Control system for improving the efficiency of wireless charging for electric vehicles with integrated GaN driver

Abstract

This invention relates to the field of electric vehicle charging technology, specifically to a wireless charging efficiency improvement control system for electric vehicles integrating a GaN driver. The system includes a signal acquisition module, a phase processing module, a misalignment calculation module, a control modulation module, and a drive output module. The system acquires the input current of the primary coil and the output voltage of the secondary coil to form a raw electrical parameter sequence. It then extracts the instantaneous phase difference angle between the primary current and the secondary voltage using a Hilbert transform, and calculates the current resonant misalignment value of the wireless charging system. This misalignment value is then input to the built-in pulse width modulation controller of the GaN driver to dynamically adjust the on-time width of the GaN power transistor and output the adapted drive signal. This method can accurately capture dynamic changes in the resonant phase, simplify the control signal transmission link, adapt to fluctuations in charging load conditions, optimize the system's resonant matching state, and smoothly adapt to various operating scenarios of wireless charging for electric vehicles.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging technology, and more particularly to a wireless charging efficiency improvement control system for electric vehicles with integrated GaN driver. Background Technology

[0002] Wireless charging for electric vehicles is gradually becoming more widespread due to its convenience and contactless nature. However, existing wireless charging systems mostly use conventional signal sampling methods to obtain the electrical parameters of the primary and secondary coils, relying on basic algorithms to compare the current and voltage phases. This only allows them to obtain phase values ​​at fixed moments, making it difficult to capture real-time phase fluctuations under dynamic changes in charging conditions. Conventional charging control architectures deploy the pulse width modulation control unit and GaN driver independently, relying on external circuits for signal interaction. This only allows setting the GaN power transistor switching duration according to preset fixed parameters, and cannot autonomously adjust based on the system's resonant state.

[0003] Conventional phase detection methods cannot analyze the deep time-domain characteristics of electrical parameter sequences, resulting in low accuracy in phase difference detection. This makes it difficult to accurately reflect the degree of resonance offset in the wireless charging system, leading to prolonged periods of uncoordinated operation and additional energy losses during transmission. Discrete control structures suffer from lengthy signal transmission links and lag in response, failing to match the resonant operating conditions in real time with changes in the electric vehicle charging load. Furthermore, the operating states of the power transistors do not match the actual operational requirements of the system, thus limiting the adaptability and operational stability of the wireless charging system. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing an integrated GaN driver-based wireless charging efficiency improvement control system for electric vehicles.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a wireless charging efficiency improvement control system for electric vehicles integrating a GaN driver, comprising: The signal acquisition module acquires the input current signal of the primary coil and the output voltage signal of the secondary coil in the wireless charging system as the original electrical parameter sequence; The phase processing module performs a Hilbert transform on the original electrical parameter sequence to extract the instantaneous phase difference angle between the primary current and the secondary voltage; The disharmony calculation module calculates the current resonance disharmony value of the wireless charging system based on the instantaneous phase difference angle; The control modulation module inputs the current resonance misalignment value to the pulse width modulation controller integrated inside the GaN driver. The pulse width modulation controller dynamically adjusts the switching conduction time width of the GaN power transistor according to the current resonance misalignment value. The drive output module applies the adjusted switch on-time width to the drive signal output terminal of the GaN driver.

[0006] As a further aspect of the present invention, the input current signal of the primary coil and the output voltage signal of the secondary coil in the wireless charging system are collected as the original electrical parameter sequence, including: A high-frequency current transformer is installed on the primary side of the wireless charging system. The high-frequency current transformer captures the AC current waveform flowing through the primary coil at a fixed sampling frequency as the original waveform of the primary current. A resistor divider network is set on the secondary side of the wireless charging system; The resistor divider network captures the AC voltage waveform across the secondary coil at the same sampling frequency as the fixed sampling frequency, which is used as the original waveform of the secondary voltage. A low-pass filter is applied to the original waveform of the primary current to remove high-frequency noise components above the switching frequency; Perform the same low-pass filtering operation on the original waveform of the secondary voltage; The filtered primary current waveform and the filtered secondary voltage waveform are organized into the original electrical parameter sequence after being aligned with time.

[0007] As a further aspect of the present invention, performing a Hilbert transform on the original electrical parameter sequence to extract the instantaneous phase difference angle between the primary current and the secondary voltage includes: Extract the primary-side current waveform sequence from the original electrical parameter sequence; Perform a Hilbert transform on the primary-side current waveform sequence to construct a primary-side complex analytic signal; Extract the instantaneous phase angle sequence of the primary current from the primary complex analytic signal; Extract the secondary voltage waveform sequence from the original electrical parameter sequence; Perform a Hilbert transform on the secondary voltage waveform sequence to construct a secondary complex analytic signal; Extract the instantaneous phase angle sequence of the secondary voltage from the secondary complex analytic signal; Calculate the difference between the instantaneous phase angle of the secondary voltage and the instantaneous phase angle of the primary current at each identical discrete time point; The calculated difference sequence is used as the instantaneous phase difference angle.

[0008] As a further aspect of the present invention, calculating the current resonance misalignment value of the wireless charging system based on the instantaneous phase difference angle includes: Extract the phase difference angle values ​​over several consecutive complete switching cycles from the instantaneous phase difference angle; A moving average filter is applied to the phase difference angle values ​​within a series of complete switching cycles to eliminate sampling noise; Divide the phase difference angle value after moving average filtering by the rated resonant angular frequency of the wireless charging system. The result of the division operation is used as the phase difference time offset. The absolute value of the phase difference time offset is taken as the current resonance misalignment value.

[0009] As a further aspect of the present invention, the current resonant misalignment value is input to a pulse width modulation controller integrated within the GaN driver, including: Convert the current resonance misalignment value into a decimal numerical format; The current resonance misalignment value in the decimal format is transmitted to the register unit inside the GaN driver via the serial peripheral interface bus. The register unit latches the received current resonance misalignment value and provides it to the input of the pulse width modulation controller; The input terminal of the pulse width modulation controller continuously reads the current resonance misalignment value latched in the register unit.

[0010] As a further aspect of the present invention, the pulse width modulation controller dynamically adjusts the switching on-time width of the GaN power transistor according to the current resonant misalignment value, including: The pulse width modulation controller has a built-in reference on-time width value; The pulse width modulation controller reads the current resonance misalignment value latched in the register unit; The pulse width modulation controller multiplies the current resonance misalignment value by a preset scaling factor to obtain an adjustment offset. When the current resonance misalignment value is positive, the pulse width modulation controller subtracts the adjustment offset from the reference conduction time width value to obtain the new conduction time width. When the current resonance misalignment value is negative, the pulse width modulation controller adds the adjustment offset to the reference conduction time width value as the new conduction time width. When the current resonance misalignment value is zero, the pulse width modulation controller keeps the reference conduction time width value unchanged as the new conduction time width.

[0011] As a further aspect of the present invention, the adjusted switch on-time width is applied to the drive signal output terminal of the GaN driver, including: The pulse width modulation controller writes the calculated new on-time width into the period register in the drive signal generation module; The drive signal generation module reconfigures the high-level duration of the output square wave signal according to the new on-time width in the period register. The drive signal generation module maintains the low-level duration of the output square wave signal unchanged; The drive signal generation module alternately outputs high-level and low-level signals according to the reconfigured high-level duration and the original low-level duration; The high-level and low-level signals output by the drive signal generation module are directly connected to the gate terminal of the GaN power transistor.

[0012] As a further aspect of the present invention, before performing a Hilbert transform on the original electrical parameter sequence, the following is also included: Amplitude normalization is performed on the filtered primary-side current waveform and the filtered secondary-side voltage waveform. Divide the amplitude-normalized primary current waveform by the root mean square value of the primary current waveform. Divide the amplitude-normalized secondary voltage waveform by the root mean square value of the secondary voltage waveform; The normalized primary-side current waveform and the normalized secondary-side voltage waveform are reorganized into a corrected sequence of original electrical parameters. The corrected original electrical parameter sequence is used to replace the original original electrical parameter sequence for subsequent Hilbert transform operations.

[0013] As a further aspect of the present invention, after applying the adjusted switch on-time width to the drive signal output of the GaN driver, the following steps are also performed: The primary coil input current signal after the adjusted switch on-time width is re-acquired is used as the feedback current signal; The secondary coil output voltage signal after the adjusted switch on-time width is re-acquired and used as the feedback voltage signal; Perform a Hilbert transform on the feedback current signal and the feedback voltage signal to extract the instantaneous feedback phase difference angle; The instantaneous feedback phase difference angle is compared with zero. When the absolute value of the instantaneous feedback phase difference angle is less than the preset convergence threshold, the current switch conduction time width remains unchanged; When the absolute value of the instantaneous feedback phase difference angle is greater than or equal to the preset convergence threshold, the dynamic adjustment process of the switch conduction time width is repeated.

[0014] As a further aspect of the present invention, the process of repeatedly performing dynamic adjustment of the switch conduction time width also includes performing: Record the current resonance misalignment value calculated during each dynamic adjustment process; Arrange the current resonance detuning values obtained from consecutive recordings in chronological order to form a detuning value sequence. Calculate the change step between adjacent two values in the detuning value sequence. Compare the change step obtained each time with the change step obtained in the previous calculation. When the absolute value of the change step shows a monotonically decreasing trend, keep the current proportionality coefficient unchanged. When the absolute value of the change step shows a monotonically increasing trend, reduce the proportionality coefficient by half and then restart the dynamic adjustment process of the switch-on time width.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: Perform Hilbert transform on the original electrical parameter sequence composed of the input current signal of the primary coil and the output voltage signal of the secondary coil in the wireless charging system, which can deeply disassemble the time-domain waveform characteristics of the electrical parameter sequence, strip the influence of clutter interference on the phase information, carefully analyze the phase dynamic change law of the primary current and the secondary voltage, accurately obtain the instantaneous phase difference angle under the real-time change of the working condition, and can completely reflect the subtle deviation trend of the resonance state of the wireless charging system and restore the real resonance detuning change process of the system.

[0016] Import the resonance detuning value into the pulse width modulation controller integrated inside the GaN driver. Relying on the controller, the switch-on time width of the GaN power tube is changed in real time. The integrated layout eliminates the signal transfer link of the external control module, simplifies the control signal transmission path, and weakens the control deviation caused by line delay. The switch-on duration of the power tube can be adjusted synchronously with the real-time change of the resonance detuning value, adapting to the charging working condition changes brought by the driving and parking of electric vehicles and load fluctuations, making the GaN power tube always match the working timing of the resonance operation of the wireless charging system and adapting to the system operation state change law under different charging scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is the timing diagram of the electric vehicle wireless charging efficiency improvement control system integrating GaN driver according to the present invention; Figure 2 It is the flowchart of the acquisition of the original electrical parameter sequence; Figure 3 It is the flowchart of the extraction of the instantaneous phase difference angle. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] See Figure 1 A wireless charging efficiency improvement control system for electric vehicles integrating a GaN driver is disclosed, the overall implementation scheme of which is as follows: The system includes a signal acquisition module, a phase processing module, a misalignment calculation module, a control modulation module, and a drive output module. The signal acquisition module acquires the input current signal of the primary coil and the output voltage signal of the secondary coil in the wireless charging system as the raw electrical parameter sequence. The phase processing module performs a Hilbert transform on the raw electrical parameter sequence to extract the instantaneous phase difference angle between the primary current and the secondary voltage. The misalignment calculation module calculates the current resonant misalignment value of the wireless charging system based on the instantaneous phase difference angle. The control modulation module inputs the current resonant misalignment value to a pulse width modulation controller integrated within the GaN driver, which dynamically adjusts the switching on-time width of the GaN power transistor according to the current resonant misalignment value. The drive output module applies the adjusted switching on-time width to the drive signal output terminal of the GaN driver.

[0021] In one embodiment of the present invention, see [reference] Figure 2A high-frequency current transformer is installed on the primary side of the wireless charging system. This high-frequency current transformer captures the AC current waveform flowing through the primary coil at a fixed sampling frequency, serving as the original primary current waveform. A resistor divider network is installed on the secondary side of the wireless charging system. This resistor divider network captures the AC voltage waveform across the secondary coil at the same sampling frequency as the fixed sampling frequency, serving as the original secondary voltage waveform. A low-pass filter is applied to the original primary current waveform to remove high-frequency noise components above the switching frequency. The same low-pass filter is applied to the original secondary voltage waveform. The filtered primary current waveform and the filtered secondary voltage waveform are then time-aligned and organized into an original electrical parameter sequence. Before performing a Hilbert transform on the original electrical parameter sequence, amplitude normalization is applied to both the filtered primary current waveform and the filtered secondary voltage waveform. The amplitude-normalized primary current waveform is divided by its root mean square (RMS) value, and the amplitude-normalized secondary voltage waveform is also divided by its RMS value. The normalized primary-side current waveform and the normalized secondary-side voltage waveform are reorganized into a corrected original electrical parameter sequence. This corrected original electrical parameter sequence is then used to replace the original original electrical parameter sequence for subsequent Hilbert transform operations.

[0022] In the specific implementation, a high-frequency current transformer is set on the primary side of the wireless charging system. This high-frequency current transformer captures the AC current waveform flowing through the primary coil at a fixed sampling frequency of 1 MHz as the original waveform of the primary current. A resistor voltage divider network is set on the secondary side of the wireless charging system. This resistor voltage divider network captures the AC voltage waveform across the secondary coil at the same 1 MHz sampling frequency as the original waveform of the secondary voltage. A low-pass filter is performed on the original waveform of the primary current to remove high-frequency noise components above the switching frequency. The cutoff frequency of the low-pass filter is set to 1.2 times the switching frequency. The same low-pass filter is performed on the original waveform of the secondary voltage. The filtered primary current waveform and the filtered secondary voltage waveform are then aligned in time and organized into an original electrical parameter sequence.

[0023] In some embodiments, amplitude normalization processing is performed on the filtered primary-side current waveform and the filtered secondary-side voltage waveform. The amplitude-normalized primary-side current waveform is divided by the root mean square (RMS) value of the primary-side current waveform, and the amplitude-normalized secondary-side voltage waveform is divided by the RMS value of the secondary-side voltage waveform. The normalization expression is as follows: in: Represents at discrete time points The normalized instantaneous value of the primary current. Represents at discrete time points The instantaneous value of the primary current after filtering. This represents the total number of sampling points contained in a complete sampling period. Indicates the first Each sampling time point; the normalization of the secondary voltage uses the same expression, only replacing the current sign with the voltage sign. Optionally, the normalized primary current waveform and the normalized secondary voltage waveform are reorganized into a corrected original electrical parameter sequence. The corrected original electrical parameter sequence contains data from two channels: the first channel is the normalized primary current waveform, and the second channel is the normalized secondary voltage waveform. It can be understood that the corrected original electrical parameter sequence replaces the original original electrical parameter sequence for subsequent Hilbert transform operations. Each discrete time point in the corrected original electrical parameter sequence simultaneously contains the normalized primary current value and the normalized secondary voltage value.

[0024] In another specific implementation, the magnetic core of the high-frequency current transformer is made of nickel-zinc ferrite material, the turns ratio of the high-frequency current transformer is set to 50:1, and the resistive voltage divider network is composed of two metal film resistors with a precision of 1% connected in series, wherein the upper voltage divider resistor has a resistance of 10 kΩ and the lower voltage divider resistor has a resistance of 1 kΩ.

[0025] In some embodiments, the low-pass filtering operation is implemented using a fourth-order Butterworth active filter. The passband gain of the fourth-order Butterworth active filter is set to 0 dB. The cutoff frequency of the fourth-order Butterworth active filter is dynamically configured according to the switching frequency of the wireless charging system. When the switching frequency is 85 kHz, the cutoff frequency can be set to 102 kHz. When the filtered primary-side current waveform and the filtered secondary-side voltage waveform are aligned in time, the same clock source is used to drive the analog-to-digital converters on the primary and secondary sides. The frequency of the clock source is set to 10 times the fixed sampling frequency. It is understood that the deviation between the sampling time of the primary-side current and the sampling time of the secondary-side voltage is less than one sampling period through a clock synchronization counter. The original electrical parameter sequence is stored in the form of data frames. Each data frame contains 1024 discrete time points of primary-side current value and corresponding 1024 discrete time points of secondary-side voltage value. The two columns of values ​​correspond one-to-one on the time axis.

[0026] In one embodiment of the present invention, see [reference] Figure 3The process involves extracting the primary-side current waveform sequence from the original electrical parameter sequence, performing a Hilbert transform on the primary-side current waveform sequence to construct a primary-side complex analytic signal, and then extracting the instantaneous phase angle sequence of the primary-side current from the primary-side complex analytic signal. Similarly, the process involves extracting the secondary-side voltage waveform sequence from the original electrical parameter sequence, performing a Hilbert transform on the secondary-side voltage waveform sequence to construct a secondary-side complex analytic signal, and then extracting the instantaneous phase angle sequence of the secondary-side voltage from the secondary-side complex analytic signal. At each identical discrete time point, the difference between the instantaneous phase angle of the secondary-side voltage and the instantaneous phase angle of the primary-side current is calculated, and the calculated difference sequence is used as the instantaneous phase difference angle. In a specific implementation, the primary-side current waveform sequence is extracted from the corrected original electrical parameter sequence, and the primary-side current waveform sequence is represented as follows: ,in For sequence length, For each equally spaced discrete time point, perform a Hilbert transform on the primary current waveform sequence to construct the primary complex analytic signal. The discrete form of the Hilbert transform is: in: Represents at discrete time points The Hilbert transform result of the primary current waveform sequence. and All are discrete time point indices, when The value is 0 when the time is taken; the complex analytic signal of the primary side is constructed as follows: ,in The imaginary unit is used to extract the instantaneous phase angle sequence of the primary current from the primary complex analytic signal. ,in The range of values ​​is In some embodiments, a secondary-side voltage waveform sequence is extracted from the corrected original electrical parameter sequence, and the secondary-side voltage waveform sequence is represented as follows: ,in Represents at discrete time points The normalized instantaneous value of the secondary voltage is obtained; a Hilbert transform is performed on the secondary voltage waveform sequence to construct the complex analytic signal of the secondary side; the Hilbert transform result of the secondary voltage waveform sequence is obtained. The complex analytic signal of the secondary side is constructed as follows: Extracting the instantaneous phase angle sequence of the secondary voltage from the secondary complex analytic signal. .

[0027] Optionally, at each identical discrete time point The difference between the instantaneous phase angle of the secondary voltage and the instantaneous phase angle of the primary current is calculated using the following formula: The calculated difference sequence As an instantaneous phase difference angle, it can be understood when the primary current is normalized to a certain value. Or the normalized value of the secondary voltage When the value is zero, the corresponding instantaneous phase angle is calculated using the four-quadrant arctangent function. The return value of the four-quadrant arctangent function is located in... In another specific implementation within the interval, the Hilbert transform is implemented using a finite impulse response (FIR) filter. The FIR filter is set to order 51, and its coefficients are generated using a window function design method, with a Hamming window selected as the window function. In some embodiments, before extracting the instantaneous phase angle sequence of the primary current, a phase unwrapping operation is performed on the primary complex analytic signal. The phase unwrapping operation will exceed... Phase transitions within a range are achieved through addition and subtraction. The compensation is performed as an integer multiple of the phase angle, making the instantaneous phase angle sequence a continuous function; the phase unwinding operation is also performed on the instantaneous phase angle sequence of the secondary voltage.

[0028] Optionally, each value in the instantaneous phase difference angle sequence is expressed in radians. The instantaneous phase difference angle sequence retains the positive and negative signs; a positive sign indicates that the secondary voltage phase leads the primary current phase, and a negative sign indicates that the secondary voltage phase lags behind the primary current phase. This can be understood as discrete time points... and The time interval between them is equal to the reciprocal of the fixed sampling frequency. When the fixed sampling frequency is 1 MHz, the time interval between adjacent discrete time points is 1 microsecond. The time resolution of adjacent values ​​in the instantaneous phase difference angle sequence is also 1 microsecond.

[0029] In one embodiment of the present invention, phase difference angle values ​​within a series of consecutive complete switching cycles are extracted from the instantaneous phase difference angle. A moving average filter is then applied to these phase difference angle values ​​within the series of consecutive complete switching cycles to eliminate sampling noise. The phase difference angle values ​​after moving average filtering are divided by the rated resonant angular frequency of the wireless charging system, and the result of the division is used as the phase difference time offset. The absolute value of the phase difference time offset is used as the current resonant misalignment value. The current resonant misalignment value is converted to a decimal value format and transmitted to a register unit inside the GaN driver via a serial peripheral interface bus. This register unit latches the received current resonant misalignment value and provides it to the input of the pulse width modulation controller. The input of the pulse width modulation controller continuously reads the current resonant misalignment value latched in the register unit. In a specific implementation, phase difference angle values ​​within a series of consecutive complete switching cycles are extracted from the instantaneous phase difference angle sequence. The number of switching cycles is set to 10, and each switching cycle contains... A discrete time point, It equals the fixed sampling frequency divided by the switching frequency. The fixed sampling frequency is 1 MHz and the switching frequency is 85 kHz. The value is 12. After adjustment, only 5 sampling points are retained in each switching cycle. The extracted phase difference angle values ​​are arranged in matrix form according to time order. The rows correspond to the switching cycle number, and the columns correspond to the phase difference angle sampling points in each switching cycle. See Table 1.

[0030] Table 1: Sample table of phase difference angle values ​​over 10 consecutive complete switching cycles In some embodiments, a moving average filter is performed on the phase difference angle values ​​over several consecutive complete switching cycles to eliminate sampling noise, and the moving window length is set to the number of sampling points included in one switching cycle. and The value is 5, and the moving average filter calculation formula is: in: Indicates the first The average phase difference angle of the output from the second moving average filter. This indicates the number of switching cycles participating in the average, and its value is 10. Indicates the switching cycle number. This indicates the sampling point number within each switching cycle. Indicates the first Within the first sliding window The first switching cycle The instantaneous phase difference angle values ​​at each sampling point are used. The sliding window moves forward by an optional switching cycle length each time, and the phase difference angle values ​​after moving average filtering are then analyzed. Divide by the rated resonant angular frequency of the wireless charging system The phase difference time offset is obtained. ,in , The rated resonant frequency of the wireless charging system is set to 85 kHz; the phase difference time offset is... The absolute value is taken as the current resonance misalignment value. In another specific implementation, the current resonance misalignment value is... The value is converted to decimal format, retaining 6 significant digits after the decimal point. This current resonant misalignment value is then transmitted via the serial peripheral interface bus to the register within the GaN driver. The serial peripheral interface bus is set to mode 0, with a clock frequency of 10 MHz and a data width of 24 bits, where the most significant bit is the sign bit and the remaining 23 bits represent the magnitude of the misalignment. In some embodiments, the register unit latches the received current resonant misalignment value and provides it to the input of the pulse width modulation controller. The register unit uses a rising edge triggered latching method, and the latching signal is generated by the falling edge of the chip select signal of the serial peripheral interface bus. The input of the pulse width modulation controller continuously reads the current resonant misalignment value latched in the register unit. The read operation is performed once at the beginning of each switching cycle, and the read time is aligned with the rising edge of the synchronous clock signal of the GaN driver. It can be understood that when the instantaneous phase difference angle... When positive, the phase difference time offset A positive value indicates that the secondary voltage phase leads the primary current phase; when the instantaneous phase difference angle is... Phase difference time offset when it is negative A negative value indicates that the secondary voltage phase lags behind the primary current phase; the misalignment value... Always a non-negative number, selectable, rated resonant angular frequency. The data is pre-stored in non-volatile memory, which shares the same power domain as the pulse width modulation controller. After system power-on, the non-volatile memory will... The value is loaded into the internal read-only register of the pulse width modulation controller.

[0031] In one embodiment of the present invention, the pulse width modulation (PWM) controller internally incorporates a reference conduction time width value. The PWM controller reads the current resonance misalignment value latched in the register unit. The PWM controller multiplies the current resonance misalignment value by a preset scaling factor to obtain an adjustment offset. When the current resonance misalignment value is positive, the PWM controller subtracts the adjustment offset from the reference conduction time width value to obtain the new conduction time width. When the current resonance misalignment value is negative, the PWM controller adds the adjustment offset to the reference conduction time width value to obtain the new conduction time width. When the current resonance misalignment value is zero, the PWM controller keeps the reference conduction time width value unchanged as the new conduction time width. The PWM controller writes the calculated new conduction time width into the period register in the drive signal generation module. The drive signal generation module reconfigures the high-level duration of the output square wave signal according to the new conduction time width in the period register.

[0032] The drive signal generation module maintains the low-level duration of the output square wave signal unchanged, and alternately outputs high-level and low-level signals according to the reconfigured high-level duration and the original low-level duration. The high-level and low-level signals output by the drive signal generation module are directly connected to the gate of the GaN power transistor. In a specific implementation, the pulse width modulation controller internally incorporates a reference on-time width value. The unit is nanoseconds; the pulse width modulation controller reads the current resonance misalignment value latched in the register unit. The unit is seconds; the pulse width modulation controller will determine the current resonant misalignment value. Multiplied by a preset scaling factor We obtain an adjustment offset: in: It is a dimensionless constant with a value of 0.5; when the current resonance misalignment value is... When the value is positive, the pulse width modulation controller will set the reference on-time width value. Subtract adjustment offset As a new conduction time width When the current resonance misalignment value When the value is negative, the pulse width modulation controller will set the reference on-time width value. Add adjustment offset As a new conduction time width When the current resonance misalignment value To maintain the reference on-time width value for the zero-time pulse width modulation controller The new on-time width remains unchanged. .

[0033] In some embodiments, the pulse width modulation controller will calculate the new on-time width. The data is written to the period register in the drive signal generation module. The period register is a 16-bit register with a unit resolution of 1 nanosecond. The write operation is completed through the internal bus, and the write cycle is one clock cycle. The drive signal generation module reconfigures the high-level duration of the output square wave signal according to the new on-time width in the period register. The high-level duration is set to equal to... The drive signal generation module maintains the low level of the output square wave signal for a certain duration. constant, The preset duration is 500 nanoseconds; the drive signal generation module follows the reconfigured high-level duration. and the original low level duration Alternating between high and low level outputs, the length of one complete switching cycle is... Optionally, the high and low level signals output by the drive signal generation module are directly connected to the gate of the GaN power transistor using differential pair routing. A 5-ohm resistor is connected in series at the gate to suppress ringing. After each new conduction time width is written to the period register, the drive signal generation module starts outputting a new waveform at the next rising edge at the end of a switching cycle, avoiding the generation of incomplete pulses during the writing process. In another specific implementation, the scaling factor... The value is determined in advance through offline calibration. During the calibration process, the wireless charging system is placed under rated load conditions, and the optimal conduction time width corresponding to different discontinuity values ​​is measured. The measurement data is then fitted into a linear relationship and calculated. Refer to Table 2, which provides the adjustment offset and new conduction time width values ​​corresponding to different current resonance misalignment values.

[0034] Table 2: Comparison Table of Resonance Discoordination Values ​​and On-Time Width Adjustment In some embodiments, the reference conduction time width value Set to 100 nanoseconds, corresponding to the optimal conduction time width when the wireless charging system is fully resonant; adjust the offset. The calculation formula is ,in This represents the absolute value of the current resonant misalignment, converted to nanoseconds and then compared with... Keep units consistent. This can be understood as the current resonance misalignment value... A positive value indicates that the system is in a capacitive detuned state, requiring a reduction in the conduction time width to compensate for phase lead; when the current resonant detuning value is... A negative value indicates that the system is in an inductively detuned state, requiring an increase in the conduction time width to compensate for the phase lag; the new conduction time width It is limited to a minimum of 10 nanoseconds and a maximum of 200 nanoseconds, enforced by the upper limit comparator and lower limit comparator inside the pulse width modulation controller.

[0035] Optionally, the high-level signal output by the drive signal generation module has an amplitude of 5 volts, and the low-level signal has an amplitude of 0 volts. The threshold voltage of the GaN power transistor is 1.5 volts, and the high-level duration directly controls the conduction time of the GaN power transistor. During the reconfiguration of the high-level duration of the output square wave signal, the state machine inside the drive signal generation module enters a waiting state. After the current switching cycle is completed, it enters an update state, loads the new value in the cycle register into the timer comparison unit, and returns to the running state to start outputting a new waveform. It can be understood that the low-level duration of the output square wave signal... Maintaining a constant frequency means that the switching frequency changes slightly with the duration of the high level, but the range of change is limited to within ±5 kHz of 85 kHz. The resonant circuit of the wireless charging system can still maintain high power transmission efficiency within this frequency range.

[0036] In one embodiment of the present invention, after applying the adjusted switch on-time width to the drive signal output of the GaN driver, the primary coil input current signal after applying the adjusted switch on-time width is re-acquired as the feedback current signal, and the secondary coil output voltage signal after applying the adjusted switch on-time width is re-acquired as the feedback voltage signal. A Hilbert transform is performed on the feedback current signal and feedback voltage signal to extract the feedback instantaneous phase difference angle, and the feedback instantaneous phase difference angle is compared with zero. When the absolute value of the feedback instantaneous phase difference angle is less than a preset convergence threshold, the current switch on-time width remains unchanged; when the absolute value of the feedback instantaneous phase difference angle is greater than or equal to the preset convergence threshold, the dynamic adjustment process of the switch on-time width is repeated. During the repeated dynamic adjustment of the switch on-time width, the current resonance misalignment value calculated in each dynamic adjustment process is recorded, and the current resonance misalignment values ​​recorded several times consecutively are arranged in chronological order into a misalignment value sequence. The change step size between two adjacent values ​​in the misalignment value sequence is calculated, and the change step size calculated each time is compared with the change step size calculated previously. When the absolute value of the step size shows a monotonically decreasing trend, the current proportional coefficient remains unchanged. When the absolute value of the step size shows a monotonically increasing trend, the proportional coefficient is reduced by half and the dynamic adjustment process of the switch conduction time width is restarted.

[0037] In specific implementations, after applying the adjusted switch conduction time width to the drive signal output of the GaN driver, the primary coil input current signal after applying the adjusted switch conduction time width is re-acquired as the feedback current signal, and the secondary coil output voltage signal after applying the adjusted switch conduction time width is re-acquired as the feedback voltage signal. The acquisition method of the feedback current signal and feedback voltage signal is the same as the method of acquiring the original electrical parameter sequence in the signal acquisition module, that is, using the same high-frequency current transformer and the same resistor voltage divider network, and the sampling frequency is also set to 1 MHz. In some embodiments, Hilbert transform is performed on the feedback current signal and feedback voltage signal to extract the feedback instantaneous phase difference angle. The execution steps of the Hilbert transform are the same as the steps of performing the Hilbert transform on the original electrical parameter sequence in the phase processing module, that is, first constructing the complex analytic signal of the feedback current and the complex analytic signal of the feedback voltage, then extracting the instantaneous phase angle sequence of the feedback current and the instantaneous phase angle sequence of the feedback voltage respectively, and finally calculating the difference between the instantaneous phase angle of the feedback voltage and the instantaneous phase angle of the feedback current at each same discrete time point as the feedback instantaneous phase difference angle sequence. .

[0038] Optionally, the instantaneous feedback phase difference angle is compared with zero, and the comparison operation is performed at each discrete time point. Upper judgment Whether it is valid, among which The preset convergence threshold is 0.01 radians; when the absolute value of the instantaneous phase difference angle is fed back... Less than the preset convergence threshold While maintaining the current switch conduction time width unchanged, when the absolute value of the instantaneous phase difference angle is fed back... Greater than or equal to the preset convergence threshold The dynamic adjustment process of the switch on-time width is repeatedly executed. This dynamic adjustment process includes recalculating the current resonant misalignment value, regenerating the new on-time width, and rewriting it to the period register. In another specific implementation, during the repeated execution of the dynamic adjustment process of the switch on-time width, the current resonant misalignment value calculated in each dynamic adjustment is recorded. ,in Index indicating the number of adjustments, Arrange the current resonance misalignment values ​​obtained from several consecutive recordings into a misalignment value sequence in chronological order. ,in The sequence length is 5; calculate the step size of the change between two adjacent values ​​in the sequence of discoconstant values. ,in The step size calculated each time is compared with the step size calculated in the previous time. The comparison formula is as follows: in: Indicates the first The absolute value of the step size of the nth change and the nth The ratio of the absolute values ​​of the step sizes of each change, dimensionless; Indicates the first value in the sequence of uncoordinated quantities The number and the first The step size between the values, in seconds; Indicates the first value in the sequence of uncoordinated quantities The number and the first The step size between the values, in seconds; and These represent the absolute values ​​of the corresponding change step sizes, in seconds; The index for the variable step size, with a value greater than or equal to 2.

[0039] It is understandable that when the absolute value of the step size exhibits a monotonically decreasing trend, that is, for all... satisfy At this point, the current ratio coefficient is maintained. Unchanged; when the absolute value of the step size shows a monotonically increasing trend, that is, for all satisfy At this point, the proportionality coefficient will be... After reducing it by half, the dynamic adjustment process of the switch conduction time width is restarted, and the proportional coefficient update formula is as follows: After restarting execution, the sequence of uncoordinated values ​​is cleared and recording begins from the beginning. In some embodiments, when repeatedly performing the dynamic adjustment process of the switch on-time width, after each adjustment, 10 complete switching cycles are waited before the feedback current signal and feedback voltage signal are re-acquired. The waiting time is used to allow the transient process of the wireless charging system to stabilize; a preset convergence threshold is set. Stored in the read-only memory inside the pulse width modulation controller, it cannot be changed after factory calibration. Optional: the length of the misalignment value sequence. Controlled by a counter inside the pulse width modulation controller, the counter increments by 1 each time a new current resonance misalignment value is recorded. When the counter reaches a certain value... The step size calculation logic is triggered when the time is right. After the calculation is completed, the counter is automatically reset to zero and the earliest value in the sequence is cleared to maintain the sequence length. It is understandable that when the absolute value of the step size is neither monotonically decreasing nor monotonically increasing, i.e., the ratio... The pulse width modulation controller maintains the current proportional coefficient, which alternates between values ​​greater than 1 and less than 1. The proportional coefficient remains unchanged, and the system continues to perform the normal dynamic adjustment process of the switch conduction time width without performing the proportional coefficient reduction operation.

[0040] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A wireless charging efficiency improvement control system for electric vehicles integrating a GaN driver, characterized in that, include: The signal acquisition module acquires the input current signal of the primary coil and the output voltage signal of the secondary coil in the wireless charging system as the original electrical parameter sequence; The phase processing module performs a Hilbert transform on the original electrical parameter sequence to extract the instantaneous phase difference angle between the primary current and the secondary voltage; The disharmony calculation module calculates the current resonance disharmony value of the wireless charging system based on the instantaneous phase difference angle; The control modulation module inputs the current resonance misalignment value to the pulse width modulation controller integrated inside the GaN driver. The pulse width modulation controller dynamically adjusts the switching conduction time width of the GaN power transistor according to the current resonance misalignment value. The drive output module applies the adjusted switch on-time width to the drive signal output terminal of the GaN driver; The input current signal of the primary coil and the output voltage signal of the secondary coil in the wireless charging system are collected as the original electrical parameter sequence, including: A high-frequency current transformer is installed on the primary side of the wireless charging system. The high-frequency current transformer captures the AC current waveform flowing through the primary coil at a fixed sampling frequency as the original waveform of the primary current. A resistor divider network is set on the secondary side of the wireless charging system; The resistor divider network captures the AC voltage waveform across the secondary coil at the same sampling frequency as the fixed sampling frequency, which is used as the original waveform of the secondary voltage. A low-pass filter is applied to the original waveform of the primary current to remove high-frequency noise components above the switching frequency; Perform the same low-pass filtering operation on the original waveform of the secondary voltage; The filtered primary-side current waveform and the filtered secondary-side voltage waveform are organized into the original electrical parameter sequence after being aligned with time. Performing a Hilbert transform on the original electrical parameter sequence to extract the instantaneous phase difference angle between the primary current and the secondary voltage includes: Extract the primary-side current waveform sequence from the original electrical parameter sequence; Perform a Hilbert transform on the primary-side current waveform sequence to construct a primary-side complex analytic signal; Extract the instantaneous phase angle sequence of the primary current from the primary complex analytic signal; Extract the secondary voltage waveform sequence from the original electrical parameter sequence; Perform a Hilbert transform on the secondary voltage waveform sequence to construct a secondary complex analytic signal; Extract the instantaneous phase angle sequence of the secondary voltage from the secondary complex analytic signal; Calculate the difference between the instantaneous phase angle of the secondary voltage and the instantaneous phase angle of the primary current at each identical discrete time point; The calculated difference sequence is used as the instantaneous phase difference angle.

2. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 1, characterized in that, The current resonance misalignment value of the wireless charging system is calculated based on the instantaneous phase difference angle, including: Extract the phase difference angle values ​​over several consecutive complete switching cycles from the instantaneous phase difference angle; A moving average filter is applied to the phase difference angle values ​​within a series of complete switching cycles to eliminate sampling noise; Divide the phase difference angle value after moving average filtering by the rated resonant angular frequency of the wireless charging system. The result of the division operation is used as the phase difference time offset. The absolute value of the phase difference time offset is taken as the current resonance misalignment value.

3. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 2, characterized in that, The current resonant misalignment value is input to the pulse width modulation controller integrated within the GaN driver, including: Convert the current resonance misalignment value into a decimal numerical format; The current resonance misalignment value in the decimal format is transmitted to the register unit inside the GaN driver via the serial peripheral interface bus. The register unit latches the received current resonance misalignment value and provides it to the input of the pulse width modulation controller; The input terminal of the pulse width modulation controller continuously reads the current resonance misalignment value latched in the register unit.

4. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 3, characterized in that, The pulse width modulation controller dynamically adjusts the switching on-time width of the GaN power transistor according to the current resonant misalignment value, including: The pulse width modulation controller has a built-in reference on-time width value; The pulse width modulation controller reads the current resonance misalignment value latched in the register unit; The pulse width modulation controller multiplies the current resonance misalignment value by a preset scaling factor to obtain an adjustment offset. When the current resonance misalignment value is positive, the pulse width modulation controller subtracts the adjustment offset from the reference conduction time width value to obtain the new conduction time width. When the current resonance misalignment value is negative, the pulse width modulation controller adds the adjustment offset to the reference conduction time width value as the new conduction time width. When the current resonance misalignment value is zero, the pulse width modulation controller keeps the reference conduction time width value unchanged as the new conduction time width.

5. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 4, characterized in that, The adjusted switch on-time width is applied to the drive signal output of the GaN driver, including: The pulse width modulation controller writes the calculated new on-time width into the period register in the drive signal generation module; The drive signal generation module reconfigures the high-level duration of the output square wave signal according to the new on-time width in the period register. The drive signal generation module maintains the low-level duration of the output square wave signal unchanged; The drive signal generation module alternately outputs high-level and low-level signals according to the reconfigured high-level duration and the original low-level duration; The high-level and low-level signals output by the drive signal generation module are directly connected to the gate terminal of the GaN power transistor.

6. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 5, characterized in that, Before performing the Hilbert transform on the original electrical parameter sequence, the following steps are also included: Amplitude normalization is performed on the filtered primary-side current waveform and the filtered secondary-side voltage waveform. Divide the amplitude-normalized primary current waveform by the root mean square value of the primary current waveform. Divide the amplitude-normalized secondary voltage waveform by the root mean square value of the secondary voltage waveform; The normalized primary-side current waveform and the normalized secondary-side voltage waveform are reorganized into a corrected sequence of original electrical parameters. The corrected original electrical parameter sequence is used to replace the original original electrical parameter sequence for subsequent Hilbert transform operations.

7. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 1, characterized in that, After applying the adjusted switch on-time width to the drive signal output of the GaN driver, the process also includes: The primary coil input current signal after the adjusted switch on-time width is re-acquired is used as the feedback current signal; The secondary coil output voltage signal after the adjusted switch on-time width is re-acquired and used as the feedback voltage signal; Perform a Hilbert transform on the feedback current signal and the feedback voltage signal to extract the instantaneous feedback phase difference angle; The instantaneous feedback phase difference angle is compared with zero. When the absolute value of the instantaneous feedback phase difference angle is less than the preset convergence threshold, the current switch conduction time width remains unchanged; When the absolute value of the instantaneous feedback phase difference angle is greater than or equal to the preset convergence threshold, the dynamic adjustment process of the switch conduction time width is repeated.

8. The electric vehicle wireless charging efficiency improvement control system with integrated GaN driver according to claim 7, characterized in that, The process of repeatedly performing dynamic adjustments to the switch on-time width also includes executing: Record the current resonance misalignment value calculated during each dynamic adjustment process; Arrange the current resonance misalignment values ​​obtained from several consecutive records into a misalignment value sequence in chronological order; Calculate the step size of change between two adjacent values ​​in the sequence of disharmony values; Compare the change step size obtained in each calculation with the change step size obtained in the previous calculation; When the absolute value of the step size shows a monotonically decreasing trend, the current proportional coefficient remains unchanged. When the absolute value of the change step size shows a monotonically increasing trend, the proportional coefficient is reduced by half and the dynamic adjustment process of the switch conduction time width is restarted.

Images