HU MEILIN (CN)
LIN ZHONGSHENG (CN)
| WO2021154836A1 | 2021-08-05 |
| US20210050136A1 | 2021-02-18 |
BAIMEL, D. ET AL.: "Modeling and Analysis of None-Series Compensation for Inductive Wireless Power Transfer Links", 2020 IEEE 29TH INTERNATIONAL SYMPOSIUM ON INDUSTRIAL ELECTRONICS (ISIE, 2020, pages 1623 - 1627, XP033800956, Retrieved from the Internet
| CLAIMS: 1. A wireless power transfer coupling apparatus, comprising: at least one conductive member configured as a layer of the first coupling member to provide a magnetic field for wireless power transfer, the conductive member having; a first end; and a second end opposite the first end, wherein: the conductive member extends from the first end to the second end along a lengthwise axis of the coupling apparatus and is configured to distribute current across the layer between the first and second ends. 2. The apparatus of claim 1 wherein the conductive member comprises a plurality of lengths of wire. 3. The apparatus of claim 1 wherein the conductive member comprises a sheet material. 4. The apparatus of any one of the preceding claims wherein a termination is provided at each end of the conductive member, each termination being configured for connecting the conductive member to a power supply. 5. The apparatus of claim 4 wherein the terminations extend transversely across a width of the conductive member. 6. The apparatus of any one of the preceding claims further comprising a layer of permeable material. 7. The apparatus of claim 6 wherein the magnetic field for power transfer is provided on a side of the apparatus opposite to the permeable material. 8. The apparatus as claimed in claim 6 or claim 7 wherein the permeable material extends either side of the conductive member. 9. The apparatus as claimed in claim 8 wherein the permeable material forms pole areas. 10. The apparatus as claimed in claim 8 or claim 9 wherein the permeable material which extends either side of the conductive member also extends in a direction toward a magnetic flux coupling region. 11. The apparatus of any one of the preceding claims wherein the apparatus is uncompensated. 12. The apparatus of any one of the preceding claims wherein the conductive member comprises a plurality of wires configured in parallel and spaced more than one wire diameter apart. 13. A wireless power transfer system comprising a wireless power transfer coupling apparatus according to any of the preceding claims and a secondary coupler comprising a flat coil. 14. The system of claim 13 wherein the flat coil is a vertical coil. 15. The system of claim 13 wherein the flat coil is a flat solenoid coil. 16. The system of any of claims 13 to 15 further comprising providing the secondary coupler with capacitor compensation. 17. A method of wireless power transfer between an uncompensated primary coupler and a capacitor compensated secondary comprising fully compensating the reactance of the primary coupler by a reflected impedance. 18. The method of claim 17 further comprising partially compensating the secondary reactance by the capacitor. |
Figure 37 shows a circuit diagram for switches of an H bridge converter used for example to drive or energise a field forming conductor, together with the waveforms for the current duty cycle D;
Figure 38 shows the waveforms for the switching process for Figure 37;
Figure 39 shows an example of an alternative converter switch topology;
Figure 40 shows a circuit diagram for an IPT system including a primary and secondary structure according to any of the examples discussed above, and showing the compensation topology for the second side;
Figure 41 shows a phasor diagram for the system of Figure 40 when Xs is fully compensated;
Figure 42 shows a phasor diagram for the system of Figure 40 when Xp is fully compensated;
Figure 43 shows a phasor diagram for series-series compensation for the system of Figure 40;
Figure 44 shows plots illustrating current, voltage, apparent power, and efficiency for case 1 ;
Figure 45 is a diagrammatic cross section of a system according to Design 3 illustrating measured Φ i in the receiver RX;
Figure 46(a)-46(c) show plots for CV, power loss (W) and normalized cost;
Figure 47 shows variation between CV, cost and P loss ;
Figure 48 shows a ccomparisons of calculated and simulated receiver flux with horizontal displacements.
Detailed Description The present invention is a wireless power transfer (WPT) apparatus including an inductive power transfer (IPT) apparatus useful for the wireless charging or powering of a large range of devices, including for example consumer devices, such as mobile communication devices for example, and electric vehicles, ranging for example from drones to cars or trucks. The IPT apparatus and control disclosed herein enable stable, efficient, economical, and safe dynamic charging for mobile devices including electric vehicles (EVs). Although vehicles are referred to herein by way of example it will be appreciated by those skilled in the art that this disclosure is applicable to many other WPT/IPT applications. The transmitter of the inductive power transfer apparatus of the present invention is unlike existing constructions in that it is not formed as a coil, so there are no turns in the conductor that produces the field for power transfer. It may instead be considered as a conductor configured as a layer i.e. a conductive region that has a longitudinal dimension and a transverse dimension, both of which are significantly greater than its depth or thickness dimension. Some examples of primary magnetic coupling structures which have a conductor layer configuration are shown in Figures 1 to 7. Referring to Figures 1 to 7, an IPT system is shown generally referenced 1 having a primary or transmitter apparatus comprising a magnetic structure 2 in the form of a pad and a secondary or receiver apparatus 4 which is disposed adjacent to, but spaced from, the pad 2. As shown in Figure 1, the primary structure is electrically connected to a converter 6 supplied by a voltage source 8. A primary controller 10 is configured to operate the converter to deliver an alternating current to drive or energise the structure 2. Similarly, still referring to Figure 1, secondary coupler structure 4 is electrically connected to a converter or rectifier 12 that may be controlled, if required, by controller 14 to provide an output power 16 which may be used to supply a load, for example to charge a battery. The power supplies 8, 16, controllers 10, 14 and converters/rectifiers 6, 12 such as those shown in Figure 1 are omitted for clarity in some other figures such as Figures 2-7, but it will be understood that these may be used as required with the apparatus described or illustrated herein to realise wireless power transfer components or systems. In Figures 1 to 4, the primary coupler 2 has a field forming conductor 20 which is formed as a thin layer of conductive sheet material which in this example comprises a foil, such a copper or aluminium foil that is laid over a layer of magnetically permeable material 20 such as ferrite which is provided as a ferrite plate. The conductor has a length L and a width W which is transverse to the length. Both L and W are much greater than the thickness of the conductor (i.e. the dimension of the conductor that is perpendicular to the length and width dimensions). Either end of conductor 20 has a termination 21, each of which enable a secure electrical connection to be made between the conductor 20 and the cables 23 that conduct current between the field forming conductor 20 and the converter 6. The terminations 21 allow the current to be distributed across the width of the conductor 20. The permeable layer 22 has regions 24 that are not covered by the conductor 20. Regions 24 can act as pole regions for the field produced in use by the conductor 20 to enter and exit the permeable layer 22 and thus guide a flux path that forms a loop or arch over the conductor 20 and into a power transfer region 26 on a side of the conductor 20 that is opposite to the side on which the permeable layer 22 is provided. The field shape is indicated by arrows 28 and 30 in Figure 2 when current is flowing through the conductor 20 into the page as indicated by arrows 32. The size of regions 24 and the width of conductor 20, or their relative dimensions of the conductor 20 and the pole regions 24 can be adjusted to provide a required field shape in use. For example, making the conductor 20 wider may provide a flatter or lower field extending across the width of the coupler. This may be advantageous for allowing a wider or broader power transfer region 26 for coupling with a secondary, meaning that there is a reduced requirement for precise alignment for effective power transfer to occur. In some examples regions 24 may not be provided i.e. the permeable material 22 may end at, or within, the side edges of the conductor 20. The foil conductor 20 in the example shown in Figures 3 and 4 is the same as that described above with reference to Figures 1 and 2, but the exposed regions 24 of the permeable layer in Figures 3 and 4 are raised to present one or more walls 36 and/or 38 which provide additional or alternative exposed surfaces by which the field may enter and exit the permeable material. The permeable material 22 in the Figure 3 and Figure 4 example may take to form of a “U” or “C” shape. As shown in Figure 4, there may in some examples be gaps 40 between the inner walls 38 of the permeable material 22 and the sides of the conductor 20. Gaps 40 may not be present in some examples. If gaps 40 are present, their dimensions may be adjusted to provide a required field shape. Turning now to Figures 5 to 8, another example is shown in which the conductor 20 comprises a plurality of individual lengths of conducting wire or cable 50, such as litz wire. The individual wires 50 are not shown in Figure 5 for clarity, however an example of the conductor 20 is illustrated in Figure 7 without permeable layer 22. The gaps between wires 50 are configured to provide a required field or flux pattern. In some embodiments or examples the wires 50 are spaced many wire diameters apart, and in others around one wire diameter apart, or less than one wire diameter apart from each other. As with the examples described above, the terminals 21 conduct and distribute current from wires 23 to the wires 50. The secondary coupler 4 in the examples discussed and shown above comprises a coil 5 which can be wound as a multi-turn coil of a suitable conductor such as litz wire. The coil 5 may in some examples be flat, i.e. wound as a spiral. As shown in Figures 1, 3 and 5 the coil 5 is vertically oriented, i.e. it is arranged or provided in a vertical plane that extend along the lengthwise axis of the field forming conductor 20. The field forming conductors 20 shown in the examples above, together with the permeable plate 22 provide magnetic fields in a coupling region above the primary structure and very little or no field on the underside of the structure. Therefore, the conductors 20 can produce a single-sided flux pattern without having the returning wires that are required when forming a coil. There is no coil arrangement that creates a return path as in the prior art in which components of current flow in opposite directions across the structure at the same time. Accordingly, the flat conductor region can generate wide and flat magnetic flux above its upper surface. The air-cored vertical receiving coil 5 can capture most of the produced flux by the primary, which significantly benefits dynamic wireless charging for applications such as moving EVs. The ferrite plates 22 is coupled with the transmitting coil to strengthen the coupling and reduce the flux leakage. In some examples or embodiments a shield (such as an aluminium pate) may be provided beneath the lower surface of the ferrite 22 to further assist with producing a required magnetic flux pattern, or to assist with reducing leakage fields around the sies and base of the coupler. For some applications use of copper foil for conductor 20 will be preferable as it is much more cost- efficient than Litz wire. In EV applications for example a foil transmitting coil 20 is mounted in the roadway while the receiver is mounted at the bottom of the car chassis. However, in other applications the transmitting coil may be mounted in a charging device, and the receiving coil in a consumer electronic device, such as a mobile phone. For example, the receiver coil 5 may be connected by a hinge or extension that allows the coil to be moved out from the surface of the phone such that it is vertically oriented in relation to the transmitter coil. Figures 8 and 9 show other examples or embodiments for the secondary coupler 4 in which the coil 5 has a different form, and in which permeable material 52 such as a ferrite may be used in conjunction with coil 5 to enhance the field coupled from the primary. Thus, in Figure 8, the coil 5 comprises a solenoid, and which has been wound in a flat form, provided a distance D above the primary structure. As shown, the solenoid winding 5 may be provided in a plane that is generally parallel to the flat field forming conductor 20. In the example shown the coil 5 is magnetically associated with permeable material 52 and may be wound around the permeable material. In some examples the permeable material may be flat and/or oriented parallel to the permeable material 22 or conductor 20 of the primary coupler. In some examples the permeable material 52 may not extend beyond the edges of coil 5, and in other examples such as that of Figure 8, the permeable material 52 has exposed regions 56 and 58 that may provide pole areas for entry and exit of magnetic flux and couple with the pole areas 24 of the primary coupler. In Figure 9, the secondary also has permeable material as described above, but the coil 5 is provided as a plurality of coils that are spread, spaced, or distributed across the width of the receiver structure. In another example, the coils 5 shown in Figure 9 are modular units which may be used with or without permeable material and added to, or removed from, the secondary as required for the particular application. In Figures 10 to 14, coils 5 that do not necessarily include permeable material i.e. air-cored coils which are shown in several exemplary configurations are shown. These coils may be multi-turn or possibly single turn depending on the required application. The various configurations may include more than one coil, as shown in Figures 11 -14. The coils shown in Figures 10 to 14 allow spatial coil arrangements that can be adopted to increase the misalignment tolerance while still possessing the advantage of vertical components of orientation that are efficient for flux capture. As can be seen most clearly in Figures 13 and 14, multiple coils can be arranged or configured spatially to provide greater tolerance to possible angular or translational misalignment between the secondary coil(s) and the primary conductor 20. Again, for each of the examples or embodiments in Figures 10 to 14 the transmitting conductor 20 does not have or require a return winding on the backside of the ferrite plate 22, therefore producing no back magnetomotive force (MMF). The flux patterns and performance of the coil pairs detailed above will be demonstrated below. The simulated coil pairs for comparison includes five types: the transmitter with a flat ferrite-vertical receiver as shown in Figures 1 and 2 (hereinafter “Design 1”); foil transmitter with U-shaped ferrite- vertical receiver as shown in Figures 3 and 4 (“Design 2”); flat Litz-paralleled transmitter with ferrite - vertical receiver as shown in Figures 5 to 7 (“Design 3”). Their simulation models are shown below in Figures 3 to 5. The excitation direction is also shown in these figures with a red arrow. The geometric parameters of these designs are given in Table 2. To fairly compare their performance, the transmitting coils have the same MMF of 24 At. Except for Design 2 with narrower C-shaped ferrite, the size of the other four transmitters is identical. The transmitting coil may be covered by a plastic cover with a thickness of 5 mm (but it is not at all necessary to have a cover). In the example application of EV car charging, the air gap between the ground and the car chassis is set as 0.21 m. The size of the vertical receiver is restricted by the height of the car chassis and its width is set to be 0.15 m. The skin depth of copper foil can be calculated using (1). Where μ is 4πE-7 H/m and σ is 5.8E7 S/m for copper. Therefore, the skin depth at 85 kHz (frequency used for EV wireless charging) is about 0.23 mm. The thickness of the copper foil was then chosen as 0.5 mm, and its width as 0.2 m to allow for an acceptable misalignment tolerance. The foil coil has only one layer. Therefore, the cross-section area of the foil winding is 0.1E-3 m 2 , which means its mean current density is 2.4E-5 A/m 2 (0.24A/mm 2 ). Table 2. Geometric parameters of different coils Litz wire with a diameter of 5 mm was selected to wind the Litz coils. The number of turns of the transmitting Litz coils is determined to be 40 to keep the width of foil coil and Litz coils the same. Therefore, the current in each turn is 0.6 A. Meanwhile, the turn number of the vertical receiver is chosen to be 2. The magnetic flux distribution in a central cut plane i.e. a cross section taken through the centre of the transmitter and receiver structures, the cut plane of the cross section being transverse to the lengthwise direction (longitudinal axis) of the conductor 20, is shown in Figures 15 to 17. Electric parameters of the designs when the receiver is aligned are given in Table 3 below. Figures 15 to 17 show the magnetic flux density B distribution in the central cut plane of these three designs. It can be seen in Figures 15 and 16 that the foil transmitting conductor can generate flat and stable horizontal flux patterns over the conductor surface. Therefore, the vertical receiving coil has a strong lateral misalignment tolerance. The flux patterns of Design 1 and Design 3 are quite similar. Moreover, the flux density of the foil coil with a U-shaped ferrite plate (Design 2) is slightly stronger than that with a flat ferrite plate (Design 1). Table 3. Electric parameter comparison when aligned Table 4 gives the loss in foil conductor of Design 1 and Design 2. The foil transmitting conductor with U-shaped ferrite has a lower loss, as the ferrite around the two ends of the copper foil guides flux away from the copper, reducing the loss. Fig.18 compares the loss density in the foil coil of these two designs. As can be seen, Design 2 with U-shaped ferrite has a smaller loss density. Table 4. Foil conductor loss comparison In practical applications, the geometric parameters of the proposed designs can be quickly optimized through the reluctance-based modelling. Reluctance models established for Design 1, Design 2, and Design 3 are discussed below, along with their corresponding flux patterns and flux tubes of the mutual coupling. According to the flux patterns, the reluctance of the mutual coupling flux tube is calculated for each design based on the magnetic equivalent circuit (MEC) method. For the reluctance calculations, the magnetomotive force (MMF) drop in ferrite is neglected to simplify the analysis due to the relatively large permeability of ferrite materials. The flux distribution of Design 1 in the side view is shown in Fig.19 (a). Fig.19 (b) is the approximate flux path shape. Fig.20 gives the geometry of the coupled flux tube of the vertical receiver. In Fig.20, H1 is the height of the receiver, and L R is the length of the receiver. L 1 is the length of the foil conductor’s section, while W is the side width of the foil conductor, which is the same as the width of the vertical receiver. The flux tube's inner and outer lengths are l 1 and l 2 , respectively, as shown in (2) and (3). The equivalent magnetic reluctance of the mutual coupling part can then be obtained through effective flux length and area, as shown in (4) and (5). For the reluctance calculation of the other two designs, the process is the same. The flux distribution of Design 2 in the side view is shown in Fig.21(a). Fig.21(b) is the approximate flux path shape. Fig.22 gives the geometry of the coupled flux tube of the vertical receiver. In Fig.22, L 5 is the section length of the C-shaped ferrite. The equivalent magnetic reluctance of the mutual coupling part can then be obtained through effective flux length and area of the tube, as shown from (6) to (9). The flux distribution of Design 3 in the side view is shown in Fig.23(a). Fig.23(b) is the approximate flux path shape. Fig.24 gives the geometry of the coupled flux tube of the vertical receiver. In Fig.24, L 7 is the section length of the inner coupling flux boundary, and L 8 is the section length of the outer coupling flux boundary. As can be seen from the flux distribution, the mutual coupling flux includes two parts: the partially coupling part and the fully coupling part. Partially coupling refers to only part of the primary is coupled with the secondary. Fully coupling means all the primary is coupled with the secondary. In the equations below, the subscript P refers to the partially coupling flux, and F refers to the fully coupling flux. For the partially coupling part, the effective number of primary turns N 1 ’ can be obtained by considering the relationship between L 7 and L 1 , as shown in (10). N1 is the total number of turns of the primary side. α is the ratio of the effective primary turns to the total turns. The equivalent magnetic reluctance of these two mutual coupling parts can then be obtained through effective flux length and area of each tube, as shown from (11) to (18). The coupled flux linkage in the vertical receiver will now be calculated through the MEC method according to the calculated reluctance. The calculated flux linkage will be compared with the results obtained through the finite element method (FEM). After obtaining the equivalent reluctance of each design, the coupling flux linkage can be obtained by (19). N 2 is the number of turns of the vertical receiver, which is 2. F is the MMF of the transmitter, which is 24 At. R is the calculated reluctance of the mutual coupling flux tube. For Design 3, its coupling flux linkage includes two parts: the fully coupling flux linkage and the partially coupling flux linkage. Therefore, its total coupling flux linkage can be obtained from (20). The obtained mutual flux linkage from MEC and FEM is given in Table 5 for the three designs. As can be seen, the reluctance models of Design 1 and Design 3 are reasonably accurate, while the error of the other one is lower than 10%, which is acceptable and can be further improved by refining the flux tubes. Table 5. Comparison of coupling flux linkage (Wb) To implement the designs described above in a dynamic charging environment for EVs, there are at least two possible arrangements according to the direction of movement of the EV. These arrangements are shown in Fig.25 (a) and (b). TX refers to the transmitting conductor, while RX represents the receiving coil. In Fig.25(a), the moving direction is the same as the excitation direction. In 25(b), the moving direction is orthogonal to the excitation direction, and the TX is placed along the lateral side. For the Fig.25(a) arrangement, as the current can be seen as continuous in the moving direction when the TX distance is small, it can naturally produce smooth and stable coupling between TX and RX when the EV is moving. For the Fig.25(b) arrangement, the coupling can be unstable when moving, as the flux is not always constant above one TX in this moving direction. However, the arrangement of Fig.25(b) can provide superior results. It is helpful to consider achieving constant flux by designing the TX width and distance. Design 1 with the geometric parameters in Table 2 is given as an example to study flux leveling. The calculated flux linkages between each TX coil and RX coil are shown in Fig.26, where Φ 1, Φ2, Φ3 are flux linkage produced by TX1, TX2, TX3, respectively. The gray line is the superimposed flux linkage produced by the three TX coils. It can be seen that by designing the TX coils size and distance between TX coils, a nearly leveled flux linkage profile can be achieved. This feature benefits the dynamic charging for the EVs since the induced voltage is proportional to the flux linkage, and less fluctuating flux leads to less fluctuation in induced voltage. We now take Design 1 as an example for implementation in an EV dynamic charging application with a high-power level. Since the inductances of the transmitting conductors are relatively small, only the receiving side compensation is considered. The required excitation level and produced reactive power are calculated to achieve 30 kW power charging. The equivalent circuit diagram is shown in Fig.27, where L p and L s are inductances of TX and RX coils, C s is the compensating capacitor resonating with Ls. I P and I S are the primary and secondary current. V P and V S are the primary input AC voltage and secondary output AC voltage. The battery load with an active full-bridge converter can be equivalent to an AC voltage source VS, as shown in Fig.27, where the V S can be modulated to achieve battery charging and the battery’s voltage is V DC , as expressed in (15): Where D is the duty cycle, ranging from 0 to 1, θ is the phase angle of VS relative to primary AC voltage source V P , ranging from 0 to 180 degrees. The primary current I P and secondary current I S are solved in (15). The vectors of currents and voltages are shown in Fig.28. Note that I S is the vector sum of the term1 I S1 and term2 I S2 . If |V P |=|V S |, | I S1 | is seven times larger than |I S2 | for the foil TX coil structure since the mutual inductance M is seven times larger than the TX coil’s self-inductance L P . Therefore, it is reasonable to use I S1 to estimate the magnitude of I S . The power flows out of the voltage source V P and V S can be calculated by numerical methods through (17). The real power P PS flowing from the primary side to the secondary side can be estimated by (18), since I S1 is sig nificantly larger than I S2 . When θ is 90 degrees, the reactive power of the primary side Q P is zero, and the reactive power of the secondary side Q S is shown in (19). The accurate numerical solution for the real powers, the reactive powers, and currents are shown in Fig.29 and Fig.30 for two different cases (Sim 1 and Sim 4). The simulation parameters of four cases with various operating frequencies and RX turns are listed in Table 6. Table 6. Simulation parameters (Foil TX conductor) In Fig.29 and Fig.30, the point with real power PP of 30 kW and minimum magnitude of VS is chosen as the operation point. For each simulation, the reactive power and current values at the operation point are listed in Table 7. Table 7. The operation point data (Foil TX conductor) It can be seen from Table 7, from sim 1 to sim 2, when the frequency is reduced from 85 kHz to 20 kHz, it leads to a smaller Q S , and a significantly larger I S . From sim 2 to sim 3 and sim 4, the turns of RX coil are increased from 30 turns to 70 and 100 turns at a frequency of 20 kHz. The increase of turns brings to a larger mutual inductance, which reduces the secondary current while keeping the Q S unchanging. An analytical analysis has been shown to verify the results above i.e.: (1) The reduction in frequency leads to a smaller Q S , and a significantly larger |I S |. (2) The increasing mutual inductance reduces |I S | while keeping the Q S unchanging. Figs.31(a) and 31(b) show the variation of the primary self-inductance and the mutual inductance of Design 1 and Design 3 with different heights of the secondary above the primary. As can be seen, the self-inductance remains constant, while the mutual inductance reduces with the increasing height. In order to compare the |I P |, |I S |, Q P , Q S of the foil and Litz TX conductor structures, a calculation is applied based on the parameters in Table 8. The frequency is 20 kHz. The required power is 2 kW. The primary and secondary voltage, V P and V S , are controlled at the same level to make sure that I P and I S have the same amplitude. θ is 90 degrees. Table 8 Simulation parameters (foil TX conductor and Litz TX conductor) The primary reactive power Q P is zero, according to the analysis above. The variation of the current amplitude |I P | or |I S | and the reactive power Q S are shown in Figures 32(a) and 32(b). As can be seen, the current amplitude and Q S increase with the increasing height. As the primary self-inductance L P and the mutual inductance M of Design 3 with paralleled Litz transmitter is larger than Design 1 with the foil conductor transmitter, |I P | or |I S | of Design 3 is relatively smaller and its Q S is also smaller. Figs.34(a) and 34(b) show the variation of the primary self-inductance and the mutual inductance of Design 1 and Design 3 with different angles theta (which is shown in Figure 33). As can be seen, the self-inductance remains constant, while the mutual inductance reduces with the increasing angle. In order to compare the |I P |, |I S |, Q P , Q S of the foil and Litz TX coil structures, a calculation is applied based on the parameters in Table 9. The frequency is 20 kHz. The required power is 2 kW. The primary and secondary voltage, V P and V S , are controlled at the same level to make sure that I P and I S have the same amplitude. θ is 90 degrees. Table 9 Simulation parameters (foil TX coil and Litz TX coil) The primary reactive power Q P is zero, according to the analysis above. The variation of the current amplitude |I P | or |I S | and the reactive power Q S are shown in Fig 35(a) and FIG 35(b). As can be seen, the current amplitude and Q S increase with the increasing angle. As the primary self-inductance L P and the mutual inductance M of Design 3 with paralleled Litz transmitter is larger than Design 1 with foil coil, |I P | or |I S | of Design 3 is relatively smaller. For small angle, Q S of Litz coil is smaller, while Q S of Litz coil is larger when angle is large. Figures show how self-inductance, mutual inductance, current and Qs vary with primary TX conductor width and the height of the secondary from the TX conductor for Design 3, based on the parameters shown in Table 10. Table 10 Simulation parameters (Litz TX coil) Designs 1 and 2 have been compared to a known commonly used IPT system having flat circular primary and secondary coils i.e. having at least a primary of TX coil that has return windings. The parameters of such a system are listed in Table 10, alongside the corresponding parameters of Design 1 and Design 3 for comparison. Table 10 Simulation parameters (foil TX coil and circular TX coil) It can be seen from Table 10 that the foil TX conductor structure requires larger energizing current since the mutual inductance M is less than that of the common (i.e. circular) design. Although the foil TX conductor structure requires larger currents |I P | and |I S |, the reactive power Q S exchanged in the secondary side is small since the primary inductance L P is small. The output power and the power loss in coils for Design 3 can be determined from the following: The power loss in converters can also be determined (refer to Figures 37 and 38): The parameters relating to the loss calculation are shown in Table 11 below: Table 11 An N-parallel converter module in which one or more additional switches are provided in parallel in one or more legs of the converter, as shown for example in Figure 39, has been found to reduce switching losses: Assuming there are N converter modules connected in parallel, the conduction loss and the switching loss in each switch can be calculated as: It can be seen that N-parallel module can reduce the conduction loss and reduce the current stress, keeping the switching loss unchanged. Therefore, an IPW65R080CFD MOSFET may for example be applied in a high power level application by considering or employing the parallel connection. As set forth earlier in this document, the relatively low inductance of the transmitter coupling structures allows for systems with no primary side compensation. A simple, cost-effective and advantageous compensation arrangement is shown in Figure 40, in which a secondary-side compensation topology is introduced. There is only one series capacitor Cs on the secondary side, while there is no compensating capacitor on the primary side. Two specific operation or control cases are discussed below, to help decrease the VA rating of the source. The system of Figure 40 may be represented mathematically as: It is assumed that the inductive reactance of L P and L S are X P = ω L P and X P = ω L S When ^^ ௌ is compensated by the capacitor C S , the total reactance on the secondary side is For the operational case 1, Xs is fully compensated by the capacitor Cs, and it is assumed that there is no reactive power provided by the secondary source. The condition can be expressed as: The angle θ between Vp and Vs can be solved as: The output active and reactive power are: The vector or phasor diagram for the voltage and current on the primary and secondary sides are shown in Figure 41. The active power can be controlled by regulating Vp and Vs, while keeping θ constant to make the reactive power Qs zero. However, as the output power demand increases, the reactive power Qp on the primary side also increases, which increases the requirement for the VA rating of the primary source. For operational case 2, Xp is fully compensated by the reflected impedance, and Xs is partially compensated by Cs but there is no reactive power provided by the secondary source. The condition can be expressed as: The angle θ and the compensation ratio where k is the coupling coefficient between TX and RX coils are given by: The reactance Xsc on the secondary side is: The output active and reactive power are: The vector or phasor diagram for the voltage and current on the primary and secondary sides for case 2 are shown in Figure 42. For the given θ and ^^ ^ defined in case 2, the active power can be controlled by regulating V P and V S . By operating under the conditions of case 2, Q P and Q S are kept zero, which will help to decrease the demand for high VA rating of sources compared to case 1. For the case of conventional series-series (S-S) compensation, X P and X S are fully compensated by series capacitors C P and C S respectively, and there is no reactive power provided by the primary and secondary sources. The condition can be expressed as The angle θ between V P and V S can be solved as: The output active and reactive power is: The phasor diagram for the voltage and current on the primary and secondary side is shown in Figure 43. With the conventional S-S compensation, the active power can be controlled by regulating V P and V S , and Q P and Q P are zero when However, the conventional S-S compensation requires two capacitors C P and C P to make Q P and Q P zero, which increases the cost of the system. Comparing case 1 and case 2 it can be seen that θ required for case 1 and case 2 is the same, which is determined by the ratio o After selectingθ , the next step is selecting Cs so that Xs is compensated to = 0 for case 1, while Xs is compensated For case 1, is only provided by the output reactive power from the voltage source V P . The output reactive power is he output active power For case 2, as a negative value, and Q P,X is only provided by the reflected reactance The reactive powe The output reactive power from the voltage source V P is 0. The output active power is A calculation can be performed using the parameters shown in Table 12 below. Table 12 The parameters for the calculation. In the calculation of case 1 and case 2, V P and are chosen as two variables to see the profiles of the power and efficiency. θ is changed according to the rati the design of C S in case 2, to simplify the analysis, it is assumed th which means adapting C S to other ratios not included in the simulation. In the calculation for S-S compensation topolo regulated to control the output power, and θ is kept constant at 90 degre es. To compare the three cases, α/ β is chosen as the variable and the reactive power, power loss, and efficiency are considered along the trajectory of an output power of 3.3 kW, which is shown in Figure 44. In Figure 44, whenα/ β increases, |I P |, | V S | decrease and |I S |, | V P | increase in three cases. The maximum efficiency in case 1, case 2, and S-S compensation are 92.97%, 92.57%, and 92.97%, respectively. The most efficient operating points are located close to α/ β = 1. At each most efficient operating point, the primary apparent power | S P | is 3730 VA, 3570 VA, and 3549 VA in case 1, case 2, and S-S compensation, respectively. The secondary apparent power | S S | is kept constant at 3.3 kVA in three cases, as it is assumed that there is no reactive power provided by the secondary source. Comparing case 1 and the S-S compensation, by removing the primary-side compensating capacitor, the cost of the IPT system can be reduced in case 1. However, the reactive power consumed by L P is provided by the primary source, which increases the VA rating of the primary source in case 1. Comparing case 1 and case 2, by partially compensating L S , the reactive power consumed by L P is only provided by the reflected reactance in case 2. With the most efficient operation, the required | S P | decreases from 3730 VA in case 1 to 3570 VA in case 2, without much loss of efficiency. Therefore, the proposed operation method in case 2 will help to decrease the VA rating of the source in a cost-efficient IPT system with only secondary-side compensation. Parameters for the number of wires/cables 50 and the spacing between them for implementing a primary coupler according to Design 3 are now considered. Referring to Figure 45, the number of wire lengths 50 and the distance between each length can be optimized in one segment for dynamic wireless power transfer (DWPT) to achieve better performance in a comprehensive manner. Thus, the optimization problem can be stated as follows: Change the number of wire lengths 50 (N) and the distance (d) between each wire length of the primary paralleled Litz wires in one segment λ of a DWPT system when the total MMF is constant, to achieve a multi-objective optimization, which includes: 1. Improve the stability of coupling flux Φ in the RX coil with the lateral displacement x; 2. Improve the magnitude of coupling flux Φ in the RX coil; 3. Reduce the loss in Litz wires of the TX coil; 4. Reduce the cost of Litz wires of the TX coil. a. Variables N and d N: number of the paralleled turns of Litz wires in the primary; d: distance between the center of each wire length. The value of d is assumed to be identical between any adjacent wire lengths. b. Optimization objectives The symbol σ Φ is the standard deviation of the coupling flux in the vertical RX coil and can measure the uniformity of the magnetic field B distribution and coupling. Ф i is the coupled flux at a position P i in the RX at a height of h, as shown in Figure 45. The symbol m represents the number of measured positions. The symbol is the average magnitude of Φ i and can measure the magnitude of coupling. Mirror image is used to calculate the magnetic field of the primary coil fitted with a ferrite plate. 2. Minimize P loss The power loss in Litz wires of the TX coil is the sum of losses of all the paralleled lengths, which includes DC loss and AC loss. The AC loss is caused by the skin effect and proximity effect. These losses of Litz wires with a unit length can be calculated according to the equations below. Where j means the j th turn of the TX coil. The symbol n is the number of strands in a Litz wire, and Rdc is the DC resistance of a single strand in the Litz wire. The parameters F R and G R are factors introduced in the loss model, which are frequency dependent. The symbol d a is the diameter of a Litz w ire, is the peak current in each length, and is the peak external magnetic field posed by other lengths and should be evaluated at the conductor centers. Assume the current is evenly distributed in each wire, as the conductor is Litz wire. Thus, 3. Minimize the normalized cost C N Where C 0 is the cost per length of Litz wires per unit length, and N max is the maximum number of the paralleled lengths of Litz wires of the TX coil. c. Parameter setting for optimization Item Symbol Value total current in the TX coil I t [A] 50 segment length λ [m] 0.2 observation height of Φ h [m] 0.01 width of the RX coil w S [m] 0.15 length of the RX coil lS [m] 1 number of turns of the RX coil NS 1 diameter of a Litz wire d a [mm] 4 number of strands in a Litz wire n 1000 maximum number of turns of Nmax 100 the TX coil Distance between lengths d [m] 0.002 ~ 0.04 c. Optimization method Parameter sweep is conducted for the two variables N and d in Matlab, according to the parameter setting. Meanwhile, these two variables need to satisfy the equation below to guarantee valid calculation. Values of the three optimization objectives CV, P loss , and C N are then obtained for each set of N and d. Figure 46 (a), (b), and (c) give the variation of CV, P loss , and C N with N and d, respectively. As can be seen, these three objectives show different variations with N and d. In Figure 46 (a), for a given distance d, a lower CV can be achieved by increasing the number of lengths N. In Figure 46 (b), the power loss P loss in the TX coil reduces with the increasing N, as the current in each length is becoming smaller. However, increasing N will increase the cost C N as shown in Figure 46 (c). In order to determine the optimum design, Figure 47 gives the variation of the three objectives CV, P loss , and C N with each other. The color bar indicates the variation of power loss. Each point in Figure 47 indicates a design with specific N and d. As can be seen, when C N decreases and CV is kept identical, Ploss increases. When CV decreases and C N is kept identical, Ploss has slight variation. Though the design with the least CV, P loss , and C N is desired, the three objectives cannot be achieved simultaneously according to Figure 47. Therefore, two designs with relatively low objectives are selected in Figure 47, which are represented by red stars. Their performance is compared to determine the optimum solution. The number of turns and turn distances of the two selected designs and their corresponding objectives are given in Table 13. The selected design 1 has lower C N and CV and a relatively higher loss compared to the selected design 2. Table 13 Performance of the selected designs. The variation of coupling flux in the RX coil of the two selected designs with lateral displacement x is shown in Figure 48. The coupling flux in design 2 is generally higher than in design 1, however, with relatively lower stability and a much higher number of lengths. Considering that the selected design 1 has only one-third of the cost compared to design 2, the selected design 1 can be chosen as the optimum design. Design 1 has spacings of up to 10 wire diameters between wires, with 6 lengths. Design 2 has spacings of 2-4 wire diameters between wires. Variations of the high performing designs above are possible. For example, a primary or transmitter structure according to Design 3 may have a 5 to 10 lengths 50 with a range of 0.2 - .06m between lengths, or 15-25 lengths with a spacing of 0.08 to 0.015m therebetween. 2D simulations conducted for the selected designs in COMSOL are shown in Figure 48 which agrees with the foregoing conclusions. It can be seen that a novel transmitter together with a vertical Litz receiver is provided that achieves stable and effective inductive coupling for a number of applications, one of which includes dynamic EV charging. Either flat ferrite plate or U-shaped ferrite can be added to the transmitter to enhance the performance, while the receiver can be air-cored. The flux density distribution and coupled flux in the receiver shows that the proposed arrangement can improve upon existing designs regarding the coupling performance, weight, and cost on the receiver side. Throughout the description like reference numerals are used to refer to like features in different embodiments. It will be understood that the primary or secondary couplers described herein may be interchanged i.e. the primary may be used as a secondary or vice versa, and in a bidirectional system the primary and secondary couplers will effectively be interchanged dependent on the direction in which power is being transferred. Unless the context clearly requires otherwise, throughout the description, the words “comprise”, “comprising”, and the like, are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense, that is to say, in the sense of “including, but not limited to”. Although this invention has been described by way of example and with reference to possible embodiments thereof, it is to be understood that modifications or improvements may be made thereto without departing from the scope of the invention. The invention may also be said broadly to consist in the parts, elements and features referred to or indicated in the specification of the application, individually or collectively, in any or all combinations of two or more of said parts, elements or features. Furthermore, where reference has been made to specific components or integers of the invention having known equivalents, then such equivalents are herein incorporated as if individually set forth. Any discussion of the prior art throughout the specification should in no way be considered as an admission that such prior art is widely known or forms part of common general knowledge in the field.