Title Page

ABSTRACT

Contents

Nomenclature 12

Chapter 1. Introduction 13

1.1. Research Motivation 14

1.2. Necessity of phase synchronization and magnitude control in multiple transmitter systems 16

1.3. Research objectives 18

1.4. Thesis outline 18

Chapter 2. Phase synchronization and magnitude control with interference between multiple TXs in wireless power transfer 20

2.1. Introduction 20

2.2. Desired operating point and its difficulties 23

2.2.1. Problems of achieving |ITX₁/ITX₂|＝M₁/M₂ ＜INTC when MTX exists[이미지참조] 26

2.2.2. Problems of tracking |ITX₁|:|ITX₂|＝M₁ :M₂ with MTX existence[이미지참조] 26

2.3. Proposed phase and magnitude control 29

2.3.1. The negative and positive modes 29

2.3.2. Negative mode operation 29

2.3.3. Positive mode operation 31

2.3.4. Implementation details 32

2.4. Measurement 35

2.5. Conclusion 41

Chapter 3. Coupling estimation and fast maximum efficiency tracking in multi-transmitter WPT 43

3.1. Introduction 43

3.2. Coupling extraction in multiple TXs scenario and maximum efficiency tracking 49

3.2.1. Coupling estimation 49

3.2.2. RX load and power estimation within TX 54

3.2.3. Maximum efficiency tracking and power regulation for multiple TXs without iteration or communication 55

3.2.4. Comparison with prior works 58

3.3. Measurements 60

3.4. Conclusion 65

Chapter 4. Summary and conclusion 66

4.1. Summary 66

4.2. Conclusion 66

Appendix: Automatic Resonance Tuning with ON/OFF Soft Switching for Push-Pull Parallel-Resonant Inverter in Wireless Power Transfer 69

Introduction. 69

Problems in conventional techniques 71

Problems of detuning in push-pull inverter 71

Drawbacks of conventional switch-controlled capacitor 71

Proposed tuning capacitor and control method 73

Proposed switching operations 73

Zero voltage turn-on and low dv/dt turn off 74

Proposed control 75

Analysis of duty cycle and effective capacitance 75

Measurements 79

Conclusion 84

References 85

국문초록 94

Curriculum Vitae 96

Table 3.1.1. Comparison with Prior Works. 44

Table 3.3.1. Component Parameters 59

Table A.1. COMPARISON AT THE SAME 200W LOAD AND COIL PARAMETERS 83

Fig. 1.1.1. Fundamental diagram of WPT system 14

Fig. 1.2.1. (a) The couplings are modelled as reflected resistances Rrefl and ZINT,TX . Phase imbalance between ITX₁ and ITX₂ causes negative resistance Re{ZINT,TX1 } at TX1 and positive...[이미지참조] 16

Fig. 1.2.2. (a) Power transfer efficiency when the phase difference exists between ITX₁ and ITX₂ while k₁/k₂＝|ITX₁|/|ITX₂|=0.68. (b) Power transfer efficiency at different coil current ratios at k₁/k₂＝0.68 where...[이미지참조] 17

Fig. 2.1.1. Two TX coil currents, ITX₁ and ITX₂, should be in-phase, while their magnitude ratio should follow the coupling ratio of M₁/M₂. Unfortunately, the coupling between TXs, MTX,...[이미지참조] 21

Fig. 2.1.2. (a) Overlapped coil structure to avoid coupling between TXs. Susceptible to mechanical error, and installation becomes complex. (b) Side-by-side TX coils used in this... 22

Fig. 2.2.1. Two transmitters coupled with a single receiver. TX2 is strongly coupled while TX1 is weakly coupled. The mutual coupling between TX coils, MTX, exists.[이미지참조] 24

Fig. 2.2.2. ITX phase and magnitude using (3) when RX is more leant to TX#1. M₁/M₂＝0.13. (a) ITX phase difference as a function of magnitude and phase of Vinv....[이미지참조] 25

Fig. 2.3.1. TX1 negative mode for ITX₁/ITX₂ lower than interference current of (4). (a) Controller action and circuit response. The relationship between voltage and current is reversed....[이미지참조] 28

Fig. 2.3.2. The positive mode for ITX₁/ITX₂ and ITX₂/ITX₁ higher than interference current (4). (a) Controller action and circuit response. The relationship between voltage phase (magnitude)...[이미지참조] 30

Fig. 2.3.3. Implementation of the proposed control. When the target|ITX₁/ITX₂| ratio is lower than interference current (INTC) of (4), TX1 and TX2 are set to negative and positive mode,...[이미지참조] 32

Fig. 2.3.4. Interference current has negligible dependency on RX misalignment (M₁ and M₂ value) and load resistance. Interference current value is decided mainly by MTX and XTX which...[이미지참조] 33

Fig. 2.3.5. Experiment setup with three TX coils. 35

Fig. 2.3.6. (a) Waveform of Mosfet Q2 and Q6 at RX location 12.85cm. (b) ZVS situation across different RX positions. Even if ZVS fails for TX#1 at 0~12.8cm, its switching loss is... 36

Fig. 2.3.7. (a) TX2's Vinv₂ and ITX₂. (b) TX1's Vinv1 and ITX₁. At RX location of 12.8cm. Both the (a) and (b) are measured simultaneously with a shared triggering signal to oscilloscopes in order...[이미지참조] 37

Fig. 2.3.8. (a) Conventional TX when TX1 duty is set to zero. (b) Proposed, M₁/M₂＝0.34. (c) Proposed method, M₁/M₂＝0.2. 37

Fig. 2.3.9. Coil currents ITX₁ and ITX₂ follows the coupling ratio when RX is gradually swept from the TX2 (0cm) to the midpoint between TX1 and TX2 (21.8cm).[이미지참조] 38

Fig. 2.3.10. Response when the output power is changed from 200W to 100W. 38

Fig. 2.3.11. Accuracy of the coupling coefficient tracking. 39

Fig. 2.3.12. Comparison of efficiencies of conventional systems and proposed system when RX is swept from 0cm to a midpoint between TX#1 and TX#2. 39

Fig. 2.3.13. Measurement with 3 TXs. (a) RX is at TX#1. (b) RX is between TX#1 and TX#2. (c) RX is at TX#2. (d) Efficiency comparisons. 40

Fig. 3.1.1. Existing approaches for MET. (a) P&O method. (b) Multiple operating modes method. (c) TX-RX communication channel method. (d) Additional... 43

Fig. 3.2.1. (a) Equivalent circuit. ZRef,n is reflected impedance from RX. ZRef,n does not physically present in circuit. (b) Equivalent model of RX coupled with multiple TXs. (c)...[이미지참조] 49

Fig. 3.2.2. Block diagram of proposed method. 57

Fig. 3.3.1. Experimental setup with 3 TXs. 58

Fig. 3.3.2. ITX,₁ -ITX,₃, VRL, VDD₁-VDD₃, and V rect waveforms when RX is moving from -33cm to 32.8cm at 6.84 km/h velocity.[이미지참조] 59

Fig. 3.3.3. Waveforms of TXs coil currents, high-side mosfet's source voltages and low-side mosfet's gate voltages. (a) At RX location -25.2cm. (b) At RX location -16.75cm. (ITX,₁, ITX,₂ :...[이미지참조] 60

Fig. 3.3.4. Estimated coupling values matches the actual coupling values. 61

Fig. 3.3.5. ITX,₁ -ITX,₃ values measured in comparison with calculated ITX,₁ -ITX,₃ optimum values.[이미지참조] 62

Fig. 3.3.6. Proposed method tracks the Rrect,opt that maximizes the overall efficiency.[이미지참조] 62

Fig. 3.3.7. Efficiency comparison of conventional system and proposed MET when RX is swept from -33cm to 32.8cm. The ITX level of 1TX conventional setup is chosen to guarantee... 63

Fig. 3.3.8. Step Change of RX power. MCU adjusts the TX coil current and settles the output in 3.8ms. (a) Load power change from 150W to 200W. (ITX,₁ : 10A/div, IRL : 2A/div, and Vrect :...[이미지참조] 64

Fig. A.1. Conventional push-pull parallel-resonant inverter. (a) Schematic, (b) When LTX is detuned to low value. The effective output voltage of inverter is low, and the voltage stress...[이미지참조] 70

Fig. A.2. Proposed tuning control for parallel-resonant inverter. (a) When Mpwm₁ is ON, LTX resonates both with CTX and Cpwm₁. Vd₁ and comparator's output Vcomp (b) at perfect resonant...[이미지참조] 72

Fig. A.3. Waveforms of the proposed autotuning inverter. 74

Fig. A.4. Switching diagram of the proposed PWM tuning for current-fed resonant inverter. (a) Before t₁ . (b) t₁-t₂. ZVS turning ON of Mpwm₁, which does not conduct yet. (c) t₂-t₃, deadtime...[이미지참조] 76

Fig. A.5. ZVS fails for Mpwm₁ if VGpwm₁ is turned on after the falling edge of Vg₁.[이미지참조] 77

Fig. A.6. (a) Equivalent circuit for effective capacitance calculation, and waveforms. (b) Ceff/Cpwm ratio with respect to duty cycle of VGpwm.[이미지참조] 78

Fig. A.7. Experimental setup. (a) TX inverter, RX rectifier, sensor, and MCU boards, (b) Variation of LTX and coupling k with respect to TX-RX distance.[이미지참조] 79

Fig. A.8. Waveform comparison between conventional and proposed. TX-RX distance is set to 6cm. (Vd₁ : 50V/div, Vg₁ : 5V/div, VGmpwm₁ : 5v/div and Time: 2.00µs/div). With the same...[이미지참조] 80

Fig. A.9. TX-RX distance is set to 4cm. (a) RX power is limited to 181.02W (b) Proposed with tuning. RX power is improved to 200W. 80

Fig. A.10. VGpwm₁ should be turned on before the falling edge of Vg₁ to ensure ZVS. I Cpwm1 starts to flow as Vg₁ turns off. (Vd₁ : 50V/div, Vg₁ : 5V/div, VGpwm₁ : 5V/div, ICpwm₁ : 4A/div and...[이미지참조] 81

Fig. A.11. Dynamic tuning of the proposed control system. (Vcomp : 500mV/div, Vg₁ : 5V/div and VGpwm₁ : 5V/div) devices.[이미지참조] 81

Fig. A.12. RX power and TX-to-RX efficiency comparison between the conventional and the proposed method. (a) RX power with respect to distance. (b) TX-to-RX efficiency with respect... 82