Determining Charging Frequency for Wireless Power Transfer

Abstract

A charging frequency between a transmitting coil (106) and a receiving coil (105) of a wireless power transfer (WPT) system is found by setting an input voltage to the transmitting coil to a fixed voltage, shunting the receiving coil, determining a first input current frequency range in which a resonant frequency is expected to be found, frequency poling within the first input current frequency range using a set of polling frequencies within the first input current frequency range, and selecting a polling frequency with a minimum input current as the charging frequency. The first input current frequency range can be found by setting the first input current frequency range based on regulations, international electromagnetic frequency allocations, or past experience with WPT between the transmitting coil and the receiving coil. A frequency range near the minimum of the input current is selected to poll a charging frequency that approximates the resonant frequency.

Description

[Technical field]

[0001] The present disclosure relates to wireless power transfer, and more particularly, to devices, systems, and methods relating to wireless power transfer to a remote system, such as a vehicle that includes a battery. More particularly, the present disclosure relates to identifying a charging frequency for wireless power transfer that minimizes charging time and provides improved charging efficiency. [Background technology]

[0002] The use of magnetic induction (wireless) power transfer systems to deliver power through an air gap has been known since 1900. A primary (transmitting) induction coil is paired with a secondary (receiving) induction coil and power is transferred through a dielectric, typically an air gap. It is also well known to apply an alternating current (AC) to the primary coil to generate the required oscillating magnetic field. Maximum power transfer efficiency is achieved with magnetically coupled inductive wireless power systems when the AC input is at the resonant frequency.

[0003] The recent trend away from internal combustion engines, along with advances in energy storage, is making battery-powered electric vehicles (BOEVs) more marketable. Wireless power transfer (WPT) systems can be used to charge BOEVs. WPT's inherent safety, reliability from no exposed wires, connectors and no moving parts are useful as high-power charging to reduce overall charging times becomes more common. Summary of the Invention

[0004] Various examples are described below to introduce in a simplified form some of the concepts that are further described in the Detailed Description. This Summary is not intended to be used to limit the scope of the claimed subject matter.

[0005] A wireless power transfer (WPT) charger is provided that includes a transmitting coil for wirelessly transmitting electrical energy to a receiving coil at a charging frequency, and a controller including a processor that executes instructions to determine the charging frequency between the transmitting coil and the receiving coil. The controller executes instructions to perform a method of performing operations including setting an input voltage to the transmitting coil to a fixed voltage, shunting the receiving coil, determining a first input current frequency range in which a resonant frequency is expected to be found, frequency polling within the first input current frequency range using a set of polling frequencies within the first input current frequency range, and selecting a polling frequency with a lowest input current as a charging frequency. The first input current frequency range in which the resonant frequency is expected to be found may be determined by setting the first input current frequency range based on at least one of regulations, international electromagnetic frequency allocations, or past experience with WPT between the transmitting coil and the receiving coil.

[0006] Polling frequencies within the first input current frequency range may include determining an input current minimum at the set of polling frequencies, determining a second input current frequency range around the input current minimum that is narrower than the first input current frequency range, and polling frequencies within the second input current frequency range using a second set of polling frequencies within the second input current frequency range. The second input current frequency range around the input current minimum may be determined by determining two polling frequencies in the set of polling frequencies at which the input current is lowest and using the two polling frequencies to define upper and lower boundaries of the second input current frequency range. Alternatively, the second input current frequency range around the input current minimum may be determined by determining two polling frequencies adjacent to a polling frequency having a lowest input current and using the two polling frequencies to define upper and lower boundaries of the second input current frequency range. In performing the polling, multiple frequency maps may be determined, the frequency maps having measurement distributions determined from the probability that the input current minimum is within a frequency range of the respective frequency maps. The polling frequencies are obtained from the respective frequency maps.

[0007] In a sample configuration, if the polling frequency with the lowest input current cannot be determined beyond an input current uncertainty limit, the processor of the controller may further execute instructions to perform operations including: segmenting an input current vs. input current frequency response curve into a lower response curve above the input current uncertainty limit and an upper response curve above the input current uncertainty limit, and selecting a highest frequency point on the lower response curve above the input current uncertainty limit as a charging frequency. Alternatively, the charging frequency may be selected as a midpoint frequency between a lower crossing frequency where the lower response curve crosses the input current uncertainty limit and an upper crossing frequency where the upper response curve crosses the input current uncertainty limit, or as a frequency between the midpoint frequency between the lower crossing frequency and the upper crossing frequency, the selected frequency being a percentage of the distance between the lower crossing frequency and the midpoint frequency.

[0008] Alternatively, if the polling frequency with the lowest input current cannot be determined beyond the input current uncertainty limit, the processor of the controller may further execute instructions to perform operations including fitting curves to at least three polling frequencies adjacent to the input current uncertainty limit for the lower response curve and the upper response curve to find an intersection between the second and second linear projections of the lower response curve based on at least two polling frequencies adjacent to the input current uncertainty limit and selecting the frequency of the intersection as the charging frequency. In another alternative, the processor of the controller may further execute instructions to perform operations including fitting curves to at least three polling frequencies adjacent to the input current uncertainty limit for the lower response curve and the upper response curve to find an intersection between the curves fitted to the lower response curve and the upper response curve and selecting the frequency of the intersection as the charging frequency.

[0009] The steps of determining the first input current frequency range, polling a frequency, and selecting a charging frequency may be performed multiple times for different input voltages for each iteration. The different input voltages for each iteration may include different percentages of a requested power level. In such a configuration, the processor may execute additional instructions to perform operations including determining a relationship between a power level and a resonant frequency at each power level, and determining a resonant frequency at the requested power level from the relationship.

[0010] In another configuration, the processor of the control device may execute additional instructions to avoid over-current in the transmitting coil due to over-coupling between the transmitting coil and the receiving coil by performing operations including taking a first current measurement at a lower wavenumber boundary of the frequency range and, if the first current measurement exceeds a safe current threshold for current detection, aborting the step of determining the charging frequency or resuming the step of determining the charging frequency at a lower fixed input voltage level to the transmitting coil.

[0011] In yet another configuration, the input voltage to the transmit coil is set to a fixed voltage by shunting the receive coil, setting the first input current frequency to a polling frequency within the first input current frequency range, incrementally varying the input voltage to the transmit coil, measuring the input current to the transmit coil at each voltage increment until the input current measurement exceeds a safety margin threshold but not an adverse condition threshold, and selecting the fixed voltage as the voltage at the last voltage increment where the input current measurement is below the safety margin threshold. Incrementally varying the input voltage to the transmit coil includes using the first voltage increment until the input current of the transmit coil exceeds a guard threshold that is less than the safety margin threshold, and using a second voltage increment less than the first voltage increment until the input current of the transmit coil exceeds the safety margin threshold or a nominal test voltage is obtained.

[0012] This Abstract is provided to introduce aspects of the inventive subject matter in a simplified form, with further description of the inventive subject matter following in the body of the Detailed Description. The particular combination and order of elements described in this Abstract are not intended to provide limitations on the elements of the claimed subject matter. Rather, it will be understood that this Abstract provides a summarized example of some of the embodiments described in the Detailed Description below. [Brief description of the drawings]

[0013] These and other beneficial features and advantages of the present invention will become apparent from the following detailed description taken in conjunction with the accompanying drawings. [Figure 1] FIG. 1 is a diagram illustrating the use and main components of a wireless power transfer system in a sample configuration. [Diagram 2] FIG. 2 is a schematic diagram of a parallel-parallel resonant inductive circuit in a sample configuration. [Diagram 3] FIG. 3 shows diagrammatically the charging procedure in a sample configuration. [Figure 4] 4A, 4B, 4C, and 4D show measurement scenarios using the frequency determination procedure in an example configuration. [Diagram 5] FIG. 5 is a diagram illustrating the main components and subsystems of a bidirectional wireless power transfer system in a sample configuration. [Figure 6] Figures 6A and 6B show the first and second stages of the frequency determination procedure in a sample configuration, and Figure 6C shows the ability to configure multiple frequency polling maps in the same efficient frequency determination in a sample configuration. [Figure 7] Figures 7A and 7B show a method for selecting an efficient frequency when the resonant frequency cannot be found deterministically, and Figures 7C and 7D show a method for estimating an efficient frequency when the resonant frequency cannot be found deterministically. [Figure 8] FIG. 8 illustrates the results of a multi-path implementation of the frequency determination procedure in a sample configuration. [Figure 9] FIG. 9 graphically illustrates the parameters and results of the safety procedure for dealing with overcoupled induction coils. [Figure 10] FIG. 10 illustrates the main subsystems and communication paths within the WPT in a sample configuration. [Figure 11] FIG. 11 is a flow diagram showing the process of determining the charging frequency performed by the ground side control device in the configuration example. [Figure 12] FIG. 12 illustrates a scenario for determining the fixed input voltage to the transmit coil without harming the WPT electronics. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Next, a detailed description of an exemplary embodiment will be described with reference to Figures 1 to 12. Although this specification provides a detailed description of possible implementations, it should be noted that these details are exemplary and do not limit the scope of the present subject matter in any way.

[0015] In this document directions are given according to their common meanings. ISO 4130:1978 "Road vehicles -- Three-dimensional reference systems and datum marks" is used as a guide for the Cartesian coordinate system, where forward is the -x direction, +x is the reverse or back direction, right is the +y direction and left is the -y direction. The horizontal z=0 plane corresponds to ground level, grade or pavement level, +z is the up direction and -z is the down direction (below grade).

[0016] As used herein, the term "battery" refers to a general chemical energy storage system that can be replaced, supplemented, or hybridized with other portable energy storage systems (e.g., solid-state batteries, reversible fuel cells, ultracapacitors, etc.) Also, while many of the examples used are of wireless power transfer (WPT) systems used to power on-board systems and charge stationary electric vehicle (EV) batteries, this application is by no means the only application.

[0017] Because the wireless power transfer system is modular and capable of bidirectional charging, the term Ground Transceiver Station (GTS) is used to refer to a grouping of Ground Transceiver Assemblies (GTAs) and Vehicle Transceiver Station (VTS) is used to refer to a grouping of Vehicle Transfer Assemblies (VTAs). A GTS is made up of one or more GTAs and a VTS is made up of one or more VTAs. To maximize power transfer capability, all VTAs in a VTS are paired and aligned with their corresponding GTAs in the GTS. Alignment refers to the primary and secondary coils being magnetically aligned, not necessarily geometrically aligned.

[0018] As described in U.S. Patent Application Serial No. 16 / 940,658, entitled "Magnetic Field Management for Efficiency Improvement," automatic determination of resonant frequency is useful even when the vehicle make and model are communicated, the characteristics of the GTA are known and the characteristics of the VTA are available, or the vehicle has previously been charged by the WPT charger and past physical and magnetic information is maintained by the GTS or charging station controller. The exact resonant frequency of the paired GTA and VTA may vary from unit to unit and EV to EV due to differences in unit construction between VTA (and GTA) manufacturers. Factors that affect the coupling coefficient (k) and therefore the resonant frequency include the fact that EVs of the same make, model, and year may have different coil gaps (e.g., Z-gap). The Z-gap may also be affected by the loaded and unloaded state of the EV, the use of adaptive suspension or kneeling, and aftermarket modifications that affect the height of the vehicle chassis. Each charge may also experience misalignment due to roll, pitch, and yaw of the VTS.

[0019] When using modular WPT, i.e. multiple primary coils and paired secondary coils, the resonant frequency of each GTA-VTA pair must be predetermined to maximize power transfer efficiency.

[0020] "Deterministic" is used when the resonant frequency is determined by measuring the GTA current response versus frequency. "Selected" is used when the resonant frequency cannot be determined by measurement, but a more efficient frequency can be selected than allowed by the limitations of the measurement data. "Estimated" is used when the resonant frequency cannot be found directly by measurement, but a more efficient frequency can be estimated from the measurement data.

[0021] Figure 1 Figure 1 shows an example of a wireless power transfer system for an EV during charging. An EV 101 rests on a ground charger 102. The ground charger (Ground Transceiver Station (GTS)) 102 is shown here protruding above the pavement 103, although other GTS 102 installations are contemplated (e.g., mounted flush with the pavement or recessed below the pavement). The Vehicle Transceiver Station (VTS) 104 is mounted on the underside of the EV 101 in this example.

[0022] The GTS 102 includes one or more coil assemblies 106 (collectively, ground transceiver assemblies (GTA)) with ancillary and structural members for mechanical support, cooling, eddy current shielding, and inductive communication link. Similarly, the VTS 104 includes one or more vehicle transceiver assemblies (VTA), each having a coil assembly 105 and ancillary and structural members for mechanical support, cooling, eddy current shielding, and inductive communication link. The GTA coil 106 and the VTA on-board coil 105 can be made with the same size in the XY plane but with different number of turns. The inter-coil gap 107 in the Z plane is a function of the type of ground charger 102, the EV configuration, and the EV load. In addition to the inter-coil gap, the GTA to VTA orientation can experience centering offsets as well as deviations from parallel coil matching in roll, pitch, and yaw. The positioning deviations can be unique for each charging session and are unknown prior to measurement. Each positioning deviation and inter-coil gap affects the value of the coupling constant (k), which in turn sets the resonant frequency of each GTA to VTA individually for each charging session. The curb 108 is a visual (and possibly tactile) indicator of the GTS position for the driver, a driver assistance system, or an autonomous driving system.

[0023] While both the GTS 102 and VTS 104 are shown as single coil assemblies in FIG. 1 (the GTS has a single GTA and the VTS has a single VTA in FIG. 1), multiple coil assembly pairs (wireless couplers) can be used for higher power transfer. The benefits of a modular approach to wireless power systems also come into play. By manufacturing a single GTA and a single standard VTA, economies of scale can be achieved by combining the GTA into a larger Ground Transmitting Station (GTS) to service an EV-mounted VTS with one or more VTAs.

[0024] The modular GTS and VTS enable a charging station that can build flexible, dynamically allocatable, dynamically configurable GTS configurations (e.g., one GTA per GTS, two parallel (2x1) GTAs per GTS, two series (1x2) GTAs per GTS, three series (1x3) GTAs per GTS, four GTAs (2x2) per GTS, six GTAs (2x3) per GTS, or a GTA configuration that supports up to 100 GTS (corresponding set of VTAs served) with the maximum number of vehicles supported by the GTA.

[0025] Also, the GTS 102 and VTS 104 can be part of a bidirectional WPT where the primary and secondary are switchable and the EV supplies energy to an electric grid (not shown). The VTS 104 paired with the GTS 102 can be operated in a bidirectional mode so that wireless power transfer can be reversed, with power originating from the power grid (through the GTS) to the vehicle (through the VRTS) or from the EV's energy storage (e.g., battery pack) sent by the VTS to the GTS to power a DC or AC load (e.g., a home).

[0026] Figure 2 FIG. 2 illustrates, using a Thevenin circuit representation, an exemplary balanced parallel resonant network (PPRN circuit) for use in an inductively coupled wireless power transfer system. A primary coil assembly 201 and a secondary coil assembly 202 serve to transfer energy via mutual induction from a primary coil 204 to a secondary coil 203. The primary coil 204 and primary capacitor 205, and the secondary coil 203 and secondary capacitor 206 have values ​​selected to form a mutually resonant circuit. An inter-coil gap 209 is present in all inductive WPT systems and may vary with each EV and each charging session. A secondary output 211 appears as an AC voltage controlled current source (VCCS) to a load (not shown), nominally the EV battery pack.

[0027] In a resonant inductive power transfer circuit, the complex impedance Z seen by the voltage source 210 varies with the angular frequency of the coupled coil pair.

[0028]

number

[0029] , j2=(-1), with a coupling coefficient k, and the primary resistance R, primary coil inductance L, and primary capacitance C are as shown in Equation 1.0:

[0030]

number

[0031] Therefore, the magnitude of the GTA input current is expressed as a voltage (V) and current (I) with impedance Z according to Equation 1.1:

[0032]

number

[0033] Although the frequency can be selected and the resistance, inductance, and capacitance are all known prior to wireless charging, the value of the coupling coefficient (k) is not known, and therefore the input current to the primary coil assembly cannot be deterministically calculated for a useful range of input voltages and input frequencies for the primary assembly.

[0034] Figure 3 FIG. 3 shows the timeline of events leading up to and including a wireless charging session.

[0035] Initialization 301 occurs when the electric vehicle (EV) and ground transceiver station (GTS) establish two-way communication. Once two-way communication is working, information can (optionally) be obtained from a data store 302. This information may include data about the EV (e.g., EV identifier, nominal height, resting or kneeling height, number of VTAs, magnetic profile of each VTA (e.g., coil characteristics such as inductance, dimensions, total number of turns, resistance, impedance, calibrated center frequency at resonance, etc.). The data store 302 may also have estimates of historical resonant frequency from previous charging sessions.

[0036] During alignment 303, the EV's driver (or driver assistance system or autonomous driver system) receives guidance information and course corrections so that the EV's VTS is well-aligned with its GTS and each VTA is paired with a GTA to keep the EV stationary. Guidance aids can be visual markers, visual light indicators or reflectors, radio antennas embedded in the pavement, radio beacons, pavement shaping (dents or grooves), pavement markers, etc.

[0037] In verification 304, the status and alignment of the EV is checked. Using GTS-EV wireless communication, the vehicle status (lowering or kneeling, battery status (voltage, SoC), VTS status (each VTA), Battery Management System (BMS) status, requested power level) is queried by the charger, as described, for example, in U.S. Pat. No. 10,135,496, “Near Field, Full Duplex Data Link for Use in Static and Dynamic Resonant Inductive Wireless Charging,” U.S. Pat. No. 10,826,565, “Near Field, Full Duplex Data Link for Resonant Inductive Wireless Charging,” and U.S. Pat. No. 11,121,740, “Near Field, Full Duplex Data Link for Resonant Inductive Wireless Charging.”

[0038] For frequency setting 305, a polling approach for resonant frequency determination as described herein is used. For each GTA paired with a corresponding aligned VTA, a fixed ground coil voltage is applied at a set of frequencies and the resulting GTA input current is measured. From the polling and the magnitude of the GTA current at the set of input frequencies, a resonant frequency is determined for each VTA / GTA pair. The set of frequencies polled can be equally spaced linearly distributed across the available bandwidth or can be stochastically or historically centered to obtain the closest estimate of the resonant frequency. Example frequency polling intervals depend on the polling distribution pattern or stage (multiple stage frequency determination) and can range from 10 Hz to 1000 Hz. Minimizing the number of samples can reduce charging delays due to frequency determination.

[0039] If the exact resonant frequency cannot be determined, an estimated or calculated frequency based on the transfer efficiency is used. Measuring the GTA coil current can be accomplished using a current sense circuit as described in U.S. patent application Ser. No. 16 / 825,624, entitled "Current Sensing in Wireless Power Transfer Systems."

[0040] In charging 306, each VTA is unshunted and then each GTA receives an AC current at a determined resonant frequency to enable efficient transfer of power to the corresponding VTA at the required (nominal 100%) level. The AC input signal may be a sine wave, ramp, or square wave.

[0041] During charging 306, the coupling coefficient may change, for example if the GTA and VTA components heat up or if the EV is loaded or unloaded. It is possible to interrupt the power transmission and restart the frequency determination procedure 307 of the set frequency 305. This is a purely optional operation, since restarting the frequency determination procedure will increase the duration of the charging session.

[0042] The frequency determination procedure may be performed simultaneously on all active GTAs within the GTS.

[0043] Although the total time of pre-charging activities (less than 5 seconds) is small compared to the total charging time, the charge start delay is noticeable as the customer focuses on charging, and is therefore a disadvantage. This perception of charge time due to charge start delays is in contrast to the increased charge time due to the selection of an inefficient charging signal frequency, which is typically not noticed by the customer.

[0044] Figures 4A-4D Figures 4A, 4B, 4C, and 4D are used to explain different results in the charging frequency determination due to the coupling coefficient (k) being unknown prior to a charging session, primarily due to misalignment between coils, differences in coil manufacturing, and variations in Z-gap between the coils.

[0045] Figure 4A 4A graphically illustrates a scenario showing the GTA input current draw versus varied input signal frequency when the GTA input voltage is set to a fixed level and the associated VTA shunts its coil. In the scenario of FIG. 4A, the minimum GTA current 406 exceeds the uncertainty limit 407 due to the accuracy of the current sense circuitry and is therefore the optimum frequency for WPT.

[0046] FIG. 4A shows an example in which the x-axis 401 is set as frequency and the y-axis 402 is set as ground coil current. The lower frequency limit 403 and the upper frequency limit 404 are set by local regulations or international electromagnetic frequency allocations (e.g., in SAE standard SAE J2954, the WPT band is centered at 85 kHz and ranges from 81.3 to 90 kHz). The illustrated WPT band 405 may use a back-off margin of 50 to 1000 Hz from the regulatory or statutory upper and lower frequency limits to ensure that the power transfer signal does not exceed the assigned band. As shown, the GTA current curve 408 reaches a current minimum above the uncertainty limit 407, which corresponds to the resonant frequency.

[0047] Figure 4B FIG. 4B illustrates an alternative scenario showing the GTA coil input current draw versus varied input signal frequency when the GTA input voltage is set to a fixed level and the associated VTA coil is shunted, and the minimum GTA coil current cannot be uniquely determined. FIG. 4B illustrates frequency on the x-axis 401 and ground coil current on the y-axis 402. The WPT band 405 is set with a lower frequency limit 403 and the upper frequency limit 404 set by local regulations or international electromagnetic frequency allocations (back-off margin not shown). A first GTA current curve 408 can be determined for a lower range of polling frequencies and a second GTA current curve 409 can be determined for an upper range of frequencies, but the minimum of the GTA current cannot be determined directly because it is below a current sensor sensitivity threshold 407, resulting in an ambiguity region 410 where the GA current cannot be determined. Since the minimum of the GTA current cannot be measured directly, estimation techniques can be applied to find an efficient, albeit suboptimal, WPT transfer frequency.

[0048] Figure 4C Figure 4C graphs an alternative scenario showing the GTA coil input current draw versus varied input signal frequency when the GTA coil input voltage is set to a fixed level and the associated VTA coil is shunted. Figure 4C sets the x-axis 401 as frequency and the y-axis 402 as the GTA coil current measurement. The lower frequency 403 and the upper frequency 404 are set by local regulations or international electromagnetic frequency allocations, plus a back-off margin, so that a WPT band 405A GTA current curve 411 can be determined for a range of polling frequencies. Since the measured GTA current minimum 412 is at the upper frequency boundary 404, the optimal WPT frequency cannot be uniquely determined and the WPT cannot radiate beyond the permitted WPT band 405, so the minimum frequency 412 is reported for use as the determined most efficient frequency.

[0049] Figure 4D FIG. 4D plots an alternative scenario showing the GTA coil input current draw versus the varied input signal frequency when the GTA coil input voltage is set to a fixed level and the associated VTA coil is shunted. In FIG. 4D, the x-axis 401 is frequency and the y-axis 402 is GTA coil current. The lower frequency limit 403 and the upper frequency limit 404 are set by local regulations or international electromagnetic frequency allocations. A partial GTA current curve 413 can be determined for a range of polling frequencies. The optimal WPT frequency cannot be determined univocally because the measured GTA current minimum 414 is at the measurement uncertainty threshold 407. The measured GTA current minimum 414 is the most efficient frequency found and can be used in subsequent charging sessions. Alternatively, a computational approach can be used that projects a partial GTA current curve 413 (either a linear projection or a curve fitting based on polled current, frequency points collected along the measured curve 413) and finds the intersection (i.e., zero current) of the curve 413 with the x-axis 401. If the calculated x-axis intersection is within the upper frequency bound 404, then the frequency of the calculated x-axis intersection can be reported as the most efficient frequency.

[0050] Figure 5 FIG. 5 is an exemplary high-level functional diagram of power flow-through and conversion by a bidirectional wireless power transfer (WPT) system in a sample configuration.

[0051] While certain components are inherently bidirectional and symmetrical (e.g., resonant inductive circuits, also known as open-core transformers) and can be shared, the forward (charging) and reverse (discharging) power transfer paths rely on a branch-simplex architecture and require switches 509, control logic (not shown), and communication links (also not shown) to initiate and complete the power transfer paths for each of the forward (charging) and reverse (discharging) usage scenarios. Although both forward and reverse paths are presented here, WPT can be implemented (and optimized) using only one path, nominally the forward path.

[0052] In the forward direction, power is nominally provided by a power grid 501. Depending on the grid connection, power can be single-phase alternating current (AC), direct current (DC), or polyphase AC. The power grid 501 includes the transformers required to step down the voltage from the high voltage transmission line. In this example, single-phase AC is provided by the power grid 501 with sufficient capacitance present so that the power factor is regulated to near unity.

[0053] Alternating current (AC) power is converted to direct current (DC) by AC / DC converter 502. This function can be realized by an active (switch-based) rectifier or a passive (diode-based) rectifier. DC / AC converter 503 takes the input DC power and converts it into a high-frequency AC (nominal 85 kHz in this example) sine wave signal. The DC / AC conversion operation by DC / AC converter 503 can be realized by using an inverter.

[0054] The AC power signal is passed to a resonant air core transformer 504 having a primary coil and a secondary coil. The AC power is converted to magnetic flux at the primary side and inductively coupled to the secondary side. The secondary coil converts the received magnetic flux into an AC power signal. The AC power signal is passed to an AC / DC converter 505. The AC / DC conversion function can be achieved by an active (switch-based) rectifier or a passive (diode-based) rectifier.

[0055] The resulting DC signal is used to charge an energy storage device 506, nominally a rechargeable chemical battery, but could also be a capacitor bank, a reversible fuel cell, a solid-state battery, or one or more of a hybrid combination of the above. The DC signal can also be used to directly power electronic devices. Being bidirectional, the energy storage device 506 can output stored power as DC on the reverse transmission line. The DC power is converted to the required AC power signal by a DC / AC converter 507.

[0056] This AC power signal is input to a resonant induction circuit 504. In this reverse path scenario, the coils operate in the opposite direction to the forward path. The AC power is converted to magnetic flux in the primary coil of an open core transformer 504, which is inductively coupled with the secondary coil. The secondary coil converts the received magnetic flux into an AC power signal. The resulting AC power is frequency regulated by an AC / DC converter 508. In one configuration, an AC / DC / AC converter is used as the AC / AC converter 508, and the AC / AC frequency regulation operation is achieved using an AC / DC rectifier, which is then converted from DC to AC at the required frequency by an inverter circuit. The power grid 501 in this example includes the necessary transformers to convert the AC power to the desired voltage and AC / DC conversion as required to interface with the power grid supply.

[0057] In an alternative configuration, if a DC power feed is available from a DC power grid (not shown), the DC / AC converter 503 can be sized to directly accept the DC feed, eliminating the need for the preceding AC / DC stage 502.

[0058] Figure 6A FIG. 6A graphically illustrates a scenario showing the GTA coil input current draw versus the GTA coil input current signal frequency when the GTA coil input voltage is set to a fixed level and the associated VTA coil is shunted in the first stage of resonant frequency determination. In FIG. 6A, the GTA coil input current signal is adjusted to a small number of preselected frequencies to determine a frequency range of interest for more detailed examination. These preselected frequencies can be applied to maximize the efficiency of the search to determine the minimum of the input current if existing data is available. For example, a random set of frequencies can be applied. A set of frequencies centered around the resonant frequency can also be applied. A first set of widely spaced frequencies can be applied, and a second set of frequencies focused around a frequency from the first set of frequencies can be applied to obtain the minimum input current to the GTA.

[0059] FIG. 6A shows an x-axis 601 set as frequency and a y-axis 602 set as GTA coil current. A lower frequency limit 603 and an upper frequency limit 604 resulting in a WPT band 605 may be set by local regulations or international electromagnetic frequency allocations, or may be a smaller, possibly offset range from the center frequency based on prior experience with the GTA / VTA pair concerned. The accuracy limit 607 of the GTA current sensor is shown. The GTA current measurement 609 is measured by adjusting the GTA input frequency with the static GTA input voltage set to a percentage of full potential. In this example, the lowest GTA current measurement 608 is above the current sensor uncertainty threshold and is reported at the resonant frequency used during the charging session.

[0060] When a two-stage efficient frequency determination process is used, only a small number of polling frequencies are used across the frequency band defined by the lower 603 and upper 604 frequency boundaries to reduce the time required to complete the first stage. In one embodiment, the preselected frequencies are only applied until a local minimum (minimum of GTA current) is detected, i.e., a measurement bracketed by a first measurement at a higher frequency with a higher GTA current and a second measurement at a lower frequency, also with a higher GTA current.

[0061] Figure 6B 6B shows the second stage of a two-stage resonant frequency determination. The x-axis 601 shows the frequency of the GTA current input and the y-axis 602 shows the GTA coil current at a set input voltage. In this second stage, the lower 610 and upper 611 frequency constraints are tightened based on the GTA current minimum 608 found in the first stage, and all frequency polling is done within these constraints 610, 611. The VTA, which corresponds geometrically to the GTA under test, is shunted while the GTA value is set to a fixed level.

[0062] In one embodiment, the constraints 610, 611 are set to the two frequency poll points having the lowest GTA current measurements determined in the first stage. In another embodiment, the constraints 610, 611 are set to the two frequency poll points adjacent to the frequency poll point having the lowest GTA current found in the first stage polling.

[0063] The second stage frequency polling occurs between the constraints 610, 611 using a selected set of frequencies. The set of frequencies may be selected to minimize a frequency decision period that delays the start of charging and increases overall charging time. The frequency 612 with the lowest GTA current that exceeds an uncertainty threshold 612 is reported as the effective WPT frequency.

[0064] For some charging sessions, if the characteristics of the GTA, VTA, and EV are known (e.g., because the EV previously used the GTA), the second stage of frequency determination can begin without the first stage, since the resonant frequency is expected to be limited to a narrower frequency range.

[0065] Figure 6C FIG. 6C illustrates the ability to set multiple frequency polling maps in the same efficient frequency determination in a sample configuration. In FIG. 6C, the x-axis 601 is set as frequency and the y-axis 602 is set as GTA coil current. The lower frequency limit 613 and the upper frequency limit 614 resulting in a WPT band of 609 may be set by local regulations or international electromagnetic frequency allocations, or may be a smaller, potentially offset range from a center frequency based on prior experience with the GTA / VTA pair involved. The accuracy limit 607 of the GTA current sensor is shown. In this example, three frequency maps 615, 616, and 617 are set, each map having a measurement distribution determined from the probability that the GTA current minimum is within the frequency range of the map. GTA current measurements are made by adjusting the GTA input frequency in increments of total potential.

[0066] In the example of FIG. 6C, a GTS current minimum 618 is found in the intermediate frequency map 616 and that frequency is reported as the effective WPT frequency.

[0067] Figure 7A FIG. 7A shows a scenario for efficient WPT frequency determination when the GTA current minimum is ambiguous. The x-axis 701 shows the frequency of the GTA current input, and the y-axis 702 shows the GTA coil current at a set input voltage. When the current uncertainty limit 706 is exceeded, the GTA current vs. frequency response is partitioned into a lower response curve 708 and an upper response curve 709. The GTA voltage is set to a percentage of the available voltage, and the geometrically corresponding VTA coil is shunted. An ambiguity area 711 exists between the measured upper response curve 709 and the measured lower response curve, so that an efficient WPT frequency cannot be determined directly by measurement. In the example shown in FIG. 7A, one solution is to select a point within or tangent to the ambiguity area 711.

[0068] Figure 7B 7B illustrates the selection of a WPT frequency from the ambiguity region 711. The x-axis 701 illustrates the frequency of the GTA current input and the y-axis 702 illustrates the GTA coil current at a set input voltage. Beyond the current uncertainty limit 706, the GTA current vs. frequency response is partitioned into a lower response curve 707 and an upper response curve 708. The ambiguity region 711 is further subdivided into lower ambiguity frequency bands 712.

[0069] One selection option is to select the highest frequency point 713 on the lower response curve 708. Another option is to estimate using the projection of the lower response curve 708 and the upper response curve 709 as shown in Figure 7B. The selection of this point 713 is as accurate an effective WPT frequency as can be determined by measurement.

[0070] Another option for efficient WPT frequency selection is to use the midpoint frequency 714 of the ambiguity region 711. Although it cannot be measured directly, the midpoint frequency 714 may be closer to the resonant frequency than is the effective measurable frequency.

[0071] Another frequency selection option, which requires more computation, is to use the known asymmetry of the GTA vs. frequency curve to report an effective WPT frequency taken from the lower ambiguous frequency band 712 that extends from the highest measured frequency of the lower current response curve 708 to the midpoint of the ambiguous region, where the frequency can be selected as a percentage of the width of the lower ambiguous frequency band 712.

[0072] Figure 7C In FIG. 7C, since the resonant frequency cannot be measured directly, a projection method is used to determine the efficient frequency. The VTA coil is shunted and the GTA input voltage is set constant. The x-axis 701 shows the frequency of the GTA current input and the y-axis 702 shows the GTA coil current at the set input voltage. Beyond the current uncertainty limit 706, the GTA current vs. frequency response is partitioned into a lower response curve 707 and an upper response curve 708. Since the minimum value of the GTA current is ambiguous, a lower linear projection 716 based on measurements from the lower response curve 707 is calculated. Similarly, an upper linear projection 717 based on measurements from the upper response curve 708 is calculated. The intersection of the two projections 716 and 717 is intersection 718, whose frequency is reported as the efficient WPT frequency.

[0073] Alternatively, if the intersection of the two projections 716 and 717 is below 0.0 A of GA current (below the x-axis 701), the process is stopped and an efficient frequency selection is performed. Alternatively, if the intersection of the two projections 716 and 717 is above 0.0 A of GA current (above the x-axis 701) and sufficient processing power and delay time are available, a secondary projection technique can be used. In the example shown in Figure 7C, the slope of each projection 716, 717 is based on the lowest point (in GA current) of the response curve 707, 708, or alternately uses the average of three frequencies polled adjacent to the uncertainty of the measurement threshold 707 of each response curve 707, 708. The intersection 718 of the predictions 711, 712 is the estimated efficient frequency.

[0074] FIG. 7D details a selection method for determining an efficient GTA frequency for WPT when the resonant frequency cannot be determined directly by measurement. There are no valid GTA current measurements below a current uncertainty 706 threshold. The x-axis 701 is the GTA input frequency and the GTA input current is shown on the y-axis 702. A frequency band 705 is defined by a lower limit 703 and an upper limit 704. The GTA current measurements are shown split into a lower response curve 707 and an upper response curve 708, both of which terminate at the current uncertainty threshold 706.

[0075] To overcome the lack of a direct measurement of minimum GTA current, a lower fitted curve projection 719 is calculated based on measurements from the lower response curve 707. Similarly, an upper fitted curve projection 720 is calculated based on measurements from the upper response curve 708. At the intersection of the projected curves 719 and 720 is the frequency of the intersection that is reported as the efficient WPT frequency.

[0076] Alternatively, if the intersection of the projected curves 719 and 720 is below 0.0 amperes (below the x-axis 791), projection is discontinued and effective frequency selection is performed.

[0077] If the GA current cannot be calculated or is calculated to be negative, a recalculation using alternative means (using alternative curve fitting algorithms or linear projection) may be performed, but it should be noted that such additional processing may lead to a delay in the start of charging and thus in the duration of the entire charging session.

[0078] Figure 8 Since accuracy in estimating the resonant frequency is essential for wireless power efficiency and therefore reducing harmful heating and harmonic radio frequency emissions at high power, multipath estimation can be used.

[0079] In Fig. 8, the result of a multi-pass estimation is shown. The x-axis 801 shows the frequency and the y-axis 802 shows the GTA current. The frequency determination described above is performed four times in this example staying within the legal or regulated frequency band 803. In the first pass, the GTA voltage is set to a specific point (e.g. 10% of full or requested power), which results in a resonant frequency estimation 804 at the measured or estimated minimum GTAA current. In each pass, the GTA voltage is changed (increased in this example), which results in a second frequency estimate 805 at 20% power and a third frequency estimate 806 at 30% power. With enough frequency measurement passes (three or more), the relationship shown by the curve 808 between power level and resonant frequency is derived and the frequency 807 at full (or requested) power is calculated.

[0080] Figure 9 FIG. 9 illustrates precautions taken during or prior to the resonant frequency determination procedure to avoid GTA overcurrent due to overcoupling between the GTA and its paired VTA.

[0081] The x-axis 901 shows the GTA input frequency and the y-axis 902 shows the GTA current. The VTA coil is shunted during this operation while the initial GTA input voltage is fixed at full power transfer level or a percentage of the requested power transfer level. The GTA input frequency is constrained by a lower frequency limit 903 and an upper frequency limit 904, resulting in a frequency search band 905. The uncertainty threshold for current detection 906 is shown as well as a maximum allowed GA current limit 907 for the resonant frequency determination procedure.

[0082] A first current measurement 908 is taken at the lower frequency boundary 903. If the first GTA current measurement 908 exceeds the maximum current limit 904, the GTA can abort the resonant frequency determination procedure or resume the procedure after lowering the fixed GTA input voltage level. A second GTA current measurement 909 can optionally be taken at the upper frequency boundary 904 at the initial GTA voltage, but due to the asymmetry of the coupling coefficients, it will be seen that the first GTA current measurement 908 at the lower frequency boundary 903 will usually be higher than the second GTA current measurement 909 taken at the upper frequency boundary 904. The second frequency point current measurement 909 can be taken to indicate the need for the controller to instruct the system to use even a lower fixed voltage to restart the measurement process (e.g., drop from 30% of the voltage range to 15% of the voltage range rather than restarting polling at 20% of the voltage range).

[0083] The GTS can then reduce the fixed GTA voltage and start the frequency determination procedure again. The VTA coil is still shunted. If a subsequent measurement 910 at the lower frequency boundary 903 is below the GTA current safety limit 907, then a subsequent second GTA current measurement 911 can optionally be made at the upper frequency boundary 904 to double check that the GTA current is below the safety limit for the entire frequency band.

[0084] Figure 10 Figure 10 shows the main subsystem components of a WPT system for EV charging.

[0085] A first subsystem 1001 includes an energy source (e.g., a power grid) and power factor correction (PFC) hardware. The first subsystem 1001 delivers power to an inverter subsystem 1002, which includes the AC / DC converter, the DC / AC converter, and an inverter. The inverter subsystem 1002 can measure the output AC current using a current sensor (e.g., Rogowski coil, split core current sensor, solid core current transducer, Hall effect sensor (either open loop or closed loop)). Each current sensor type has an accuracy level based on its design. For example, the current sensor provides a sufficient level of measurement accuracy defined across the frequency band of interest. Regardless of the type of current sensor, each measurement is degraded by factors that affect the uncertainty, such as output noise, quantization error, temperature variations, conductor position, application of external magnetic fields, magnetization, and common mode voltage of the current sensor. Thus, a limit of the uncertainty in the current measurement exists for any selected current sensor.

[0086] The inverter subsystem 1002 provides an AC signal to a primary assembly subsystem 1003, which includes a primary compensation network (ie, a resonant capacitor) and a primary / transmitter induction coil (Tx).

[0087] A ground side controller 1004 oversees the ground side subsystems 1001, 1002, and 1003 via a network of control and monitoring data lines 1012. The ground side controller controls the power level and frequency of inverter 1002 to control the AC signal to the primary assembly 1003, as well as monitoring status and alarms. The ground side controller 1004 can also communicate with an external control and data system (not shown).

[0088] Installed in the EV, a secondary / receiving coil (Rx) and secondary compensation network 1007 sends an AC power signal to a rectifier 1008, which provides the resulting DC power to a filter and battery subsystem 1009. The EV side controller 1010 oversees the EV side subsystems 1007, 1008, and 1009 via a network of control and monitoring data lines 1013. The ground side controller 1010 also controls the switching frequency of the rectifier 1008 if it uses active rectification.

[0089] Both transmission power and duplex communication between the stationary charger and the EV pass through the air gap 1006. In this example, short-range loop antennas 1005 and 1011 are used for data signal transmission through said air gap 1006.

[0090] In such a configuration, the installation control device 1004 may include a processor programmed to implement the charging frequency determination techniques described herein.

[0091] A Station or Depot Controller 1013 is connected to the Ground Side Controller 1010. The Station Controller 1013 allows coordination between GTSs and is a local store of EV information which can include magnetic profiles. The Station Controller 1013 also acts as a gateway for secure communications for billing, diagnostics and EV log upload.

[0092] Figure 11 FIG. 11 is a flow chart showing the process of determining the charging frequency performed by the ground side control device 1004.

[0093] In an initialization step 1101, information regarding a vehicle's WPT (a profile of the vehicle) is obtained either from the vehicle or an external storage site via the wireless communications system. The WPT information may include, for each VTA, details of the make, model and manufacturer of the VTA, the magnetic properties of the VTA coil assembly, and details of previous charging experiences which may include the charging signal frequency used. If no VTA data is available, a default profile is used and a new record file is started to establish a new profile.

[0094] The controller then calculates, at a calculation step 1102, the set GTA voltage to use, the number of current measurements to take, and the frequency map (the number, order, and interval of frequencies at which to measure current).

[0095] A measuring step 1103 includes shunting the VTA coil, applying a set voltage to the corresponding GTA input, and applying an alternating current using the frequency map.

[0096] The determining step 1104 may include calculating a second measurement value 1103 with the recalculated frequency map. The determining step 1104 may include calculating an effective frequency if the resonant frequency cannot be determined within the allowable WPT bandwidth or exceeds the current sensor uncertainty threshold.

[0097] In a reporting step 1105, the measured resonant frequency or calculated effective frequency for the GTA-VTA pair is reported to the controller 1104, the vehicle, and an offset data store (if present).

[0098] In a pre-charge step 1106, the VTA coil is unshunted, the GTA input is regulated to the reported frequency, and the GTA voltage is set to supply the requested power to the VTA and thus the EV.

[0099] Figure 12 Rather than starting the efficient frequency determination procedure with a fixed GTA input voltage and varying the GTA current while the VTA coil is shunted, a more conservative method can be implemented that avoids excessively high currents in the GTA, such as if the GTA and VTA coils are unexpectedly over-coupled. Figure 12 shows scenarios for setting the GA input voltage without harming the GTA or coupled VTA-WPT electronics.

[0100] The x-axis 1201 shows the frequency of the GTA current input, and the y-axis 1202 shows the GTA coil current at varied GTA input voltages, with 1203 being the lower frequency boundary and 1204 being the upper frequency boundary. Above a current measurement uncertainty limit 1206, the GTA current is shown in response to a range of GTA input voltages at set frequencies 1205. The VTA (not shown), which corresponds geometrically to the GTA under test, has its coil shunted during testing.

[0101] With GTA and VTA aligned and the VTA coil shunted, a conservative method for arriving at a normal GTS input test voltage begins by setting the GTA input voltage to a first voltage (e.g., 0.0 volts). The GTA input voltage is increased in a series of incremental (percentage or fraction of nominal test voltage) voltage steps. Once the GTA input current exceeds the current measurement uncertainty limit 1206, at each subsequent voltage step the GTA current at the selected GTA voltage and set frequency is measured and compared to the GTA current guard threshold 1207, safety limit threshold 1208, and adverse condition threshold 1209.

[0102] GTA current measurements 1209, 1210, and 1211 obtained from a first GTA input voltage series are shown at a selected frequency 1205. The selected frequency 1205 may be selected at the lower frequency boundary 1203 to take advantage of the asymmetry of the effect of the coupling coefficient (k) on the GTA input current. This first series of voltages stops increasing when either the nominal GTS test voltage is reached or a GTA current measurement 1212 that exceeds the guard threshold 1207 is produced in response to a GTA voltage input (as shown in FIG. 12).

[0103] Once the GTA current exceeds guard threshold 1207, the GTA input voltage increment is reduced and a second series of GTA input voltages is applied and the resulting GTA input currents 1213, 1214, and 1215 are measured. The second GTA voltage series is applied incrementally until either the nominal GTS test voltage is reached or the GTA input current measurement 1216 exceeds the safety limit threshold 1208. Once the GTA input current measurement 1216 exceeds the safety limit threshold, the GTA input voltage at current measurement 121 is used as the test voltage and an efficient frequency determination procedure can begin.

[0104] In FIG. 12, no GTA input current measurements are shown that exceed the adversity threshold 1209. Such currents could damage the GTA or the corresponding VTA. Note that the safety margin threshold 1208 is set well below the adversity threshold 1209. If a GTA current measurement exceeds the adversity threshold 1209, the GTA input voltage 1209 from a previous step (where the resulting current was measured below the adversity threshold value) will be reported as the GTA test voltage to be used in the efficient frequency determination procedure.

[0105] conclusion Although various embodiments have been described above, it should be understood that they are presented by way of example and not limitation. For example, any of the elements associated with the systems and methods described above may employ any of the desired functionality described herein. Thus, the breadth and scope of the preferred embodiments should not be limited by any of the sample embodiments described above.

[0106] As discussed herein, logic, commands, or instructions implementing aspects of the methods described herein may be provided in a computer system including any number of form factors for computer systems, such as desktop or notebook personal computers, mobile devices such as tablets, netbooks, and smartphones, client terminals, and examples of server host machines. Other examples discussed herein include incorporating the techniques discussed herein into other forms of programmed logic, hardware configurations, or specialized components or modules, including apparatuses having respective means for performing the functions of such techniques. Each algorithm used to implement the functions of such techniques may include some or all of the sequences of electronic operations described herein, or other aspects depicted in the accompanying drawings and the detailed description below. Such systems and computer-readable media including instructions for implementing the methods described herein also constitute sample configurations.

[0107] The functionality described herein with respect to Figures 1-11 may be implemented in software in one configuration. The software may be comprised of computer-executable instructions stored in a computer-readable medium or computer-readable storage device, such as one or more local or networked non-transitory memories or other types of hardware-based storage devices. Furthermore, such functionality corresponds to modules, which may be software, hardware, firmware, or any combination thereof. Functions may be performed in one or more modules as desired, and the configurations described are merely examples. The software may be executed on a digital signal processor, ASIC, microprocessor, or other type of processor operating on a computer system, such as a personal computer, server, or other computer system, turning such a computer system into a specifically programmed machine.

[0108] The examples described herein may include or operate on a processor, logic, or a number of components, modules, or mechanisms (herein "modules"). A module is a tangible entity (e.g., hardware) capable of performing specified operations and may be configured or arranged in a particular manner. As an example, a circuit may be arranged (e.g., internally or with respect to external entities such as other circuits) in a specified manner as a module. In an example, one or more computer systems (e.g., stand-alone, client or server computer systems) or one or more hardware processors may be configured in whole or in part by firmware or software (e.g., instructions, application portions, or applications) as modules that operate to perform specified operations. As an example, the software may reside on a machine-readable medium. The software, when executed by the hardware underlying the module, causes the hardware to perform the specified operations.

[0109] Thus, the term "module" is understood to encompass tangible hardware and / or software entities that are physically configured, specifically configured (e.g., built into hardware), or temporarily (e.g., temporarily) configured (e.g., programmed) to operate in a specified manner or to perform some or all of the operations described herein. Considering examples in which modules are temporarily configured, each of said modules need not be instantiated at any given moment. For example, if the modules consist of a general-purpose hardware processor configured using software, the general-purpose hardware processor may be configured as each different module at different times. Accordingly, the software may configure the hardware processor, for example, to configure a particular module at one time and a different module at a different time.

[0110] Those skilled in the art will appreciate that while the disclosure contained herein relates to powering vehicles, this is only one of many possible applications and other examples are possible, including non-vehicle applications. For example, those skilled in the art will appreciate that there are numerous applications in which it is desirable to provide charging to a customer's electronic devices as the customer waits in a queue and moves through the queue. For example, inductive portable consumer electronic device chargers such as those used to charge toothbrushes, cell phones, and other devices (e.g., PowerMat®) can be managed as described herein. These and other such applications are therefore within the scope of the following claims.

Claims

1. 1. A wireless power transfer (WPT) charger, comprising: a transmitting coil for wirelessly transmitting electrical energy at a charging frequency to a receiving coil; A control device, setting an input voltage to the transmitting coil to a fixed voltage; shunting the receive coil; determining a first input current frequency range within which a resonant frequency is expected to be found; frequency polling within the first input current frequency range using a set of polling frequencies in the first input current frequency range; selecting the polling frequency having the lowest input current as the charging frequency; and executing instructions to determine the charging frequency between the transmitting coil and the receiving coil by performing operations including: A control device including a processor; A wireless power transfer (WPT) charger comprising:

2. 2. The WPT charger of claim 1, wherein the processor of the control device further comprises: setting the first input current frequency range based on at least one of regulations, international electromagnetic frequency allocations, or past experience with WPT between the transmit coil and the receive coil; determining the first input current frequency range in which the resonant frequency is expected to be found by performing operations comprising:

3. 2. The WPT charger of claim 1, wherein the processor of the control device further comprises: determining an input current minimum at said set of polling frequencies; determining a second input current frequency range about the input current minimum that is narrower than the first input current frequency range; frequency polling within the second input current frequency range using a second set of polling frequencies for the second input current frequency range; and executing instructions to poll frequencies within the first input current frequency range by performing operations comprising:

4. 2. The WPT charger of claim 1, wherein the processor of the control device further comprises: determining two polling frequencies among said set of polling frequencies that have said lowest input current; using the two polling frequencies to define upper and lower limits of the second input current frequency range; and executing instructions to determine the second input current frequency range around the input current minimum by performing operations comprising:

5. 2. The WPT charger of claim 1, wherein the processor of the control device further comprises: determining two polling frequencies adjacent to the polling frequency having the lowest input current; using the two polling frequencies to define upper and lower limits of the second input current frequency range; determining the second input current frequency range around the input current minimum by performing operations comprising:

6. 6. The WPT charger of claim 5, wherein the processor of the control device further comprises: polling within the second input current frequency range around the input current minimum by performing operations including defining the plurality of frequency maps having measurement distributions determined from a probability that the input current minimum is within a frequency range of the respective frequency map and a polling frequency within the respective frequency map.

7. 2. In the WPT charger of claim 1, if the polling frequency with the lowest input current cannot be determined beyond an input current uncertainty limit, the processor of the controller further comprises: segmenting an input current versus input current frequency response curve into a lower response curve above said input current uncertainty limit and an upper response curve above said input current uncertainty limit; selecting the highest frequency point on the lower response curve above the input current uncertainty limit as the charging frequency; A WPT charger that performs the following operations.

8. 2. In the WPT charger of claim 1, if the polling frequency with the lowest input current cannot be determined beyond an input current uncertainty limit, the processor of the controller further comprises: segmenting an input current versus frequency response curve into a lower response curve above said input current uncertainty limit and an upper response curve above said input current uncertainty limit; selecting as the charging frequency a midpoint frequency between a lower crossover frequency where the lower response curve crosses the input current uncertainty limit and an upper crossover frequency where the upper response curve crosses the input current uncertainty limit; A WPT charger that executes instructions to perform operations having the following:

9. 2. In the WPT charger of claim 1, if the polling frequency with the lowest input current cannot be determined beyond an input current uncertainty limit, the processor of the controller further comprises: segmenting an input current vs. input current frequency response curve into a lower response curve above said input current uncertainty limit and an upper response curve above said input current uncertainty limit; selecting as a charging frequency a frequency between a lower crossover frequency where a lower response curve crosses the input current uncertainty limit and a midpoint frequency between the lower crossover frequency and the upper crossover frequency where an upper response curve crosses the input current uncertainty limit, the selected frequency being an offset from the lower crossover frequency, the offset being a percentage of the distance between the lower crossover frequency and the midpoint frequency; A WPT charger that executes instructions to perform operations having the following:

10. 2. In the WPT charger of claim 1, if the polling frequency with the lowest input current cannot be determined beyond an input current uncertainty limit, the processor of the controller further comprises: segmenting an input current versus input current frequency response curve into a lower response curve above said input current uncertainty limit and an upper response curve above said input current uncertainty limit; using a first linear projection of the lower response curve based on at least two polling frequencies adjacent to the input current uncertainty limit for the lower response curve and a second linear projection of the upper response curve based on at least two polling frequencies adjacent to the input current uncertainty limit for the lower response curve to find an intersection of the first and second linear projections; selecting the frequency of the intersection as the charging frequency; A WPT charger that executes instructions to perform operations having the following:

11. 2. In the WPT charger of claim 1, if the polling frequency with the lowest input current cannot be determined beyond an input current uncertainty limit, the processor of the controller further comprises: segmenting an input current versus input current frequency response curve into a lower response curve above said input current uncertainty limit and an upper response curve above said input current uncertainty limit; fitting curves to at least three polling frequencies adjacent to the input current uncertainty limit for the lower response curve and the upper response curve to find an intersection between the curves fitted to the lower response curve and the upper response curve; selecting the frequency of the intersection as the charging frequency; A WPT charger that executes instructions to perform operations having the following:

12. 2. The WPT charger of claim 1, wherein the processor of the control device further comprises: repeating the steps of determining a first input current frequency range, polling a frequency, and selecting a charging frequency a plurality of times with a different input voltage for each iteration, the different input voltage for each iteration being a different percentage of a requested power level; determining a relationship between power level and resonant frequency at each power level; determining a resonant frequency at said desired power level from said relationship; A WPT charger that executes instructions to perform operations having the following:

13. 2. The WPT charger of claim 1, wherein the processor: taking an initial current measurement at a lower frequency boundary of the frequency range; if the first current measurement is greater than or equal to a safe current threshold for current detection, ceasing to determine a charging frequency or resuming to determine the charging frequency at a lower fixed input voltage level to the transmit coil; and executing additional instructions to avoid over-current in a transmitting coil due to over-coupling between the transmitting coil and the receiving coil by executing an operation having the following steps:

14. 2. The WPT charger of claim 1, wherein the processor: shunting the receive coil; setting the first input current frequency to a polling frequency within the first input current frequency range; incrementally varying the input voltage to the transmit coil; measuring an input current to the transmit coil at each voltage increment until the input current measurement exceeds a safety margin threshold but does not exceed an adverse condition threshold; selecting the fixed voltage as the voltage at the last voltage increment where the input current measurement falls below the safety margin threshold; and executing additional instructions to set the input voltage to the transmit coil to a fixed voltage by executing operations having the following:

15. 15. The WPT charger of claim 14, wherein the processor: using a first voltage increment until an input current of the transmit coil exceeds a guard threshold that is less than the safety margin threshold; using a second voltage increment less than the first voltage increment until the input current of the transmitting coil exceeds the safety margin threshold or a nominal test voltage is obtained; and executing additional instructions to incrementally vary the input voltage to the transmit coil by performing operations having the following:

Images