Misalignment-Tolerant Series Hybrid with Active Adjustable Constant Current and Constant Voltage Output Wireless Charging System

: This paper presents a series hybrid wireless charging system with an active adjustable circuitry offering constant current and constant voltage output characteristics. The series hybrid system consists of the inductor–capacitor–capacitor ( LCC ) and series-series (SS) networks are used for improving charging pad misalignment tolerance. An active switch is employed to provide an adjustable CC and CV output for different battery charging stages. To demonstrate the performance of the proposed method, a 310 W prototype was built. A systematic optimization in the parameter of the proposed topology to achieve relative constant output was analyzed within a certain range of the designed operating region. The experimental results indicate that the output current ﬂuctuation is less than 5% with load variations, and the output voltage ﬂuctuation is less than 5% with load varying from 19 to 70 Ω , as the pick-up pads misaligned within 50% of the pad outer diameter. pad moves out of the operating region. Besides, the function switch is employed to transfer CC output to CV output without complicated control schemes. Moreover, a parameter optimization design method is presented to provide high misalignment tolerance. The experimental results demonstrate that the system can maintain the output current ﬂuctuation to be less than 5% with the load varying from 15 to 18 Ω , and the output voltage ﬂuctuation less than 5% with the load varying from 19 to 70 Ω , when the pick-up pads are within 50% misalignment. The results demonstrate the theoretical analysis, and indicate that the series hybrid and conﬁgurable wireless charging system offers a reliable solution to wireless EV charging applications. manuscript.


Introduction
Wireless power transfer (WPT) systems using an alternating magnetic field to transfer power across a relatively large air gap have been adopted in numerous industrial and commercial applications. In comparison to traditional plug-in charging technology, WPT is safe offering galvanic isolation, maintenance free with dirt and moisture environments [1][2][3][4][5][6], and has great potential to be used in, for example, electric vehicle (EV) charging, portable electronics, autonomous underwater vehicles (AUVs), and implantable biomedical devices [7][8][9][10][11], etc.
As for practical battery charging applications, the equivalent resistance of the battery varies significantly when considering the entire charging process. Thus, charging with a wide range of loading conditions to achieve the constant current (CC) output and constant voltage (CV) output is essential for most of the li-ion battery charging applications [12]. Specific to wireless charging scenarios, misalignment between magnetic couplers results in variation of the self and mutual inductance, and poses a negative impact on the current, voltage outputs, and transfer efficiency. Thus, the motivation of this paper is to design a WPT system, which is able to provide CC output and CV output with a wide range of loads and high misalignment tolerance between the primary and pickup magnetic couplers.
In the past few years, many approaches have been investigated to realize the desired current or voltage under different loading conditions for a WPT system. The DC-DC converter at the primary side or the secondary side is employed to regulate the current or voltage output [13,14]. However, this additional cascading DC-DC converter will result in extra converter loss and cost, and need extra space for setting up the device. Variable frequency control and phase shift control of the WPT system can also be used to realize current and protect the inverter, it may significantly reduce the system efficiency and reliability.
This paper presents a novel series hybrid and configurable compensation topology to achieve CC and CV outputs with high misalignment tolerance. Additionally, the main contributions of this paper are summarized as follows: (1) The proposed series hybrid compensation topology can achieve load-independent CC output, and realize operation without pickup side coils, which can overcome the aforementioned drawbacks associated with [34], and limit the current of the inverter without extra controllers. In addition, a set of parameter optimization is proposed to improve the constancy in output power when misalignment occurs.
(2) A variable structure T-circuit is employed to transfer CC output to CV output without an additional cascade converter, and owns a native load-independent character, which can simplify the control schemes.
Specifically, Section 2 analyzes the mathematical model of the series hybrid and configurable topology. The changes in the inductances of the DDQ pads between primary and pick-up sides, and the parameter optimization are presented in Section 3. In Section 4, a 310 W prototype that was built to verify the theoretical analysis is described. Finally, the conclusion is drawn in Section 5. Figure 1 shows the proposed series hybrid and configurable topology. The high frequency inverter and full-bridge rectifier consists of four MOSFETs (Q 1 -Q 4 ) and four diodes (D 1 -D 4 ), respectively. An inductor L 0 and capacitors C 0 and C 1 (L 5 , C 5 and C 2 ) constitute the LCC-LCC compensation topology, while capacitor C 3 (C 4 ) is part of the SS compensation topology. LCC-LCC and SS topologies of the primary and pick-up sides are both connected in series, together making up the LCC-S series hybrid topology, and the configurable topology is formed with inductors L f1 and L f2 and capacitor C f , together with switches S 1 and S 2 . M 12 and M 34 are the main mutual inductances, while M 13 , M 14 , M 23 , and M 24 are the cross mutual inductances. The analysis of the proposed system is based on the fundamental harmonic approximation method and the influence of the higher harmonics is ignored. alignment. However, when the pick-up pads move out of the operating region, the current of the parallel SS compensation network in the primary side may increase too much to break down the inverter, which is caused by the decrease in reflected impedance due to the decreasing coupling. Although using a closed-loop controller can suppress this increase current and protect the inverter, it may significantly reduce the system efficiency and reliability. This paper presents a novel series hybrid and configurable compensation topology to achieve CC and CV outputs with high misalignment tolerance. Additionally, the main contributions of this paper are summarized as follows:

Theoretical Analysis
(1) The proposed series hybrid compensation topology can achieve load-independent CC output, and realize operation without pickup side coils, which can overcome the aforementioned drawbacks associated with [34], and limit the current of the inverter without extra controllers. In addition, a set of parameter optimization is proposed to improve the constancy in output power when misalignment occurs.
(2) A variable structure T-circuit is employed to transfer CC output to CV output without an additional cascade converter, and owns a native load-independent character, which can simplify the control schemes.
Specifically, Section II analyzes the mathematical model of the series hybrid and configurable topology. The changes in the inductances of the DDQ pads between primary and pick-up sides, and the parameter optimization are presented in Section III. In Section IV, a 310 W prototype that was built to verify the theoretical analysis is described. Finally, the conclusion is drawn in Section V. Figure 1 shows the proposed series hybrid and configurable topology. The high frequency inverter and full-bridge rectifier consists of four MOSFETs (Q1-Q4) and four diodes (D1-D4), respectively. An inductor L0 and capacitors C0 and C1 (L5, C5 and C2) constitute the LCC-LCC compensation topology, while capacitor C3 (C4) is part of the SS compensation topology. LCC-LCC and SS topologies of the primary and pick-up sides are both connected in series, together making up the LCC-S series hybrid topology, and the configurable topology is formed with inductors Lf1 and Lf2 and capacitor Cf, together with switches S1 and S2. M12 and M34 are the main mutual inductances, while M13, M14, M23, and M24 are the cross mutual inductances. The analysis of the proposed system is based on the fundamental harmonic approximation method and the influence of the higher harmonics is ignored.  Figure 2 illustrates the series hybrid topology, which is driven by a resonant inverter with angular frequency ω. In order to tune the LCC-S series hybrid compensation net-   work in the primary and pick-up sides to the constant resonant frequency, the resonant parameters are designed to satisfy the following equations: REVIEW 4 of 19 work in the primary and pick-up sides to the constant resonant frequency, the resonant parameters are designed to satisfy the following equations: Figure 2. Equivalent circuit of the series hybrid topology.

Analysis of the Series Hybrid Topology
According to Figure 2, Kirchhoff's voltage law is used to describe the relationship of voltages in the series hybrid circuit; therefore, we can obtain: where: In order to simplify the analysis, only the main couplings (M12 and M34) are taken into consideration, while the cross couplings (M13, M14, M23, and M24) can be ignored by designing proper coupling structures, which will be discussed in Section III. Hence, the current of the inverter, the current of the L1 transmitter, the current of the L2 receiver, and the current of the RAB load are obtained as:  According to Figure 2, Kirchhoff's voltage law is used to describe the relationship of voltages in the series hybrid circuit; therefore, we can obtain: where: In order to simplify the analysis, only the main couplings (M 12 and M 34 ) are taken into consideration, while the cross couplings (M 13 , M 14 , M 23 , and M 24 ) can be ignored by designing proper coupling structures, which will be discussed in Section 3. Hence, the current of the inverter, the current of the L 1 transmitter, the current of the L 2 receiver, and the current of the R AB load are obtained as: Then, according to the current . I 0 of the inverter, the input equivalent impedance Z in of the system can be deduced as: According to (3) and (4), the current . I 0 is in the same phase with the inverter output voltage . U in , and the input impedance of the system is pure resistance. When the whole Energies 2021, 14, 7594 5 of 17 system works in the full tuned condition, the input reactive power is zero, which can improve the transmission efficiency of the system. In addition, the current . I 0 and the input equivalent impedance Z in clearly indicate that the inverter current is close to zero when the pick-up sides move far away, which means the mutual inductance is close to zero. Therefore, the series hybrid system can avoid the extreme situation that the inverter output current is too high to burn down due to the no-load operation of the parallel SS structure, and improve the reliability of the system.
Equation (3) shows that the output current . I 3 is load independent and lags the inverter output voltage by 90 • . Figure 3 shows the equivalent circuit of the configurable topology is driven by a current source, because the series hybrid topology circuit can obtain load-independent constant current output. When the function switch S 1 is ON and S 2 is OFF, the system provides CC output, and the output current . I CD is shown as:

Analysis of the Configurable Topology
.
Zin of the system can be deduced as: According to (3) and (4), the current 0 I  is in the same phase with the inverter output voltage in U  , and the input impedance of the system is pure resistance. When the whole system works in the full tuned condition, the input reactive power is zero, which can improve the transmission efficiency of the system. In addition, the current 0 I  and the input equivalent impedance Zin clearly indicate that the inverter current is close to zero when the pick-up sides move far away, which means the mutual inductance is close to zero. Therefore, the series hybrid system can avoid the extreme situation that the inverter output current is too high to burn down due to the no-load operation of the parallel SS structure, and improve the reliability of the system. Equation (3) shows that the output current 3 I  is load independent and lags the inverter output voltage by 90°. Figure 3 shows the equivalent circuit of the configurable topology is driven by a current source, because the series hybrid topology circuit can obtain load-independent constant current output. When the function switch S1 is ON and S2 is OFF, the system provides CC output, and the output current CD I  is shown as: The input equivalent impedance Zcc-AB is expressed as:

Analysis of the Configurable Topology
When the S1 is OFF and S2 is ON, the system provides CV output, as is shown in Figure 4. The relationship between inductors Lf1 and Lf2 and capacitor Cf satisfies the following equations, i.e., 0 Then, according to Thevenin-Norton theorems, the input voltage shown in Figure 5 can be described by: The input equivalent impedance Z cc-AB is expressed as: When the S 1 is OFF and S 2 is ON, the system provides CV output, as is shown in Figure 4. The relationship between inductors L f 1 and L f 2 and capacitor C f satisfies the following equations, i.e., Equation (6) indicates that the inductor Lf2 and capacitor Cf form the resonant network tank; therefore, the output voltage is: The input equivalent impedance Zcv-AB is expressed as:  Then, according to Thevenin-Norton theorems, the input voltage shown in Figure 5 can be described by:   Figure 6 shows the equivalent circuit with the series hybrid and the configurable topology. Comparing Figure 1 with Figure 6, the capacitor CE is used to substitute the inductor L0 (L5) and capacitor C3 (C4). The component CE can be expressed by:

The Proposed Combination of Series Hybrid and Configurable Topology
or: From (3) and (5), when the system operates in CC output, the output current is: Additionally, the corresponding input impedance Zin-cc is expressed by: Equation (6) indicates that the inductor L f 2 and capacitor C f form the resonant network tank; therefore, the output voltage is: The input equivalent impedance Z cv-AB is expressed as: Figure 6 shows the equivalent circuit with the series hybrid and the configurable topology. Comparing Figure 1 with Figure 6, the capacitor C E is used to substitute the inductor L 0 (L 5 ) and capacitor C 3 (C 4 ). The component C E can be expressed by:

The Proposed Combination of Series Hybrid and Configurable Topology
or: From (3) and (8), when the system operates in CV output, the output voltage is: Additionally, the corresponding input impedance Zin-cv is expressed by: ( ) According to (12) and (14), the value of L0L5/M12 will increase, while the value of M34 will decrease, when the main mutual inductances M12 and M34 decrease with pad misalignment. Thus, the sum of L0L5/M12 and M34 can remain relatively constant within a certain range of misalignment when the parameters are properly designed. Then, a constant current and voltage output with pad misalignment can be achieved.
From (13) and (15), it illustrates that the input impedance of the system is pure resistance when the series hybrid and configurable topology works in CC output and CV output, which means the output voltage and current of the inverter can achieve a zero phase angle (ZPA).

Coupler Design
According to (3), constant current output can be achieved only when the cross cou- .
Additionally, the corresponding input impedance Z in-cc is expressed by: From (3) and (8), when the system operates in CV output, the output voltage is: .
Additionally, the corresponding input impedance Z in-cv is expressed by: According to (12) and (14), the value of L 0 L 5 /M 12 will increase, while the value of M 34 will decrease, when the main mutual inductances M 12 and M 34 decrease with pad misalignment. Thus, the sum of L 0 L 5 /M 12 and M 34 can remain relatively constant within a certain range of misalignment when the parameters are properly designed. Then, a constant current and voltage output with pad misalignment can be achieved.
From (13) and (15), it illustrates that the input impedance of the system is pure resistance when the series hybrid and configurable topology works in CC output and CV output, which means the output voltage and current of the inverter can achieve a zero phase angle (ZPA).

Coupler Design
According to (3), constant current output can be achieved only when the cross couplings (M 13 , M 14 , M 23 , and M 24 ) are too small to be negligible. Therefore, only the main mutual inductances M 12 and M 34 are taken into consideration when the magnetic coupler is in misalignment. Additionally, the DDQ and DD structures can satisfy these desired requirements as discussed in [32,33]. Hence, the DDQ structure is used and shown in Figure 7. The Q pad structure is formed by the transmitter L 1 and receiver L 2 , while the DD pad structure is formed by the transmitter L 3 and receiver L 4 . Meanwhile, the misalignment occurrence between the primary and pick-up pads is unavoidable in the charging system, including X-axis misalignment, Y-axis misalignment, and Z-axis misalignment. All the mutual inductances of the DDQ coils are measured when the pick-up pads are moved along the X-axis, Y-axis, and Z-axis, separately.   As is shown in Figure 8, the main mutual inductances M 12 and M 34 decrease apparently with the increase of X-axis and Z-axis misalignments, while the cross coupling mutual inductances are too small to be neglected, because the Q pad and the DD pad are symmetrically placed in the primary and pick-up side, so that the amount of the magnetic flux that flows into the Q (DD) pad equals the flows out of it. However, when Y-axis misalignment occurs, the main mutual inductances M 12 and M 34 and the cross coupling mutual inductances M 23 and M 14 vary significantly, so the proposed system cannot achieve CC output and CV output in the Y-axis misalignment.

Parameter Optimization Design
In accordance with the analysis in Section 2, designing proper compensation parameters is of great significance to achieve the relative constant output current and voltage within a certain range of misalignment. In this article, a parameter optimization design method based on inductance L 0 and L 5 is proposed to ensure relative constant current output of the system.

Parameter Optimization Design
In accordance with the analysis in Section II, designing p rameters is of great significance to achieve the relative const voltage within a certain range of misalignment. In this article, a design method based on inductance L0 and L5 is proposed to current output of the system. According to Figure 8a of the X-axis misalignment curves, M12 and M34 can be approximately regarded as a linear function: 34 12 where a and b are the coefficients. When the structural paramete system, such as the material, size, and spacing height of the coil, the change trend of main mutual inductances M12 and M34, the pa be recalculated, respectively. From Figure 8a of the X-axis misalignment curves, the calcu are 0.52 and 2.17 × 10 −6 , respectively, and the variation range of ance M12 is [14 uH, 30 uH], then the output current can be expres To simplify the analysis, it is assumed that the parameter and L5 are equal. Figure 9 shows the output current curves of t According to Figure 8a of the X-axis misalignment curves, the relationship between M 12 and M 34 can be approximately regarded as a linear function: where a and b are the coefficients. When the structural parameters of the DDQ coil of the system, such as the material, size, and spacing height of the coil, are changed, resulting in the change trend of main mutual inductances M 12 and M 34 , the parameters a and b need to be recalculated, respectively. From Figure 8a of the X-axis misalignment curves, the calculated parameters a and b are 0.52 and 2.17 × 10 −6 , respectively, and the variation range of the main mutual inductance M 12 is [14 uH, 30 uH], then the output current can be expressed as: To simplify the analysis, it is assumed that the parameter values of inductance L 0 and L 5 are equal. Figure 9 shows the output current curves of the system with different inductors L 0 . It can be found that when the inductor L 0 decreases, the output current of the system will increase. In addition, when the main mutual inductance M 12 decreases, the system output current shows a trend of increasing first and then decreasing. In this article, the maximum output current of the system is designed to be 4 A, and the allowable deviation of the current is 5%. In other words, the area in the red region of Figure 9 can meet the constant current output under the condition of 50% X-axis offset. Therefore, the inductors L 0 and L 5 are selected as 16 uH.
According to the measured self-inductance L 1 and L 2 of Q coils, self-inductance L 3 and L 4 of DD coils, system resonant angular frequency ω, and inductance L 0 and L 5 with optimized parameters, the parameters of components, such as capacitors C 0 , C 1 , C 2 , C 3 , C 4 , and C 5 , can be obtained from (1).
inductors L0. It can be found that when the inductor L0 decreases, the output current of the system will increase. In addition, when the main mutual inductance M12 decreases, the system output current shows a trend of increasing first and then decreasing. In this article, the maximum output current of the system is designed to be 4 A, and the allowable deviation of the current is 5%. In other words, the area in the red region of Figure 9 can meet the constant current output under the condition of 50% X-axis offset. Therefore, the inductors L0 and L5 are selected as 16 uH. According to the measured self-inductance L1 and L2 of Q coils, self-inductance L3 and L4 of DD coils, system resonant angular frequency ω, and inductance L0 and L5 with optimized parameters, the parameters of components, such as capacitors C0, C1, C2, C3, C4, and C5, can be obtained from (1). Figure 10 shows the proposed series hybrid and configurable wireless charging system. The inverter switching devices Q1-Q4 MOSFETs use C2M0080120 and the function switching devices S1-S2 MOSFETs use SPW47N60C3. An electronic load is used to verify the performance of the CC and CV output. The system parameters are listed in Table 1.  Figure 10 shows the proposed series hybrid and configurable wireless charging system. The inverter switching devices Q 1 -Q 4 MOSFETs use C2M0080120 and the function switching devices S 1 -S 2 MOSFETs use SPW47N60C3. An electronic load is used to verify the performance of the CC and CV output. The system parameters are listed in Table 1.   Figure 11 plots the measured output current varying with the loads and X-axis misalignment. Within 140 mm X-axis misalignment, the output current of the system is between 3.85 and 4.25 A, meeting the requirement of 5% deviation. Under the same misalignment condition, the output current of 15 Ω is largest, and the output current of 18 Ω slightly decreased. If the load becomes lighter, the current fluctuation will exceed the limitation of 5%. In addition, the system output current is minimum when the receiving coil is offset to 140 mm. Additionally, it is clearly found that the load current increases first and then decreases with the increase of the offset distance, which is beneficial to improve the anti-offset ability of the system.

Experimental Results and Discussion
As is shown in Figure 12, the output voltage Uin and the output current Iin of the in-   Figure 11 plots the measured output current varying with the loads and X-axis misalignment. Within 140 mm X-axis misalignment, the output current of the system is between 3.85 and 4.25 A, meeting the requirement of 5% deviation. Under the same misalignment condition, the output current of 15 Ω is largest, and the output current of 18 Ω slightly decreased. If the load becomes lighter, the current fluctuation will exceed the limitation of 5%. In addition, the system output current is minimum when the receiving coil is offset to 140 mm. Additionally, it is clearly found that the load current increases first and then decreases with the increase of the offset distance, which is beneficial to improve the anti-offset ability of the system.  Figure 11. Measured output current varying with the RL loads and X-axis misalignment. Figure 11. Measured output current varying with the R L loads and X-axis misalignment.
As is shown in Figure 12, the output voltage U in and the output current I in of the inverter are almost in the same phase, which indicates that near zero reactive power is achieved. Additionally, the output voltage U L and output current I L , indicate that the system can achieve a relative constant current output within 50% X-axis misalignment.
The function switches consist of two anti-series-connected MOSFETs, which are shown in Figure 13. When the S 1 switch is ON and the S 2 switch is OFF, the system works in CC mode. Otherwise, the system works in CV mode. Figure 14 shows the sudden transient waveforms. There are some oscillations when the charging mode changes from CC to CV output. In CC mode, the output current is around 4 A, and the output voltage is about 72 V in CV mode. In addition, the system only takes about 4 ms to reach the new steady state. Figure 11 plots the measured output current varying with the loads and Xalignment. Within 140 mm X-axis misalignment, the output current of the syst tween 3.85 and 4.25 A, meeting the requirement of 5% deviation. Under the sa lignment condition, the output current of 15 Ω is largest, and the output curren slightly decreased. If the load becomes lighter, the current fluctuation will ex limitation of 5%. In addition, the system output current is minimum when the coil is offset to 140 mm. Additionally, it is clearly found that the load current first and then decreases with the increase of the offset distance, which is ben improve the anti-offset ability of the system.
As is shown in Figure 12, the output voltage Uin and the output current Iin verter are almost in the same phase, which indicates that near zero reactive achieved. Additionally, the output voltage UL and output current IL, indicate system can achieve a relative constant current output within 50% X-axis misalig The function switches consist of two anti-series-connected MOSFETs, w shown in Figure 13. When the S1 switch is ON and the S2 switch is OFF, the syste in CC mode. Otherwise, the system works in CV mode. Figure 14 shows the transient waveforms. There are some oscillations when the charging mode chan CC to CV output. In CC mode, the output current is around 4 A, and the output v about 72 V in CV mode. In addition, the system only takes about 4 ms to reach steady state.

CV Output Performance of the System
As is shown in Figure 15, the output voltage of the system is between 68.5 and 75.5 V within 140 mm X-axis misalignment, meeting the requirement of 5% deviation, when the loads vary from 19 to 70 Ω. In addition, if the load becomes lighter, the voltage fluctuation will exceed the limitation of 5%. Additionally, the system output voltage climbs to the peak when the receiving coil is offset to 80 mm, which is consistent with the trend of the output current.

CV Output Performance of the System
As is shown in Figure 15, the output voltage of the system is between 68.5 and 75.5 V within 140 mm X-axis misalignment, meeting the requirement of 5% deviation, when the loads vary from 19 to 70 Ω. In addition, if the load becomes lighter, the voltage fluctuation will exceed the limitation of 5%. Additionally, the system output voltage climbs to the peak when the receiving coil is offset to 80 mm, which is consistent with the trend of the output current.

CV Output Performance of the System
As is shown in Figure 15, the output voltage of the system is between 68.5 and 75.5 V within 140 mm X-axis misalignment, meeting the requirement of 5% deviation, when the loads vary from 19 to 70 Ω. In addition, if the load becomes lighter, the voltage fluctuation will exceed the limitation of 5%. Additionally, the system output voltage climbs to the peak when the receiving coil is offset to 80 mm, which is consistent with the trend of the output current.   Figure 16 shows that there is a small phase angle between the voltage and the current of the inverter, which indicates that ZVS is achieved within 50% X-axis misalignment. Additionally, the output voltage U L and output current I L indicate that the system can achieve relative constant voltage output within 50% X-axis misalignment. Figure 17 shows the experiment waveforms when the pick-up sides moved far away. The output voltage of the system is zero, and the current of the inverter is near zero, which indicates that the proposed system can work without the pick-up sides, and increase the stability of the system without any extra controllers. Figure 16 shows that there is a small phase angle between the voltage and th rent of the inverter, which indicates that ZVS is achieved within 50% X-axis mis ment. Additionally, the output voltage UL and output current IL indicate that the s can achieve relative constant voltage output within 50% X-axis misalignment.  Figure 17 shows the experiment waveforms when the pick-up sides mov away. The output voltage of the system is zero, and the current of the inverter i  zero, which indicates that the proposed system can work without the pick-up sides, and increase the stability of the system without any extra controllers.  Figure 18 shows that the efficiency in CC mode is between 91.0% and 89.2%, while the efficiency in CV mode is between 90.5% and 85.2%. In addition, the efficiency drops with the increase of the X-axis misalignment, because the increasing current flows into the L1 coil, resulting in increasing corresponding conductance loss.

Comparison with Other Methods
The performance of the proposed approach was compared with methods using control schemes and compensation topologies, as listed in Table 2. Compared with methods using control schemes, such as [14,16,17], the proposed WPT system can realize CC-CV output without an additional control scheme. Compared with methods using variable compensation topologies, such as [19,20], the proposed WPT system can both realize CC-CV output and misalignment tolerance. Among [31][32][33], the hybrid topologies are employed to improve misalignment tolerance. However, they can only achieve CC or CV output. Compared with [34], the proposed WPT system uses fewer components, and can operate without the pickup side, which improves the reliability of the system. Therefore, the proposed WPT system is superior to the other approaches in terms of CC Figure 17. Experimental waveforms of U in , I in , U L , I LCC when the pick-up sides moved far away (mutual inductance M 12 and M 34 are zero) (U in : 50 V/div; I in : 1 A/div; U L : 50 V/div; I LCC : 5 A/div; t: 5 us). Figure 18 shows that the efficiency in CC mode is between 91.0% and 89.2%, while the efficiency in CV mode is between 90.5% and 85.2%. In addition, the efficiency drops with the increase of the X-axis misalignment, because the increasing current flows into the L 1 coil, resulting in increasing corresponding conductance loss. zero, which indicates that the proposed system can work without the pick-up sides, and increase the stability of the system without any extra controllers.  Figure 18 shows that the efficiency in CC mode is between 91.0% and 89.2%, while the efficiency in CV mode is between 90.5% and 85.2%. In addition, the efficiency drops with the increase of the X-axis misalignment, because the increasing current flows into the L1 coil, resulting in increasing corresponding conductance loss.

Comparison with Other Methods
The performance of the proposed approach was compared with methods using control schemes and compensation topologies, as listed in Table 2. Compared with methods using control schemes, such as [14,16,17], the proposed WPT system can realize CC-CV output without an additional control scheme. Compared with methods using variable compensation topologies, such as [19,20], the proposed WPT system can both realize CC-CV output and misalignment tolerance. Among [31][32][33], the hybrid topologies are employed to improve misalignment tolerance. However, they can only achieve CC or CV output. Compared with [34], the proposed WPT system uses fewer components, and can operate without the pickup side, which improves the reliability of the system. Therefore, the proposed WPT system is superior to the other approaches in terms of CC

Comparison with Other Methods
The performance of the proposed approach was compared with methods using control schemes and compensation topologies, as listed in Table 2. Compared with methods using control schemes, such as [14,16,17], the proposed WPT system can realize CC-CV output without an additional control scheme. Compared with methods using variable compensation topologies, such as [19,20], the proposed WPT system can both realize CC-CV output and misalignment tolerance. Among [31][32][33], the hybrid topologies are employed to improve misalignment tolerance. However, they can only achieve CC or CV output. Compared with [34], the proposed WPT system uses fewer components, and can operate without the pickup side, which improves the reliability of the system. Therefore, the proposed WPT system is superior to the other approaches in terms of CC and CV output, misalignment tolerance, and the ability of operating without a pickup pad.