Efficiency Enhancement for Dynamic Wireless Power Transfer System ...

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2015.2508721, IEEE Transactions on Transportation Electrification

Efficiency Enhancement for Dynamic Wireless Power Transfer System with Segmented Transmitter Array J.P.K. Sampath1, D.M. Vilathgamuwa2, A. Alphones1 School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 2 School of Electrical Engineering and Computer Science, Queensland University of Technology, Brisbane, Australia 1

Abstract—Achieving high efficiency with improved power transfer distance and misalignment tolerance is the major design challenge in realizing dynamic Wireless Power Transfer (D-WPT) systems. This paper provides an analysis on designing D-WPT systems. Design parameters such as number of Tx coils, separation between Tx, operating frequency and load characteristics are analyzed with respect to efficiency for the D-WPT system with segmented transmitter array. A double spiral repeater (DSR) is proposed for improving efficiency, enhancing transfer distance and misalignment tolerance. Experimental results of the proposed topology with DSRs show efficiencies of 81% and 60% at normalized transfer distances (normalized to geometric-mean of Tx and Rx sizes) of 0.74 and 2.2 respectively. The proposed topology can be effectively used to alleviate efficiency deterioration against transfer distance and misalignment in D-WPT systems. Index Terms— Dynamic wireless power transfer, Magnetic resonance coupling, Double spiral repeater, segmented transmitter coil array.

I. INTRODUCTION WIRELESS POWER TRANSFER (WPT) systems have been extensively investigated in the last few years. WPT technology can be used in a wide range of industrial applications such as biomedical implants, robotics, consumer electronics and electric vehicle (EV) charging to eliminate charging hazards and drawbacks related to cables. Dynamic WPT (D-WPT) enables the receiver to be charged continuously while in motion. D-WPT applications include dynamic charging for EVs [1-6], biomedical implants, robots [7] and overhead conveyers. D-WPT for EV charging has many advantages compared to wired-EV charging. Inconveniences in wired-EV charging process are a major impediment in their widespread adoption. From the receiver viewpoint, D-WPT theoretically solves the battery problem as it can provide an unlimited driving range. However, the employment of such systems is reliant on the infrastructure development, which in turn, is limited by its cost. Although many researchers have been working on D-WPT technology, still there are numerous challenges to overcome in the process of bringing it to commercial level. Acceptable power transfer efficiency (PTE), transferred power (TP) at high transfer range, misalignment tolerance and safety considerations are major technical challenges. Both the maximum power transfer operation and the maximum efficiency conditions are used as design goals in the literature [8]. However, both PTE and TP cannot be optimized simultaneously in classical WPT systems. If TP is chosen as the first optimization goal in the design process, the maximum achievable efficiency will be limited, whereas, if PTE is

prioritized, some other techniques such as adaptive impedance matching (IM) [9] can be used to improve the TP. Therefore, the efficiency optimization is considered as the preliminary design objective of this study.

Receiver Transmitter array Fig. 1. Example of D-WPT application: Dynamic EV charging

D-WPT for EVs can be implemented using either single coil transmitter track [5] or segmented coils in a transmitter array [1]. Transmitter track based systems are easier to control as the track is powered from a single source. Coupling coefficient along the track is nearly constant as the receiver (Rx) moves along the track. The transmitter track can be a few meters to several tens of meters long [5]. However, this type of design suffers from a fairly low coupling coefficient because of the smaller transmitter region covered by the Rx thus giving rise to a lower efficiency. Moreover, electromagnetic field emitted in the uncoupled region has to be suppressed to mitigate harmful exposure. On the other hand, segmented transmitter (Tx) coils give higher coupling coefficient and efficiency compared to Tx track based methods. Nevertheless, misalignment tolerance, position tracking, transmitter separation and controlling the power flow are design challenges that need to be solved in segmented Tx designs. There are two methods in connecting the power source to segmented Tx coils. Several Tx coils can be connected to a single power source in parallel, or dedicated sources can be used to supply power to each coil. Phase difference between adjacent source inverters degrades the system performance [10] in case of dedicated multi-source systems, introducing additional complexity. Furthermore, the connection of dedicated power source for each coil is not cost effective. In contrast, connecting multiple Tx coils to a single source is relatively simpler and cost effective. Another approach of dedicated source configuration is to use single source connected to multiple coils, and switch between each coil when the vehicle moves along the Tx array using a switch box [2]. In addition to source configuration, separation between transmitters needs to be carefully optimized. If coils are placed far apart, efficiency reduces drastically when the Rx moves away from the transmitter and the power transfer will not be continuous. However, coils cannot be kept too close to each other because the system cost will be increased with the

0885-8993 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

requires

IEEE

permission.

See


increased number of transmitters placed in a given length of the track. In addition, placing Tx coils too close to each other may experience lower performance due to the cross coupling effect. Fig. 1 shows a classical arrangement of Tx coils connected as an array in the implementation of a segmented D-WPT system. Misalignment problem dominates the technical challenges in designing D-WPT systems with segmented Tx arrays [1]. In [11], the authors present a method for segmentation and compensation of Tx coils for D-WPT. However, literature still lacks a comprehensive analysis on design parameters such as number of Tx coils powered simultaneously, separation between Tx coils and, frequency and load characteristics. Therefore, the requirement of such an analysis on segmented D-WPT system is a vital aspect in optimal design process. This paper presents detailed analysis on designing D-WPT system using segmented Tx array. High frequency operation is preferred as it generates a strong magnetic field and induced emf for a given primary current. In addition, high frequency operation is desireable in achieving high quality factor coils for highly efficient WPT. However, operating frequency is mainly restricted by the device limitations. The advent of modern semiconductor devices such as wide band gap transistors has paved way for increasing the operating frequency of power converters tremendously. Therefore, cost effective and higher power devices operating in high frequency for EV applications will be available in near future [12-14]. Therefore, the frequency of operation of the proposed D-WPT system is chosen to be around 1MHz. This paper analyses the frequency and load characteristic of D-WPT system with segmented Tx array. The effect of number of Tx (NTX) coils powered at a time is studied. Frequency splitting phenomena that is investigated for NTX >1 has not been studied previously. A novel double spiral repeater (DSR) scheme is proposed for efficiency and misalignment tolerance improvement. Recommendations for designing future D-WPT systems are given. The paper is organized as follows. Section II introduces analysis of D-WPT system using equivalent circuit theory and coil parameter calculations. D-WPT using an array of Tx resonators is discussed in Section III. Frequency and load characteristic of the system is also evaluated. Subsequently, the system with focusing DSR is introduced and its performance is analyzed in Section IV. Detailed analysis of the proposed system is presented along with the experimental results followed by the conclusions. II. PARAMETER CALCULATION AND ANALYSIS A. Equivalent circuit analysis For the system analysis, the equivalent circuit method with multiple transmitters, repeaters and receiver coils has been used as it is well established in the literature [15]. Fig. 2(a) shows the equivalent circuit of dedicated power source configuration for each coil and Fig. 2(b) illustrates the single power source configuration whereby three nearby transmitters are powered simultaneously. Power sources are modeled as voltage sources and all excitation voltages are chosen to be equal in the analysis (Vt1 = Vt2 = Vt3 = Vs). Equivalent source resistance represents the output impedance of the power source. The notion of source resistance is

sometimes misleading in literature, particularly studies focus on IM or two port network model [9] use 50Ω source impedance. However, source impedance is much smaller in value when power converters are used as the source [16]. Therefore, equivalent circuits are considered to be having 0Ω source resistance. The equivalent circuits with either source connection types become identical with the assumption of 0Ω source. Therefore, (1) can be used to study the system performance for both parallel single-source and dedicated multi-source configurations with 0Ω source impedance. PTE is defined as Tx coils-to-Rx coil efficiency and the main objective of this paper is to study coil configurations for improved efficiency.

(a)

(b) Fig. 2. Equivalent circuit for three adjacent transmitters and a receiver. (a) – Separate sources connected to each transmitter, (b) – Transmitter coils are connected in parallel with a single source.

𝑍𝑡1 𝑋𝑡2𝑡1 𝑋𝑡3𝑡1 [ 𝑋𝑟𝑡1

𝑋𝑡1𝑡2 𝑍𝑡2 𝑋𝑡3𝑡2 𝑋𝑟𝑡2

𝑋𝑡1𝑡3 𝑋𝑡2𝑡3 𝑍𝑡3 𝑋𝑟𝑡3 𝑃𝑜𝑢𝑡

𝑋𝑡1𝑟 𝑋𝑡3𝑟 𝑋𝑡3𝑟 (𝑍𝑟 + 𝑅𝐿 )] = |𝐼𝑟 |2 𝑅𝐿

𝐼𝑡1 𝑉𝑡1 𝐼𝑡2 𝑉 = [ 𝑡2 ] 𝐼𝑡3 𝑉𝑡3 𝐼𝑟 0

(1)

|𝐼𝑟 |2 𝑅𝐿 𝑃𝑜𝑢𝑡 𝑃𝑇𝐸 = = 2 2 2 𝑃𝑖𝑛 |𝐼𝑡1 | 𝑅𝑡1 + |𝐼𝑡2 | 𝑅𝑡2 + |𝐼𝑡3 | 𝑅𝑡3 + |𝐼𝑟 |2 (𝑅𝐿 +𝑅𝑟 )

Zm = Rm + j (ωLm -1/ωCm), (m = t1,2,3,r) are the selfimpedances of Tx and Rx coils where Lm , Cm , and Rm represent inductance, capacitance and parasitic resistance of the coil. Capacitance includes both distributed capacitance and externally added capacitance. Xmn = Xnm = jωMmn = jωMmn (m = t1,2,3,r) where Mmn (= Mnm) is mutual inductance between mth and nth coils. Coupling coefficient (k) has linear relationship with mutual inductance (kmn=Mmn/√(LmLn)). ω is the operating frequency of the source. Subscripts t1,2,3 represent three nearest Tx coils to the Rx coil concerned (represented by subscript r). Solving for currents and PTE explicitly from (1) results in a very complex expression. Therefore, system performance is numerically analyzed and validated with the experiments. Equation (1) will be used extensively for the analyses in the rest of the paper. Measurement of scattering parameters (S-parameter) is easier compared to voltage and current measurements in the laboratory prototype. In particular, S-parameter measurement is suitable at the initial stage of the design and for analyzing the frequency characteristic. Therefore, we use network analyzer measurements of parallel connected resonators for experiments. PTE of parallel connected Tx configuration can be expressed as in (2) using impedance parameters [6]. Regardless of the source impedance, it can be seen from (2) that coil-to-coil PTE calculated using (2) for parallel Tx configuration is identical to that of 0Ω dedicated source connection configuration. This is because PTE is calculated after the source impedance. Therefore, PTE measurements from a 50 Ω network analyzer


requires

IEEE

permission.

See


|𝑉𝑠 |2 𝑅 |𝑍𝑠 +𝑍𝑖𝑛 |2 𝑖𝑛 2 2 |𝑍21 | |𝑉𝑠 | 𝑃𝑜𝑢𝑡 = 𝑅 |(𝑍𝑠 +𝑍𝑖𝑛 )2 (𝑍𝐿 +𝑍22 )2 | 𝐿 |𝑍21 |2 𝑅𝐿 𝑃𝑜𝑢𝑡 𝑃𝑇𝐸 = = |𝑍𝐿 +𝑍22 |2 𝑅𝑖𝑛 𝑃𝑖𝑛 𝑃𝑖𝑛 =

where 𝑍𝑖𝑛 = 𝑍11 −

(2)

AC Resistance ()

0.6

𝑍22 +𝑍𝐿

B. Coil parameter calculation System performance is largely dependent on the coil parameters. For the coil construction, Litz wire windings consisting of 1162 46AWG strands are used. Tightly wound coils with very small separation between turns result in very high resistance that is dominated by proximity effect [17]. Design parameters chosen according to design guidelines presented in [18] are depicted in Table I. AC resistance of Litz wire coils is dependent on the operating frequency (fs). Therefore, frequency dependence of coil resistances is taken into account in analyzing the frequency response of all the proposed topologies. AC resistance of the coils are calculated using the method proposed in [19, 20], and validated with experiments as shown in Fig. 3(a). Selfinductances of the coils are calculated using circular loop approximation [20, 21], and they agree well with measured values as illustrated in Table I. Stray capacitance of the designed coils are experimentally evaluated by measuring natural resonance frequency of the coils alone. Table I. Design parameters of helical shaped and spiral shaped coils Helical shaped 8.9cm 4.2mm 5cm 11 30µH 32µH ~3pF ~16.5MHz

0.5 0.4

Spiral Shaped 8.9cm 4.2mm 0.2cm 14 20µH 21µH ~0.8pF ~39.8MHz

C. Mutual Inductance Calculation Mutual inductance between circular resonators can be calculated using filament method [22, 23] by modeling the coils as a set of filamentary circular loops. Calculations are experimentally verified (shown in Fig. 3(b)) using the measurement technique presented in [24]. Furthermore, coupling coefficient between designed helical and spiral shaped coils are examined. The coupling coefficient of helical shaped coils is found to be higher than that of the spiral shaped coils at higher axial and radial distances. Therefore, helical shaped coils are preferred for WPT applications. However, helical shaped coils occupy larger space due to their height, making them unsuitable for design of planar repeaters. Therefore, helical

Helical - Calc. Helical - Exp. Spiral - Calc. Spiral - Exp.

0.3 0.2 0.1 0.2

𝑍21 𝑍12

𝑅𝑖𝑛 = 𝑟𝑒𝑎𝑙(𝑍𝑖𝑛 ), 𝑅𝐿 = 𝑟𝑒𝑎𝑙(𝑍𝐿 )

Outer radius Turns separation Thickness Number of turns Inductance - Calculated Inductance - Measured Stray-Capacitance Natural Resonance of coils

shaped coils are used as Tx and Rx coils while spiral shape is used for the design of repeaters. With the above calculation procedure, (1) can be solved for given Tx and Rx arrangements. Characteristics of D-WPT system with a segmented Tx array is analyzed in subsequent sections. Mutual Inductance (H)

are valid for both types of source connection configurations. Therefore, (2) is used for the experimental validations of this paper. However, it can also be seen that the output power of dedicated source connection is different from that of parallel connected Tx configuration.

0.4

0.6 0.8 1 Frequency (MHz)

1.2

1.4

6

Yrx =0 Calc. Yrx =0 Exp.

4

Yrx =5cm Calc. Yrx=5cm Exp.

Yrx=10cm Calc.

2

0

Yrx =10cm Exp. 5

10 15 20 25 Axial seperation (Zrx - cm)

(b) (a) Fig. 3. (a) – AC Resistance vs. Frequency, (b) – Calculated and experimental values of mutual inductance between helical shaped coils.

III. ANALYSIS OF THE CLASSICAL TRANSMITTER ARRAY PTE of the classical Tx array with helical shaped coils is analyzed in this section. Resonance frequency (f0) of Tx and Rx resonators is chosen to be 0.99MHz (f0=1/(2π√(LC)). Series lumped capacitors are used to tune each coil to the resonance frequency. Fig. 4 illustrates a section of a segmented Tx array. System performance is analyzed for different number of Tx coils (NTX) powered at a time. In case of one or three Tx coils are powered (NTX =1 or 3), the segment from A to C is repeated when Rx moves along the Tx array, and segment B to D is repeated for NTX =2. Next Tx coil has to be powered on and last Tx coil has to be disconnected once the Rx moves away from respective segment. Therefore, in all three cases (NTX=1,2, and 3), length of the repeated segment is equal to the separation between Tx coils (ytx). Position B is considered as the aligned position (yrx=0), while position C is considered as the maximum misaligned position (yrx=ytx/2) in subsequent analyses. Equation (1) is used extensively for the analysis of this section, Fig. 5, 6 and 7 are generated from (1). A. Frequency and Load Characteristic Fig. 5 illustrates PTE variation calculated at the aligned position and 20cm transfer distance (yrx = 0, zrx = 20cm) with respect to operating frequency and load resistance for NTX = 1, 2 and 3. As expected, PTE is always maximum at selfresonance frequency for single Tx scenario (NTX=1). In contrast, when two or three Tx coils are powered a dip can be observed at self-resonance (fs=f0) whereas, the maximum PTE is achieved when operating at frequencies away from f0. A frequency splitting phenomena, where the maximum PTE occurs at two different frequencies, can be observed. These two frequencies can be considered as the resonance frequencies of the overall system, lower resonance frequency (LRF), and higher resonance frequency (HRF). Rx A

D

yrx

zrx Tx1

C

B

ytx

Tx2

Tx3

Fig. 4. D-WPT system with segmented transmitter array


requires

IEEE

permission.

See


The frequency splitting phenomena can be explained by examining current ratios and input impedances seen by the source at each Tx coil. Equation (1) can be simplified to (3) assuming identical Tx and Rx coils. Transfer current ratio (χ) is defined as the ratio between summation of absolute squares of Tx currents and absolute square of Rx current. It can be seen from (3) that PTE decreases with the increase of χ. The input impedance characteristics shown in Fig. 6(a) is evaluated at aligned position for NTX=3. The input impedance is determined by the resonator impedance and the reflected impedance from the surrounding coils. At self-resonance frequency (fs=f0), all resonator impedances are equal to their respective coil resistances. As the middle transmitter (Tx2) is nearest to the Rx, reflected impedance is higher compared to that of the other two resonators (Tx1 & Tx3 for NTX=3; Tx3 for NTX=2). This explains higher input impedance seen at Tx2 near resonance frequency. Consequently, there is a significantly lower current in Tx2 compared to the transmitters farther away (Tx1 & Tx3) leading to high χ near self-resonance frequency as seen in Fig. 6(b). On the other hand, at LRF and HRF, input impedance of Tx2 becomes lower than that of the Tx1 and Tx3 (Fig. 6(a)). This regulates the currents in Tx1 and Tx3, causing χ to be minimized at HRF and LRF. 2

2

2

(3)

|𝐼𝑡1 | + |𝐼𝑡2 | + |𝐼𝑡3 | 𝑅𝐿 ); 𝜉 = 2 |𝐼𝑟 | 𝑅

Z

10

rx

Z

2

rx

= 5cm = 20cm

Solid: Tx 2 Dot: Tx /Tx .

1 0.5

1

0.95 1 Operating frequency (f ) - MHz

3

1.05

s

(a) 10

4

Z Z

3

Z

rx rx

= 5cm = 10cm = 20cm



10

rx

10 10

2

1

0.9

LRF z=5cm

0.95 1 LRF HRF LRF z=10cm z=20cm z=20cm

1.05 HRF z=10cm

NTX =2

0.4

zrx=50cm

HRF z=5cm

Operating frequency (f_s)

(b) Fig. 6. (a) – Input impedance of each transmitter coil at aligned position and (b) – Transfer current ratio variation with respect to operating frequency. (NTx=3).

Furthermore, it can be seen from Fig. 5 that highest PTE occurs when the load resistance is lower than 10Ω. However,

0 0

5

10

15

20

25

NTX =1

0.6 0.4

Misalignment (Yrx - cm)

zrx=20cm

0.8

0.2

Zrx = 10cm

5

1

NTX =1 NTX =3

max

20

0.9

zrx=20cm

0.6

𝑅 = 𝑅𝑡1 = 𝑅𝑡2 = 𝑅𝑡3 = 𝑅𝑟

40

1 0.8

PTE

𝜉 𝜒+𝜉+1

PTE

Input Impedance |Z in|

𝜒=(

𝑃𝑇𝐸 =

B. Transmitter separation and Number of transmitter coils The effect of number of Tx coils powered at a time (NTX) and separation between Tx coils (ytx) is analyzed in this section. Numerically calculated results in Fig. 7 illustrate the maximum PTE at ytx values of 50cm and 20cm when load resistance and operating frequency are at their optimum values. If Tx coils are far apart, power transfer contribution from Tx coils placed farther away is not very significant at lower axial distances. This is because coupling between Rx and nearer Tx is much higher than the coupling between Rx and Tx coils farther away. This attributes to the highest PTE observed with NTX =1 at Zrx =20cm and NTX=2 at Zrx=50cm in Fig. 7(a). In contrast, when Tx coils are placed very close to each other (ytx=20cm in Fig. 7(b)) coils farther away also contribute to the power transfer. Results in Fig. 7(b) show that PTE is almost identical for all NTX (1,2 or 3) for Zrx =20cm whereas, NTX =3 shows the best performance for Zrx =50cm. However, Tx coils that are too close result in higher implementation cost as the number of Tx coils to be placed in a given length of Tx array is higher. Therefore, it would be favorable if PTE and misalignment tolerance can be improved with a larger separation between Tx coils. The proposed receiver structure of this paper allows enhanced PTE and misalignment tolerance even with higher separation (ytx=50cm) between Tx coils.

max

(c) (a) (b) Fig. 5. PTE with respect to operating frequency and load resistance, (a) – NTX = 1, (b) – NTX = 2 and (c) – NTX = 3; (yrx = 0, ytx = 50 cm, zrx = 20cm)

equivalent load is dependent on the state of charge (SOC) of the EV battery, and the configuration of battery charging circuitry. Therefore, D-WPT system should be capable of delivering power with an acceptable PTE in a wide load range. It is clear that the classical segmented Tx array experiences lower efficiency at higher load. However, the proposed receiver structure and the tuning method of this paper allows higher PTE in a wide range of load variation.

0.2 0

NTX =2 NTX =3

zrx=50cm 2

4

6

8

Misalignment (Yrx - cm)

(a)

10

(b)

Fig. 7.Maximum PTE (when load resistance and operating frequency is optimum) with respect to misalignment. (a) – ytx=50cm and (b) – ytx=20cm

C. Experimental Results Network analyzer measurements are used along with (2) for the experimental verification. Tx coils are connected parallel with the network analyzer (Agilent E5062A) Port 01, while Rx is connected to the Port 02. Measured S-parameters are converted to Z-parameters to evaluate PTE using (2) for the comparison [6]. The load resistance is chosen to be 50Ω for the subsequent comparisons with experimental results as the characteristic impedance of the network analyzer is 50Ω. 1) Frequency Response Fig. 8(a) shows frequency characteristic at the aligned position for different axial distances. Both LRF and HRF are dependent on the position of the Rx (coupling among coils). The maximum PTEs (at LRF and HRF) and respective frequencies are illustrated in Fig. 8(b). PTE at HRF is higher


requires

IEEE

permission.

See


0.6

Zrx =15cm Zrx =20cm

0.4

Dash:NTX=1 Dot :NTX=2 Solid:NTX=3

0.2 0

0.9

PTE at LRF PTE at HRF LRF HRF Solid: NTX = 2 Dot: NTX = 3

0.8 0.6 0.4 0.2 0

1 1.1 Frequency(MHz)

1.2 1.1 1 0.9

5

0.8 30

10 15 20 25 Axial Seperation Zrx

(b) (a) Fig. 8. (a) – PTE variation with respect to operating frequency at different axial distances at the aligned position, (b) – Maximum achievable PTE

Fig. 9 shows frequency response observed in the experiment. Due to manufacturing tolerances, coil resonances are slightly higher than 0.99MHz as depicted in Table II. Frequency splitting phenomena can be clearly observed in the experimental results as well. Such a frequency characteristic has not been reported in previous research efforts on D-WPT with a moving load. This could be attributed to low frequency (in kHz range) operation where resultant lower reflected impedance is not significant in the realization of splitting phenomena. D-WPT system tested for EV applications in [1] has reported a slight efficiency dip at self-resonance frequency (Fig 4. of [1]). Therefore, careful frequency selection must be carried out in designing a D-WPT system in high frequency. 0.8

zrx =

10cm

15cm

20cm

0.8

25cm

0.6

Increase zrx

0.4

10cm

15cm

20cm

25cm

Increase zrx

0.4 0.2

0.2 0

PTE

PTE

0.6

zrx =

0.9

1 1.1 Frequency(MHz)

1.2

0

0.9

0.95 1 1.05 Frequency(MHz)

1.1

Zrx=20cm

0.4

Solid: Calc. Dot: Exp.

0.2

NTX=2 0

5

10 15 20 Misallignment (cm)

Zrx=15cm

0.4


Zrx=20cm

0.2 0

25

Zrx=15cm Zrx=20cm

1.02


1

NTX=2 0

5


NTX=3 0

5

10 15 20 Misalignment (cm)

25

Zrx=10cm

1.01

Zrx=15cm Zrx=20cm

1.005


1 0.995 0.99

25

(b)

1.015

Zrx=10cm

1.04

Zrx=10cm

0.6

(a)

1.06

0.98

Maximum PTE

Zrx=15cm

Optimum Freq. (MHz)

Zrx =10cm

0.8

Zrx=10cm

0.6

0

Optimum Freq. (MHz)

0.8

1 0.8

1.3

Frequency(MHz)

1

Maximum PTE

PTE

1

is because the system performance is sensitive to coupling changes at lower distances (i.e. high coupling). Even with very low axial separation, PTE drops to less than 2% with the maximum misalignment. A similar efficiency deterioration has been reported in previous research efforts [1]. Therefore, designing D-WPT system with classical receiver arrangement (Topology-) is not desirable as power transfer is not continuous. A Rx coil arrangement to improve the PTE and misalignment tolerance is introduced in the following section. Maximum PTE

than PTE at LRF for NTX = 3, while both PTEs (at LRF and HRF) are almost same for NTX = 2. It can be seen that both LRF and HRF are close to self-resonance frequency for higher axial distances (i.e. lower coupling) whereas, LRF and HRF are further apart for low axial distances (i.e. higher coupling). This is because the input impedance seen at Tx2 is reduced at higher axial distances (Fig. 6(a)), and becomes comparable with the input impedances seen at Tx1 and Tx3. This reduction in input impedance is due to the decrease in the reflected impedance at Tx2 with the coupling deterioration. Therefore, the input impedance of Tx2 is lower at higher axial distances and it results in a lower frequency separation between HRF and LRF at higher axial distances. Frequency splitting disappears when axial distance is further away, due to insignificant reflected impedance between Tx coils.

NTX=3 0

5


25

(c)

(d) Fig. 10. Comparison of calculations and experimental results: (a), (b) – The maximum PTE with respect to misalignment, (c), (d) – Optimum resonance frequencies (HRFs) with respect to misalignment. (a) & (c) – NTX=2, (b) & (d) – NTX=3.

IV. PROPOSED RECEIVER STRUCTURE In this section, DSR with two spiral resonators is proposed to enhance the PTE against axial distance and misalignment. DSR is composed of two spiral coils placed side by side as shown in Fig. 11. Two spiral coils are not connected to each other. Four resonator arrangement topologies are proposed and compared. Topologies with the proposed DSR are shown in Fig. 12. Topology  is the classical receiver arrangement as shown in Fig. 12(a) without any repeater structures, and is used as a reference arrangement to compare the performance of the proposed arrangements. Topology  consists of a DSR placed in front of the Rx at the Tx side. DSR placed behind the Rx is considered as topology . The topology  is a combination of both topologies  and . Axial distance (zrx) refers to unobstructed air gap between the Tx array and the receiver structure (Fig. 12).

1.15

(a) (b) Fig. 9. Experimental results: PTE variation with respect to operating frequency at the aligned position. (a) – NTX =2 , (b) – NTX=3.

2) Misalignment tolerance Misalignment tolerance is crucial in designing D-WPT systems in order to maintain continuous power availability while receiver moves along the Tx array. Fig. 10 compares the calculated maximum PTE and experimental results with respect to misalignment when operating at the optimum resonance frequency. Fig. 10 shows the maximum PTE at HRF with respect to misalignment and respective resonance frequencies. Experimental results follow theoretical model well with exceptions at low axial distances and low misalignments. This

(a)

Compensation capacitor

(b)

Fig. 11. The proposed DSR (a) – Drawing of DSR, (b) – Experimental realization of DSR

Fig. 13 shows the proposed Rx arrangement with DSR for topology . The DSR placed behind the Rx coil is named as DSR1 (RxR1 & RxR2) and the DSR placed in front of Rx coil is named as DSR2 (RxR3 & RxR4) (refer Fig. 13). DSR1 can be viewed as a magnetic reflector whereas, DSR2 can be viewed as a magnetic lens which congregates magnetic flux towards Rx. Therefore, performance improvements from topologies  and  can be analyzed as lensing and reflecting


requires

IEEE

permission.

See


effects respectively. Topology  exhibits both lensing and reflecting effects. YDSR1 (or YDSR2) is lateral separation between Rx and DSR1 (or DSR2) while ZDSR1 (or ZDSR2) is axial separation between Rx and DSR1 (or DSR2) (refer Fig. 13). Rx zrx Tx1

yrx Tx3

Tx2

ytx (a)

Rx DSR2

zrx (b)

DSR1 Rx

zrx (c)

DSR1 Rx DSR2

zrx

yrx (d)

Fig. 12. Proposed topologies with DSR. (a) – Topology , (b) – Topology , (c) – Topology  and (d) – Topology . YDSR1

RxR1

Rx

RxR2 ZDSR1

nominal axial distance of 20cm is chosen for the experimental validation. The optimum values for YDSR1,2 and ZDSR1,2 aligned position are less than few centimeters in all three topologies. This results in a compact version of DSR. However, misalignment tolerance is poor with low YDSR1,2. The optimum YDSR1,2 for the maximum misalignment (25cm for ytx=50cm) is found to be higher than 10cm. Therefore, YDSR1,2 is chosen to be higher than 9cm in the optimization process. B. Efficiency Enhancement with DSR PTE can be improved up to 76% using the proposed topologies whereas PTE is around 30% in the classical arrangement at nominal transfer distance of 20cm. Table II shows both theoretical and experimentally tuned optimum selfresonance frequencies which produces maximum PTE for each topology. Each coil is separately tuned with high quality factor lumped capacitors. It can be seen from Table II that the optimum transmitter resonance (fTx) is much lower than that of Rx (fRx) and DSRs (fRxR1-fRxR4). Optimum operating frequency (fs) is slightly higher than DSR resonances and lower than fRx. Therefore, it is apparent that the reactive impedance of Tx coils farther away limits the currents, thus improving the system efficiency at the optimum operating frequency. The optimum frequency characteristics of all topologies follow similar behavior in all Rx positions. Table II. Theoretical and experimental optimum self-resonance frequencies (in MHz) at aligned position for 20cm axial separation.

fTx1 Rx

fTx2

ZDSR2 RxR3

YDSR2 (a)

RxR4

DSR1

fTx3

DSR2 (b)

Fig. 13. (a) – 2-D view and dimensions of the proposed Receiver Structure with double spiral repeater, (b) – experimental realization of receiver for topology 

A. Optimization of DSR In this section, enhanced PTE and misalignment tolerance of the proposed receiver topologies with DSRs are analyzed. Equation (1) is extended to analyze the proposed topologies. Rank of the impedance matrix and the equation is equal to the number of coils present in each topology. As identified in the previous section, frequency characteristic of the system is dependent on the position of the Rx and the self-resonance of each coil. The optimum resonance changes with coupling between coils and thus, with the position of the Rx and arrangement of DSRs. Matlab® optimization toolbox is used to find the optimum design parameters. PTE calculated from (1) is implemented as a function of resonance frequency of each coil, operating frequency, dimensions of the DSR and receiver position. PTE is employed as the maximization objective function in Matlab® using interior point algorithm. Optimum design variables can be executed for each receiver position. A set of switched capacitor banks can be used to tune each coil’s self-resonance at a particular position to obtain the maximum PTE. However, it would not be practically feasible due to the complex nature of the D-WPT system. Therefore, a

fRx fRxR1 fRxR2 fRxR3 fRxR4

Topology Theoretical Experimental Theoretical Experimental Theoretical Experimental Theoretical Experimental Theoretical Experimental Theoretical Experimental Theoretical Experimental Theoretical Experimental

 0.99 0.9925 0.99 0.9905 0.99 0.9925 0.99 0.9915

 0.4 0.403 0.4 0.407 0.4 0.407 1.079 1.071

NA

NA

NA

NA

NA NA

 0.4 0.403 0.4 0.407 0.4 0.407 1.175 1.6875 1.0871 1.0869 1.0871 1.0875

1.0150 1.0159 1.0150 1.0155

NA NA

 0.4002 0.403 0.4002 0.407 0.4002 0.407 1.1104 1.1003 1.0028 1.0013 1.0028 1.0025 1.0084 1.0103 1.0084 1.0035

Simulation results illustrated in Fig. 14 show dependence of PTE on each coil’s self-resonance. In this analysis, individually selected frequency is varied while other frequencies are kept at their optimal values. It can be seen that the bandwidth of the operating frequency (fs) is very narrow in all the topologies. Therefore, the selection of proper operating frequency is very important in designing D-WPT system with the proposed topologies. Conversely, bandwidths of fTx and fRx are wider than fs. A larger bandwidth allows relatively easier tuning during the experiment. Tx coils are connected in parallel with the network analyzer (Agilent E5062A) to experimentally validate the proposed topologies. The experimental setup is illustrated in Fig. 15. Double peak phenomena can be observed in the experimental results shown in Fig. 16 as well. It should be noted that the optimum experimental operating frequency is slightly


requires

IEEE

permission.

See


different from that of the theoretical analysis. This is because the assumption of identical coils in the analysis does not stand due to experimental disparities as seen in Table II. However, maximum efficiencies and frequency responses still follow the theoretical analysis. 0.4

PTE

f f

0.2

f 0

0.5

1 1.5 2 2.5 Normalized Frequency (f/fopt)

f

PTE

f

0.4

f

0.2

f 0.5

1 1.5 2 Normalized Frequency (f/f

(b)

0.8

)

2.5

PTE

f

0.4

f

0.2

f 1 1.5 2 Normalized Frequency (f/f

(c)

PTE

0.8

s RxR1

)

2.5

Tx s RxR3

&f

RxR4

Rx

3

f f

0.4

f f

0.2

f 1 1.5 2 Normalized Frequency (f/f

RxR2

Rx

opt

0.6

0.5

&f

3

f

0.5

Tx

opt

0.6

0

Rx

3

0.6

0

s

(a)

0.8

0

Tx

)

2.5

Tx

s RxR1 RxR3

&f &f

RxR2 RxR4

Rx

3

opt

(d) Fig. 14. Frequency characteristic of proposed topologies at aligned position at 20cm axial distance. Each frequency is normalized to its optimum value. (a) – Topology , (b) – Topology , (c) – Topology  and (d) – Topology .

Network Analyser To Port 02

From Port 01 Fig. 15. The experimental setup – topology  0.8

Topology (1)

0.6

PTE

Topology (2)

0.4

Topology (3)

0.2 0 0.9

Topology (4) 0.95

1 1.05 Frequency (MHz)

1.1

1.15

Fig. 16. Experimental results: PTE vs. operating frequency at aligned position at 20cm axial separation.

Next, PTEs of the proposed topologies, optimized at 20cm nominal distance are compared with the classical arrangement in terms of transfer distance and misalignment. It can be seen from Fig. 17(a) that the proposed topologies show greater efficiency improvement against transfer distance. Topology  with two DSRs in front and behind Rx shows 81% and 60% efficiency at 10cm and 35cm, respectively. It should be noted that the Rx size of the proposed topologies are larger than that

of the classical topology. Normalized distance (znorm) has been introduced to compare dissimilar designs with different sizes as defined in (4) (DTx and DRx are maximum dimension of Tx and Rx respectively) [20]. The performance improvement is further compared using znorm to incorporate the size difference. Fig. 17 (b) compares experimental results with respect to the normalized distance. It can be seen that, proposed topologies allow higher PTE with respect to normalized distance except for the topology . PTE profile for Topology  is similar to the classical topology when znorm >1.7. Apart from Topology  at high normalized distances, all other cases with proposed topologies show higher PTE than classical topology. For example, when distances are normalized to Tx and Rx sizes, 20cm nominal distance for proposed topologies is equivalent to 13.2cm for classical topology (respective znorm is around 1.5). PTEs obtained for normalized distance of 1.5 in topologies , ,  and  are 51%, 70%, 66% and 76% respectively. 𝑧𝑟𝑥 𝑍𝑛𝑜𝑟𝑚 = (4) √𝐷𝑇𝑥 𝐷𝑅𝑥 2 2 Fig. 17(c) illustrates the variation of experimental PTE with respect to misalignment at 20cm axial distance. All three topologies with DSR show significant efficiency improvement against misalignment. PTE of Topology  drops only 1.5% from 76% with the maximum misalignment even with fixed frequency operation. Maximum and minimum PTE pairs for topologies ,  and  against misalignments are 32%-1.2%, 70%-66.7% and 68.5%-56.7% respectively. It is apparent that PTE improvement that can be obtained by the lensing effect (Topology ) is higher than that from the reflecting effect (Topology ). The highest PTE improvement results in Topology  can be viewed as combination of both lensing and reflecting effects. In addition to y-direction misalignment which is unavoidable in D-WPT system, lateral displacement along x-direction is also examined. The numerical results in Fig. 18 compare performance variation against lateral displacement at nominal transfer distance of 20cm (znorm ~ 1.5 for topologies -). It can be seen that the proposed topology allows higher misalignment tolerance in lateral displacement compared to the classical topology. Nevertheless, PTE declines in a faster rate against lateral displacement in all topologies compared to y-direction misalignment. It should be noted that, the lateral displacement along x-direction can be minimized using techniques such as tracking and autonomous driving. Therefore, it is apparent that misalignment problem of the DWPT technology can be solved to a greater extent by using the proposed topologies with DSR. The system resonance frequency changes with the position of the Rx. Frequency tuning has been understood as a promising approach to provide enhanced misalignment tolerance. Fig. 19 illustrates the maximum achievable PTEs with variable frequency operation. It can be seen from the experimental results in Fig. 19(b) that the efficiency improvements in topologies ,  and  with the variable frequency operation are around 4%, 1% and 2% respectively. The cost effectiveness of frequency tuning has to be decided based on the application as the communication, control and location needed for tracking makes the design more complex. Other than the operating frequency tuning, resonance of the DSR resonators can also be tuned using switched


requires

IEEE

permission.

See


capacitor bank to improve the PTE. Adaptive tuning of the DSR coils is beyond the scope of this paper. The transient response of the 2-coil WPT system is characterized with an exponential decay constant of ω/Q [25]. A similar behavior is also observed with the proposed topologies. For example, Fig. 20 shows simulation and experimental results of the transient response for normalized receiver current in topology . (The simulation is performed in PLECS using multi-terminal mutual inductance model, while the measured S-parameters are used to generate experimental transient response along with Agilent ADS tool [26]) It can be seen that steady state condition is reached before 200μs time. Therefore, adaptive tuning technique can be implemented for an EV moving in highway speed.

results in Fig. 21(a) show that proposed topologies allow higher PTE in a wider load variations compared to the classical topology. Theoretical results are verified with experiments using S-parameter based technique [26] as shown in Fig. 21(b).

(b) (a) Fig. 20. Normalized output current - transient response for topology . (a) – Simulation results, (b) – Experimental results

0.8 0.8

0.7 0.6

PTE

PTE

0.6 0.4 Topology Topology Topology Topology

0.2 0

10

0.5 Topology Topology Topology Topology

0.4

(1) (2) (3) (4)

0.3

15 20 25 30 Axial Seperation (z-cm)

0.2 1

35

(1) (2) (3) (4)

1.5 2 Normalized distance

(a)0.8

2.5

(a)

(b)

PTE

0.6 Topology Topology Topology Topology

0.4 0.2 0

0

5

(1) (2) (3) (4)


V. CONCLUSIONS

25

(c) Fig. 17. Experimental results: (a) – PTE with respect to axial separation at the aligned position, (b) – PTE variation with respect to normalized distance, (c) – PTE with respect to misalignment when optimized to aligned position at zrx=20cm. 0.8 0.6

PTE

Topology Topology Topology Topology

0.4

(1) (2) (3) (4)

0.2 0

0

5 10 Lateral Displacement - xrx (cm)

15

Fig. 18. PTE variation with respect to lateral displacement (x-direction) for zrx=20cm. 0.8

0.7

PTE

(2) (2) (3) (3) (4) (4)

0.5

0.4

0

-

Fixed Freq. Freq. Tuning Fixed Freq. Freq. Tuning Fixed Freq. Freq. Tuning 5


(a)

PTE

0.7

0.6

(b)

Fig. 21. PTE variation against load resistance. (a) – Theoretical analysis, (b) Experimental results (zrx=20cm, yrx=0)

0.6

(2) (2) (3) (3) (4) (4)

0.5

25

0.4

0

-

Fixed Freq. Freq. Tuning Fixed Freq. Freq. Tuning Fixed Freq. Freq. Tuning 5

10 15 Misallignment (cm)

20

25

(b)

This paper presented numerical analysis on the design of dynamic wireless power transfer (D-WPT) system with segmented transmitter (Tx) array. Efficiency characteristics against design parameters such as, operating frequency, load resistance, number of simultaneously powered Tx coils (NTX) and Tx coil spacing were presented. A frequency splitting phenomena was observed in efficiency characteristic for NTX>1. In addition, a double spiral repeater (DSR) for dynamic wireless power transfer for EV was introduced. The DSR placed at the receiver coil was found to be a promising solution to obtain enhanced PTE with higher transfer distance and misalignment tolerance. The performance deterioration effect that occurs when the load is moving away from a source resonator can be compensated by congregating magnetic flux towards the load. Efficiency enhancement due to lensing DSR placed in front of the Rx was found to be higher than that of the reflecting DSR placed behind Rx. Quasi static condition was analyzed using equivalent circuit model and validated with the experimental results. Furthermore, proposed topologies showed high efficiency in a wide load range. It was also found that the optimal operating frequency is not identical to the selfresonance of the individual resonators, and frequency splitting phenomenon was investigated in all resonator arrangements. Therefore, frequency characteristic must be considered with respect to the whole system, instead of individual resonators.

Fig. 19. PTE comparison, operating frequency tuning vs. fixed frequency operation. (a) – Theoretical analysis, (b) - Experimental results

C. Load Variation The equivalent load resistance changes with the battery SOC while the battery is charged by D-WPT. Hence, performance against load variation is also a vital design consideration. This section compares PTE against load variation in proposed topologies optimized to nominal load of 50 ohm. Numerical

REFERENCES [1]

[2]

O. C. Onar, J. M. Miller, S. L. Campbell, C. Coomer, C. P. White, and L. E. Seiber, "A novel wireless power transfer for in-motion EV/PHEV charging," in Twenty-Eighth Annual IEEE Applied Power Electronics Conference and Exposition (APEC), 2013 2013, pp. 3073-3080. S. Choi, J. Huh, W. Y. Lee, S. W. Lee, and C. T. Rim, "New CrossSegmented Power Supply Rails for Roadway-Powered Electric Vehicles," IEEE Transactions on Power Electronics, vol. 28, pp. 5832-5841, 2013.


requires

IEEE

permission.

See

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TTE.2015.2508721, IEEE Transactions on Transportation Electrification [3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] [18]

[19]

[20]

[21]

[22]

[23] [24]

[25]

J. Huh, S. W. Lee, W. Y. Lee, G. H. Cho, and C. T. Rim, "Narrow-Width Inductive Power Transfer System for Online Electrical Vehicles," IEEE Transactions on Power Electronics, vol. 26, pp. 3666-3679, 2011. H. Rakouth, J. Absmeier, A. Brown Jr, I.-S. Suh, M. John M, R. Sumner, et al., "EV Charging Through Wireless Power Transfer: Analysis of Efficiency Optimization and Technology Trends," in Proceedings of the FISITA 2012 World Automotive Congress. vol. 192, ed: Springer Berlin Heidelberg, 2013, pp. 871-884. N. P. Suh, D. H. Cho, and C. T. Rim, "Design of On-Line Electric Vehicle (OLEV)," in Global Product Development, A. Bernard, Ed., ed: Springer Berlin Heidelberg, 2011, pp. 3-8. L. Siqi and C. C. Mi, "Wireless Power Transfer for Electric Vehicle Applications," IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 3, pp. 4-17, 2015. J. P. C. Smeets, T. T. Overboom, J. W. Jansen, and E. A. Lomonova, "Comparison of Position-Independent Contactless Energy Transfer Systems," IEEE Transactions on Power Electronics, vol. 28, pp. 20592067, 2013. S. Y. R. Hui, Z. Wenxing, and C. K. Lee, "A Critical Review of Recent Progress in Mid-Range Wireless Power Transfer," IEEE Transactions on Power Electronics, vol. 29, pp. 4500-4511, 2014. B. Teck Chuan, M. Kato, T. Imura, O. Sehoon, and Y. Hori, "Automated Impedance Matching System for Robust Wireless Power Transfer via Magnetic Resonance Coupling," IEEE Transactions on Industrial Electronics, vol. 60, pp. 3689-3698, 2013. R. Johari, J. V. Krogmeier, and D. J. Love, "Analysis and Practical Considerations in Implementing Multiple Transmitters for Wireless Power Transfer via Coupled Magnetic Resonance," IEEE Transactions on Industrial Electronics, vol. 61, pp. 1774-1783, 2014. L. Kibok, Z. Pantic, and S. M. Lukic, "Reflexive Field Containment in Dynamic Inductive Power Transfer Systems," IEEE Transactions on Power Electronics, vol. 29, pp. 4592-4602, 2014. A. Bindra, "Wide-Bandgap-Based Power Devices: Reshaping the power electronics landscape," IEEE Power Electronics Magazine, vol. 2, pp. 4247, 2015. L. Seung-Hwan and R. D. Lorenz, "A design methodology for multi-kW, large air-gap, MHz frequency, wireless power transfer systems," in Energy Conversion Congress and Exposition (ECCE), 2011 IEEE, 2011, pp. 3503-3510. J. Millan, P. Godignon, X. Perpina, A. Perez-Tomas, and J. Rebollo, "A Survey of Wide Bandgap Power Semiconductor Devices," IEEE Transactions on Power Electronics, vol. 29, pp. 2155-2163, 2014. Z. Wenxing, L. Chi Kwan, and S. Y. R. Hui, "General Analysis on the Use of Tesla's Resonators in Domino Forms for Wireless Power Transfer," IEEE Transactions on Industrial Electronics, vol. 60, pp. 261270, 2013. I. Awai, "Basic characteristics of "Magnetic resonance" wireless power transfer system excited by a 0 ohm power source," IEICE Electronics Express, vol. 10, pp. 20132008-20132008, 2013. G. S. Smith, "Proximity effect in systems of parallel conductors," Journal of Applied Physics, vol. 43, pp. 2196-2203, 1972. J. P. K. Sampath, A. Alphones, D. M. Vilathgamuwa, A. Ong, and B. Nguyen Xuan, "Coil Enhancements for High Efficiency Wireless Power Transfer Applications " in 40th Annual Conference of IEEE Industrial Electronics Society. (IECON). , 2014, pp. 2978-2983. Z. Yang, L. Wentai, and E. Basham, "Inductor Modeling in Wireless Links for Implantable Electronics," IEEE Transactions on Magnetics, vol. 43, pp. 3851-3860, 2007. A. K. RamRakhyani, S. Mirabbasi, and C. Mu, "Design and Optimization of Resonance-Based Efficient Wireless Power Delivery Systems for Biomedical Implants," IEEE Transactions on Biomedical Circuits and Systems, vol. 5, pp. 48-63, 2011. C. M. Zierhofer and E. S. Hochmair, "Geometric approach for coupling enhancement of magnetically coupled coils," IEEE Transactions on Biomedical Engineering, vol. 43, pp. 708-714, 1996. C. Akyel, S. I. Babic, and M.-M. Mahmoudi, "Mutual inductance calculation for non-coaxial circular air coils with parallel axes," Progress In Electromagnetics Research, vol. PIER 91, pp. 287–301, 2009. F. W. Grover, Inductance Calculations: New York: Dover, 1964. Y. P. Su, L. Xun, and S. Y. R. Hui, "Mutual inductance calculation of movable planar coils on parallel surfaces," IEEE Transactions on Power Electronics, vol. 24, pp. 1115-1123, 2009. M. Kiani and M. Ghovanloo, "The Circuit Theory Behind Coupled-Mode Magnetic Resonance-Based Wireless Power Transmission," IEEE

Transactions on Circuits and Systems I: Regular Papers, vol. 59, pp. 2065-2074, 2012. [26] K. Jinwook, K. Do-Hyeon, and P. Young-Jin, "Analysis of Capacitive Impedance Matching Networks for Simultaneous Wireless Power Transfer to Multiple Devices," IEEE Transactions on Industrial Electronics, vol. 62, pp. 2807-2813, 2015.

J. P. K. Sampath (S’12) B.Sc. degree in electronics and telecommunications engineering from the University of Moratuwa, Sri Lanka in 2009. He is currently pursuing Ph.D. degree in the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. His research interests include design and optimization of wireless power transfer systems. D. Mahinda Vilathgamuwa (S’90– M’93–SM’99) received the B.Sc. degree in electrical engineering from the University of Moratuwa, Moratuwa, Sri Lanka, in 1985 and the Ph.D. degree in electrical engineering from Cambridge University, Cambridge, U.K., in 1993. In 1993, he joined the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, as a Lecturer and then became an Associate Professor. He is currently a Professor with the Queensland University of Technology, Brisbane, Australia. His current research interests include power electronic converters, electrical drives, and power quality. Dr. Vilathgamuwa was the Chairman of the IEEE Section, Singapore, and a member of the Power Electronics Technical Committee of the IEEE Industrial Electronics Society. He is an associate editor of IEEE Transactions on Industry Applications. A Alphones (SM’98) received his B.Tech. from Madras Institute of Technology in 1982, M.Tech. from Indian Institute of Technology Kharagpur in 1984 and Ph.D. degree in Optically Controlled Millimeter wave Circuits from Kyoto Institute of Technology (Japan) in 1992. He was a JSPS visiting fellow from 1996-97 at Japan. During 1997-2001, he was with Centre for Wireless Communications, National University of Singapore involved in the research on optically controlled passive/active devices. Since 2001 he is with the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. He is also Program coordinator for research. He has 30 years of research experience. He has published and presented over 250 technical papers in peer reviewed International Journals/ Conferences. His current interests are electro-magnetic analysis on planar RF circuits and integrated optics, microwave photonics, metamaterial based leaky wave antennas and wireless power transfer technologies. He was involved many IEEE flagship conferences held in Singapore and General Chair


requires

IEEE

permission.

See


of APMC 2009, and MWP 2011. He is currently the chairman of IEEE Singapore section and a senior member of IEEE.


requires

IEEE

permission.

See