IEEE Power Electronics Magazine Compendium - March 2018 - 66

Component
Models
xi
Parameters

η (%)
Evaluation
Pareto
Surface

xj
xk

System
Model
σ
(kW/∈)

Design
Space

α
(kW/dm2)
Performance
Space

(a)

Magnetic
Coupling k
Realized Prototype
0.350
η-α-Pareto Front
99
0.325
0.300
98
0.275
97
0.250
0.225
96
0.200
0.175
95
1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.150

Efficiency η (%)

100

Area-Related Power Density α (kW/dm2)
(b)

Power Losses (kW)

1.6
Realized
Prototype
1.2
1.0
0.8
0.6
0.4

Pareto Front
4

10 12 14 16 18 20
Stray Field (µT)
(c)

Coil Area
(dm2)
40.0
37.5
35.0
32.5
30.0
27.5
25.0
22.5
20.5
17.5
15.0

FIG 4 (a) The multiobjective power electronics optimization:
component and system models map the system parameters of
the design space onto the performance space, where a multidimensional Pareto surface limits the performance of the technology. The results of the multiobjective IPT coil optimization:
(b) the transmission efficiency h of a forced-air cooled
Pout = 50 -kW system with an air gap of 160 mm and a resonant
frequency of 85 kHz as a function of the area-related power
density a = Pout/Acoil; (c) the tradeoff between the total coil
losses and the magnetic stray field at 1.10-m distance from the
coil center.

transmission efficiency h is reduced as expected from
(1). In return, a smaller amount of active materials is
required for the more compact IPT coils, which results
in a lower coil weight and reduced material costs. At
the performance limit, the area-related power density a
can be increased only at the price of a lower transmission efficiency h. This design tradeoff is described by
the h-a -Pareto front.
Not captured by this analysis is the fact that the magnetic coupling and therefore the transmission efficiency

IEEE PowEr ElEctronIcs MagazInE

z September 2016

of smaller coils become more sensitive to the positioning of the EV. Only an overdimensioning of the transmitter coil and the power electronics or the use of an
array of multiple, selectively activated transmitters can
provide sufficient coil-misalignment tolerance. Consequently, there also exists a tradeoff between the coil
size, the complexity of the transmitter, and the positioning tolerance.
A low magnetic stray field presents the third key performance factor for a contactless EV charger. The IPT
system must comply with the relevant safety standards
for the magnetic field [31], [33], [34] in all regions that are
accessible to humans, i.e., in the passenger cabin and at
all sides of the vehicle. Because the passenger cabin is
shielded by the EV's metallic chassis, the main concern
is the stray field around the vehicle. In practice, the magnetic field would be measured with a field probe at test
locations within this area, at a fixed distance from the EV
specified by a standard.
For the multiobjective optimization of the 50-kW/
85-kHz IPT system, the influence of the EV chassis is
not included in the sense of a worst-case analysis. In the
compliance test measurements, the magnetic stray field is
sampled at a fixed distance of 1.10 m from the coil center
(the critical distance in the industry application at hand),
assuming that the IPT coil is mounted centrally below the
EV. The obtained results are shown in Figure 4(c). On the
indicated Pareto front, the stray field can be reduced only
if the IPT coils, i.e., the magnetic field source, are made
smaller to increase the distance from the current-carrying
conductors to the observation point [compare the coloring
in Figure 4(c)]. However, the smaller IPT coils lead to a
lower magnetic coupling and a reduced transmission
efficiency. Therefore, a design tradeoff results among
the magnetic stray field, the coil size, and the power
losses. To overcome this limit, additional shielding of
the IPT coils must be implemented, at the cost of additional system complexity.
The presented results do not include the power electronic converter, from which additional volume, weight,
and cost arise. At higher power levels, off-board conductive
dc chargers are typically connected to the vehicle at the
receiver-side dc link [compare U 2, dc in Figure 3(a)] for minimizing the number of on-board power conversion stages.
For the contactless charger, a larger fraction of the power
electronics must be located on the vehicle due to the intrinsic division of the system at the air gap between the IPT
coils. In addition, controlling power transfer with high efficiency over a wide load range requires a comparably high
level of complexity. Significant simplifications typically
lead to lower efficiency and, consequently, higher energy
cost for the EV charger, e.g., caused by switching losses
in the power converter [24]. Thus, for the on-board power
electronics of a high-power contactless charger, a higher
volume, weight, and cost are expected than for an offboard conductive dc charger. The effort for the vehicle-side

Table of Contents for the Digital Edition of IEEE Power Electronics Magazine Compendium - March 2018