Electric power conversion device

Abstract

The present invention addresses the problem of performing soft switching more reliably. Provided is an electric power conversion device (100) in which a control device (50) can execute a first control operation and a second control operation when it is determined that resonance currents passing through two or more switches (8), respectively, flow simultaneously in a resonance inductor (L1). The first control operation causes a high-level period of a control signal for each of the two or more switches (8) to overlap, for a predetermined period, with a dead-time period corresponding to each of two or more switching circuits (10) connected to the two or more switches (8). The second control operation determines a starting time of the high-level period of the control signal for at least one switch (8), in accordance with a load current.

Description

Power Conversion Device

[0001] The present disclosure relates to a power conversion device, and more particularly to a power conversion device capable of converting DC power into AC power.

[0002] Patent Document 1 discloses a power conversion device that converts direct current into multi-phase alternating current.

[0003] The power conversion device disclosed in Patent Document 1 includes a main switching means (power conversion circuit), two capacitors, one coil (resonant inductor), multiple auxiliary switching elements, and a control means. The main switching means is composed of a pair of main switching elements connected in series between both terminals of a DC power supply, and a main switching circuit is provided for each phase of the multi-phase AC, with the interconnection point of the pair of main switching elements serving as the output point for each phase. The two capacitors divide the voltage of the DC power supply. One end of the coil is connected to the voltage division point between the two capacitors. The multiple auxiliary switching elements connect the other end of the coil to the output points for each phase. When the control means determines that multiple phase currents are flowing through the coil, it controls the multiple auxiliary switching elements so that the current flowing in at least one phase is smaller than a preset magnitude.

[0004] In the power conversion device described in Patent Document 1, when the control means determines that multiple phase currents are flowing through the coil, it controls multiple auxiliary switch elements so that the current flowing through at least one phase is smaller than a preset magnitude, and therefore soft switching of the main switch corresponding to the at least one phase is not performed.

[0005] JP 2010-233306 A

[0006] An object of the present disclosure is to provide a power conversion device that can perform soft switching more reliably.

[0007] A power conversion device according to one aspect of the present disclosure includes a first DC terminal and a second DC terminal, a power conversion circuit, multiple AC terminals, multiple switches, multiple resonant capacitors, a resonant inductor, a regenerative capacitor, and a control device. The power conversion circuit includes multiple first switching elements and multiple second switching elements. In the power conversion circuit, multiple switching circuits are connected in parallel, with the multiple first switching elements and the multiple second switching elements connected in series in a one-to-one relationship. In the power conversion circuit, the multiple first switching elements are connected to the first DC terminals, and the multiple second switching elements are connected to the second DC terminals. The multiple AC terminals correspond one-to-one to the multiple switching circuits. Each of the multiple AC terminals is connected to a connection point between the first switching element and the second switching element in the corresponding switching circuit. The multiple switches correspond one-to-one to the multiple switching circuits. Each of the multiple switches has a first end connected to the connection point between the first switching element and the second switching element in the corresponding switching circuit, and a second end connected in common to a common connection point. The resonant capacitors correspond one-to-one to the multiple switches. Each of the multiple resonant capacitors is connected between the first end and the second DC terminal of the corresponding switch. The resonant inductor has a first end and a second end. The first end of the resonant inductor is connected to the common connection point. The regenerative capacitor has a third end and a fourth end. The third end of the regenerative capacitor is connected to the first DC terminal or the second DC terminal. The control device provides control signals whose potential changes between a high level and a low level to each of the multiple first switching elements, the multiple second switching elements, and the multiple switches. The control device is capable of executing a first control operation and a second control operation when it determines that resonant currents passing through two or more of the multiple switches simultaneously flow through the resonant inductor.The first control operation overlaps a high-level period of a control signal to each of the two or more switches with a dead-time period corresponding to each of two or more switching circuits connected to the two or more switches among the plurality of switching circuits by a predetermined period, and the second control operation determines a start point of a high-level period of a control signal to at least one switch among the plurality of switches in accordance with at least one phase of load current flowing through an AC load connected to the plurality of AC terminals.

[0008] FIG. 1 is a circuit diagram of a system including a power conversion device according to a first embodiment. FIG. 2 is a diagram illustrating time variations in duty and load current corresponding to voltage commands for each of three phases in an AC load connected to multiple AC terminals of the power conversion device. FIG. 3 is an explanatory diagram illustrating the operation of a control device in the power conversion device when performing a basic operation in a case where the load current is greater than 0 and the resonant capacitor is charging. FIG. 4 is another explanatory diagram illustrating the operation of a control device in the power conversion device when performing a basic operation in a case where the load current is greater than 0 and the resonant capacitor is charging. FIG. 5 is an explanatory diagram illustrating a first current threshold and a second current threshold used by the control device in the power conversion device. FIG. 6 is an explanatory diagram illustrating the operation of a control device in the power conversion device when performing a basic operation in a case where the load current is greater than 0 and the resonant capacitor is discharging. FIG. 7 is an explanatory diagram illustrating the operation of a control device in the power conversion device when performing a basic operation in a case where the load current is less than 0 and the resonant capacitor is discharging. FIG. 8 is an explanatory diagram illustrating the operation of a control device in the power conversion device when performing a basic operation in a case where the load current is less than 0 and the resonant capacitor is charging. Fig. 9 is a timing chart showing a case where the control device executes the first control operation, the second control operation, and the third control operation in the power conversion device of the same. Fig. 10 is a timing chart showing a case where the control device does not execute the first control operation, the second control operation, and the third control operation in the power conversion device of the same. Fig. 11 is a timing chart for explaining an example where the control device estimates that two-phase resonant currents overlap in the power conversion device of the same. Fig. 12 is a timing chart for explaining another example where the control device estimates that two-phase resonant currents overlap in the power conversion device of the same. Fig. 13 is a timing chart for explaining another example where the control device estimates that two-phase resonant currents overlap in the power conversion device of the same. Fig. 14 is a diagram showing another example of time changes in duties and time changes in load currents corresponding to voltage commands for each of three phases in an AC load connected to multiple AC terminals of the power conversion device of the same. Fig. 15 is a timing chart for explaining yet another example where the control device estimates that two-phase resonant currents overlap in the power conversion device of the same.Fig. 16 is a timing chart showing a case where the control device executes the first control operation, the second control operation, and the third control operation in the power conversion device of the same. Fig. 17 is a timing chart showing a case where the control device does not execute the first control operation, the second control operation, and the third control operation in the power conversion device of the same. Fig. 18 is a timing chart showing a case where the control device executes the first control operation and the third control operation in the power conversion device of the same. Fig. 19 is a timing chart showing a case where the control device does not execute the first control operation and the third control operation in the power conversion device of the same. Fig. 20 is a timing chart for explaining another example where the control device estimates that three-phase resonant currents overlap in the power conversion device of the same. Fig. 21 is a timing chart for explaining yet another example where the control device estimates that three-phase resonant currents overlap in the power conversion device of the same. Fig. 22 is a timing chart for explaining another example where the control device estimates that two-phase resonant currents overlap in the power conversion device of the same. Fig. 23 is a timing chart for explaining yet another example in which the control device in the power conversion device estimates that two-phase resonant currents overlap. Fig. 24 is a timing chart in a case in which the control device in the power conversion device executes the first control operation and the third control operation. Fig. 25 is a timing chart in a case in which the control device in the power conversion device does not execute the first control operation and the third control operation. Fig. 26 is a timing chart in a case in which the control device executes the first control operation and the second control operation in the power conversion device according to Modification 1 of Embodiment 1. Fig. 27 is a timing chart in a case in which the control device executes the first control operation and the second control operation in the power conversion device according to Modification 2 of Embodiment 1. Fig. 28 is a timing chart in a case in which the control device does not execute the first control operation and the second control operation in the power conversion device according to Modification 3 of Embodiment 1. Fig. 29 is a timing chart in a case in which the control device executes the first control operation and the second control operation in the power conversion device according to Modification 3 of Embodiment 1. FIG. 30 is a timing chart when the control device executes the first control operation, the second control operation, and the third control operation in the power conversion device according to the fourth modification of the first embodiment.FIG. 31 is a timing chart showing a case where a control device executes a first control operation and a second control operation in a power conversion device according to a fifth modification of the first embodiment. FIG. 32 is a timing chart showing a case where a control device executes a first control operation and a second control operation in the same power conversion device. FIG. 33 is a diagram illustrating an operation that is a prerequisite for a power conversion device according to a sixth modification of the first embodiment. FIG. 34 is a timing chart showing a case where a control device executes a first control operation, a second control operation, and a third control operation in the same power conversion device. FIG. 35 is a timing chart showing a case where a control device executes a first control operation and a second control operation in a power conversion device according to a seventh modification of the first embodiment. FIG. 36 is a circuit diagram of a system including a power conversion device according to an eighth modification of the first embodiment. FIG. 37 is a circuit diagram of a system including a power conversion device according to a ninth modification of the first embodiment. FIG. 38 is a circuit diagram of a system including a power conversion device according to a tenth modification of the first embodiment. FIG. 39 is a circuit diagram of a system including a power conversion device according to an eleventh modification of the first embodiment. FIG. 40 is a circuit diagram of a system including a power conversion device according to a twelfth modification of the first embodiment. Fig. 41 is a circuit diagram of a system including a power conversion device according to a thirteenth modification of embodiment 1. Fig. 42 is a circuit diagram of a system including a power conversion device according to embodiment 2. Fig. 43 is a circuit diagram of a system including a power conversion device according to embodiment 3.

[0009] First Embodiment A power conversion device 100 according to a first embodiment will be described below with reference to FIGS.

[0010] (1) Overall Configuration of the Power Conversion Device As shown in FIG. 1 , the power conversion device 100 includes a first DC terminal 31, a second DC terminal 32, and multiple (e.g., three) AC terminals 41. A DC power source E1 is connected between the first DC terminal 31 and the second DC terminal 32, and an AC load RA1 is connected to the multiple AC terminals 41. The AC load RA1 is, for example, a three-phase motor. The power conversion device 100 converts DC output from the DC power source E1 into AC power and outputs it to the AC load RA1. The DC power source E1 includes, for example, a solar cell or a fuel cell. The DC power source E1 may also include a DC-DC converter. In the power conversion device 100, when the multiple AC terminals 41 are three AC terminals 41, the AC power is, for example, three-phase AC power having a U phase, a V phase, and a W phase.

[0011] The power conversion device 100 includes a power conversion circuit 11, a plurality of (e.g., three) switches 8, a plurality of (e.g., three) resonant capacitors 9, a regenerative capacitor 15, a resonant inductor L1, and a control device 50. The power conversion device 100 also includes a protection circuit 17 and a capacitor C10. Each of the plurality of switches 8 is, for example, a bidirectional switch.

[0012] The power conversion circuit 11 has a plurality of (e.g., three) first switching elements 1 and a plurality of (e.g., three) second switching elements 2. In the power conversion circuit 11, a plurality of (e.g., three) switching circuits 10, each having a plurality of first switching elements 1 and a plurality of second switching elements 2 connected in series in a one-to-one relationship, are connected in parallel to one another. In the power conversion circuit 11, the plurality of first switching elements 1 are connected to first DC terminals 31, and the plurality of second switching elements 2 are connected to second DC terminals 32. The plurality of AC terminals 41 correspond one-to-one to the plurality of switching circuits 10. Each of the plurality of AC terminals 41 is connected to a connection point 3 between the first switching elements 1 and the second switching elements 2 in the corresponding switching circuit 10. The plurality of switches 8 correspond one-to-one to the plurality of switching circuits 10. A first end 81 of each of the plurality of switches 8 is connected to the connection point 3 between the first switching elements 1 and the second switching elements 2 in the corresponding switching circuit 10. The plurality of resonant capacitors 9 correspond one-to-one to the plurality of switches 8. Each of the multiple resonant capacitors 9 is connected between the first end 81 and the second DC terminal 32 of the corresponding switch 8. The resonant inductor L1 has a first end and a second end, and the first end is connected to the common connection point 25. The regenerative capacitor 15 has a third end 153 and a fourth end 154. The regenerative capacitor 15 has the third end 153 connected to the second DC terminal 32, and the fourth end 154 connected to the common connection point 25 via the resonant inductor L1. The control device 50 controls the multiple first switching elements 1, the multiple second switching elements 2, and the multiple switches 8.

[0013] (2) Details of the Power Conversion Device Hereinafter, for convenience of explanation, the switching circuits 10 corresponding to the U phase, V phase, and W phase of the multiple switching circuits 10 may be referred to as switching circuit 10U, switching circuit 10V, and switching circuit 10W, respectively. Hereinafter, the first switching element 1 and the second switching element 2 of switching circuit 10U may be referred to as first switching element 1U and second switching element 2U. Hereinafter, the first switching element 1 and the second switching element 2 of switching circuit 10V may be referred to as first switching element 1V and second switching element 2V. Hereinafter, the first switching element 1 and the second switching element 2 of switching circuit 10W may be referred to as first switching element 1W and second switching element 2W. In the following, the connection point 3 between the first switching element 1U and the second switching element 2U will be referred to as connection point 3U, the connection point 3 between the first switching element 1V and the second switching element 2V will be referred to as connection point 3V, and the connection point 3 between the first switching element 1W and the second switching element 2W will be referred to as connection point 3W. In the following, the AC terminal 41 connected to connection point 3U will be referred to as AC terminal 41U, the AC terminal 41 connected to connection point 3V will be referred to as AC terminal 41V, and the AC terminal 41 connected to connection point 3W will be referred to as AC terminal 41W. In the following, the resonant capacitor 9 connected in parallel to the second switching element 2U will be referred to as resonant capacitor 9U, the resonant capacitor 9 connected in parallel to the second switching element 2V will be referred to as resonant capacitor 9V, and the resonant capacitor 9 connected in parallel to the second switching element 2W will be referred to as resonant capacitor 9W. In the following, the switch 8 connected to connection point 3U may be referred to as switch 8U, the switch 8 connected to connection point 3V may be referred to as switch 8V, and the switch 8 connected to connection point 3W may be referred to as switch 8W.

[0014] In the power conversion device 100, a high-potential output terminal (positive electrode) of the DC power supply E1 is connected to the first DC terminal 31, and a low-potential output terminal (negative electrode) of the DC power supply E1 is connected to the second DC terminal 32. In addition, in the power conversion device 100, for example, a U-phase terminal, a V-phase terminal, and a W-phase terminal of an AC load RA1 are connected to three AC terminals 41U, 41V, and 41W, respectively.

[0015] In the power conversion circuit 11, each of a plurality of (e.g., three) first switching elements 1 and a plurality of (e.g., three) second switching elements 2 has a control terminal, a first main terminal, and a second main terminal. The control terminals of the plurality of first switching elements 1 and the plurality of second switching elements 2 are connected to a control device 50. In each of the plurality of switching circuits 10 of the power conversion device 100, a first main terminal of the first switching element 1 is connected to a first DC terminal 31, a second main terminal of the first switching element 1 is connected to a first main terminal of the second switching element 2, and a second main terminal of the second switching element 2 is connected to a second DC terminal 32. In each of the plurality of switching circuits 10, the first switching element 1 is a high-side switching element (P-side switching element), and the second switching element 2 is a low-side switching element (N-side switching element). Each of the plurality of first switching elements 1 and the plurality of second switching elements 2 is, for example, an IGBT (Insulated Gate Bipolar Transistor). Therefore, the control terminal, the first main terminal and the second main terminal of each of the plurality of first switching elements 1 and the plurality of second switching elements 2 are the gate terminal, the collector terminal and the emitter terminal, respectively.

[0016] The power conversion circuit 11 further includes a plurality of (three) first diodes 4 connected in anti-parallel to the plurality of (three) first switching elements 1 in a one-to-one relationship, and a plurality of (three) second diodes 5 connected in anti-parallel to the plurality of (three) second switching elements 2 in a one-to-one relationship. In each of the plurality of first diodes 4, the anode of the first diode 4 is connected to the second main terminal (emitter terminal) of the first switching element 1 corresponding to the first diode 4, and the cathode of the first diode 4 is connected to the first main terminal (collector terminal) of the first switching element 1 corresponding to the first diode 4. In each of the plurality of second diodes 5, the anode of the second diode 5 is connected to the second main terminal (emitter terminal) of the second switching element 2 corresponding to the second diode 5, and the cathode of the second diode 5 is connected to the first main terminal (collector terminal) of the second switching element 2 corresponding to the second diode 5.

[0017] A connection point 3U between the first switching element 1U and the second switching element 2U is connected to, for example, the U-phase terminal of the AC load RA1 via an AC terminal 41U. A connection point 3V between the first switching element 1V and the second switching element 2V is connected to, for example, the V-phase of the AC load RA1 via an AC terminal 41V. A connection point 3W between the first switching element 1W and the second switching element 2W is connected to, for example, the W-phase of the AC load RA1 via an AC terminal 41W.

[0018] The multiple resonant capacitors 9 correspond one-to-one to the multiple switches 8. Each of the multiple resonant capacitors 9 is connected between the first end 81 and the second DC terminal 32 of the corresponding switch 8. The power conversion device 100 has multiple resonant circuits. The multiple resonant circuits include a resonant circuit having a resonant capacitor 9U and a resonant inductor L1, a resonant circuit having a resonant capacitor 9V and a resonant inductor L1, and a resonant circuit having a resonant capacitor 9W and a resonant inductor L1. The multiple resonant circuits share the resonant inductor L1 in common.

[0019] Each of the multiple switches 8 has, for example, two IGBTs, a first IGBT 6 and a second IGBT 7, connected in anti-parallel. In each of the switches 8, the collector terminal of the first IGBT 6 is connected to the emitter terminal of the second IGBT 7, and the emitter terminal of the first IGBT 6 is connected to the collector terminal of the second IGBT 7. In each of the multiple switches 8, the emitter terminal of the first IGBT 6 is connected to the connection point 3 of the switching circuit 10 corresponding to the switch 8 having the first IGBT 6. In each of the multiple switches 8, the collector terminal of the second IGBT 7 is connected to the connection point 3 of the switching circuit 10 corresponding to the switch 8 having the second IGBT 7. The switch 8U is connected to the connection point 3U between the first switching element 1U and the second switching element 2U. The switch 8V is connected to the connection point 3V between the first switching element 1V and the second switching element 2V. The switch 8W is connected to a connection point 3W between the first switching element 1W and the second switching element 2W. Hereinafter, for convenience of explanation, the first IGBT 6 and the second IGBT 7 of the switch 8U may be referred to as the first IGBT 6U and the second IGBT 7U, the first IGBT 6 and the second IGBT 7 of the switch 8V may be referred to as the first IGBT 6V and the second IGBT 7V, and the first IGBT 6 and the second IGBT 7 of the switch 8W may be referred to as the first IGBT 6W and the second IGBT 7W, respectively.

[0020] The plurality of switches 8 are controlled by the control device 50. In other words, the first IGBT 6U, the second IGBT 7U, the first IGBT 6V, the second IGBT 7V, the first IGBT 6W, and the second IGBT 7W are controlled by the control device 50.

[0021] The resonant inductor L1 has a first end and a second end. The first end of the resonant inductor L1 is connected to the common connection point 25. The second end of the resonant inductor L1 is connected to the fourth end 154 of the regenerative capacitor 15.

[0022] The regenerative capacitor 15 is connected between the second end of the resonant inductor L1 and the second DC terminal 32. The regenerative capacitor 15 is, for example, a film capacitor.

[0023] The protection circuit 17 includes a third diode 13 and a fourth diode 14. The third diode 13 is connected between the common connection point 25 and the first DC terminal 31. In the third diode 13, the anode of the third diode 13 is connected to the common connection point 25. In addition, in the third diode 13, the cathode of the third diode 13 is connected to the first DC terminal 31. The fourth diode 14 is connected between the common connection point 25 and the second DC terminal 32. In the fourth diode 14, the anode of the fourth diode 14 is connected to the second DC terminal 32. In the fourth diode 14, the cathode of the fourth diode 14 is connected to the common connection point 25. Therefore, the fourth diode 14 is connected in series with the third diode 13.

[0024] The capacitor C10 is connected between the first DC terminal 31 and the second DC terminal 32, and is connected in parallel to the power conversion circuit 11. The capacitor C10 is, for example, an electrolytic capacitor.

[0025] The control device 50 controls a plurality of first switching elements 1, a plurality of second switching elements 2, and a plurality of switches 8. The execution entity of the control device 50 includes a computer system. The computer system has one or more computers. The computer system is mainly composed of a processor and memory as hardware. The processor executes a program recorded in the memory of the computer system, thereby realizing the function of the control device 50 as the execution entity in the present disclosure. The program may be pre-recorded in the memory of the computer system, provided via a telecommunications line, or recorded and provided on a non-transitory recording medium such as a memory card, optical disk, or hard disk drive (magnetic disk) readable by the computer system. The processor of the computer system is composed of one or more electronic circuits including a semiconductor integrated circuit (IC) or a large-scale integrated circuit (LSI). The multiple electronic circuits may be integrated into a single chip or distributed across multiple chips. The multiple chips may be integrated into a single device or distributed across multiple devices.

[0026] The control device 50 outputs control signals SU1, SV1, and SW1 that control the on / off of the multiple first switching elements 1U, 1V, and 1W, respectively. Each of the control signals SU1, SV1, and SW1 is, for example, a PWM (Pulse Width Modulation) signal whose potential level changes between a first potential level (hereinafter also referred to as a low level) and a second potential level (hereinafter also referred to as a high level) that is higher than the first potential level. The first switching elements 1U, 1V, and 1W are turned on when the control signals SU1, SV1, and SW1 are at a high level, and turned off when the control signals SU1, SV1, and SW1 are at a low level. The control device 50 also outputs control signals SU2, SV2, and SW2 that control the on / off of the multiple second switching elements 2U, 2V, and 2W, respectively. Each of the control signals SU2, SV2, and SW2 is, for example, a PWM signal whose potential level changes between a first potential level (hereinafter also referred to as a low level) and a second potential level (hereinafter also referred to as a high level) that is higher than the first potential level. The second switching elements 2U, 2V, and 2W are turned on when the control signals SU2, SV2, and SW2 are at a high level, and turned off when they are at a low level.

[0027] The control device 50 uses a sawtooth carrier signal (see FIG. 3 ) to generate control signals SU1, SV1, and SW1 corresponding to the multiple first switching elements 1U, 1V, and 1W, respectively, and control signals SU2, SV2, and SW2 corresponding to the multiple second switching elements 2U, 2V, and 2W, respectively. More specifically, the control device 50 generates the control signals SU1 and SU2 to be provided to the first switching element 1U and the second switching element 2U, respectively, based on at least the carrier signal and a U-phase voltage command. The control device 50 also generates the control signals SV1 and SV2 to be provided to the first switching element 1V and the second switching element 2V, respectively, based on at least the carrier signal and a V-phase voltage command. The control device 50 also generates the control signals SW1 and SW2 to be provided to the first switching element 1W and the second switching element 2W, respectively, based on at least the carrier signal and a W-phase voltage command. The U-phase voltage command, V-phase voltage command, and W-phase voltage command are, for example, sinusoidal signals whose phases differ by 120° from each other, and whose amplitudes (voltage command values) change over time. The waveform of the carrier signal is not limited to a sawtooth waveform and may be, for example, a triangular wave or a sawtooth wave obtained by inverting the sawtooth wave shown in FIG. 3 . The U-phase voltage command, V-phase voltage command, and W-phase voltage command each have the same cycle length. The U-phase voltage command, V-phase voltage command, and W-phase voltage command each have a longer cycle length than the carrier signal.

[0028] The duty of the control signals SU1 and SU2 provided by the control device 50 to the first switching element 1U and the second switching element 2U, respectively, varies based on the U-phase voltage command. In FIG. 2, the duty of the control signal SU1 is shown as the U-phase duty. The control device 50 (see FIG. 1) compares the U-phase voltage command with a carrier signal to generate the control signal SU1 provided to the first switching element 1U. The control device 50 also inverts the control signal SU1 provided to the first switching element 1U to generate the control signal SU2 provided to the second switching element 2U. The control device 50 also sets a dead time Td (see FIG. 3) between the period when the control signal SU1 is at a high level and the period when the control signal SU2 is at a high level so that the on periods of the first switching element 1U and the second switching element 2U do not overlap.

[0029] The duties of the control signals SV1 and SV2 provided by the control device 50 to the first switching element 1V and the second switching element 2V, respectively, vary based on the V-phase voltage command. In FIG. 2, the duty of the control signal SV1 is shown as the V-phase duty. The control device 50 (see FIG. 1) compares the V-phase voltage command with a carrier signal to generate the control signal SV1 provided to the first switching element 1V. The control device 50 also inverts the control signal SV1 provided to the first switching element 1V to generate the control signal SV2 provided to the second switching element 2V. The control device 50 also sets a dead time period Td (see FIG. 3) between the period when the control signal SV1 is at a high level and the period when the control signal SV2 is at a high level so that the on periods of the first switching element 1V and the second switching element 2V do not overlap.

[0030] The duties of the control signals SW1 and SW2 provided by the control device 50 to the first switching element 1W and the second switching element 2W, respectively, vary based on the W-phase voltage command. In FIG. 2, the duty of the control signal SW1 is shown as the W-phase duty. The control device 50 (see FIG. 1) compares the W-phase voltage command with a carrier signal to generate the control signal SW1 provided to the first switching element 1W. The control device 50 also inverts the control signal SW1 provided to the first switching element 1W to generate the control signal SW2 provided to the second switching element 2W. The control device 50 also sets a dead time Td (see FIG. 4) between the period when the control signal SW1 is at a high level and the period when the control signal SW2 is at a high level so that the on periods of the first switching element 1W and the second switching element 2W do not overlap.

[0031] The U-phase voltage command, V-phase voltage command, and W-phase voltage command are, for example, sinusoidal signals that are 120° out of phase with each other, and their amplitudes change over time. Therefore, the duty of control signal SU1 (U-phase duty), the duty of control signal SV1 (V-phase duty), and the duty of control signal SW1 (W-phase duty) change like sinusoidal waves that are 120° out of phase with each other, as shown in Figure 2. Similarly, the duty of control signal SU2, the duty of control signal SV2, and the duty of control signal SW2 change like sinusoidal waves that are 120° out of phase with each other.

[0032] The control device 50 generates the control signals SU1, SU2, SV1, SV2, SW1, and SW2 based on the carrier signal, the voltage commands, and information about the state of the AC load RA1. For example, if the AC load RA1 is a three-phase motor, the information about the state of the AC load RA1 includes, for example, detection values ​​from a plurality of current sensors that detect output currents (hereinafter also referred to as load currents) iU, iV, and iW flowing through the U-phase, V-phase, and W-phase of the AC load RA1, respectively.

[0033] The plurality of switches 8, the resonant inductor L1, the plurality of resonant capacitors 9, and the regenerative capacitor 15 are provided to perform zero-voltage soft switching of the plurality of first switching elements 1 and the plurality of second switching elements 2.

[0034] In the power conversion device 100 , the control device 50 controls the plurality of switches 8 in addition to the plurality of first switching elements 1 and second switching elements 2 of the power conversion circuit 11 .

[0035] The control device 50 generates control signals SU6, SU7, SV6, SV7, SW6, and SW7 that control the on / off of the first IGBT6U, the second IGBT7U, the first IGBT6V, the second IGBT7V, the first IGBT6W, and the second IGBT7W, respectively, and outputs them to the gate terminals of the first IGBT6U, the second IGBT7U, the first IGBT6V, the second IGBT7V, the first IGBT6W, and the second IGBT7W, respectively.

[0036] When the first IGBT 6U is in the ON state and the second IGBT 7U is in the OFF state, the switch 8U allows a charging current to pass through the path of the regenerative capacitor 15, resonant inductor L1, switch 8U, and resonant capacitor 9U. The charging current is a current that charges the resonant capacitor 9U. When the first IGBT 6U is in the OFF state and the second IGBT 7U is in the ON state, the switch 8U allows a discharging current to pass through the path of the resonant capacitor 9U, switch 8U, resonant inductor L1, and regenerative capacitor 15. The discharging current is a current that discharges the charge of the resonant capacitor 9U.

[0037] When the first IGBT 6V is in the ON state and the second IGBT 7V is in the OFF state, the switch 8V allows a charging current to pass through the path of the regenerative capacitor 15-resonant inductor L1-switch 8V-resonant capacitor 9V. The charging current is a current that charges the resonant capacitor 9V. When the first IGBT 6V is in the OFF state and the second IGBT 7V is in the ON state, the switch 8V allows a discharging current to pass through the path of the resonant capacitor 9V-switch 8V-resonant inductor L1-regenerative capacitor 15. The discharging current is a current that discharges the charge of the resonant capacitor 9V.

[0038] When the first IGBT 6W is in the ON state and the second IGBT 7W is in the OFF state, the switch 8W allows a charging current to pass through the path of the regenerative capacitor 15, resonant inductor L1, switch 8W, and resonant capacitor 9W. The charging current is a current that charges the resonant capacitor 9W. When the first IGBT 6W is in the OFF state and the second IGBT 7W is in the ON state, the switch 8W allows a discharging current to pass through the path of the resonant capacitor 9W, switch 8W, resonant inductor L1, and regenerative capacitor 15. The discharging current is a current that discharges the charge of the resonant capacitor 9W.

[0039] (3) Operation of the Power Conversion Device In the following description, the polarity of current iL1 flowing through resonant inductor L1 will be defined as positive when flowing in the direction of the arrow in Fig. 1, and as negative when flowing in the direction opposite to the direction of the arrow in Fig. 1. Also, in the following description, the polarity of load currents iU, iV, and iW flowing through the U, V, and W phases of AC load RA1 will be defined as positive when flowing in the direction of the arrow in Fig. 1, and as negative when flowing in the direction opposite to the direction of the arrow in Fig. 1. Also, in the following description, the polarity of currents i9U, i9V, and i9W flowing through resonant capacitors 9U, 9V, and 9W will be defined as positive when flowing in the direction of the arrow in Fig. 1, and as negative when flowing in the direction opposite to the direction of the arrow in Fig. 1. Therefore, in the case of a discharge operation in which the resonant capacitors 9U, 9V, and 9W are discharged, the polarity of the currents i9U, i9V, and i9W is positive, and in the case of a charge operation in which the resonant capacitors 9U, 9V, and 9W are charged, the polarity of the currents i9U, i9V, and i9W is negative.

[0040] In the power conversion device 100, for example, the first IGBT 6U of the switch 8U may change from an ON state in which the first IGBT 6U of the switch 8U is in an ON state and current iL1 is flowing through the resonant inductor L1 with positive polarity to an OFF state. In this case, the current iL1 flowing through the resonant inductor L1 is regenerated to the power conversion circuit 11 via the third diode 13 until the energy of the resonant inductor L1 is consumed and the current iL1 becomes zero. In addition, in the power conversion device 100, for example, the second IGBT 7U of the switch 8U may change from an ON state in which the second IGBT 7U of the switch 8U is in an ON state and current iL1 is flowing through the resonant inductor L1 with negative polarity to an OFF state. In this case, the current iL1 flowing through the resonant inductor L1 flows through the path of the fourth diode 14, the resonant inductor L1, and the regenerative capacitor 15 until the energy of the resonant inductor L1 is consumed and the current iL1 becomes zero.

[0041] Furthermore, in the power conversion device 100, for example, a state in which the first IGBT 6V of the switch 8V is in the on state and current iL1 is flowing through the resonant inductor L1 with positive polarity may change to an off state, where the first IGBT 6V of the switch 8V is in the on state. In this case, the current iL1 flowing through the resonant inductor L1 is regenerated to the power conversion circuit 11 via the third diode 13 until the energy of the resonant inductor L1 is consumed and the current iL1 becomes zero. Also, in the power conversion device 100, for example, a state in which the second IGBT 7V of the switch 8V is in the on state and current iL1 is flowing through the resonant inductor L1 with negative polarity may change to an off state, where the current iL1 flowing through the resonant inductor L1 flows through the path of the fourth diode 14, resonant inductor L1, and regenerative capacitor 15 until the energy of the resonant inductor L1 is consumed and the current iL1 becomes zero.

[0042] Furthermore, in the power conversion device 100, for example, the first IGBT 6W of the switch 8W may be switched from an ON state in which the current iL1 is flowing through the resonant inductor L1 with positive polarity to an OFF state in which the first IGBT 6W of the switch 8W is switched. In this case, the current iL1 flowing through the resonant inductor L1 is regenerated to the power conversion circuit 11 via the third diode 13 until the energy of the resonant inductor L1 is consumed and the current iL1 becomes zero. Furthermore, in the power conversion device 100, for example, the second IGBT 7W of the switch 8W may be switched from an ON state in which the current iL1 is flowing through the resonant inductor L1 with negative polarity to an OFF state in which the second IGBT 7W of the switch 8W is switched. In this case, the current iL1 flowing through the resonant inductor L1 flows through the path of the fourth diode 14, the resonant inductor L1, and the regenerative capacitor 15 until the energy of the resonant inductor L1 is consumed and the current iL1 becomes zero.

[0043] The control device 50 sets a dead time period Td between the high level period of the control signals SU1, SV1, SW1 to the first switching elements 1U, 1V, 1W and the high level period of the control signals SU2, SV2, SW2 to the second switching elements 2U, 2V, 2W for each of the multiple switching circuits 10.

[0044] 1 to 25, the basic operation of the zero voltage soft switching of each of the plurality of first switching elements 1 and the plurality of second switching elements 2 is described below. The basic operation is an operation performed when a resonant current passing through two or more of the plurality of switches 8 does not flow simultaneously through the resonant inductor L1. After describing the basic operation, the operation performed when the control device 50 determines that a resonant current passing through two or more of the plurality of switches 8 will flow simultaneously will be described.

[0045] (3.1) Basic Operation In zero voltage soft switching of the first switching element 1, the voltage across the first switching element 1 must be set to zero immediately before the first switching element 1, which is the target of zero voltage soft switching, is turned on. In addition, in zero voltage soft switching of the second switching element 2, the voltage across the second switching element 2 must be set to zero immediately before the second switching element 2, which is the target of zero voltage soft switching, is turned on. Hereinafter, the switching element (first switching element 1 or second switching element 2) that is the target of zero voltage soft switching is also referred to as the target switching element.

[0046] The basic operation of the control device 50 differs depending on the polarity (positive / negative) of the load current flowing through the AC terminal 41 connected to the target switching element and the operation (charging operation / discharging operation) of the resonant capacitor 9 connected in series or parallel to the target switching element. The load current is positive when flowing from the AC terminal 41 to the AC load RA1, and negative when flowing from the AC load RA1 to the AC terminal 41. When the resonant capacitor 9 is charging, the voltage across the resonant capacitor 9 increases. When the resonant capacitor 9 is discharging, the voltage across the resonant capacitor 9 decreases. The voltage across each of the multiple second switching elements 2 is the same as the voltage across the resonant capacitor 9 connected in parallel to the second switching element 2.

[0047] (3.1.1) Operation for Soft-Switching the First Switching Element When Load Current > 0 When the target of soft switching is the first switching element 1 (hereinafter also referred to as the target first switching element 1), and the polarity of the load current flowing through the AC terminal 41 connected to the target first switching element 1 is positive, the control device 50 turns on the first IGBT 6 corresponding to the target first switching element 1. As a result, the control device 50 causes resonance between the resonant inductor L1 and the resonant capacitor 9 connected to the target first switching element 1, charging the resonant capacitor 9 from the regenerative capacitor 15 and setting the voltage across the target first switching element 1 to zero. In this way, the power conversion device 100 can achieve zero-voltage soft switching of the target first switching element 1.

[0048] 3 shows the control signals SU1 and SU2 given from the control device 50 to the first switching element 1U and the second switching element 2U of the switching circuit 10U, respectively, when the target first switching element is the first switching element 1U of the switching circuit 10U. Also shown in FIG. 3 are the control signal SU6 given from the control device 50 to the first IGBT 6U of the switch 8U, the load current iU flowing in the U-phase of the AC load RA1, the current iL1 flowing in the resonant inductor L1, and the voltage V across the first switching element 1U. １Ｕ and the voltage V across the second switching element 2U. ２Ｕ 3 also illustrates the control signals SV1 and SV2 given from the control device 50 to the first switching element 1V and the second switching element 2V of the switching circuit 10V, respectively, in the case where the target first switching element is the first switching element 1V of the switching circuit 10V. Also, FIG. 3 also illustrates the control signal SV6 given from the control device 50 to the first IGBT 6V of the switch 8V, the load current iV flowing in the V-phase of the AC load RA1, the current iL1 flowing in the resonant inductor L1, and the voltage V across the first switching element 1V. １Ｖ and the voltage V across the second switching element 2V. ２Ｖ and are illustrated.

[0049] 3 also shows a dead time period Td set in the control device 50 to prevent the first switching element 1 and the second switching element 2, which are in phase, from being turned on simultaneously. Also shown in FIG. 3 is an additional time Tau set in the control device 50 for the control signal SU6 of the first IGBT 6U of the switch 8U, and an additional time Tav set in the control device 50 for the control signal SV6 of the first IGBT 6V of the switch 8V. The additional time Tau and the additional time Tav will be described later.

[0050] 4 illustrates the control signals SW1 and SW2 given from the control device 50 to the first switching element 1W and the second switching element 2W of the switching circuit 10W, respectively, when the target first switching element is the first switching element 1W of the switching circuit 10W. Also, FIG. 4 illustrates the control signal SW6 given from the control device 50 to the first IGBT 6W of the switch 8W, and the load current iW flowing through the W-phase of the AC load RA1. Also, FIG. 4 illustrates the current iL1 flowing through the resonant inductor L1. Also, FIG. 4 illustrates the voltage V across the first switching element 1W. １Ｗ and the voltage V across the second switching element 2W. ２Ｗ 4, the voltage value of the DC power supply E1 is shown as Vd.

[0051] 4 also shows a dead time period Td set in the control device 50 to prevent the first switching element 1W and the second switching element 2W from being turned on simultaneously. Also, Fig. 4 shows an additional time Taw set in the control device 50 for the control signal SW6 of the first IGBT 6W of the switch 8W. The additional time Taw will be described later.

[0052] As shown in FIG. 3 , the additional time Tau is set to advance the start point (time t1) of the high-level period of the control signal SU6 relative to the start point (time t2) of the dead time period Td, thereby making the high-level period of the control signal SU6 longer than the dead time period Td. The length of the additional time Tau is set based on the value of the load current iU. To start LC resonance at the start point (time t2) of the dead time period Td, it is desirable that the value of the current iL1 match the value of the load current iU at the start point (time t2) of the dead time period Td. This is because, while iL1 < iU, all of the current iL1 flows through the AC load RA1, preventing the resonant capacitor 9U from being charged. The end point of the high-level period of the control signal SU6 may be the same as or later than the end point (time t3) of the dead time period Td. FIG. 3 shows an example in which the end point of the high-level period of the control signal SU6 is set to the same as the end point (time t3) of the dead time period Td. The control device 50 sets the high level period of the control signal SU6 to Tau+Td. In the switching circuit 10U, the voltage V across the second switching element 2U is ２Ｕ becomes Vd at the end of the dead time period Td (time t3), and the voltage V across the first switching element 1U １Ｕ becomes zero at the end of the dead time period Td (time t3). In the example of FIG. 3 , the current iL1 flowing through the resonant inductor L1 begins to flow at the start of the high-level period of the control signal SU6 (time t1) and becomes zero at time t4, when the additional time Tau has elapsed since the end of the dead time period Td (time t3). With regard to the current iL1, since iL1≧iU holds true from the start of the dead time period Td (time t2), the current iL1 in the shaded region of the current waveform in the fifth row from the top in FIG. 3 flows into the resonant capacitor 9U, causing LC resonance. After the end of the dead time period Td (time t3), the current iL1 is regenerated in the power conversion circuit 11 via the third diode 13 directly connected to the resonant inductor L1.

[0053] As described above, the control device 50 determines the additional time Tau based on the load current iU so that iL1 = iU at the start of the dead time period Td (time t2) and the end of the resonant half cycle at the end of the dead time period Td. More specifically, the control device 50 determines the additional time Tau by calculating Tau = iU × (L / V15) using, for example, the detection result of the load current iU by a current sensor, its signal-processed value, or an estimated value of the load current iU, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. The detection result of the load current iU or its signal-processed value at this time is the detection value at the carrier cycle to which the additional time Tau is added, or the timing closest to that carrier cycle. The estimated value of the load current iU at this time is, for example, the load current iU estimated for the carrier cycle to which the additional time Tau is added. The resonance half period in the case of basic operation is half the resonance period, which is the reciprocal of the resonance frequency of the resonance circuit including the resonance inductor L1 and one resonance capacitor 9. Therefore, if the inductance of the resonance inductor L1 is L and the capacitance of the resonance capacitor 9 is C, the resonance half period is π × (L C) １／２ In the control device 50, the resonance half cycle during basic operation is set to be equal to the length of the dead time period Td, for example.

[0054] As shown in FIG. 3 , the additional time Tav is set to advance the start point (time t5) of the high-level period of the control signal SV6 relative to the start point (time t6) of the dead time period Td, thereby making the high-level period of the control signal SV6 longer than the dead time period Td. The length of the additional time Tav is set based on the value of the load current iV. To start LC resonance at the start point (time t6) of the dead time period Td, it is desirable that the value of the current iL1 match the value of the load current iV at the start point (time t6) of the dead time period Td. This is because, while iL1 < iV, all of the current iL1 flows through the AC load RA1, preventing the resonant capacitor 9V from being charged. The end point of the high-level period of the control signal SV6 may be the same as or later than the end point (time t7) of the dead time period Td. FIG. 3 shows an example in which the end point of the high-level period of the control signal SV6 is set to the same as the end point (time t7) of the dead time period Td. The control device 50 sets the high level period of the control signal SV6 to Tav+Td. １Ｖ becomes zero at the end of the dead time period Td (time t7). In the example of FIG. 3 , current iL1 flowing through resonant inductor L1 begins to flow at the start of the high-level period of control signal SV6 (time t5) and becomes zero at time t8, when additional time Tav has elapsed since the end of dead time period Td (time t7). With regard to current iL1, since iL1≧iV holds true from the start of dead time period Td (time t6), current iL1 in the shaded region of the current waveform in the tenth row from the top in FIG. 3 flows into resonant capacitor 9V, causing LC resonance. After the end of dead time period Td (time t7), current iL1 is regenerated in power conversion circuit 11 via third diode 13, which is directly connected to resonant inductor L1.

[0055] As described above, in order to start LC resonance at the start of the dead time period Td (time t6), the control device 50 determines the additional time Tav based on the load current iV so that iL1 = iV at the start of the dead time period Td (time t6). More specifically, the control device 50 determines the additional time Tav by calculating Tav = iV × (L / V15) using, for example, the detection result of the load current iV by a current sensor or its signal-processed value, or an estimated value of the load current iV, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. The detection result of the load current iV or its signal-processed value at this time uses a detection value at the carrier cycle to which the additional time Tav is added, or at a timing closest to that carrier cycle. Furthermore, the estimated value of the load current iV at this time uses, for example, a value estimated from the load current iV at the carrier cycle to which the additional time Tav is added.

[0056] As shown in FIG. 4 , the additional time Taw is set to advance the start point (time t9) of the high-level period of the control signal SW6 to earlier than the start point (time t10) of the dead time period Td, thereby making the high-level period of the control signal SW6 longer than the dead time period Td. The length of the additional time Taw is set based on the value of the load current iW. To start LC resonance at the start point (time t10) of the dead time period Td, it is desirable that the value of the current iL1 match the value of the load current iW at the start point (time t10) of the dead time period Td. This is because, while iL1 < iW, all of the current iL1 flows through the AC load RA1, preventing the resonant capacitor 9W from being charged. The end point of the high-level period of the control signal SW6 may be the same as or later than the end point (time t11) of the dead time period Td. FIG. 4 illustrates an example in which the end point of the high-level period of the control signal SW6 is set to the same as the end point (time t11) of the dead time period Td. The control device 50 sets the high level period of the control signal SW6 to Taw+Td. １Ｗbecomes zero at the end of the dead time period Td (time t11). In the example of FIG. 4 , the current iL1 flowing through the resonant inductor L1 starts flowing at the start of the high-level period of the control signal SW6 (time t9) and becomes zero at time t12, when the additional time Taw has elapsed since the end of the dead time period Td (time t11). With regard to the current iL1, since iL1≧iW holds from the start of the dead time period Td (time t10), the current iL1 in the shaded region of the current waveform in the fourth row from the top in FIG. 4 flows into the resonant capacitor 9W, causing LC resonance. After the end of the dead time period Td (time t11), the current iL1 is regenerated in the power conversion circuit 11 via the third diode 13 directly connected to the resonant inductor L1.

[0057] The control device 50 determines the additional time Taw based on the load current iW. More specifically, the control device 50 determines the additional time Taw by calculating Taw = iW × (L / V15) using the detection result of the load current iW by the current sensor, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. The detection result of the load current iW or its signal processed value at this time uses a detection value at the carrier cycle to which the additional time Taw is added or at a timing closest to that carrier cycle. Furthermore, the estimated value of the load current iW at this time uses, for example, a value estimated from the load current iW at the carrier cycle to which the additional time Taw is added.

[0058] (3.1.2) Operation for Soft-Switching the Second Switching Element When Load Current > 0 When the soft-switching target is the second switching element 2 (hereinafter also referred to as the target second switching element 2), and when the polarity of the load current (load current iU, load current iV, or load current iW) flowing through the AC terminal 41 connected to the target second switching element 2 is positive, the control device 50 compares the current value of the load current with the first current threshold I1 (=Ith, see FIG. 5 ). When the current value of the load current is greater than the first current threshold I1, the control device 50 does not turn on the switch 8, and when the current value of the load current is less than the first current threshold I1, the control device 50 turns on the switch 8 during the dead time period Td. In the power conversion device 100, when the current value of the load current is greater than the first current threshold I1, the control device 50 can discharge the resonant capacitor 9U connected in parallel to the target second switching element 2 using the load current iU without turning on the switch 8 corresponding to the target second switching element 2. This allows the power conversion device 100 to achieve zero voltage soft switching of the target second switching element 2.

[0059] 6 shows a case where the target second switching element 2 is the second switching element 2U of the switching circuit 10U, and the current value of the load current is greater than the first current threshold I1. The control signals SU1, SU2, and SU7, the load current iU, the current i9U flowing from the resonance capacitor 9U, and the voltage V across the second switching element 2U are ２Ｕ 6 also shows the dead time period Td and the additional time Tau that is set in the control device 50 for the control signal SU7 of the second IGBT 7U of the switch 8U.

[0060] When the current value of the load current iU is greater than the first current threshold I1, the control device 50 does not provide a high-level period for the control signal SU7. In this case, in the power conversion device 100, the current iU9 starts to flow from the resonant capacitor 9U at the start of the dead time period Td (time t22), the current i9U drops to zero before the end of the dead time period Td (time t23), and the voltage V across the second switching element 2U decreases before the end of the dead time period Td (time t23).２Ｕ becomes zero. As a result, in the power conversion device 100, when the control signal SU2 changes from low level to high level at the end of the dead time period Td (time t23), the second switching element 2U is subjected to zero-voltage soft switching.

[0061] When the current value of the load current iU is smaller than the first current threshold I1, the control device 50 provides a high-level period for the control signal SU7, as shown by the two-dot chain line in FIG. 6 . The start point of the high-level period of the control signal SU7 at this time is, for example, the same as the start point (time t22) of the dead time period Td. The end point of the high-level period of the control signal SU7 is the same as the end point (time t23) of the dead time period Td. As a result, in the power conversion device 100, the voltage V across the second switching element 2U is lowered before the end point (time t23) of the dead time period Td. ２Ｕ becomes zero. Therefore, in the power conversion device 100, when the control signal SU2 changes from low level to high level at the end of the dead time period Td (time t23), the second switching element 2U is subjected to zero-voltage soft switching. The start of the high-level period of the control signal SU7 may be time t21, which is earlier than the start of the dead time period Td by the additional time Tau. The end of the high-level period of the control signal SU7 may be time t24, which is later than the end of the dead time period Td (time t23) by the additional time Tau. Note that the time before and after the period that overlaps with the dead time period Td in the high-level period is not limited to the additional time Tau and may be another set time.

[0062] (3.1.3) Operation for Soft-Switching the Second Switching Element When Load Current<0 When the polarity of the load current (load current iU, load current iV, or load current iW) flowing through the AC terminal 41 connected to the target second switching element 2 is negative, the control device 50 turns on the second IGBT 7 corresponding to the target second switching element 2. As a result, the control device 50 causes the resonant capacitor 9 and the resonant inductor L1 connected to the target second switching element 2 to resonate, discharging the resonant capacitor 9 and setting the voltage across the target second switching element 2 to zero. This allows the power conversion device 100 to achieve zero-voltage soft switching of the target second switching element 2.

[0063] 7 shows the relationship between the control signals SU1, SU2, and SU7, the load current iU, the current iL1 flowing through the resonance inductor L1, and the voltage V across the second switching element 2U when the target second switching element 2 is the second switching element 2U of the switching circuit 10U. ２Ｕ and are illustrated.

[0064] FIG. 7 also illustrates a dead time period Td set in the control device 50 to prevent the first switching element 1 and the second switching element 2, which are in phase, from being turned on simultaneously. FIG. 7 also illustrates an additional time Tau set in the control device 50 for the control signal SU7 of the second IGBT 7U of the switch 8U. The end point of the high-level period of the control signal SU7 may be the same as or later than the end point of the dead time period Td (time t33). FIG. 7 illustrates an example in which the end point of the high-level period of the control signal SU7 is set to the same as the end point of the dead time period Td (time t33). The control device 50 sets the high-level period of the control signal SU7 to Tau+Td. In the switching circuit 10U, the voltage V across the second switching element 2U is ２Ｕbecomes zero at the end of the dead time period Td (time t33). In the example of FIG. 7 , the current iL1 flowing through the resonant inductor L1 starts flowing at the start of the high-level period of the control signal SU7 (time t31) and becomes zero at time t34, when the additional time Tau has elapsed since the end of the dead time period Td (time t33). With regard to the current iL1, since iL1≦iU is established from the start of the dead time period Td (time t32), LC resonance occurs, and a resonant current (discharge current of the resonant capacitor 9U) flows from the resonant capacitor 9U toward the resonant inductor L1. After the end of the dead time period Td (time t33), the current iL1 is regenerated in the power conversion circuit 11 via the fourth diode 14, which is directly connected to the resonant inductor L1.

[0065] The control device 50 determines the additional time Tau based on the load current iU so that iL1 = iU at the start of the dead time period Td (time t32) and the end of the dead time period Td (time t33) so that LC resonance starts at the start of the dead time period Td (time t32) and the resonant half cycle ends at the end of the dead time period Td (time t33). More specifically, the control device 50 determines the additional time Tau by calculating Tau = iU × (L / V15) using, for example, the detection result of the output current iU by a current sensor or its signal-processed value, or an estimated value of the load current iU, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. The detection result of the load current iU or its signal-processed value at this time uses a detection value at the carrier cycle to which the additional time Tau is added or at a timing closest to that carrier cycle. The estimated value of the load current iU at this time uses, for example, an estimated value of the load current iU at the carrier cycle to which the additional time Tau is added. The resonance half period in the case of basic operation is half the resonance period, which is the reciprocal of the resonance frequency of the resonance circuit including the resonance inductor L1 and one resonance capacitor 9. Therefore, if the inductance of the resonance inductor L1 is L and the capacitance of the resonance capacitor 9 is C, the resonance half period is π × (L C) １／２ In the control device 50, the resonance half cycle during basic operation is set to be equal to the length of the dead time period Td, for example.

[0066] (3.1.4) Operation for Soft-Switching the First Switching Element When Load Current < 0 When the polarity of the load current (load current iU, load current iV, or load current iW) flowing through the AC terminal 41 connected to the target first switching element 1 is negative, the control device 50 compares the current value of the load current with the second current threshold I2 (= −Ith, see FIG. 5 ). When the current value of the load current is smaller than the second current threshold I2, the control device 50 does not turn on the switch 8. When the current value of the load current is larger than the second current threshold I2, the control device 50 turns on the switch 8 during the dead time period Td. When the current value of the load current is smaller than the second current threshold I2, the power conversion device 100 can charge the resonant capacitor 9U connected in series to the target first switching element 1 with the load current without turning on the switch 8 corresponding to the target first switching element 1. This allows the power conversion device 100 to achieve zero-voltage soft switching of the target first switching element 1.

[0067] 8 shows the relationship between the control signals SU1, SU2, and SU6, the load current iU, the current i9U flowing from the resonant capacitor 9U, and the voltage V across the second switching element 2U when the target first switching element 1 is the first switching element 1U of the switching circuit 10U and the current value of the load current is greater than the second current threshold I2 (in other words, when the absolute value of the current value of the load current is smaller than the absolute value of the second current threshold I2). ２Ｕ 8 also shows the dead time period Td.

[0068] When the current value of the load current is smaller than the second current threshold I2 (in other words, when the absolute value of the load current is larger than the absolute value of the second current threshold I2), the control device 50 does not provide a high-level period for the control signal SU6. In this case, in the power conversion device 100, the current iU9 starts to flow through the resonant capacitor 9U at the start of the dead time period Td (time t41). As a result, in the power conversion device 100, the resonant capacitor 9U is charged, and the voltage V across the second switching element 2U ２Ｕincreases, the current i9U becomes zero before the end of the dead time period Td (time t23), and the voltage V across the first switching element 1U decreases before the end of the dead time period Td (time t42). １Ｕ becomes zero. As a result, in the power conversion device 100, when the control signal SU1 changes from low level to high level at the end of the dead time period Td (time t42), the first switching element 1U is subjected to zero-voltage soft switching.

[0069] When the current value of the load current is greater than the second current threshold I2 (in other words, when the absolute value of the load current is smaller than the absolute value of the second current threshold), the control device 50 sets a high-level period for the control signal SU6, as shown by the two-dot chain line in FIG. 8 . In this case, the start point of the high-level period of the control signal SU6 is the same as the start point (time t41) of the dead time period Td. Furthermore, the end point of the high-level period of the control signal SU6 is the same as the end point (time t42) of the dead time period Td. As a result, in the power conversion device 100, the voltage V across the first switching element 1U is set to a high level before the end point (time t42) of the dead time period Td. １Ｕ becomes zero. Therefore, in the power conversion device 100, when the control signal SU1 changes from low level to high level at the end of the dead time period Td (time t42), the first switching element 1U is subjected to zero voltage soft switching.

[0070] (3.2) Operations when it is determined that two-phase resonant currents flow simultaneously When it is determined that resonant currents passing through two of the multiple switches 8 flow simultaneously through the resonant inductor L1, the control device 50 can execute a first control operation, a second control operation, and a third control operation. "When it is determined that resonant currents passing through two of the multiple switches 8 flow simultaneously" means a case where it is estimated in advance that resonant currents passing through the two switches 8 flow simultaneously through the resonant inductor L1.

[0071] (3.2.1) Determination of whether resonant currents flow simultaneously In the power conversion device 100, the phases of the voltage commands for three phases (U phase, V phase, and W phase) differ from each other by 120°, but the command values ​​of the voltage commands for two phases approach each other by an electrical angle of 60°, and the duties of the control signals for the two phases approach each other (see areas A1 and A2 in FIG. 2 ). In area A1 in FIG. 2 , the duties of the control signals for the U phase and the V phase are near 0.75. In area A2 in FIG. 2 , the duties of the control signals for the U phase and the V phase are near 0.25. The polarity of the resonant current is the same as the polarity of the current iL1, and in area A1, the polarity of the resonant current is positive, and in area A2, the polarity of the resonant current is negative. In the case of region A1, for example, during one cycle of the carrier signal, the time difference between the start point of the high-level period of the control signal SU6 supplied to the first IGBT 6U (time t1: see FIG. 3) and the start point of the high-level period of the control signal SV6 supplied to the first IGBT 6V (time t5: see FIG. 3) becomes short, and there is a possibility that the U-phase resonant current and the V-phase resonant current will flow simultaneously through the resonant inductor L1. In the power conversion device 100, in the case of region A2, the direction of the resonant current is opposite to that in region A1, but there is a possibility that the U-phase resonant current and the V-phase resonant current will flow simultaneously through the resonant inductor L1.

[0072] Assuming that the capacitance of each of the multiple resonant capacitors 9U, 9V, and 9W is C, if a U-phase current and a V-phase current flow simultaneously through the resonant inductor L1, then in terms of an equivalent circuit, a capacitor having a combined capacitance (=2×C) of the resonant capacitors 9U and 9V is connected in series to the resonant inductor L1. Therefore, in the power conversion device 100, if two-phase currents flow simultaneously through the resonant inductor L1, the resonant frequency of the resonant circuit including the resonant inductor L1 will change compared to when a single-phase current flows through the resonant inductor L1, and zero-voltage soft switching may not be achieved.

[0073] (3.2.1.1) In the case of charging the resonant capacitor Figure 3 is a diagram showing an example of the boundary conditions between the case where the U-phase resonant current and the V-phase resonant current do not overlap (flow simultaneously) and the case where they overlap (flow simultaneously). The boundary conditions will be described with reference to Figure 3.

[0074] In the power conversion device 100, if the time difference ΔT between the start point (time t3) of the high-level period of the control signal SU1 and the start point (time t7) of the high-level period of the control signal SV1 is equal to or greater than (Tau + Tav + Td), the U-phase resonant current and the V-phase resonant current do not overlap, and if the time difference ΔT is less than (Tau + Tav + Td), the U-phase resonant current and the V-phase resonant current overlap. The control device 50 sets a threshold for the time difference ΔT to, for example, (Tau + Tav + Td), and if the time difference ΔT is less than the threshold, it estimates that resonant currents corresponding to two phases, switching circuit 10U and switching circuit 10V, of the multiple switching circuits 10 will flow simultaneously through the resonant inductor L1. The above threshold setting is merely an example, and other values ​​may also be considered. For example, the threshold value may be set to a value greater than (Tau + Tav + Td) taking into account errors in the additional time Tau and the additional time Tav. Furthermore, in the control device 50, the threshold value for the time difference ΔT may be set to, for example, the same value as the resonant half cycle (in this embodiment, the resonant half cycle = the dead time period Td). In this case, if the time difference ΔT is less than the length of the dead time period Td, the control device 50 estimates that resonant currents corresponding to the two phases of the switching circuit 10U and the switching circuit 10V flow simultaneously in the resonant inductor L1. Furthermore, the method for calculating the time difference used to determine whether the two-phase resonant currents flow simultaneously is not limited to the above example, and any other calculation method may be used as long as it can calculate a time difference equivalent to the time difference. For example, the time difference used to determine whether the two-phase resonant currents flow simultaneously may be the time difference between the end point of the high-level period of the control signal SU2 (time t2) and the end point of the high-level period of the control signal SV2 (time t6).

[0075] Furthermore, in the power conversion device 100, if the time difference between the start point (time t3) of the high-level period of the control signal SU1 and the start point (time t11) of the high-level period of the control signal SW1 is equal to or greater than (Tau + Taw + Td), the U-phase resonant current and the W-phase resonant current do not overlap, and if the time difference is less than (Tau + Taw + Td), the U-phase resonant current and the W-phase resonant current overlap. The control device 50 sets a threshold for the time difference to, for example, (Tau + Taw + Td), and if the time difference is less than the threshold, it estimates that resonant currents corresponding to two phases, switching circuit 10U and switching circuit 10W, of the multiple switching circuits 10, will flow simultaneously through the resonant inductor L1. The above threshold setting is merely an example, and other values ​​may also be considered. For example, it is possible to set the threshold value greater than (Tau + Taw + Td) by taking into account errors in the additional time Tau and the additional time Taw. Furthermore, in the control device 50, the threshold value for the time difference may be set to, for example, the same value as the resonant half cycle (in this embodiment, the resonant half cycle = the dead time period Td). In this case, if the time difference is less than the length of the dead time period Td, the control device 50 estimates that resonant currents corresponding to the two phases of the switching circuit 10U and the switching circuit 10W flow simultaneously in the resonant inductor L1. Furthermore, the method for calculating the time difference used to determine whether the two-phase resonant currents flow simultaneously is not limited to the above example; any other calculation method may be used as long as it can calculate a time difference equivalent to the time difference. For example, the time difference used to determine whether the two-phase resonant currents flow simultaneously may be the time difference between the end point of the high-level period of the control signal SU2 (time t2) and the end point of the high-level period of the control signal SW2 (time t10).

[0076] Furthermore, in the power conversion device 100, if the time difference between the start point (time t7) of the high-level period of the control signal SV1 provided to the first switching element 1V of the switching circuit 10V and the start point (time t11) of the high-level period of the control signal SW1 provided to the first switching element 1W of the switching circuit 10W is (Tav + Taw + Td) or more, the V-phase resonant current and the W-phase resonant current do not overlap, and if the time difference is less than (Tav + Taw + Td), the V-phase resonant current and the W-phase resonant current overlap. The control device 50 sets a threshold for the time difference to, for example, (Tav + Taw + Td), and if the time difference is less than the threshold, it estimates that resonant currents corresponding to two phases, the switching circuit 10V and the switching circuit 10W, of the multiple switching circuits 10 will flow simultaneously through the resonant inductor L1. The above threshold setting is merely an example, and other values ​​may also be considered. For example, it is possible to set the threshold value greater than (Tav + Taw + Td) in consideration of errors in the additional times Tav and Taw. Furthermore, in the control device 50, the threshold value for the time difference may be set to, for example, the same value as the resonant half cycle (in this embodiment, the resonant half cycle = the dead time period Td). In this case, if the time difference is less than the length of the dead time period Td, the control device 50 estimates that resonant currents corresponding to the two phases of the switching circuit 10V and the switching circuit 10W flow simultaneously in the resonant inductor L1. Furthermore, the method for calculating the time difference used to determine whether the two-phase resonant currents flow simultaneously is not limited to the above example, and any other calculation method may be used as long as it can calculate a time difference equivalent to the time difference. For example, the time difference used to determine whether the two-phase resonant currents flow simultaneously may be the time difference between the end point of the high-level period of the control signal SV2 (time t6) and the end point of the high-level period of the control signal SW2 (time t10).

[0077] (3.2.1.2) In the case of discharging operation of the resonant capacitor In the case of discharging operation of the resonant capacitor 9, the control device 50 can determine whether two-phase resonant currents flow simultaneously using the same time difference and threshold value as in the case of charging operation of the resonant capacitor 9.

[0078] For example, the control device 50 estimates that the U-phase resonant current and the V-phase resonant current overlap if the time difference between the start point of the high-level period of the control signal SU2 and the start point of the high-level period of the control signal SV2 is less than a threshold value (e.g., Tau + Tav + Td).

[0079] In addition, the control device 50 estimates that the U-phase resonant current and the W-phase resonant current overlap if the time difference between the start point of the high-level period of the control signal SU2 and the start point of the high-level period of the control signal SW2 is less than a threshold value (e.g., Tau + Taw + Td).

[0080] In addition, the control device 50 estimates that the V-phase resonant current and the W-phase resonant current overlap if the time difference between the start of the high-level period of the control signal SV2 and the start of the high-level period of the control signal SW2 is less than a threshold value (e.g., Tav + Taw + Td).

[0081] (3.2.2) First Control Operation, Second Control Operation, and Third Control Operation (3.2.2.1) Operation for Soft Switching of First Switching Element In the first control operation, the control device 50 overlaps the high-level period of the control signal to each of the two switches 8 with the dead time period Td corresponding to each of the two switching circuits 10 connected to the two switches 8 for a predetermined period of time.

[0082] In the second control operation, the control device 50 determines the start point of the high-level period of the control signal to at least one switch 8 of the multiple switches 8 in accordance with the load current flowing through the AC load RA1. Here, in the second control operation of the control device 50, the start point of the high-level period of the control signal to two switches 8 of the multiple switches 8 is changed in accordance with the total value of two-phase load currents flowing through each of the two AC terminals 41 connected to those two switches 8 among the multiple AC terminals 41. In the example of Fig. 9, the start point of the high-level period of the control signals SU6, SV6 to the two switches 8 is changed in accordance with the total value of the U-phase load current iU flowing through AC terminal 41U and the V-phase load current iV flowing through AC terminal 41V.

[0083] In the third control operation, the control device 50 can execute a third control operation to make the dead time periods Td1, Td2 corresponding to each of two or more switching circuits 10 connected to two or more switches 8 among the plurality of switching circuits 10 longer than the predetermined dead time period Td by an additional time Tad. The predetermined dead time period Td is the dead time period Td in the basic operation.

[0084] 9 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation, the second control operation, and the third control operation, and FIG. 10 shows a timing chart of the power conversion device 100 when the control device 50 does not execute the first control operation, the second control operation, and the third control operation. The predetermined period is, for example, at least a part of a resonant half cycle of a resonant circuit including the resonant inductor L1 and two resonant capacitors 9 (here, resonant capacitors 9U and 9V). For example, if the resonant half cycle is Tr2 when the resonant circuit includes two resonant capacitors 9 (in other words, when the current iL1 flowing through the resonant inductor L1 includes a two-phase resonant current), the resonant half cycle Tr2 is half the resonant cycle, which is the reciprocal of the resonant frequency of the resonant circuit including the resonant inductor L1 and the two resonant capacitors 9. When the resonant circuit includes two resonant capacitors 9, the inductance of the resonant inductor L1 is L, and the capacitance of each of the two resonant capacitors 9 is C, the resonant half cycle Tr2 of the resonant circuit is Tr2=2 １／２ ×π×(L C) １／２ 9, the predetermined period is the entire period of the resonance half cycle Tr2. In other words, the length of the predetermined period is 100% of the resonance half cycle Tr2.

[0085] The operation of the control device 50 will now be described in more detail.

[0086] When the control device 50 determines that the resonant currents of the two phases overlap, it executes the first step, the second step, and the third step in this order. Note that the following describes an example in which it is estimated that the resonant current flowing through the U-phase switch 8U and the V-phase switch 8V in the resonant inductor L1 overlap, but the same applies to the cases of the U and W phases and the V and W phases.

[0087] In the first step, the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2. In the example of FIG. 9 , the start point of the high-level period of the V-phase control signal SV1 and the end point of the high-level period of the V-phase control signal SV2 are shifted earlier by ΔT1, thereby synchronizing the U-phase control signals SU1 and SU2 with the V-phase control signals SV1 and SV2. In the example of FIG. 10 , ΔT1 is the time difference between the start point of the high-level period of the control signal SU1 and the start point of the high-level period of the control signal SV1, or the time difference between the end point of the high-level period of the control signal SU2 and the end point of the control signal SV2. In the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the U-phase control signals SU1 and SU2 by ΔT1 in a direction that delays the high-level periods of the U-phase control signals. In addition, in the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting them by a total of ΔT1.

[0088] In the second step, an additional time Tad corresponding to the current value |iU+iV| of the total current of the load currents iU and iV of the two phases is added to the high-level periods of the control signals SU6 and SV6 to the switches 8 of the two phases, corresponding to the dead time periods Td of each of the two phases (see FIG. 10 ). The control device 50 determines the additional time Tad by calculating Tad=L×|iU+iV| / V15 using, for example, the detection results of the load currents iU and iV by the current sensors, or their signal-processed values, or the estimated values ​​of the load currents iU and iV, the inductance L of the resonance inductor L1 stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. In the power conversion device 100 according to the first embodiment, the second step corresponds to the second control operation of the control device 50.

[0089] In the third step, the lengths of the high level periods and the dead time periods Td of the control signals SU6 and SV6 are set to Tr2=2, which is the resonant half cycle of the resonant circuit. １／２ ×π×(L C) １／２ In more detail, the control device 50 sets the length of the period obtained by subtracting the additional time Tad from the high level period of each of the control signals SU6 and SV6 to the resonant half cycle Tr2 of the resonant circuit, and changes the dead time period Td to a dead time period Td1 equal to the resonant half cycle Tr2 of the resonant circuit. In the power conversion device 100 according to the first embodiment, the third step corresponds to the third control operation of the control device 50. Note that the end point of the control signals SU6 and SV6 may be any point after the end point of the resonant half cycle Tr2.

[0090] In the power conversion device 100, when the control device 50 does not execute the first control operation, the second control operation, and the third control operation, as shown in FIG. 10, the voltages V across the second switching elements 2U and 2V are equal to or higher than the voltages V across the second switching elements 2U and 2V at the time when the control signals SU1 and SV1 change from a low level period to a high level period (the end time of the dead time period Td corresponding to the U phase and the V phase, respectively). ２Ｕ , V ２Ｖdoes not rise to Vd. In other words, if the control device 50 does not execute the first control operation, the second control operation, or the third control operation, the charging of the resonant capacitors 9U and 9V does not finish at the end of the dead time period Td corresponding to the U phase and the V phase, respectively. Therefore, if the first control operation, the second control operation, or the third control operation is not executed, the voltages across the first switching elements 1U and 1V do not decrease to zero at the end of the dead time period Td corresponding to the U phase and the V phase, respectively. As a result, in the power conversion device 100, the switching of the first switching elements 1U and 1V becomes hard switching.

[0091] In contrast, when the control device 50 executes the first control operation, the second control operation, and the third control operation, as shown in FIG. 9, the voltages V across the second switching elements 2U and 2V at the time when the control signals SU1 and SV1 change from a low level period to a high level period (the end time of the dead time period Td1 corresponding to the U phase and the V phase, respectively) ２Ｕ , V ２Ｖ rises to Vd. In other words, when the control device 50 executes the first control operation, the second control operation, and the third control operation, charging of the resonant capacitors 9U and 9V ends at the end of the dead time period Td1 corresponding to the U phase and the V phase, respectively. Therefore, in the power conversion device 100, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the switching of the first switching elements 1U and 1V becomes zero-voltage soft switching.

[0092] 10 illustrates the relationship between the control signals SU1, SU2, SV1, SV2, SU6, SV6, etc. in an example where the control signal SU6 to the U-phase switch 8 and the control signal SV6 to the V-phase switch 8 overlap, but the present invention is not limited to this. For example, in both cases where the V-phase and the W-phase overlap and where the W-phase and the U-phase overlap, the control device 50 performs the first control operation, the second control operation, and the third control operation, thereby enabling zero-voltage soft switching. The example of FIG. 11 illustrates the relationship between the control signals SV1, SV2, SW1, SW2, SV6, and SW6 in a case where the control signal SV6 to the V-phase switch 8 and the control signal SW6 to the W-phase switch 8 overlap.

[0093] Furthermore, when it is determined that the two-phase resonant currents overlap, it is sufficient that at least a portion of the control signals to the two switches 8 overlap, and the temporal relationship between the control signals to the first switching element 1 and the second switching element 2 is not limited. For example, as shown in Fig. 12, the start point of the high-level period of the control signal SV1 may be earlier than the start point of the high-level period of the control signal SU1, and the start point of the high-level period of the control signal SV6 may be earlier than the start point of the high-level period of the control signal SU6. Furthermore, as shown in Fig. 13, the dead time period Td between the control signals SU1 and SU2 and the dead time period Td between the control signals SV1 and SV2 may not overlap, and the high-level periods of the control signals SU6 and SV6 may partially overlap.

[0094] Furthermore, the relationship in polarity and magnitude of the two-phase load currents when the two-phase resonant currents overlap is not limited to the relationship iU>iV>0 in region A1 in the example of Fig. 2, but may be, for example, the relationship iU>0>iV in region A1 in the example of Fig. 14. In this case, for example, as shown in Fig. 15, the control signal SU6 and the control signal SV6 overlap, and the two-phase resonant currents overlap.

[0095] (3.2.2.2) Operation for soft-switching the second switching element In the first control operation, the control device 50 overlaps the high-level period of the control signal to each of the two or more switches 8 with the dead time period Td (see, for example, FIG. 17) corresponding to each of the two switching circuits 10 connected to two of the multiple switching circuits 10 for a predetermined period. FIG. 16 shows a timing chart when the first control operation, the second control operation, and the third control operation are executed, and FIG. 17 shows a timing chart when the first control operation, the second control operation, and the third control operation are not executed. The predetermined period is, for example, at least a part of the resonant half cycle Tr2 of the resonant circuit including the resonant inductor L1 and the two resonant capacitors 9 connected to the two switches 8, respectively. The resonant half cycle Tr2 is, for example, Tr2=2 １／２ ×π×(L C) １／２16, the predetermined period is the entire period of the resonance half cycle Tr2. In other words, the length of the predetermined period is 100% of the resonance half cycle.

[0096] The operation of the control device 50 will now be described in more detail.

[0097] If the control device 50 determines that the resonant currents overlap, it executes the first step, the second step, and the third step in this order. Note that the following describes an example in which it is estimated that the resonant current flowing through the U-phase switch 8U and the V-phase switch 8V overlap in the resonant inductor L1, but the same applies to the cases of the U-phase and the W-phase, and the V-phase and the W-phase.

[0098] In the first step, the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2. In the example of FIG. 16 , the end point of the high-level period of the V-phase control signal SV1 and the start point of the high-level period of the V-phase control signal SV2 are shifted earlier by ΔT1 in the example of FIG. 17 , thereby synchronizing the U-phase control signals SU1 and SU2 with the V-phase control signals SV1 and SV2. In the power conversion device 100 according to the first embodiment, the first step corresponds to the first control operation of the control device 50. Note that in the first step, the U-phase control signals SU1 and SU2 may also be synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the U-phase control signals SU1 and SU2 later by ΔT1. In the first step, the U-phase control signals SU1 and SU2 and the V-phase control signals SV1 and SV2 may be synchronized by shifting them by a total of ΔT1.

[0099] In the second step, an additional time Tad corresponding to the total current of the load currents iU and iV of the two phases is added to the high-level periods of the control signals SU7 and SV7 sent to the switches 8 of the two phases, corresponding to the dead time periods Td of each of the two phases. The control device 50 determines the additional time Tad by calculating Tad = L × |iU + iV| / V15 using, for example, the detection results of the load currents iU and iV by the current sensors, or their signal processed values, or the estimated values ​​of the load currents iU and iV, the inductance L of the resonance inductor L1 stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. In the power conversion device 100 according to the first embodiment, the second step corresponds to the second control operation of the control device 50.

[0100] In the third step, the lengths of the high level periods and the dead time periods Td of the control signals SU7 and SV7 are set to Tr2=2, which is the resonant half period of the resonant circuit. １／２ ×π×(L C) １／２ In more detail, the control device 50 sets the length of the period obtained by subtracting the additional time Tad from the high level period of each of the control signals SU7 and SV7 to the resonant half cycle Tr2 of the resonant circuit, and changes the dead time period Td to the dead time period Td1 equal to the resonant half cycle Tr2 of the resonant circuit. In the power conversion device 100 according to the first embodiment, the third step corresponds to the third control operation of the control device 50. Note that the end point of the control signals SU7 and SV7 may be any point after the end point of the resonant half cycle Tr2.

[0101] In the power conversion device 100, when the control device 50 does not execute the first control operation, the second control operation, and the third control operation, as shown in FIG. 17, the voltages V across the second switching elements 2U and 2V are equal to or higher than the voltages V across the second switching elements 2U and 2V at the time when the control signals SU2 and SV2 change from a low level period to a high level period (the end time of the dead time period Td corresponding to the U phase and the V phase, respectively). ２Ｕ , V ２Ｖdoes not decrease to zero. In other words, in the power conversion device 100, when the control device 50 does not perform the first control operation, the second control operation, or the third control operation, the discharge of the resonant capacitors 9U and 9V does not finish at the end of the dead time period Td corresponding to the U phase and the V phase, respectively. Therefore, in the power conversion device 100, when the control device 50 does not perform the first control operation, the second control operation, or the third control operation, the switching of the second switching elements 2U and 2V becomes hard switching.

[0102] In contrast, when the control device 50 executes the first control operation, the second control operation, and the third control operation, as shown in FIG. 16, the voltages V across the second switching elements 2U and 2V decrease at the time when the control signals SU2 and SV2 change from a low level period to a high level period (at the end of the dead time periods Td1 corresponding to the U phase and the V phase, respectively). ２Ｕ , V ２Ｖ becomes zero. In other words, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the discharge of the resonant capacitors 9U and 9V ends at the end of the dead time period Td1 corresponding to the U phase and the V phase, respectively. Therefore, in the power conversion device 100, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the switching of the second switching elements 2U and 2V becomes zero-voltage soft switching.

[0103] (3.3) Operation When It is Determined that Three-Phase Resonant Currents Simultaneously Flow Hereinafter, the operation of the control device 50 when it is determined that resonant currents passing through three of the multiple switches 8 simultaneously flow through the resonant inductor L1 will be described with reference to FIG. 18 . The control device 50 executes a first control operation when it is determined that resonant currents passing through three of the multiple switches 8 simultaneously flow through the resonant inductor L1. "When it is determined that resonant currents passing through three of the multiple switches 8 simultaneously flow through the resonant inductor L1" refers to a case where it has been estimated in advance that resonant currents passing through the three switches 8 simultaneously flow through the resonant inductor L1. An example of a case where three-phase resonant currents overlap is when the AC load RA1 is in a light load state, where iU = 0, iV = 0, and iW = 0. This situation occurs, for example, when the AC load RA1 is a motor, particularly when the motor is rotating at a low speed or when the motor's rotational speed is zero (e.g., when the motor is locked). Therefore, the control device 50 determines that "three-phase resonant currents flow simultaneously" when, for example, the rotational speed of the motor (e.g., the number of rotations [rpm]) falls below a rotational speed threshold. Here, the control device 50 determines that "three-phase resonant currents flow simultaneously" when, for example, the rotational speed calculated from sensor information output from a sensor device (e.g., an encoder or resolver) for detecting the rotational speed of the motor, or the estimated rotational speed, falls below the rotational speed threshold.

[0104] Assuming that the capacitance of each of the multiple resonant capacitors 9U, 9V, and 9W is C, if a U-phase current, a V-phase current, and a W-phase current simultaneously flow through the resonant inductor L1, then in terms of an equivalent circuit, a capacitor having a combined capacitance (=3×C) of the resonant capacitors 9U, 9V, and 9W is connected in series to the resonant inductor L1. Therefore, in the power conversion device 100, if three-phase currents simultaneously flow through the resonant inductor L1, the resonant frequency of the resonant circuit including the resonant inductor L1 will change compared to when a single-phase current flows through the resonant inductor L1, and zero-voltage soft switching may not be achieved.

[0105] (3.3.1) First Control Operation (3.3.1.1) Operation for Soft-Switching First Switching Element In the first control operation, the control device 50 overlaps, for a predetermined period, the high-level periods of the control signals to the three switches 8, respectively, with the dead time periods Td (see, for example, FIG. 19 ) corresponding to the three switching circuits 10 connected to the three switches 8 among the multiple switching circuits 10. FIG. 18 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation, and FIG. 19 shows a timing chart of the power conversion device 100 when the control device 50 does not execute the first control operation. The predetermined period is, for example, at least a portion of a resonant half cycle of a resonant circuit including the resonant inductor L1 and the three resonant capacitors 9 connected to the three switches 8, respectively. For example, if the resonant half period is Tr3 when the resonant circuit includes three resonant capacitors 9 (in other words, when the current iL1 flowing through the resonant inductor L1 includes a three-phase resonant current), the resonant half period Tr3 is half the resonant period, which is the reciprocal of the resonant frequency of the resonant circuit including the resonant inductor L1 and the three resonant capacitors 9. When the resonant circuit includes three resonant capacitors 9, if the inductance of the resonant inductor L1 is L and the capacitance of each of the three resonant capacitors 9 is C, the resonant half period Tr3 of the resonant circuit is Tr3=3 １／２ ×π×(L C) １／２ 18, the predetermined period is the entire period of the resonance half cycle Tr3. In other words, the length of the predetermined period is 100% of the resonance half cycle.

[0106] In the first control operation, the control device 50 sets the length of the dead time period Td2 to the same length as the resonance half cycle Tr3. Therefore, in the first control operation, the control device 50 sets the length of the dead time period Td2 to 3 times the length of the dead time period Td in the basic operation. １／２ Furthermore, the control device 50 matches the start and end times of the high level periods of the control signals SU6, SV6, and SW6 to the first IGBTs 6U, 6V, and 6W of the three switches 8 through which the resonant current flows.

[0107] The operation of the control device 50 will now be described in more detail.

[0108] When the control device 50 determines that the three-phase resonant currents overlap, it executes the first step and the second step in this order.

[0109] In the first step, the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 are synchronized. In the example of Fig. 18, the start point of the high-level period of the V-phase control signal SV1 and the end point of the high-level period of the V-phase control signal SV2 are shifted earlier in the example of Fig. 19, and the start point of the high-level period of the W-phase control signal SW1 and the end point of the high-level period of the W-phase control signal SW2 are shifted earlier, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed and the high-level periods of the W-phase control signals SW1 and SW2 may be advanced, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In addition, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be shifted in a later direction, the high-level periods of the V-phase control signals SV1 and SV2 may be shifted in a later direction, and the high-level periods of the W-phase control signals SW1 and SW2 may be shifted in an earlier direction, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2.

[0110] In the second step, the high level period and the dead time period Td of each of the control signals SU6, SV6, and SW6 are set to the resonant half period Tr3=3 of the resonant circuit. １／２ ×π×(L C) １／２ In more detail, the control device 50 sets the length of the high-level period of each of the control signals SU6, SV6, and SW6 to the resonant half cycle Tr3 of the resonant circuit, and sets the dead time period Td to a dead time period Td2 equal to the resonant half cycle Tr3 of the resonant circuit. Note that the control device 50 sets the additional times Tau, Tav, and Taw to 0 when the load current iU=0, the load current iV=0, and the load current iW=0.

[0111] In the power conversion device 100, when the control device 50 does not execute the first control operation, as shown in FIG. 19, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time when the control signals SU1, SV1, and SW1 change from a low level period to a high level period (the end time of the dead time period Td corresponding to each of the U phase, V phase, and W phase). ２Ｕ , V ２Ｖ , V ２Ｗ does not rise to Vd. In other words, in the power conversion device 100, when the control device 50 does not execute the first control operation, charging of the resonant capacitors 9U, 9V, and 9W does not finish at the end of the dead time period Td corresponding to each of the U phase, V phase, and W phase. For this reason, in the power conversion device 100, when the control device 50 does not execute the first control operation, the voltages across each of the first switching elements 1U, 1V, and 1W do not decrease to zero at the end of the dead time period Td corresponding to each of the U phase, V phase, and W phase, and the switching of the first switching elements 1U, 1V, and 1W becomes hard switching.

[0112] In contrast, in the power conversion device 100, when the control device 50 executes the first control operation, as shown in FIG. 18, the voltages V across the second switching elements 2U, 2V, and 2W decrease at the time when the control signals SU1, SV1, and SW1 change from a low level period to a high level period (at the end of the dead time period Td2 corresponding to each of the U phase, V phase, and W phase). ２Ｕ , V ２Ｖ , V ２Ｗrises to Vd. That is, in the power conversion device 100, when the control device 50 executes the first control operation, charging of the resonant capacitors 9U, 9V, and 9W ends at the end of the dead time period Td2 corresponding to each of the U phase, V phase, and W phase. Therefore, in the power conversion device 100, when the control device 50 executes the first control operation, the switching of the first switching elements 1U, 1V, and 1W becomes zero-voltage soft switching.

[0113] Incidentally, when it is determined that the three-phase resonant currents overlap, the relationship between the polarities and magnitudes of the three-phase load currents is not limited to the relationship iU = 0, iV = 0, iW = 0 as shown in Figure 19, and may be, for example, a relationship iU > 0 > iV as shown in Figure 20, or may be, for example, a relationship iW > iV > 0 > iU as shown in Figure 21.

[0114] Furthermore, as long as the control signals to the three switches 8 at least partially overlap, the temporal relationship between the control signals to the first switching element 1 and the temporal relationship between the control signals to the second switching element 2 are not limited. For example, as shown in Figure 22, the start point of the high-level period of the control signal SV1 may be earlier than the start point of the high-level period of the control signal SU1, the start point of the high-level period of the control signal SU1 may be earlier than the start point of the high-level period of the control signal SW1, the start point of the high-level period of the control signal SV6 may be earlier than the start point of the high-level period of the control signal SU6, and the start point of the control signal SU6 may be earlier than the start point of the control signal SW6. Also, for example, as shown in Figure 23, the dead time period Td between the control signals SU1 and SU2 and the dead time period Td between the control signals SV1 and SV2 may not overlap, and the high-level period of the control signal SU6, the high-level period of the control signal SV6, and the high-level period of the control signal SW may partially overlap. In the case of an operation for soft-switching the first switching element (in the case of charging the resonant capacitor 9), when three-phase resonant currents flow simultaneously, it is desirable that, when the load current is positive, additional times Tau, Tav, and Taw are added to the high-level period of the switch 8 corresponding to the phase through which the positive load current flows, as in the basic operation.

[0115] (3.3.1.2) Operation for Soft-Switching the Second Switching Element In the first control operation, the high-level periods of the control signals SU7, SV7, and SW7 to the three switches 8 are overlapped for a predetermined period with the dead time periods Td2 (see, for example, FIG. 24 ) corresponding to each of the three switching circuits 10. FIG. 24 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation, and FIG. 25 shows a timing chart of the power conversion device 100 when the control device 50 does not execute the first control operation. FIGS. 24 and 25 show timing charts for a period in which the resonant circuit includes the resonant capacitors 9U, 9V, and 9W. The predetermined period is, for example, at least a portion of the resonant half cycle Tr3 of the resonant circuit including the resonant inductor L1 and the three resonant capacitors 9. The resonant half cycle Tr3 is, for example, Tr3=3. １／２ ×π×(L C) １／２ 24, the predetermined period is the entire period of the resonance half cycle Tr3. In other words, the length of the predetermined period is 100% of the resonance half cycle Tr3.

[0116] In the first control operation, the control device 50 sets the length of the dead time period Td2 to the same length as the resonance half cycle Tr3. Therefore, in the first control operation, the control device 50 sets the length of the dead time period Td2 to 3 times the length of the dead time period Td in the basic operation. １／２ Furthermore, the control device 50 matches the start and end points of the high level periods of the control signals SU7, SV7, and SW7 to the three second IGBTs 7U, 7V, and 7W.

[0117] The operation of the control device 50 will now be described in more detail.

[0118] When the control device 50 determines that the three-phase resonant currents overlap, it executes the first step and the second step in this order.

[0119] In the first step, the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 are synchronized. In the example of Fig. 24, the end point of the high-level period of the V-phase control signal SV1 and the start point of the high-level period of the control signal SV2 are shifted earlier in the example of Fig. 25, and the end point of the high-level period of the W-phase control signal SW1 and the start point of the high-level period of the control signal SW2 are shifted earlier, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed and the high-level periods of the W-phase control signals SW1 and SW2 may be advanced, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In addition, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be shifted in a later direction, the high-level periods of the V-phase control signals SV1 and SV2 may be shifted in a later direction, and the high-level periods of the W-phase control signals SW1 and SW2 may be shifted in an earlier direction, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2.

[0120] In the second step, the high level period and the dead time period Td of each of the control signals SU6, SV6, and SW6 are set to the resonant half period Tr3=3 of the resonant circuit. １／２ ×π×(L C) １／２In more detail, the control device 50 sets the length of the high-level period of each of the control signals SU7, SV7, and SW7 to the resonant half cycle Tr3 of the resonant circuit, and sets the dead time period Td to a dead time period Td2 equal to the resonant half cycle Tr3 of the resonant circuit. Note that the control device 50 sets the additional times Tau, Tav, and Taw to 0 when the load current iU=0, the load current iV=0, and the load current iW=0.

[0121] When the control device 50 does not execute the first control operation, as shown in FIG. 25, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time when the control signals SU2, SV2, and SW2 change from a low level period to a high level period (the end time of the dead time period Td corresponding to each of the U phase, V phase, and W phase). ２Ｕ , V ２Ｖ , V ２Ｗ does not decrease to zero. In other words, in the power conversion device 100, when the control device 50 does not execute the first control operation, the discharge of the resonant capacitors 9U, 9V, and 9W does not finish at the end of the dead time period Td corresponding to each of the U phase, V phase, and W phase. For this reason, in the power conversion device 100, when the control device 50 does not execute the first control operation, the switching of the second switching elements 2U, 2V, and 2W becomes hard switching.

[0122] In contrast, in the power conversion device 100, when the control device 50 executes the first control operation, as shown in FIG. 24, the voltages V across the second switching elements 2U, 2V, and 2W decrease at the time when the control signals SU2, SV2, and SW2 change from a low level period to a high level period (at the end of the dead time period Td2 for each of the U phase, V phase, and W phase). ２Ｕ , V ２Ｖ , V ２Ｗ decreases to zero. That is, in the power conversion device 100, when the control device 50 executes the first control operation, the discharge of the resonant capacitors 9U, 9V, and 9W ends at the end of the dead time period Td2. Therefore, in the power conversion device 100, when the control device 50 executes the first control operation, the switching of the second switching elements 2U, 2V, and 2W becomes zero-voltage soft switching.

[0123] (4) Summary In the power conversion device 100 according to the first embodiment, the control device 50 can execute a first control operation and a second control operation when it determines that resonant currents flowing through two or more switches 8 simultaneously through the resonant inductor L1. The first control operation overlaps the high-level period of a control signal for each of the two or more switches 8 with the dead time periods (Td1, Td2) corresponding to the two or more switching circuits 10 connected to the two or more switches 8 for a predetermined period of time. The second control operation determines the start point of the high-level period of a control signal for at least one switch 8 among the multiple switches 8 in accordance with at least one phase of load current flowing through the AC load RA1 connected to the multiple AC terminals 41. This enables the power conversion device 100 to more reliably achieve soft switching.

[0124] In the power conversion device 100 according to the first embodiment, the predetermined period is the entire period of the resonant half cycle, which allows the power conversion device 100 according to the first embodiment to more reliably achieve zero voltage soft switching.

[0125] Furthermore, in the power conversion device 100 according to the first embodiment, in the second control operation of the control device 50, the start time of the high-level period of the control signal to at least one switch 8 of the multiple switches 8 is changed in accordance with the total value of two or more phases of load current flowing through each of two or more AC terminals 41 connected to two or more switches 8 of the multiple AC terminals 41. This enables the power conversion device 100 to start resonance at the start times of the dead time periods Td1 and Td2.

[0126] Furthermore, in the power conversion device 100 according to the first embodiment, the control device 50 can execute a third control operation of making the dead time periods Td1, Td2 corresponding to each of two or more switching circuits 10 connected to two or more switches 8 among the multiple switching circuits 10 longer than the predetermined dead time period Td by an additional time Tad. This enables the power conversion device 100 to achieve zero-voltage soft switching even when the resonant half cycles Tr2, Tr3 are longer than the dead time period Td.

[0127] (Modification 1) The circuit configuration of the power conversion device 100 according to Modification 1 of the first embodiment is the same as that of the power conversion device 100 according to the first embodiment (see FIG. 1), and therefore will not be illustrated or described again.

[0128] In the power conversion device 100 according to the first modification, part of the operation of the control device 50 when it is determined that the two-phase resonant currents overlap is different from the operation of the control device 50 according to the first embodiment. The operation of the control device 50 for soft-switching the first switching element 1 will be described below with reference to Fig. 26 and other figures.

[0129] In the first control operation, the control device 50 overlaps the high-level period of the control signal for each of the two or more switches 8 with the dead time period Td (see, for example, FIG. 26 ) corresponding to each of the two switching circuits 10 connected to two of the multiple switching circuits 10 by a predetermined period. FIG. 26 shows a timing chart when the first control operation and the second control operation are executed. The predetermined period is, for example, a portion of the resonant half-cycle Tr2 of the resonant circuit including the resonant inductor L1 and the two resonant capacitors 9 connected to the two switches 8, respectively. In the example of FIG. 26 , the predetermined period is 60% of the resonant half-cycle Tr2. In other words, the length of the predetermined period is 60% of the resonant half-cycle.

[0130] Furthermore, the control device 50 matches the start and end points of the high-level periods of the two control signals to the first IGBTs 6 of the two switches 8 through which the resonant current flows. In the example of Fig. 26, the control device 50 matches the start and end points of the high-level periods of the control signal SU6 to the first IGBT 6U and the control signal SV6 to the first IGBT 6V.

[0131] The operation of the control device 50 will now be described in more detail.

[0132] When the control device 50 determines that the resonant currents of the two phases overlap, it executes the first step, the second step, and the third step in this order. Note that the following describes an example in which it is estimated that the resonant current flowing through the U-phase switch 8U and the V-phase switch 8 overlap in the resonant inductor L1, but the same applies to the cases of the U and W phases and the V and W phases.

[0133] In the first step, the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2. In the example of FIG. 26 , the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the V-phase control signals SV1 and SV2 in the example of FIG. 10 earlier by ΔT1. The first step corresponds to the first control operation of the control device 50. Note that in the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the U-phase control signals SU1 and SU2 in a later direction by ΔT1. Alternatively, in the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting them by a total of ΔT1.

[0134] In the second step, an additional time Tad corresponding to the total current of the load currents iU and iV of the two phases is added to the high-level periods of the control signals SU6 and SV6 sent to the switches 8 of the two phases, corresponding to the dead time periods Td of each of the two phases. The control device 50 determines the additional time Tad by calculating Tad = L × |iU + iV| / V15 using, for example, the detection results of the load currents iU and iV by the current sensors, or their signal processed values, or estimates of the load currents iU and iV, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. The second step corresponds to the second control operation of the control device 50.

[0135] In the third step, the high level periods of the control signals SU6 and SV6 are set to the resonant half period Tr2=2 of the resonant circuit. １／２ ×π×(L C) １／２ More specifically, the control device 50 sets the length of the period obtained by subtracting the additional time Tad from the high level period of each of the control signals SU6 and SV6 to a value that is 60% of the resonant half cycle Tr2 of the resonant circuit. Note that the end point of the control signals SU6 and SV6 may be any time after the end point of the resonant half cycle Tr2.

[0136] In the power conversion device 100 according to the first modification, when the control device 50 does not execute the first control operation and the second control operation, as shown in FIG. 10, the voltages V across the second switching elements 2U and 2V are equal to or greater than the voltages V across the second switching elements 2U and 2V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end point of the dead time period Td). ２Ｕ , V ２Ｖ does not rise to Vd. In other words, if the control device 50 does not execute the first control operation and the second control operation, the charging of the resonant capacitors 9U and 9V does not finish at the end of the dead time period Td. Therefore, if the first control operation and the second control operation are not executed, the voltages across the first switching elements 1U and 1V do not decrease to zero at the end of the dead time period Td, and the switching of the first switching elements 1U and 1V becomes hard switching.

[0137] In contrast, in the power conversion device 100 according to the first modification, when the control device 50 executes the first control operation and the second control operation, as shown in FIG. 26, the voltages V across the second switching elements 2U and 2V are equal to or higher than the voltages V across the second switching elements 2U and 2V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end time of the dead time period Td). ２Ｕ , V ２Ｖ rises to a voltage closer to Vd. For this reason, in the power conversion device 100 according to the first modification, when the control device 50 executes the first control operation and the second control operation, the switching of the first switching elements 1U, 1V becomes slightly imperfect soft switching, but it is possible to reduce loss and noise compared to complete hard switching.

[0138] (Modification 2) The circuit configuration of the power conversion device 100 according to Modification 2 of Embodiment 1 is the same as that of the power conversion device 100 according to Embodiment 1 (see FIG. 1), and therefore will not be illustrated or described again. The operation of the control device 50 will be described below with reference to FIG. 27 and other figures.

[0139] In the power conversion device 100 according to the second modification, the resonant half-cycle during basic operation is half the dead time period Td during basic operation, and part of the operation of the control device 50 when it is determined that the two-phase resonant currents overlap differs from the operation of the control device 50 according to the first embodiment.

[0140] In the first control operation, the control device 50 overlaps, for a predetermined period, the high-level period of the control signal to each of the two or more switches 8 with the dead time period Td corresponding to each of the two switching circuits 10 connected to two of the multiple switching circuits 10. FIG. 27 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation and the second control operation, and FIG. 28 shows a timing chart of the power conversion device 100 when the control device 50 does not execute the first control operation or the second control operation. In the second modification, the resonant half cycle in the case of basic operation is Td / 2. The predetermined period is, for example, the entire resonant half cycle Tr2 of the resonant circuit including the resonant inductor L1 and the two resonant capacitors 9 connected to the two switches 8, respectively. The resonant half cycle Tr2 is expressed as Tr2=2１／２ ×Td / 2. In the example of Fig. 27, the length of the resonance half cycle Tr2 is shorter than the length of the dead time period Td, and the high level periods of the control signals SU6 and SV6, excluding the additional time Tad, overlap the entire resonance half cycle Tr2. In the second modification, if the resonance half cycle Tr2 is within the dead time period Td in the basic operation, there is no need to lengthen the dead time period Td in the first control operation from the dead time period Td in the basic operation.

[0141] The operation of the control device 50 will now be described in more detail.

[0142] If the control device 50 determines that the resonant currents overlap, it executes the first step, the second step, and the third step in this order. Note that the following describes an example in which it is estimated that the resonant current flowing through the U-phase switch 8U and the V-phase switch 8V overlap in the resonant inductor L1, but the same applies to the cases of the U-phase and the W-phase, and the V-phase and the W-phase.

[0143] In the first step, the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2. In the example of FIG. 27 , the start point of the high-level period of the V-phase control signal SV1 and the end point of the high-level period of the control signal SV2 are shifted earlier by ΔT1 in the example of FIG. 28 , thereby synchronizing the U-phase control signals SU1 and SU2 with the V-phase control signals SV1 and SV2. The first step corresponds to the first control operation of the control device 50. Note that in the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the U-phase control signals SU1 and SU2 later by ΔT1. Alternatively, in the first step, the U-phase control signals SU1 and SU2 and the V-phase control signals SV1 and SV2 may be synchronized by shifting them by a total of ΔT1.

[0144] In the second step, an additional time Tad corresponding to the total current of the load currents iU and iV of the two phases is added to the high-level periods of the control signals SU7 and SV7 sent to the switches 8 of the two phases, corresponding to the dead time periods Td of each of the two phases. The control device 50 determines the additional time Tad by calculating Tad = L × |iU + iV| / V15 using, for example, the detection results of the load currents iU and iV by the current sensors, or their signal processed values, or the estimated values ​​of the load currents iU and iV, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15. The second step corresponds to the second control operation of the control device 50.

[0145] In the third step, the high level periods of the control signals SU7 and SV7 are set to the resonant half period Tr2=2 of the resonant circuit. １／２ ×Td / 2. More specifically, the control device 50 sets the length of the period obtained by subtracting the additional time Tad from the high level period of each of the control signals SU7 and SV7 as the resonant half cycle Tr2 of the resonant circuit. The end points of the control signals SU7 and SV7 may be any time after the end point of the resonant half cycle Tr2.

[0146] When the control device 50 does not execute the first control operation and the second control operation, as shown in FIG. 28, the voltages V across the second switching elements 2U and 2V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end point of the dead time period Td) ２Ｕ , V ２Ｖ does not rise to Vd. In other words, when the control device 50 does not execute the first control operation or the second control operation, charging of the resonant capacitors 9U, 9V does not finish at the end of the dead time period Td. Therefore, in the power conversion device 100 according to the second modification, when the control device 50 does not execute the first control operation or the second control operation, the switching of the first switching elements 1U, 1V becomes hard switching.

[0147] In contrast, in the power conversion device 100 according to the second modification, when the control device 50 executes the first control operation and the second control operation, as shown in FIG. 27, the voltages V across the first switching elements 1U and 1V are equal to or greater than the voltages V across the first switching elements 1U and 1V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end time of the dead time period Td). １Ｕ , V １Ｖ becomes zero. That is, in the power conversion device 100 according to the second modification, when the control device 50 executes the first control operation and the second control operation, the charging of the resonant capacitors 9U, 9V ends at the end of the dead time period Td. Therefore, in the power conversion device 100 according to the second modification, when the control device 50 executes the first control operation and the second control operation, the switching of the first switching elements 1U, 1V becomes zero-voltage soft switching.

[0148] (Modification 3) The circuit configuration of the power conversion device 100 according to Modification 3 of Embodiment 1 is the same as that of the power conversion device 100 according to Embodiment 1 (see FIG. 1), and therefore will not be illustrated or described again. Below, the operation of the control device 50 for soft-switching the first switching element 1 will be described with reference to FIG. 29 and other figures.

[0149] In the power conversion device 100 according to the third modification, the resonant half-cycle during basic operation is half the dead time period Td during basic operation, and part of the operation of the control device 50 when it is determined that the three-phase resonant currents overlap differs from the operation of the control device 50 of the first embodiment.

[0150] In the first control operation, the control device 50 overlaps the high-level period of the control signal to each of the three switches 8 with the dead time period Td corresponding to each of the three switching circuits 10 connected to the three switches 8 by a predetermined period. FIG. 29 shows a timing chart when the first control operation and the second control operation are executed. The predetermined period is, for example, a portion of the resonant half-cycle Tr3 of the resonant circuit including the resonant inductor L1 and the three resonant capacitors 9. In the example of FIG. 29 , the predetermined period is 60% of the resonant half-cycle Tr2. In other words, the length of the predetermined period is 60% of the resonant half-cycle Tr2.

[0151] Furthermore, the control device 50 causes the start and end times of the high level periods of the three control signals SU6, SV6, and SW6 to the first IGBTs 6 of the three switches 8 through which the resonant current flows to coincide with each other.

[0152] The operation of the control device 50 will now be described in more detail.

[0153] When the control device 50 determines that the three-phase resonant currents overlap, it executes the first step and the second step in this order.

[0154] In the first step, the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 are synchronized. In the example of Fig. 29, in the example of Fig. 19, the high-level periods of the V-phase control signals SV1 and SV2 are shifted earlier, and the high-level periods of the W-phase control signals SW1 and SW2 are shifted earlier, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. Note that in the first step, the high-level periods of the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 may also be synchronized by shifting the high-level periods of the U-phase control signals SU1 and SU2 later, and shifting the high-level periods of the W-phase control signals SW1 and SW2 earlier. Alternatively, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed, and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. Alternatively, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed, and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, and the high-level periods of the W-phase control signals SW1 and SW2 may be advanced, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2.

[0155] In the second step, the high level period and the dead time period Td of each of the control signals SU6, SV6, and SW6 are set to the resonant half period Tr3=3 of the resonant circuit. １／２ ×π×(L C) １／２ In more detail, the control device 50 sets the length of each high-level period of the control signals SU6, SV6, and SW6, which is a predetermined period overlapped with the dead time period Td, to a value that is 60% of the resonant half cycle Tr3 of the resonant circuit, for each high-level period of the control signals SU6, SV6, and SW6. Note that the control device 50 sets the additional times Tau, Tav, and Taw to 0 when the load current iU=0, the load current iV=0, and the load current iW=0.

[0156] When the control device 50 does not execute the first control operation and the second control operation, as shown in FIG. 19, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time when the control signals SU1, SV1, and SW1 change from a low level period to a high level period (the end time of the dead time period Td corresponding to the U phase, V phase, and W phase, respectively). ２Ｕ , V ２Ｖ , V ２Ｗ does not rise to Vd. In other words, in the power conversion device 100, when the control device 50 does not execute the first control operation or the second control operation, the charging of the resonant capacitors 9U, 9V, and 9W does not finish at the end of the dead time period Td corresponding to the U phase, V phase, and W phase, respectively. For this reason, in the power conversion device 100, when the control device 50 does not execute the first control operation or the second control operation, the switching of the first switching elements 1U, 1V, and 1W becomes hard switching.

[0157] In contrast, in the power conversion device 100, when the control device 50 executes the first control operation and the second control operation, as shown in FIG. 29, the voltages V across the first switching elements 1U, 1V, and 1W decrease at the time points when the control signals SU1, SV1, and SW1 change from a low level period to a high level period (the end points of the dead time periods Td for the U phase, V phase, and W phase, respectively). １Ｕ , V １Ｖ , V １Ｗrises to a value closer to Vd. For this reason, in the power conversion device 100 according to the third modification, when the control device 50 executes the first control operation and the second control operation, although the switching of the first switching elements 1U, 1V becomes somewhat imperfect soft switching, it is possible to reduce loss and noise more than in the case of complete hard switching.

[0158] (Modification 4) The circuit configuration of the power conversion device 100 according to Modification 4 of Embodiment 1 is the same as that of the power conversion device 100 according to Embodiment 1 (see FIG. 1), and therefore will not be illustrated or described again. Below, the operation of the control device 50 for soft-switching the first switching element 1 will be described with reference to FIG. 30 and the like.

[0159] In the power conversion device 100 according to the fourth modification, the resonant half-cycle during basic operation is half the dead time period Td during basic operation, and part of the operation of the control device 50 when it is determined that the three-phase resonant currents overlap differs from the operation of the control device 50 of the first embodiment.

[0160] In the first control operation, the control device 50 overlaps the high-level period of the control signal to each of the three switches 8 with the dead time period Td corresponding to each of the three switching circuits 10 connected to the three switches 8 for a predetermined period. Fig. 30 shows timing charts when the first control operation, the second control operation, and the third control operation are executed. The predetermined period is, for example, the entire resonant half-cycle Tr3 of the resonant circuit including the resonant inductor L1 and the three resonant capacitors 9. In the example of Fig. 30, the predetermined period is 100% of the resonant half-cycle Tr3.

[0161] Furthermore, the control device 50 causes the start and end times of the high level periods of the three control signals SU6, SV6, and SW6 to the first IGBTs 6 of the three switches 8 through which the resonant current flows to coincide with each other.

[0162] The operation of the control device 50 will now be described in more detail.

[0163] When the control device 50 determines that the three-phase resonant currents overlap, it executes the first step, the second step, and the third step in this order.

[0164] In the first step, the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 are synchronized. In the example of Fig. 30, when iU ≠ 0, iV ≠ 0, iW = 0, and iU > 0 > iV in the example of Fig. 19, the high-level periods of the V-phase control signals SV1 and SV2 are shifted earlier, and the high-level periods of the W-phase control signals SW1 and SW2 are shifted earlier, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed and the high-level periods of the W-phase control signals SW1 and SW2 may be advanced, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. In addition, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be shifted in a later direction, the high-level periods of the V-phase control signals SV1 and SV2 may be shifted in a later direction, and the high-level periods of the W-phase control signals SW1 and SW2 may be shifted in an earlier direction, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2.

[0165] In the second step, the additional time Tad of the high level period of each of the control signals SU6, SV6, and SW6 is changed to a value calculated using the value of the total load current (0 in the example of Figure 30) as Tad = |iU + iV + iW| / V15.

[0166] In the third step, the high level period and the dead time period Td of each of the control signals SU6, SV6, and SW6 are set to the resonant half period Tr3=3 of the resonant circuit. １／２ ×π×(L C) １／２ More specifically, the control device 50 sets the length of the high level period of each of the control signals SU6, SV6, and SW6 to the resonant half cycle Tr3 of the resonant circuit, and sets the dead time period Td to a dead time period Td2 equal to the resonant half cycle Tr3 of the resonant circuit.

[0167] In the power conversion device 100 according to the fourth modification, when the control device 50 does not execute the first control operation, the second control operation, and the third control operation, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time when the control signals SU1, SV1, and SW1 change from a low level period to a high level period (the end time of the dead time period Td corresponding to the U phase, V phase, and W phase, respectively). １Ｕ , V １Ｖ , V １Ｗ does not rise to Vd. In other words, in the power conversion device 100, when the control device 50 does not execute the first control operation, the second control operation, or the third control operation, the charging of the resonant capacitors 9U, 9V, and 9W does not finish at the end of the dead time period Td corresponding to the U phase, the V phase, or the W phase. For this reason, in the power conversion device 100, when the control device 50 does not execute the first control operation, the second control operation, or the third control operation, the switching of the first switching elements 1U, 1V, and 1W becomes hard switching.

[0168] In contrast to this, in the power conversion device 100 according to the fourth modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, as shown in FIG. 30, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time points when the control signals SU1, SV1, and SW1 change from the low level period to the high level period (the end points of the dead time periods Td2 for the U phase, V phase, and W phase, respectively). ２Ｕ , V ２Ｖ , V ２Ｗrises to Vd. Therefore, in the power conversion device 100 according to the fourth modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, it is possible to realize zero-voltage soft switching of the first switching elements 1U and 1V.

[0169] (Modification 5) The circuit configuration of the power conversion device 100 according to Modification 5 of Embodiment 1 is the same as that of the power conversion device 100 according to Embodiment 1 (see FIG. 1), and therefore will not be illustrated or described again. Below, the operation of the control device 50 for soft-switching the first switching element 1 will be described with reference to FIG. 31 and the like.

[0170] In the power conversion device 100 according to the fifth modification, the resonant half-cycle during basic operation is half the dead time period Td during basic operation, and part of the operation of the control device 50 when it is determined that the three-phase resonant currents overlap differs from the operation of the control device 50 of the first embodiment.

[0171] In the first control operation, the control device 50 overlaps, for a predetermined period, the high-level periods of the control signals to the three switches 8 with the dead time periods Td corresponding to the three switching circuits 10 connected to the three switches 8. Fig. 31 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation and the second control operation. Fig. 32 shows a timing chart of the power conversion device 100 when the control device 50 does not execute the first control operation or the second control operation. The predetermined period is, for example, the entire resonant half cycle Tr3 of the resonant circuit including the resonant inductor L1 and the three resonant capacitors 9.

[0172] Furthermore, the control device 50 causes the start and end times of the high level periods of the three control signals SU6, SV6, and SW6 to the first IGBTs 6 of the three switches 8 through which the resonant current flows to coincide with each other.

[0173] The operation of the control device 50 will now be described in more detail.

[0174] When the control device 50 determines that the three-phase resonant currents overlap, it executes the first step and the second step in this order.

[0175] In the first step, the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 are synchronized. In the example of Fig. 31, the high-level periods of the V-phase control signals SV1 and SV2 are shifted earlier and the high-level periods of the W-phase control signals SW1 and SW2 are shifted earlier in the example of Fig. 32, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. Note that in the first step, the high-level periods of the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2 may also be synchronized by shifting the high-level periods of the U-phase control signals SU1 and SU2 later and shifting the high-level periods of the W-phase control signals SW1 and SW2 earlier. Alternatively, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed, and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2. Alternatively, in the first step, the high-level periods of the U-phase control signals SU1 and SU2 may be delayed, and the high-level periods of the V-phase control signals SV1 and SV2 may be delayed, and the high-level periods of the W-phase control signals SW1 and SW2 may be advanced, thereby synchronizing the U-phase control signals SU1 and SU2, the V-phase control signals SV1 and SV2, and the W-phase control signals SW1 and SW2.

[0176] In the second step, the length of the high level period of each of the control signals SU6, SV6, and SW6 is set to Tr3=3, which is the resonant half period of the resonant circuit. １／２ ×Td / 2.

[0177] In the power conversion device 100 according to the fifth modification, when the control device 50 does not execute the first control operation and the second control operation, as shown in FIG. 32, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time points when the control signals SU1, SV1, and SW1 change from a low level period to a high level period (the end points of the dead time periods Td corresponding to the U phase, V phase, and W phase, respectively). ２Ｕ , V ２Ｖ , V ２Ｗ does not rise to Vd. In other words, in the power conversion device 100, when the control device 50 does not execute the first control operation or the second control operation, the charging of the resonant capacitors 9U, 9V, and 9W does not finish at the end of the dead time period Td corresponding to the U phase, V phase, and W phase, respectively. For this reason, in the power conversion device 100, when the control device 50 does not execute the first control operation or the second control operation, the switching of the first switching elements 1U, 1V, and 1W becomes hard switching.

[0178] In contrast, in the power conversion device 100 according to the fifth modification, when the control device 50 executes the first control operation and the second control operation, as shown in FIG. 31, the voltages V across the second switching elements 2U, 2V, and 2W are equal to or higher than the voltages V across the second switching elements 2U, 2V, and 2W at the time points when the control signals SU1, SV1, and SW1 change from the low level period to the high level period (the end points of the dead time periods Td for the U phase, V phase, and W phase, respectively). ２Ｕ , V ２Ｖ , V ２Ｗ rises to Vd. Therefore, in the power conversion device 100 according to the fifth modification, when the control device 50 executes the first control operation and the second control operation, it is possible to realize zero-voltage soft switching of the first switching elements 1U and 1V.

[0179] (Modification 6) The circuit configuration of the power conversion device 100 according to Modification 6 of Embodiment 1 is the same as that of the power conversion device 100 according to Embodiment 1 (see FIG. 1), and therefore will not be illustrated or described again. The operation of the control device 50 will be described below with reference to FIGS. 33 and 34 etc.

[0180] In the power conversion device 100, when the control device 50 determines that the two-phase resonant currents overlap, it is not essential that the high-level periods of the control signals to the switches 8 of the two phases completely overlap, and it is sufficient to output control signals SV6, SW6 such that the switch 8 of one of the two phases is conductive until the current iL1 flowing through the resonant inductor L1 reaches iV + iW, which is the sum of the load currents of the two phases, as shown in Fig. 33. Fig. 33 shows an example in which the resonant current of the V phase and the resonant current of the W phase overlap, but this is not limiting, and the same applies to the case in which the resonant current of the U phase and the resonant current of the V phase overlap, and the case in which the resonant current of the U phase and the resonant current of the W phase overlap.

[0181] In the power conversion device 100 according to the sixth modification, the operation of the control device 50 when it is determined that the two-phase resonant currents overlap is partially different from the operation of the control device 50 according to the first embodiment.

[0182] In the first control operation, the control device 50 overlaps the high-level period of the control signal to each of the two switches 8 with the dead time period Td corresponding to each of the two switching circuits 10 connected to the two switches 8 for a predetermined period. Fig. 34 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation, the second control operation, and the third control operation. In the sixth modification, the resonant half cycle in the case of basic operation is the same as the length of the dead time period Td, and the predetermined period is, for example, the entire resonant half cycle Tr2 of the resonant circuit including the resonant inductor L1 and the two resonant capacitors 9 connected to each of the two switches 8. The resonant half cycle Tr2 is expressed as Tr2=2 １／２ 34, the dead time period Td2 is the same as the length of the resonance half cycle Tr2, and the high level periods of the control signals SU6 and SV6 overlap the entire resonance half cycle Tr2.

[0183] The operation of the control device 50 will now be described in more detail.

[0184] If the control device 50 determines that the resonant currents overlap, it executes the first step, the second step, and the third step in this order. Note that the following describes an example in which it is estimated that the resonant current flowing through the U-phase switch 8U and the V-phase switch 8V overlap in the resonant inductor L1, but the same applies to the cases of the U-phase and the W-phase, and the V-phase and the W-phase.

[0185] In the first step, the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2. In the example of FIG. 34 , the U-phase control signals SU1 and SU2 are synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the V-phase control signals SV1 and SV2 in the example of FIG. 10 earlier by ΔT1. Note that in the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting the high-level periods of the U-phase control signals SU1 and SU2 in a later direction by ΔT1. Alternatively, in the first step, the U-phase control signals SU1 and SU2 may be synchronized with the V-phase control signals SV1 and SV2 by shifting them a total of ΔT1.

[0186] In the second step, an additional time Tad corresponding to the total current of the two-phase load currents iU and iV is added to the high-level period of the control signal SU6 sent to the U-phase switch 8U. The control device 50 determines the additional time Tad by calculating Tad=L×|iU+iV| / V15 using, for example, the detection results of the load currents iU and iV by the current sensors or their signal processed values, or the estimated values ​​of the load currents iU and iV, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15.

[0187] In the third step, the high level period and the dead time period Td of each of the control signals SU6 and SV6 are set to Tr2=2, which is the resonant half period of the resonant circuit. １／２More specifically, the control device 50 sets the length of the period obtained by subtracting the additional time Tad from the high level period of the control signal SU6 to the resonance half period Tr2, sets the length of the high level period of the control signal SV6 to the resonance half period Tr2, and changes the dead time period Td1 to Td1=2 １／２ The end points of the control signals SU6 and SV6 may be any point after the end point of the resonance half cycle Tr2.

[0188] When the control device 50 does not execute the first control operation, the second control operation, and the third control operation, as shown in FIG. 10, the voltages V across the second switching elements 2U and 2V are equal to or higher than the voltages V across the second switching elements 2U and 2V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end point of the dead time period Td). ２Ｕ , V ２Ｖ does not rise to Vd. In other words, when the control device 50 does not perform the first control operation, the second control operation, or the third control operation, the charging of the resonant capacitors 9U, 9V does not finish at the end of the dead time period Td. Therefore, in the power conversion device 100 according to the sixth modification, when the control device 50 does not perform the first control operation, the second control operation, or the third control operation, the switching of the first switching elements 1U, 1V becomes hard switching.

[0189] In contrast, in the power conversion device 100 according to the sixth modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, as shown in FIG. 34, the voltages V across the first switching elements 1U and 1V are equal to or greater than the voltages V across the first switching elements 1U and 1V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end time of the dead time period Td). １Ｕ , V １Ｖ becomes zero. That is, in the power conversion device 100 according to the sixth modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the charging of the resonant capacitors 9U, 9V ends at the end of the dead time period Td. Therefore, in the power conversion device 100 according to the sixth modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the switching of the first switching elements 1U, 1V becomes zero-voltage soft switching.

[0190] (Variation 7) The circuit configuration of the power conversion device 100 according to Variation 7 of Embodiment 1 is the same as that of the power conversion device 100 according to Embodiment 1 (see FIG. 1 ), and therefore will not be illustrated or described again. The operation of the control device 50 will be described below with reference to FIG. 35 and the like.

[0191] In the power conversion device 100 according to the seventh modification, the resonant half-cycle during basic operation is half the dead time period Td during basic operation, and part of the operation of the control device 50 when it is determined that the two-phase resonant currents overlap differs from the operation of the control device 50 according to the first embodiment.

[0192] In the first control operation, the control device 50 overlaps the high-level period of the control signal to each of the two switches 8 with the dead time period Td corresponding to each of the two switching circuits 10 connected to the two switches 8 for a predetermined period. Fig. 35 shows a timing chart of the power conversion device 100 when the control device 50 executes the first control operation, the second control operation, and the third control operation. In the seventh modification, the resonant half cycle in the case of basic operation is the same as half the length of the dead time period Td, and the predetermined period is, for example, the entire resonant half cycle Tr2 of the resonant circuit including the resonant inductor L1 and the two resonant capacitors 9 connected to each of the two switches 8. The resonant half cycle Tr2 is expressed as Tr2=2 １／２ ×Td / 2. In the example of Figure 35, the length of the resonance half cycle Tr2 is shorter than the length of the dead time period Td, the high level period of the control signal SU6 excluding the additional time Tad overlaps the entire resonance half cycle Tr2, and the high level period of the control signal SV6 overlaps the entire resonance half cycle Tr2. In the seventh modification, if the resonance half cycle Tr2 is within the dead time period Td in the basic operation, there is no need to lengthen the dead time period Td in the first control operation from the dead time period Td in the basic operation.

[0193] The operation of the control device 50 will now be described in more detail.

[0194] If the control device 50 determines that the resonant currents overlap, it executes the first step, the second step, and the third step in this order. Note that the following describes an example in which it is estimated that the resonant current flowing through the U-phase switch 8U and the V-phase switch 8V overlap in the resonant inductor L1, but the same applies to the cases of the U-phase and the W-phase, and the V-phase and the W-phase.

[0195] In the first step, the end points of the U-phase control signal SU6 and the V-phase control signal SV6 are synchronized. In the example of Fig. 35, the high-level period of the V-phase control signal SV6 in the example of Fig. 28 is shifted forward by ΔT1, thereby synchronizing the U-phase control signal SU6 and the V-phase control signal SV6.

[0196] In the second step, an additional time Tad corresponding to the total current of the two-phase load currents iU and iV is added to the high-level period of the control signal SU6 sent to the U-phase switch 8U. The control device 50 determines the additional time Tad by calculating Tad=L×|iU+iV| / V15 using, for example, the detection results of the load currents iU and iV by the current sensors or their signal processed values, or the estimated values ​​of the load currents iU and iV, the inductance L of the resonance inductor L1 that is stored in advance, and the detection result of the potential V15 of the regenerative capacitor 15.

[0197] In the third step, the high level periods of the control signals SU6 and SV6 are set to the resonant half period Tr2=2 of the resonant circuit. １／２ ×Td. More specifically, the control device 50 sets the length of the high-level period of the control signal SU6 minus the additional time Tad to the resonance half cycle Tr2, and sets the length of the high-level period of the control signal SV6 to the resonance half cycle Tr2. Note that the end points of the control signals SU6 and SV6 may be any point after the end point of the resonance half cycle Tr2.

[0198] When the control device 50 does not execute the first control operation, the second control operation, and the third control operation, as shown in FIG. 28, the voltages V across the second switching elements 2U and 2V are equal to or greater than V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end point of the dead time period Td). ２Ｕ , V ２Ｖ does not rise to Vd. In other words, when the control device 50 does not perform the first control operation, the second control operation, or the third control operation, the charging of the resonant capacitors 9U, 9V does not finish at the end of the dead time period Td. For this reason, in the power conversion device 100 according to the seventh modification, when the control device 50 does not perform the first control operation, the second control operation, or the third control operation, the switching of the first switching elements 1U, 1V becomes hard switching.

[0199] In contrast, in the power conversion device 100 according to the seventh modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, as shown in FIG. 35, the voltages V across the first switching elements 1U and 1V are equal to or greater than the voltages V across the first switching elements 1U and 1V at the time when the control signals SU1 and SV1 change from the low level period to the high level period (the end time of the dead time period Td). １Ｕ , V １Ｖ becomes zero. That is, in the power conversion device 100 according to the seventh modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the charging of the resonant capacitors 9U, 9V ends at the end of the dead time period Td. Therefore, in the power conversion device 100 according to the seventh modification, when the control device 50 executes the first control operation, the second control operation, and the third control operation, the switching of the first switching elements 1U, 1V becomes zero-voltage soft switching.

[0200] (Modification 8) A power conversion device 100A according to Modification 8 of Embodiment 1 will be described with reference to Fig. 36. With regard to the power conversion device 100A according to Modification 8 of Embodiment 1, components similar to those of the power conversion device 100 according to Embodiment 1 are denoted by the same reference numerals, and description thereof will be omitted.

[0201] In a power conversion device 100A according to the eighth modification, the first IGBT 6 and the second IGBT 7 are connected in anti-series in each of the plurality of switches 8. In the power conversion device 100A according to the eighth modification, the collector terminal of the first IGBT 6 and the collector terminal of the second IGBT 7 are connected in each of the plurality of switches 8, the emitter terminal of the first IGBT 6 is connected to the connection point 3 of a corresponding one of the plurality of switching circuits 10, and the emitter terminal of the second IGBT 7 is connected to the common connection point 25. Each of the plurality of switches 8 further includes a diode 61 connected in anti-parallel to the first IGBT 6 and a diode 71 connected in anti-parallel to the second IGBT 7.

[0202] In the power conversion device 100A according to the eighth modification, each of the first IGBT 6 and the second IGBT 7 may be replaced with a MOSFET or a bipolar transistor. In this case, the diode 61 and the diode 71 in FIG. 36 may be substituted with a parasitic diode of the replaced element, or an element built into the chip of the replaced element. Furthermore, in the power conversion device 100A according to the first modification, the diode 61 and the diode 71 are not limited to being externally connected to the first IGBT 6 and the second IGBT 7, but may be an element built into the chip.

[0203] (Modification 9) A power conversion device 100A according to Modification 9 of Embodiment 1 will be described with reference to Fig. 37. Regarding the power conversion device 100A according to Modification 10 of Embodiment 1, components similar to those of the power conversion device 100 according to Embodiment 1 are denoted by the same reference numerals, and description thereof will be omitted.

[0204] In a power conversion device 100A according to the 9th modification, the first IGBT 6 and the second IGBT 7 are connected in anti-series in each of the plurality of switches 8. In the power conversion device 100A according to the 9th modification, the emitter terminal of the first IGBT 6 and the emitter terminal of the second IGBT 7 are connected in each of the plurality of switches 8, the collector terminal of the first IGBT 6 is connected to the connection point 3 of a corresponding one of the plurality of switching circuits 10, and the collector terminal of the second IGBT 7 is connected to the common connection point 25. Each of the plurality of switches 8 further includes a diode 61 connected in anti-parallel to the first IGBT 6 and a diode 71 connected in anti-parallel to the second IGBT 7.

[0205] In the power conversion device 100A according to the 9th modification, each of the first IGBT 6 and the second IGBT 7 may be replaced with a MOSFET or a bipolar transistor. In this case, the diode 61 and the diode 71 in FIG. 37 may be substituted with a parasitic diode of the replaced element, or an element built into the chip of the replaced element. Furthermore, in the power conversion device 100A according to the 9th modification, the diode 61 and the diode 71 are not limited to being externally attached to the first IGBT 6 and the second IGBT 7, but may also be an element built into the chip.

[0206] (Modification 10) A power conversion device 100A according to Modification 10 of Embodiment 1 will be described with reference to Fig. 38. Regarding the power conversion device 100A according to Modification 10 of Embodiment 1, components that are the same as those of the power conversion device 100 according to Embodiment 1 are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0207] In a power conversion device 100A according to Modification 10, a first MOSFET 6A and a second MOSFET 7A are connected in anti-series in each of a plurality of switches 8. In the power conversion device 100A according to Modification 10, the drain terminal of the first MOSFET 6A and the drain terminal of the second MOSFET 7A are connected in anti-parallel in each of the plurality of switches 8. Each of the plurality of switches 8 further includes a diode 61 connected in anti-parallel to the first MOSFET 6A and a diode 71 connected in anti-parallel to the second MOSFET 7A. In each of the plurality of switches 8, the source terminal of the second MOSFET 7A is connected to a common connection point 25. In each of the plurality of switches 8, the source terminal of the first MOSFET 6A is connected to a connection point 3 of a switching circuit 10 corresponding to the switch 8 having the first MOSFET 6A. Control signals SU6 and SU7 are provided from a control device 50 to the first MOSFET 6A and the second MOSFET 7A of a switch 8U. The first MOSFET 6A and the second MOSFET 7A of the switch 8V are supplied with control signals SV6 and SV7 from the control device 50. The first MOSFET 6A and the second MOSFET 7A of the switch 8W are supplied with control signals SW6 and SW7 from the control device 50.

[0208] (Modification 11) A power conversion device 100A according to Modification 11 of Embodiment 1 will be described with reference to Fig. 39. Regarding the power conversion device 100A according to Modification 11 of Embodiment 1, components similar to those of the power conversion device 100 according to Embodiment 1 will be denoted by the same reference numerals and descriptions thereof will be omitted.

[0209] In a power conversion device 100A according to the eleventh modification, a diode 63 is connected in series to the first MOSFET 6A, and a diode 73 is connected in series to the second MOSFET 7A in each of the multiple switches 8. In the power conversion device 100A according to the eleventh modification, a series circuit of the first MOSFET 6A and the diode 63 and a series circuit of the second MOSFET 7A and the diode 73 are connected in anti-parallel.

[0210] (Modification 12) A power conversion device 100A according to Modification 12 of Embodiment 1 will be described with reference to Fig. 40. Regarding the power conversion device 100A according to Modification 12 of Embodiment 1, components similar to those of the power conversion device 100 according to Embodiment 1 are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0211] In a power conversion device 100A according to the twelfth modification, each of the multiple switches 8 includes one MOSFET 80, a diode 83 connected in anti-parallel to the MOSFET 80, a series circuit of two diodes 84 and 85 connected in anti-parallel to the MOSFET 80, and a series circuit of two diodes 86 and 87 connected in anti-parallel to the MOSFET 80. In each of the multiple switches 8, a connection point between the diode 84 and the diode 85 in the switch 8 (a first end 81 of the switch 8) is connected to a connection point 3 of a corresponding switching circuit 10 among the multiple switching circuits 10, and a connection point between the diode 86 and the diode 87 (a second end 82 of the switch 8) is connected to the common connection point 25. In each of the switches 8, when the MOSFET 80 is in an on state, the switch 8 is in an on state, and when the MOSFET 80 is in an off state, the switch 8 is in an off state.

[0212] The MOSFETs 80 of the multiple switches 8 are controlled by a control device 50. The control device 50 outputs a control signal SU8 that controls the on / off state of the MOSFET 80 of the switch 8U, a control signal SV8 that controls the on / off state of the MOSFET 80 of the switch 8V, and a control signal SW8 that controls the on / off state of the MOSFET 80 of the switch 8W.

[0213] In the switch 8, when the MOSFET 80 is in the on state, a resonant current flows due to a resonant circuit including the resonant inductor L1 and the resonant capacitor 9. In the power conversion device 100A, when one of the multiple switches 8 is in the on state, a charging current including the resonant current flows through the path of the regenerative capacitor 15, the resonant inductor L1, the diode 86, the MOSFET 80, the diode 85, and the resonant capacitor 9. In addition, in the power conversion device 100A, when one of the multiple switches 8 is in the on state, a discharging current including the resonant current flows through the path of the resonant capacitor 9, the diode 84, the MOSFET 80, the diode 87, the resonant inductor L1, and the regenerative capacitor 15.

[0214] In the power conversion device 100A according to the twelfth modification, each of the plurality of MOSFETs 80 may be replaced with an IGBT. Furthermore, in the power conversion device 100A according to the twelfth modification, each of the plurality of switches 8 may have, instead of the MOSFET 80, a bipolar transistor or a GaN-based GIT (Gate Injection Transistor), for example.

[0215] (Modification 13) A power conversion device 100A according to Modification 13 of Embodiment 1 will be described with reference to Fig. 41. Regarding the power conversion device 100A according to Modification 13 of Embodiment 1, components similar to those of the power conversion device 100 according to Embodiment 1 are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0216] In the power conversion device 100 according to the thirteenth modification, each of the multiple switches 8 is a dual-gate GaN-based GIT having a first source terminal, a first gate terminal, a second gate terminal, and a second source terminal. In the power conversion device 100A according to the thirteenth modification, a control signal SU6 is applied between the first gate terminal and the first source terminal of the dual-gate GaN-based GIT constituting the switch 8U, and a control signal SU7 is applied between the second gate terminal and the second source terminal. Furthermore, a control signal SV6 is applied between the first gate terminal and the first source terminal of the dual-gate GaN-based GIT constituting the switch 8V, and a control signal SV7 is applied between the second gate terminal and the second source terminal. Furthermore, a control signal SW6 is applied between the first gate terminal and the first source terminal of the dual-gate GaN-based GIT constituting the switch 8W, and a control signal SW7 is applied between the second gate terminal and the second source terminal.

[0217] (Embodiment 2) A power conversion device 100B according to embodiment 2 will be described below with reference to Fig. 42. The power conversion device 100B according to embodiment 2 differs from the power conversion device 100 according to embodiment 1 (see Fig. 1) in that it further includes a capacitor 16 connected between the second end of the resonance inductor L1 and the first DC terminal 31. In the power conversion device 100B according to embodiment 2, components similar to those of the power conversion device 100 according to embodiment 1 are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0218] The power conversion device 100B does not include the capacitor C10 in the power conversion device 100 according to the first embodiment. The capacitor 16 is connected in series with the regenerative capacitor 15. Therefore, in the power conversion device 100B, the series circuit of the capacitor 16 and the regenerative capacitor 15 is connected between the first DC terminal 31 and the second DC terminal 32. The capacitance of the capacitor 16 is the same as the capacitance of the regenerative capacitor 15. The phrase "the capacitance of the capacitor 16 is the same as the capacitance of the regenerative capacitor 15" does not necessarily mean that the capacitance of the capacitor 16 exactly matches the capacitance of the regenerative capacitor 15, but may mean that the capacitance of the capacitor 16 is within a range of 95% to 105% of the capacitance of the regenerative capacitor 15.

[0219] In the power conversion device 100B according to the second embodiment, the potential V15 at the fourth terminal 154 of the regenerative capacitor 15 is equal to the voltage value Vd of the DC power supply E1 divided by the capacitor 16 and the regenerative capacitor 15. Therefore, the potential V15 at the fourth terminal 154 of the regenerative capacitor 15 is Vd / 2. In the power conversion device 100B according to the second embodiment, the control device 50 may store the value of the potential V15 at the fourth terminal 154 of the regenerative capacitor 15 in advance.

[0220] The control device 50 of the power conversion device 100B according to the second embodiment performs a first control operation, a second control operation, and a third control operation, similar to the control device 50 of the power conversion device 100 according to the first embodiment. Therefore, similar to the power conversion device 100 according to the first embodiment, the power conversion device 100B according to the second embodiment can realize zero voltage soft switching of each of the plurality of first switching elements 1 and the plurality of second switching elements 2.

[0221] (Third Embodiment) A power conversion device 100C according to a third embodiment will be described below with reference to Fig. 43. The power conversion device 100C according to the third embodiment differs from the power conversion device 100 according to the first embodiment (see Fig. 1) in that a regenerative capacitor 15 is connected between the second end of the resonance inductor L1 and the first DC terminal 31. In the power conversion device 100C according to the third embodiment, components similar to those of the power conversion device 100 according to the first embodiment are denoted by the same reference numerals, and descriptions thereof will be omitted.

[0222] The control device 50 of the power conversion device 100C according to the third embodiment performs the first control operation, the second control operation, and the third control operation, similar to the control device 50 of the power conversion device 100 according to the first embodiment. Therefore, the power conversion device 100C according to the third embodiment can perform soft switching more reliably, similar to the power conversion device 100 according to the first embodiment.

[0223] (Other Modifications) The above-described first to third embodiments are merely examples of various embodiments of the present disclosure. The above-described first to third embodiments can be modified in various ways depending on the design, etc., as long as the object of the present disclosure can be achieved.

[0224] In the control device 50, the operation of "determining that multiple resonant currents are flowing simultaneously" is not limited to the operation of "determining that multiple resonant currents are flowing simultaneously" when the above-mentioned time difference is less than a threshold value, or the operation of "determining that three-phase resonant currents are flowing simultaneously" when the rotation speed of the above-mentioned motor is less than a rotation speed threshold value, etc.

[0225] For example, the control device 50 may determine that three-phase resonant currents are flowing simultaneously when all of the following are less than a threshold value: the time difference between the start of the high-level period of the control signal corresponding to the U phase and the start of the high-level period of the control signal corresponding to the V phase; the time difference between the start of the high-level period of the control signal corresponding to the V phase and the start of the high-level period of the control signal corresponding to the W phase; and the time difference between the start of the high-level period of the control signal corresponding to the W phase and the start of the high-level period of the control signal corresponding to the U phase.

[0226] In addition, the control device 50 may determine that two-phase resonant currents are flowing simultaneously when any one of the current difference between the U-phase load current iU and the V-phase load current iV, the current difference between the V-phase load current iV and the W-phase load current iW, and the current difference between the W-phase load current iW and the U-phase load current iU is less than a current difference threshold.

[0227] In addition, the control device 50 may determine that three-phase resonant currents are flowing simultaneously when any of the current differences between the U-phase load current iU and the V-phase load current iV, the V-phase load current iV and the W-phase load current iW, and the W-phase load current iW and the U-phase load current iU are less than the current difference threshold.

[0228] In addition, the control device 50 may determine that "two-phase resonant currents flow simultaneously" when the electrical angle calculated from sensor information output from a sensor device (e.g., an encoder or resolver) for detecting the rotation speed of the motor, or the estimated electrical angle, is within a first rotation angle range (e.g., 55 degrees or more and 65 degrees or less), a second rotation angle range (e.g., 115 degrees or more and 125 degrees or less), a third rotation angle range (e.g., 175 degrees or more and 185 degrees or less), a fourth rotation angle range (e.g., 235 degrees or more and 245 degrees or less), a fifth rotation angle range (295 degrees or more and 305 degrees or less), or a sixth rotation angle range (e.g., 355 degrees or more and 365 degrees or less).

[0229] Furthermore, each of the plurality of first switching elements 1 and the plurality of second switching elements 2 is not limited to an IGBT, but may be a MOSFET. In this case, each of the plurality of first diodes 4 may be substituted with a parasitic diode of the MOSFET constituting the corresponding first switching element 1. Also, each of the plurality of second diodes 5 may be substituted with a parasitic diode of the MOSFET constituting the corresponding second switching element 2. The MOSFET is, for example, a Si-based MOSFET or a SiC-based MOSFET. Each of the plurality of first switching elements 1 and the plurality of second switching elements 2 may be, for example, a bipolar transistor or a GaN-based GIT.

[0230] Furthermore, in the power conversion devices 100, 100A, 100B, and 100C, if the capacitance of each of the multiple resonant capacitors 9 is relatively small, instead of attaching the multiple resonant capacitors 9 externally, the parasitic capacitance between both ends of the multiple second switching elements 2 may also serve as the multiple resonant capacitors 9.

[0231] Furthermore, each of the switches 8 in the second and third embodiments other than the first embodiment may have the configuration shown in any one of the examples of FIGS.

[0232] The length of the dead time period Td is not limited to being set to be the same as the resonance half cycle, but may be set to a length different from the resonance half cycle. In either case, however, it is preferable that the end point of the dead time period Td coincides with the end point of the resonance half cycle.

[0233] The dead time period Td may be set by a dead time generation circuit included in a gate driver IC (Integrated Circuit) or the like that is provided separately from the control device 50. Alternatively, the control device 50 may include a gate driver IC, and the dead time period Td may be set by a dead time generation circuit included in the gate driver IC.

[0234] Furthermore, the power conversion devices 100, 100A, 100B, and 100C are not limited to being configured to output three-phase AC, but may be configured to output polyphase AC with three or more phases.

[0235] (Aspects) The following aspects are disclosed in this specification.

[0236] A power conversion device (100; 100A; 100B; 100C) according to a first aspect includes a first DC terminal (31), a second DC terminal (32), a power conversion circuit (11), a plurality of AC terminals (41), a plurality of switches (8), a plurality of resonant capacitors (9), a resonant inductor (L1), a regenerative capacitor (15), and a control device (50). The power conversion circuit (11) has a plurality of first switching elements (1) and a plurality of second switching elements (2). In the power conversion circuit (11), a plurality of switching circuits (10) are connected in parallel to each other, each of which has a plurality of first switching elements (1) and a plurality of second switching elements (2) connected in series in a one-to-one relationship. In the power conversion circuit (11), the plurality of first switching elements (1) are connected to the first DC terminal (31), and the plurality of second switching elements (2) are connected to the second DC terminal (32). The plurality of AC terminals (41) correspond one-to-one to the plurality of switching circuits (10). Each of the plurality of AC terminals (41) is connected to a connection point (3) between a first switching element (1) and a second switching element (2) in the corresponding switching circuit (10). The plurality of switches (8) correspond one-to-one to the plurality of switching circuits (10). Each of the plurality of switches (8) has a first end (81) connected to a connection point (3) between the first switching element (1) and the second switching element (2) in the corresponding switching circuit (10), and a second end (82) commonly connected to a common connection point (25). The resonant capacitors (9) correspond one-to-one to the plurality of switches (8). Each of the plurality of resonant capacitors (9) is connected between a first end (81) of the corresponding switch (8) and a second DC terminal (32). The resonant inductor (L1) has a first end and a second end. In the resonant inductor (L1), a first end of the resonant inductor (L1) is connected to the common connection point (25). The regenerative capacitor (15) has a third end (153) and a fourth end (154). In the regenerative capacitor (15), the third end (153) is connected to the first DC terminal (31) or the second DC terminal (32).The control device (50) provides a control signal whose potential changes between a high level and a low level to each of a plurality of first switching elements (1), a plurality of second switching elements (2), and a plurality of switches (8). The control device (50) is capable of executing a first control operation and a second control operation when it determines that a resonant current flows simultaneously through two or more switches (8) of the plurality of switches (8) through the resonant inductor (L1). The first control operation overlaps a high-level period of the control signal to each of the two or more switches (8) by a predetermined period with a dead time period (Td) corresponding to each of two or more switching circuits (10) connected to the two or more switches (8) of the plurality of switching circuits (10). The second control operation determines a start point of the high-level period of the control signal to at least one switch (8) of the plurality of switches (8) in accordance with at least one phase of load current flowing through an AC load (RA1) connected to a plurality of AC terminals (41).

[0237] According to this aspect, soft switching can be performed more reliably.

[0238] In the power conversion device (100; 100A; 100B; 100C) according to the second aspect, in the first aspect, the predetermined period is at least a portion of the resonant half cycle (Tr2, Tr3) of a resonant circuit including a resonant inductor (L1) and two or more resonant capacitors (9) connected to two or more switches (8), respectively.

[0239] In the power conversion device (100; 100A; 100B; 100C) according to the third aspect, in the second aspect, the predetermined period is the entire period of the resonance half cycle (Tr2, Tr3).

[0240] According to this embodiment, it is possible to more reliably achieve zero voltage soft switching.

[0241] A power conversion device (100; 100A; 100B; 100C) according to a fourth aspect is based on any one of the first to third aspects. In a second control operation of the control device (50), a start point of a high-level period of a control signal to one switch (8) among the plurality of switches (8) is changed according to a total value of two or more phases of load current flowing through two or more AC terminals (41) connected to two or more switches (8) among the plurality of AC terminals (41).

[0242] According to this aspect, it is possible to start resonance at the start of the dead time period (Td1, Td2).

[0243] In the power conversion device (100; 100A; 100B; 100C) according to the fifth aspect, in any one of the first to fourth aspects, the control device (50) is capable of executing a third control operation of making the dead time periods (Td1, Td2) corresponding to each of two or more switching circuits (10) connected to two or more switches (8) among the plurality of switching circuits (10) longer than the predetermined dead time period (Td) by an additional time (Tad).

[0244] According to this aspect, even if the resonance half cycle (Tr2, Tr3) is longer than the dead time period (Td), it is possible to achieve zero voltage soft switching.

[0245] 1 First switching element 2 Second switching element 3 Connection point 8 Switch 81 First terminal 82 Second terminal 9 Resonant capacitor 10 Switching circuit 11 Power conversion circuit 15 Regenerative capacitor 153 Third terminal 154 Fourth terminal 25 Common connection point 31 First DC terminal 32 Second DC terminal 41 AC terminal 50 Control device 100, 100A, 100B, 100C Power conversion device iU, iV, iW Output current (load current) L1 Resonant inductor RA1 AC load Tad Additional time Td Dead time period Td1 Dead time period Td2 Dead time period Tr2 Resonant half cycle Tr3 Resonant half cycle

Claims

1. A first DC terminal and a second DC terminal; a power conversion circuit including a plurality of first switching elements and a plurality of second switching elements, the plurality of first switching elements being connected in series to the plurality of second switching elements in a one-to-one relationship, the plurality of switching circuits being connected in parallel to each other, the plurality of first switching elements being connected to the first DC terminal, and the plurality of second switching elements being connected to the second DC terminal; a plurality of AC terminals in one-to-one correspondence with the plurality of switching circuits, each AC terminal being connected to a connection point of the first switching element and the second switching element in the corresponding switching circuit; a plurality of switches each corresponding to the plurality of switching circuits, each having a first end connected to the connection point of the first switching element and the second switching element in the corresponding switching circuit and a second end commonly connected to a common connection point; a plurality of resonance capacitors corresponding to the plurality of switches one-to-one, each of the resonance capacitors being connected between the first end and the second DC terminal of the corresponding switch; a resonant inductor having a first end and a second end, the first end being connected to the common connection point; a regenerative capacitor having a third end and a fourth end, the third end being connected to the first DC terminal or the second DC terminal; a control device that provides a control signal whose potential changes between a high level and a low level to each of the first switching elements, the second switching elements, and the switches; The control device includes: When it is determined that a resonant current flows through each of two or more switches among the plurality of switches simultaneously through the resonant inductor, a first control operation for overlapping a high level period of a control signal for each of the two or more switches with a dead time period corresponding to each of the two or more switching circuits connected to the two or more switches among the plurality of switching circuits for a predetermined period; a second control operation of determining a start time point of a high level period of a control signal to at least one switch among the plurality of switches in response to at least one phase of a load current flowing through an AC load connected to the plurality of AC terminals. Power conversion equipment.

2. the predetermined period is at least a part of a resonance half cycle of a resonance circuit including the resonance inductor and two or more resonance capacitors connected to the two or more switches, respectively; The power conversion device according to claim 1 .

3. The predetermined period is the entire period of the resonance half cycle. The power conversion device according to claim 2 .

4. In the second control operation of the control device, changing a start time point of a high level period of a control signal to one of the plurality of switches according to a total value of two or more phases of load current flowing through two or more AC terminals connected to the two or more switches among the plurality of AC terminals; The power conversion device according to any one of claims 1 to 3.

5. The control device includes: a third control action can be executed to make the dead time periods corresponding to two or more switching circuits connected to the two or more switches among the plurality of switching circuits longer than a predetermined dead time period by an additional time. The power conversion device according to any one of claims 1 to 3.

6. The control device a third control action can be executed to make the dead time periods corresponding to two or more switching circuits connected to the two or more switches among the plurality of switching circuits longer than a predetermined dead time period by an additional time. The power conversion device according to claim 4.