Motor Design

Flux Conference 2012
High Efficiency Motor Design for Electric Vehicles

L. Chen, J. Wang, P.
Lombard, P. Lazari and
V. Leconte
University of Sheffield,
CEDRAT Date : 18 October 2012
Presented by: P. Lazari
Outline
High-efficiency electric motor for electric vehicle
2012 Flux Conference - Rome, Italy, October 17-18, 2012
2
Requirements and design challenges for EV traction machines
Computationally efficient design technique for EV traction machines
against driving cycle
Optimisation objective and constraints
Optimisation methodology and tools
Design case studies
Conclusions
Requirements and design challenges for EV
traction machines
3
Requirements:
High energy efficiency and cost-effectiveness.
High torque, 3~4 times the nominal value, required for acceleration and
hill climbing.
Peak power twice the rated value at high speeds.
Design Challenges:
Wide operating torque-speed range, imposes significant constraints on
achievable machine efficiency.
Machine design should be targeted for overall energy saving over
driving cycle.
Enormous computation time is required for FE evaluation over a driving
cycle. This makes multi-parameter optimisations with FE models
practically impossible.
Solutions
4
Derivation of representative driving cycle operating
points - dramatically reduce the number of FE
computations.
Use of efficient FE coupled optimisation tool
--Flux & GOT-IT provides an ideal and effective design
optimisation environment for EV traction machines.
Requirements of EV traction
5
Max speed 120 km/h
Base speed
Peak power
Max continuous power at
max speed From NEDC
Contin. Torque 10% slope
at 20km/h
Peak Torque 20% slope
at given acceleration
Investigated vehicle data
6
Parameter (Unit) Value
Tire size 165/50 R15
Radius of wheels (m) 0.273
Vehicle mass (kg) 800
Gravitational acceleration (m/s
2
) 9.807
Rolling resistance 0.007
Product of drag coefficient and
front area (m
2
)
0.35
Air density (kg/m
3
) 1.25
Efficiency of differential 0.980
Gear Ratio 4.0 & 7.0
Vehicle power-train employs a distributed drive with three motors coupled
to the front and rear axle.
40% and 60% of traction power is shared between front and rear motors
respectively.
Vehicle characteristics over NEDC
7
Speed, torque and power plots of front motor derived from vehicle
specification.
0 200 400 600 800 1000 1200
0
200
400
600
S
p
e
e
d
(
r
a
d
/
s
)
0 200 400 600 800 1000 1200
0
10
20
30
40
T
o
r
q
u
e
(
N
m
)
0 200 400 600 800 1000 1200
0
2000
4000
6000
8000
Time (s)
P
o
w
e
r
(
W
)
Motor design specification
8
Traction machine should be optimized against NEDC, while satisfying
thermal and volumetric constraints.
Parameter (Unit) Value
Base speed (rpm) 1350
Maximum cruise speed (rpm) 4500
Tpeak at base speed (Nm) for 120 sec 70.0
Tcont. at base speed (Nm) 35.0
Tpeak at maximum cruise speed (Nm) 18.0
Peak power (kW) 8.8
Continuous power (kW) 5.0
Nominal DC link voltage (V) 120
Max line-to-line voltage (V) < 250 (for safety)
Cooling Air-cooled
Problem Require enormous amount of calculation time
viz. 600 operating points over NEDC, ~ 200 hours just for one design
Energy distribution over NEDC
9
Energy consumption on the motor output shaft over NEDC of
reference vehicle.
There exist six dominant points where energy consumption is
significantly high.
0
100
200
300
400
500
0
5
10
15
20
25
30
35
0
2
4
6
8
10
12
x 10
4
Speed (rad/s) Torque (Nm)
E
n
e
r
g
y
c
o
n
s
u
m
p
t
i
o
n
(
J
)
Energy distribution over NEDC
10
The remaining points scatter around and can be represented by
another six points according to the energy centre of gravity.
i
N
j
ij i
E E
,... 2 , 1
i
i
N
j
mij ij
i
mci
N
j
ij ij
i
mci
T E
E
T
E
E
,... 2 , 1
,... 2 , 1
1
1

0
100
200
300
400
500
0
5
10
15
20
25
30
35
0
5000
10000
15000
Speed (rad/s)
Torque (Nm)
E
n
e
r
g
y
c
o
n
s
u
m
p
t
i
o
n
(
J
)
Validation of 12 NEDC points
11
i i
N
i
i i m i m
t t T t EL / ) 1 ( ) ( ) (
,... 2 , 1

j j
j
nj
E EL / ) 1 (
12
,... 2 , 1

Motor speed (rpm)

T
o
r
q
u
e
(
N
m
)
0 1000 2000 3000 4000 5000
10
20
30
40
50
60
70
0.8
0.82
0.84
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
0.96
0.94
0.95
0.93
0.92 0.91 0.90
Energy loss over all points in NEDC cycle (kJ) 55.104
Energy loss over 12 points (kJ) 53.387
Difference (%) 3.1%
Design technique for EV traction machines
12
Calculation time limitation
effectively removed using 12
representative points.
Rated
Point
NEDC high energy
consumption is dominated by
low torque/middle-high
speed points.
Mismatch between NEDC high
operating duty and rated
operating points Significance
of optimising against driving
cycle.
Design technique on ARTEMIS Urban
13
Frequent variation of Artemis Urban speed.
55 representative points with loss
calculation error of 5%.
Artemis Urban is dominated by variable
torque/ low to middle speed points.
Optimisation objective and constraints
14
Objective:
Minimisation of energy loss over 12 NEDC points.
Constraints:
Dimensional: stack length and outside diameter.
Electrical: current, voltage limits imposed by converter.
Thermal: temperatures of windings and magnets.
Mechanical: stress level of rotor.
Demagnetization withstand of magnets.
Torque ripple.
Optimisation tool: GOT-It
15
GOT-It is a powerful and efficient optimisation tool developed by
CEDRAT.
Geo or phy
parameters
Objective
Constraints
Genetic, SQP, SSO
Flux
Portunus
.
GOT-It coupling with Flux 2D/3D
16
Objectives, constrains and selection of algorithms are defined in
Got-it which performs optimisation by passing parameters to and
retrieve results from Flux FE model.
Sequential Surrogate Optimiser
17
Global optimisation
Response Surfaces (RS)
Iteratively converging to
optimum
F
x
Optimisation case studies for EV traction motors
18
Micro-sized electric vehicle with front and rear wheel drives.
Two rear motors -- 36 slot/6 pole PMa-SynRM (230%, gear ratio 7:1)
Front motor Rear motor
Ferrite magnets
Front motor -- 36 slot/4 pole PMa-SynRM (1x40%, gear ratio 4:1)
Optimisation design parameters
19
2 parameters for front motor optimisation:
Rotor radius (Rr) and tooth width (TW).
5 parameters for rear motor optimisation:
Rotor radius (Rr), insulation ratio (IR), tooth width (TW), back iron
thickness (BI) and number of turns per coil (TN).
Torque-speed operating range requirements
20
Parameter Front Rear
Base speed 1350 rpm 2100 rpm
Max speed 4500 rpm 8200 rpm
Rated torque 35 Nm 17 Nm
Peak torque at base speed 70 Nm 30 Nm
Peak torque at maximum
speed 18 Nm 7.5 Nm
Optimisation design constraints
21
Parameter Front Rear
Stack length 118 mm 105 mm
Outer diameter 150 mm 120 mm
Max current 170 A 170 A
DC link voltage 120 V 120 V
Max line-to-line no-load voltage (for safety) 240 V 240 V
Dimensional and electrical constraints:
Demagnetisation:
Magnets withstand during deep flux weakening.
Mechanical strength:
Stress level on rotor bridges < 450 MPa at 1.5 * Nmax.
Thermal constraints:
Winding and magnet temperature should be < 120 C.
Optimisation design constraints and objective
22
Static thermal model employed in optimisation process:
Optimisation objective:
Minimization of energy loss over 12 NEDC points while satisfying
required torque-speed range and other thermal and volumetric
constraints.
Rotor Stator
sh
R
1 ry
R
2 ry
R
1 mag
R
2 mag
R
ag
R
1 st
R
2 st
R
1 sy
R
2 sy
R
hs
R
0
R
ambient
T
cu
P
fe ry
P

fe st
P

fe sy
P

Optimisation workflow
23
Optimisation evolution
24
For the front motor: For the rear motor:
Iteration
1
samples
Iteration
2
samples
Iteration
3
samples
Shrinking of parameter region
during optimisation
Rotor radius (mm)
16 samples, 8 hours for the
2-parameter optimisation
78 samples, 36 hours for the
5-parameter optimisation
Optimised design case studies
25
12 NEDC points
Front Rear
W
cu
kJ 56.97kJ
37.13 kJ
W
fe
kJ 15.0 kJ
19.45 kJ
W
fe
: W
cu
3.79
1.91
Efficiency % 93.67%
94.56%
Front motor
Rear motor
Conclusions
26
Energy distribution of NEDC can be represented by 12 points.
FE based optimization for EV traction machines can be made
against 12 representative points, leading to large reduction of
computation time.
Optimum designs with multiple parameters can be efficiently
obtained using GOT-It & Flux.
For PMa-SynRM copper loss is the dominant loss component over
NEDC. Optimisation against rated point may also lead to good
efficiency over NEDC.
Optimised design case studies Rear Motor
27
Base Speed rpm 2100
Peak Torque N-m 30
Torque density (peak) kN-m/m
3
26.6
Armature current A (peak) 80.5
Cu Loss W 681.1
Iron Loss W 40.5
Total Loss W 721.6
Mechanical power W 6600
Efficiency % 90.16%
Rated Torque N-m 17
Torque density (continuous) kN-m/m
3
15.08
Cu Loss W 235.9
Iron Loss W 33.4
Total Loss W 269.3
Mechanical power W 3750
Efficiency % 93.29%
Maximum Speed rpm 8200
Peak Torque N-m 7.5
Cu Loss W 498.0
Iron Loss W 136.7
Efficiency % 91.00%
Continuous Torque N-m 4.4
Cu Loss W 184.5
Iron Loss W 79.2
Efficiency % 93.31%
Optimised design case studies Rear Motor
28
NEDC
Cycle Wcu J 3.64E+04
Cycle Wfe J 2.17E+04
Wcu:Wfe 1.67
Cycle loss J 5.81E+04
Cycle Wout J 9.52E+05
NEDC Energy Efficiency % 94.30%
ARTEMIS
Cycle Wcu J 4.74E+04
Cycle Wfe J 1.13E+04
Wcu:Wfe 4.19
Cycle loss J 5.85E+04
Cycle Wout J 7.59E+05
NEDC Energy Efficiency % 92.83%
Optimised design case studies Front Motor
29
Items
Final optimal
design
Peak torque density 35.4 kNm/m
3
Peak current 153 A
Peak current density (rms) 16.38 A/mm
2
Voltage at peak torque/base speed 102 V
Copper Loss at peak torque/base speed 2517 W
Iron Loss at peak torque/ base speed 36.20 W
Efficiency at peak torque/base speed 79.47%
Peak torque/max speed 15 Nm/4500 rpm
Current at peak torque/max speed 68.77 A
Voltage at peak torque/max speed 120 V
Copper Loss at peak torque/max speed 508.33 W
Iron Loss at peak torque/ max speed 165.75 W
Efficiency at peak torque/max speed 91.19%
Copper loss over NEDC 56.97kJ
Iron loss over NEDC 15.0 kJ
Wfe : Wcu 2.8
Driving cycle efficiency (NEDC) 93.67%

Motor Design

Uploaded by

Copyright:

Available Formats

Motor Design

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Motor Design

Uploaded by

Copyright:

Available Formats

Flux Conference 2012

High Efficiency Motor Design for Electric Vehicles

Motor speed (rpm)

You might also like