A Review on the Recent Development of Capacitive Wireless Power

energies

Review Review AA ReviewReview onon thethe RecentRecent DevelopmentDevelopment ofof CapacitiveCapacitive WirelessWireless PowerPower TransferTransfer TechnologyTechnology Fei Lu, Hua Zhang and Chris Mi * Fei Lu ID , Hua Zhang and Chris Mi * ID Department of Electrical and Computer Engineering, San Diego State University, San Diego, CA 92182, USA; [email protected] of Electrical (F.L.); [email protected] and Computer Engineering, (H.Z.) San Diego State University, San Diego, CA 92182, USA; [email protected]* Correspondence: (F.L.); [email protected]; [email protected] Tel.: +1-619-594-2654 (H.Z.) * Correspondence: [email protected]; Tel.: +1-619-594-2654 Received: 17 October 2017; Accepted: 30 October 2017; Published: 1 November 2017 Received: 17 October 2017; Accepted: 30 October 2017; Published: 1 November 2017

Abstract:Abstract: Capacitive power transfer (CPT) technology is an effective andand importantimportant alternativealternative to thethe conventionalconventional inductiveinductive power transfertransfer (IPT). It utilizes high-frequency electric fieldsfields to transfer electric power, whichwhich hashas threethree distinguishingdistinguishing advantages:advantages: negligiblenegligible eddy-current loss, relatively lowlow costcost andand weight, weight, and and excellent excellent misalignment misalignment performance. performance. In recentIn recent years, years, the powerthe power level level and efficiencyand efficiency of CPT of systems CPT systems has been has significantly been signif improvedicantly improved and has reached and has the reached power levelthe power suitable level for electricsuitable vehicle for electric charging vehicle applications. charging Thisapplicatio paperns. reviews This paper the latest reviews developments the latest in developments CPT technology, in focusingCPT technology, on two keyfocusing technologies: on two thekey compensationtechnologies: the circuit compensation topology and circuit the capacitivetopology and coupler the structure.capacitive Thecoupler comparison structure. with The the comparison IPT system with and somethe IPT critical system issues and in some practical critical applications issues in arepractical also discussed. applications Based are onalso these discussed. analyses, Based the on future these research analyses, direction the future can research be developed direction and can the applicationsbe developed of and the the CPT applications technology of can the be CPT promoted. technology can be promoted.

Keywords: capacitive power transfer; wireless powerpower transfer;transfer; electricelectric ﬁelds;fields; electricelectric vehiclevehicle

1. Introduction Wireless powerpower transfertransfer (WPT)(WPT) technologytechnology utilizesutilizes energy-containedenergy-contained ﬁeldsfields to deliverdeliver electricelectric powerpower from thethe transmittertransmitter to the receiver side, removing the direct metal-to-metal contact [1–4]. [1–4]. TheThe typicaltypical structurestructure ofof aa WPTWPT systemsystem isis shownshown inin FigureFigure1 .1.

Figure 1. Typical structure of a wireless powerpower transfertransfer system.system.

At the primaryprimary side, an auxiliary circuit is usuallyusually required to convertconvert the electricelectric power into a properproper formform to to excite excite the the transmitter transmitter device, device, which which can generate can generate the corresponding the corresponding ﬁelds and fields transmit and totransmit the secondary to the secondary side. In the side. power In the transmission power transm process,ission process, a medium a medium is not always is not required,always required, but its appearancebut its appearance usually contributesusually contributes to improve to theimprove system the performance. system performance. At the secondary At the side,secondary the receiver side, devicethe receiver can convert device can the convert ﬁelds back the fields into electric back into power electric and power provide and it provide to the load it to throughthe load through another auxiliaryanother auxiliary circuit. circuit.

Energies 2017, 10, 1752; doi:10.3390/en10111752 www.mdpi.com/journal/energies Energies 2017, 10, 1752; doi:10.3390/en10111752 www.mdpi.com/journal/energies Energies 2017, 10, 1752 2 of 30 EnergiesEnergies 2017 2017, ,10 10, ,1752 1752 2 2of of 30 30

AccordingAccording to toto the thethe category categorycategory of ofof ﬁelds, fields, the the current current WPT WPTWPT systems systemssystems can becancan classiﬁed bebe classifiedclassified intoacoustic intointo acoustic acoustic power transferpowerpower transfer transfer (APT) [ 5 (APT)(APT)–7], optical [5–7],[5–7], power opticaloptical transfer power (OPT)transfer [8 ],(OPT) microwave [8],[8], microwavemicrowave power transfer powerpower (MPT)transfer transfer [9 ],(MPT) (MPT) inductive [9], [9], powerinductiveinductive transfer power power (IPT) transfertransfer [10], (IPT) and(IPT) capacitive [10],[10], andand capacitive power transfer power (CPT) transfertransfer [11 ], (CPT)(CPT) as shown [11],[11], as inas shown Figureshown 2in ,in which Figure Figure are 2, 2, suitablewhichwhich are are for suitable suitable different forfor applications. differentdifferent applications.applications.

FigureFigure 2.2. CategoryCategory of wireless power transfer transfer systems. systems.

Among all the power transfer systems, the CPT technology has been widely studied in recent Among all the power transfer systems, the CPT technologytechnology has been widely studied in recent years. The typical structure of a CPT system is shown in Figure 3. years. The typical structure of a CPTCPT system is shownshown inin FigureFigure3 3..

Figure 3. Typical structure of a capacitive power transfer (CPT) system. Figure 3. Typical structure of a capacitive power transfer (CPT) system.system. The CPT system in Figure 3 is also a practical implementation of the WPT system in Figure 1. The CPT system in Figure 3 is also a practical implementation of the WPT system in Figure 1. The Theauxiliary CPT system circuits in Figureare realized3 is also by a practicalpower electronics implementation converters of the WPTand the system corresponding in Figure1. The auxiliary circuits are realized by power electronics converters and the corresponding Thecompensation auxiliary circuits networks, are realized including by power inductors electronics and converterscapacitors. andThe the transmitter corresponding and compensationreceiver are compensation networks, including inductors and capacitors. The transmitter and receiver are networks,realized by including metal inductorsplates. Usually, and capacitors. four metal The pl transmitterates are required and receiver in a are CPT realized system by metalto form plates. a realized by metal plates. Usually, four metal plates are required in a CPT system to form a Usually,capacitive four coupler. metal platesTwo plates are required are used in at a CPTthe primary system toside form as a capacitivepower transmitter, coupler. Twoand the plates other are usedcapacitivetwo atplates the coupler. primaryat the secondary Two side asplates a side power are act transmitter,usedas a power at the rece andprimaryiver, the other resultingside twoas a platesin power at least at transmitter, the two secondary coupling andside capacitors the act other as a powertwoto provide plates receiver, at a thepower resulting secondary flow inloop. atside least acttwo as a couplingpower rece capacitorsiver, resulting to provide in at aleast power two flow coupling loop. capacitors to provideDifferentDifferent a power CPTCPT systemsflowsystems loop. cancan bebe realizedrealized by different coupler structures. structures. For For example, example, in in Figure Figure 3,3 , sincesinceDifferent therethere areare CPT twotwo systems pairspairs ofof can platesplates be inrealizedin thethe coupler,coupler, by different itit isis alsoalso coupler called structures. a bipolar Forstructure. structure. example, In In some somein Figure other other 3, applications,sinceapplications, there are oneone two pairpair pairs of plates of plates can can in be be the realized realized coupler, by by theit the is stray also stray capacitancecalled capacitance a bipolar from from structure. a receiver a receiver Interminal terminalsome otherto a to aapplications,transmitter transmitter terminal, one terminal, pair ofresulting resulting plates canin in a be aunipolar unipolarrealized struct by structure theure stray that that capacitance is is similar similar to from to a asingle-wiresingle-wire a receiver systemterminal system [12]. [to12 a ]. ThetransmitterThe valuevalue of ofterminal, couplingcoupling resulting capacitancecapacitance in a dependsdependsunipolar on struct the ure plateplate that area, is similardistance, to and anda single-wire the the dielectric dielectric system material material [12]. The value of coupling capacitance depends on the plate area, distance, and the −dielectric−12 material betweenbetween the the plates. plates. Since Since the the permittivity permittivity constant constantε 0ε0of of air air is is as as small small as as 8.85 8.85× ×10 10 F/m,F/m, the the value value −12 ofbetweenof coupling coupling the capacitance plates.capacitance Since is is limited.th limited.e permittivity Therefore, Therefore, constant it it usually usually ε0 of requires airrequires is as resonances smallresonances as 8.85 in in the × the10 circuit circuit F/m, between betweenthe value the compensationofthe coupling compensation capacitance networks networks andis limited. theand coupler, the Therefore, coupler, producing it producing usually high requires voltages high voltages resonances on the plateon thein to the generateplate circuit to generate sufficientbetween electricthesufficient compensation fields electric for power fields networks transfer.for power and transfer.the coupler, producing high voltages on the plate to generate sufficientTheTheadvantages electricadvantages fields of of thefor the power CPT CPT technology technologytransfer. lie lie in in its its low low cost, cost, low low weight, weight, and and low low eddy-current eddy-current loss inloss nearby Thein nearby advantages metals metals [13]. of According[13]. the AccordingCPT technology to recent to recent studies, lie instudies, its it canlow it be cost,can used below inused weight, short in distanceshort and distancelow and eddy-current low and power low applications,losspower in nearbyapplications, such metals as such integrated[13]. as According integrated circuits to circuits (IC) recent [14 (IC)],studies, biomedical [14], itbiomedical can devices be used devices [15 in,16 short], [15,16], LED distance lighting LED lightingand [17 ,low18], USBpower[17,18], and applications, USB mobile and devicemobile such chargingdevice as integrated charging [19–21 circuits ].[19–21]. In addition, (IC) In addition, [14], its biomedical power its power can be devicescan increased be increased [15,16], to kWLED to kW level lighting level and applied[17,18],and applied USB to high and to power high mobile power areas, device suchareas, charging as such synchronous as [19–21]. synchronous machineIn addition, machine excitation its power excita [22– tioncan24]. be[22–24]. As increased a duality As a to toduality inductivekW level to andinductive applied machine, to high thepower Lorentz areas, force such in asa CPT synchronous system can machine be used excitato realizetion an[22–24]. electrostatic As a duality rotating to inductive machine, the Lorentz force in a CPT system can be used to realize an electrostatic rotating

Energies 2017, 10, 1752 3 of 30

Energies 2017, 10, 1752 3 of 30 machine, the Lorentz force in a CPT system can be used to realize an electrostatic rotating machine machine for motor driving [25,26]. Moreover, the transfer distance and power can be further for motor driving [25,26]. Moreover, the transfer distance and power can be further increased to increased to charge the electric vehicles, or even realize dynamic charging while the vehicle is charge the electric vehicles, or even realize dynamic charging while the vehicle is moving along the moving along the roadway [27–30]. Based on these developments, this paper will review all the roadway [27–30]. Based on these developments, this paper will review all the latest progresses in the latest progresses in the CPT technology, especially for the long-distance and high-power electric CPT technology, especially for the long-distance and high-power electric vehicle charging applications. vehicle charging applications. The theoretical analysis and detailed implementations will be The theoretical analysis and detailed implementations will be presented, aiming to promote the presented, aiming to promote the practical applications of the CPT technology. practical applications of the CPT technology.

2. Fundamental Working Principle

2.1. General Mathematical Model of CPT System To analyze its fundamental working principle, the structure of a CPT system can be further simplifiedsimpliﬁed as the circuit topology in Figure4 4..

Figure 4. SimplifiedSimpliﬁed circuit topology of a CPT system.

The input and output sides are represented by two sinusoidal sources V1 and V2, neglecting the The input and output sides are represented by two sinusoidal sources V1 and V2, neglecting high-order harmonics in the circuit. The input and output side currents are defined as I1 and I2, the high-order harmonics in the circuit. The input and output side currents are defined as I and I , respectively. The capacitive coupler acts as a two-port network for the primary and secondary1 sides.2 respectively. The capacitive coupler acts as a two-port network for the primary and secondary sides. Generally, the capacitive coupler can be represented by its equivalent behavior source model, Generally, the capacitive coupler can be represented by its equivalent behavior source model, including including two capacitors and two voltage-controlled current sources, as provided in [31]. In a two capacitors and two voltage-controlled current sources, as provided in [31]. In a capacitive coupler, capacitive coupler, there are multiple coupling capacitances between plates, which can increase the there are multiple coupling capacitances between plates, which can increase the complexity of the complexity of the working principle analysis. Therefore, it needs to be simplified as an equivalent working principle analysis. Therefore, it needs to be simplified as an equivalent mutual capacitance CM mutual capacitance CM and two equivalent internal self-capacitances Cin1 and Cin2, as shown in and two equivalent internal self-capacitances C and C , as shown in Figure4. The simplification Figure 4. The simplification process and definitionsin1 of thein2 equivalent capacitances will be provided process and definitions of the equivalent capacitances will be provided in Section4 in this paper. in Section 4 in this paper. In Figure4, the capacitive coupling between the primary and secondary sides is represented by In Figure 4, the capacitive coupling between the primary and secondary sides is represented by two voltage-controlled current sources IM1 and IM2, where IM1 is used to transfer power and IM2 is two voltage-controlled current sources IM1 and IM2, where IM1 is used to transfer power and IM2 is used to receive power [32]. Therefore, the apparent power SM1 absorbed by IM1 is expressed as used to receive power [32]. Therefore, the apparent power SM1 absorbed by IM1 is expressed as ∗ ∗ SM1 = V=⋅−=⋅−C1 · (−IM1) **= VC1 · (−ωjωCMVC2) (1) SVMC11() I M 1 V C 1 ( jCV MC 2 ) (1) where thethe asteriskasterisk notation notation means means the the complex complex conjugate; conjugate;SM S1 Mis1 ais complex a complex power, power, including including real real and andimaginary imaginary part definedpart defined as PM 1asand PMQ1 Mand1, respectively; QM1, respectively;VC1 and VCV1 Cand2 are V theC2 are voltages the voltages across the across primary the primaryand secondary and secondary capacitance capacitanceCin1 and C Cinin21; ωandis theCin2; angular ω is the switching angular switching frequency offrequency the input of and the output input andsources. output The sources. primary voltageThe primaryVC1 is voltage selected V asC1 the is referenceselected as phasor, the reference and the phase phasor, difference and the of VphaseC2 is differencedefined as ofθ, resultingVC2 is defined in as θ, resulting in V = |V | · (cos θ + j sin θ) (2) C2 =⋅C2 θθ + VVCC22(cos j sin ) (2) According to Equation (2), the apparent power SM1 is further expressed as According to Equation (2), the apparent power SM1 is further expressed as SM1 = PM=+1 + jQM1 = ω =CωθωθM · | ⋅⋅VC1| · |VC2 ⋅| · sin θ ++ jωC ⋅⋅M · |VC1| · | ⋅VC2| · cos θ (3) SMM11 P jQCV M 1 MCC 12 Vsin jCVV MCC 12 cos (3)

Therefore, the active and reactive power absorbed by the current source IM1 is expressed as

Energies 2017, 10, 1752 4 of 30

Energies 2017, 10, 1752 4 of 30 Therefore, the active and reactive power absorbed by the current source IM1 is expressed as (  =⋅⋅⋅ωθ PPCVVMMCC=112ωC · |V | · |V |sin· sin θ M 1 M C1 C2 (4)(4) Q ==⋅⋅⋅ωθC · |V | · |V | · MQCVV1MMCC112ω M C1 C2coscos θ

At thethe receiverreceiver side,side, thethe apparentapparent powerpowerS SMM22 thatthat providedprovided byby thethe secondarysecondary currentcurrent source sourceI MIM22 isis expressedexpressed asas = · ∗ = · ( )∗ SM2 V=⋅=⋅C2 IM**2 VC2 ωjωCMVC1 (5) SVIVjCVMCMC2222() MC 1 (5) Similarly, the real and imaginary part of SM2 is deﬁned as PM2 and QM2, respectively. Then, the Similarly, the real and imaginary part of SM2 is defined as PM2 and QM2, respectively. Then, the apparent power SM2 is further expressed as apparent power SM2 is further expressed as

SM2 = PM=+2 + jQM2 = ω =CωθωθM · | ⋅⋅⋅VC1| · |VC2| · sin θ −− jωCM ⋅⋅⋅· |VC1| · |VC2| · cos θ (6) SMM22 P jQCV M 2 MCC 12 Vsin jCVV MCC 12 cos (6)

Therefore, thethe activeactive andand reactivereactive powerpower providedprovided byby thethe currentcurrent sourcesourceI IMM22 is expressed as (  =⋅⋅⋅ωθ PMPCVV2 MMCC=212ωCM · |VC1| · |VC2|sin· sin θ  (7)(7) QM2 = =−−ωωθCM · ⋅|VC1| ⋅· |VC2 ⋅| · cos θ QCVVMMCC212cos

Comparing Equations (1) and (5), the magnitudemagnitude of apparent power can be expressedexpressed asas |SM|=||=|SMS2M|=|2|=|SM1S|.M 1In|. addition, In addition, comparing comparing Equations Equations (3) (3)and and (6), (6), it itshows shows that that the the active powerpower absorbed byby IIMM1 isis equal equal to the active powerpower providedprovided byby IIMM2,, which which is consistent with thethe universaluniversal energyenergy conservationconservation law. Then,Then, thethe activeactive powerpower is furtherfurther deﬁneddefined asas PM = PPMM1 1= =PMP2M. It2. also It also needs needs to tobe beemphasized emphasized that that the the reactive reactive power power QQMM1 1absorbedabsorbed by by IMIM1 1doesdoes not not equal equal to to the the reactive power providedprovided byby QQMM2,, but but they they are are opposite opposite to to each other. Then, the reactive power isis furtherfurther deﬁneddefined asas QQMM = QM11 == −−QQMM2. 2To. To summarize, summarize, the the apparent apparent power power SSMM cancan be be illustrated illustrated in in Figure Figure 55,, wherewhere thethe x-axis-axis isis thethe reactivereactive powerpower andand yy-axis-axis isis thethe activeactive power.power.

FigureFigure 5.5. TheThe apparentapparent powerpower SSMM transferred to the secondary side.

Figure 5 shows that, when θ is in the range of (0°,◦ 180°),◦ there is PM1 > 0, which means active Figure5 shows that, when θ is in the range of (0 , 180 ), there is PM > 0, which means active power is transferred from the primary to secondary side. When θ is in the1 range of (−180°, 0°), there power is transferred from the primary to secondary side. When θ is in the range of (−180◦, 0◦), there is is PM1 < 0 and the power transfer direction is reversed. In a CPT system design process, the active PM1 < 0 and the power transfer direction is reversed. In a CPT system design process, the active power power PM is usually known to satisfy the power requirement. Then, the corresponding reactive PM is usually known to satisfy the power requirement. Then, the corresponding reactive power QM is power QM is expressed as expressed as ==⋅· θ QQPMMMPM cotcot θ (8)(8)

Then, the relationship between the active power PM andand reactive reactive power power QQMM isis shown shown in in Figure Figure 6.6 .

Energies 2017, 10, 1752 5 of 30 Energies 2017, 10, 1752 5 of 30 Energies 2017, 10, 1752 5 of 30

Figure 6. The relationship between reactive power QM and active power PM. FigureFigure 6. 6.The The relationship relationship between between reactive reactive power powerQ MQMand and active active power powerP MP.M.

Figure 6 shows that the reactive power QM relates to the phase angle θ between VC1 and VC2. Figure 6 shows that the reactive power QM relates to the phase angle θ between VC1 and VC2. For example,Figure6 shows when thatθ = the90°, reactive QM = 0, power which QmeansM relates there to theis no phase reactive angle powerθ between transferredVC1 and to VtheC2. For example, when θ = 90°,◦ QM = 0, which means there is no reactive power transferred to the Forsecondary example, side. when Whenθ =θ = 90 45°,, Q QMM == 0,PM which, which means meansthere the transferred is no reactive reactive power power transferred is equal toto the secondary side. When θ = 45°,◦ QM = PM, which means the transferred reactive power is equal to the secondaryactive power. side. When When θ approachesθ = 45 , QM to= P0°,M ,the which reactive means power the transferred QM is dramatically reactive increased power is equalto be tomuch the active power. When θ approaches to◦ 0°, the reactive power QM is dramatically increased to be much activelarger power.than the When activeθ powerapproaches PM. to 0 , the reactive power QM is dramatically increased to be much larger than the active power PM. largerAccording than the activeto Equations power P (4)M. and (7), when the required active power PM is known, increasing According to Equations (4) and (7), when the required active power PM is known, increasing sinθ Accordingcan help reduce to Equations the voltages (4) and |V (7),C1| whenand |V theC2|. required Therefore, active in a power practicalPM systemis known, design, increasing the phase sinθ sinθ can help reduce the voltages |VC1| and |VC2|. Therefore, in a practical system design, the phase can help reduce the voltages |VC | and |VC |. Therefore, in a practical system design, the phase angleangle θ θcan can be be designed designed to to be be close close1 to to 90°. 90°. In In this2 this way, way, the the reactive reactive power power circulating circulating in in the the circuit circuit is is angle θ can be designed to be close to 90◦. In this way, the reactive power circulating in the circuit reducedreduced to to almost almost zero, zero, and and the the corresponding corresponding conduction conduction loss loss is is also also minimized. minimized. It Itneeds needs to to be be is reduced to almost zero, and the corresponding conduction loss is also minimized. It needs to be clarifiedclarified that that the the phase phase angle angle θ θcan can be be adjusted adjusted through through compensation compensation circuit circuit design, design, which which will will be be clariﬁed that the phase angle θ can be adjusted through compensation circuit design, which will be presentedpresented in in Section Section 3. 3. presented in Section3. Meanwhile,Meanwhile, Equations Equations (4) (4) and and (7) (7) also also show show that that the the system system power power is is proportional proportional to to the the switchingMeanwhile, frequency Equations ω, the (4) mutual and (7) coupling also show capacitance that the system CM, and power the is voltages proportional |VC1 to| and the switching|VC2| on switching frequency ω, the mutual coupling capacitance CM, and the voltages |VC1| and |VC2| on frequency ω, the mutual coupling capacitance CM, and the voltages |VC | and |VC | on the metal thethe metal metal plates. plates. In In a aCPT CPT system, system, these these values values can can be be increased increased to to achieve achieve1 high high power power2 transfer. transfer. The The plates. In a CPT system, these values can be increased to achieve high power transfer. The parameter parameterparameter sin sinθ θcan can be be designed designed to to be be 1.0 1.0 to to maximize maximize the the system system power. power. As As an an example, example, the the sincouplingθ can be capacitance designed to C beM is 1.0 10pF, to maximize the maximum the system achievable power. system As an example, power P theM at coupling different capacitance switching coupling capacitance CM is 10pF, the maximum achievable system power PM at different switching π CfrequencyM is 10pF, f thes (fs maximum= ω/2π), and achievable the plate system voltage power |VCP1M| atis differentshown in switching Figure 7. frequency In this analysis,fs (fs = ω /2it is), frequency fs (fs = ω/2π), and the plate voltage |VC1| is shown in Figure 7. In this analysis, it is andassumed the plate that voltage|VC1| = | VVCC2|1| for is simplification shown in Figure purpose.7. In this analysis, it is assumed that | VC1| = |VC2| assumed that |VC1| = |VC2| for simplification purpose. for simpliﬁcation purpose.

Figure 7. The maximum achievable system power P at different f and |V |, when C = 10 pF and Figure 7. The maximum achievable system power PMM at different sfs and |VCC11|, when CM = 10 pF and Figure 7. The maximum achievable system power PM at different fs and |VC1|, when CM = 10 pF and |V | = |V |. |VCC1|1 = |VC2|.C2 |VC1| = |VC2|.

Figure 7 shows that the magnitude of the plate voltage |VC1| needs to be increased to kV level Figure 7 shows that the magnitude of the plate voltage |VC1| needs to be increased to kV level to achieve kW power transfer. For example, in a vehicle charging application, when fs is 1 MHz, 3 to achieve kW power transfer. For example, in a vehicle charging application, when fs is 1 MHz, 3 kW system power requires |VC1| to be 7 kV. This high voltage can induce potential safety issues in a kW system power requires |VC1| to be 7 kV. This high voltage can induce potential safety issues in a

Energies 2017, 10, 1752 6 of 30

Figure7 shows that the magnitude of the plate voltage | VC1| needs to be increased to kV level to EnergiesachieveEnergies 2017 2017 kW, 10, power10, 1752, 1752 transfer. For example, in a vehicle charging application, when fs is 1 MHz,6 3 6of kW of 30 30 system power requires |VC1| to be 7 kV. This high voltage can induce potential safety issues in a practicalpractical application, application, which which will will be be discussed discussed in in Section Section 66. .6. Theoretically,Theoretically, Theoretically, accordingaccording according toto to EquationsEquations Equations (4)(4) (4) andand (7), (7), the the CPT CPT technology technology can can also also be be used used in in an an extremelyextremely extremely highhigh high powerpower power system.system. system. For For example, example, the the plateplate voltage voltage | V|VC1C1|1| |at atat different differentdifferent C CCMMM, ,and and f sf fsofs of a a200 200 kW kWkW CPT CPTCPT system systemsystem is isis shown shownshown in inin FigureFigure Figure 88. .8. Similarly,Similarly, Similarly, itit it isis assumed assumed sin sinθθ == = 1 1 and and | V|VCC1C1|1| |= =| V| VCV2C|.C2|.2 |.

IncreasingIncreasing the the switching switching frequency frequency f s fiss is an an effective effective way way to to acquire acquire higher higher power. power. With With limited limited Increasing the switching frequency fs is an effective way to acquire higher power. With limited couplingcoupling capacitance capacitance C CMM, ,f s fsneeds needs to to be be increased increased to to 10s 10s of of MHz MHz to to achieve achieve 10s 10s or or even even 100s 100s of of kW kW coupling capacitance CM, fs needs to be increased to 10s of MHz to achieve 10s or even 100s of kW powerpower transfer.transfer. transfer. The The The high high high frequency frequency frequency and and highand high powerhigh power power working working working condition condition condition usually usually requireusually widerequire require bandgap wide wide (WBG)bandgapbandgap power (WBG) (WBG) electronics power power electronics devices,electronics such devices, devices, as silicon such such carbide as as silicon silicon (SiC) carbide carbide and gallium (SiC) (SiC) and nitride and gallium gallium (GaN) nitride MOSFETs,nitride (GaN) (GaN) to MOSFETs,buildMOSFETs, up the to to power build build up converters. up the the power power converters. converters.

2.2.2.2. Maximum Maximum Achievable Achievable E Efficiency Efficiencyfficiency of of a a CPT CPT System System TheThe previous previous section section presented presented the the system system powe power power rproperty property for for an an ideal ideal CPT CPT system system where where the the losseslosses are are neglected. neglected. However, However, However, the the efficiencyefficiency efficiency property property is is also also anan an importantimportant important concern.concern. concern. Especially Especially in in a a long-distancelong-distance and and loosely-coupled loosely-coupled CPT CPT CPT system, system, system, it it it is is is meaningful meaningful to to study study its its efficiencyefficiency efficiency to to maintain maintain effectiveeffective power power transfer. transfer. Since Since the the coupling coupling capacitance capacitance C CMM isis is limited, limited, limited, it it it is is is necessary necessary to to increase increase the the plateplate voltage voltage for for sufficientsufficient sufficient power power transfer. transfer. Auxilia Auxiliary Auxiliaryry circuits, circuits, either either resonant resonant or or non-resonant, non-resonant, are are requiredrequired toto to workwork work withwith with thethe the capacitivecapacitive capacitive coupler. coupler. coupler. According Accord Accordinging to to Figureto Figure Figure4, its4, 4, its equivalentits equivalent equivalent circuit circuit circuit is shownis is shown shown in inFigurein Figure Figure9 for 9 efficiency9for for efficiency efficiency analysis. analysis. analysis. This equivalentThis This equivalent equivalent circuit cancircuit circuit be realizedcan can be be realized byrealized multiple by by methods,multiple multiple resultingmethods, methods, resultinginresulting different in in compensationdifferent different compensation compensation circuit topologies, circuit circuit topo topo whichlogies,logies, will which which be discussed will will be be discussed in discussed Section 3in .in Section Section 3. 3.

FigureFigure 9. 9. Equivalent Equivalent circuit circuit topology topology of of a aCPT CPT system system with with inductive inductive networks. networks.

ComparingComparing FiguresFigures 4 4 andand 9,9, thethe externalexternal capacitancescapacitances CCexex1 1 andand CCexex2 2 inin thethe compensationcompensation networksnetworks areare combinedcombined withwith thethe internalinternal capacitancescapacitances CCin1in 1 andand CCin2in 2 ofof thethe capacitivecapacitive coupler,coupler, resultingresulting in in an an equivalent equivalent capacitive capacitive coupler coupler with with C C1 1= =C Cin1in 1+ +C Cexex1 1and and C C2 2= =C Cin2in 2+ +C Cexex2.2 .The The capacitive capacitive couplingcoupling coefficient coefficient k Ck Cis is further further defined defined as as

Energies 2017, 10, 1752 7 of 30

Comparing Figures4 and9, the external capacitances Cex1 and Cex2 in the compensation networks Energies 2017, 10, 1752 7 of 30 are combined with the internal capacitances Cin1 and Cin2 of the capacitive coupler, resulting in an equivalent capacitive coupler with C1 = Cin1 + Cex1 and C2 = Cin2 + Cex2. The capacitive coupling C coefficient kC is further defined as = M kC (9) CCCM kC = √ 12 (9) C1C2 Then, two inductive networks are used at both the primary and secondary sides to form Then, two inductive networks are used at both the primary and secondary sides to form resonances resonances with the capacitances. The output voltage source V2 is represented by a resistive load RL with the capacitances. The output voltage source V2 is represented by a resistive load RL to analyze to analyze the system efficiency. Considering the power losses of the resonant circuit components, the system efficiency. Considering the power losses of the resonant circuit components, Figure9 is Figure 9 is further simplified as Figure 10. further simplified as Figure 10.

Figure 10. Equivalent circuit topology of a CPT system considering power losses of components.

In Figure 10, the input source V1 is represented by an equivalent source V1eq, and the output In Figure 10, the input source V1 is represented by an equivalent source V1eq, and the output L Leq load RL is represented by anan equivalentequivalent conductanceconductance GGLeq.. The The power power losses losses of of the capacitances are represented by parallel-connected conductances GC1 and GC2. The inductive networks in Figure 9 are represented by parallel-connected conductances GC1 and GC2. The inductive networks in Figure9 are represented by two equivalent inductances L1 and L2, and their power losses are also represented by represented by two equivalent inductances L1 and L2, and their power losses are also represented by two parallel-connected conductances GL1 and GL2, respectively. It needs to be emphasized that the two parallel-connected conductances GL1 and GL2, respectively. It needs to be emphasized that the derivation of the equivalent inductances is a simplificationsimplification of the inductive networks. Regardless of the complexitycomplexity ofof the the inductive inductive networks, networks, they they can ca ben simplifiedbe simplified as equivalent as equivalent inductances inductances with theirwith losses,their losses, which which mean mean this analysis this analysis is independent is independent of the of circuit the circuit topology topology [33–35 [33–35].]. According to Figure 1010,, the systemsystem efficiencyefficiency isis expressedexpressed asas

2 VG2 η = |VCCLeq2 | GLeq η = 222 (10) 2 ++2 2 (10) |VCVGVGVG1CC|11G1 + |VC 222| G2 + |V CLeq 2C2| GLeq where G 1 and G2 areare defined deﬁned as as 1 2 ( G = G + G GG1 =+C1 GL1 (11) 111CL G2 ==+GC2 + GL2 (11) GG222CL G Generally, at both the primary and secondary sides, the matching networks are designed to Generally, at both the primary and secondary sides, the matching networks are designed to compensate the impedance of the capacitive coupler. To eliminate the reactive power circulating in the compensate the impedance of the capacitive coupler. To eliminate the reactive power circulating in circuit and maximize the system efﬁciency, L1 and L2 need to be designed to perfectly resonate with C1 the circuit and maximize the system efficiency, L1 and L2 need to be designed to perfectly resonate and C2, respectively. In this way, the secondary side voltage VC2 is expressed as with C1 and C2, respectively. In this way, the secondary side voltage VC2 is expressed as

IM2 ==+ (G2 + GLeq)VC2 == ωjωCMVC1 (12) IGGVjCVMLeqCMC22() 2 1 (12)

By substituting Equation (12) into Equation (10), the system efﬁciencyefficiency isis rewrittenrewritten asas 1 η = 1 η = 2 2 (13) ()(G2++GLeq) G G GG2 Leq GG1 + 2 + 1 (13) ω2C2 GLeq 12++GLeq 1 ω 22M CM GGLeq Leq

By substituting Equation (9) into Equation (13), the efficiency is simplified as

Energies 2017, 10, 1752 8 of 30 Energies 2017, 10, 1752 8 of 30

By substituting Equation (9) into Equation (13), the1 efficiency is simplified as η = 1 α ++12 η = α 1 (14) α+ 1 +2 ++ 2α 1 1 2kQQ+ α +α 1 kC QC1Q122 wherewhereα αis is defined defined as as load load ratio, ratio, and andQ Q1 and1 andQ Q2 2are are defined defined as as the the quality quality factor factor of of the the primary primary and and secondarysecondary circuit, circuit, respectively. respectively. It needsIt needs to be to clarified be clarified that Qthat1 and Q1Q and2 have Q2 includedhave included the power the lossespower inlosses both thein both capacitors the capacitors and inductors. and inductors. They can They be expressed can be expressed as as ωω GLeq CC αG===Leq ,,QQωC112ωC2 (15) α = GGG, Q1 =12, Q2 = (15) G2 21G1 G 22

Equation (14) shows that the system efficiency is determined by the coupling coefficient kC, the Equation (14) shows that the system efficiency is determined by the coupling coefficient kC, the quality factor Q1 and Q2, and the load condition α. According to Equation (14), for a certain quality factor Q1 and Q2, and the load condition α. According to Equation (14), for a certain coupling coupling coefficient and quality factor, there exists an optimum αηmax that can maximize the system coefficient and quality factor, there exists an optimum αηmax that can maximize the system efficiency, whichefficiency, can be which expressed can be as expressed as 2 kCQ2 1Q2 = kQQC 12 ηmaxη = 2 (16) max q (16) (1 + ++1 + k222Q Q ) (1 1kQQC 121 2 )

TheThe optimum optimumα ηαmaxηmax isis furtherfurther expressedexpressed asas

α q=+2 η 1 kQQ2 (17) αηmax max= 1 + kCCQ 1Q 22 (17)

In a practical CPT system, the quality factors Q1 and Q2 at the primary and secondary sides are In a practical CPT system, the quality factors Q1 and Q2 at the primary and secondary sides usually close to each other, which can be assumed to be Q = Q1 = Q2. In this way, the efficiency are usually close to each other, which can be assumed to be Q = Q1 = Q2. In this way, the efficiency Equation (14) can be simplified, and the product of kCQ can be considered together. At different Equation (14) can be simplified, and the product of kCQ can be considered together. At different values values of kCQ and load condition α, the system efficiency is shown in Figure 11. of kCQ and load condition α, the system efficiency is shown in Figure 11.

Figure 11. Efﬁciency of a CPT system at different values of k Q and α. Figure 11. Efficiency of a CPT system at different values ofC kCQ and α.

FigureFigure 11 11 shows shows that that increasing increasing the valuethe value of kCQ ofhelps kCQ improvehelps improve the system the efficiency. system efficiency. In addition, In foraddition, a certain for value a certain of kCQ ,value a maximum of kCQ achievable, a maximum efficiency achievable that relates efficiency to the loadthat conditionrelates to αthe, which load iscondition consistent α, withwhich Equation is consistent (16), with exists. Equation To clarify, (16), at exists. different To clarify, values at of differentkC and valuesQ, the of maximum kC and Q, achievablethe maximum efficiency achievable is shown efficien in Figurecy is shown 12. in Figure 12.

Energies 2017, 10, 1752 9 of 30 Energies 2017, 10, 1752 9 of 30

Figure 12. Maximum achievable efficiency of a CPT system at different values of kC and Q. Figure 12. Maximum achievable efﬁciency of a CPT system at different values of kC and Q.

Figure 12 shows that a CPT system can achieve a high efficiency even with limited value of kC, Figure 12 shows that a CPT system can achieve a high efficiency even with limited value of k , as as long as the quality factor Q of the components is high enough. For example, when Q is higherC long as the quality factor Q of the components is high enough. For example, when Q is higher than than 400, the system can realize 90% efficiency with a coupling coefficient kC as low as 0.05. 400, the system can realize 90% efficiency with a coupling coefficient k as low as 0.05. Therefore, in Therefore, in a practical CPT system, the quality factor of the capacitorsC and inductors needs to be a practical CPT system, the quality factor of the capacitors and inductors needs to be increased to increased to improve the system efficiency. It needs to be pointed out that the efficiency discussed improve the system efficiency. It needs to be pointed out that the efficiency discussed in this section in this section only considers the power losses in the resonant components, including the capacitive only considers the power losses in the resonant components, including the capacitive coupler and coupler and compensation networks. The efficiency of a practical system should be slightly lower, compensation networks. The efficiency of a practical system should be slightly lower, considering the considering the losses in the power electronics converters. losses in the power electronics converters. It also needs to be clarified that the above derivation of the power and efficiency in a CPT It also needs to be clarified that the above derivation of the power and efficiency in a CPT system system is in a general form and it does not rely on a specific circuit topology. For example, is in a general form and it does not rely on a specific circuit topology. For example, Equations (4) Equations (4) and (7) show that the system power depends on the voltage stresses on the plates, and (7) show that the system power depends on the voltage stresses on the plates, regardless how these regardless how these voltages are generated by the compensation circuits. In addition, Figure 11 voltages are generated by the compensation circuits. In addition, Figure 11 shows that the topology of shows that the topology of the compensation circuits does not affect the calculation process of the the compensation circuits does not affect the calculation process of the system efficiency as in Equation system efficiency as in Equation (14), as long as the reactive power in the circuit is fully (14), as long as the reactive power in the circuit is fully compensated. compensated. The power and efficiency analysis of a CPT system indicates a close relation with the conventional The power and efficiency analysis of a CPT system indicates a close relation with the IPT system [33]. In fact, a CPT system performs as a duality of an IPT system, such as the electric conventional IPT system [33]. In fact, a CPT system performs as a duality of an IPT system, such as fields are used to transfer power instead of the magnetic fields. Therefore, the fundamental working the electric fields are used to transfer power instead of the magnetic fields. Therefore, the principle of a CPT system is similar to an IPT system, and the previous research methods and results fundamental working principle of a CPT system is similar to an IPT system, and the previous of an IPT system can also be applied in a CPT system, which can promote the theoretical study and research methods and results of an IPT system can also be applied in a CPT system, which can practical application of the CPT technology. promote the theoretical study and practical application of the CPT technology. 3. Compensation Circuit Topology 3. Compensation Circuit Topology Different CPT systems can be distinguished by their compensation circuit topologies, which Different CPT systems can be distinguished by their compensation circuit topologies, which determine the system power capability, efficiency, and frequency properties with specified input and determine the system power capability, efficiency, and frequency properties with specified input output conditions. For example, a good compensation circuit should be able to increase the voltages and output conditions. For example, a good compensation circuit should be able to increase the on metal plates for high power transfer [34]. In addition, it needs to achieve a high coupling coefficient voltages on metal plates for high power transfer [34]. In addition, it needs to achieve a high kC and an optimum load ratio α to maximize the system efficiency [35]. Moreover, it is better to be coupling coefficient kC and an optimum load ratio α to maximize the system efficiency [35]. robust to frequency and parameter variations for practical applications [36]. Moreover, it is better to be robust to frequency and parameter variations for practical applications The recent compensation circuit can be classified as non-resonant and resonant topologies. It needs [36]. to be emphasized that the compensation networks in Figure4 do not necessarily need to be resonant The recent compensation circuit can be classified as non-resonant and resonant topologies. It circuits, as long as the voltages on plates can be established for power transfer. The non-resonant needs to be emphasized that the compensation networks in Figure 4 do not necessarily need to be circuit is usually a pulse width modulation (PWM) converter with intermediate capacitors as a resonant circuits, as long as the voltages on plates can be established for power transfer. The capacitive coupler, and its working principle is similar to a regular PWM converter. The resonant non-resonant circuit is usually a pulse width modulation (PWM) converter with intermediate topology can be realized by high-frequency power amplifiers or a full-bridge inverter with auxiliary capacitors as a capacitive coupler, and its working principle is similar to a regular PWM converter. components. Therefore, the compensation circuit topologies are summarized in three categories: PWM The resonant topology can be realized by high-frequency power amplifiers or a full-bridge inverter converter based topology, power amplifier based topology, and full-bridge inverter based topology, as with auxiliary components. Therefore, the compensation circuit topologies are summarized in three presented below. categories: PWM converter based topology, power amplifier based topology, and full-bridge inverter based topology, as presented below.

Energies 2017, 10, 1752 10 of 30 Energies 2017, 10, 1752 10 of 30

3.1. PWM Converter Based Topology The PWM converterconverter with aa singlesingle activeactive switch,switch, suchsuch asas Buck-boost,Buck-boost, Cuk,Cuk, Sepic, and Zeta converter, cancan be be used used to to realize realize a CPT a CPT system system [37]. [37]. For example,For exampl thee, circuit the circuit topology topology of a CPT of systema CPT basedsystem on based sepic on converter sepic converter is shown is inshown Figure in 13 Figure. 13.

Figure 13. Circuit topology of a CPT system based on sepic converter.

In Figure 13, there are two coupling capacitances, CM1 and CM2, as a capacitive coupler, which is In Figure 13, there are two coupling capacitances, CM1 and CM2, as a capacitive coupler, which isdifferent different from from the the conventional conventional sepic sepic converter. converter. These These capacitances capacitances work work as as the the energy storagestorage components and also provide thethe isolationisolation betweenbetween thethe primaryprimary andand secondarysecondary sides.sides. When switch S turns off, the mutual capacitances CM1 and CM2 are charged by the input source and inductor L1. S turns off, the mutual capacitances CM1 and CM2 are charged by the input source and inductor When S turns on, the capacitances are discharged, and power is transferred to the load. In this L1. When S turns on, the capacitances are discharged, and power is transferred to the load. In this design, thethe coupling coupling capacitance capacitance and and switching switching frequency frequency should should be large be enough large toenough realize significantto realize powersignificant transfer power and transfer avoid discontinuous and avoid discontinuou working mode.s working For example, mode. For when example, the coupling when the capacitance coupling increasescapacitance to 24increases nF and theto 24 switching nF and the frequency switching is 200 frequency kHz, the is sepic 200 converterkHz, the realizessepic converter more than realizes 1 kW powermore than transfer 1 kW with power 90.3% transfer DC-DC with efficiency 90.3% DC-DC [38]. efficiency [38]. The advantage of the PWM converter based CPT systemsystem is that the systemsystem performanceperformance is not sensitive to circuit parameter variation and the system efficiencyefficiency is relatively high due to the large coupling capacitances. In In real real applications, the the ph physicalysical size and distance between the plates can change thethe coupling coupling capacitance. capacitance. In In addition, addition, it is it common is common to have to have up to up±10% to ±10% variations variations in the valuesin the ofvalues resonant of resonant components, components, such as su inductorsch as inductors and capacitors and capacitors [39]. In [3 a9]. PWM In a converter,PWM converter, as long as as long the parametersas the parameters are large are enough large to maintainenough to a continuousmaintain currenta continuous working current mode, theseworking variations mode, cannot these affectvariations the system cannot power. affect the system power. However, the limitation of this circuit topology is that it can only be usedused inin shortshort distancedistance applications. SinceSince the the relatively relatively large large coupling coupling capacitances capacitances are required are required for energy for storageenergy purpose,storage thepurpose, distance the should distance be should even lower be ev thanen lower 1 mm, than and 1 dielectric mm, and coating dielectric is used coating on theis used surface on ofthe the surface plate toof furtherthe plate increase to further the capacitancesincrease the [capacitances40]. These factors [40]. These affect thefactors convenience affect the in convenience the usage of in WPT the technology.usage of WPT In addition,technology. the soft-switchingIn addition, the condition soft-switching of the MOSFETs condition cannot of the be MOSFETs achieved cannot at all load be conditionsachieved at in all a PWMload conditions converter, unlessin a PWM special converter, soft-switching unless techniquespecial soft-switching is applied [41 ].technique Therefore, is theapplied system [41]. efficiency Therefore, is significantly the system reduced efficiency and is the significantly conductive reduced EMI problem and the increases. conductive Moreover, EMI sinceproblem there increases. is only oneMoreover, active switch since there in the is circuit, only one the active power switch level in of the a CPT circuit, system the ispower limited level by of the a performanceCPT system is of limited the switch. by the Therefore, performance to increase of the the switch. system Therefore, power level, to increase multiple the PWM system converters power canlevel, be multiple connected PWM in parallel, converters andthe can interleaved be connected control in parallel, strategy and can bethe applied interleaved to improve control the strategy system performancecan be applied [42 to]. Inimprove addition, the a push-pullsystem performanc PWM convertere [42]. withIn addition, two power a push-pull switches PWM can be converter used as a powerwith two excitation power switches to increase can the be system used as power a power [43 ].excitation to increase the system power [43].

3.2. Power AmpliﬁerAmplifier Based Topology High frequency power amplified, ampliﬁed, such as class D, class E, E, class EF, EF, and class φ converters [44], [44], can alsoalso bebe usedused toto realize realize a a CPT CPT system. system. For For example, example, the the circuit circuit topology topology of of a CPT a CPT system system based based on theon the class class E converter E converter is shown is shown in Figure in Figure 14. 14.

Energies 2017, 10, 1752 11 of 30 Energies 2017, 10, 1752 11 of 30

Figure 14. Circuit topology of a CPT system based on class E converter.

In Figure 14, the coupling capacitors CM1 and CM2 can provide isolation between the primary In Figure 14, the coupling capacitors C and C can provide isolation between the primary and and secondary sides and also act as resonantM1 componentsM2 in the CPT system. The circuit working secondary sides and also act as resonant components in the CPT system. The circuit working principle principle is similar to a regular class E converter under the control of switch S. It indicates that any is similar to a regular class E converter under the control of switch S. It indicates that any resonant resonant converter with series-connected capacitors can be used to realize a CPT system, for converter with series-connected capacitors can be used to realize a CPT system, for example, the class example, the class E converter can be combined with a LCCL resonant network to establish E converter can be combined with a LCCL resonant network to establish resonances in the circuit, and resonances in the circuit, and effective power transfer can be realized [45]. effective power transfer can be realized [45]. The main advantage of the power amplifier based topology is that the transfer distance can be The main advantage of the power amplifier based topology is that the transfer distance can be significantly increased [46–49]. There are usually resonances in a power amplifier, and the significantly increased [46–49]. There are usually resonances in a power amplifier, and the switching switching frequency can increase to a very high value, which helps reduce the required inductances frequency can increase to a very high value, which helps reduce the required inductances and and capacitances. The high-frequency property not only contributes to increase the system power capacitances. The high-frequency property not only contributes to increase the system power capacity, capacity, but also reduce the size of the passive components. In addition, since the soft-switching but also reduce the size of the passive components. In addition, since the soft-switching condition condition can be further realized in a power amplifier, the system efficiency can be maintained at a can be further realized in a power amplifier, the system efficiency can be maintained at a high level. high level. This circuit topology has been applied in vehicle charging applications, in which a This circuit topology has been applied in vehicle charging applications, in which a conformal bumper conformal bumper is installed in front of a vehicle to form capacitances with the charging pad [38]. is installed in front of a vehicle to form capacitances with the charging pad [38]. When the switching When the switching frequency is 530 kHz, the system can achieve 1 kW power transfer with a frequency is 530 kHz, the system can achieve 1 kW power transfer with a DC-DC efficiency as high as DC-DC efficiency as high as 92%. With the help of wide bandgap (WBG) devices, when the 92%. With the help of wide bandgap (WBG) devices, when the switching frequency increases to 10s of switching frequency increases to 10s of MHz range, the system performance can be further MHz range, the system performance can be further improved. improved. However, the disadvantage of the power amplifier based topology is its sensitivity to parameter However, the disadvantage of the power amplifier based topology is its sensitivity to variations. In practical applications, it is common to have variations on the plate distance of the parameter variations. In practical applications, it is common to have variations on the plate distance capacitive coupler. In addition, the primary and secondary plates could have misalignments, which of the capacitive coupler. In addition, the primary and secondary plates could have misalignments, cause a significant decrease of the capacitance value. Since the coupling capacitance is series connected which cause a significant decrease of the capacitance value. Since the coupling capacitance is series in the circuit, all of these variations can affect the resonances in the circuit and induce a dramatic connected in the circuit, all of these variations can affect the resonances in the circuit and induce a reduction of the system power. Meanwhile, the soft-switching condition can also be disturbed that dramatic reduction of the system power. Meanwhile, the soft-switching condition can also be might result in extra power losses in the semiconductor devices. Since the converter works at high disturbed that might result in extra power losses in the semiconductor devices. Since the converter frequency mode, the losses can impact the safe operation of the system. Moreover, since there is usually works at high frequency mode, the losses can impact the safe operation of the system. Moreover, only one active power switch in a power amplifier, it is challenging to increase the system power level. since there is usually only one active power switch in a power amplifier, it is challenging to increase 3.3.the system Full-Bridge power Inverter level. Based Topology

3.3. Full-BridgeSimilar to aInverter regular Based resonant Topology circuit, the full-bridge inverter is an effective method to provide ac excitation to a CPT system. Usually, at the primary side, four MOSFET switches are used to form the Similar to a regular resonant circuit, the full-bridge inverter is an effective method to provide inverter, which are driven by PWM signals. It is convenient to adjust the switching frequency and ac excitation to a CPT system. Usually, at the primary side, four MOSFET switches are used to form duty ratio to regulate the system power. At the secondary side, a diode rectiﬁer is used to provide dc the inverter, which are driven by PWM signals. It is convenient to adjust the switching frequency current to the load, either a resistor or a battery. The inverter is the key part to provide ac excitation and duty ratio to regulate the system power. At the secondary side, a diode rectifier is used to to the resonant circuit. With the help of this full-bridge inverter, multiple compensation circuits can provide dc current to the load, either a resistor or a battery. The inverter is the key part to provide be developed to realize capacitive power transfer, for example series L, LC, LCL, LCLC, CLLC, etc., ac excitation to the resonant circuit. With the help of this full-bridge inverter, multiple which will be explained below. compensation circuits can be developed to realize capacitive power transfer, for example series L, LC, LCL, LCLC, CLLC, etc., which will be explained below.

Series L Compensation

Energies 2017, 10, 1752 12 of 30

EnergiesSeries L2017 Compensation, 10, 1752 12 of 30

Since the capacitive coupler perf performsorms as capacitors, it is commoncommon to use inductors connected in series with the capacitors to compensate them [[50–55],50–55], resulting in the seriesseries LL compensationcompensation circuit,circuit, as shown in Figure 1515..

Figure 15. CircuitCircuit topology topology of of a aCPT CPT system system based based on on full-bridge full-bridge inverter inverter with with double-sided double-sided L compensation.L compensation.

In Figure 15, two inductors, L1 and L2, are used at both the primary and secondary sides to In Figure 15, two inductors, L1 and L2, are used at both the primary and secondary sides to compensate the mutual capacitors CM1 and CM2. The inductors can also provide EMI suppression compensate the mutual capacitors CM1 and CM2. The inductors can also provide EMI suppression function to the input side [56]. The inductance value and resonant frequency are determined by the function to the input side [56]. The inductance value and resonant frequency are determined by the value of capacitances. Usually, to reduce the inductance size and volume, the switching frequency value of capacitances. Usually, to reduce the inductance size and volume, the switching frequency is is very high. In practical applications, to reduce the complexity of the secondary side, the two very high. In practical applications, to reduce the complexity of the secondary side, the two inductors inductors can be combined to one and placed at the primary side [57]. In addition, to increase the can be combined to one and placed at the primary side [57]. In addition, to increase the system power system power level, the full-bridge inverter can be upgraded to a three-phase inverter with three level, the full-bridge inverter can be upgraded to a three-phase inverter with three legs, and more series legs, and more series inductors are required for the extra leg [58]. Moreover, active circuit with inductors are required for the extra leg [58]. Moreover, active circuit with MOSFETs and capacitors can MOSFETs and capacitors can be used to replace the series inductors, which can perform as be used to replace the series inductors, which can perform as inductive impedance and reduce the inductive impedance and reduce the system size [59–61]. Since there are more extra active switches system size [59–61]. Since there are more extra active switches in the circuit, the system efficiency is in the circuit, the system efficiency is then reduced. then reduced. The advantage of series L compensation is its simplicity, and it can be applied in both low and The advantage of series L compensation is its simplicity, and it can be applied in both low and high power applications. When it is used in biomedical devices to transfer power through human high power applications. When it is used in biomedical devices to transfer power through human tissue, the frequency can reach hundreds of MHz, and the power is relatively low [62]. It can also be tissue, the frequency can reach hundreds of MHz, and the power is relatively low [62]. It can also be applied in the excitation of a synchronous motor, and the system dc-dc efficiency can achieve over applied in the excitation of a synchronous motor, and the system dc-dc efficiency can achieve over 90% [22,23]. Compared to the PWM converter based CPT systems, the coupling capacitances can be 90% [22,23]. Compared to the PWM converter based CPT systems, the coupling capacitances can be relatively small and the switching frequency can be very high. The series compensation can also be relatively small and the switching frequency can be very high. The series compensation can also be used in the electric vehicle charging applications, where the tires are used as the dielectric material used in the electric vehicle charging applications, where the tires are used as the dielectric material to increase the coupling capacitances [27]. However, it still requires the switching frequency to be in to increase the coupling capacitances [27]. However, it still requires the switching frequency to be in MHz range to provide sufficient power capability. MHz range to provide sufficient power capability. The disadvantage of series L compensation lies in its requirements on inductance size and its The disadvantage of series L compensation lies in its requirements on inductance size and its sensitivity with parameter variations. Especially in long distance and high power scenarios, the sensitivity with parameter variations. Especially in long distance and high power scenarios, the inductances should be large enough to establish resonance or the switching frequency needs to be inductances should be large enough to establish resonance or the switching frequency needs to be high high enough. Similar to the power amplifier based CPT system, when the coupler capacitance enough. Similar to the power amplifier based CPT system, when the coupler capacitance changes, the changes, the resonances in the circuit can be disturbed and the power transfer process is then resonances in the circuit can be disturbed and the power transfer process is then terminated, which terminated, which limits its practical applications. limits its practical applications.

LC Compensation The main challenge in capacitive power transfer comes from the conflict conﬂict between the extremely small couplingcoupling capacitances capacitances and and the systemthe system power po requirements.wer requirements. Therefore, Therefore, to increase to theincrease equivalent the equivalentcoupling capacitance, coupling capacitance, external capacitors external can capacitors be used can to connect be used in to parallel connect with in parallel them, resulting with them, in a resultingdouble-sided in a double-sided LC compensation LC compensati circuit [35],on as circuit shown [35], in Figure as shown 16. in Figure 16.

Energies 2017, 10, 1752 13 of 30 Energies 2017, 10, 1752 13 of 30 Energies 2017, 10, 1752 13 of 30

Figure 16. Circuit topology of a CPT system based on full-bridge inverter with double-sided LC Figure 16. CircuitCircuit topology topology of ofa CPT a CPT system system based based on onfull-bridge full-bridge inverter inverter with with double-sided double-sided LC compensation. compensation.LC compensation.

In Figure 16, the parallel capacitors Cex1 and Cex2 are usually much larger than the mutual In Figure 1616,, thethe parallelparallel capacitorscapacitors CCexex1 and Cex22 areare usually usually much much larger larger than the mutual capacitances CM1 and CM2, which dominate the equivalent capacitance of the network. The inductor capacitances CM11 andand CCMM2,2 ,which which dominate dominate the the equivalent equivalent capaci capacitancetance of of the the network. network. The The inductor L1 and L2 resonate with the equivalent capacitances at the primary and secondary sides, respectively. L1 andand L22 resonateresonate with with the the equivalent equivalent capacitances capacitances at at the the primary primary and and secondary secondary sides, sides, respectively. Since the capacitances are increased, the required inductances L1 and L2 are also decreased, so as Since thethe capacitancescapacitances are are increased, increased, the the required required inductances inductancesL1 andL1 andL2 areL2 alsoare also decreased, decreased, so as so their as their size and volume. theirsize and size volume. and volume. The advantage of the double-sided LC compensation circuit is its feasibility in long distance The advantageadvantage ofof thethe double-sided double-sided LC LC compensation compensation circuit circuit is itsis feasibilityits feasibility in long in long distance distance and and high power applications [63–65]. Regardless of the mutual capacitances CM1 and CM2, the andhigh high power power applications applications [63–65 [63–65].]. Regardless Regardless of the mutualof the capacitancesmutual capacitancesCM1 and CCMM1 2and, the C externalM2, the external capacitances Cex1 and Cex2 can be relatively large. In addition, the resonances in this system externalcapacitances capacitancesCex1 and CexC1ex and2 can Cex be2 can relatively be relatively large. large. In addition, In additi theon, the resonances resonances in thisin this system system is is not sensitive to distance and misalignment variations in the capacitive coupler. Since the mutual isnot not sensitive sensitive to to distance distance and and misalignment misalignment variations variations in in the the capacitive capacitive coupler.coupler. SinceSince the mutual capacitances CM1 and CM2 are small, their variations do not have significant influence to the capacitances CCMM11 andandC MC2Mare2 are small, small, their their variations variations do not do have not significant have significant influence influence to the resonances to the resonances in the circuit and the system performance can be maintained. resonancesin the circuit in and the thecircuit system and performancethe system performance can be maintained. can be maintained. The disadvantage of double-sided LC compensation is that its system power is reversely The disadvantage of double-sided LC compensation is that its system power is reversely proportional to the coupling coefficient [66]. The system power increases with an increasing proportional toto the the coupling coupling coefficient coefficient [66 ].[66]. The systemThe system power power increases increases with an with increasing an increasing distance, distance, however, the system efficiency decrease dramatically with the distance. In addition, a distance,however, thehowever, system the efficiency system decreaseefficiency dramatically decrease dramatically with the distance. with the In addition,distance. aIn higher addition, power a higher power means a higher voltage and current stress on the circuit components, which can affect highermeans apower higher means voltage a andhigher current voltage stress and on current the circuit stress components, on the circuit which components, can affect the which safe can operation affect the safe operation of the CPT system. In addition, since the system power relates to the coupling theof the safe CPT operation system. of In the addition, CPT system. since the In system addition power, since relates the system to the coupling power relates coefficient, to the the coupling system coefficient, the system efficiency cannot be improved through circuit parameter optimization [35]. coefficient,efficiency cannot the system be improved efficiency through cannot circuit be improved parameter through optimization circuit parameter [35]. optimization [35]. By combining the series L and LC compensation together, it can result in a double-sided LCL By combining the series L and LC compensation together, it can result in a double-sided LCL compensation topology [67], as shown in Figure 17. compensation topology [[67],67], as shown in Figure 1717..

Figure 17. Circuit topology of a CPT system based on full-bridge inverter with double-sided LCL Figure 17. 17. CircuitCircuit topology topology of ofa CPT a CPT system system based based on full-bridge on full-bridge inverter inverter with withdouble-sided double-sided LCL compensation. compensation.LCL compensation.

In Figure 17, the series inductances Ls1 and Ls2 compensate only parts of the mutual In Figure 17, the series inductances Ls1 and Ls2 compensate only parts of the mutual capacitancesIn Figure C 17M1, theand series CM2, inductancesand the remainingLs1 and Lpartss2 compensate are compensated only parts by of the the LC mutual networks. capacitances It can capacitances CM1 and CM2, and the remaining parts are compensated by the LC networks. It can CprovideM1 and theCM flexibility2, and the to remaining tune the system parts arepower compensated through designing by the LC the networks. inductance It ratio can provideof L1 and the Ls1, provide the flexibility to tune the system power through designing the inductance ratio of L1 and Ls1, ﬂexibilityso as the ratio to tune of L the2 and system Ls2. However, power through the system designing power the is inductancestill reversely ratio proportional of L1 and L sto1, coupling so as the so as the ratio of L2 and Ls2. However, the system power is still reversely proportional to coupling ratiocoefficient, of L2 and andL its2 still. However, requires the a relatively system power large isvalue still of reversely series inductances. proportional to coupling coefﬁcient, coefficient, and it still requires a relatively large value of series inductances. and it still requires a relatively large value of series inductances. LCLC Compensation LCLC Compensation To overcome the limitation of LC and LCL compensation networks, a double-sided LCLC To overcome the limitation of LC and LCL compensation networks, a double-sided LCLC compensation topology [68] is proposed, as shown in Figure 18. compensation topology [68] is proposed, as shown in Figure 18.

Energies 2017, 10, 1752 14 of 30

LCLC Compensation To overcome the limitation of LC and LCL compensation networks, a double-sided LCLC Energies 2017, 10, 1752 14 of 30 compensationEnergies 2017, 10, 1752 topology [68] is proposed, as shown in Figure 18. 14 of 30

Figure 18. Circuit topology of a CPT system based on full-bridge inverter with double-sided LCLC Figure 18. Circuit topology of a CPT system based on full-bridge inverter with double-sided LCLC compensation.Figure 18. Circuit topology of a CPT system based on full-bridge inverter with double-sided LCLCcompensation. compensation.

In Figure 18, two external capacitors Cex1 and Cex2 remain the same with the previous LC In Figure 18, two external capacitors Cex1 and Cex2 remain the same with the previous LC compensationIn Figure circuit.18, two Extra external LC networks capacitors Lf1C-Cexf1 andand LCf2ex-C2f2remain are used the at samethe primary with the and previous secondary LC compensation circuit. Extra LC networks Lf1-Cf1 and Lf2-Cf2 are used at the primary and secondary sidescompensation to convert circuit. the voltage Extra LCsources networks V1 andLf 1V-C2 finto1 and currentLf2-Cf2 sourcesare used for at the the resonant primary andcircuits. secondary There sides to convert the voltage sources V1 and V2 into current sources for the resonant circuits. There aresides multiple to convert resonances the voltage in the sources circuit;V1 andfor example,V2 into current Cf1 can sources resonate for with the resonantboth Lf1 and circuits. L1, which There are are are multiple resonances in the circuit; for example, Cf1 can resonate with both Lf1 and L1, which are usedmultiple to increase resonances the voltage in the circuit; in the forcapacitive example, couplerCf1 can for resonate sufficient with power both transfer.Lf1 and L 1, which are used used to increase the voltage in the capacitive coupler for sufficient power transfer. to increaseThe advantage the voltage of inthe the double-sided capacitive coupler LCLC for compen sufficientsation power circuit transfer. is that the system power is The advantage of the double-sided LCLC compensation circuit is that the system power is proportionalThe advantage to the coupling of the double-sided coefficient, and LCLC the compensation system power circuit can be is regulated that the system through power circuit is proportional to the coupling coefficient, and the system power can be regulated through circuit parameterproportional design to the without coupling affecting coefficient, the coupling and the coefficient system power [68]. Therefore, can be regulated it is possible through to maintain circuit parameter design without affecting the coupling coefficient [68]. Therefore, it is possible to maintain aparameter high coupling design coefficient without affecting for efficiency the coupling consideration coefficient while [68]. fulfilling Therefore, the it ispower possible requirements. to maintain It a a high coupling coefficient for efficiency consideration while fulfilling the power requirements. It alsohigh has coupling the merit coefficient of the forLC efficiencycompensation consideration circuit that while the fulfilling resonant the inductances power requirements. L1 and L2 can It also be also has the merit of the LC compensation circuit that the resonant inductances L1 and L2 can be significantlyhas the merit reduced of the LC by compensation the external circuitcapacitors that theCex1resonant and Cex2.inductances L1 and L2 can be significantly significantly reduced by the external capacitors Cex1 and Cex2. reducedThe bydisadvantage the external of capacitors the double-sidedCex1 and LCLCCex2. compensation circuit is its complexity. Since there The disadvantage of the double-sided LCLC compensation circuit is its complexity. Since there are eightThe disadvantagepassive components of the double-sided in the circuit, LCLC the compensation system cost circuitand weight is its complexity. is then increased. Since there In are eight passive components in the circuit, the system cost and weight is then increased. In addition,are eight passivemore components components can in thecause circuit, extra the power system losses cost in and the weight circuit is that then can increased. reduce Inthe addition, system addition, more components can cause extra power losses in the circuit that can reduce the system efficiency.more components Therefore, can the cause system extra design power should losses pa iny the attention circuit thatto the can efficiency reduce theoptimization. system efficiency. As an efficiency. Therefore, the system design should pay attention to the efficiency optimization. As an alternative,Therefore, the the system LCLC designcompensation should pay can attention also be re toconfigured the efficiency as a optimization. CLLC topology As an [69] alternative, as shown the in alternative, the LCLC compensation can also be reconfigured as a CLLC topology [69] as shown in FigureLCLC compensation19. can also be reconfigured as a CLLC topology [69] as shown in Figure 19. Figure 19.

Figure 19. 19. CircuitCircuit topology topology of a of CPT a CPT system system based based on full-bridge on full-bridge inverter inverter with double-sided with double-sided CLLC Figure 19. Circuit topology of a CPT system based on full-bridge inverter with double-sided CLLC compensation.CLLC compensation. compensation.

In Figure 19, the positions of Lf1 and Cf1 are exchanged at the primary side, and the position of In FigureFigure 1919,, thethe positionspositions ofof LLff11 and Cf1 areare exchanged at at the primary side, and the position of Lf2 and Cf2 are exchanged at the secondary side. The other component positions remain the same. In Lff2 andand CCf2f 2areare exchanged exchanged at at the the secondary secondary side. side. The The other other component component positions positions remain remain the the same. same. In addition, the resonant relationship is not affected in the circuit. Usually, the inductors L1 and L2 Inaddition, addition, the the resonant resonant relationship relationship is isnot not affected affected in in the the circuit. circuit. Usually, Usually, the inductors L11 and LL22 have the largest volume, and the CLLC compensation can help reduce the required inductances of have thethe largestlargest volume,volume, andand the the CLLC CLLC compensation compensation can can help help reduce reduce the the required required inductances inductances of Lof1 L1 and L2 which can reduce the system volume. andL1 andL2 Lwhich2 which can can reduce reduce the the system system volume. volume. By comparing the L, LC, and LCLC networks and their variations, it shows that there are By comparing the L, LC, and LCLC networks and their variations, it shows that there are different compensation topologies that can be used to realize a CPT system. The only principle is different compensation topologies that can be used to realize a CPT system. The only principle is that their equivalent inductance should provide resonant condition with the mutual capacitances. that their equivalent inductance should provide resonant condition with the mutual capacitances. Since the LC circuit can be treated as a basic cell of the compensation circuit, the network can be Since the LC circuit can be treated as a basic cell of the compensation circuit, the network can be extended to m × n LC compensation [70–73], as shown in Figure 20. extended to m × n LC compensation [70–73], as shown in Figure 20.

Energies 2017, 10, 1752 15 of 30

By comparing the L, LC, and LCLC networks and their variations, it shows that there are different compensation topologies that can be used to realize a CPT system. The only principle is that their equivalent inductance should provide resonant condition with the mutual capacitances. Since the LC circuit can be treated as a basic cell of the compensation circuit, the network can be extended to m × n EnergiesLC compensation 2017, 10, 1752 [ 70–73], as shown in Figure 20. 15 of 30

Figure 20.20.Circuit Circuit topology topology of aof CPT a systemCPT system based onbased full-bridge on full-bridge inverter with inverterm × n LCwith compensation. m × n LC compensation. In Figure 20, there are m LC cells at the primary side and n LC cells at the secondary side. It should be pointedIn Figure out 20, that therem and aren dom LC not cells need at to the be equal,primary and side an asymmetricaland n LC cells circuit at the topology secondary can side. also beIt shouldused to be realize pointed a CPT out system. that m Inand addition, n do not the need position to be of equal, inductor and and an asymmetrical capacitor can alsocircuit be swapped,topology whichcan also can be resultsused to in realize different a CPT system system. performance In addition, [74 the–77 position]. The number of induct ofor LC and cells capacitor determines can also the besystem swapped, power which and efficiency, can results which in different needs to system be optimized performance according [74–77]. to practical The number requirements of LC cells [78]. determinesFor example, the a largersystem number power of and LC networkefficiency, can whic provideh needs more to designbe optimized flexibility, according but the power to practical losses requirementsmight increase [78]. and For the example, system efficiency a larger number can be lower.of LC network Besides, thecan LCprovide network more can design also beflexibility, twisted, butresulting the power in a Z-impedance losses might network increase similar and tothe the syst regularem efficiency resonant convertercan be lower. which hasBesides, been studiedthe LC networkin a CPT systemcan also [79 be]. twisted, resulting in a Z-impedance network similar to the regular resonant converterConsidering which has different been studied compensation in a CPT circuit system topologies [79]. presented above, they are compared from differentConsidering aspects, asdifferent shown incompensati Table1. on circuit topologies presented above, they are compared from different aspects, as shown in Table 1. Table 1. Comparison of different compensation circuits. Table 1. Comparison of different compensation circuits. Category Inductance Output Type Output Expression Output Power Category Inductance Output Type Output Expression Output Power 2 1 D D 2 1 Sepic CPT [42]* L1, L2 >> 21 Voltage source Vout = − · Vin Pout = · Vin · >> ω CM =⋅1 DD =⋅⋅1−DD R1L Sepic CPT [42] * LL12, Voltage source VV PV ω 2C out− in out− in 1 M 1 1 D 1 DR2 1L Class E CPT [80] ** L >> 2 , L ≈ 2 Voltage source V = 0.76V P = (0.76 · V ) · 1 ω CM 2 ω CM out in out in RL >>111 ≈ = =⋅⋅()2 12 1 ClassSeries EL CPTCPT [80] [22] ** LL12L1, L2 ≈ , 2 VoltageVoltage source source VVVout = 0.76V PVoutPout 0.76= V · in ωω22ω CM outin in in RL CCM M RL 2 kC 1 1 L L ≈ I = ωC · 2 · V LC CPT [35] 1, 2 ω2C1 Current source 2 M k 1 Pout = ωCM · 2 · V11 · RL LL, ≈ M = C PV=⋅kC Series L CPT [22] 12 ω 2 Voltage source VVout in out in CM RL 2 kC Cf 1Cf 2 Cf 1Cf 2 LCLC CPT [68] L , L ≈ 2 Current source I = ωC · · V P = C · · V · R 1 2 ω CM 2 M C C 1 out ω M C C 12 L k 1 21 1 2 ≈ C =⋅⋅ω =⋅⋅⋅ω 1 LC CPT [35] LL, Current source IC21M 2 V PCVR * In a sepic CPT system, D12denotesω 2C the duty ratio of the PWM signal. ** In a class-Ek amplifierout with 50% M duty2 1 ratio, L M C kC zero voltage and zero derivative at switch on instant, there is Vout = 0.76 Vin. 2 k CC CC ≈ C =⋅ω ff12 ⋅ =⋅ω ff12 ⋅⋅ LCLC CPT [68] LL12, Current source IC V PC VR ω 2C 21M CC out MCC 1 L Table1 provides the comparisonM according to the inductance design12 and output properties.12 Since the mutual* In a sepic capacitance CPT system,CM Dis denotes usually the very duty small, ratio of a the good PWM CPT sign systemal. ** In shoulda class-E be amplifier able to with reduce 50% the requiredduty resonantratio, zero inductances.voltage and zero It deriva showstive that at switch the sepic, on instant, class there E, and is V seriesout = 0.76 L V compensationin. need relatively large inductances, which means they are more suitable for the short distance applications. In theTable LC and 1 provides LCLC compensation, the comparison since according the coupling to the coefficient inductancekC is design usually and much output smaller properties. than 1.0, Sincethe required the mutual resonant capacitance inductance CM is can usually be significantly very small,reduced. a good CPT Besides, system the should output be property able to reduce shows thethat required they perform resonant as current inductances. sources It for shows the output, that th whiche sepic, is class suitable E, and for vehicleseries L charging compensation applications need withrelatively a battery large load. inductances, The comparison which means of LC andthey LCLCare more shows suitable that C forf1 and the Cshortf2 provide distance the applications. flexibility to Inregulate the LC the and system LCLC power compensation, without affect since the the coupling coupling coefficient, coefficient which kC is meansusually the much LCLC smaller system than can 1.0,achieve the arequired higher couplingresonant coefficient, inductance so can as thebe systemsignificantly efficiency. reduced. Besides, the output property shows that they perform as current sources for the output, which is suitable for vehicle charging applications with a battery load. The comparison of LC and LCLC shows that Cf1 and Cf2 provide the flexibility to regulate the system power without affect the coupling coefficient, which means the LCLC system can achieve a higher coupling coefficient, so as the system efficiency.

Energies 2017, 10, 1752 16 of 30 Energies 2017, 10, 1752 16 of 30

4. Capacitive Coupler Structure The capacitivecapacitive couplercoupler consistsconsists of of multiple multiple metal metal plates plates that that are are used used to to generate generate electric electric ﬁelds fields to transferto transfer power. power. There There are coupling are coupling capacitances capacitanc betweenes between each pair each of plates, pair andof plates, different and structure different of thestructure capacitive of the coupler capacitive results coupler in different results coupling in different model. coupling The coupling model. capacitancesThe coupling determine capacitances the powerdetermine transfer the power performance, transfer whichperformance, will be which discussed will inbe thisdiscussed section. in Dependingthis section. onDepending the coupler on structure,the coupler there structure, are two-plate there are couplertwo-plate structure, coupler stru four-platecture, four-plate parallel structure, parallel structure, four-plate four-plate stacked structure,stacked structure, six-plate six-plate structure, structure, and repeater and repeater structure, structure, which will which be discussed will be discussed in following. in following.

4.1. Two-Plate Structure In aa CPTCPT system,system, at at least least two two metal metal plates plates need need to to be be used used to formto form a capacitive a capacitive coupler, coupler, and and it can it alsocan bealso called be called a unipolar a unipolar structure. structure. One plate One is placed plate atis the placed primary at sidethe primary as a power side transmitter as a power and thetransmitter other plate and is the placed other at theplate secondary is placed side at the as asecondary power receiver. side as The a power mutual receiver. capacitance The between mutual thecapacitance two plates between provides the the two path plates for theprovides current the to ﬂowpath forwardfor the current to the load, to flow and forward a conductive to the path load, is thenand a required conductive to allow path the is currentthen required ﬂowing backto allow to the the primary current side. flowing Therefore, back ato typical the primary structure side. of a two-plateTherefore, capacitive a typical structure coupler is of shown a two-plate in Figure capacitive 21. coupler is shown in Figure 21.

Figure 21. Structure of a two-plate capacitive coupler.

In Figure 21, the capacitive coupler can be applied in both short and long distance applications. In Figure 21, the capacitive coupler can be applied in both short and long distance applications. It also needs to be pointed out that the plate size can be different, especially in an asymmetric It also needs to be pointed out that the plate size can be different, especially in an asymmetric coupler structure. The conductive path can be realized by metal-to-metal connection, resulting in a coupler structure. The conductive path can be realized by metal-to-metal connection, resulting in quasi-wireless CPT system. For example, one receiver terminal can be directly grounded. In the a quasi-wireless CPT system. For example, one receiver terminal can be directly grounded. In the application of the propulsion of a high-speed train, the metal wheels can be used as the returning application of the propulsion of a high-speed train, the metal wheels can be used as the returning path, path, which can replace the conventional pantograph and improve the system reliability [34]. In which can replace the conventional pantograph and improve the system reliability [34]. In addition, addition, since it is easy to get access to the earth ground in most applications and there are usually since it is easy to get access to the earth ground in most applications and there are usually parasitic parasitic capacitances between the plates to the earth ground, the earth ground is therefore used as capacitances between the plates to the earth ground, the earth ground is therefore used as the the conductive returning path. For example, the receiver terminals can be floating from the ground, conductive returning path. For example, the receiver terminals can be ﬂoating from the ground, and the stray capacitance can then be used to transfer power. In the scenario of electric vehicle and the stray capacitance can then be used to transfer power. In the scenario of electric vehicle charging, the parasitic capacitance between the vehicle chassis and the earth ground is used to charging, the parasitic capacitance between the vehicle chassis and the earth ground is used to conduct conduct current. One critical concern is to increase the conductivity of earth ground, and reduce the current. One critical concern is to increase the conductivity of earth ground, and reduce the system system losses [81]. losses [81]. The advantage of the two-plate coupler structure is its simplicity. The recent study also shows The advantage of the two-plate coupler structure is its simplicity. The recent study also shows that it has a good tolerance to large misalignment, which is named as a single-wire CPT system that it has a good tolerance to large misalignment, which is named as a single-wire CPT system [82,83]. [82,83]. In addition, experiments show that effective power transfer can be achieved by two plates In addition, experiments show that effective power transfer can be achieved by two plates even when even when their parasitic capacitances with the ground are extremely small. All of these promote their parasitic capacitances with the ground are extremely small. All of these promote the theoretical the theoretical study in CPT system, which attempts to explain the phenomenon from the aspect of study in CPT system, which attempts to explain the phenomenon from the aspect of fundamental fundamental physics. For example, the standing-wave theory assumes that the length of the physics. For example, the standing-wave theory assumes that the length of the inductor is equal to the inductor is equal to the quarter wavelength of the electromagnetic wave, which means the CPT quarter wavelength of the electromagnetic wave, which means the CPT system works as an antenna system works as an antenna system [84]. Although this two-plate structure is the most basic method system [84]. Although this two-plate structure is the most basic method to realize a CPT system, the to realize a CPT system, the recent investigation is far from sufficient. recent investigation is far from sufﬁcient.

Energies 2017, 10, 1752 17 of 30 Energies 2017, 10, 1752 17 of 30 Energies 2017, 10, 1752 17 of 30 4.2.4.2. Four-Plate Parallel Structure 4.2. Four-Plate Parallel Structure TheThe four-platefour-plate parallelparallel structure structure is is the the most most common common way way to realizeto realize a capacitive a capacitive coupler coupler [85], [85], and The four-plate parallel structure is the most common way to realize a capacitive coupler [85], itand can it also can bealso called be called a bipolar a bipolar structure, structure, which includeswhich includes two pairs two of metalpairs platesof metal as shownplates as in Figureshown 22in. and it can also be called a bipolar structure, which includes two pairs of metal plates as shown in ItFigure is named 22. It because is named the because two pairs the oftwo plates pairs are of parallelplates are arranged. parallel arranged. Considering Considering the vehicle the charging vehicle Figure 22. It is named because the two pairs of plates are parallel arranged. Considering the vehicle applications,charging applications, the transmitter the transmitter plates P1 andplatesP2 Pare1 and placed P2 are on placed the ground on the side ground in the side same in horizontal the same charging applications, the transmitter plates P1 and P2 are placed on the ground side in the same plane,horizontal and theplane, receiver and the plates receiverP3 and platesP4 are P3 placedand P4 atare the placed vehicle at the side vehicle and also side in and the samealso in horizontal the same horizontal plane, and the receiver plates P3 and P4 are placed at the vehicle side and also in the same plane.horizontal Therefore, plane. thisTherefor structuree, this can structure also be can called also a four-placebe called a horizontalfour-place structurehorizontal [32 structure]. [32]. horizontal plane. Therefore, this structure can also be called a four-place horizontal structure [32].

Figure 22. Structure of a four-plate parallel capacitive coupler. Figure 22. Structure of a four-plate parallel capacitive coupler. In Figure 22, there is a coupling capacitance between each pair of plates, totally resulting in six InIn FigureFigure 2222,, therethere isis aa couplingcoupling capacitancecapacitance betweenbetween eacheach pairpair ofof plates,plates, totallytotally resultingresulting inin sixsix capacitances, in which C13 and C24 are defined as main coupling capacitances; C14 and C23 are capacitances, inin which whichC 13C13and andC24 Care24 are defined defined as main as main coupling coupling capacitances; capacitances;C14 and CC1423 andare definedC23 are defined as cross coupling capacitances; C12 and C34 are defined as the self-coupling capacitances. defined as cross coupling capacitances;C C C12 and C34 are defined as the self-coupling capacitances. asDifferent cross coupling from the capacitances; conventional 12couplingand 34 model,are defined this six-capacitance as the self-coupling model capacitances. is more effective Different and Different from the conventional coupling model, this six-capacitance model is more effective and fromaccurate. the conventionalUsually, when coupling the plates model, are thiswell-align six-capacitanceed and the model plate is moredistance effective is short, and the accurate. main accurate. Usually, when the plates are well-aligned and the plate distance is short, the main Usually,couplings when dominate the plates the arecapacitive well-aligned model and and the the plate other distance couplings is short, can thebe mainneglected couplings to simplify dominate the couplings dominate the capacitive model and the other couplings can be neglected to simplify the thecircuit capacitive analysis. model However, and the when other the couplings plates canare bemisaligned neglected and to simplify the plate the distance circuit analysis. is long, However,the cross circuit analysis. However, when the plates are misaligned and the plate distance is long, the cross whencouplings the plates are relatively are misaligned large andand the they plate need distance to be is long,considered the cross in couplings circuit analysis. are relatively Since large the couplings are relatively large and they need to be considered in circuit analysis. Since the andsix-capacitance they need to model be considered is very incomplicated, circuit analysis. the Sinceequivalent the six-capacitance behavior source model model is very [86] complicated, is a more six-capacitance model is very complicated, the equivalent behavior source model [86] is a more theeffective equivalent and convenient behavior sourcemethod model to model [86] the is a capacitive more effective coupler and as convenient shown in Figure method 23. to model the capacitiveeffective and coupler convenient as shown method in Figure to model 23. the capacitive coupler as shown in Figure 23.

Figure 23. Equivalent behavior source model of a capacitive coupler. Figure 23. Equivalent behavior source model of a capacitive coupler. In Figure 23, the capacitive coupler is treated as a two-port network with the input port at the In Figure 23, the capacitive coupler is treated as a two-port network with the input port at the primaryIn Figure side and 23, the output capacitive port coupler at the secondary is treated si asde. a two-portIt needs to network be emphasized with the that input this port two-port at the primary side and the output port at the secondary side. It needs to be emphasized that this two-port primarynetwork sidecan andbe used the output as the portequivalence at the secondary of differ side.ent coupler It needs structures. to be emphasized For example, that this it two-porthas been network can be used as the equivalence of different coupler structures. For example, it has been networkused in Figure can be 4 used as a as general the equivalence capacitive of coupler different to coupleranalyze structures.the fundamental For example, working it hasprinciple been used of a used in Figure 4 as a general capacitive coupler to analyze the fundamental working principle of a inCPT Figure system.4 as aSpecifically, general capacitive in this coupler four-plate to analyze coupler the fundamentalstructure, the working relationship principle between of a CPT the CPT system. Specifically, in this four-plate coupler structure, the relationship between the system.capacitances Speciﬁcally, in Figure in 22 this and four-plate the equivalent coupler capa structure,citances the in relationship Figure 23 is betweenshown below the capacitances [31,87]. in Figurecapacitances 22 and in the Figure equivalent 22 and capacitances the equivalent in Figurecapacitances 23 is shown in Figure below 23 is [31 shown,87]. below [31,87].  CC− CC C = CC12 34− CC 14 23 CM == C+++1212C 3434−C14 14C23 23 CMM CCCC12C +C 34+C 14+C 23  CCCC12+++34 14 23  ()()CC12( +⋅++ 34)·( 14CC+ 23 )  =+13C13 C 1414 C23 23C24 24 CCCin1 =()()CC+⋅++ + CC+ + C12 (18)(18)  in112=+13C12+++C 1434 C14 23C23 24 CCin112CCCC12 34 14 23 (18)  CCCC(C13++++C23)·(C14+C24)  C = 12+⋅+ 34 14 23 + C  in2 ()()CCC12+C34+C CC14+C23 34 CC=+()()CC13+⋅+ 23 CC 14 24 CCin234=+CCCC13+++ 23 14 24  in23412+++ 34 14 23  CCCC12 34 14 23

Energies 2017, 10, 1752 18 of 30 Energies 2017, 10, 1752 18 of 30

It showsshows that that the the equivalent equivalent mutual mutual capacitance capacitanceCM is C determinedM is determined by the by main the and main cross and coupling cross capacitances,coupling capacitances, while, the self-capacitanceswhile, the self-capacitancesCin1 and Cin 2Calsoin1 and relate Cin to2 also the self-couplings.relate to the self-couplings. In this four-plate In parallelthis four-plate structure, parallel the capacitances structure, Cthe12 andcapacitancesC34 are usually C12 and very C34 small, are usually which limits very thesmall, self-capacitances. which limits Therefore,the self-capacitances. external capacitors Therefore, are external required capacitors to be connected are required with the to couplerbe connected to increase with thethe equivalentcoupler to self-capacitances,increase the equivalent as shown self-capacitances, in Figure9. However, as shown the in external Figure capacitances 9. However, usually the external take extra capacitances space in theusually system take [68 extra]. Moreover, space in thisthe parallelsystem [68]. structure Moreover, is sensitive this parallel to angular structure misalignment. is sensitive For to example, angular whenmisalignment. there is 90 For◦ angular example, shift when between there the is transmitter90° angular and shift receiver, between the the system transmitter power canand reducereceiver, to zero,the system which power limits itscan practical reduce to applications. zero, which limits its practical applications.

4.3. Four-Plate Stacked Structure

To increaseincrease thethe self-capacitancesself-capacitances andand eliminateeliminate thethe externalexternal capacitances,capacitances, thethe primaryprimary platesplates PP11 and PP22 areare placedplaced close close to to each each other other [88 [88,89].,89]. Similarly, Similarly, the the distance distance between between secondary secondary plates platesP3 and P3 Pand4 is P also4 is also reduced, reduced, as shown as shown in Figure in Figure 24. Since 24. Since all the all platesthe plates are stackedare stacked together, together, it is namedit is named as a four-plateas a four-plate stacked stacked structure. structure. When When it is it used is used in the in vehiclethe vehicle charging charging application, application, all theall platesthe plates are verticallyare vertically arranged arranged that that is also is also called called a four-plate a four-plate vertical vertical structure structure [31]. [31].

Figure 24. Structure of a four-plate stacked capacitive coupler.

In Figure 24, the size of P1 and P3 is larger than that of P2 and P4 to ensure the coupling between In Figure 24, the size of P1 and P3 is larger than that of P2 and P4 to ensure the coupling between each pair of plates. Since the distance between the same-side plates is small, the capacitances C12 each pair of plates. Since the distance between the same-side plates is small, the capacitances C12 and and C34 can be significantly increased. The coupling model of this stacked structure is the same as C can be significantly increased. The coupling model of this stacked structure is the same as that of that34 of the parallel structure in Figure 22, and it can also be simplified as the two-port behavior the parallel structure in Figure 22, and it can also be simplified as the two-port behavior source model source model in Figure 23 using Equation (18). Compared to the four-plate parallel structure, since in Figure 23 using Equation (18). Compared to the four-plate parallel structure, since the plates P2 and the plates P2 and P4 are inserted between P1 and P3, this stacked coupler is much more compact. In P are inserted between P and P , this stacked coupler is much more compact. In addition, since all addition,4 since all the plates1 are3 center aligned, it is robust to angular misalignment. Specifically, the plates are center aligned, it is robust to angular misalignment. Specifically, when the plates are in when the plates are in circular shape, the angular rotation has no influence to the coupling circular shape, the angular rotation has no influence to the coupling capacitances. capacitances. However, one limitation of the stacked coupler is its relatively small mutual capacitance. Since However, one limitation of the stacked coupler is its relatively small mutual capacitance. Since the coupler is more compact, the cross-coupling capacitances C14 and C23 are increased. According to the coupler is more compact, the cross-coupling capacitances C14 and C23 are increased. According Equation (18), the equivalent mutual capacitance CM is therefore reduced. Another important concern to Equation (18), the equivalent mutual capacitance CM is therefore reduced. Another important in the stacked coupler is the voltage stress between the same-side plates. Especially in long-distance concern in the stacked coupler is the voltage stress between the same-side plates. Especially in power transfer scenario, the voltage stress between the adjacent plates is usually very high. Therefore, long-distance power transfer scenario, the voltage stress between the adjacent plates is usually very reliable insulation should be applied on the surface of the plates. high. Therefore, reliable insulation should be applied on the surface of the plates. 4.4. Six-Plate Structure 4.4. Six-Plate Structure The four-plate parallel and stacked structures can be combined together [90,91], resulting in a The four-plate parallel and stacked structures can be combined together [90,91], resulting in a six-plate coupler structure, as shown in Figure 25. six-plate coupler structure, as shown in Figure 25.

Energies 2017, 10, 1752 19 of 30 Energies 2017, 10, 1752 19 of 30 Energies 2017, 10, 1752 19 of 30

Figure 25. Structure of a six-plate capacitive coupler. Figure 25. Structure of a six-plate capacitive coupler.

In Figure 25, a four-plate parallel structure is placed between two large plates P5 and P6. The In FigureFigure 25 25,, a a four-plate four-plate parallel parallel structure structure is placed is placed between between two largetwo large plates platesP5 and PP5 6and. The P plates6. The plates P1, P2, P3, and P4 work as active plates to transfer power, and the plates P5 and P6 work as Pplates, P , P1,, P and2, PP3, andwork P4as work active as platesactiveto plates transfer to transfer power, andpower, the platesand theP platesand P Pwork5 and asP6 auxiliarywork as auxiliary1 2 3 plates 4to increase the equivalent self-capacitance and serve as5 electric6 field shielding. platesauxiliary to increaseplates to the increase equivalent the self-capacitanceequivalent self-capacitance and serve as electricand serve field as shielding. electric field When shielding. it is used When it is used in a CPT system, the active plates have direct connection with the compensation inWhen a CPT it is system, used in the a activeCPT system, plates havethe active direct plat connectiones have direct with the connection compensation with the components, compensation and components, and the large plates are in floating status. If the primary side plate P5 is connected to thecomponents, large plates and are the in large floating plates status. are in If thefloating primary status. side If platethe primaryP5 is connected side plate to P the5 is earthconnected ground, to the earth ground, the secondary side plate P6 is also equivalently connected to the ground, resulting the earth secondary ground, side the plate secondaryP is also side equivalently plate P6 is also connected equivalently to the connected ground, resulting to the ground, in a significant resulting in a significant reduction of6 the electric fields emission to the surroundings. Therefore, the large reductionin a significant of the reduction electric fields of the emission electric to fields the surroundings. emission to the Therefore, surroundings. the large Therefore, plates also the work large as plates also work as the shielding to electric fields, which is an important advantage of this structure. theplates shielding also work to electric as the shielding fields, which to electric is an important fields, which advantage is an important of this structure. advantage of this structure. Since there is capacitive coupling between each two plates, there are totally 15 capacitances in Since therethere isis capacitivecapacitive coupling coupling between between each each two two plates, plates, there there are are totally totally 15 capacitances15 capacitances in this in this six-plate coupler. In an approximate model, only the main and self-couplings are considered six-platethis six-plate coupler. coupler. In an In approximate an approximate model, model, only the only main the and main self-couplings and self-couplings are considered are considered and the and the cross couplings can be neglected. In an accurate model, all the capacitances should be crossand the couplings cross couplings can be neglected. can be neglected. In an accurate In an model, accurate all the model, capacitances all the shouldcapacitances be included should and be included and the two-port behavior source model can be used to represent the coupler. The theincluded two-port and behavior the two-port source behavior model can source be used mode to representl can be theused coupler. to represent The relationship the coupler. between The relationship between the full-capacitance and two-port model is discussed in [90]. therelationship full-capacitance between and the two-port full-capacitance model is and discussed two-port in model [90]. is discussed in [90]. 4.5. Electric Field Repeater 4.5. Electric Field Repeater The aforementioned capacitive couplers provide the fundamental method to realize a CPT The aforementionedaforementioned capacitive capacitive couplers couplers provide provide the fundamentalthe fundamental method method to realize to realize a CPT system,a CPT system, and they can be used in both short and long distance applications. When the transfer andsystem, they and can bethey used can in be both used short in and both long short distance and applications.long distance When applications. the transfer When distance the needstransfer to distance needs to be further increased, the concept of electric field repeater is then proposed. For bedistance further needs increased, to be thefurther concept increased, of electric the field concept repeater of electric is then proposed.field repeater For example,is then proposed. the two-plate For example, the two-plate structure can be used to realize a repeater system as shown in Figure 26. structureexample, canthe two-plate be used to structure realize a repeatercan be used system to re asalize shown a repeater in Figure system 26. as shown in Figure 26.

Figure 26. An electric field repeater system based on two-plate coupler structure. Figure 26. An electric field ﬁeld repeater system based on two-plate coupler structure.

In Figure 26, the metal plate P1 is used as a power transmitter, P2 is used as power receiver #1, In Figure 26, the metal plate P1 is used as a power transmitter, P2 is used as power receiver #1, and PIn3 is Figure used 26as, power the metal receiver plate #2.P1 isElectric used asfield a power is used transmitter, between platesP2 is to used transfer as power power. receiver It needs #1, and P3 is used as power receiver #2. Electric field is used between plates to transfer power. It needs andto beP pointed3 is used out as powerthat the receiver receiver #2. can Electric also work ﬁeld as is useda repeater between to continuously plates to transfer increase power. the Ittransfer needs to be pointed out that the receiver can also work as a repeater to continuously increase the transfer todistance. be pointed In this out example, that the each receiver stage can has also direct work connection as a repeater with to the continuously earth ground, increase which theutilizes transfer the distance. In this example, each stage has direct connection with the earth ground, which utilizes the distance.earth as the In thiscurrent example, returning each stagepath. hasBecause direct of connection the connection with the with earth earth, ground, this two-plate which utilizes coupler the earth as the current returning path. Because of the connection with earth, this two-plate coupler earthbased assystem the current is a quasi-wireless returning path. repeater Because system. of the connection When the withfour-plate earth, parallel this two-plate coupler coupler structure based is based system is a quasi-wireless repeater system. When the four-plate parallel coupler structure is systemused to isrealize a quasi-wireless a repeater system repeater [92,93], system. its Whenstructur thee is four-plate shown in parallel Figure 27. coupler structure is used to used to realize a repeater system [92,93], its structure is shown in Figure 27. realize a repeater system [92,93], its structure is shown in Figure 27.

Energies 2017, 10, 1752 20 of 30 Energies 2017, 10, 1752 20 of 30

FigureFigure 27. 27. AnAn electric electric field ﬁeld repeater repeater system based on four-plate parallel coupler structure.

Figure 27 shows that the connection between the transmitter and receiver is eliminated, which Figure 27 shows that the connection between the transmitter and receiver is eliminated, which is a is a pure wireless repeater system. The most critical challenge in a repeater system is the transfer pure wireless repeater system. The most critical challenge in a repeater system is the transfer efficiency. efficiency. Especially in a long-distance application, when the number of repeater stage increases, Especially in a long-distance application, when the number of repeater stage increases, the power loss the power loss in the system could increase dramatically. The other important research topic in a in the system could increase dramatically. The other important research topic in a repeater system repeater system is to balance the consumed power and transferred power of each stage. It is is to balance the consumed power and transferred power of each stage. It is important to ensure important to ensure that sufficient power is transferred to the next stage without affecting the that sufficient power is transferred to the next stage without affecting the power requirement of the power requirement of the local stage. local stage. The repeater system represents a newly generated research direction in the WPT technology, The repeater system represents a newly generated research direction in the WPT technology, which which is the multiple inputs and multiple outputs WPT system (not only CPT, but also IPT). There is the multiple inputs and multiple outputs WPT system (not only CPT, but also IPT). There are already are already some pioneering works, such as the matrix transmitter and multiple pickup CPT some pioneering works, such as the matrix transmitter and multiple pickup CPT systems as presented systems as presented in [94,95]. In the future, the study on these systems can be a very promising in [94,95]. In the future, the study on these systems can be a very promising research direction. research direction. 5. Comparison of CPT and IPT Technology 5. Comparison of CPT and IPT Technology The CPT technology is the duality of the conventional IPT technology [96], and the CPT technology has beenThe developedCPT technology from achievementsis the duality in of IPT the systems. conventional The IPT IPT system technology utilizes [96], magnetic and the fields CPT to technologytransfer power, has while been the developed CPT system from uses achievemen electric fields.ts Therein IPT is inductivesystems. couplingThe IPT insystem an IPT utilizes system, magneticwhile the capacitivefields to transfer coupling power, dominates while athe CPT CPT system. system Capacitive uses electric matching fields. networks There is are inductive used in couplingan IPT system, in an whileIPT system, inductive while compensations the capacitive are coupling used in adominates CPT system a CPT to establish system. resonancesCapacitive matchingwith the coupler.networks For are example, used in an the IPT series-series system, while compensation inductive circuitcompensations in an IPT are system used isin similara CPT systemto the double-sided to establish resonances LC compensation with the circuitcoupler. in For a CPT example, system, the and series-ser they haveies compensation the same frequency circuit inproperty an IPT [system97]. The is doublesimilar sidedto the LCCdouble-sided compensation LC compensation in an IPT system circuit also in a has CPT the system, same input and they and haveoutput the performance same frequency with property the double [97]. sided The double LCLC compensationsided LCC compensation in a CPT system in an IPT [98]. system Although also hasthe CPTthe same and IPTinput systems and output have manyperformance similarities, with therethe double are also sided differences LCLC compensation between them in that a CPT can systemdistinguish [98].them Although from differentthe CPT applications.and IPT systems have many similarities, there are also differences between them that can distinguish them from different applications. 5.1. Advantages of CPT Technology 5.1. Advantages of CPT Technology Compared to the conventional IPT technology, the CPT system mainly has three advantages: negligibleCompared eddy-current to the conventional loss, low cost IPT and technology, weight, and the good CPT misalignment system mainly performance, has three which advantages: will be negligibleexplained aseddy-current follows. loss, low cost and weight, and good misalignment performance, which will be explainedFirst, in anas IPTfollows. system, the high-frequency magnetic fields can generate eddy-current losses in the nearbyFirst, metal, in an causing IPT system, significant the high-frequency temperature rise magn in bothetic low fields and can high generate power applications.eddy-current Especially losses in thewhen nearby it is used metal, in thecausing vehicle significant charging temperature application, therise system in both power low canand reachhigh uppower to 10s applications. of kW, and Especiallythe magnetic when fields it is can used induce in the a largevehicle amount charging of heat application, in the nearby the system metal object, power which can reach is a dangerous up to 10s offire kW, hazard. and the However, magnetic in fields a CPT can system, induce where a large the amount electric fieldsof heat are in usedthe nearby to transfer metal power, object, there which is isno a concerndangerous about fire the hazard. eddy-current However, loss. in a Therefore, CPT system it, can where beused the electric in the areafields that arehas used unavoidable to transfer power,metal materials. there is no concern about the eddy-current loss. Therefore, it can be used in the area that has unavoidableSecond, metal in an IPTmaterials. system, the inductive coupler contains two coils that work as a loosely coupled transformer.Second, in Since an theIPT couplingsystem, the coefficient inductive is usuallycoupler verycontains small, two we coils need that to increase work as the a currentloosely coupledcirculating transformer. in the coils Since to transfer the coupling enough coefficient power. Considering is usually very the skinsmall, depth we need of the to conductorincrease the at current circulating in the coils to transfer enough power. Considering the skin depth of the conductor at very high frequency, a large amount of Litz-wire is required to build the coils, which

Energies 2017, 10, 1752 21 of 30 very high frequency, a large amount of Litz-wire is required to build the coils, which increases the system cost and weight. In addition, the conductive losses in the coil can be significant in high power application. Moreover, to enhance the magnetic coupling, magnetic plates made by ferrite iron are used at both the primary and secondary sides, which increase the system cost and weight even further. However, in a CPT system, only several pieces of metal plates are used to build the capacitive coupler. The material and thickness of the plates do not affect the system performance very much. For example, the aluminum sheets can be used to realize a CPT system [68], and a thickness of 2 mm is sufficient to transfer 2.4 kW power. Since the aluminum plate is much cheaper and lighter than the Litz-wire made by copper, the system cost and weight can therefore be significantly reduced. In low power applications, the PCB board can be used as the plate. The shape of the coupler plate does not affect the power transfer performance, as long as the total area is achieved. Third, the experimental results show that the CPT system has much better misalignment performance than the IPT system. For example, in a CPT system with 610 mm × 610 mm metal plates, the system can maintain 89.4% of the well-aligned power with 300 mm misalignment [68]. However, in a comparable IPT system with a 600 mm × 800 mm inductive coupler, the system power drops to 56% of the well-aligned value with 310 mm misalignment [99]. Therefore, it can be concluded that the CPT technology is suitable for electric vehicle charging applications where the misalignment is relatively large and unavoidable.

5.2. Limits of CPT Technology Compared to the IPT system, the recent CPT technology still needs a long way to mature and there are some limitations to overcome: low power density, low efficiency, and strong magnetic field emissions, which will be explained as follows. First, in an IPT system, a 450 mm × 450 mm inductive coupler can achieve 7.7 kW power transfer for vehicle charging application, resulting in a power density of 38.0 kW/m2 [100]. However, in a CPT system, when the size of the capacitive coupler is 914 mm × 914 mm, it can achieve 1.87 kW power transfer for vehicle charging application, which means the power density is only 2.2 kW/m2. Therefore, in long distance applications, it shows that the power density of a CPT system is much smaller than that of an IPT system. It is because that the coupling capacitance is usually in the pF range when the transfer distance is in 100s of mm range. According to the power analysis in Equations (4) and (7), for a given capacitive coupler, the effective method to increase the power density is to increase the plate voltage and the switching frequency. Therefore, in future research, the compensation circuit topology and parameters can be optimized to further increase the plate voltage, and the insulation between plates needs to be improved. In addition, the switching frequency can be increased to 10s of MHz level to acquire high power. Second, in an IPT system [99], the DC-DC efficiency from the dc source to dc load can achieve 97%, while a CPT system can realize a DC-DC efficiency of 91.6% for electric vehicle charging application [90]. The power loss in a CPT system mainly comes from the conductive loss in the compensation inductors. Since the switching frequency is usually higher than 1 MHz, the skin effect of the wire becomes apparent and extra losses can be induced. If the switching frequency increases even further, the power losses can be more significant. This is an important concern in the system design. In addition, in the recent design, the inductors have air core to eliminate the magnetic losses [68], but the volume is too large, which is difficult for practice use. In future design, high frequency and low loss magnetic material can be adopted to reduce the power loss and volume of the inductors. Third, in an IPT system, the magnetic fields can be easily shielded by ferrite and aluminum plates, and the leakage fields to the surrounding area can be reduced below safety level. For example, the safe range of a 3.5 kW IPT system is about 200 mm away from the coupler [101]. However, the electric field emission in a CPT system is difficult to be shielded, because the electric fields can easily pass through metal material. Since a CPT system usually requires high voltages on the plates to transfer power, either the four-plate parallel or four-plate stacked capacitive coupler cannot reduce the electric field Energies 2017, 10, 1752 22 of 30

Energies 2017, 10, 1752 22 of 30 emissions. For example, in a 2.4 kW CPT system with a four-plate parallel capacitive coupler, the safe rangeparallel is capacitive about 900 mmcoupler, away the from safe the range coupler, is about which 900 is mm much away larger from than the the coupler, IPT system. which Although is much thelarger aforementioned than the IPT system. six-plate Although coupler can the reduce aforementi electriconed ﬁeld six-plate emissions, coupler its application can reduce area electric is limited field byemissions, the complicated its application structure. area Therefore, is limited in by future the complicated research, more structure. effort should Therefore, be spent in future on the research, study of electricmore effort ﬁeld should shielding. be spent on the study of electric field shielding.

5.3. Combination of IPT and CPT Technologies The previous analysis shows that the IPT system has higher efﬁciencyefficiency and the CPT system has lower cost, it is meaningfulmeaningful to combine these two technologies together, which can promote their applications [[102].102]. The structure of an IPT-CPTIPT-CPT combinedcombined systemsystem isis shownshown inin FigureFigure 2828..

Figure 28. Circuit topology of an IPT-CPT combined system.

In Figure 28, the primary and secondary sides are inductively coupled by L1 and L2, and the In Figure 28, the primary and secondary sides are inductively coupled by L1 and L2, and the coupling coefficient is defined as kI. In addition, they are capacitively coupled by CM1 and CM2. This coupling coefficient is defined as k . In addition, they are capacitively coupled by C and C . circuit topology is a combinationI of the LCC compensated IPT system [99] andM 1the LCLCM2 This circuit topology is a combination of the LCC compensated IPT system [99] and the LCLC compensated CPT system [68]. The circuit parameters can be designed to maintain comparable compensated CPT system [68]. The circuit parameters can be designed to maintain comparable contribution from both the IPT and CPT parts. Since the IPT technology has been developed for a contribution from both the IPT and CPT parts. Since the IPT technology has been developed for a long long time and is becoming available in the market, the combination can help to promote the real time and is becoming available in the market, the combination can help to promote the real application application of the CPT technology. For example, after combination, the system DC-DC efficiency is of the CPT technology. For example, after combination, the system DC-DC efficiency is improved from improved from 93.04% to 94.45%, and the system power is increased from 0.86 kW to 2.84 kW [103]. 93.04% to 94.45%, and the system power is increased from 0.86 kW to 2.84 kW [103]. To improve the system power density, the inductive and capacitive coupler can be further To improve the system power density, the inductive and capacitive coupler can be further integrated together [104], resulting in a single coupler that can generate both magnetic and electric integrated together [104], resulting in a single coupler that can generate both magnetic and electric fields [105]. In addition, the integrated coupler can be used to transfer power, which realizes fields [105]. In addition, the integrated coupler can be used to transfer power, which realizes simultaneous power and data transmission [106]. Besides, the inductive and capacitive hybrid simultaneous power and data transmission [106]. Besides, the inductive and capacitive hybrid system system can also be used in other applications, such as remote sensing [107], electric vehicle charging can also be used in other applications, such as remote sensing [107], electric vehicle charging [108], [108], DC-DC converter [109], and electric motor diagnosing [110]. DC-DC converter [109], and electric motor diagnosing [110].

6. Discussion: Practical Issues Although the CPT technology has been rapidly deve developedloped in recent years, there are still some important issues need to be further studied before it can be widely applied in real life, such as the safety concerns and dynamicdynamic charging applications.applications.

6.1. Safety Concerns Safety is always an important concern in the WPT technology; it exists in both IPT and CPT systems, andand isis becoming becoming a majora major constraint constraint in the in practicalthe practical applications. applications. Although Although it has beenit has proven been thatproven a CPT that system a CPT cansystem realize can effectiverealize effective power transfer power transfer through through a considerable a considerable distance, distance, this process this mustprocess be safemust enough be safe to enough be used to for be the used general for public.the general In CPT public. system, In thereCPT aresystem, mainly there three are important mainly safetythree important concerns: highsafety voltage concerns: stress high on voltage the plates, stress strong on the electric plates, ﬁelds strong emissions electric to fields the surroundings, emissions to andthe surroundings, foreign object inﬂuence.and foreign object influence. First, the calculation in Section 2 has shown that the plate voltage can increase to kV level to achieve high power transfer, which can be a potential danger to the users. Therefore, reliable insulation has to be applied on the surface of the capacitive coupler to prevent electric shock. This is the primary safety concern in the high power CPT applications.

Energies 2017, 10, 1752 23 of 30

First, the calculation in Section2 has shown that the plate voltage can increase to kV level to achieve high power transfer, which can be a potential danger to the users. Therefore, reliable insulation hasEnergies to be2017 applied, 10, 1752 on the surface of the capacitive coupler to prevent electric shock. This is the primary23 of 30 safety concern in the high power CPT applications. Second, thethe electricelectric fieldfield emissions emissions should should be be limited limited below below the the safe safe limitation limitation for bothfor both high high and lowand powerlow power CPT CPT systems. systems. For example,For example, when when the CPTthe CPT technology technology is used is used in biomedical in biomedical devices, devices, the effectthe effect of electric of electric fields fields on the on humanthe human body, body, such su asch the as heart the heart and brain,and brain, should should be considered be considered [111]. In[111]. addition, In addition, in the in high the powerhigh power application application of electric of electric vehicle vehicle charging, charging, the systemthe system power power is in is kW in range,kW range, the electricthe electric field field emission emission to the to driver the driv ander passengersand passengers should should be considered be considered and limited and limited [112]. For[112]. example, For example, the aforementioned the aforemention six-plateed six-plate coupler coupler structure structure [90] is[90] an is effective an effective way way to reduce to reduce the emissionsthe emissions as shown as shown in Figure in Figure 29. 29.

Figure 29. ElectricElectric field ﬁeld emissions of a 1.97 kW CPT system with six-plate capacitive coupler.

According the IEEE C95.1 standard, the electric field strength should be lower than 614 V/m at According the IEEE C95.1 standard, the electric field strength should be lower than 614 V/m at 1 MHz for human safety consideration [113]. It shows that although the electric field strength can be 1 MHz for human safety consideration [113]. It shows that although the electric field strength can be as high as 170 kV/m in a 1.97 kW system, the leakage electric fields outside the capacitive coupler is as high as 170 kV/m in a 1.97 kW system, the leakage electric fields outside the capacitive coupler is significantly reduced by the shielding plates P5 and P6, and the safe range is about 100 mm away significantly reduced by the shielding plates P and P , and the safe range is about 100 mm away from from the coupler. Besides, another effective method5 6 to reduce the electric field emissions is to use the coupler. Besides, another effective method to reduce the electric field emissions is to use active active plates adjacent to the power transfer plates with opposite voltage potential. In this way, the plates adjacent to the power transfer plates with opposite voltage potential. In this way, the leakage leakage electric fields to the surroundings can be limited [114]. electric fields to the surroundings can be limited [114]. Third, foreign objects, including conductive and dielectric materials, can have significant Third, foreign objects, including conductive and dielectric materials, can have significant influence influence to the operating of a CPT system. In practical applications, the foreign object can change to the operating of a CPT system. In practical applications, the foreign object can change the coupling the coupling capacitances between the plates and affects the resonances in the circuit. The capacitances between the plates and affects the resonances in the circuit. The transferred power can transferred power can therefore be affected. Specifically, if a living object, such as an animal, gets therefore be affected. Specifically, if a living object, such as an animal, gets access to the capacitive access to the capacitive coupler or run underneath an electric vehicle, the CPT system should have coupler or run underneath an electric vehicle, the CPT system should have the ability to shut down the ability to shut down the power transfer process and protect the living object. These aspects the power transfer process and protect the living object. These aspects should be solved before the should be solved before the CPT technology can enter the real life. CPT technology can enter the real life. 6.2. Dynamic CPT Application 6.2. Dynamic CPT Application Most of the recent CPT systems focus on the stationary charging scenarios. An electric vehicle Most of the recent CPT systems focus on the stationary charging scenarios. An electric vehicle should be parked over a charging pad, and the receiver should be well-aligned with the transmitter should be parked over a charging pad, and the receiver should be well-aligned with the transmitter to to maximize the system power and efficiency. If there is a large misalignment between the receiver maximize the system power and efficiency. If there is a large misalignment between the receiver and and transmitter, the system performance can be significantly affected. However, in the future transmitter, the system performance can be significantly affected. However, in the future application, application, it is common to have relative movement between the transmitter and receiver. For it is common to have relative movement between the transmitter and receiver. For example, in the example, in the vehicle charging applications, the transmitter is embedded on the road side and the vehicle charging applications, the transmitter is embedded on the road side and the receiver is installed receiver is installed on the vehicle side, or even the vehicle chassis can be used as part of the on the vehicle side, or even the vehicle chassis can be used as part of the receiver. The structure of a receiver. The structure of a dynamic CPT system can be shown in Figure 30. dynamic CPT system can be shown in Figure 30.

Energies 2017, 10, 1752 24 of 30 Energies 2017, 10, 1752 24 of 30

Figure 30. Structure of a dynamic CPT system.

In Figure 30, a long track of metal plates is used as the power transmitter, and the receivers are In Figure 30, a long track of metal plates is used as the power transmitter, and the receivers are made by relatively small plates [83,115]. In the practical applications, there are several challenges made by relatively small plates [83,115]. In the practical applications, there are several challenges that that need to be considered: high power implementation, dynamic transition process, and the need to be considered: high power implementation, dynamic transition process, and the stand-by stand-by power loss. For example, when multiple receivers are powered simultaneously, the power power loss. For example, when multiple receivers are powered simultaneously, the power requirement requirement from the primary side will increase. Considering the switching frequency of a CPT from the primary side will increase. Considering the switching frequency of a CPT system is usually system is usually in MHz range, it is challenging to realize a high power and high frequency in MHz range, it is challenging to realize a high power and high frequency converter even using wide converter even using wide bandgap devices. bandgap devices. In addition, when the receiver switches from one transmitter to the other, the dynamic In addition, when the receiver switches from one transmitter to the other, the dynamic transition transition process can affect both the primary and secondary sides. Considering the moving speed process can affect both the primary and secondary sides. Considering the moving speed of the of the receiver, the system dynamic should be stabilized before the receiver moves away from the receiver, the system dynamic should be stabilized before the receiver moves away from the transmitter. transmitter. Otherwise, effective power transfer cannot be realized and the oscillation can damage Otherwise, effective power transfer cannot be realized and the oscillation can damage the circuit. These the circuit. These will need to model the dynamic performance of the resonant circuit accurately will need to model the dynamic performance of the resonant circuit accurately and acquire the rapid and acquire the rapid response solution. response solution. Moreover, when the transmitter is becoming longer, the power pulsation at the receiver side Moreover, when the transmitter is becoming longer, the power pulsation at the receiver side can can be reduced. However, the stand-by losses at the transmitter side could be more significant. be reduced. However, the stand-by losses at the transmitter side could be more significant. Besides, Besides, the self-inductance of the transmitter can affect the resonance and power transfer process the self-inductance of the transmitter can affect the resonance and power transfer process in a CPT in a CPT system. Therefore, there is a trade-off between the transmitter length and the system system. Therefore, there is a trade-off between the transmitter length and the system performance, performance, which needs to be considered in practical implementations. which needs to be considered in practical implementations. 7. Conclusions 7. Conclusions Research on CPT technology is becoming popular in recent years. This paper reviews and Research on CPT technology is becoming popular in recent years. This paper reviews and summarizes the latest progresses and developments in the CPT technology, indicating that it has summarizes the latest progresses and developments in the CPT technology, indicating that it has been been well prepared for practical applications in the near future. Specifically, the analysis of its well prepared for practical applications in the near future. Specifically, the analysis of its fundamental fundamental working principle presents that the power of a CPT system is proportional to the working principle presents that the power of a CPT system is proportional to the switching frequency, switching frequency, the mutual inductance, and voltages between the metal plates. the mutual inductance, and voltages between the metal plates. Therefore, the compensation circuit design aims to increase the voltages on plates through Therefore, the compensation circuit design aims to increase the voltages on plates through resonances, resulting in series L, LC, and LCLC compensation topologies, while the capacitive resonances, resulting in series L, LC, and LCLC compensation topologies, while the capacitive coupler coupler design aims to increase the equivalent mutual capacitance, resulting in two-plate, four-plate design aims to increase the equivalent mutual capacitance, resulting in two-plate, four-plate parallel, parallel, four-plate stacked, six-plate, and repeater coupler structures. The CPT system is also four-plate stacked, six-plate, and repeater coupler structures. The CPT system is also compared with compared with the IPT counterpart, showing that it has the merit of lower cost and weight, but the the IPT counterpart, showing that it has the merit of lower cost and weight, but the power density of power density of the recent CPT system is still not comparable to the IPT system. the recent CPT system is still not comparable to the IPT system. Finally, some practical issues in CPT system implementation are also discussed. The safety Finally, some practical issues in CPT system implementation are also discussed. The safety analysis presents the methods to avoid potential dangerous hazard in a CPT system, including the analysis presents the methods to avoid potential dangerous hazard in a CPT system, including the high voltage stress, electric field emissions, and foreign object detection. Moreover, the discussion high voltage stress, electric field emissions, and foreign object detection. Moreover, the discussion on the dynamic CPT system provides a new application area of the CPT technology, which is also on the dynamic CPT system provides a new application area of the CPT technology, which is also an an important development direction in future research. important development direction in future research.

Acknowledgments: This work work is supported by the U.S. DepartmentDepartment of Energy (DOE) Graduate Automotive Technology EducationEducation (GATE) (GATE) Center Center for Electricfor Electric Drive Dr Transportation,ive Transportation, DOEU.S.-China DOE U.S.-China Clean Energy Clean Research Energy ResearchCenter (CERC), Center and (CERC), Huawei and Technology. Huawei Technology. The authors The would authors like towould thank like them to forthank their them ﬁnancial for their assistantship. financial assistantship.The authors would The authors also like would to thank also all like the to reviewers thank all for the their reviewers kindly corrections for their kindly and helpful corrections suggestions. and helpful suggestions.

Energies 2017, 10, 1752 25 of 30

Author Contributions: Chris Mi proposed the main idea of this paper; Fei Lu contribute the compensation circuit analysis part and wrote the paper; Hua Zhang contribute to the capacitive coupler analysis part. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Kazmierkowski, M.; Moradewicz, A. Unplugged but connected: Review of contactless energy transfer systems. IEEE Ind. Electron. Mag. 2012, 6, 47–55. [CrossRef] 2. Popovic, Z. Cut the cord: Low-power far-field wireless powering. IEEE Microw. Mag. 2013, 14, 55–62. [CrossRef] 3. Hui, S.; Lee, C. A critical review of recent progress in mid-range wireless power transfer. IEEE Trans. Power Electron. 2014, 9, 4500–4511. [CrossRef] 4. Jawad, A.M.; Nordin, R.; Gharghan, S.K.; Jawad, H.M.; Ismail, M. Opportunities and challenges for near-field wireless power transfer: A review. Energies 2017, 10, 1022. [CrossRef] 5. Ottman, G.K.; Hofmann, H.; Bhatt, A.; Lesieutre, G. Adaptive piezoelectric energy harvesting circuit for wireless remote power supply. IEEE Trans. Power Electron. 2002, 17, 669–676. [CrossRef] 6. Zaid, T.; Saat, S.; Yusop, Y.; Jamal, N. Contactless energy transfer using acoustic approach—A Review. In Proceedings of the Conference on Computer, Communications, Control Tech, Langkawi, Malaysia, 2–4 September 2014; pp. 376–381. 7. Leung, H.; Hu, A.P. Modeling the contact interface of ultrasonic power transfer system based on mechanical and electrical equivalence. IEEE J. Emerg. Sel. Top. Power Electron. 2017, in press. [CrossRef] 8. Jarvis, S.; Mukherjee, J.; Perren, M.; Sweeney, S. Development and characterization of laser power converters for optical power transfer applications. IET Optoelectron. 2014, 8, 64–70. [CrossRef] 9. Huang, K.; Zhou, X. Cutting the last wires for mobile communications by microwave power transfer. IEEE Commun. Mag. 2015, 53, 86–93. [CrossRef] 10. Covic, G.; Boys, J. Inductive power transfer. Proc. IEEE 2014, 6, 1276–1289. [CrossRef] 11. Dai, J.; Ludois, D. A survey of wireless power transfer and a critical comparison of inductive and capacitive coupling for small gap applications. IEEE Trans. Power Electron. 2015, 30, 6017–6029. [CrossRef] 12. Bank, M. One-wire line system for transmitting energy or information. Int. J. Commun. 2012, 6, 55–63. 13. Huang, L.; Hu, A.P.; Swain, A.; Kim, S.; Ren, Y. An overview of capacitively coupled power transfer—A new contactless power transfer solution. In Proceedings of the IEEE Conference on Industrial Electronics and Applications, Melbourne, Australia, 19–21 June 2013; pp. 461–465. 14. Culurciello, E.; Andreou, A. Capacitive inter-chip data and power transfer for 3-D VLSI. IEEE Trans. Circuit Syst. II 2006, 53, 1348–1352. [CrossRef] 15. Sodagar, A.; Amiri, P. Capacitive coupling for power and data telemetry to implantable biomedical microsystems. In Proceedings of the IEEE Conference on Neural Engineering, Antalya, Turkey, 29 April– 2 May 2009; pp. 411–414. 16. Jegadeesan, R.; Agarwal, K.; Guo, Y.; Yen, S.; Thakor, N. Wireless power delivery to flexible subcutaneous implants using capacitive coupling. IEEE Trans. Microw. Theory Tech. 2017, 65, 280–292. [CrossRef] 17. Shmilovitz, D.; Abramovitz, A.; Reichman, I. Quasi-Resonant LED driver with capacitive isolation and high PF. IEEE J. Emerg. Sel. Top. Power Electron. 2015, 3, 633–641. [CrossRef] 18. Liu, C.; Hu, A.P.; Nair, N. Coupling study of a rotary capacitive power transfer system. In Proceedings of the 2009 IEEE International Conference on Industrial Technology, Gippsland, Australia, 10–13 February 2009; pp. 1–6. 19. Wang, K.; Sanders, S. Contactless USB—A capacitive power and bidirectional data transfer system. In Proceedings of the 2014 IEEE Applied Power Electronics Conference and Exposition, Fort Worth, TX, USA, 16–20 March 2014; pp. 1342–1347. 20. Mostafa, T.; Muharam, A.; Hattori, R. Wireless battery charging system for drones via capacitive power transfer. In Proceedings of the IEEE Workshop on Emerging Technologies: Wireless Power Transfer, Chongqing, China, 20–22 May 2017; pp. 1–6. 21. Hu, A.P.; Liu, C.; Li, H. A novel contactless battery charging system for soccer playing robot. In Proceedings of the International Conference on Mechatronics and Machine Vision in Practice, Auckland, New Zealand, 2–4 December 2008; pp. 646–650. Energies 2017, 10, 1752 26 of 30

22. Ludois, D.; Reed, J.; Hanson, K. Capacitive power transfer for rotor field current in synchronous machines. IEEE Trans. Power Electron. 2012, 27, 4638–4645. [CrossRef] 23. Ludois, D.; Erickson, M.; Reed, J. Aerodynamic fluid bearings for translational and rotating capacitors in noncontact capacitive power transfer systems. IEEE Trans. Ind. Appl. 2014, 50, 1025–1033. [CrossRef] 24. Ludois, D.; Reed, J. Brushless mitigation of bearing currents in electric machines via capacitively coupled shunting. IEEE Trans. Ind. Appl. 2015, 51, 3783–3790. [CrossRef] 25. Ge, B.; Ghule, A.; Ludois, D. Three-dimensional printed fluid-filled electrostatic rotating machine designed with conformal mapping methods. IEEE Trans. Ind. Appl. 2017, 53, 4348–4359. [CrossRef] 26. Ge, B.; Ludois, D. Design concepts for a fluid-filled three-phase axial-peg-style electrostatic rotating machine utilizing variable elastance. IEEE Trans. Ind. Appl. 2016, 52, 2156–2166. [CrossRef] 27. Kim, J.; Bien, F. Electric field coupling technique of wireless power transfer for electric vehicles. In Proceedings of the IEEE 2013 Tencon-Spring Conference, Sydney, Australia, 17–19 April 2013; pp. 267–271. 28. Sakai, N.; Itokazu, D.; Suzuki, Y.; Sakihara, S.; Ohira, T. One-kilowatt capacitive power transfer via wheels of a compact electric vehicle. In Proceedings of the Wireless Power Transfer Conference, Aveiro, Portugal, 5–6 May 2016; pp. 1–3. 29. Vu, V.; Kamal, L.; Tay, J.; Pickert, V.; Dahidah, M.; Logenthiran, T.; Phan, V. A multi-output capacitive charger for electric vehicles. In Proceedings of the 2017 IEEE 26th International Symposium on Industrial Electronics (ISIE), Edinburgh, UK, 19–21 June 2017; pp. 565–569. 30. Rim, C.; Mi, C. Wireless Power Transfer for Electric Vehicles and Mobile Devices; Wiley-IEEE Press: Hoboken, NJ, USA, 2017. 31. Zhang, H.; Lu, F.; Hofmann, H.; Liu, W.; Mi, C. A four-plate compact capacitive coupler design and LCL-compensated topology for capacitive power transfer in electric vehicle charging application. IEEE Trans. Power Electron. 2016, 31, 8541–8551. 32. Lu, F. High Power Capacitive Power Transfer for Electric Vehicle Charging Applications; University of Michigan: Ann Arbor, MI, USA, 2017. 33. Li, S.; Mi, C. Wireless power transfer for electric vehicle applications. IEEE J. Emerg. Sel. Top. Power Electron. 2015, 3, 4–17. 34. Li, S.; Zhao, H.; Zhu, L.; Shuai, C.; Chen, Z. Wireless power transfer by electric field resonance and its application in dynamic charging. IEEE Trans. Ind. Electron. 2016, 63, 6602–6612. [CrossRef] 35. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. A double-sided LC compensation circuit for loosely-coupled capacitive power transfer. IEEE Trans. Power Electron. 2017, in press. [CrossRef] 36. Wang, C.; Covic, G.; Stielau, O. Power transfer capability and bifurcation phenomena of loosely coupled inductive power transfer systems. IEEE Trans. Power Electron. 2004, 51, 148–157. [CrossRef] 37. Dai, J.; Hagen, S.; Ludois, D.; Brown, I. Synchronous generator brushless field excitation and voltage regulation via capacitive coupling through journal bearing. IEEE Trans. Ind. Appl. 2017, 4, 3317–3326. [CrossRef] 38. Dai, J.; Ludois, D. Capacitive power transfer through a conformal bumper for electric vehicle charging. IEEE J. Emerg. Sel. Top. Power Electron. 2016, 4, 1015–1025. [CrossRef] 39. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. Output power and efficiency sensitivity to circuit parameter variations in double-sided LCC-compensated wireless power transfer system. In Proceedings of the IEEE Applied Power Electronics Conference and Exposition, Charlotte, NC, USA, 15–19 March 2015; pp. 597–601. 40. Ge, B.; Ludois, D.; Perez, R. The use of dielectric coatings in capacitive power transfer systems. In Proceedings of the IEEE 2014 Energy Conversion Congress and Exposition, Pittsburgh, PA, USA, 14–18 September 2014; pp. 2193–2199. 41. Lee, I.; Kim, J.; Lee, W. A high-efficient low-cost converter for capacitive wireless power transfer systems. Energies 2017, 10, 1473. [CrossRef] 42. Dai, J.; Ludois, D. Single active switch power electronics for kilowatt scale capacitive power transfer. IEEE J. Emerg. Sel. Top. Power Electron. 2016, 3, 315–323. 43. Huang, L.; Hu, A.P.; Swain, A.; Dai, X. Comparison of two high frequency converters for capacitive power transfer. In Proceedings of the IEEE 2014 Energy Conversion Congress and Exposition, Pittsburgh, PA, USA, 14–18 September 2014; pp. 5437–5443. 44. Choi, J.; Tsukiyama, D.; Tsuruda, Y.; Rivas, J. High-frequency, high-power resonant inverter with eGaN FET for wireless power transfer. IEEE Trans. Power Electron. 2017, in press. [CrossRef] Energies 2017, 10, 1752 27 of 30

45. Yusop, Y.; Husin, H.; Saat, S.; Nguang, S.; Ghani, Z. Class-E LCCL for capacitive power transfer system. In Proceedings of the IEEE 2016 Power and Energy Conference, Melaka, Malaysia, 28–29 November 2016; pp. 428–433. 46. Huang, L.; Hu, A.P.; Swain, A. A resonant compensation method for improving the performance of capacitively coupled power transfer system. In Proceedings of the IEEE 2014 Energy Conversion Congress and Exposition, Pittsburgh, PA, USA, 14–18 September 2014; pp. 870–875. 47. Choi, B.; Nguyen, D.; Yoo, S.; Kim, J.; Rim, C.T. A novel source-side monitored capacitive power transfer system for contactless mobile charger using class-E converter. In Proceedings of the 2014 IEEE 79th Vehicular Technology Conference, Seoul, Korea, 18–21 May 2014; pp. 1–5. 48. Narayanamoorthi, R.; Juliet, A.V.; Chokkalingam, B.; Padmanaban, S.; Leonowicz, Z.M. Class E power amplifier design and optimization for the capacitive coupled wireless power transfer system in biomedical implants. Energies 2017, 10, 1409. 49. Yusop, Y.; Saat, S.; Ghani, Z.; Husin, H.; Nguang, S. Capacitive power transfer with impedance matching network. In Proceedings of the 2016 IEEE 12th International Colloquium on Signal Processing & Its Applications, Malacca City, Malaysia, 4–6 March 2016; pp. 124–129. 50. Minnaert, B.; Stevens, N. Conjugate image theory applied on capacitive wireless power transfer. Energies 2017, 10, 46. [CrossRef] 51. Kline, M.; Izyumin, I.; Boser, B.; Sanders, S. Capacitive Power transfer for contactless charging. In Proceedings of the 2011 IEEE Applied Power Electronics Conference and Exposition, Fort Worth, TX, USA, 6–11 March 2011; pp. 1398–1404. 52. Antivachis, M.; Kasper, M.; Bortis, D.; Kolar, J. Analysis of capacitive power transfer GaN ISOP multi-cell DC/DC converter systems for single-phase telecom power supply modules. In Proceedings of the IECON 2016—42nd Annual Conference of the IEEE Industrial Electronics Society, Florence, Italy, 23–26 October 2016; pp. 1280–1287. 53. Mishra, S.; Adda, R.; Sekhar, S.; Joshi, A.; Rathore, A. Power transfer using portable surfaces in capacitively coupled power transfer technology. IET Power Electron. 2016, 9, 997–1008. [CrossRef] 54. Rozario, D.; Pathipati, V.; Ram, A.; Azeez, N.; Williamson, S. Modified resonant converters for contactless capacitive power transfer system used in EV charging applications. In Proceedings of the IECON 2016— 42nd Annual Conference of the IEEE Industrial Electronics Society, Florence, Italy, 23–26 October 2016; pp. 4510–4517. 55. Choi, S.; Choi, H. Capacitive wireless power transfer system with double matching transformers for reduced stress and extended ZVS range. In Proceedings of the IECON 2016—42nd Annual Conference of the IEEE Industrial Electronics Society, Florence, Italy, 23–26 October 2016; pp. 1–6. 56. Zhu, J.; Xu, M.; Sun, J.; Wang, C. Novel capacitor-isolated power converter. In Proceedings of the IEEE 2014 Energy Conversion Congress and Exposition, Pittsburgh, PA, USA, 14–18 September 2014; pp. 1824–1829. 57. Liu, C.; Hu, A.P.; Covic, G.; Nair, N. Comparative study of CCPT systems with two different inductor tuning positions. IEEE Trans. Power Electron. 2012, 27, 294–306. 58. Dai, J.; Hagen, S.; Ludois, D. Linear motion system cable elimination via multiphase capacitive power transfer through sliding journal bearing. In Proceedings of the IEEE 2017 Applied Power Electronics Conference, Tampa, FL, USA, 26–30 March 2017; pp. 2157–2164. 59. Kobayashi, H.; Funato, H.; Chiku, Y. Enhancement of transfer power of capacitive power transfer system using cascaded one pulse switching active capacitor with three-level operation. In Proceedings of the 7th International Power Electronics and Motion Control Conference, Harbin, China, 2–5 June 2012; pp. 884–888. 60. Funato, H.; Kobayashi, H.; Kitabayshi, T. Analysis of transfer power of capacitive power transfer system. In Proceedings of the IEEE 10th International Conference on Power Electronics and Drive Systems, Kitakyushu, Japan, 22–25 April 2013; pp. 1015–1020. 61. Amano, K.; Funato, H.; Haruna, J. Proposal of isolated outlet socket and plug using capacitive power transfer. In Proceedings of the Intern. Future Energy Electronics Conference and ECCE Asia, Kaohsiung, Taiwan, 3–7 June 2017; pp. 648–653. 62. Jegadeesan, R.; Guo, Y.; Je, M. Electric near-field coupling for wireless power transfer in biomedical applications. In Proceedings of the 2013 IEEE MTT-S International Microwave Workshop Series on RF and Wireless Technologies for Biomedical and Healthcare Applications, Singapore, 9–11 December 2013; pp. 1–3. Energies 2017, 10, 1752 28 of 30

63. Kusunoki, M.; Obara, D.; Masuda, M. Wireless power transfer via electric field resonance coupling. In Proceedings of the IEEE Asia-Pacific Microwave Conference (APMC), Sendai, Japan, 4–7 November 2014; pp. 1360–1362. 64. Fernandes, R.D.; Matos, J.N.; Carvalho, N.B. Wireless power transmission based on resonant electrical coupling. In Proceedings of the IEEE European Microwave Conference (EuMC), Rome, Italy, 6–9 October 2014; pp. 17–20. 65. Regensburger, B.; Kumar, A.; Sinha, S.; Doubleday, K.; Pervaiz, S.; Popovic, Z.; Afridi, K. High performance large air-gap capacitive wireless power transfer system for electric vehicle charging. In Proceedings of the IEEE 2017 Transportation Electrification Conference and Expo, Chicago, IL, USA, 22–24 June 2017; pp. 638–643. 66. Zhang, H.; Lu, F.; Hofmann, H.; Mi, C. A loosely coupled capacitive power transfer system with LC compensation circuit topology. In Proceedings of the IEEE Energy Conversation Congress Exposition (ECCE), Milwaukee, WI, USA, 18–22 September 2016; pp. 1–5. 67. Theodoridis, M.P. Effective capacitive power transfer. IEEE Trans. Power Electron. 2012, 27, 4906–4913. [CrossRef] 68. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. A double-sided LCLC-compensated capacitive power transfer system for electric vehicle charging. IEEE Trans. Power Electron. 2015, 30, 6011–6014. [CrossRef] 69. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. A CLLC-compensated high power and large air-gap capacitive power transfer system for electric vehicle charging applications. In Proceedings of the IEEE Applied Power Electronics Conference and Exposition, Long Beach, CA, USA, 20–24 March 2016; pp. 1721–1725. 70. Doubleday, K.; Kumar, A.; Sinha, S.; Regensburger, B.; Pervaiz, S.; Afridi, K. Design tradeoff in a multi-modular capacitive wireless power transfer system. In Proceedings of the IEEE 2016 PELS Workshop on Emerging Technologies: Wireless Power Transfer, Knoxville, TN, USA, 4–6 October 2016; pp. 35–41. 71. Sinha, S.; Kumar, A.; Pervaiz, S.; Regensburgr, B.; Afridi, K. Design of efficient matching networks for capacitive wireless power transfer systems. In Proceedings of the IEEE 2016 Workshop on Control and Modeling for Power Electronics, Trondheim, Norway, 27–30 June 2016; pp. 1–7. 72. Su, Y.; Zhou, W.; Hu, A.P.; Tang, C.; Xie, S.; Sun, Y. Full-duplex communication on the shared channel of a capacitively coupled power transfer system. IEEE Trans. Power Electron. 2017, 32, 3229–3239. [CrossRef] 73. Rozadio, D.; Azeez, N.; Williamson, S. Comprehensive review and comparative analysis of compensation networks for capacitive power transfer systems. In Proceedings of the IEEE 2016 International Symposium on Industrial Electronics, Santa Clara, CA, USA, 8–10 June 2016; pp. 823–829. 74. Sepahvand, A.; Kumar, A.; Afridi, K.; Maksimovic, D. High power transfer density and high efficiency 100 MHz capacitive wireless power transfer system. In Proceedings of the IEEE Workshop on Control and Modeling for Power Electronics, Vancouver, BC, Canada, 12–15 July 2015; pp. 1–4. 75. Kumar, A.; Pervaiz, S.; Chang, C.; Korhummel, S.; Popovic, Z.; Afridi, K. Investigation of power transfer density enhancement in large air-gap capacitive wireless power transfer systems. In Proceedings of the IEEE Wireless Power Transfer Conference, Boulder, CO, USA, 13–15 May 2015; pp. 1–4. 76. Minnaert, B.; Stevens, N. Optimal analytical solution for a capacitive wireless power transfer system with one transmitter and two receivers. Energies 2017, 10, 1444. [CrossRef] 77. Yi, K. High frequency capacitive coupling wireless power transfer using glass dielectric layers. In Proceedings of the IEEE Wireless Power Transfer Conference, Aveiro, Portugal, 5–6 May 2016; pp. 1–3. 78. Kumar, A.; Sinha, S.; Sepahvand, A.; Afridi, K. Improved design optimization for high-efficiency matching networks. IEEE Trans. Power Electron. 2017, in press. [CrossRef] 79. Huang, L.; Hu, A.P.; Swain, A.; Su, Y. Z-impedance compensation for wireless power transfer based on electric field. IEEE Trans. Power Electron. 2016, 31, 7556–7563. [CrossRef] 80. Wilkinson, A.J.; Everard, J.K.A. Transmission line load network topology for class E power amplifier. IEEE Trans. Microw. Theory Tech. 2001, 49, 1202–1210. [CrossRef] 81. Lu, F.; Zhang, H.; Mi, C. A two-plate capacitive wireless power transfer system for electric vehicle charging applications. IEEE Trans. Power Electron. 2017, in press. [CrossRef] 82. Zou, L.; Hu, A.P.; Su, Y. A single-wire capacitive power transfer system with large coupling alignment tolerance. In Proceedings of the IEEE 2017 PELS Workshop on Emerging Technologies: Wireless Power Transfer, Chongqing, China, 20–22 May 2017; pp. 93–98. Energies 2017, 10, 1752 29 of 30

83. Van Neste, C.; Phani, A.; Hull, R.; Hawk, J.; Thundat, T. Quasi-wireless capacitive energy transfer for the dynamic charging of personal mobility vehicles. In Proceedings of the IEEE 2016 PELS Workshop on Emerging Technologies: Wireless Power Transfer, Knoxville, TN, USA, 4–6 October 2016; pp. 196–199. 84. Van Neste, C.; Phani, A.; Larocque, A.; Hawk, J.; Kalra, R.; Banaag, M.; Wu, M.; Thundat, T. Quarter wavelength resonators for use in wireless capacitive power transfer. In Proceedings of the IEEE 2017 PELS Workshop on Emerging Technologies: Wireless Power Transfer, Chongqing, China, 20–22 May 2017; pp. 229–234. 85. Rozario, D.; Azeez, N.; Williamson, S. Analysis and design of coupling capacitors for contactless capacitive power transfer systems. In Proceedings of the IEEE 2016 Transportation Electrification Conference and Expo, Dearborn, MI, USA, 27–29 June 2016; pp. 1–7. 86. Liu, C.; Hu, A.P.; Budhia, M. A generalized coupling model for capacitive power transfer systems. In Proceedings of the 36th Annual Conference on IEEE Industrial Electronics Society, Glendale, AZ, USA, 7–10 November 2010; pp. 271–279. 87. Huang, L.; Hu, A.P. Defining the mutual coupling of capacitive power transfer for wireless power transfer. Electron. Lett. 2015, 51, 1806–1807. [CrossRef] 88. Komaru, T.; Akita, H. Positional characteristics of capacitive power transfer as a resonance coupling system. In Proceedings of the IEEE Wireless Power Transfer Conference, Perugia, Italy, 15–16 May 2013; pp. 218–221. 89. Liu, Y.; Li, C.; Chi, Y.; Lee, Y.; Yuan, Q.; Chen, Q. Novel design of pixelated charging electrodes for capacitive power transfer. In Proceedings of the IEEE Wireless Power Transfer Conference, Taipei, Taiwan, 10–12 May 2017; pp. 1–4. 90. Zhang, H.; Lu, F.; Hofmann, H.; Liu, W.; Mi, C. A six-plate capacitive coupler to reduce electric field emission in large air-gap capacitive power transfer. IEEE Trans. Power Electron. 2017, in press. [CrossRef] 91. Doubleday, K.; Kumar, A.; Regensburger, B.; Pervaiz, S.; Sinha, S.; Popovic, Z.; Afridi, K. Multi-objective optimization of capacitive wireless power transfer systems for electric vehicle charging. In Proceedings of the IEEE 2017 Workshop on Control and Modeling for Power Electronics, Stanford, CA, USA, 9–12 July 2017; pp. 1–8. 92. Zhang, H.; Lu, F.; Hofmann, H.; Liu, W.; Mi, C. An LC compensated electric field repeater for long distance capacitive power transfer. IEEE Trans. Ind. Appl. 2017, in press. [CrossRef] 93. Miyazaki, M.; Abe, S.; Suzuki, Y.; Sakai, N.; Ohira, T.; Sugino, M. Sandwiched parallel plate capacitive coupler for wireless power transfer tolerant of electrode displacement. In Proceedings of the IEEE 2017 Microwaves for Intelligent Mobility Conference, Nagoya, Japan, 19–21 March 2017; pp. 29–32. 94. Liu, C.; Hu, A.P.; Wang, B.; Nair, N. A capacitively coupled contactless matrix charging platform with soft switched transformer control. IEEE Trans. Ind. Electron. 2013, 60, 249–260. [CrossRef] 95. Su, Y.; Xie, S.; Hu, A.P.; Tang, C.; Zhou, W.; Huang, L. Capacitive power transfer system with a mixed-resonant topology for constant-current multiple-pickup applications. IEEE Trans. Power Electron. 2017, 32, 8778–8786. [CrossRef] 96. Zargham, M.; Gulak, P. Maximum achievable efficiency in near-field coupled power transfer systems. IEEE Trans. Biomed. Circuits Syst. 2012, 6, 228–245. [CrossRef][PubMed] 97. Huang, Z.; Wong, S.; Tse, C.K. Design methodology of a series-series inductive power transfer system for electric vehicle battery charger application. In Proceedings of the IEEE Energy Conversion Congress and Exposition (ECCE), Pittsburgh, PA, USA, 14–18 September 2014; pp. 1778–1782. 98. Deng, J.; Lu, F.; Li, S.; Nguyen, T.; Mi, C. Development of a high efficiency primary side controlled 7 kW wireless power charger. In Proceedings of the 2014 IEEE International Electric Vehicle Conference, Florence, Italy, 17 December 2014; pp. 1–6. 99. Li, S.; Li, W.; Deng, J.; Nguyen, T.D.; Mi, C. A double sided LCC compensation network and its tuning method for wireless power transfer. IEEE Trans. Veh. Technol. 2015, 64, 2261–2273. [CrossRef] 100. Lu, F.; Zhang, H.; Kan, T.; Hofmann, H.; Mei, Y.; Cai, L.; Mi, C. A high efficiency and compact inductive power transfer system compatible with both 3.3 kW and 7.7 kW receivers. In Proceedings of the IEEE Applied Power Electronics Conference and Exposition, Tampa, FL, USA, 26–30 March 2017; pp. 3669–3673. 101. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. A dual-coupled LCC-Compensated IPT system with a compact magnetic coupler. IEEE Trans. Power Electron. 2017, in press. [CrossRef] Energies 2017, 10, 1752 30 of 30

102. Minnaert, B.; Stevens, N. Conjugate image theory for non-symmetric inductive, capacitive and mixed coupling. In Proceedings of the IEEE Wireless Power Transfer Conference, Taipei, Taiwan, 10–12 May 2017; pp. 1–4. 103. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. An Inductive and capacitive combined wireless power transfer system with LC-compensated topology. IEEE Trans. Power Electron. 2016, 31, 8471–8482. [CrossRef] 104. Lu, F.; Zhang, H.; Hofmann, H.; Mi, C. An inductive and capacitive integrated coupler and its LCL compensation design for wireless power transfer. IEEE Trans. Ind. Appl. 2017, in press. [CrossRef] 105. Schroedermeier, A.; Ludois, D. Integrated inductor and capacitor with co-located electric and magnetic fields. IEEE Trans. Ind. Appl. 2017, 53, 380–390. [CrossRef] 106. Li, X.; Tang, C.; Dai, X.; Deng, P.; Su, Y. An inductive and capacitive combined parallel transmission of power and data for wireless power transfer systems. IEEE Trans. Power Electron. 2017, in press. [CrossRef] 107. Trevisan, R.; Costanzo, A. Exploitation of capacitive coupling at UHF for remote sensing in a kW WPT system. In Proceedings of the 2015 European Microwave Conference, Paris, France, 7–10 September 2015; pp. 490–493. 108. Vincent, D.; Sang, P.; Williamson, S. Feasibility study of hybrid inductive and capacitive wireless power transfer for future transportation. In Proceedings of the IEEE 2017 Transportation Electrification Conference and Expo, Chicago, IL, USA, 22–24 June 2017; pp. 229–233. 109. Ishigaki, M.; Shin, J.; Dede, E. A novel soft switching bidirectional DC-DC converter using magnetic and capacitive hybrid power transfer. IEEE Trans. Power Electron. 2017, 32, 6961–6970. [CrossRef] 110. Behjat, V.; Bahedi, A.; Setayeshmehr, A.; Borsi, H.; Gockenbach, E. Diagnosing shorted turns on the windings of power transformers based upon online fra using capacitive and inductive couplings. IEEE Trans. Power Deliv. 2011, 26, 2123–2133. [CrossRef] 111. Erfani, R.; Marefat, F.; Sodagar, A.; Mohseni, P. Modeling and experimental validation of a capacitive link for wireless power transfer to biomedical implants. IEEE Trans. Circuits Syst. II 2017, in press. [CrossRef] 112. Bartlett, A.; Arsenault, P. Wireless power for vehicle lightweighting, reducing costs, and improving manufacturing efficiencies. In Proceedings of the IEEE Vehicular Technology Conference, Boston, MA, USA, 6–9 September 2015; pp. 1–6. 113. IEEE Standards Coordinating Committee. IEEE Standard for Safety Levels with Respect to Human Exposure to Radio Frequency Electromagnetic Fields, 3 kHz to 300 GHz; IEEE Standard C95.1-2005; IEEE Standards Coordinating Committee: Piscataway, NJ, USA, 2006. 114. Ramos, I.; Afridi, K.; Estrada, J.; Popovic, Z. Nearfield capacitive wireless power transfer array with external field cancellation. In Proceedings of the IEEE Wireless Power Transfer Conference, Aveiro, Portugal, 5–6 May 2016; pp. 1–4. 115. Lu, F.; Zhang, H.; Hofmann, H.; Mei, Y.; Mi, C. A dynamic capacitive power transfer system with reduced power pulsation. In Proceedings of the IEEE 2016 PELS Workshop on Emerging Technologies: Wireless Power Transfer, Knoxville, TN, USA, 4–6 October 2016; pp. 60–64.

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).