JP2011010380A - Dc power converter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a DC power converter for suppressing transient current generated in switching an operation mode, such as at the start of power conversion.SOLUTION: The DC power converter has an upper snubber circuit 51 and a lower snubber circuit 52, in which serial circuits of a snubber capacitor 6 and a snubber diode 7 are provided, in parallel with an upper main switch 31 and a lower main switch 32, respectively and an auxiliary switch 8 is connected between a connection point of the snubber capacitor 6 and the snubber diode 7 and a DC power supply 1 via a reactor 21. Of the upper snubber circuit and the lower snubber circuit, one of the main switches is switched periodically, from a state where an inter-terminal voltage of the snubber capacitor 6 of the one snubber circuit 5 is the output of the DC power converter and an inter-erminal voltage of the snubber capacitor 6 of the other snubber circuit 5 is zero; and in this case, the switching starts at an activation frequency lower than a predetermined switching frequency in steady state.

Description

  The present invention relates to a DC power converter that converts power between DC and DC.

  A rotating electrical machine that functions as an electric motor (motor) operates on the principle of generating a force (torque) by a magnetic field and an electric current. However, while the motor is rotating, a force acts in the magnetic field, so that the rotating electrical machine also functions as a generator, and so-called counter electromotive force is generated. The counter electromotive force is generated in a direction that hinders the flow of current that generates torque. Since the counter electromotive force increases as the motor rotation speed increases, when the rotation speed reaches a certain value, the current generated by the counter electromotive force reaches the drive current, and the motor cannot be controlled. As a countermeasure, “field weakening control” is often performed to weaken the field force that generates the magnetic field and suppress the generation of the back electromotive force. However, in order to weaken the field force, the strength of the magnetic field also decreases, and the maximum torque that can be obtained decreases. In Japanese Patent Laid-Open No. 10-66383 (Patent Document 1), the voltage of the battery is boosted by a booster circuit (converter) and the voltage for driving the motor is increased, so that the number of revolutions for shifting to field weakening control is increased. A technique for shifting to the rotational speed has been proposed.

  Generally, a converter includes a reactor and a switching element connected between the reactor and the negative power supply side. The switching element is intermittently driven by, for example, PWM (pulse width modulation) control. Switching in a situation where a voltage is applied to the switching element or a current flows is called hard switching. A hard switching converter has a simple structure, but produces a transient response during switching. This transient response causes large switching noise and switching loss. In general-purpose PWM control, the number of times of switching per unit time is large, and accordingly, switching noise and switching loss increase. For hard switching, there is a method called soft switching in which switching is performed in a state where the voltage applied to the switching element and the flowing current are substantially zero. The basic configuration of soft switching is to provide an LC resonance circuit outside the switching element. Resonance is generated in the LC resonance circuit by controlling the switching timing of the switching element, and the voltage and current at the time of switching are set to zero. Thereby, the switching noise and switching loss accompanying switching are greatly reduced compared with hard switching. International Publication No. WO2006 / 098376 (Patent Document 2) discloses a technique of such a soft switching type boost converter.

JP-A-10-66383 (3rd to 12th paragraphs, FIGS. 1 and 2 etc.) International Publication No. 2006/098376 (paragraphs 60-67, FIG. 7, 8 etc.)

By the way, in recent years, an automobile having a smaller environmental load than an automobile driven by an internal combustion engine that burns fossil fuel has been proposed. An electric vehicle driven by a motor that is a rotating electric machine and a hybrid vehicle driven by an internal combustion engine and a motor are examples. In an electric vehicle or a hybrid vehicle, the rotating electrical machine functions as a generator by braking or application of force to the rotating electrical machine by an internal combustion engine, and the generated power is regenerated to the battery. For this reason, the converter has a bidirectional configuration. That is, like the booster circuit (24) shown in FIG. 1 of Patent Document 1, two switching elements are connected in series between the positive and negative power supplies on the output side so that the reactor is connected to the middle point and can be complementarily switched. (Main switch). In order to perform soft switching of this converter, a snubber circuit including an auxiliary switch (S ₂ ), a snubber diode (D ₁ ), and a snubber capacitor (C ₁ ) as shown in FIG. The switching element is provided.

  When boosting is not required and the converter is operating in a direct connection mode that simply outputs battery voltage, the two switching elements as the main switch are not switched. The converter outputs the battery voltage via the flywheel diode of the upper switching element that functions as an output diode. At this time, since the switching element on the upper stage side is equivalent to the conductive state by the flywheel diode, the voltage between the terminals of the snubber capacitor of the snubber circuit of this switching element is also substantially zero. Here, when the converter starts boosting, the following problem occurs as an example. For example, when the lower switching element starts switching for boosting, a potential difference is generated between both ends of the upper snubber capacitor, and a large transient current is generated. Then, since this transient current flows into the lower switching element, switching noise and switching loss accompanying switching increase. In addition, in order to provide resistance to the transient current, it is necessary to increase the current capacity of the switching element and the converter, and the converter is enlarged. A similar problem may occur when the converter operating mode is switched between a power running operation and a regenerative operation.

  In view of the above problems, it is desired to provide a DC power conversion device in which a transient current generated at the time of switching an operation mode such as at the start of power conversion is suppressed.

The characteristic configuration of the DC power converter according to the present invention created in view of the above problems is as follows.
A DC power conversion device provided between an inverter interposed between a rotating electrical machine and a DC power source that supplies power to the rotating electrical machine and the DC power source,
A main reactor in which the first reactor and the second reactor are equivalently configured as one reactor by being connected in series, and one terminal is connected to the positive electrode of the DC power supply,
An upper main switch that connects the other terminal of the main reactor and the positive output terminal of the DC power converter;
A lower main switch that connects the other terminal of the main reactor and the negative output terminal of the DC power converter;
A series circuit of a snubber capacitor and a snubber diode is provided in parallel to the upper-stage main switch, and the upper-stage auxiliary switch is connected to a connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. An upper snubber circuit provided between the connection points of
A series circuit of a snubber capacitor and a snubber diode is provided in parallel to the lower-side main switch, and the lower-stage auxiliary switch is connected to the connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. A lower snubber circuit provided between the connection points of
The voltage between the terminals of the snubber capacitor of the snubber circuit with respect to one of the upper main switch and the lower main switch is an output voltage of the DC power converter, and the snubber capacitor terminal of the other snubber circuit. When the one main switch is periodically switched from a state in which the inter-voltage is zero, switching is started at a startup frequency lower than a predetermined switching frequency in the steady state.

  According to this configuration, since the switching of the main switch is started at the startup frequency lower than the predetermined switching frequency in the steady state, the on-duty is substantially reduced. Therefore, the snubber capacitor whose terminal voltage is zero before the start of switching is gradually charged. Since the charging current is gradually generated accordingly, the transient current can be suppressed.

  Further, in the DC power converter according to the present invention, the main switch is switched at the predetermined switching frequency during the ON state in one cycle when the one main switch is switched at the startup frequency. It is preferable to set the period in the ON state in one cycle.

  According to the above configuration, the ON period when periodically switched is a value corresponding to the predetermined switching frequency in the steady state regardless of the switching frequency. Therefore, it is not necessary to greatly change the circuit configuration or program of the conventional switching control. In addition, since only the switching period varies while maintaining the ON period, the ON duty can be substantially changed by a simple circuit configuration or program.

  The frequency at which the one main switch of the DC power converter according to the present invention is switched is such that the voltage across the snubber capacitor of the snubber circuit with respect to the other main switch is the output voltage of the DC power converter. In this case, it is preferable that the frequency is changed from the startup frequency to the predetermined switching frequency.

  If the voltage between the terminals of the snubber capacitor becomes the output voltage of the DC power converter, no further charging current is generated. Therefore, no transient current is generated, and the original function of the DC power converter can be exhibited at an early stage by changing the switching frequency to perform switching at a predetermined switching frequency.

  In addition, the frequency for switching the one main switch of the DC power converter according to the present invention is changed from the startup frequency to the predetermined switching frequency after a lapse of a predetermined time after switching of the one main switch is started. It is preferred that it be changed.

  Since the circuit constant of the DC power converter and the rated voltage of the DC power supply are known, it is possible to estimate the time until the snubber capacitor in question is sufficiently charged and the charging current is eliminated. Since it is not necessary to measure the voltage between the terminals of the snubber capacitor, the DC power converter can be configured easily.

  In the DC power converter according to the present invention, it is preferable that the frequency for switching the one main switch is changed to a high frequency stepwise or continuously from the startup frequency to the predetermined switching frequency. is there.

  If the start-up frequency is set in stages, such as a predetermined switching frequency being doubled or quadrupled, the circuit configuration and program can be simplified. Moreover, if the switching frequency is continuously changed from the startup frequency to a predetermined switching frequency, smooth transition is possible.

  In the DC power converter according to the present invention, it is preferable that the predetermined switching frequency in the steady state is a modulation frequency in pulse width modulation control.

  Generally, pulse width modulation control is performed in a DC power converter or the like. In the pulse width modulation control, the control is performed with a duty in a modulation period corresponding to a predetermined modulation frequency. As a result of the transition from the startup frequency to the predetermined switching frequency, when the modulation frequency is reached, it is possible to smoothly transition to a steady control state. Further, in the case where only the switching cycle is changed while maintaining the ON state period in the switching cycle, the ON state period can be calculated by the same circuit and program as in the steady control. . Therefore, the direct-current power converter can be configured easily and can be smoothly shifted to a steady control state.

Another characteristic configuration of the DC power converter according to the present invention is as follows.
A DC power conversion device provided between an inverter interposed between a rotating electrical machine and a DC power source that supplies power to the rotating electrical machine and the DC power source,
A main reactor in which the first reactor and the second reactor are equivalently configured as one reactor by being connected in series, and one terminal is connected to the positive electrode of the DC power supply,
An upper main switch that connects the other terminal of the main reactor and the positive output terminal of the DC power converter;
A lower main switch that connects the other terminal of the main reactor and the negative output terminal of the DC power converter;
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the upper main switch, and the upper auxiliary switch is connected to the connection point of the snubber capacitor and the snubber diode, the first reactor, and the second reactor. An upper snubber circuit provided between the connection points of
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the lower-stage main switch, and the lower-stage auxiliary switch includes a connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. A lower snubber circuit provided between the connection points of
Both the upper stage main switch and the lower stage main switch are in an off state, a flywheel diode connected in parallel to the upper stage main switch functions as an output diode, and the output voltage of the DC power supply is the DC power conversion From the direct connection mode that is output as the output voltage of the device, the lower-side main switch is periodically switched to boost the output voltage of the DC power supply and to the boost mode that is output as the output voltage of the DC power converter. In this case, switching is started at a startup frequency lower than a predetermined switching frequency in a steady state.

  According to this configuration, when the operation mode is switched from the direct connection mode to the boost mode, switching is started at a startup frequency lower than the predetermined switching frequency in the steady state. That is, since the on-duty is substantially reduced at the start of boosting, the snubber capacitor of the upper snubber circuit whose inter-terminal voltage was zero before the start of switching is gradually charged. Since the charging current is gradually generated accordingly, the transient current can be suppressed.

Further, another characteristic configuration of the DC power converter according to the present invention is as follows:
A DC power conversion device provided between an inverter interposed between a rotating electrical machine and a DC power source that supplies power to the rotating electrical machine and the DC power source,
A main reactor in which the first reactor and the second reactor are equivalently configured as one reactor by being connected in series, and one terminal is connected to the positive electrode of the DC power supply,
An upper main switch that connects the other terminal of the main reactor and the positive output terminal of the DC power converter;
A lower main switch that connects the other terminal of the main reactor and the negative output terminal of the DC power converter;
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the upper main switch, and the upper auxiliary switch is connected to the connection point of the snubber capacitor and the snubber diode, the first reactor, and the second reactor. An upper snubber circuit provided between the connection points of
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the lower-stage main switch, and the lower-stage auxiliary switch includes a connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. A lower snubber circuit provided between the connection points of
The operation mode is switched between a power running mode and a regeneration mode, which are two operation modes in which one of the upper main switch and the lower main switch is in an OFF state and the other is periodically switched. In this case, the switching of any one of the upper-side main switch and the lower-side main switch that is periodically switched from the OFF state is lower than the predetermined switching frequency in the steady state. It is at a point that starts at the time frequency.

  A transient current may also be generated when the operation mode is switched between the power running mode and the regenerative mode. However, when the switching is started at a startup frequency lower than a predetermined switching frequency, the transient current is reduced. It is suppressed.

  Further, the DC power conversion device according to the present invention is configured to switch the operation mode among the upper main switch and the lower main switch when the operation mode is switched between the power running mode and the regeneration mode. It is preferable that any one of the main switches that are turned off by the switching is turned off through switching at the startup frequency from the predetermined switching frequency.

  The switching of the operation mode between the power running mode and the regeneration mode can be predicted to some extent. Therefore, the main switch can be smoothly turned off by substantially reducing the on-duty by reducing the switching frequency of the main switch that is periodically switching to the start-up frequency. .

  Here, when the power in the power running mode and the power in the regenerative mode become equal to or less than a predetermined threshold power in the flow direction of each power, it becomes a transition period for switching the operation mode, and in this transition period, It is preferable that complementary control in which both the upper stage main switch and the lower stage main switch are periodically switched is executed.

  When the operation mode is switched, the power consumption and generated power are reduced and go through zero. Therefore, the transition period can be easily set based on the magnitude of power. When the sign of power, that is, powering and regeneration are not distinguished, the transition period can be set in each power flow direction by using the absolute value of power. Complementary control executed in the transition period is based on the direction of power flow even when the input voltage and output voltage are determined to be in a certain state and the load fluctuates or the power flow direction changes. The switch that flows naturally switches. That is, it is not necessary to determine the polarity of the current flowing through the main reactor, and the main switch only needs to maintain a certain state, so that control is very easy. In addition, stable control is possible when the operation mode is switched. When the operation mode is switched, the switching of any one of the main switches that are periodically switched from the OFF state is started at a startup frequency lower than a predetermined switching frequency in the steady state. The other main switch that is turned off is turned off through switching at a startup frequency from a predetermined switching frequency. Therefore, in the transition period, both the upper and lower main switches are switched at the startup frequency. Since both the main switches are switched at a frequency lower than a predetermined switching frequency, it is possible to smoothly switch between power running and regeneration without providing a dead time again. During the dead time, the regenerated electric power is stored in the output capacitor of the DC power converter, and the output capacitor, the inverter, or the like may be overvoltage. However, a period in which both the main switches are both turned off during the switching control appears, and it is not necessary to provide a dead time. Therefore, the switching control can be started at an early stage, and the problem of the overvoltage is suppressed. For the determination based on power, a power value (watts) or a current value (amperes) may be used.

Moreover, the DC power converter according to the present invention includes, as the rotating electric machine, a first rotating electric machine that mainly functions as a generator and a second rotating electric machine that mainly functions as an electric motor,
When the difference between the power generated by the first rotating electrical machine and the power consumed by the second rotating electrical machine is equal to or less than the predetermined threshold power, the transition period is preferably entered.

  When two rotating electric machines, that is, a rotating electric machine that functions as a generator and a rotating electric machine that functions as an electric motor, are connected to the DC power converter, the operation state of both rotating electric machines can be monitored as a whole by moving to a powering state. It can be determined whether it is in a regenerative state, and it can be predicted in which state it is changing. In other words, by comparing the generated power of the rotating electrical machine functioning as a generator with the power consumption of the rotating electrical machine functioning as an electric motor, it is possible to determine the current state and predict the changing state. When the difference between the generated power and the consumed power is less than or equal to the predetermined threshold power, there is a high possibility that the power running and regeneration are reversed. Therefore, it is preferable to transition to the transition period in preparation for this reversal. .

  In the transition period, it is preferable that the start-up frequency is set for each of the upper-side main switch and the lower-side main switch based on a power flow direction and a power magnitude.

  The startup frequency is not limited to one frequency, and may change stepwise or continuously. The possibility of switching between power running and regeneration can be estimated according to the direction of power flow and the magnitude of power. That is, if the magnitude of power is small, there is a high possibility that switching will occur, and if it is large, it can be estimated that switching will not be easy. For example, if switching is unlikely to occur, switching may be performed at a high frequency close to a predetermined switching frequency, and if switching is likely to occur, switching is performed at a frequency significantly lower than the predetermined switching frequency. Better. Based on the direction of power flow and the magnitude of power, such setting of the startup frequency can be performed accurately.

The block diagram which shows typically an example of the rotary electric machine control apparatus containing a direct-current power converter Explanatory drawing which shows operation | movement of DC power converter device at the time of direct connection mode Explanatory drawing which shows operation | movement of DC power converter device at the time of pressure | voltage rise mode starting Waveform diagram conceptually showing how the converter output voltage and reactor current are started Waveform diagram showing the switching sequence when boost mode is started Waveform diagram showing a control example that changes the frequency step by step Waveform diagram showing a control example that continuously changes the frequency Explanatory diagram schematically showing the state of boost control during power running Explanatory drawing schematically showing the state of complementary control during power running Explanatory drawing schematically showing the state of complementary control during regeneration Explanatory diagram schematically showing the state of step-down control during regeneration Timing chart schematically showing an example of the gate drive signals of the upper and lower IGBTs when the operation state transitions from power running to regeneration Block diagram schematically showing another configuration example of the rotating electrical machine control device

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Here, a converter (DC power conversion device) applied to a control device that controls a motor (rotating electric machine) serving as a drive source of a vehicle such as a hybrid vehicle or an electric vehicle will be described as an example. FIG. 1 is a block diagram schematically showing an example of the system configuration of such a control device for the motor 30. The motor 30 is electrically connected to the battery 1 and the converter 10 that boosts the output voltage of the battery 1 through the inverter 20, and generates a driving force when supplied with electric power. In the present embodiment, a system for controlling one rotating electrical machine will be described as an example in order to simplify the description. However, for example, as shown in FIG. 13, two inverters 20 (20A and 20B) are connected to one converter 10, and a rotating electrical machine 30 (30A and 30B) is connected to each inverter 20A and 20B. It may be a configuration.

  As shown in FIGS. 1 and 13, the inverter 20 includes a plurality of switching elements. It is preferable to apply an insulated gate bipolar transistor (IGBT) or a metal oxide semiconductor field effect transistor (MOSFET) to the switching element. In this example, an IGBT is used as the switching element. The inverter 20 is configured by a three-phase bridge circuit as is well known. Two IGBTs are connected in series between the DC positive side and the DC negative side of the inverter 20, and this series circuit is connected in parallel in three lines. That is, a bridge circuit in which a set of series circuits corresponds to each of the stator coils corresponding to the u-phase, v-phase, and w-phase of the motor 30 is configured. The collector of the IGBT on the upper stage side of each phase is connected to the DC plus side of the inverter 20, and the emitter is connected to the collector of the IGBT on the lower stage side of each phase. Moreover, the emitter of the IGBT on the lower side of each phase is connected to the DC negative side (for example, ground) of the inverter 20. The intermediate point of the series circuit of IGBTs of each phase that forms a pair, that is, the connection point of the IGBT, is connected to the stator coil of each phase of the motor 30.

  As shown in FIGS. 1 and 13, flywheel diodes (regenerative diodes) are connected in parallel to the IGBTs. The flywheel diode is connected in parallel to the IGBT such that the cathode terminal is connected to the collector terminal of the IGBT and the anode terminal is connected to the emitter terminal of the IGBT. The gate of each IGBT is connected to a TCU (trans-axle control unit) 50 which is one of ECUs (electronic control units) mounted on the vehicle via a driver circuit 40, and is individually controlled to be switched. . The TCU 50 is configured with a logic circuit such as a microcomputer as a core. As shown in FIGS. 1 and 13, the current flowing through each phase of the motor 30 is detected by a current sensor 35, and the rotation state such as the rotation speed is detected by a rotation sensor 37 such as a resolver. The TCU 50 performs feedback control based on the detection results of these sensors, and generates a drive signal for switching each IGBT. A drive signal input to the gate of an IGBT or MOSFET that switches a high voltage requires a voltage higher than the drive voltage of a general electronic circuit such as a microcomputer. Therefore, after being boosted through the driver circuit 40 , And input to the inverter 20.

  The converter 10 includes a main reactor 2, a pair of upper and lower switching elements (main switches) 3 (31 and 32) connected in series, a discharge resistor 11, an output capacitor 12, an input capacitor 13, and upper and lower corresponding to the switching element 3. It has a pair of snubber circuits 5 (51 and 52). As the switching element 3, it is preferable to apply an IGBT or a MOSFET as in the inverter 20. In this embodiment, the case where it comprises using IGBT is illustrated as an example. Both the output capacitor 12 and the input capacitor 13 are smoothing capacitors. The names of output and input correspond to the case where the converter 10 is powered.

  The main reactor 2 is configured by a first reactor 21 and a second reactor 22 that are equivalently configured as one reactor by being connected in series. One terminal of the main reactor 2 as one reactor is connected to the plus side (positive electrode output end) of the battery 1 (DC power supply). The emitter of the upper IGBT 31 (upper main switch) is connected to the collector of the lower IGBT 32 (lower main switch) and is connected to the positive side of the battery 1 via the reactor 2. The collector of the upper IGBT 31 is connected to the DC positive side of the inverter 20. The emitter of the lower IGBT 32 is connected to the minus side (negative output terminal) of the battery 1 and the DC minus side of the inverter 20.

  The gates of the upper IGBT 31 and the lower IGBT 32 of the converter 10 are connected to the TCU 50 via the driver circuit 40. The IGBTs 31 and 32 are controlled by the TCU 50, boost the output of the battery 1 as necessary, and supply it to the inverter 20. Further, when the motor 30 functions as a generator, the generated power supplied via the inverter 20 is stepped down as needed to regenerate the battery 1. The TCU 50 controls the IGBT 31 and the IGBT 32 based on a command value set according to the target torque of the motor 30.

  Specifically, the TCU 50 boosts the voltage of the battery 1 and outputs it to the inverter 20 by switching the IGBT 31 on the upper stage OFF and switching the IGBT 32 on the lower stage at a predetermined frequency by, for example, PWM control ( Boost mode). When boosting is not necessary, both IGBTs 3 are controlled to be in an OFF state, whereby the output of the battery 1 is output to the inverter 20 via the flywheel diode 41 of the upper IGBT 31 (direct connection mode). Here, the boost mode and the direct connection mode are collectively referred to as a power running mode.

  When the motor 30 performs a regenerative operation, the TCU 50 switches the lower IGBT 32 to the OFF state and switches the upper IGBT 31 at a predetermined frequency, thereby reducing the generated power and regenerating it to the battery 1 (step-down regeneration). mode). When step-down is not required, the lower IGBT 32 is turned off and the upper IGBT 32 is controlled to be turned on to regenerate power (direct regenerative mode). Here, the step-down regeneration mode and the direct connection regeneration mode are collectively referred to as a regeneration mode.

  As shown in FIGS. 1 and 13, the converter 10 includes a snubber circuit 5 in order to perform soft switching of the upper IGBT 31 and the lower IGBT 32, respectively. The snubber circuit 5 for the upper IGBT 31 is an upper snubber circuit 51, and the snubber circuit 5 for the lower IGBT 32 is a lower snubber circuit 52. The upper-side snubber circuit 51 and the lower-side snubber circuit 52 are only different in polarity and are equivalent in configuration and function. Therefore, both are hereinafter simply referred to as the snubber circuit 5 unless otherwise distinguished. explain. In addition, except for the case where it is particularly necessary to distinguish, one of the snubber circuits 5 will be described as a representative.

  The snubber circuit 5 includes a snubber capacitor 6 (61, 62), a snubber diode 7 (71, 72), a switching element (auxiliary switch) 8 (81, 82), and a rectifier diode 9 (91, 92). Configured. The switching element 8 is preferably an IGBT or a MOSFET. In this embodiment, the case where it comprises using IGBT is illustrated as an example. As shown in FIG. 1, the series circuit of the snubber capacitor 6 and the snubber diode 7 is provided in parallel with the IGBT 3 as the main switch. The IGBT 8 as an auxiliary switch is provided between a connection point between the snubber capacitor 6 and the snubber diode 7 and a connection point between the first reactor 21 and the second reactor 22. Flywheel diodes 8a and 8b are connected in parallel to IGBTs 81 and 82, respectively. An upper-stage snubber circuit 51 and a lower-stage snubber circuit 52 are provided to soft-switch each of the upper-stage IGBT 31 and the lower-stage IGBT 32.

  Since soft switching is a well-known technique as described in Patent Document 2 and the like, only a minimal explanation is given here. For example, when the converter 10 operates in the boost mode, as described above, the upper IGBT 31 is maintained in the OFF state, and the lower IGBT 32 is switched by PWM control. The flywheel diode 41 of the upper IGBT 31 functions as an output diode of the converter 10. The snubber capacitor 62 of the lower snubber circuit 52 is charged, and the voltage between its terminals is substantially the output voltage of the battery 1.

  Before the IGBT 32 that is the main switch is turned on, the IGBT 82 that is the auxiliary switch of the lower snubber circuit 52 is turned on. The charge accumulated in the flywheel diode 41 as the output diode flows to the first reactor 21 via the IGBT 82. When the charge of the flywheel diode 41 disappears, the flywheel diode 41 reversely recovers and becomes non-conductive. Here, the snubber capacitor 62 resonates and discharges the accumulated charge. When all the charges are discharged, the voltage between the terminals of the snubber capacitor 62 becomes zero. At this timing, the IGBT 32 which is the main switch is turned on and turned on at zero voltage. Since the regenerative current due to resonance flows in a direction that cancels the current of the IGBT 32, the IGBT 32 is turned on from zero current. Thus, soft switching with zero voltage and zero current is realized.

  The main reactor 2 is equivalently configured as one reactor by connecting the first reactor 21 and the second reactor 22 in series. The charge accumulated in the output diode (flywheel diode 41) is passed through the first reactor 21 on the battery 1 side of the two main reactors in the direction opposite to the main current of the main reactor 2, and the battery Regenerated to 1. Thereby, the overvoltage due to the reverse recovery of the output diode (flywheel diode 41) is suppressed, and the heat loss generated in the main reactor 2 is also reduced. If the reactor corresponding to the first reactor 21 is large, the charge stored in the output diode cannot be eliminated, and a large surge voltage is generated during reverse recovery. However, by configuring the main reactor 2 by connecting the first reactor 21 and the second reactor 22 in series, it is possible to appropriately set the value of the first reactor 21 while maintaining the size of the main reactor 2. It becomes.

  Here, consider a case where the boost mode is started from the direct connection mode in which the converter 10 outputs the output of the battery 1 to the inverter 20 as it is. During operation in the direct connection mode, as shown in FIG. 2, the upper flywheel diode 41 functions as an output diode, and outputs the output voltage E of the battery 1 to the inverter 20 almost as it is. At this time, the IGBT 31 and IGBT 32 that are the main switches of the converter 10 and the IGBT 81 and IGBT 82 that are the auxiliary switches are all in the OFF state as shown by the broken lines. In the figure, G31, G32, G81, and G82 indicate gate drive signals of the IGBT 31, IGBT 32, IGBT 81, and IGBT 82, respectively.

  In the lower snubber circuit 52, the snubber capacitor 62 is charged via the snubber diode 72, and the voltage V2 between the terminals of the snubber capacitor 62 becomes substantially the output voltage E of the battery 1. Further, since it is in the direct connection mode, the voltage V3 between the terminals of the output capacitor 12 is also substantially the output voltage E of the battery 1. In the upper-side snubber circuit 51, since the snubber diode 71 and the flywheel diode 41 are both connected in the forward direction, the snubber capacitor 61 is not charged, and the inter-terminal voltage V1 is substantially zero.

  When the boost mode is started from the direct connection mode, there are the following two cases. The first case is an idling state in which the vehicle is stopped. At this time, since the inverter 20 and the motor 30 are almost in a no-load state, the boost mode is started from the direct connection mode while the output power of the converter 10 remains at a very low load. The second case is a case where the vehicle has already started moving in the so-called EV travel mode by the driving force of the motor 30, and the boost mode is started from the direct connection mode for acceleration.

  In common with these two cases, when complementary control involving soft switching operation with zero voltage and zero current based on the principle as in Patent Document 2 is performed, a large current flows through the IGBT 32. That is, each time the IGBT 81 that is the upper auxiliary switch is switched by complementary control, the charging voltage of the upper snubber capacitor 61 is discharged each time. As a result, a large charging current flows to the IGBT 32 via the snubber capacitor 61 and the snubber diode 71 every switching (see FIG. 3). For this reason, the first problem that complementary control accompanied by soft switching operation by zero voltage zero current cannot be performed at the start.

  Next, when the vehicle is stopped in the first case, the inverter 20 and the motor 30 are in a no-load state, so the second problem occurs. FIG. 4B conceptually shows waveform changes at the start of the output voltage Vout and the current I22 of the reactor 22 when the output voltage is correctly controlled. The output voltage Vout of the converter 10 is controlled from the input voltage E to the set value V0 and rises flexibly. It is activated by so-called soft start. However, when the inverter 20 and the motor 30 are in a no-load state, the gate drive signal G32 of the IGBT 32, which is the main switch, is fixed in a state where the minimum on-width (minimum on-duty) is set. For this reason, as shown in FIG. 4A, the output voltage Vout is not properly controlled, and the output voltage Vout rises to an overvoltage Vp. As a result, an excessive charging current of the output capacitor 12 flows to the IGBT 32.

  In the second case, when the vehicle has already started moving in the EV travel mode, the inverter 20 and the motor 30 are in a constant load state. Therefore, when the output voltage is correctly controlled and the start from the direct connection to the boost mode is possible, the output voltage is not correctly controlled, and the output voltage is overvoltage as in the first case (second problem). May be Vp.

  As described above, when the boost mode is started from the direct connection mode, there are three problems related to soft start. That is, the first problem common to the no-load state and the load state in which complementary control cannot be performed at the start, the second problem when starting with no load, and the third problem when starting with a constant load There are three issues.

  Therefore, when starting the boost mode from the direct connection mode, the TCU 50 switches the IGBT 32 at a startup frequency lower than a predetermined switching frequency (a modulation frequency of PWM control) as shown in FIG. As described above, the IGBT 82 serving as the auxiliary switch is turned on prior to the IGBT 32 serving as the main switch. Therefore, the gate drive signal G82 becomes effective before the gate drive signal G32. A symbol T38 in the figure indicates the preceding time difference. Based on the boost command value, the TCU 50 calculates a pulse width T3 for boosting the output of the battery 1 by PWM control at the modulation frequency F30, which is a predetermined switching frequency. The switching cycle T30 during PWM control is 1 / F30, which is the inverse of the modulation frequency.

  The TCU 50 generates the gate drive signals G32 and G82 at the start-up frequency (F31 and F32) lower than the modulation frequency F30 while maintaining the pulse width T3 of the effective period when starting the boost mode operation. Here, the pulse width T3 is a narrowed pulse width corresponding to a case where the step-up rate is low as a pulse width in PWM control during steady operation in the step-up mode. Since the gate drive signals G32 and G82 are synchronized, the gate drive signal G32 will be described below as a representative. FIG. 5 illustrates a case where the frequency is switched in a stepwise manner from the startup frequency to the modulation frequency F30 which is a predetermined switching frequency. Here, the case where there are two stages of the first startup frequency F31 and the second startup frequency F32 is illustrated as the startup frequency. In FIG. 5, the second startup frequency F32 is a half of the predetermined switching frequency F30, and the first startup frequency F31 is a quarter of the predetermined switching frequency F30. The start-up frequency is not limited to two stages, and may be one stage or three or more stages.

  As shown in FIG. 5, the on-duty when switching is performed at the modulation frequency F30 is T3 / T30. The on-duty when switching is performed at the first startup frequency F31 is T3 / T31. Since the first startup frequency F31 is ¼ of the modulation frequency F30, T31 is a period four times T30. Therefore, the on-duty when switched at the first startup frequency F31 is substantially ¼ compared to when switched at the modulation frequency F30. Further, the on-duty when switching is performed at the second startup frequency F32 is T3 / T32. Since the second startup frequency F32 is ½ of the modulation frequency F30, T32 is twice as long as T30. Therefore, the on-duty when switching at the second startup frequency F32 is substantially ½ compared to when switching at the modulation frequency F30.

  Whether switching is performed at the modulation frequency F30 or switching at the startup frequencies F31 and F32, the time during which the IGBT 32 is turned on in one switching does not change at T3. However, the on-duty is substantially reduced because the ratio of the on state in one switching cycle decreases. As a result, the charging current (transient current) of the snubber capacitor 61 flowing into the IGBT 32 per unit time is reduced. Since the PWM control pulse width T3 calculated based on the boost command value and the modulation frequency does not fluctuate, problems in starting the boost mode can be suppressed without greatly changing the configuration or program of the converter 10 so far. It becomes possible to do. In other words, since the switching frequency is changed in a state where the pulse width T3 is maintained, it can be said that pulse frequency modulation is performed during control using the startup frequency.

  The same applies to the gate drive signal G82 for switching the IGBT 82 as the auxiliary switch. In FIG. 5, T80 matches T30, T81 matches T31, and T82 matches T32. The pulse width T8 of the effective period of the gate drive signal G82 is set according to the circuit constants of the converter 10 including the snubber circuit 52 and the ON time T3 of the IGBT 32. Further, as described above, the IGBT 82 of the snubber circuit 52 is turned on before the IGBT 32 is turned on, but the time difference T38 is a predetermined value determined by the operating state. In the present embodiment, the time difference T38 is the same value in the course of switching the switching frequency from the starting frequency to the modulation frequency. That is, the gate drive signal G32 and the gate drive signal G82 are output while maintaining the phase relationship in the course of switching the switching frequency.

  When the TCU 50 switches the switching frequency from the starting frequency to the modulation frequency, there are a method of switching to a higher frequency step by step as shown in FIG. 5 and a method of switching to a higher frequency continuously. First, a control example when the TCU 50 switches the switching frequency in stages will be described with reference to FIG.

  As described above, the TCU 50 is configured by a microcomputer, an application specific integrated circuit (ASIC), or the like. These electronic circuits generally operate using a system clock as a minimum control unit. Here, as shown in FIG. 6, a clock cp having a frequency of 100 MHz is used as a system clock. The TCU 50 includes a counter or a timer that operates according to the clock cp. For example, counting is started on the basis of the timing of the reset RST.

  Here, it is assumed that the modulation frequency F30 which is a predetermined switching frequency is 200 kHz. A signal cp500 corresponding to 500 clocks cp is generated, and the signal cp500 becomes a reference for PWM control. The first startup frequency F31, which is a quarter of the modulation frequency F30, is 50 kHz. Therefore, a signal cp2000 corresponding to 2000 clocks cp is generated. Similarly, since the second startup frequency F32 is 100 kHz, a signal cp1000 corresponding to 1000 clocks cp is generated.

  These cp500, cp1000, and cp2000 signals indicate the timing for temporarily clearing the counter in order to periodically generate the respective signals. Therefore, when switching is performed at the modulation frequency F30, the counter value is 1 to 500 (0 to 499). When switching is performed at the first startup frequency, the counter is 1 to 2000 (0 to 1999). When the switching is performed at the value and the second startup frequency, the gate drive signals G32 and G82 are generated using the counter value of 1 to 1000 (0 to 999). FIG. 6 exemplifies a case where the gate drive signals G32 and G82 are generated with reference to the signals cp475, cp1975, and cp975 generated at the counter values 475, 1975, and 975, respectively, before being temporarily cleared. .

  When the boost mode is started from the direct connection mode, the boost mode signal st is enabled. This signal st becomes effective in synchronization with the signal cp1975 having the slowest frequency. When the boost mode is terminated and the mode is switched to the direct connection mode, or when the mode is switched between power running and regeneration, it is preferable that the signal cp 1975 changes in synchronization with the reference signal having the slowest frequency at the time of switching. Many other methods can be adopted as a method of synchronizing the boost mode signal st and the gate drive signal (G32, G82, etc.). A person skilled in the art will be able to realize the present invention by using a logic circuit or a program different from the present embodiment. Here, it is important that the status signal such as the boost mode signal st is synchronized with the gate drive signal. If both signals are generated asynchronously and then output via a combinational circuit, the pulse width of the gate drive signal is disturbed, or only one of the main switch and auxiliary switch is output. May occur. By generating and outputting both signals in synchronization, stable switching can be achieved both at startup and at stop.

  In the effective period of the boost mode signal st, first, gate drive signals G82 and G32 are generated with reference to the signal cp1975 (time t10). The gate drive signal G82 of the IGBT 82 which is an auxiliary switch rises after 20 cycles of the clock cp with reference to the signal cp1975 and becomes effective over 100 cycles of the clock cp. The gate drive signal G32 of the IGBT 32, which is the main switch, rises 50 cycles after the clock cp with reference to the signal cp1975 and becomes effective over 150 cycles of the clock cp. In this way, first, switching is started at the first startup frequency F31.

  If the predetermined condition is satisfied after the start of the start, that is, after the start of the switching at the first start-up frequency F31, the switching frequency is switched from the first start-up frequency F31 to the second start-up frequency F32 (time t20). ). The predetermined condition is that a predetermined time has elapsed after the start of starting, or that the voltage between the terminals of the snubber capacitor 61 of the upper-stage snubber circuit 51 has increased to a predetermined voltage. The predetermined time and the predetermined voltage are set based on the rated voltage of the battery 1, the capacity of the snubber capacitor 61, and the like. When switching from the first startup frequency F31 to the second startup frequency F32, it is preferable to synchronize with the signal cp975. In this example, the second start-up frequency F32 is twice the first start-up frequency F31, and the count is started based on the timing of the reset RST, so that the second start-up frequency F32 is switched in synchronization with the signal cp1975. In this case, switching is performed in synchronization with the signal cp975.

  Here, switching of the switching frequency corresponds to changing a signal serving as a reference for generating the gate drive signals G32 and G82. When switching at the first startup frequency F31, the signal cp1975 was used as a reference. After switching the switching frequency, the gate drive signals G32 and G82 are generated using the signal cp975 as a reference. However, as shown in the figure, the generation timing of the gate drive signals G32 and G82 with respect to the reference signal is not changed. That is, as before the change of the reference signal, the gate drive signal G82 of the IGBT 82 which is the auxiliary switch rises after 20 cycles of the clock cp with the signal cp975 as a reference, and becomes effective over 100 cycles of the clock cp. The gate drive signal G32 of the IGBT 32, which is the main switch, rises after 50 cycles of the clock cp with reference to the signal cp975 and becomes effective over 150 cycles of the clock cp.

  When a predetermined condition is satisfied after the start of switching at the second start-up frequency F32 or after the start of start-up, the switching frequency is switched from the second start-up frequency F32 to the modulation frequency F30 which is a steady-state switching frequency ( Time t30). As described above, the predetermined condition is that a predetermined time has elapsed since the start of starting, or that the voltage between the terminals of the snubber capacitor 61 of the upper snubber circuit 51 has increased to a predetermined voltage. The predetermined time is set based on the rated voltage of the battery 1, the capacity of the snubber capacitor 61, and the like. It is preferable that the predetermined voltage is substantially equivalent to the output voltage of the converter 10, which is a voltage in a state where the snubber capacitor 61 of the upper snubber circuit 51 is sufficiently charged. When switching the switching frequency, it is preferable to synchronize with the signal cp475 as described above. In this example, the modulation frequency F30 is twice the frequency of the second startup frequency F32, and the count is started on the basis of the timing of the reset RST, so that if the signal is switched in synchronization with the signal cp975, Switching is also performed in synchronization with cp475.

  Similar to switching from the first startup frequency F31 to the second startup frequency F32, switching of the switching frequency corresponds to changing a signal that is a reference for generating the gate drive signals G32 and G82. After the change, gate drive signals G32 and G82 are generated based on the signal cp475 instead of the signal cp975. In this case, the generation timing for the reference signal is not changed. That is, the gate drive signal G82 of the IGBT 82 that is the auxiliary switch rises after 20 cycles of the clock cp with reference to the signal cp475 and becomes effective over 100 cycles of the clock cp. The gate drive signal G32 of the IGBT 32, which is the main switch, rises after 50 cycles of the clock cp with reference to the signal cp475 and becomes effective over 150 cycles of the clock cp.

  As described above, the effective pulse widths T3 and T8 of the gate drive signals G32 and G82, the relationship between the two signals (time difference T38), and the like do not depend on the switching frequency. A value set based on the command value for boosting and the modulation frequency which is the switching frequency in the steady state is used. Therefore, it is possible to satisfactorily suppress problems at the time of starting the boost mode without greatly changing the configuration or program of the TCU 50.

  In addition, in the above, the case where it switched to the high frequency in steps from the starting frequency to the predetermined switching frequency was illustrated. However, the present invention is not limited to this, and the frequency may be continuously switched from a startup frequency to a predetermined switching frequency. Hereinafter, a control example when the TCU 50 continuously switches the switching frequency will be described with reference to FIG.

  In the example described above based on FIG. 6, the signals cp500, cp1000, and cp2000 for temporarily clearing the counter are generated in order to switch the frequency step by step. In this example, in order to continuously switch to a higher frequency, a signal cpx in which the count value of the clear signal varies is used. The count value of the signal cpx is set by another down counter x. Here, in order to clearly show the change, a down counter x in which the count value is decreased by 250 and fixed at the count value 500 is illustrated. The signal cpx generated according to the value specified by the down counter x is used as a reference signal in the same manner as the signals cp975 and cp1000 described above, so that the frequency is changed continuously from the startup frequency to the modulation frequency. Is possible. In FIG. 7, cp2250, cp2000, cp1750, cp1500, cp1250, cp1000, cp750, and cp500 are generated as the signal cpx corresponding to the down counter x that is decreased by 250 and fixed at the count value 500. An example is shown. Of course, the present invention is not limited to the example shown in FIG. 7, and it will be easily understood that the continuity can be further increased by appropriately setting the reduction width of the down counter x. For example, in the example shown in FIG. 7, an example in which the period changes by a constant value ΔT (= 250) is shown. However, the present invention is not limited to this, and the frequency may be changed by a constant value Δf.

  Incidentally, the motor 30 as a rotating electrical machine also functions as a generator, and the generated power is regenerated to the battery 1. For example, in the system shown in FIG. 1, when the vehicle brakes, the motor 30 functions as a generator, and the generated power is regenerated to the battery 1. In the system having a plurality of motors 30 as shown in FIG. 13, one motor 30A (first rotating electrical machine) is mechanically rotated by an internal combustion engine or the like to function as a generator, and the generated power and the battery 1 May be used to drive the other motor 30B (second rotating electrical machine). When the power consumption of the motor 30B is lower than the power generated by the motor 30A, the surplus power is regenerated to the battery 1.

  Not only when the converter 10 starts boosting, the converter 10 switches between a power running mode in which power is supplied from the battery 1 to the inverter 20 and a regeneration mode in which power is regenerated from the inverter 20 to the battery 1. Even when switching, a large transient current may occur in the snubber capacitor 6. Therefore, when switching between the power running mode and the regenerative mode, it is preferable to perform the pulse frequency modulation by changing the switching frequency using the startup frequency.

  For example, when switching between the power running mode and the regeneration mode, one of the IGBT 31 that is the upper main switch and the IGBT 32 that is the lower main switch is turned off according to the power running and regeneration mode. Then, the switching control of the other IGBT is started from the start-up frequency in the same manner as the start-up of the boost mode described above.

  Further, when the IGBT on the turn-off side at the time of switching is turned off, the modulation frequency may be changed to the startup frequency in a sequence reverse to the startup time, and the IGBT may be shifted to a continuous OFF state. For example, as shown in FIG. 13, when a plurality of rotating electrical machines are provided, one functioning as a generator and the other functioning as an electric motor, the difference between the electric power generated by the electric generator and the electric power consumed by the electric motor is equal to or less than a predetermined threshold power. In this case, it is preferable to change from a predetermined switching frequency to a startup frequency in a sequence reverse to that at the time of starting. As this “predetermined threshold value”, it is preferable to use a rated current value Ith described later. As shown in FIG. 13, the current flowing through each phase of the motor 30 is detected by a current sensor 35, and the rotation state such as the rotation speed is detected by a rotation sensor 37 such as a resolver. The TCU 50 calculates generated power and power consumption based on the detection results of these sensors. When the switching frequency is continuously changed, the down counter x shown in FIG. 7 may be an up counter.

  In general, when switching between the power running mode and the regenerative mode, both IGBTs 3 are turned off so that the IGBTs 31 and IGBTs 32 that are complementarily connected in series as main switches are simultaneously turned on and the output of the converter 10 is not short-circuited. A dead time to be in a state is provided. However, the regenerated power is stored in the output capacitor 12 during the dead time, and the output capacitor 12 and the inverter 20 may be overvoltage. In the pulse frequency modulation by the start-up frequency, the on-duty is substantially reduced. Therefore, it is possible to provide a period during which both IGBTs 3 are in the OFF state during the switching control without setting the dead time again. . The regenerative current stays in the output capacitor 12 during the dead time, and a decrease in which the potential of the output capacitor 12 rapidly rises can be suppressed, and switching control can be started early.

  Note that the control method may be switched from PWM control to pulse frequency modulation during operation in the boost mode or the step-down regeneration mode, for example, when switching between the power running mode and the regeneration mode or when starting the boost mode from the direct connection mode. Good. When the modulation frequency is constant and the on-duty in PWM control is reduced, the on-time of the gate drive signal of the IGBT 3 is shortened. Accordingly, it is necessary to adjust the ON time of the gate drive signal of the IGBT 8 of the snubber circuit 5. However, if the ON time of the IGBT 8 of the snubber circuit becomes too short, there is a possibility that the effect of soft switching becomes limited. Therefore, when the on-duty falls below the predetermined lower limit duty, the on-time T3 (T8) of the gate drive signals G32 (G31) and G82 (G81) is fixed and the modulation frequency is changed to the startup frequency. However, it is preferable to substantially reduce the on-duty.

  Further, when control called complementary control is performed on IGBT 41 and IGBT 42 that are the main switches, control when the operation mode of converter 10 changes can be facilitated. Hereinafter, an example of performing complementary control when switching from the power running mode to the regeneration mode will be described. 8 to 11 will be described focusing on a single pulse of the gate drive signal or a period corresponding to 2 to 3 pulses, and therefore, modulation schemes such as PWM and pulse frequency modulation are not specifically mentioned. . Referring to the supplementary explanation using FIG. 12, it is possible to obtain the same effect as that provided the dead time without substantially setting the dead time by applying the pulse frequency modulation in the complementary control. Will be easily understood. In addition, of course, pulse frequency modulation can be applied at the time of transition from power running control to complementary control and vice versa, and at the time of transition from complementary control to regenerative control and vice versa. It is.

  FIG. 8A shows the operating state in the boost mode (powering mode), and the hollow arrows indicate the direction of power flow. As described above, the gate drive signal G32 of the IGBT 32 and the gate drive signal G82 of the IGBT 82 are given with a predetermined phase difference T38, and the converter 10 operates with high efficiency by soft switching. A waveform V <b> 22 shown in FIG. 8B shows the voltage between the terminals of the second reactor 22, and a waveform I <b> 22 shows the current flowing through the second reactor 22. The symbol Ith is a positive rated value (positive predetermined value) of the current I22. When current I22 is larger than positive rated value Ith on the positive side, converter 10 operates in a boost mode with soft switching.

  FIG. 9A shows the operating state when the power consumption of powering is decreasing and shifting to the regeneration side. The hollow arrow indicates the direction of power flow, and at this point, it is still in a power running state. However, as shown in FIG. 9B, the current I22 is lower than the positive rated value Ith, and the converter 10 operates by complementary control. As shown in FIG. 9B, the complementary control is a control for switching the IGBT 32 as the upper main switch and the IGBT 31 as the lower main switch alternately, that is, in a complementary manner. At this time, a dead time dt is provided so that the IGBT 32 and the IGBT 31 are not simultaneously turned on. Here, if the pulse frequency modulation is applied as described above, the same effect as that provided by the dead time dt can be obtained without substantially setting the dead time dt. As shown in FIG. 9B, since the current I22 flowing through the reactor 22 is small, the cut-off currents of the IGBT 31 and the IGBT 32 are also small. Therefore, even if hard switching is performed, the loss is small, and the snubber circuit 5 is stopped during this complementary control.

  In complementary control, a switch that flows naturally according to the power flow direction even if the load fluctuates or the power flow direction changes while the input voltage and output voltage are determined to be in a certain state. Switches. That is, it is not necessary to determine the polarity of the current I22 flowing through the reactor 22, and the gate drive signals G31 and G32 of the IGBT 31 and IGBT 32 only have to maintain the state of FIG. is there.

  FIG. 10 shows a state where the polarity of the current I22 of the reactor 22 is reversed. The hollow arrows indicate the direction of power flow, and the state is from power running to regeneration. However, as shown in FIG. 10B, the current I22 is lower than the negative rated value Ith (higher than the rated value −Ith). Therefore, the converter 10 continues to operate under complementary control. In the state of FIGS. 9 and 10, the power consumption of powering is reduced and the state shifts to regeneration, and the state where the regenerative power is still low or the state where the regenerative power is low shifts to powering and the power consumption of powering is still small. For example, power running and regeneration are easily interchanged. At this time, since complementary control is executed, the same control can be continued regardless of the state of power running and regeneration.

  FIG. 11 shows the operation state in the step-down mode (regeneration mode), and the hollow arrow indicates the direction of power flow. Since the current I22 is higher than the negative rated value Ith (lower than the rated value −Ith), the lower IGBT 32 is maintained in the OFF state, not the complementary control, and the upper IGBT 31 is soft-switched. Switched. Similar to the lower-stage control during power running, the gate drive signal G31 of the IGBT 31 and the gate drive signal G81 of the IGBT 81 are given with a predetermined phase difference T38. The converter 10 operates with high efficiency in a step-down mode with soft switching.

  8 to 11, in order to facilitate understanding, the case where there is one rotating electric machine as in FIG. 1 has been described as an example. However, for example, as shown in FIG. 13, even when a plurality of rotating electrical machines are provided, one functioning as a generator and the other functioning as an electric motor, when the operating state of the converter 10 is switched between power running and regeneration, it is complementary. It is preferable that the control is executed. In the case of the configuration shown in FIG. 13, the power flow direction of converter 10 is determined by the difference between the power generated by the generator and the power consumed by the motor. Therefore, it is preferable to execute complementary control when the difference in power consumption is equal to or less than a predetermined threshold value. The predetermined threshold value may be a power value (watt) or a current value (ampere), and for example, a positive / negative rated value Ith can naturally be used.

  Hereinafter, a supplementary description will be given with reference to FIG. 12, which is a timing chart schematically showing an example of the gate drive signals G32 and G31 of the upper IGBT 32 and the lower IGBT 31 when the operation state transitions from power running to regeneration. To do. The “predetermined threshold value” described above is not limited to one value, and may be set in a plurality of stages, for example, two or more stages. In this example, a case where two stages of positive and negative are set is shown. During powering before entering the transition period, lower-stage control is performed in which only the lower-stage IGBT 32 is switching-controlled. When the power flow direction is the power running direction shown in FIGS. 8 and 9, and the current I22 of the reactor 22 is larger than the first threshold value that is larger than the rated value Ith, the gate drive signal G32 has a predetermined switching frequency. It is output at a certain PWM modulation frequency.

  When the current I22 of the reactor 22 decreases and becomes equal to or less than the first threshold value, a transition period in which the operation mode may be switched from power running to regeneration. The gate drive signal G32 is output at the second startup frequency instead of the PWM modulation frequency that is a predetermined switching frequency. Further, when the current I22 of the reactor 22 decreases and becomes equal to or lower than the positive rated value Ith (second threshold value), the gate drive signal G32 is output at the first starting frequency that is a lower frequency. Furthermore, the gate drive signal G31 of the upper-stage IGBT 31 also starts to be output at the first activation frequency. At this time, both the upper and lower IGBTs 31 and 32 perform complementary control in which switching control is performed. As shown in FIG. 12, the pulse width of the high period of the gate drive signals G31 and G32 is constant, but the off period is substantially increased by increasing the output period. As a result, the dead time dt is automatically set without having to set it again.

  As described above with reference to FIGS. 9 and 10, when the absolute value of the current I22 of the reactor 22 is equal to or less than the absolute value of the positive / negative rated value Ith (second threshold value), complementary control is executed. During this time, even if the power flow direction causes hunting between power running and regeneration, the power running operation and the regeneration operation are well maintained. When the power flow direction is switched and the current I22 of the reactor 22 becomes a regenerative current and exceeds the negative rated value Ith (second threshold) (below −Ith), the switching of the lower IGBT 32 is stopped. . The gate drive signal G31 of the upper IGBT 31 is output at a second activation frequency that is higher than the first activation frequency. That is, transition from complementary control to upper stage control (regeneration). When the regenerative current further increases and the current I22 of the reactor 22 exceeds the first threshold value in the negative direction, the transition period is completed. The gate drive signal G31 is output at a PWM modulation frequency that is a predetermined switching frequency.

  Thus, in the transition period, it is preferable that the startup frequency is set for each of the upper-side main switch and the lower-side main switch based on the power flow direction and the power magnitude. The direction of power flow and the magnitude of power may be a power value (watts) or a current value (amperes) as in the above example. In addition, FIG. 12 illustrates a case where the startup frequency changes stepwise, but a mode in which it continuously changes may be used as a matter of course.

  The DC power conversion device of the present invention can be used to convert DC power between a DC power source and an inverter in a rotating electrical machine control device such as an electric vehicle or a hybrid vehicle.

1: Battery (DC power supply)
2: Main reactor 3: Main switch 4: Flywheel diode 5: Snubber circuit 6: Snubber capacitor 7: Snubber diode 8: Auxiliary switch 21: First reactor 22: Second reactor 31: Upper IGBT (upper main switch)
32: Lower IGBT (upper main switch)
41: Upper stage flywheel diode 42: Lower stage side flywheel diode 51: Upper stage side snubber circuit 52: Lower stage side snubber circuit 61: Upper stage side snubber capacitor 62: Lower stage side snubber capacitor 71: Upper stage side snubber diode 72: Lower stage side snubber diode 81: Upper stage side auxiliary switch 82: Lower stage side auxiliary switch 10: Converter (DC power converter)
20: Inverter 30: Motor (rotary electric machine)
30A: Motor (first rotating electrical machine)
30B: Motor (second rotating electrical machine)

Claims (12)

A DC power conversion device provided between an inverter interposed between a rotating electrical machine and a DC power source that supplies power to the rotating electrical machine and the DC power source,
A main reactor in which the first reactor and the second reactor are equivalently configured as one reactor by being connected in series, and one terminal is connected to the positive electrode of the DC power supply,
An upper main switch that connects the other terminal of the main reactor and the positive output terminal of the DC power converter;
A lower main switch that connects the other terminal of the main reactor and the negative output terminal of the DC power converter;
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the upper main switch, and the upper auxiliary switch is connected to the connection point of the snubber capacitor and the snubber diode, the first reactor, and the second reactor. An upper snubber circuit provided between the connection points of
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the lower-stage main switch, and the lower-stage auxiliary switch includes a connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. A lower snubber circuit provided between the connection points of
The voltage between the terminals of the snubber capacitor of the snubber circuit for one of the upper main switch and the lower main switch is the output voltage of the DC power converter, and the snubber capacitor terminal of the other snubber circuit A DC power conversion device in which switching is started at a startup frequency lower than a predetermined switching frequency in a steady state when the one main switch is periodically switched from a state in which the inter-voltage is zero.   The on-state period in one cycle when the one main switch is switched at the start-up frequency is set to the on-state period in one cycle when the main switch is switched at the predetermined switching frequency. The DC power converter according to claim 1.   The frequency at which the one main switch is switched is determined from the start-up frequency when the voltage between the terminals of the snubber capacitor of the snubber circuit with respect to the other main switch becomes the output voltage of the DC power converter. The direct-current power converter according to claim 1, wherein the direct-current power converter is changed to the switching frequency.   The frequency at which the one main switch is switched is changed from the startup frequency to the predetermined switching frequency after a lapse of a predetermined time after switching of the one main switch is started. The direct-current power converter according to one item.   The DC power according to any one of claims 1 to 4, wherein a frequency for switching the one main switch is changed to a high frequency stepwise or continuously from the startup frequency to the predetermined switching frequency. Conversion device.   The DC power conversion device according to any one of claims 1 to 5, wherein the predetermined switching frequency in the steady state is a modulation frequency in pulse width modulation control. A DC power conversion device provided between an inverter interposed between a rotating electrical machine and a DC power source that supplies power to the rotating electrical machine and the DC power source,
A main reactor in which the first reactor and the second reactor are equivalently configured as one reactor by being connected in series, and one terminal is connected to the positive electrode of the DC power supply,
An upper main switch that connects the other terminal of the main reactor and the positive output terminal of the DC power converter;
A lower main switch that connects the other terminal of the main reactor and the negative output terminal of the DC power converter;
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the upper main switch, and the upper auxiliary switch is connected to the connection point of the snubber capacitor and the snubber diode, the first reactor, and the second reactor. An upper snubber circuit provided between the connection points of
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the lower-stage main switch, and the lower-stage auxiliary switch includes a connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. A lower snubber circuit provided between the connection points of
Both the upper stage main switch and the lower stage main switch are in an off state, a flywheel diode connected in parallel to the upper stage main switch functions as an output diode, and the output voltage of the DC power supply is the DC power conversion From the direct connection mode that is output as the output voltage of the device, the lower-side main switch is periodically switched to boost the output voltage of the DC power supply and to the boost mode that is output as the output voltage of the DC power converter. A DC power conversion device in which switching is started at a startup frequency lower than a predetermined switching frequency in a steady state. A DC power conversion device provided between an inverter interposed between a rotating electrical machine and a DC power source that supplies power to the rotating electrical machine and the DC power source,
A main reactor in which the first reactor and the second reactor are equivalently configured as one reactor by being connected in series, and one terminal is connected to the positive electrode of the DC power supply,
An upper main switch that connects the other terminal of the main reactor and the positive output terminal of the DC power converter;
A lower main switch that connects the other terminal of the main reactor and the negative output terminal of the DC power converter;
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the upper main switch, and the upper auxiliary switch is connected to the connection point of the snubber capacitor and the snubber diode, the first reactor, and the second reactor. An upper snubber circuit provided between the connection points of
A series circuit of a snubber capacitor and a snubber diode is provided in parallel with the lower-stage main switch, and the lower-stage auxiliary switch includes a connection point between the snubber capacitor and the snubber diode, the first reactor, and the second reactor. A lower snubber circuit provided between the connection points of
The operation mode is switched between a power running mode and a regeneration mode, which are two operation modes in which one of the upper main switch and the lower main switch is in an OFF state and the other is periodically switched. In this case, the switching of any one of the upper-side main switch and the lower-side main switch that is periodically switched from the OFF state is lower than the predetermined switching frequency in the steady state. DC power converter started at hourly frequency.   Any one of the upper-side main switch and the lower-side main switch that is turned off by the switching of the operation mode is switched from the predetermined switching frequency to the off-state through switching at the startup frequency. The DC power converter according to claim 8 to be performed. When the power in the power running mode and the power in the regenerative mode become a predetermined threshold power or less in the flow direction of each power, it is a transition period for switching the operation mode,
The DC power converter according to claim 8 or 9, wherein in the transition period, complementary control in which both the upper stage main switch and the lower stage main switch are periodically switched is executed. As the rotating electrical machine, a first rotating electrical machine that mainly functions as a generator, and a second rotating electrical machine that mainly functions as an electric motor,
11. The DC power conversion according to claim 10, wherein when the difference between the power generated by the first rotating electrical machine and the power consumed by the second rotating electrical machine becomes equal to or less than the predetermined threshold power, the transition period is reached. apparatus.   The said starting frequency is set based on the power flow direction and the magnitude | size of electric power with respect to each of the said upper stage side main switch and the said lower stage side main switch in the said transition period. DC power converter.

Images