CWIEME Berlin, May 11, 2016
SPARK Introduction | February, 2016 | 1
Dr. Dan M. Ionel
Dan M. Ionel is Professor of Electrical Engineering and L. Stanley
Pigman Chair in Power at University of Kentucky in Lexington, KY.
Previously, he held dual appointments in industry, as Chief Engineer
with Regal Beloit Corp and before as Chief Scientist with Vestas
Wind Turbines, and in academia, as Visiting and Research Professor
with University of Wisconsin and Marquette University in
Milwaukee, WI.
Dr. Ionel has more than 25 years of engineering experience and has
designed electric machines and drives with power ratings between
0.002 and 10,000hp. He holds more than 30 patents and has
published more than 100 journal and conference papers, including
two winners of IEEE best paper awards. Dr. Ionel is an IEEE Fellow,
the Chair of the IEEE Power and Energy Society Electric Motor
Subcommittee, and the General Chair of the 2017 anniversary
edition of the IEEE IEMDC Conference.
dan.ionel@uky.edu
SPARK
Laboratory
New Frontiers of Brushless PM Motor Technology | CWIEME Berlin, May 2016 | 2
SPARK and PEIK at University of Kentucky (UK)
• UK enjoys a longstanding tradition in electric machines and drives
• Early developments on linear and PM motors, and vector control
• Many learned machines using the Nasar and Boldea classic books
• PEIK - Power and Energy Institute of Kentucky, launched with large DOE grant in 2010
• Core faculty in electric power engineering and many others in related fields
• Endowment established and inaugural L. Stanley Pigman Chair started in 2015
• On-going research on electric machines and drives, power electronics and systems,
renewable and alternative energy technologies
• SPARK and other laboratories
• Motor Design Ltd. and ANSYS Inc. strategic partnerships.
SPARK
Laboratory
Motor Design Ltd. and Center for Applied Energy Research
Motor Design Ltd. (MDL)
• Founded in 1998
• Developers of the Motor-CAD & Motor-LAB
software
• Engineering and R&D projects for companies
and EU programs
• Strategic partner of ANSYS.
Center for Applied Energy Research (CAER)
• Established in 1975
• One of University of Kentucky’s largest,
stand-alone, multidisciplinary research
centers with more than 100 staff
• Fiber Development Laboratory with the
largest solution spinning line found in an
academic setting in North America
• Renowned research program on nano
carbon composites.
SPARK
Laboratory
Outline
1. Introduction
2. How far can we reach with existent materials?
•
•
•
•
Record-braking motor – Formula E racing cars
Design considerations for industrial applications
Large scale optimization and innovation
Complex duty cycles – multi-physics analysis, including thermal
3. How far will we be able to reach with new materials?
•
•
•
•
Nanocarbon wires and windings
Material characteristics and manufacturing issues
Electric motor concepts – solar planes and cars
US DOE research program “Next Generation Electric Machines”.
4. Conclusions.
SPARK
Laboratory
AIM Motor Introduction
• FIA Formula E started in 2014
• 1 and 5 – Identic cars: Spark-Renault SRT_01E
• 2 – Powertrain and electronics: McLaren Electronics
• 3 – Gearbox: Hewland
• 4 – Battery 200kW: Williams Advanced Engineering
• 6 – Tyres: Michelin.
Source: www.formulae.com
Typical racing car driving cycle for one lap - LeMans
SPARK
Laboratory
Comparative Performance
Motor
Type
Torque Mass
(Nm)
(kg)
TRW
(Nm/kg)
Toyota Prius (2004)
Interior PM
400
51
7.8
Nissan Leaf
Tesla S
Interior PM
Induction
300
430
46
90
6.5
4.8
YASA 400
Axial PM
360
24
15.0
AIM
Spoke PM
110
9
12.2
Note: Values listed are for peak torque and active material mass.
SPARK
Laboratory
Variable Speed Motor Drives – Drive Cycles
•
•
•
•
•
Design considerations
Constant torque
Fan or pump load curve
Duty / driving cycle, incl. transients
Analysis of each design requires analysis for
multiple operating conditions.
Shaft torque, pu
92
0.8
93
0.9
94. 5
95
94
1
94. 5
0.7
efficiency map
2 Hp
4 Hp
6 Hp
8 Hp
10 Hp
Fan
4 load
90
9
0.6
0.5
94
93
92
0.4
93
91
90
0.3
0.2
92 0.3
SPARK
Laboratory
0.491
0.5
90
9
0.6 8 0.7
speed, pu
0.887
0.9
1
85
Multi-objective Optimization
• Example – Minimize cost and
minimize losses (i.e. maximize
efficiency)
• Impose constraints
• Quantify the effects on other
performance indices
• Thousands of design “candidates”
(variations) may have to be
analyzed
• Collection of “best compromise”
designs
• Definition of a Pareto front:
improvement in one objective can
only be achieved through a
deterioration in another
objective, e.g. cost vs. efficiency
tradeoffs
SPARK
Laboratory
Example Comparative Study – SynRel and PMSynRel
• Systematic optimization study with
10,000 candidate designs
• ultra-fast electromagnetic FEA
• differential evolution (DE)
• Typical rating 10hp 1,800rpm
• Induction motor stator core and
winding pattern
• Independent variables for rotor
geometry (8 or 9) and torque angle
• Comprehensive performance
evaluation
• Two objectives: max power factor
and min. badness
• One constraint: torque ripple
smaller than 20%.
SPARK
Laboratory
Synchronous Reluctance (SynRel)
SPARK
Laboratory
PM Assisted Synchronous Reluctance (PMSynRel)
SPARK
Laboratory
Coupled Electromagnetic and Thermal Analysis
PM machine specific characteristics
• Variation of PM characteristics with temperature
• Torque per amp (torque constant) changes
• Risk of demagnetization, especially in PM hot spots
• Variation of winding conductor characteristics with temperature.
Model
SPARK
Laboratory
Losses
Temperature
Coupled Electromagnetic and Thermal Analysis – Motor-CAD
Electromagnetics
• Ultra-fast 2D FEA in the abc reference
frame
• Seconds on a state of the art PC
workstation
• Analytical calculations for end effects
Thermal and air-flow
• Equivalent 3D networks
• Calculation time one order of magnitude shorter
than for electromagnetics
Coupling methods
• “Serial”, typ. 6 iterations for each
of Emag and thermal
• “Weak”, typ. only 2 iterations
for Emag and 6-10 for thermal
“Cutting edge”
• Design for complex duty cycles
• Large-scale optimization studies.
SPARK
Laboratory
Fast Duty Cycle Analysis
Maxwell
Motor-LAB
Motor-CAD
Therm
SPARK
Laboratory
Multi-objective Design Optimization
SPARK
Laboratory
AIM Motor Factsheet
•
•
•
•
•
•
•
•
•
•
•
E-motor for racing applications
Peak torque = 110Nm
Cont. torque = 70Nm
Max speed = 12,000rpm
Power = 73/60 kW
Max. current = 325Arms
Field oriented control
Rare-earth magnets
Non-oriented thin gage silicon steel
Liquid and air cooled
Active materials mass = 9kg
• Manufactured by Equipmake
• Electromagnetic & Thermal design by Motor Design Ltd.
SPARK
Laboratory
AIM Motor Stator and Rotor
SPARK
Laboratory
AIM Motor Modeling
SPARK
Laboratory
Magnetic Field and Copper Losses
90
Pac/Pdc
60
30
0
Flux-lines and flux-density
distribution on (peak) load
SPARK
Laboratory
0
1000
Frequency [Hz]
2000
AC losses in the stator winding
Core Losses
• Most core losses located in the stator
• Rotor surface can also have high loss density
• Highest loss concentration occurs in stator teeth and winding.
Pt [W/Kg]
863
431
0
Core losses on (peak) load
SPARK
Laboratory
Magnet Losses - Sources
•
•
•
•
•
•
PMs may be electrically conductive
Conductivity increases 6%-10% for 100C temperature rise
Higher PM losses in surface SPM than interior IPM
Surface BPM may require a retainer sleeve with additional losses
Retainers of glass fibre or carbon fibre
Resistivity
Material
Eddy-current induced by:
(ohm*m)
• Space MMF harmonics
• Permeance variation
• Time current harmonics
Copper
1.7 x 10-8
Aluminum
2.8 x 10-8
Steel
10 x 10-8
• Induced eddy-currents create PM losses
SmCo 1-5 Alloys
• Magnet losses mitigation methods:
• Integer or fractional slots/pole
• Segmentation.
SPARK
Laboratory
50 x 10-8
SmCo 2-17 Alloys
90 x 10-8
NdFeB – sintered
160 x 10-8
NdFeB – bonded
14,000 x 10-8
Ferrite
105
Magnet Losses - Mitigation
Distribution of specific on-load PM losses
Effect of segmentation on the magnet losses
m = transversal segments
n = axial segments
Psegmented
Pmonolithic
SPARK
Laboratory

L 
mL  n
Efficiency maps (MTPA)
Imax = 325Arms, Twdg = 160C0, Tmag = 120C0
SPARK
Laboratory
Duty Cycle
Loss model of the typical racing car driving cycle for one lap
SPARK
Laboratory
Transient Thermal Analysis
Results for 20 laps drive cycle
SPARK
Laboratory
Experimental Data
Measured and calculated torque
(q-axis current excitation, PMs at 40C0)
Measured total loss vs. current and
speed at 40C0
SPARK
Laboratory
Carbon Nanotubes (CNT)
• Individual carbon nanotubes (CNT) can have a conductivity up to 100 MS/m and very low
mass density
• Conductivity substantially decreases when individual CNTs are assembled to form
macroscopic conductors
• Conductivity may be enhanced by plating nanotubes with copper (Cu).
SPARK
Laboratory
Fiber Development Laboratory at University of Kentucky (UK)
• Part of the large Center for Applied Energy Research (CAER).
• Largest solution spinning line found in an academic setting in North America.
SPARK
Laboratory
Cu-plated Aligned MWCNT Wires
Bore
dispersion
MWCNTs
Cu-plated MWCNTs
Alignment
by drawing
Cu-MWCNT
conductor
core wire
SPARK
Laboratory
Coreless Multi-Disc Axial Flux PM machine
• Systematic study on a
suitable benchmark
• Coreless topology –
windings account for
most active weight and
losses
• Conventional
counterpart originally
developed at University
of Bath for the European
HELINET solar airplane
• Multiple stator modules
and rotor discs
• Stator with coils and a
light supporting
structure
• PM rotor.
SPARK
Laboratory
Example Design Studies
Performance improvement achieved by replacing copper coils with carbon
nanotube (CNT) windings. Machines with CNT windings maybe larger in
size and lighter. AC supplementary losses in NCT windings are negligible.
Further improvements possible due to better heath transfer.
SPARK
Laboratory
US DOE Research Program for NCT Wires and Machines
High performance conductors to minimize losses in the stator windings
• To be demonstrated on a 28 AWG NCT round wire with 1 m length
• Minimum of 33% reduction in I2R losses per unit weight or volume over a 28 AWG round
copper or Al wire at 1500C
• Use the new wire to demonstrate a 1 hp single phase induction motor, including
windings with electric insulation.
Source: US DOE-FOA-0001467, Next Generation Electric
Machines: Enabling Technologies, 2016.
Photo courtesy of Regal Beloit Corp.
SPARK
Laboratory
Conclusion
• Things that we can do today
• Example ultra-high torque density (12Nm/kg) AIM motor developed
for Formula E racing cars
• Innovate with existent materials
• Automated design optimization
• Motor performance under duty / driving cycles simulated using
electromagnetic and thermal coupled analysis
• Things that we maybe able to do in the future
• New materials such as nanocarbon tubes and wires
• It may take some time to deploy the new technology in conventional
industrial applications
• First in line to benefit may be high-tech applications, such as those for
aerospace.
SPARK
Laboratory
References
• Wang, Yi, Ionel, D. M., Staton, D. A., “Ultrafast Steady-State Multiphysics Model for PM and Synchronous
Reluctance Machines”, IEEE Transactions on Industry Applications, Vol. 51, No. 5, Sep/Oct 2015, pp. 3639-3646.
• Zhang Peng, Sizov, G. Y., Ionel, D. M., Demerdash, N.A.O., “Establishing the Relative Merits of Interior and SpokeType Permanent-Magnet Machines with Ferrite or NdFeB through Systematic Design Optimization”, IEEE
Transactions on Industry Applications, Vol. 51, No. 4, Jul/Aug 2015, pp. 2940-2948.
• Wrobel, R., Simpson, N., Mellor, P., Goss, J., Staton, D. A., ”Design of a Brushless PM Starter-Generator for Low-Cost
Manufacture and a High-Aspect-Ratio Mechanical Space Envelope“, Proceedings IEEE ECCE 2015 Congress,
Montreal, Canada, Sept. 2015, pp. 813-820
• A. Fatemi, N. Demerdash, T. Nehl, D.M. Ionel, "Large-scale Design Optimization of PM Machines Over a Target
Operating Cycle," in press, IEEE Transactions on Industry Applications, Early Access DOI: 10.1109/TIA.2016.2563383
• Yi Wang, D. M. Ionel, M. Jiang, S. J. Stretz, “Establishing the Relative Merits of Synchronous Reluctance and PM
Assisted Technology Through Systematic Design Optimization”, in press, IEEE Transactions on Industry Applications,
Early Access DOI: 10.1109/TIA.2016.2544831
• Popescu, M., Foley, I., Staton, D. A., Goss, J. E., “Multi-physics Analysis of a High Torque Density Motor for Electric
Cars”, Proceedings IEEE ECCE 2015 Congress, Montreal, Canada, Sept. 2015, pp.6537-6544
• Fatemi A., Ionel D. M., Demerdash N.A.O., Popescu, M., “Design Optimization of Spoke-Type PM Motors for Formula
E Racing Cars”, accepted for publication in Proc. of ECCE 2016
• Vandana Rallabandi, Narges Taran, D. M. Ionel and J. F. Eastham, “On the Feasibility of Carbon Nanotube Windings
for Electrical Machines: Case Study for a Coreless Axial Flux Motor”, accepted for publication in Proc. of ECCE 2016.
SPARK
Laboratory
University of Kentucky – Solar Car
SPARK
Laboratory
Consortium and Acknowledgments
• The SPARK - SEMPEED Consortium was recently established to
further the art of electric machine design
• Two academic sites at University of Kentucky and Marquette
University and one partner software company, Motor Design Ltd
(MDL)
• Four inaugural members: Grundfos, Kollmorgen, MTS, and Regal
Beloit
• Inspired by the legacy of the SPEED Consortium
• The continued support of ANSYS for our academic research is
gratefully acknowledged.
SPARK
Laboratory

Multilayer winding arrangements in axial flux PM machines

Transcription

Similar documents

T-REX Manual - RC Helitechnics

I`m helping to end MS by participating in the 2016 Jayman BUILT MS

The Board of Directors of Qatar Fuel (WOQOD) is pleased to invite

COMMERCIAL BOOTHS

5 `S` Training Program at ABV-IIITM, Gwalior on 21.07.2016

Porky Expo Flyer

The NeuLion Report

Sept 30-Oct 2, 2016 Nov 4-6, 2016 @Camp Tadmor

Initial Announcement

4 star ALOE HOTEL, PAPHOS