Politecnico di Torino
Porto Institutional Repository
[Proceeding] Load identification in dynamic wireless power transfer system
utilizing current injection in the transmitting coil
Original Citation:
Khalilian, Mojtaba; Rosu, Stefan George; Cirimele, Vincenzo; Guglielmi, Paolo; Ruffo, Riccardo
(2016). Load identification in dynamic wireless power transfer system utilizing current injection in
the transmitting coil. In: 2016 IEEE Wireless Power Transfer Conference (WPTC), 5-6 May 2016.
pp. 1-4
Availability:
This version is available at : http://porto.polito.it/2644844/ since: July 2016
Publisher:
IEEE
Published version:
DOI:10.1109/WPT.2016.7498793
Terms of use:
This article is made available under terms and conditions applicable to Open Access Policy Article
("Public - All rights reserved") , as described at http://porto.polito.it/terms_and_conditions.
html
Porto, the institutional repository of the Politecnico di Torino, is provided by the University Library
and the IT-Services. The aim is to enable open access to all the world. Please share with us how
this access benefits you. Your story matters.
(Article begins on next page)
Load Identification in Dynamic Wireless Power Transfer System
Utilizing Current Injection in the Transmitting Coil
Mojtaba Khalilian1, Stefan George Rosu2, Vincenzo Cirimele1, Paolo Guglielmi1 and Riccardo Ruffo1
1
Politecnico di Torino, Department of Energy, Torino, Italy
University Politehnica of Bucharest, Bucharest, Romania
mojtaba.khalilian@polito.it, stefan.g.rosu@ieee.org, vincenzo.cirimele@polito.it, paolo.guglielmi@polito.it
2
Abstract—In this paper, a load identification approach for
dynamic wireless power transfer for electric vehicles is
introduced. In the proposed method, determination of the load
parameters and position are achieved without requiring any
communication link between the transmitter and receiver or
using any extra coil. The load detection procedure is achievable
using two modes: the energy injection mode and the free resonant
mode. The operating principles and theoretical analysis of this
method are presented in this paper, together with experimental
results.
Keywords—Load
resonant converter.
identification;
I.
wireless
power
transfer;
INTRODUCTION
Electric vehicles (EVs) have not yet been widely accepted
by the consumer due to the several problems such as price,
driving range, and charging time. Wireless Power Transfer
(WPT) technology is a practical way to mitigate these
problems. This technology allows the power transfer between a
fixed coil named transmitter placed on or under the road
pavement and a second one placed under the vehicle floor
named receiver [1]-[3].
Dynamic IPT/WPT is the extension of this technology
toward the charge of the vehicle during the motion. This is
achieved by spreading out the transmitter coils over a dedicated
lane. Three different generic roadbed geometries, long wire
loop, sectioned wire loops, and spaced loops, are investigated
in [4]. The overall efficiency of the system is strongly affected
by the coupling coefficient. A small loop size increases the
system efficiency and can reduce the supply voltage and
reactive power requirements [4] but the power transfer must
begin at each transmitter coil in a safe way. One major problem
in dynamic wireless power transfer is accurately detecting the
electric vehicle as it approaches and passes along the
transmitter coils. When the electric vehicle is approaching a
transmitter, the power supply controller has to enable power
transfer by energizing the receiver coil in correspondence to its
arrival over the transmitter. On the other hand, when no load is
detected, the transmitter is no longer energized and the power
transfer has to be stopped or avoided. This aspect influences
the continuity of the power transmission and represents an
important part of the charging operation. Furthermore, if a
conductive material is near the energized primary coil it must
be detected in order to avoid additional losses or damages.
One possible solution for load identification is to use
communications between the transmitter and the receiver using
a Radio Frequency (RF) link in order to determine the position.
However, this method may have a delay that will result in large
current peaks in the coils or decreasing the time that the EV
can be charged.
Several works have dealt with these aspects [5]-[9]. In [5]
the authors treat a similar problem by applying the Particle
Swarm Optimization (PSO) algorithm to identify the load
parameters in an induction heating application. In [6], multiple
coils mounted on the primary and secondary couplers are used
to detect the position of the EV. Paper [7] introduces a load
detection method for the voltage-fed IPT used in induction
heating application systems by using the energy injection
mode. In [8], a load detection scheme has been proposed for a
wireless power transfer system based on the observation of the
transmitter coil voltage and supply current. In this paper, a load
position and identification approach for dynamic WPT is
proposed. The determination of the operating mode is achieved
without requiring any communication link between the
transmitter and the receiver or using any extra coil but adopting
a specific detection method directly using the power electronics
connected to each transmitter. The evolution of the envelope of
the peak value of the transmitter resonant current is used to
detect the position or type of load. The operating principles and
theoretical analysis are presented in this paper together with an
experimental validation.
II.
WPT OPERATION DESCRIPTION
A simplified scheme of the WPT system is illustrated in
Fig. 1. It is composed of an input DC voltage source, a full
bridge inverter, transmitter and receiver coils, compensation
capacitors C1 and C2, a diode rectifier at the receiver side with
a filter capacitor Cf connected to the EV battery. The capacitors
C1 and C2 are selected in order to obtain the same resonant
frequency for the primary and the secondary sides. This choice
allows compensating the impedance of the coils improving the
power transfer capability. The equivalent impedance of an IPT
transformer is defined based on the following equation:
Z eq


1
ω2M 2
= R1 + jωL1 +
+
jωc1  R + R + jωL + 1
2
L
 2
jωC 2


 (1)


Where R1 is the winding resistance of the transmitter, L1 is
its self-inductance, M is the mutual inductance of coupled
inductors, R2 is the winding resistance of the receiver, L2 is
self-inductance and RL is the equivalent load resistance [10].
978-1-4673-7986-1/16/$31.00 ©2016 IEEE
S1
S3
iL1
VDC
iL2
L1
L2
S4
Do3
Cf
C2
C1
S2
Do1
Do2
Vout
Do4
Fig. 1. Scheme of the proposed series-series compensated IPT system
If the converter works at resonant frequency, the equivalent
resistance seen from the primary is:
Req = R1 +
III.
ω2M 2
R2 + RL
(2)
Fig. 2. Simulation of energy injection mode and free resonant mode
LOAD DETECTON METHOD
The load detection process is composed of two parts; the
energy injection mode and observation of the envelope of the
peak values of the resonant current in the free resonant mode.
The current in the transmitter during the energy injection mode
and free resonant mode is presented in Fig. 2. The injection
mode is realized by applying a voltage impulse to the
transmitter. Based on Fig. 2, the energy injection starts at t0. In
this mode, the inverter switches are working in the
conventional mode and the energy is stored in the reactive
elements of the system. When the current injection mode ends,
the free resonant mode starts at t1. For a typical WPT system
the time interval t0 - t1 is a fraction of the resonant period. In
free resonant mode, the load can be identified. In this mode,
lower switches of the inverter turn on at the same time. S1 and
S3 turn off and S2 and S4 turn on and the current freewheels in
the circuit. In free resonant mode, the injected power is
dissipated in the resistances of the primary circuit or, if the load
is present, transferred to the load. By writing the differential
equation of the primary circuit in free resonant mode, the
primary current equation can be written as follows [7]:
i1 (t ) =
α=
ω 0 − αt
e cos(ω1t + θ)
L1ω1
(3)
ω L
ω1
ω
, ω1 = ω0 2 − α 2 , θ = arccos 1 , Q1 = 0 1 (4)
Req
2Q1
ω0
Where, ω1 is the operating frequency in free resonant mode.
The waveform of the primary current is shown in Fig. 3. As
visible, the current decreases exponentially with the behavior
described by (3). If the values of R1, R2 and L1 remain constant,
equations (3) and (4) indicate that the damping coefficient is
related only to RL and M. If the load resistance increases, the
damping coefficient increases and thus, the time needed to
reach a minimum set value of the peak current decreases.
Furthermore, the damping coefficient increases when mutual
inductance increases.
When the primary side and the secondary side are separated
from each other, M is equal to zero so the damping coefficient
is in the minimum value and thus, envelope waveform of i1 will
decay in a longer time.
Fig. 3. Simulation of primary current in free resonant mode
At this time, no load position detection exists, but a foreign
material that increases the equivalent resistance could be
detected. Thus, the detection process continues until a load is
detected. When the receiver reaches the transmitter, M
increases causing the increasing of the damping coefficient.
According to this, i1 envelope waveform declines. This
reduction is used by the control circuit to identify the presence
of the load. The equation of the peak values of the current i1
can be written as:
i1 (n) =
1 − αt n
2π(n + 1)
, tn =
e
ω0
L1
(5)
Where n is an integer defined as the nth oscillation of the
primary current. If (5) is used for two different n, then:
−α(
2 π ( n1 +1)
)
ω0
i1 (n1 ) e
=
2 π( n2 +1)
i1 (n2 )
−α(
)
ω0
e
(6)
By simplifying (6), α can be rewritten as:
α=
ω0 (ln(i1 (n1 )) − ln(i1 (n2 )))
2π(n1 − n2 )
(7)
For foreign material identification, the peak values of i1
should be obtained and then, the damping coefficient is
calculated by (7). By using (4), Q1 and Req are obtained. Finally
by (2), the load resistance equation can be obtained:
RL =
ω0 2 M 2
− R2
(ln(i1 (n1 )) − ln(i1 (n2 )))ω0 L1
− R1
π(n1 − n2 )
I.
(8)
EXPERIMENTAL RESULTS
In order to validate the theoretical analysis, experimental
tests have been conducted. In these tests, the supply voltage for
the full bridge inverter is 200 V DC. A combination of Nchannel SiC power MOSFET SCT2080KE and IGBT
IRG7PH46U is used for inverter switches to reduce conduction
losses. The experimental parameters are: C1=15 nF, C2=30 nF,
L1= 240 μH, L2= 120 μH, M= 18 μH, R1=0.5 Ω, R2=0.35 Ω,
and inverter switching frequency is equal to 84 kHz. The
pictures of the implemented system are shown in Fig. 4 and
Fig. 5. Two sets of experimental test are done to show the
effectiveness of the proposed load identification method. For
all tested situation a single pulse small amplitude squared
voltage has been applied to the primary side.
Fig. 4. Implemented power converter
In the first test, the output of the diode rectifier connected to
the receiver is short circuited. The primary and secondary side
currents for different positions of the load are shown in Fig. 6
and Fig. 7 respectively. When receiver coil is far from the
transmitter (M=0), no energy is transferred to the secondary
and thus, change of the current envelope is not high.
Fig. 5. Picture of transmitter and receiver
1
M=0 coils are partially coupled
0.5
0
-0.5
M=Max
-1
0
0.5
1
Time (s)
1.5
10-4
Fig. 6. Experimental results: first configuration, primary current with diode
bridge
Fig. 8. Experimental results: second configuration, primary current without
diode bridge
Fig. 7. Experimental results: first configuration, secondary current with
diode bridge
Fig. 9. Experimental results: second configuration, secondary current without
diode bridge
If a load is present, the envelope of the peak current has a
fast variation after the first oscillations. The first oscillations
will transfer some power to the secondary however, because
of the presence of the diode bridge, there will be no power
sent back to the transmitter. The amplitude of the envelope of
the remaining oscillations in the primary will depend on the
coupling factor. Based on this, a minimum coupling factor
could be defined at which the power transfer will begin.
In the second test, full bridge diode in the secondary is
removed and the secondary is short circuited. Thus, in the
secondary, there is only a resonant circuit composed of L2, C2
and parasitic resistance. Currents of primary and secondary
coils for two different positions of the load for this test are
shown in Fig. 8 and Fig. 9. The absence of the diode bridge
allows the power that is not lost as heat in the parasitic
resistance of the receiver circuit to be fed back to the
transmitter. The slope of the envelope of the peak current
oscillations will depend on the coupling factor. It should be
mentioned that the same behavior is obtained if the short
circuit is made with two MOSFET transistors in a full
MOSFET bridge. The small oscillations of the secondary
current visible in Fig. 9, appear also in the case M=0
according to the proximity of the receiver due to the
dimensions of the laboratory.
II.
CONCLUSION
In this paper, a new load identification method for wireless
power transfer system is introduced. The advantages of this
method are that it does not require any communication link
between the primary and secondary coil and it does not use any
extra coil. By using the full bridge inverter, a square-wave
voltage impulse at output of the inverter is generated. Then,
primary coil and primary capacitor start resonating. The
evolution of the envelope of the peak values of the primary
current in the free resonant mode is used by the control circuit
to detect the position or type of load. The operating principles
and theoretical analysis was presented in this paper.
Experimental results were also presented to prove the validity
of theoretical analysis.
ACKNOWLEDGMENT
This work has been co-funded by the European Union’s 7th
Framework Program for Research through the FABRIC project
(No. FP7 605405).
REFERENCES
[1]
F. Musavi and W. Eberle, “Overview of wireless power transfer
technologies for electric vehicle battery charging”, IET Power
Electronics, vol. 7, no. 1, pp.60-66, 2014.
[2] S. Y. Choi, B. W. Gu, S. Y. Jeong and C. T. Rim, “Advances in wireless
power transfer systems for roadway-powered electric vehicles”, IEEE J.
Emerg. Select. Topics Power Electron., vol. 3, no. 1, pp.18 -36, 2015.
[3] S. Li and C. C. Mi, “Wireless power transfer for electric vehicle
applications” , IEEE Journal of Emerging and Selected Topics in Power
Electronics, vol. 3, no. 1, pp.4 -17, 2015.
[4] M. Yilmaz, V.T. Buyukdegirmenci and P. T. Krein, "General design
requirements and analysis of roadbed inductive power transfer system
for dynamic electric vehicle charging", Proc. IEEE ITEC , pp.1-6, 2012.
[5] A. Dominguez, A. Otin, I. Urriza, L. A. Barragan, D. Navarro, and J. I.
Artigas, ‘‘Load identification of domestic induction heating based on
Particle Swarm Optimization," Proc IEEE COMPEL, pp. 1-6, 2014.
[6] G. Nagendra, G. Covic and J. Boys , "Detection of EVs on IPT
highways", IEEE J. Emerg. Sel. Topics Power Electron., vol. 2, no.
3, pp.584 -597 , 2014.
[7] Z.-H. Wang, Y.-P. Li, Y. Sun, C.-S. Tang and L. Xiao , “Load detection
model of voltage-fed inductive power transfer system”, IEEE Trans.
Power Electron. , vol. 28 , no. 11 , pp.5233 -5243 , 2013.
[8] D. Lin, J. Yin and S. Y. Hui , “Parameter identification of wireless
power transfer systems using input voltage and current” , Proc. IEEE
ECCE 2014 , pp.832 -836, 2014.
[9] M.C. Stevens, A.C. Knill, and J.R. Dunton,, “Controlling inductive
power transfer systems”, U.S. Patent 8,610,400 B2, Dec. 17, 2013
[10] V. Cirimele, S.G. Rosu, P. Guglielmi, F. Freschi, “Performance
evaluation of wireless power transfer systems for electric vehicles using
the opposition method,” Proc IEEE RTSI, pp.546-550, 16-18 Sept. 2015.