Power conversion system

Abstract

To suppress the bias of the magnetic core in magnetic components such as isolation transformers connected between bridge circuits. [Solution] A power conversion system comprising a control device that causes a first bridge circuit to output the primary AC voltage by comparing a first pulse width command for determining the pulse width of the primary AC voltage applied to the primary side of an isolation transformer with a first carrier wave, and a control device that causes a second bridge circuit to output the secondary AC voltage by comparing a second pulse width command for determining the pulse width of the secondary AC voltage applied to the secondary side of the isolation transformer with a second carrier wave, wherein the control device corrects the first pulse width command or the second pulse width command to reduce the DC component of the excitation current of the isolation transformer and the DC component of the load current obtained by subtracting the excitation current from the primary AC current flowing on the primary side of the isolation transformer or the secondary AC current flowing on the secondary side of the isolation transformer.

Description

Technical Field

[0001] The present disclosure relates to a power conversion system.

Background Art

[0002] Conventionally, as a type of DC / DC converter, a bidirectional isolated DC / DC converter (DAB converter) called DAB (Dual Active Bridge) is known. This DAB converter has a configuration in which two single-phase bridge circuits are connected via a high-frequency isolation transformer (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in a conventional DAB converter, when a voltage including a DC component is applied to a magnetic component such as an isolation transformer connected between bridge circuits, the magnetic core in the magnetic component connected between the bridge circuits may be magnetically biased, and power transmission may become impossible.

[0005] An object of the present disclosure is to suppress magnetic biasing of a magnetic core in a magnetic component such as an isolation transformer connected between bridge circuits.

Means for Solving the Problems

[0006] The present disclosure is an isolation transformer having a magnetic core, a first bridge circuit connected to the primary side of the isolation transformer and having a plurality of switching elements, a second bridge circuit connected to the secondary side of the isolation transformer and having a plurality of switching elements, The system includes a control device that, by comparing a first pulse width command for determining the pulse width of the primary AC voltage applied to the primary side of the isolation transformer with a first carrier wave, causes the primary AC voltage to be output to the first bridge circuit, and by comparing a second pulse width command for determining the pulse width of the secondary AC voltage applied to the secondary side of the isolation transformer with a second carrier wave, causes the secondary AC voltage to be output to the second bridge circuit, The control device provides a power conversion system that reduces the DC component of the excitation current of the isolation transformer and the DC component of the load current obtained by subtracting the excitation current from the primary AC current flowing on the primary side of the isolation transformer or the secondary AC current flowing on the secondary side of the isolation transformer, by correcting the first pulse width command or the second pulse width command. [Effects of the Invention]

[0007] According to this disclosure, it is possible to suppress the bias of the magnetic core in magnetic components such as isolation transformers connected between bridge circuits. [Brief explanation of the drawing]

[0008] [Figure 1] This is a circuit diagram showing one example configuration of a power conversion system according to the first embodiment. [Figure 2] This timing chart illustrates the waveform of the first drive pulse supplied by the control device to the primary drive circuit and the switching waveform of the primary switching element. [Figure 3] This is a block diagram showing an example of a control device configuration. [Figure 4] This is a block diagram showing the first example of a correction processing unit within a control device. [Figure 5] This is a block diagram showing a second example of a correction processing unit within a control device. [Figure 6] This figure shows an example of the operating waveform of the power conversion system according to the first embodiment. [Figure 7]This figure shows an example where the control device corrects the primary side U-phase pulse width command P1U based on the DC component (bias) of the excitation current, thereby widening the negative pulse width of the primary side AC voltage V1. [Figure 8] This figure shows an example where the control device corrects the primary side V-phase pulse width command P1V based on the DC component (bias) of the excitation current, thereby widening the negative pulse width of the primary side AC voltage V1. [Figure 9] This is a block diagram showing a third example of a correction processing unit within a control device. [Figure 10] This figure illustrates an example where the control device corrects the primary side U-phase pulse width command P1U based on the DC component (bias) of the excitation current, and corrects the primary side V-phase pulse width command P1V based on the DC component of the load current, thereby changing the negative pulse width of the primary side AC voltage V1. [Figure 11] This is a block diagram showing a fourth example of a correction processing unit within a control device. [Figure 12] This is a block diagram showing the fifth example of a correction processing unit within a control device. [Figure 13] This is a block diagram showing an example configuration of a control device including a current controller. [Figure 14] This timing chart illustrates the correctable timing in continuous current mode. [Figure 15] This timing chart illustrates the correctable timing in discontinuous current mode. [Figure 16] This timing chart illustrates a case where the timing of edges other than the first or last occurring within a carrier cycle is corrected for the primary AC voltage. [Figure 17] This timing chart illustrates a case where the timing of the last edge occurring within one carrier cycle is corrected for the primary AC voltage. [Figure 18] This timing chart illustrates a case where the timing of edges other than the first or last occurring within a carrier cycle in the secondary AC voltage is corrected. [Figure 19]This is a timing chart illustrating the case of correcting the timing of the first edge occurring within one carrier period in the secondary-side AC voltage. [Figure 20] This is a block diagram showing an example of a leg determination unit in the control device. [Figure 21] This is a diagram illustrating the relationship between the phase shift amount and the edge. [Figure 22] This is a functional block diagram showing an example of the leg determination unit. [Figure 23] This is a timing chart illustrating the case of correcting the timing of the last edge occurring within one carrier period in the primary-side AC voltage by the primary-side V-phase drive pulse. [Figure 24] This is a timing chart illustrating the case of correcting the timing of the first edge occurring within one carrier period in the secondary-side AC voltage by the secondary-side V-phase drive pulse. [Figure 25] This is a timing chart illustrating the case of correcting the timing of the last edge occurring within one carrier period in the primary-side AC voltage by the primary-side U-phase drive pulse. [Figure 26] This is a timing chart illustrating the case of correcting the timing of the last edge occurring within one carrier period in the primary-side AC voltage by the primary-side V-phase drive pulse in the discontinuous current mode.

Embodiments for Carrying Out the Invention

[0009] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0010] <First Embodiment> FIG. 1 is a diagram showing a configuration example of a power conversion system according to the first embodiment. The power conversion system 100 shown in FIG. 1 includes a bidirectional isolated DC / DC converter (isolated DC / DC converter 110) in which bridge circuits are provided on both sides of an isolation transformer 102. The power conversion system 100 supplies power bidirectionally between the first bridge circuit 111 and the second bridge circuit 112.

[0011] The power conversion system 100 includes an isolation transformer 102, a first bridge circuit 111, a second bridge circuit 112, and a control device 106.

[0012] The isolation transformer 102 has a primary winding 31 and a secondary winding 32, and is a transformer in which the primary winding 31 and the secondary winding 32 are magnetically coupled. The turns ratio of the primary winding 31 and the secondary winding 32 is set as appropriate. The isolation transformer 102 has a magnetic core 33 around which the primary winding 31 and the secondary winding 32 are wound.

[0013] In this specification, unless otherwise specified, the turns ratio of the primary winding 31 and the secondary winding 32 may be considered 1:1. However, when the turns ratio of the primary winding 31 and the secondary winding 32 is other than 1:1, the voltage value on the secondary side or the primary side may be converted to the voltage value on the primary side or the secondary side, and the current value on the secondary side or the primary side may be converted to the current value on the primary side or the secondary side. For example, in the following description, the secondary-side DC voltage E2 and the secondary-side AC voltage V2 mean the voltage values converted to the primary side. That is, when the number of turns of the primary winding 31 of the isolation transformer 102 is n1 and the number of turns of the secondary winding 32 is n2, the voltage value obtained by multiplying the actual secondary-side DC voltage by the coefficient n1 / n2 (the voltage value converted to the primary side) is the secondary-side DC voltage E2. The same applies to the secondary-side AC voltage V2. Also, in the following description, the low-voltage side and the high-voltage side mean the side where low voltage occurs and the side where high voltage occurs among the primary side and the secondary side of the isolation DC / DC converter 110. For example, if there is a relationship of E1 < E2 between the primary-side DC voltage E1 and the secondary-side DC voltage E2 converted to the primary-side value, the primary side is the low-voltage side and the secondary side is the high-voltage side, and if there is a relationship of E1 > E2, the primary side is the high-voltage side and the secondary side is the low-voltage side.

[0014] The first bridge circuit 111 is a primary-side bridge circuit connected to the primary side of the isolation transformer 102, and exchanges power with the primary winding 31 of the isolation transformer 102. The first bridge circuit 111 has a positive terminal 41p and a negative terminal 41n as primary-side DC terminals that are electrically connected to an external device (not shown). The first bridge circuit 111 exchanges power with an external device connected to its primary-side DC terminals.

[0015] The first bridge circuit 111 has a positive bus 43p and a negative bus 43n as a primary-side DC bus pair. The positive bus 43p is connected to the positive terminal 41p. The negative bus 43n is connected to the negative terminal 41n. The first bridge circuit 111 switches the polarity of the voltage V1 applied to the primary winding 31 of the isolation transformer 102 by the primary-side DC bus pair 43p,43n.

[0016] The first bridge circuit 111 is a full bridge circuit having multiple legs 11 and 12 in parallel.

[0017] The first bridge circuit 111 has, for example, a leg 11 in which a high-side arm Q1 and a low-side arm Q2 are connected in series, and a leg 12 in which a high-side arm Q3 and a low-side arm Q4 are connected in series. Arm Q1 is an example of a first arm, arm Q2 is an example of a second arm, arm Q3 is an example of a third arm, and arm Q4 is an example of a fourth arm. Leg 11 is an example of a first leg, and leg 12 is an example of a second leg. The high-side arm and the low-side arm are sometimes collectively referred to as the upper and lower arms.

[0018] The first bridge circuit 111 is a full bridge circuit in which the primary winding 31 of the isolation transformer 102 is provided in the bridge section 21 that connects the intermediate connection point a1 between arm Q1 and arm Q2 and the intermediate connection point b1 between arm Q3 and arm Q4. The first bridge circuit 111 may also have a reactor 104a connected in series with the primary winding 31 of the isolation transformer 102 in the bridge section 21. Intermediate connection point a1 is an example of a first connection point. Intermediate connection point b1 is an example of a second connection point. Bridge section 21 is an example of a first bridge section.

[0019] The first bridge circuit 111 includes a capacitor C1 and arms Q1 to Q4.

[0020] Capacitor C1 is connected between the primary DC bus pair 43p and 43n, and smooths the voltage between the DC bus pair 43p and 43n (the voltage across capacitor C1).

[0021] Arms Q1 to Q4 are primary-side switching elements. Specific examples include semiconductor switching elements such as MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) and IGBTs (Insulated Gate Bipolar Transistors).

[0022] Leg 11 includes a configuration in which arms Q1 and Q2 are connected in series between DC bus pairs 43p and 43n, and Leg 12 includes a configuration in which arms Q3 and Q4 are connected in series between DC bus pairs 43p and 43n. Arms Q1 to Q4 each have a first main terminal, a second main terminal, and a control terminal. For example, the first main terminal corresponds to the drain or collector, the second main terminal corresponds to the source or emitter, and the control terminal corresponds to the gate. Arms Q1 to Q4 may include diodes connected in reverse between the main terminals. If arms Q1 to Q4 are MOSFETs, these diodes may be parasitic diodes. Figure 1 illustrates freewheeling diodes D1, D2, D3, and D4.

[0023] When arms Q1 and Q4 are ON and arms Q2 and Q3 are OFF, the first bridge circuit 111 electrically connects intermediate connection point a1 to the positive bus 43p and intermediate connection point b1 to the negative bus 43n. As a result, the first bridge circuit 111 sets the voltage V1 to a positive voltage "E1". Voltage V1 is the primary AC voltage between intermediate connection points a1 and b1. E1 is the voltage value (primary DC voltage) between DC bus pairs 43p and 43n. When arms Q1 and Q4 are OFF and arms Q2 and Q3 are ON, the first bridge circuit 111 electrically connects intermediate connection point a1 to the negative bus 43n and intermediate connection point b1 to the positive bus 43p. As a result, the first bridge circuit 111 sets the voltage V1 to a negative voltage "-E1". The first bridge circuit 111 operates in this manner to switch the polarity of the voltage V1 applied to the primary winding 31 of the isolation transformer 102 by the primary DC bus pair 43p,43n.

[0024] When arms Q1 and Q3 are turned on and arms Q2 and Q4 are turned off, the first bridge circuit 111 electrically connects both intermediate connection point a1 and intermediate connection point b1 to the positive bus 43p, thereby making the voltage V1 substantially zero. When arms Q1 and Q3 are turned off and arms Q2 and Q4 are turned on, the first bridge circuit 111 electrically connects both intermediate connection point a1 and intermediate connection point b1 to the negative bus 43n, thereby making the voltage V1 substantially zero.

[0025] The second bridge circuit 112 is a secondary bridge circuit connected to the secondary side of the isolation transformer 102, and exchanges power with the secondary winding 32 of the isolation transformer 102. The second bridge circuit 112 has a positive terminal 42p and a negative terminal 42n as secondary DC terminals that are electrically connected to an external device (not shown). The second bridge circuit 112 exchanges power with an external device connected to its secondary DC terminals.

[0026] The second bridge circuit 112 has a positive bus 44p and a negative bus 44n as a secondary DC bus pair. The positive bus 44p is connected to the positive terminal 42p. The negative bus 43n is connected to the negative terminal 42n. The second bridge circuit 112 switches the polarity of the voltage V2 applied to the secondary winding 32 of the isolation transformer 102 by the secondary DC bus pair 44p,44n.

[0027] The second bridge circuit 112 is a full bridge circuit having multiple legs 13 and 14 in parallel.

[0028] The second bridge circuit 112 has, for example, a leg 13 in which a high-side arm Q5 and a low-side arm Q6 are connected in series, and a leg 14 in which a high-side arm Q7 and a low-side arm Q8 are connected in series. Arm Q5 is an example of a fifth arm, arm Q6 is an example of a sixth arm, arm Q7 is an example of a seventh arm, and arm Q8 is an example of an eighth arm. Leg 13 is an example of a third leg, and leg 14 is an example of a fourth leg. The high-side arm and the low-side arm are sometimes collectively referred to as the upper and lower arms.

[0029] The second bridge circuit 112 is a full bridge circuit in which the secondary winding 32 of the isolation transformer 102 is provided in the bridge section 23 that connects the intermediate connection point a2 between arm Q5 and arm Q6 and the intermediate connection point b2 between arm Q7 and arm Q8. The second bridge circuit 112 may also have a reactor 104b connected in series with the secondary winding 32 of the isolation transformer 102 in the bridge section 23. Intermediate connection point a2 is an example of a third connection point. Intermediate connection point b2 is an example of a fourth connection point. Bridge section 23 is an example of a second bridge section.

[0030] The second bridge circuit 112 includes a capacitor C2 and arms Q5 to Q8.

[0031] Capacitor C2 is connected between the 44p and 44n DC bus pairs on the secondary side, and smooths the voltage between the 44p and 44n DC bus pairs (the voltage across capacitor C2).

[0032] Arms Q5 to Q8 are secondary-side switching elements. Specific examples include semiconductor switching elements such as MOSFETs and IGBTs, similar to arms Q1 to Q4.

[0033] Leg 13 includes a configuration in which arms Q5 and Q6 are connected in series between DC bus pairs 44p and 44n, and Leg 14 includes a configuration in which arms Q7 and Q8 are connected in series between DC bus pairs 44p and 44n. Arms Q5 to Q8, like arms Q1 to Q4, each have a first main terminal, a second main terminal, a control terminal, and a diode. Figure 1 illustrates freewheeling diodes D5, D6, D7, and D8.

[0034] When arms Q5 and Q8 are ON and arms Q6 and Q7 are OFF, the second bridge circuit 112 electrically connects intermediate connection point a2 to the positive bus 44p and intermediate connection point b2 to the negative bus 44n. As a result, the second bridge circuit 112 sets the voltage V2 to a positive voltage "E2". Voltage V2 is the secondary AC voltage between intermediate connection points a2 and b2. E2 is the voltage value (secondary DC voltage) between the DC bus pairs 44p and 44n. When arms Q5 and Q8 are OFF and arms Q6 and Q7 are ON, the second bridge circuit 112 electrically connects intermediate connection point a2 to the negative bus 44n and intermediate connection point b2 to the positive bus 44p. As a result, the second bridge circuit 112 sets the voltage V2 to a negative voltage "-E2". The second bridge circuit 112 operates in this manner to switch the polarity of the voltage V2 applied to the secondary winding 32 of the isolation transformer 102 by the secondary DC bus pair 44p,44n.

[0035] When arms Q5 and Q7 are turned on and arms Q6 and Q8 are turned off, the second bridge circuit 112 electrically connects both intermediate connection point a2 and intermediate connection point b2 to the positive bus 44p, thereby making the voltage V2 substantially zero. When arms Q5 and Q7 are turned off and arms Q6 and Q8 are turned on, the second bridge circuit 112 electrically connects both intermediate connection point a2 and intermediate connection point b2 to the negative bus 43n, thereby making the voltage V2 substantially zero.

[0036] An external device (not shown) connected to the positive terminal 41p and the negative terminal 41n is, for example, a charge / discharge device such as a battery. An external device (not shown) connected to the positive terminal 42p and the negative terminal 42n is, for example, a DC system capable of charging and discharging the energy of the charge / discharge device connected to the primary side.

[0037] The control device 106 controls the first bridge circuit 111 and the second bridge circuit 112. The control device 106 generates drive pulses g1 to g4 to drive arms Q1 to Q4 of the first bridge circuit 111, and drive pulses g5 to g8 to drive arms Q5 to Q8 of the second bridge circuit 112, respectively. Drive pulses g1 to g8 are drive signals for controlling the on or off of the corresponding arms among arms Q1 to Q8. Drive pulses g1 to g4 are commands output from the control device 106 as first drive pulses that cause the control device 106 to switch the first bridge circuit 111. Drive pulses g5 to g8 are commands output from the control device 106 as second drive pulses that cause the control device 106 to switch the second bridge circuit 112.

[0038] The power conversion system 100 includes drive circuits 105a and 105b. Drive circuit 105a is a primary drive circuit (first drive circuit) that drives the first bridge circuit 111 based on drive pulses g1 to g4 supplied from the control device 106. Drive circuit 105a controls the on / off switching of arms Q1 to Q4 of the first bridge circuit 111 according to the drive pulses g1 to g4. Drive circuit 105b is a secondary drive circuit (second drive circuit) that drives the second bridge circuit 112 based on drive pulses g5 to g8 supplied from the control device 106. Drive circuit 105b controls the on / off switching of arms Q5 to Q8 of the second bridge circuit 112 according to the drive pulses g5 to g8.

[0039] The power conversion system 100 includes a DC voltage detection unit 107a and a DC voltage detection unit 107b. The DC voltage detection unit 107a is a circuit that detects the primary DC voltage E1 applied to the DC bus pair 43p, 43n of the first bridge circuit 111. The DC voltage detection unit 107b is a circuit that detects the secondary DC voltage E2 applied to the DC bus pair 44p, 44n of the second bridge circuit 112.

[0040] The power conversion system 100 includes a DC current detection unit 108a and a DC current detection unit 108b. The DC current detection unit 108a is a circuit that detects the primary DC current I1 flowing through the positive bus 43p of the first bridge circuit 111. The DC current detection unit 108b is a circuit that detects the secondary DC current I2 flowing through the positive bus 44p of the second bridge circuit 112.

[0041] The control device 106 controls the phase of the edges of the drive pulses g1 to g8 for driving the arms Q1 to Q8 based on the primary DC voltage E1 detected by the DC voltage detection unit 107a and the secondary DC voltage E2 detected by the DC voltage detection unit 107b. By controlling the phase of the edges of the drive pulses g1 to g8, the control device 106 controls the power transmission of the isolated DC / DC converter 110 (power transmission between the first bridge circuit 111 and the second bridge circuit 112).

[0042] The control device 106 generates drive pulses g1 to g4 that cause the primary AC voltage V1 to be output to the first bridge circuit 111 by comparing one or more first pulse width commands P1, which determine the pulse width of the primary AC voltage V1 applied to the primary side of the isolation transformer 102, with one or more first carrier waves Ca1. The first carrier waves Ca1 are signals that determine the switching frequencies of arms Q1 to Q4.

[0043] The control device 106 generates drive pulses g5 to g8 that cause the secondary AC voltage V2 to be output to the second bridge circuit 112 by comparing one or more second pulse width commands P2, which determine the pulse width of the secondary AC voltage V2 applied to the secondary side of the isolation transformer 102, with one or more second carrier waves Ca2. The second carrier waves Ca2 are signals that determine the switching frequencies of arms Q5 to Q8.

[0044] Figure 2 is a timing chart illustrating the waveform of the first drive pulse supplied by the control device to the primary drive circuit and the switching waveform of the primary switching element. As shown in Figure 2, the switch timing in which the state of arms Q1 to Q4 changes lags behind the timing in which the edges of drive pulses g1 to g4 change. Although not specifically shown, similarly, for the waveform of the second drive pulse supplied by the control device to the secondary drive circuit and the switching waveform of the secondary switching element, the switch timing in which the state of arms Q5 to Q8 changes lags behind the timing in which the edges of drive pulses g5 to g8 change.

[0045] Let ΔQ* be the delay time between the timing when the drive pulse g* changes from on to off and the timing when the state of arm Q* changes from on to off (* is a number from 1 to 8). These delay times ΔQ1 to ΔQ8 vary for each of the arms Q1 to Q8. These variations depend on the individual differences in the characteristics of the electronic components used in the arms Q1 to Q8 and the drive circuits 105a and 105b.

[0046] Due to variations in the delay time across each arm, a primary AC voltage V1 containing a DC component DC(V1) is applied to magnetic components such as the isolation transformer 102. V1' represents the primary AC voltage V1 actually applied due to the variation in delay time. T represents one cycle of arm switching. When a primary AC voltage V1 containing a DC component DC(V1) is applied, the primary AC current Iac1 flowing on the primary side of the isolation transformer 102 is superimposed with the DC component DC(Iac1). Iac1' represents the primary AC current Iac1 actually flowing due to the variation in delay time. ΔDC(Iac1) represents the amount of change in the DC component DC(Iac1) over one switching cycle. Similarly, on the secondary side, due to variations in the delay time across each arm, the secondary AC voltage V2 is superimposed with the DC component DC(V2), and the secondary AC current Iac2 flowing on the secondary side of the isolation transformer 102 is superimposed with the DC component DC(Iac2). Thus, when an AC voltage containing a DC component is applied to a magnetic component such as the isolation transformer 102, the magnetic core of the magnetic component may become demagnetized, potentially leading to a failure to transmit power or the flow of overcurrent.

[0047] To prevent such problems, the power conversion system 100 shown in Figure 1 has a function to suppress the bias of the isolation transformer 102 (magnetic core 33). Specifically, the control device 106 corrects the first pulse width command P1 or the second pulse width command P2 to reduce the DC component IdcA of the excitation current Im of the isolation transformer 102 and the DC component IdcB of the load current Ib, which is obtained by subtracting the excitation current Im from the primary AC current Iac1 or the secondary AC current Iac2. By reducing both the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib, the bias of the isolation transformer 102 can be suppressed more quickly than by reducing only one of them. By reducing the DC component IdcB of the load current Ib, an excessive load current Ib (overcurrent) can be prevented.

[0048] The DC component IdcA of the excitation current Im of the isolation transformer 102 is an example of the bias amount of the isolation transformer 102 (a quantity representing the degree of bias of the magnetic core 33).

[0049] The load current Ib is approximately equal to the current obtained by subtracting the excitation current Im of the isolation transformer 102 from the primary AC current Iac1 or secondary AC current Iac2 flowing on the primary side of the isolation transformer 102. The load current Ib corresponds to the reactor current flowing through reactor 104a if reactor 104a is present, and to the reactor current flowing through reactor 104b if reactor 104b is present.

[0050] The power conversion system 100 shown in Figure 1 has a DC component detection unit 109 as a circuit for detecting the bias of the isolation transformer 102 (DC component IdcA of the excitation current Im) and the DC component IdcB of the load current Ib. The DC component detection unit 109 detects the DC component DC(Iac1) of the primary AC current Iac1 and the DC component DC(Iac2) of the secondary AC current Iac2. For example, the DC component detection unit 109 extracts the DC component DC(Iac1) from the detected value of the primary AC current Iac1 using a low-pass filter, and also extracts the DC component DC(Iac2) from the detected value of the secondary AC current Iac2 using a low-pass filter.

[0051] The degree of magnetic bias of the isolation transformer 102 increases as the DC component of the excitation current Im of the isolation transformer 102 increases. Therefore, the DC component detection unit 109 detects the DC component IdcA of the excitation current Im of the isolation transformer 102 as the degree of magnetic bias of the isolation transformer 102. Alternatively, the detected value of the degree of magnetic bias of the isolation transformer 102 may be obtained by detecting the magnetic flux of the magnetic core 33.

[0052] The DC component detection unit 109 detects the excitation current Im based, for example, on the primary AC current Iac1 and the secondary AC current Iac2. For example, the DC component detection unit 109 detects the excitation current Im based on the relationship "Im = Iac1 - Iac2 × N", where N represents the winding ratio (transformation ratio) of the isolation transformer 102. Note that when N=1, "Im = Iac1 - Iac2". The DC component detection unit 109 calculates (detects) the DC component IdcA of the excitation current Im by subtracting the DC component DC(Iac2) of the secondary AC current Iac2 from the DC component DC(Iac1) of the primary AC current Iac1.

[0053] The DC component detection unit 109 calculates (detects) the DC component IdcB of the load current Ib by subtracting the DC component IdcA of the excitation current Im of the isolation transformer 102 from the DC component DC(Iac1) of the primary AC current Iac1 or the DC component DC(Idc2) of the secondary AC current Iac2.

[0054] Figure 3 is a block diagram showing an example of the control device configuration. The control device 106 corrects the first pulse width command P1 or the second pulse width command P2 based on the detected value of the DC component IdcA of the excitation current Im and the detected value of the DC component IdcB of the load current Ib.

[0055] The first pulse width command P1 includes a primary U-phase pulse width command P1U for determining the gate pulse widths of arms Q1 and Q2 included in the primary U-phase leg 11, and a primary V-phase pulse width command P1V for determining the gate pulse widths of arms Q3 and Q4 included in the primary V-phase leg 12. The second pulse width command P2 includes a secondary U-phase pulse width command P2U for determining the gate pulse widths of arms Q5 and Q6 included in the secondary U-phase leg 13, and a secondary V-phase pulse width command P2V for determining the gate pulse widths of arms Q7 and Q8 included in the secondary V-phase leg 14. The U-phase is an example of the first phase. The V-phase is an example of the second phase.

[0056] The control device 106 includes a bias correction unit 71, a DC correction unit 72, a correction processing unit 70, and a modulation module 60.

[0057] The bias correction unit 71 derives a correction amount (bias correction amount ΔIdcA) to suppress bias of the isolation transformer 102 by reducing the DC component IdcA, based on the detected value of the bias amount of the isolation transformer 102 (DC component IdcA of the excitation current Im). The bias correction unit 71 has a subtractor 61 that calculates the deviation (excitation current deviation) between the DC component IdcA of the excitation current Im and the command value of the DC component IdcA. Since the objective is to suppress bias of the isolation transformer 102 by reducing the DC component IdcA, the command value of the DC component IdcA is set to zero. The bias correction unit 71 has a regulator 62 that derives the bias correction amount ΔIdcA by P control or PI control that converges the excitation current deviation to zero. In P control or PI control, P represents proportional control and I represents integral control.

[0058] The DC correction unit 72 derives a correction amount (DC correction amount ΔIdcB) to suppress the bias of the isolation transformer 102 by reducing the DC component IdcB based on the detected value of the DC component IdcB of the load current Ib. The DC correction unit 72 has a subtractor 63 that calculates the deviation (DC component deviation) between the DC component IdcB of the load current Ib and the command value of the DC component IdcB. Since the objective is to suppress the bias of the isolation transformer 102 by reducing the DC component IdcB of the load current Ib, the command value of the DC component IdcB is set to zero. The DC correction unit 72 has a regulator 64 that derives the DC correction amount ΔIdcB by P control or PI control that converges the DC component deviation to zero.

[0059] The correction processing unit 70 corrects the first pulse width command P1 or the second pulse width command P2 based on the bias correction amount ΔIdcA and the DC correction amount ΔIdcB. Several examples of the correction processing by the correction processing unit 70 will be described later.

[0060] The modulation module 60 performs pulse width modulation to control the pulse widths of each drive pulse g1 to g4 by comparing the first pulse width command P1, corrected by the correction processing unit 70, with one or more first carrier waves Ca1. Alternatively, the modulation module 60 performs pulse width modulation to control the pulse widths of each drive pulse g5 to g8 by comparing the second pulse width command P2, corrected by the correction processing unit 70, with one or more second carrier waves Ca2. The primary U-phase carrier wave Ca1U is one of the first carrier waves Ca1 compared with the primary U-phase pulse width command P1U. The primary V-phase carrier wave Ca1V is one of the first carrier waves Ca1 compared with the primary V-phase pulse width command P1V. The secondary U-phase carrier wave Ca2U is one of the second carrier waves Ca2 compared with the secondary U-phase pulse width command P2U. The secondary V-phase carrier wave Ca2V is one of the second carrier waves Ca2 compared with the secondary V-phase pulse width command P2V.

[0061] Figure 4 is a block diagram showing a first example of a correction processing unit within the control device. The correction processing unit 70A is an example of the correction processing unit 70 described above. The correction processing unit 70A corrects one of the first pulse width commands P1 and the second pulse width command P2 based on the detected value of the DC component IdcA of the excitation current Im, and corrects the other pulse width command based on the detected value of the DC component IdcB of the load current Ib. As a result, the correction processing unit 70A can correct one of the pulse width commands so that the DC component IdcA of the excitation current Im becomes smaller, and correct the other pulse width command so that the DC component IdcB of the load current Ib becomes smaller. As a result, by reducing both the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib, the bias of the isolation transformer 102 can be suppressed more quickly than by reducing only one of them.

[0062] Figure 4 illustrates a configuration in which the correction processing unit 70A corrects the first pulse width command P1 based on the detected value of the DC component IdcA of the excitation current Im, and corrects the second pulse width command P2 based on the detected value of the DC component IdcB of the load current Ib. However, the correction processing unit 70A may also be configured to correct the second pulse width command P2 based on the detected value of the DC component IdcA of the excitation current Im, and correct the first pulse width command P1 based on the detected value of the DC component IdcB of the load current Ib.

[0063] The correction processing unit 70A corrects the pulse width commands of the same phase from among the primary U-phase pulse width command P1U, primary V-phase pulse width command P1V, secondary U-phase pulse width command P2U, and secondary V-phase pulse width command P2V, based on the detected value of the DC component IdcA of the excitation current Im and the detected value of the DC component IdcB of the load current Ib. Figure 4 illustrates a configuration in which the correction processing unit 70A corrects the U-phase pulse width commands P1U and P2U. However, as shown in Figure 5, the correction processing unit 70B may also be configured to correct the V-phase pulse width commands P1V and P2V.

[0064] The correction processing units 70A and 70B set the sign of the correction amount reflected in the first pulse width command P1 or the second pulse width command P2 to a sign that reduces the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib. As a result, the correction processing units 70A and 70B can adjust the pulse width of the primary AC voltage V1 or the secondary AC voltage V2 so that the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib are reduced.

[0065] In the example shown in Figure 4, the correction processing unit 70A includes a polarity switching unit 75, a polarity switching unit 76, an adder 73, and an adder 74. The polarity switching unit 75 switches whether the sign of the bias correction amount ΔIdcA, which is reflected in the primary side U-phase pulse width command P1U by the adder 73, is positive or negative. The polarity switching unit 76 switches whether the sign of the DC correction amount ΔIdcB, which is reflected in the secondary side U-phase pulse width command P2U by the adder 74, is positive or negative.

[0066] In the example shown in Figure 5, the correction processing unit 70B includes a polarity switching unit 79, a polarity switching unit 80, an adder 77, and an adder 78. The polarity switching unit 79 switches whether the sign of the bias correction amount ΔIdcA, which is reflected in the primary side V-phase pulse width command P1V by the adder 77, is positive or negative. The polarity switching unit 80 switches whether the sign of the DC correction amount ΔIdcB, which is reflected in the secondary side V-phase pulse width command P2V by the adder 78, is positive or negative.

[0067] Figure 6 shows an example of the operating waveform of the power conversion system according to the first embodiment. Figure 6 illustrates a case in which the control device 106 performs discontinuous current mode phase control that generates a period (interval) in which both the primary AC voltage V1 and the secondary AC voltage V2 of the isolation transformer 102 are at zero voltage.

[0068] When the primary U-phase drive pulse is 1, drive pulse g1 (arm Q1) is on and drive pulse g2 (arm Q2) is off. When the primary U-phase drive pulse is 0, drive pulse g1 is off and drive pulse g2 is on. When the primary V-phase drive pulse is 1, drive pulse g3 (arm Q3) is on and drive pulse g4 (arm Q4) is off. When the primary V-phase drive pulse is 0, drive pulse g3 is off and drive pulse g4 is on. When the secondary U-phase drive pulse is 1, drive pulse g5 (arm Q5) is on and drive pulse g6 (arm Q6) is off. When the secondary U-phase drive pulse is 0, drive pulse g5 is off and drive pulse g6 is on. When the secondary V-phase drive pulse is 1, drive pulse g7 (arm Q7) is on and drive pulse g8 (arm Q8) is off. When the secondary V-phase drive pulse is 0, drive pulse g7 is off and drive pulse g8 is on.

[0069] The control device 106 controls the first bridge circuit 111 and the second bridge circuit 112 in the operating mode shown in Figure 6, so that the primary AC voltage V1 and the secondary AC voltage V2 have positive voltage intervals, zero voltage intervals, and negative voltage intervals, respectively.

[0070] Figure 7 shows an example in which, in the operating mode of Figure 6, the control device 106 corrects the primary side U-phase pulse width command P1U based on the DC component IdcA (bias) of the excitation current Im, thereby widening the negative pulse width of the primary side AC voltage V1. Figure 7 shows an example of the operating waveform by the correction processing unit 70A in Figure 4.

[0071] In Figure 7, when the DC component IdcA (bias) of the excitation current Im is a positive value, the sign of the bias correction amount ΔIdcA supplied from the regulator 62 to the polarity switching unit 75 is negative. At this time, the polarity switching unit 75 sets the sign of the bias correction amount ΔIdcA, which is reflected in the primary side U-phase pulse width command P1U by the adder 73, to a negative sign, and outputs the input bias correction amount ΔIdcA with its original sign. In other words, the polarity switching unit 75 outputs a negative bias correction amount ΔIdcA. Since the negative bias correction amount ΔIdcA is reflected in the primary side U-phase pulse width command P1U by the adder 73, the primary side U-phase pulse width command P1U decreases compared to before the reflection.

[0072] The modulation module 60 performs pulse width modulation to control the pulse widths of the drive pulses g1 and g2 by comparing the primary side U-phase pulse width command P1U with the primary side U-phase carrier wave Ca1U. The modulation module 60 also performs pulse width modulation to control the pulse widths of the drive pulses g3 and g4 by comparing the primary side V-phase pulse width command P1V with the primary side V-phase carrier wave Ca1V.

[0073] The modulation module 60 compares the primary U-phase pulse width command P1U, which reflects the negative bias correction amount ΔIdcA, with the primary U-phase carrier wave Ca1U, thereby shortening the pulse width of the drive pulse g1 compared to before the correction. As a result, the period during which arm Q1 is turned on and arm Q2 is turned off is shortened, so the negative pulse width of the primary AC voltage V1 widens compared to before the correction. As the negative pulse width of the primary AC voltage V1 widens, the positive DC component IdcA (positive bias) of the excitation current Im becomes smaller compared to before the widening. Therefore, when the DC component IdcA (bias) of the excitation current Im is a positive value, it is corrected so that its value approaches zero, thus suppressing the bias.

[0074] Figure 8 shows an example in which, in the operating mode of Figure 6, the control device 106 corrects the primary side V-phase pulse width command P1V based on the DC component IdcA (bias) of the excitation current Im, thereby widening the negative pulse width of the primary side AC voltage V1. Figure 8 shows an example of the operating waveform by the correction processing unit 70B in Figure 5.

[0075] In Figure 8, when the DC component IdcA (bias) of the excitation current Im is a positive value, the sign of the bias correction amount ΔIdcA supplied from the regulator 62 to the polarity switching unit 79 is negative. At this time, the polarity switching unit 79 sets the sign of the bias correction amount ΔIdcA, which is reflected in the primary side V-phase pulse width command P1V by the adder 77, to a positive sign, and outputs the bias correction amount ΔIdcA by inverting the sign of the input bias correction amount ΔIdcA. In other words, the polarity switching unit 79 outputs a positive bias correction amount ΔIdcA. Since the positive bias correction amount ΔIdcA is reflected in the primary side V-phase pulse width command P1V by the adder 77, the primary side V-phase pulse width command P1V increases compared to before the reflection.

[0076] The modulation module 60 compares the primary side V-phase pulse width command P1V, which reflects the positive bias correction amount ΔIdcA, with the primary side V-phase carrier wave Ca1V, thereby making the pulse width of the drive pulse g3 longer than before the correction. As a result, the period during which arm Q3 is turned on and arm Q4 is turned off is extended, so the negative pulse width of the primary side AC voltage V1 widens compared to before the correction. Because the negative pulse width of the primary side AC voltage V1 widens, the positive DC component IdcA (positive bias amount) of the excitation current Im becomes smaller compared to before the widening. Therefore, when the DC component IdcA (bias amount) of the excitation current Im is a positive value, it is corrected so that its value approaches zero, thus suppressing the bias amount.

[0077] Figure 9 is a block diagram showing a third example of a correction processing unit within the control device. The correction processing unit 70C is an example of the correction processing unit 70 described above. The correction processing unit 70C corrects only one of the two pulse width commands, the first pulse width command P1 and the second pulse width command P2. This allows the correction processing unit 70C to correct the first pulse width command P1 so that the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib become smaller. As a result, by reducing both the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib, the bias of the isolation transformer 102 can be suppressed more quickly than by reducing only one of them. The correction processing unit 70C may also correct only one of the two pulse width commands P2, the second pulse width command P2 and the first pulse width command P1.

[0078] Figure 10 shows an example in which, in the operating mode of Figure 6, the control device 106 corrects the primary side U-phase pulse width command P1U based on the DC component IdcA (bias) of the excitation current Im, and corrects the primary side V-phase pulse width command P1V based on the DC component IdcB of the load current Ib, thereby changing the negative pulse width of the primary side AC voltage V1. Figure 10 shows an example of the operating waveform by the correction processing unit 70C in Figure 9.

[0079] In Figure 10, when the DC component IdcA (bias) of the excitation current Im is a positive value, the polarity switching unit 75 outputs a negative bias correction amount ΔIdcA, similar to Figure 7. Since the negative bias correction amount ΔIdcA is reflected in the primary side U-phase pulse width command P1U by the adder 73, the primary side U-phase pulse width command P1U decreases compared to before the reflection. The modulation module 60 compares the primary side U-phase pulse width command P1U, which reflects the negative bias correction amount ΔIdcA, with the primary side U-phase carrier wave Ca1U, and shortens the pulse width of the drive pulse g1 compared to before the reflection. As a result, the period during which arm Q1 is turned on and arm Q2 is turned off is shortened, so the phase of the starting point of the negative pulse width of the primary side AC voltage V1 becomes earlier than before the reflection, so that the negative pulse width of the primary side AC voltage V1 widens compared to before the reflection. Therefore, when the DC component IdcA (bias) of the excitation current Im is a positive value, it is corrected so that its value approaches zero, thus suppressing the bias.

[0080] On the other hand, in Figure 10, when the DC component IdcB of the load current Ib is a negative value, the polarity switching unit 79 reverses the sign of the DC correction amount ΔIdcB and outputs a positive DC correction amount ΔIdcB. Since the positive DC correction amount ΔIdcB is reflected in the primary side V-phase pulse width command P1V by the adder 77, the primary side V-phase pulse width command P1V decreases compared to before the reflection. The modulation module 60 compares the primary side V-phase pulse width command P1V, which reflects the negative DC correction amount ΔIdcB, with the primary side V-phase carrier wave Ca1V, and shortens the pulse width of the drive pulse g3 compared to before the reflection. As a result, the period during which arm Q3 is turned on and arm Q4 is turned off is shortened, so the phase of the endpoint of the negative pulse width of the primary side AC voltage V1 becomes earlier than before the reflection, so that the negative pulse width of the primary side AC voltage V1 becomes narrower than before the reflection. Therefore, when the DC component IdcB of the load current Ib is negative, it is corrected so that its value approaches zero, thus suppressing magnetic bias.

[0081] Figure 11 is a block diagram showing a fourth example of a correction processing unit within the control device. The correction processing unit 70D is an example of the correction processing unit 70 described above. The correction processing unit 70D corrects both the first pulse width command P1 and the second pulse width command P2. As a result, the correction processing unit 70D can correct the first pulse width command P1 and the second pulse width command P2 so that the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib become smaller. As a result, by reducing both the DC component IdcA of the excitation current Im and the DC component IdcB of the load current Ib, the bias of the isolation transformer 102 can be suppressed more quickly than by reducing only one of them.

[0082] Figure 12 is a block diagram showing a fifth example of a correction processing unit within a control device. The correction processing unit 70E is an example of the correction processing unit 70 described above. Figure 12 illustrates a configuration in which the correction processing unit 70E corrects the first pulse width command P1 based on the detected value of the DC voltage E1 of the first bridge circuit 111, and corrects the second pulse width command P2 based on the detected value of the DC voltage E2 of the second bridge circuit 112. As a result, the correction processing unit 70E can correct the first pulse width command P1 to a magnitude corresponding to the change in DC voltage E1, and correct the second pulse width command P2 to a magnitude corresponding to the change in DC voltage E2, thereby improving the accuracy of the correction.

[0083] When the DC voltage E1 decreases, the DC component IdcA of the excitation current Im decreases, and therefore the bias correction amount ΔIdcA required to suppress the bias generated by the DC component IdcA of the excitation current Im also decreases. Therefore, when the detected value of the DC voltage E1 decreases, the correction processing unit 70E reduces the bias correction amount ΔIdcA reflected in the primary side U-phase pulse width command P1U compared to before the detected value of the DC voltage E1 decreased. On the other hand, when the detected value of the DC voltage E1 increases, the correction processing unit 70E increases the bias correction amount ΔIdcA reflected in the primary side U-phase pulse width command P1U compared to before the detected value of the DC voltage E1 increased. As a result, bias can be suppressed by an appropriate bias correction amount ΔIdcA corresponding to the decrease in the DC voltage E1.

[0084] The correction processing unit 70E may adjust the correction gain 81 multiplied by the bias correction amount ΔIdcA according to the detected value of the DC voltage E1. When the detected value of the DC voltage E1 decreases, the correction processing unit 70E reduces the correction gain 81 compared to before the detected value of the DC voltage E1 decreased. On the other hand, when the detected value of the DC voltage E1 increases, the correction processing unit 70E increases the correction gain 81 compared to before the detected value of the DC voltage E1 increased. This allows bias to be suppressed by an appropriate bias correction amount ΔIdcA in accordance with the change in the DC voltage E1. The correction processing unit 70E may determine the value of the correction gain 81 corresponding to the detected value of the DC voltage E1 based on a relational rule (e.g., calculation formula, map, etc.) that defines the correspondence between the correction gain 81 and the DC voltage E1.

[0085] When the DC voltage E2 decreases, the DC component IdcA of the load current Ib decreases, and therefore the DC correction amount ΔIdcB required to suppress the magnetic bias generated by the DC component IdcB of the load current Ib also decreases. Therefore, when the detected value of the DC voltage E2 decreases, the correction processing unit 70E reduces the DC correction amount ΔIdcB reflected in the secondary U-phase pulse width command P2U compared to before the detected value of the DC voltage E2 decreased. On the other hand, when the detected value of the DC voltage E2 increases, the correction processing unit 70E increases the DC correction amount ΔIdcB reflected in the secondary U-phase pulse width command P2U compared to before the detected value of the DC voltage E2 increased. As a result, magnetic bias can be suppressed by an appropriate DC correction amount ΔIdcB corresponding to the decrease in the DC voltage E2.

[0086] The correction processing unit 70E may adjust the correction gain 82 multiplied by the DC correction amount ΔIdcB according to the detected value of the DC voltage E2. When the detected value of the DC voltage E2 decreases, the correction processing unit 70E reduces the correction gain 82 compared to before the detected value of the DC voltage E2 decreased. On the other hand, when the detected value of the DC voltage E2 increases, the correction processing unit 70E increases the correction gain 82 compared to before the detected value of the DC voltage E2 increased. This allows for the suppression of magnetic bias by an appropriate DC correction amount ΔIdcB in response to the change in the DC voltage E2. The correction processing unit 70E may determine the value of the correction gain 82 corresponding to the detected value of the DC voltage E2 based on a relational rule (e.g., calculation formula, map, etc.) that defines the correspondence between the correction gain 82 and the DC voltage E2.

[0087] Furthermore, the processing details of the correction processing unit 70E shown in Figure 12 can also be applied to other correction processing units such as those shown in Figure 5.

[0088] Figure 13 is a block diagram showing an example configuration of a control device including a current controller. The control device 106 has a current controller 50 that performs DC current control using the primary DC current I1 detected by the DC current detection unit 108a (see Figure 1). In Figure 13, the current controller 50 performs DC current control that determines the phase shift amount of the first carrier wave Ca1 and the second carrier wave Ca2, for example, based on the average value of the primary DC current I1 detected by the DC current detection unit 108a (for example, the average value for each carrier period). DC current control is not limited to this, as long as it utilizes the primary DC current I1.

[0089] On the other hand, as described above, the control device 106 performs bias suppression control by correcting the first pulse width command P1 or the second pulse width command P2, thereby reducing the DC component IdcA of the first excitation current Im and the DC component IdcB of the load current Ib. The pulse width of the primary AC voltage V1 or the secondary AC voltage V2 is adjusted by correcting the first pulse width command P1 or the second pulse width command P2.

[0090] However, the pulse width of the primary AC voltage V1 or secondary AC voltage V2 is adjusted by the bias suppression control, which causes disturbances in the primary DC current I1. If the disturbance in the primary DC current I1 is large, the DC current control, which uses the detected value of the primary DC current I1, may react to the disturbance. In this way, interference between bias suppression control and DC current control may reduce the controllability of the DC current control.

[0091] To prevent such problems, the control device 106 may perform timing-specific control to specify the timing for adjusting the pulse width of the primary AC voltage V1 and / or secondary AC voltage V2 to a timing that reduces disturbances occurring in the primary DC current I1. Through timing-specific control, the control device 106 suppresses disturbances occurring in the primary DC current I1 by bias suppression control (adjustment of the pulse width of the primary AC voltage V1 or secondary AC voltage V2).

[0092] Figure 14 is a timing chart illustrating correctable timing in continuous current mode. Continuous current mode is one of the control modes of the control device 106, and is a mode in which the primary AC voltage V1 and secondary AC voltage V2 are reversed in polarity by controlling the phase of each edge of the primary drive pulses g1 to g4 and secondary drive pulses g5 to g8. Figure 15 is a timing chart illustrating correctable timing in discontinuous current mode. Discontinuous current mode is one of the control modes of the control device 106, and is a mode in which the primary AC voltage V1 and secondary AC voltage V2 are both zero voltage by controlling the phase of each edge of the primary drive pulses g1 to g4 and secondary drive pulses g5 to g8. Figures 14 and 15 illustrate the cases of power transmission mode, where power is transmitted from the primary to the secondary, and regenerative mode, where power is transmitted from the secondary to the primary, respectively.

[0093] In both the continuous current mode shown in Figure 14 and the discontinuous current mode shown in Figure 15, there are multiple timings (rising or falling edge timings) in which the pulse width of the primary AC voltage V1 or secondary AC voltage V2 can be adjusted. The control device 106 performs timing-specific control to adjust the pulse width of the primary AC voltage V1 or secondary AC voltage V2 at a timing closer to the start or end of a carrier cycle than at a timing closer to the center of the carrier cycle. In timing-specific control, the control device 106 adjusts the pulse width of the primary AC voltage V1 and the pulse width of the secondary AC voltage V2 independently.

[0094] In Figure 14, there are four edges that occur in the primary AC voltage V1 within one carrier cycle. Edges t12 and t13 are examples of the first edge and occur closer to the center of the carrier cycle. Edge t11 is an example of the second edge, which is closer to the start of the carrier cycle than the first edge. Edge t14 is an example of the third edge, which is closer to the end of the carrier cycle than the first edge. The control device 106 performs first timing specific control to reduce the change in the primary DC current I1 by adjusting the timing of edge t11 or edge t14 without adjusting the timing of edges t12 and t13.

[0095] In Figure 14, there are two edges that occur in the secondary AC voltage V2 within one carrier cycle. Edge t22 is an example of a fourth edge and occurs closer to the center of the carrier cycle. Edge t21 is an example of a fifth edge that is closer to the start of the same carrier cycle than the fourth edge. The control device 106 performs a second timing specific control that reduces the change in the primary DC current I1 by adjusting the timing of edge t21 without adjusting the timing of edge t22.

[0096] In the continuous operation mode shown in Figure 14, the control device 106 suppresses disturbances generated in the primary DC current I1 by performing first timing specific control and / or second timing specific control, thereby suppressing bias suppression control.

[0097] In Figure 15, there are four edges in the primary AC voltage V1 within one carrier cycle. Similar to the case in Figure 14, the control device 106 performs a first timing specific control to reduce the change in the primary DC current I1 by adjusting the timing of edge t11 or edge t14, without adjusting the timing of edges t12 and t13.

[0098] In Figure 15, there are four edges that occur in the secondary AC voltage V2 within one carrier cycle. Edges t22 and t23 are examples of the fourth edge and occur closer to the center of the carrier cycle. Edge t21 is an example of the fifth edge, which is closer to the start of the carrier cycle than the fourth edge. Edge t24 is an example of the fifth edge, which is closer to the end of the carrier cycle than the fourth edge. The control device 106 performs a second timing specific control to reduce the change in the primary DC current I1 by adjusting the timing of edge t21 or edge t24 without adjusting the timing of edges t22 and t23.

[0099] In the discontinuous operation mode shown in Figure 15, the control device 106 suppresses disturbances generated in the primary DC current I1 by performing first timing identification control and / or second timing identification control, thereby suppressing bias suppression control.

[0100] Next, referring to Figures 16-19, we will explain the reason for adjusting the pulse width of the primary AC voltage V1 or secondary AC voltage V2 at a timing closer to the start or end point of a carrier cycle, rather than at a timing closer to the center of the carrier cycle.

[0101] Figure 16 is a timing chart illustrating the case where the timing of edges other than the first or last occurring in one carrier cycle is corrected for the primary AC voltage V1. As shown in Figure 16, when the timing of edges occurring near the center of one carrier cycle is corrected for the primary AC voltage V1, the primary AC current Iac begins to change from that timing near the center of one carrier cycle. When the primary AC current Iac begins to change, the primary DC current I1 also begins to change. As a result, the average value of the primary DC current I1 in that carrier cycle changes significantly compared to when this correction is not performed.

[0102] In contrast, Figure 17 is a timing chart illustrating the case where the timing of the last edge (the edge closest to the end of the carrier cycle) occurring in the primary AC voltage V1 is corrected. As shown in Figure 16, when the timing of the last edge occurring in the primary AC voltage V1 is corrected, the primary AC current Iac begins to change from that timing, which is closer to the end of the carrier cycle. When the primary AC current Iac begins to change, the primary DC current I1 also begins to change. However, both the primary AC current Iac and the primary DC current I1 begin to change at timings closer to the end of the carrier cycle. Therefore, the range by which the average value of the primary DC current I1 in that carrier cycle changes compared to the case without such correction is smaller than in the case of Figure 16. Thus, the more the timing of the edge closest to the end of the carrier cycle among the multiple edges of the primary AC voltage V1 is corrected, the smaller the change in the primary DC current I1 becomes.

[0103] Figure 17 shows a case where the change in the primary DC current I1 becomes smaller as the timing of the edge closest to the end of one carrier cycle among the multiple edges of the primary AC voltage V1 is corrected. Although not specifically shown in the figure, the change in the primary DC current I1 becomes smaller as the timing of the edge closest to the end of one carrier cycle among the multiple edges of the secondary AC voltage V2 is corrected.

[0104] Figure 18 is a timing chart illustrating the case where the timing of edges other than the first or last occurring in one carrier cycle is corrected for the secondary AC voltage V2. As shown in Figure 18, when the timing of edges occurring near the center of one carrier cycle is corrected for the secondary AC voltage V2, the primary AC current Iac begins to change from that timing near the center of one carrier cycle. When the primary AC current Iac begins to change, the primary DC current I1 also begins to change. As a result, the average value of the primary DC current I1 in that carrier cycle changes significantly compared to when this correction is not performed.

[0105] In contrast, Figure 19 is a timing chart illustrating the case where the timing of the first edge (the edge closest to the start of the carrier cycle) occurring in the secondary AC voltage V2 is corrected. As shown in Figure 19, when the timing of the first edge occurring in the carrier cycle is corrected in the secondary AC voltage V2, the primary AC current Iac begins to change from that timing closer to the start of the carrier cycle. When the primary AC current Iac begins to change, the primary DC current I1 also begins to change. However, since the primary AC current Iac begins to change at a timing closer to the start of the carrier cycle, the primary AC current Iac has an offset waveform over the entire carrier cycle. At this time, the change in the negative polarity waveform and the change in the positive polarity waveform of the primary AC current Iac cancel each other out. Therefore, the range by which the average value of the primary DC current I1 in that carrier cycle changes compared to the case without this correction is smaller than in the case of Figure 18. Thus, the more the timing of the edge of the secondary AC voltage V2 that is closest to the start of one carrier cycle is corrected, the smaller the change in the primary DC current I1 becomes.

[0106] Figure 19 shows a case where the change in the primary DC current I1 becomes smaller as the timing of the edge of the secondary AC voltage V2 closest to the start of one carrier cycle is corrected. Although not specifically illustrated, the change in the primary DC current I1 becomes smaller as the timing of the edge of the primary AC voltage V1 closest to the start of one carrier cycle is corrected.

[0107] Next, we will explain the leg determination unit, which selects a timing closer to the start or end point of a carrier cycle than a timing closer to the center of the carrier cycle.

[0108] Figure 20 is a block diagram showing an example of a leg determination unit in the control device. The leg determination unit 91 selects an edge (an edge near the start or end of one carrier cycle) from among multiple edges generated in the primary AC voltage V1 to correct the timing. The leg determination unit 92 selects an edge (an edge near the start or end of one carrier cycle) from among multiple edges generated in the secondary AC voltage V2 to correct the timing.

[0109] The leg determination unit 91 selects whether to reflect the bias correction amount ΔIdcA in the primary U-phase pulse width command P1U or in the primary V-phase pulse width command P1V, based on the primary phase shift commands θ1U and θ1V. The phase shift command θ1U is a value that specifies the amount of phase shift of the primary U-phase carrier wave Ca1U. The phase shift command θ1V is a value that specifies the amount of phase shift of the primary V-phase carrier wave Ca1V.

[0110] The leg determination unit 92 selects whether to reflect the DC correction amount ΔIdcB in the secondary U-phase pulse width command P2U or the secondary V-phase pulse width command P2V, based on the secondary phase shift commands θ2U and θ2V. The phase shift command θ2U is a value that specifies the amount of phase shift of the secondary U-phase carrier wave Ca2U. The phase shift command θ2V is a value that specifies the amount of phase shift of the secondary V-phase carrier wave Ca2V.

[0111] Figure 21 shows the relationship between the phase shift amount and the edge. Figure 21 shows the case for the primary side, but the explanation for the secondary side is omitted by referring to the explanation for the primary side which will be described later.

[0112] As described above, the control device 106 in the first embodiment adjusts the primary side U-phase pulse width command P1U or the primary side V-phase pulse width command P1V up or down. Therefore, among the multiple edges generated in the primary side AC voltage V1, the edges whose timing (phase) can be adjusted are limited to two edges that occur when the triangular wave carrier wave and the pulse width command intersect at intersection r1 or intersection r2. Therefore, the leg determination unit 91 selects the edge that is closer to the start or end of one carrier period from among the two edges. The control device 106 reduces the change in the primary side DC current I1 by correcting the timing of the selected edge.

[0113] Assuming the pulse width command is near 50%, the timing of the intersection points r1 and r2 where the edges of the primary AC voltage V1 are generated is determined by the phase shift commands θ1U and θ1V that shift the phase of the carrier wave. Phase shift command θ1U specifies the amount of phase shift of the primary U-phase carrier wave Ca1U. Phase shift command θ1V specifies the amount of phase shift of the primary V-phase carrier wave Ca1V. The leg determination unit 91 determines which of the two legs (phases) on the primary side will have its timing corrected, based on the phase shift commands θ1U and θ1V. Similarly, although not specifically shown in Figure 21, the leg determination unit 92 determines which of the two legs (phases) on the secondary side will have its timing corrected, based on the phase shift commands θ2U and θ2V.

[0114] Figure 22 is a functional block diagram showing an example of a leg determination unit. The primary leg determination unit 91 and the secondary leg determination unit 92 each have the configuration shown in Figure 22. The closer the phase shift command is to ±180°, the closer the carrier wave and pulse width command intersect at the start or end of one carrier period. The leg determination unit shown in Figure 22 compares the difference between the absolute value of the U-phase phase shift command and 180° with the difference between the absolute value of the V-phase phase shift command and 180°, and determines which of the two legs (phases) requires timing correction based on which difference is larger. In the case of Figure 21, since the phase shift command θ1V is closer to 180° than the phase shift command θ1U, the determination unit adjusts the timing of the edge of the primary AC voltage V1 by adjusting the timing of the edge of the primary V-phase drive pulse generated at the intersection point r2.

[0115] Furthermore, the leg determination unit shown in Figure 22 performs two absolute value calculations to accommodate both cases where the phase shift command is defined as 0 to 360° and cases where it is defined as -180° to +180°.

[0116] The first edge is defined as an edge that occurs closer to the center of one carrier cycle, the second edge as an edge closer to the start of the one carrier cycle than the first edge, and the third edge as an edge closer to the end of the one carrier cycle than the first edge. In this case, if the time from the start of one carrier cycle to the second edge is shorter than the time from the third edge to the end of the one carrier cycle, the leg determination unit 91 adjusts the timing of the second edge without adjusting the timing of the third edge. On the other hand, if the time from the third edge to the end of the one carrier cycle is shorter than the time from the start of one carrier cycle to the second edge, the leg determination unit 91 adjusts the timing of the third edge without adjusting the timing of the second edge. The control device 106 performs a first timing specification control to reduce the change in the primary DC current I1 according to the determination result of the leg determination unit 91.

[0117] The fourth edge is defined as an edge that occurs closer to the center of one carrier cycle, the fifth edge is defined as an edge closer to the start of the one carrier cycle than the fourth edge, and the sixth edge is defined as an edge closer to the end of the one carrier cycle than the fourth edge. In this case, if the time from the start of one carrier cycle to the fifth edge is shorter than the time from the sixth edge to the end of the one carrier cycle, the leg determination unit 92 adjusts the timing of the fifth edge without adjusting the timing of the sixth edge. On the other hand, if the time from the sixth edge to the end of the one carrier cycle is shorter than the time from the start of one carrier cycle to the fifth edge, the leg determination unit 92 adjusts the timing of the sixth edge without adjusting the timing of the fifth edge. The control device 106 performs a second timing specification control to reduce the change in the primary DC current I1 according to the determination result of the leg determination unit 92.

[0118] Figure 23 is a timing chart illustrating a case where the timing of the last edge occurring in one carrier cycle of the primary AC voltage V1 is corrected by a primary V-phase drive pulse. In Figure 23, the phase shift command θ1V is close to 180° compared to the phase shift command θ1U. Therefore, the leg determination unit 91 adjusts the timing of the edge of the primary AC voltage V1 by adjusting the timing of the edge of the primary V-phase drive pulse (in this case, drive pulse g3) that occurs at the intersection r2.

[0119] Figure 24 is a timing chart illustrating a case where the timing of the first edge occurring in one carrier cycle of the secondary AC voltage V2 is corrected by a secondary V-phase drive pulse. In Figure 24, the phase shift command θ2V is close to 180° compared to the phase shift command θ2U. Therefore, the leg determination unit 92 adjusts the timing of the edge of the secondary AC voltage V2 by adjusting the timing of the edge of the secondary V-phase drive pulse (in this case, drive pulse g7) that occurs at the intersection r2.

[0120] Figure 25 is a timing chart illustrating a case where the timing of the last edge occurring in one carrier cycle of the primary AC voltage V1 is corrected by a primary U-phase drive pulse. In Figure 25, the voltage waveform and current waveform are the same as in Figure 23, but the case where the primary AC current Iac changes in the negative direction from the start of each carrier cycle is illustrated. Similar to Figure 23, the voltage waveform and current waveform have matching polarity and indicate the power mode. In the case of Figure 25, the phase shift command θ1U is close to 180° compared to the phase shift command θ1V. Therefore, the leg determination unit 91 adjusts the timing of the edge of the primary AC voltage V1 by adjusting the timing of the edge of the primary U-phase drive pulse (in this case, drive pulse g1) that occurs at the intersection r1.

[0121] Figure 26 is a timing chart illustrating a case in discontinuous current mode where the timing of the last edge occurring in one carrier cycle of the primary AC voltage V1 is corrected by the primary V-phase drive pulse. Even in the discontinuous current mode shown in Figure 26, the timing of the edge of the primary AC voltage V1 can be adjusted, similar to the continuous current mode shown in Figures 23-25. In Figure 26, the phase shift command θ1U is close to 180° compared to the phase shift command θ1V. Therefore, the leg determination unit 91 adjusts the timing of the edge of the primary AC voltage V1 by adjusting the timing of the edge of the primary U-phase drive pulse (in this case, drive pulse g3) that occurs at the intersection r2.

[0122] The present invention is not limited by the embodiments described above. The embodiments can be implemented in various other forms, and various combinations, omissions, substitutions, and modifications are possible without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents.

[0123] In this disclosure, the control device includes electronic circuits such as a CPU (Central Processing Unit), FPGA (Field Programmable Gate Array), or ASIC (Application Specific Integrated Circuit). The control device may also be a computer having memory and a processor. The control device performs the various control operations described in this specification by executing a program such as instruction code stored in memory, or by being circuit-designed for a special application. [Explanation of symbols]

[0124] 33 Magnetic Cores 100 Power Conversion Systems 102 Isolation Transformer 106 Control device 109 DC component detection unit 110 Isolated DC / DC Converter 111 First Bridge Circuit 112 Second Bridge Circuit

Claims

1. An isolation transformer having a magnetic core, A first bridge circuit having multiple switching elements is connected to the primary side of the isolation transformer, A second bridge circuit, which is connected to the secondary side of the isolation transformer and has a plurality of switching elements, The system includes a control device that, by comparing a first pulse width command for determining the pulse width of the primary AC voltage applied to the primary side of the isolation transformer with a first carrier wave, causes the primary AC voltage to be output to the first bridge circuit, and by comparing a second pulse width command for determining the pulse width of the secondary AC voltage applied to the secondary side of the isolation transformer with a second carrier wave, causes the secondary AC voltage to be output to the second bridge circuit. The control device is a power conversion system that reduces the DC component of the excitation current of the isolation transformer and the DC component of the load current obtained by subtracting the excitation current from the primary AC current flowing on the primary side of the isolation transformer or the secondary AC current flowing on the secondary side of the isolation transformer, by correcting the first pulse width command or the second pulse width command.

2. The power conversion system according to claim 1, wherein the control device corrects the first pulse width command or the second pulse width command based on the detected value of the DC component of the excitation current and the detected value of the DC component of the load current.

3. The power conversion system according to claim 2, wherein the control device corrects one of the first pulse width command and the second pulse width command based on the detected value of the DC component of the excitation current, and corrects the other pulse width command based on the detected value of the DC component of the load current.

4. The aforementioned pulse width command includes a primary-side first-phase pulse width command for determining the gate pulse width of the first-phase switching element among the plurality of switching elements of the first bridge circuit, and a primary-side second-phase pulse width command for determining the gate pulse width of the second-phase switching element. The other pulse width command includes a secondary-side first-phase pulse width command for determining the gate pulse width of the first-phase switching element among the plurality of switching elements of the second bridge circuit, and a secondary-side second-phase pulse width command for determining the gate pulse width of the second-phase switching element. The power conversion system according to claim 3, wherein the control device corrects the pulse width command of the same phase among the primary side first phase pulse width command, the primary side second phase pulse width command, the secondary side first phase pulse width command, and the secondary side second phase pulse width command based on the detected value of the DC component of the excitation current and the detected value of the DC component of the load current.

5. The power conversion system according to any one of claims 1 to 4, wherein the control device sets the sign of the correction amount reflected in the first pulse width command or the second pulse width command to a sign that reduces the DC component of the excitation current and the DC component of the load current.

6. The power conversion system according to any one of claims 1 to 4, wherein the control device corrects only one of the first pulse width command and the second pulse width command.

7. The power conversion system according to any one of claims 1 to 4, wherein the control device corrects both the first pulse width command and the second pulse width command.

8. The power conversion system according to any one of claims 1 to 4, wherein the control device corrects the first pulse width command based on the DC voltage of the first bridge circuit and corrects the second pulse width command based on the DC voltage of the second bridge circuit.

9. The control device is By adjusting the timing of the second edge, which is closer to the start of the carrier cycle than the first edge, or the third edge, which is closer to the end of the carrier cycle than the first edge, without adjusting the timing of the first edge, the change in the primary DC current flowing through the first bridge circuit is reduced. and / or, By adjusting the timing of the fifth edge, which is closer to the start of the carrier cycle than the fourth edge, or the sixth edge, which is closer to the end of the carrier cycle than the fourth edge, without adjusting the timing of the fourth edge, the change in the primary DC current flowing through the first bridge circuit is reduced. The power conversion system according to claim 1.

10. The control device is If the time from the starting point to the second edge is shorter than the time from the third edge to the ending point, the change in the primary DC current can be reduced by adjusting the timing of the second edge without adjusting the timing of the third edge. If the time from the third edge to the endpoint is shorter than the time from the starting point to the second edge, the change in the primary DC current is reduced by adjusting the timing of the third edge without adjusting the timing of the second edge. and / or, If the time from the starting point to the fifth edge is shorter than the time from the sixth edge to the ending point, the change in the primary DC current can be reduced by adjusting the timing of the fifth edge without adjusting the timing of the sixth edge. If the time from the sixth edge to the endpoint is shorter than the time from the starting point to the fifth edge, the change in the primary DC current is reduced by adjusting the timing of the sixth edge without adjusting the timing of the fifth edge. The power conversion system according to claim 9.

11. The second edge is the closest to the starting point among the multiple edges that occur in the primary AC voltage within one carrier cycle, and / or the third edge is the closest to the ending point among the multiple edges that occur in the primary AC voltage within one carrier cycle. Furthermore / or, The fifth edge is the closest to the starting point among the multiple edges that occur in the secondary AC voltage within one carrier cycle, and / or the sixth edge is the closest to the ending point among the multiple edges that occur in the secondary AC voltage within one carrier cycle. The power conversion system according to claim 10.

Images