IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 3, NO.
1, MARCH 2015 191
Coupling Coefficients Estimation of Wireless Power
Transfer System via Magnetic Resonance Coupling
Using Information From Either Side of the System
Vissuta Jiwariyavej, Member, IEEE, Takehiro Imura, Member, IEEE, and Yoichi Hori, Fellow, IEEE
Abstract Wireless power transfer via magnetic resonance
coupling method has opened a new possibility to the electric
vehicle system. It allows the wireless charging system of moving
vehicles, using charging lanes. However, although the efficiency of
power transmission is relatively high, the efficiency still depends
on displacement of coils. There have been several researches on
methods to maintain power transmission at the highest efficiency.
However, in such systems, the information on system parameter
especially coupling coefficients is needed, and in the charging lane
system, such information is unlikely to be obtainable without a
communication system. Therefore, it has come to attention that
parameter estimation is a crucial factor to implement a charging
lane system. This paper presents derivations of equations for esti-
mating coupling coefficients in several configurations of wireless
power transfer system, using information from only one side,
either the transmitting side or the receiving side, of the system.
The presented equations are both applicable to the case of single
receiving coil and are also generalized for the case of multiple
receiving coils. Each equation is verified by both simulations and
experiments. An experimental system of the coupling coefficient
estimation system is constructed for estimation from the receiving
side using a dc/dc converter.
Index Terms Coupling coefficient, magnetic resonance
coupling, parameter estimation, wireless power transfer.
I. I NTRODUCTION
R ECENTLY, the development of electric vehicles (EVs)
has been gaining more attentions from both consumers
and research community as a potential alternative to the tradi-
tional combustion engine cars. However, the main drawback of
EVs that still remains is their energy storage. Currently, used
batteries or capacitors have relatively small energy capacity;
and hence, EVs cannot travel across long distance. Researchers
have come up with various methods to compensate such a
drawback. One of the possible methods is to incorporate
wireless power transfer system into EV charging systems.
It will make the process of charging more convenient and
safer since it helps reduce the risk of electric shock. Wireless
Manuscript received January 15, 2014; revised April 3, 2014 and
May 20, 2014; accepted May 31, 2014. Date of publication June 19, 2014;
date of current version January 29, 2015. Recommended for publication by
Associate Editor Grant A. Covic.
V. Jiwariyavej is with the Department of Electrical Engineering, University
of Tokyo, Tokyo 852-8131, Japan (e-mail: pauhtar@gmail.com).
T. Imura and Y. Hori are with the Department of Advanced
Energy, University of Tokyo, Tokyo 852-8131, Japan (e-mail:
imura@hori.k.u-tokyo.ac.jp; hori@k.u-tokyo.ac.jp).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JESTPE.2014.2332056
2168-6777 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
power transmission also permits new solutions of dynamically
charging running vehicles.
There are several methods for wireless power transmission.
However, the magnetic resonance coupling method is believed
to be the most appropriate for EV charging system due to
this methods characteristics which allows power transmission,
that is nonradiative across longer power transfer distance with
relatively high efficiency while being robust to the coils
positional shift [1], [2]. Several methods, such as impedance
matching, have been proposed to maximize efficiency of
wireless power transfer system; and however, most of them
require the knowledge of systems parameters [3][5]. And
those that do not, due to the use of search algorithm, are still
not fast enough for the purpose of improving efficiency of
moving pick-ups. Therefore, it appeared that parameter esti-
mation is a crucial factor to make the system of charging lane
implementable. This paper presents derivations of equations
for estimation of coupling coefficients in several configurations
of wireless power transfer system, using information, such as
voltage and current from only one side, either the transmitting
side or the receiving side, of the system. In Section II, wireless
power transfer system topology used in this paper will be
described. Sections III and IV will present derivations of
estimation methods from transmitting side and from receiving
side, respectively. The presented equations are both applicable
to the case of single receiving coil and are also generalized
for the case of multiple receiving coils. Each equation is
verified by simulation results and experimental results shown
in Section V. In addition, finally, the system implementation
of coupling coefficient estimation system, for the case of
estimation from the receiving side, using a dc/dc converter
is discussed in Section VI.
II. W IRELESS P OWER T RANSFER S YSTEM T OPOLOGY
In this paper, wireless power transfer system is modeled
by equivalent circuit, in which the coil is represented as
the combination of an inductor and a capacitor in series
with an internal resistance [2]. The equivalent circuit, for the
configuration with one transmitting coil and one receiving
coil, is shown in Fig. 1(a). This equivalent circuit can also be
rewritten as the T-type equivalent circuit shown in Fig. 1(b).
Parameter L 1 , C1 , R1 , L 2 , C2 , and R2 represent inductances,
capacitances, and resistances of the transmitting coil and the
receiving coil, respectively. Power source impedance is Z s and
192 IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 3, NO. 1, MARCH 2015
Fig. 1. T-type equivalent circuit for wireless power transfer. (a) Equivalent
circuit. (b) T-type equivalent circuit.
load impedance is Z L . Note that this paper will only consider
the case when the load impedance is purely resistive, in other
words, when Z L does not have an imaginary part. Z in is the
input impedance when looking from the power source. Mutual
inductance, which is the main factor that decides how much
power is transferred, is represented by L m . And finally, V1 , I1
and V2 , I2 are defined as the voltage across and the current
going through the power source and the load, respectively,
immediately before the transmitting coil, with as the phase
difference between V1 and I1 .
Estimation of coupling coefficient can be done either using
only information from the source side or only information
from the load side, depending on the application. In the next
two sections, estimation equations will be presented.
III. E STIMATION E QUATION F ROM
THE T RANSMITTING S IDE
In this section, estimation equation using only information
from the source side will be presented. The advantage of
estimating coupling coefficient from the source side is that,
in the case of multiple EVs or receivers having entered the
charging system, if the coupling coefficients between the
transmitting coil and each receiving coils are known, not only
charging efficiency can be improved, but how much power to
distribute to each EV can also be decided using the method
introduced in [6].
Looking from the source side, the information we can
obtain are V1 , I1 , and . From these parameters, given
V1 = |V1 |e j (t +) and I1 = |I1 |e j t , another information that
Fig. 2. Equivalent circuit for a system with multiple receivers.
can be retrieved is the input impedance Z in
V1
Z in = (cos + j sin ) (1)
I1
where Re{Z in } = |V1 |/|I1 | cos and Im{Z in } = |V1 |/
|I1 | sin .
This Z in parameter can be calculated regardless of the
number of receivers. Therefore, estimation equation from the
source side will be derived mainly in terms of Z in .
A. With a Single Receiver
From the equivalent circuit, the input impedance Z in can be
calculated, and the derived relations are shown in (2) and (3)
when Re{Z in } and I m{Z in } represent real part and imaginary
part of Z in , and is defined as power sources operating
frequency, respectively
(L m )2 (Z L + R2 )
Re{Z in } = R1 + (2)
(Z L + R2 )2 + (Z A2 )2
(L m )2 (Z A2 )
Im{Z in } = Z A1 (3)
(Z L + R2 )2 + (Z A2 )2
where Z A1 = L 1 (1/C1 ) and Z A2 = L 2 (1/C2 ).
However, considering measurement taking, a reliable value
of imaginary part of Z in is quite difficult to obtain since
it varies quite drastically with respect to slightly shifted
frequency point. Therefore, it is more practical to use a
measurement of the real part in the estimation process. Hence,
deriving from (2), L m can be estimated using (4), assuming
load value and coil parameters are known

 
1 [Re{Z in } R1 ] (Z L + R2 )2 + Z 2A2
Lm = . (4)
Z L + R2
B. With Multiple Receivers
For the case of a system with multiple receivers, the input
impedance can be derived similarly. Equivalent circuit for
the system with multiple receivers is shown in Fig. 2, where
subscription 2 and 3 represent components of load 1 and 2,
respectively, and so on. In this paper, cross coupling between
each load will not be considered due to the fact that this system
JIWARIYAVEJ et al.: COUPLING COEFFICIENTS ESTIMATION OF WIRELESS POWER TRANSFER SYSTEM 193

is aiming for EV charging system, in which receiving coils in

each EVs will have negligible effect on each other. Z in can
be expressed as
  m+1

1 (L 1i )2
Z in = R1 + j L 1 + +
j C1 Z i + Ri + j L i + j C
1
i=2 i
(5)
where m = the number of receiving coils.
To solve (5) for all mutual inductance values, the necessary
number of linearly independent equations equals to the number
of loads. The controllable parameter in the transmitting side Fig. 3. V2 characteristics.
is the source frequency. Therefore, once enough information
is obtained by performing frequency sweep, mutual induc-
tance values can be calculated by solving equation systems
expressed as
Q

Z in(1) P1 11 Q 12 L 212
..
.. 2
Z in(2) = P2 + . . L 13 (6)

Q 21
.
.. .. .. ..
. . . Q mm
Fig. 4. 900 kHz experimental coil.
where
1 to limitation of estimation from load side using only a single
Pv = R1 + j v L 1 +
j v C1 set of information. However, estimation of coupling coefficient
2 from load side can still be performed with the same method as
Q uv = the case of multiple receivers, although additional information
Z u + Ru + j j L u + 1
j v C u is required.
and m = total number of receivers.
B. With Multiple Receivers
IV. E STIMATION E QUATION F ROM THE R ECEIVING S IDE In the case of multiple receiving coils, the estimation is
In this section, an estimation equation using only informa- currently done for a constant voltage source, assuming the
tion from the load side will be presented. The advantage of source voltage V1 is known. However, information from only
estimating coupling coefficient from the load side is when one instance is not enough to estimate the mutual inductance
there is more than one receiver in the system. With only parameter. However, if two sets of information, in this case,
one available source, it is more practical to put control from two voltages across the load for different load values, are
the load side so all loads can match with that one source. provided, the estimation can be done by
Although estimating from the load side can only achieve the
V1 Z 2a V2b (R2 + Z 2b ) Z 2b V2a (R2 + Z 2a )
information of mutual coupling between the source and itself L 12 = (8)
but not with other load, the estimation can be done with much V2a V2b (Z 2b Z 2a )
less information comparing with estimation from the source where V2a = voltage across load when load is Z 2a and
side when the number of loads increases. V2b = voltage across load when load is Z 2b .

A. With a Single Receiver V. S IMULATION AND E XPERIMENTAL R ESULTS

In the case of a single receiver, the relation between the
mutual coupling L m and the voltage across the load are
expressed, as shown in (7), when V1 is the source voltage
and V2 is the voltage across load Z 2
 shown in Table I.
 2
1 V1 V1
Lm = Z2 Z 2 + 4R1 (R2 + Z 2 ) . (7)
2 V2 V2
However, due to characteristics of V2 , which is that it has
a peak, as shown in Fig. 3, and which sign in (7) should
be used to perform estimation depends on which side of the
peak the system is currently at. Therefore, it is not possible
to determine L m with only one set of input sample that leads
To verify the proposed equation, a simple simulation is
performed. The coil that will be used in future experiments
is a short-type with self-resonant frequency of approximately
900 kHz, as shown in Fig. 4. The detailed coil parameters are
The simulation is performed for the following setup, which
is shown in Fig. 5. The transmitting coil is fixed in place,
while the receiving coil is placed at several distances away
from the transmitting coil. For the case of multiple receivers,
the second receiving coil is placed on the opposite side of
the transmitting coil to avoid cross coupling between the
two receivers. Note that in the actual application of charging
lane, two receivers equipped in different vehicles will be too
194 IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 3, NO. 1, MARCH 2015
TABLE I
E XPERIMENTAL C OIL PARAMETERS
Fig. 5. Wireless power transfer system setup.
Fig. 6. Reference value.
far apart, therefore cross coupling between them should be
negligible.
A. Estimation Result From the Transmitting Side With
a Single Receiver (Reference Value)
For this paper, since the coupling coefficient or k estimation
equation from the transmitting side is derived from direct back-
calculation, the k value obtained by calculation from this case
will be used as a reference value. The calculation is performed
for the case when resonant frequency is 900 kHz and load
impedance is 50 . The expected k value is shown in Fig. 6.
B. Estimation Result From the Transmitting Side
With Multiple Receivers
For this case, with enough information obtained by
frequency sweep, k values between the source and each
Fig. 7. Simulation result for k12 and k13 estimation from the transmitting
side.
Fig. 8. Experiment result for k12 and k13 estimation from the transmitting
side using frequency sweep. (a) k12 estimation result. (b) k13 estimation result.
receiver can be estimated. When there are two receivers,
information from two frequency points is necessary. The
frequencies used in simulation and experiments are 898, 900,
and 902 kHz, whereas load impedance is fixed at 50 . The
expected value of k12 and k13 are calculated in a simulation
whose result is shown in Fig. 7. The plot shows k12 and k13
versus the distance between the transmitting coil and the first
receiving coil which is moved to several distances. The result
of k12 matches with the reference value, while k13 is constant
as expected since the second receiving coil is fixed in place.
The experimental result is shown in Fig. 8. The experimental
result, except for the range of extremely short transmitting
gap, is mostly consistent with the expected value from the
simulation.
JIWARIYAVEJ et al.: COUPLING COEFFICIENTS ESTIMATION OF WIRELESS POWER TRANSFER SYSTEM
Fig. 9. Simulation result for k12 estimation from the receiving side.
Fig. 10. Experiment result for k12 estimation from the receiving side using
impedance sweep.
C. Estimation Result From the Receiving Side
Due to the fact that estimation equations are the same
whether there are only one receiver or multiple receivers,
only estimation result in the case of multiple receivers are
presented. More information on the receiving side can be
gained by changing load, or impedance sweep. The impedance
values used in the simulation and the experiment are 25, 50,
and 100 , whereas the transmitting frequency is fixed at
900 kHz. As mentioned in the earlier section that by estimation
from the receiving side, only the k between itself and the
source can be obtained. The simulation results in Fig. 9 shows
that the estimated k value is consistent with the reference
value regardless of the existence of an extra receiving coil. The
experimental result shown in Fig. 10 is also mostly consistent
with the simulation result. The estimation errors occur in
experiments at larger transmission gap is believed to be due
to radiation loss.
VI. S YSTEM I MPLEMENTATION U SING
A DC/DC C ONVERTER
In actual application of efficiency improving method, it is
more practical to have the load adjust its condition to match
with the source, especially when there is more than one
receiver existing in the system. In addition, as discussed in
Section V, among the proposed k estimation equations, the
estimation equation from the receiving side remains sim-
ple regardless of the number of receivers in the system
and is robust against stray capacitance, hence a promising
k estimation approach. Therefore, this paper will focus on
195
Fig. 11. Schematics of buck (top), boost (middle), and buckboost (bottom)
converters.
TABLE II
C OMPARISON B ETWEEN B UCK , B OOST,
B UCK B OOST DC/DC C ONVERTER
implementation of k estimator from the receiving side of the
wireless power transfer system.
As derived in Section IV, for k from the receiving side, the
necessary measurement is V2 , and the controllable parameter
is Z 2 . In actual application, it is not possible to have users
manually change load value of the vehicles. However, in terms
of power electronics, dc/dc converter is a simple circuit that
can achieve the task of changing voltage, hence the change in
load value. The possibility of using dc/dc converter to build k
estimating system will be investigated in this section.
A. Concept of DC/DC Converter
DC/DC converter is a circuit that converts a source dc
voltage to another dc voltage. There are three types of
switching mode dc/dc converters: buck, boost, and buckboost
converters. Buck converter produces the output voltage that is
lower than the input voltage, while boost converter produces
the output voltage that is higher than the input voltage.
Buckboost converter can produce the output voltage either
higher or lower than the source voltage. Simple structures
of these dc/dc converters are shown in Fig. 11. A compari-
son between the three types of converters is also shown in
Table II.
To enable load variation across broader range for estimation
purpose, buckboost converter is chosen for implementation of
k estimation system.
196 IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 3, NO. 1, MARCH 2015

Fig. 12. System simulation in Simulink. (a) Simulation setup for the case with one receiver. (b) Simulation setup for the case with two receivers.

B. System Simulation in MATLAB/SimPowerSystem

1) Simulation Setup: The simulation of ac system for
estimation from the load side using a dc/dc converter to
sweep load impedance is conducted in MATLAB/Simulink.
Load impedance is changed by adjusting the duty cycle
of the dc/dc converter. Simulink diagram is shown in
Fig. 12(a).
As mentioned in Section IV, for k estimation from the
receiving side, at least two sets of (V2 , Z L ) pair information
are needed. In this section, estimation result using different
pairs, one pair each time is presented.
In this system, wireless power transfer system has resonant
frequency of 900 kHz. The voltage on the receiving side
V2 is passed through diode-bridge rectifier before being fed
into the buckboost dc/dc converter. The duty cycle D of
dc/dc converter is changed to change Z L . For buckboost
converter, equivalent impedance Z L can be simply calculated
by the relation Z L = 50(1 D)2 /D 2 , but in nonideal system Fig. 13. Response of system simulation when using D = 0.3 and D = 0.7
where there are losses in components, the efficiency of dc/dc to estimate k12 .
converter will decrease, hence Z L obtained by such equation
will be significantly inaccurate. Therefore, in this system V2 and the calculated Z L are then substituted into (8) to
implementation, load value will be calculated by Z L = V2 /I2 . estimate coupling coefficient k. Since each pair of (V2 , Z 2 ) is
Duty cycle D is set to change every 2 ms. For each interval corresponding with one duty cycle value, for simpler reference,
of the same D value, the estimator will wait for a certain from this point on, the (V2 , Z 2 ) pair will be discussed in
amount of time, in this simulation for 1 ms, before reading V2 terms of duty cycle instead. Sample response of this system
value. Once two pairs of (V2 , Z 2 ) are obtained, the measured simulation is shown in Fig. 13.
JIWARIYAVEJ et al.: COUPLING COEFFICIENTS ESTIMATION OF WIRELESS POWER TRANSFER SYSTEM
Fig. 14. Simulation result of k12 estimation system using an ideal dc/dc
converter. (a) k12 estimation result with single receiver. (b) k12 estimation
result with two receivers. (c) k12 estimation result with three receivers.
Simulations for multiple receivers are also setup, as shown
in Fig. 12(b), for a system with two receivers and similarly
for a system with three receivers.
2) Simulation Result: Following, in Fig. 14, are simulation
results for the estimation system using an ideal buckboost
dc/dc converter. Each plane is the expected value of k for
k = 0.05 0.4, and cross marks represent estimated value
of k with respect to each pair of duty cycles. The range of
duty cycle used in the simulation is chosen to be between
197
Fig. 15. Experimental system diagram.
Fig. 16. Experimental system built on a test board.
0.3 and 0.7 as it gives the range equivalent impedance of
approximately between 10 and 300 . The result shows that
estimated values match well with the expected value of k.
Hence, the feasibility of constructing estimation system using
dc/dc converter is confirmed.
C. Experimental System
The system is realized using a dc/dc converter. A controller,
both for switching signal and for k estimation, is implemented
using Arduino. System diagram is shown in Fig. 15. The test
system is first built on a test board, as shown in Fig. 16.
Estimation result for the case of the system with one receiver
and two receivers are shown in Fig. 18(a) and (b), respectively.
Estimation of coupling coefficient has been conducted with
the constructed experimental system. Experimental setup is
shown in Fig. 17 to avoid cross coupling between different
receivers. The distance of the coil was set for when k is 0.1,
0.2, and 0.3. Results in this section shows calculation using
direct measurement of V and I from the receiving side of
the wireless power transfer system without any measurement
filter.
Although the estimated values seem to be consistent regard-
less of which duty cycle pair is used for estimation, estimation
results are significantly inaccurate. The reason for estimation
198 IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 3, NO. 1, MARCH 2015

Fig. 17. Experimental setup with k estimating system using a dc/dc converter.

Fig. 18. Experimental result of k12 estimation system using experimental

system. (a) k12 estimation result with single receiver. (b) k12 estimation result
with two receivers.

error is believed to be due to nonideal component in the

system. Possible causes of estimation error are discussed as
follows.
1) Diode Rectifier: Rectifier, in this paper, is implemented
using SiC diodes for high-frequency input. The output voltage
from rectifier is expected to have voltage drop of approxi-
mately two times of the forward voltage of the diode (V f ). The
measurement for when input voltage is from 1 to 10 V shows
Fig. 19. Diode rectifier characteristics. (a) Diode rectifier output versus input.
(b) Voltage drop in a diode rectifier.
the expected voltage drop, as shown in Fig. 19. Such voltage
drop will also cause the change in equivalent impedance seen
by the source, which is also one of the parameter in the
estimation equation.
JIWARIYAVEJ et al.: COUPLING COEFFICIENTS ESTIMATION OF WIRELESS POWER TRANSFER SYSTEM
Fig. 20. Buck-boost converter characteristics.
2) DC/DC Converter: For functionality test of the
buckboost converter, Vout /Vin at different duty cycle is mea-
sured and calculated, when Vin and Vout are the input to and
the output from the dc/dc converter, respectively. Up until now,
ideal condition has been used for calculation; and therefore,
the theoretical ratio of Vout /Vin should be according to (9),
where D is duty cycle
Vout D using magnetic resonant couplings, in Proc. 35th Annu. Conf. IEEE
= . (9)
Vin 1 D
However, the actual measurements for when Vin is from
1 to 10 V show that the ratio tends to decrease when duty
cycle is high. Such a drop in voltage ratio is due to inductors
winding resistance R L . By waveform analysis, the voltage
ratio for nonideal buckboost converter can be expressed
in (10). Fig. 20 shows voltage ratio of ideal case, voltage
ratio obtained from the actual measurement of the experi-
mental setup, and the calculated voltage ratio from nonideal
converter. The measurements is consistent with the case when
R L = 1.8
Vout D 1
= . (10)
Vin 1 D 1 + D 2 R L2
(1D) R
However, such deterioration in efficiency should not cause
deviation in relation between V2 and Z 2 which is crucial to
estimation equation.
3) Measurement Error: In addition to hardware imperfec-
tion, other possible cause for estimation error is measurement
error. This is the most likely cause of error since the values of
current is very small, therefore even small measurement error
can result in large percentage comparing with the absolute
value of the current. The countermeasure, such as using
least square method to filter out measurement noise, will be
investigated in future works.
VII. C ONCLUSION
This paper presents derivations of equations for estimating
coefficients in several configurations of a wireless power
199
transfer system, mainly for a single transmitting coil with
single and multiple receiving coils. Since the estimation is
aiming for usage in dynamic EV charging system, the ideal
solution is to be able to back-calculate the value of coupling
coefficient directly from the measurement. Such calculation
can be done when estimation is done from the source side in
one transmitting and one receiving coil configuration. How-
ever, in other cases, more information is needed. The derived
relations show that, estimation of coupling coefficient is still
possible theoretically by sweeping controllable parameter to
obtain more linearly independent information. The proposed
equations are verified with simulation results and basic exper-
iments were conducted. Since estimation from the receiving
side is more practical for actual implementation, a test system
was implemented using a dc/dc converter. The implementation
of the system is proved feasible. However, the measurement
error has caused some inaccuracy in estimation of coupling
coefficient.
Future work will include investigation of integration of fil-
tering techniques, such as least squares method into the system
to improve estimation accuracy, and more detailed comparison
between the proposed method and other k estimation methods.
R EFERENCES
[1] A. Kurs, A. Karalis, R. Moffatt, J. D. Joannopoulos, P. Fisher, and
M. Soljacic, Wireless power transfer via strongly coupled magnetic
resonances, Science, vol. 317, no. 5834, pp. 8386, Jun. 2007.
[2] T. Imura, H. Okabe, T. Uchida, and Y. Hori, Study on open and short
end helical antennas with capacitor in series of wireless power transfer
Ind. Electron., Nov. 2009, pp. 38483853.
[3] C.-S. Wang, O. H. Stielau, and G. A. Covic, Design considerations for a
contactless electric vehicle battery charger, IEEE Trans. Ind. Electron.,
vol. 52, no. 5, pp. 13081314, Oct. 2005.
[4] O. H. Stielau and G. A. Covic, Design of loosely coupled inductive
power transfer systems, in Proc. Int. Conf. Power Syst. Technol., vol. 1.
2000, pp. 8590.
[5] Y. Moriwaki, T. Imura, and Y. Hori, Basic study on reduction of
reflected power using DC/DC converters in wireless power transfer
system via magnetic resonant coupling, in Proc. IEEE 33rd Int.
Telecommun. Energy Conf. (INTELEC), Oct. 2011, pp. 15.
[6] K. K. Ean, B. T. Chuan, T. Imura, and Y. Hori, Novel band-pass
filter model for multi-receiver wireless power transfer via magnetic
resonance coupling and power division, in Proc. IEEE 13th Annu.
Wireless Microw. Technol. Conf. (WAMICON), Apr. 2012, pp. 16.
[7] F. Z. Shen, W. Z. Cui, W. Ma, J. T. Huangfu, and L. X. Ran, Circuit
analysis of wireless power transfer by coupled magnetic resonance,
in Proc. IET Int. Commun. Conf. Wireless Mobile Comput. (CCWMC),
Dec. 2009, pp. 602605.
[8] K. Onizuka, N. Oodachi, H. Shoki, and O. Watanabe, Coupling factor
estimation for efficiency optimization on wireless power transfer, in
Proc. IEICE General Conf., Mar. 2010, no. B-1-29, p. 29.
[9] H. Shoki, Trend of wireless power transmission technology and
approaches for commercialization, The Institute of Electronics,
Information and Communication Engineers, Tokyo, Japan, Tech.
Rep. WPT2010-07, 2010.
[10] S. Valtchev, B. Borges, K. Brandisky, and J. B. Klaassens, Resonant
contactless energy transfer with improved efficiency, IEEE Trans.
Power Electron., vol. 24, no. 3, pp. 685699, Mar. 2009.
[11] G. Huotao, Z. Xia, and L. Jie, Estimation of mutual coupling coefficient
by simulated annealing algorithm, in Proc. 6th Int. Symp. Antennas,
Propag. EM Theory, Oct./Nov. 2003, pp. 275278.
[12] S. Abe and Y. Kaneko, Contactless power transfer system,
(in Japanese), J. Inst. Electr. Eng. Jpn., vol. 128, no. 12, pp. 796799,
2008.
[13] Z. Bingyi, L. Hongbin, Z. Yisong, Y. Yong, and F. Guihong, Contactless
electrical energy transmission system using separable transformer,
in Proc. 8th Int. Conf. Electr. Mach. Syst. (ICEMS), vol. 3. 2005,
pp. 17211724.
200 IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 3, NO. 1, MARCH 2015
[14] H. Ayano, K. Yamamoto, N. Hino, and I. Yamato, Highly effi-
cient contactless electrical energy transmission system, in Proc. IEEE
28th Annu. Conf. Ind. Electron. Soc. (IECON), vol. 2. Nov. 2002,
pp. 13641369.
[15] A. J. van den Boom, System Identification: On the Variety and
Coherence in Parameter- and Order Estimation Methods. Eindhoven,
The Netherlands: Technische Hogeschool Eindhoven, 1982.
[16] F. Sato, J. Murakami, H. Matsuki, S. Kikuchi, K. Harakawa, and
T. Satoh, Stable energy transmission to moving loads utilizing new
CLPS, IEEE Trans. Magn., vol. 32, no. 2, pp. 50345036, Sep. 1996.
[17] D. Kim, D. Son, and B. Choi, Input impedance analysis of PWM DC-
to-DC converters, in Proc. 21st Annu. IEEE Appl. Power Electron. Conf.
Expo. (APEC), Mar. 2006, p. 8.
[18] S. G. Kriventsov and J. S. Mayer, An exact expression for the input
impedance of the buck converter in continuous conduction mode, in
Proc. IEEE 32nd Annu. Power Electron. Specialists Conf. (PESC), vol. 1.
2001, pp. 351356.
Vissuta Jiwariyavej (S09M14) received the B.E.
degree in electrical engineering and computer sci-
ence from Dartmouth College, Hanover, NH, USA,
in 2009, and the M.E. degree in electrical engineer-
ing from the University of Tokyo, Tokyo, Japan, in
2009 and 2013, respectively.
Her current research interests include wireless
power transfer, in particular, electric vehicles appli-
cation.
Takehiro Imura (S09M11) received the B.S.
degree in electrical and electronics engineering from
Sophia University, Tokyo, Japan, in 2005, and the
M.S. degree in electronic engineering and the Ph.D.
degree in electrical engineering from the University
of Tokyo, Tokyo, in 2007 and 2010, respectively.
He joined the Department of Advanced Energy,
Graduate School of Frontier Sciences, University of
Tokyo, as a Research Associate in 2010. His current
research interests include the wireless power transfer
for electric vehicles using electromagnetic resonant
couplings.
Yoichi Hori (S81M83SM00F05) received the
B.S., M.S., and Ph.D. degrees in electrical engineer-
ing from the University of Tokyo, Tokyo, Japan, in
1978, 1980, and 1983, respectively.
He joined the Department of Electrical Engineer-
ing, University of Tokyo, as a Research Associate
in 1983, where he became an Assistant Professor,
an Associate Professor, and a Professor in 2000.
In 2002, he joined the Information and System
Division, Institute of Industrial Science, University
of Tokyo, as a Professor, and the Department of
Advanced Energy, Graduate School of Frontier Sciences, University of Tokyo,
in 2008. From 1991 to 1992, he was a Visiting Researcher with the University
of California at Berkeley, Berkeley, CA, USA. His current research interests
include control theory and its industrial applications to motion control,
mechatronics, robotics, and electric vehicles.
Prof. Hori is a past AdCom member of the IEEE Industrial Electronics
Society. He has been the Treasurer of the IEEE Japan Council and Tokyo
Section since 2001. He is also a member of the Society of Instrument
and Control Engineers, the Robotics Society of Japan, the Japan Society
of Mechanical Engineers, and the Society of Automotive Engineers of
Japan. He is the Past President of the Industry Applications Society of the
Institute of Electrical Engineers of Japan (IEEJ), the President of Capacitors
Forum, the Chairman of Motor Technology Symposium of Japan Management
Association, the Director on Technological Development of SAE-Japan, and
the Director of the Japan Automobile Research Institute. He was a recipient
of the Best Transactions Paper Award from the IEEE T RANSACTIONS ON
I NDUSTRIAL E LECTRONICS in 1993 and 2001, the Best Transactions Paper
Award from the IEEJ in 2000, and the Achievement Award of the IEEJ
in 2011.