DESCRIPTION POWER CONVERSION DEVICE AND REFRIGERATION/AIR-CONDITIONING SYSTEM Field [0001] The present invention relates to a power conversion device and a refrigeration/air-conditioning system using the power conversion device. Background [0002] Application fields of various power conversion devices have been developed as variable voltage/variable frequency inverters are put to practical use. [0003] With regard to power conversion devices, development of an application technique of a step-up/step-down converter has been widely in progress. Along with this development, in recent years, development of wide bandgap semiconductors has been widely in progress. The wide bandgap semiconductor has characteristics such as a higher voltage resistance, a lower power loss, and being capable of operating under a higher temperature, as compared with conventional semiconductors, and thus has been put to practical use mainly in rectifiers with regard to an element having a small current capacity (see, for example, Patent Literatures 1 to 4). Citation List Patent Literatures [0004] Patent Literature 1: Japanese Patent Application Laid-open No. 2005-160284 Patent Literature 2: Japanese Patent Application Laid-open No. 2006-067696 Patent Literature 3: Japanese Patent Application Laid-open No. 2006-006061 Patent Literature 4: Japanese Patent Application Laid-open No. 2008-061403 Summary Technical Problem [0005] However, among new devices having a low power loss and high efficiency, an element having a large current capacity has many problems in order to put it to practical use, the problems including a high cost, crystal defects, and the like. Therefore, it is considered that more time is required for popularization thereof, and it is difficult to realize high efficiency of an element having a large current capacity by applying a new device having a large current capacity to the power conversion device. [0006] The present invention has been achieved to solve the above problems, and an object of the present invention is to provide a power conversion device and a refrigeration/air-conditioning system that can ensure high efficiency and high reliability without using any new device having a large current capacity. Solution to Problem [0007] In order to solve the abovementioned problem and achieve the object, a power conversion device according to the present invention includes a power supply unit; a booster unit that boosts a voltage supplied from the power supply unit by switching control; a smoothing unit that smoothes an output voltage from the booster unit; a backflow prevention element that is arranged between the booster unit and the smoothing unit to prevent a current reverse flow toward the booster unit; and a commutation unit that is connected to the backflow prevention element in parallel to commutate a current flowing through the backflow prevention element toward the commutation unit itself. Advantageous Effects of Invention [0008] According to the present invention, control can be executed so that a reverse bias is applied to a rectifier, which is a backflow prevention element, after a forward current flowing through the rectifier is commutated toward a commutation unit. With this configuration, a recovery current of the rectifier can be suppressed, thereby enabling to realize a highly reliable and highly efficient power conversion device. Brief Description of Drawings [0009] FIG. 1 is a configuration example of a power conversion device according to a first embodiment. FIG. 2A is an explanatory diagram of an operation mode in the power conversion device. FIG. 2B is an explanatory diagram of an operation mode in the power conversion device. FIG. 2C is an explanatory diagram of an operation mode in the power conversion device. FIG. 2D is an explanatory diagram of an operation mode in the power conversion device. FIG. 3 is an example of a commutation control operation. FIG. 4 is an example of a switch control unit. FIG. 5A is an example of a drive signal generated by the switch control unit. FIG. 5B is an example of a drive signal generated by the switch control unit. FIG. 6 is a modification of the switch control unit shown in FIG. 4. FIG. 7A is an example of a drive signal generated by the switch control unit shown in FIG. 6. FIG. 7B is an example of a drive signal generated by the switch control unit shown in FIG. 6. FIG. 8 is an example of a control operation using a sawtooth signal. FIG. 9 is an example of a control operation using a sawtooth signal. FIG. 10A depicts a relation between a forward current and a recovery current. FIG. 10B depicts a relation between a forward current and a recovery current. FIG. 11A is an example of a relation between a current and 0FS1. FIG. 11B is an example of a relation between a current and 0FS2. FIG. 12 is a configuration example of a power conversion device according to a second embodiment. FIG. 13 is a configuration example of a power conversion device according to a third embodiment. Description of Embodiments [0010] Exemplary embodiments of a power conversion device and a refrigeration/air-conditioning system according to the present invention will be explained below in detail with reference to the accompanying drawings. The present invention is not limited to the embodiments. [0011] First embodiment. FIG. 1 is a configuration example of a power conversion device according to a first embodiment of the present invention. For example, the power conversion device is used in a refrigeration/air-conditioning system. The configuration of the power conversion device is explained first with reference to FIG. 1. [0012] As shown in FIG. 1, the power conversion device according to the present embodiment includes a power supply 1 that supplies power, a booster circuit 2 that boosts power supplied from the power supply 1, a smoothing circuit 3 that smoothes an output voltage from the booster circuit 2 or a commutation unit 4, which is described later, the commutation unit 4 that commutates a current flowing through the booster circuit 2 to a different path at a required timing, a voltage detection unit 5 that detects a voltage after being smoothed by the smoothing circuit 3, a control unit 6 that controls the booster circuit 2 and the commutation unit 4, a drive-signal transmission unit 7 that transmits a drive signal Sa of the booster circuit 2 generated by the control unit 6 to the booster circuit 2, a commutation-signal transmission unit 8 that transmits a drive signal Sb (also referred to as "commutation signal") of the commutation unit 4 generated by the control unit 6 to the commutation unit 4, a load 9 that is connected to the smoothing circuit 3 at the subsequent stage, a current detection element 10 that detects the current flowing through the booster circuit 2, and a current detection unit 11 that converts a detection result obtained by the current detection element 10 to a signal in a format usable by the control unit 6. [0013] An ACCT (current transformer) or a DCCT (using a Hall element, a Hall IC, or the like) is mainly used as the current detection element 10. The current detection unit 11 includes an amplifier circuit, a level-shift circuit, a filter circuit, and/or the like for converting a value detected by the current detection element 10 to an appropriate value (Idc) that can be processed in the control unit 6 to allow it to be acquired. When the function of the current detection unit 11 is included in the control unit 6, the current detection unit 11 may be omitted as appropriate. When any current control is not executed (when being applied to a device that does not require any control taking into consideration the current value flowing through the booster circuit 2), the current detection element 10 and the current detection unit 11 may be omitted as appropriate. [0014] The booster circuit 2 includes a reactor 21 connected to the positive side of the power supply 1, and a switch 22 as a switching element and a rectifier 23 as a backflow prevention element (the point B side is the anode side, and the point C side is the cathode side) that are connected at the reactor 21 at the subsequent stage. The reactor 21 can be connected to the negative side of the power supply 1. The switching state of the switch 22 is changed by a drive signal SA input from the drive-signal transmission unit 7. The booster circuit 2 boosts input power from the power supply 1 in accordance with an "ON" time to "OFF" time ratio (duty ratio) of the drive signal SA. The drive-signal transmission unit 7 is normally formed of a buffer, a logic IC, a level-shift circuit, and/or the like. However, the drive-signal transmission unit 7 can be omitted as appropriate, for example when the function of the drive-signal transmission unit 7 is provided in the control unit 6. In this case, the drive signal Sa generated by the control unit 6 serves as the drive signal SA, thereby directly performing a switching operation of the switch 22. [0015] The commutation unit 4 includes a transformer 41, a rectifier 42 serially connected to the transformer 41, and a transformer drive circuit 43 that drives the transformer 41. In FIG. 1, the polarity of a primary winding and the polarity of a secondary winding of the transformer 41 are made the same. The secondary winding of the transformer 41 is serially connected to the rectifier 42. The rectifier 42 is further connected to the rectifier 23 of the booster circuit 2 in parallel. The rectifier 42 operates as a backflow prevention element in the commutation unit 4. [0016] The transformer drive circuit 43 includes, for example, a power supply 45 and a switch 4 4 for driving the transformer 41. A limiting resistor, a high-frequency capacitor, a snubber circuit, a protection circuit, and/or the like can be inserted, as required, into a path of the power supply 45, the switch 44, and the primary winding of the transformer 41, taking into consideration a measure against noise and protection at the time of failure. In the example shown in FIG. 1, a reset winding that resets an excitation current is not provided in the transformer 41; however, the reset winding can be added to the primary winding as required, and a rectifier or the like can be provided to regenerate excitation energy to the power supply side. This can enhance efficiency. [0017] The switching state of the switch 44 is changed by a commutation signal SB input from the commutationT signal transmission unit 8. The commutation-signal transmission unit 8 is normally formed of a buffer, a logic IC, a level-shift circuit, and/or the like as in the drive-signal transmission unit 7. However, when the function of the commutation-signal transmission unit 8 is provided in the control unit 6, the commutation-signal transmission unit 8 may be omitted as appropriate. In this case, the commutation signal Sb generated by the control unit 6 serves as the commutation signal SB, thereby directly performing a switching operation of the switch 44. [0018] The voltage detection unit 5 is formed of a level-shift circuit by a partial resistance or the like. An analog/digital converter can be added as required so that a detection value can be computed in the control unit 6. [0019] The control unit 6 controls the booster circuit 2 and the commutation unit 4 based on at least one of a voltage value Vdc indicating a voltage detection result obtained by the voltage detection unit 5 and a current ■ value Idc indicating a current detection value obtained by the current detection unit 11. The control unit 6 can be formed of an arithmetic device such as a microcomputer or a digital signal processor, or a device having the same function inside thereof. [0020] An operation of the power conversion device shown in FIG. 1 is explained next. In the following explanations, it is assumed that the control unit 6 includes functions similar to those of the drive-signal transmission unit 7 and the commutation-signal transmission unit 8 (the drive-signal transmission unit 7 and the commutation-signal transmission unit 8 are omitted) to simplify the explanations. [0021] The operation of the power conversion device according to the present embodiment is such that a commutation operation of the rectifier is added to a boost chopper. There are four operation modes in total corresponding to combinations of the switching states of the switch 22 and the switch 44. It is assumed here that a recovery characteristic of the rectifier 42 is better than that of the rectifier 23. [0022] (First mode) A case where the switch 22 is ON and the switch 44 is OFF is considered. An element having a lower forward voltage is used as the rectifier 23, as compared with the rectifier 42 having the better recovery characteristic. Because the winding of the transformer 41 is an inductor component, when an excitation current does not flow, the current does not flow. Accordingly, in the case where the switch 44 is OFF, the current does not flow into the path of the commutation unit 4. Because the switch 22 is ON, energy is accumulated in the reactor 21 by a path shown in FIG. 2A. [0023] (Second mode) A case where the switch 22 is OFF and the switch 44 is OFF is considered. Also in this case, the switch 44 is OFF as in the first mode, and thus the current does not flow into the path of the commutation unit 4. Furthermore, because the switch 22 is OFF, the energy of the reactor 21 is supplied to the load 9 in a path shown in FIG. 2B. [0024] (Third mode) A case where the switch 22 is ON and the switch 44 is ON is considered. In this case, although the switch 44 is ON, the switch 22 is ON simultaneously. Because the impedance is lower on the power supply 1 side, almost no current flows into the path of the commutation unit 4, and energy is accumulated in the reactor 21 in a path shown in FIG. 2C. This mode may be generated instantaneously due to transmission delay of the commutation signal SB or the like; however, it does not cause any problem in use. [0025] (Fourth mode) A case where the switch 22 is OFF and the switch 44 is ON is considered. In this case, the switch 22 is OFF and the current flows toward the load 9 via the rectifier 23 (a current path #1 shown in FIG. 2D). Furthermore, because the switch 44 is ON, the transformer 41 is excited, and the current also flows into the path of the commutation unit 4 (a current path #2 shown in FIG. 2D). When a certain time has passed after the switch 22 and the switch 24 are turned on, the current flowing in the current path #1 (the rectifier 23) is completely commutated toward the rectifier 42. [0026] As described above, although a commutation operation is generated in the case of the fourth mode (the switch 22 is OFF and the switch 44 is ON), an energy accumulation operation by switching of the switch 22 follows a pattern of the booster chopper. Therefore, when switching is repeatedly performed by the switch 22 such that the switch 22 is turned on during an ON time Ton and off during an OFF time Toff, an average voltage Ec expressed by the following equation (1) is obtained at the point C shown in FIG. 1. Here, for simplification, the voltage of the power supply 1 is assumed to be a DC power supply Ei. [0027] [0028] FIG. 3 is an example of a commutation control operation performed by the power conversion device. Specifically, FIG. 3 represents a relation among the drive signal Sa (a drive signal for controlling the booster circuit 2) and the drive signal Sb (a commutation signal for controlling the commutation unit 4) output by the control unit 6, and respective current waveforms I1 to I5 shown in FIG. 1. In the output signals Sa and Sb from the control unit 6, the HI side corresponds to the active direction (ON direction). Furthermore, the respective waveforms are shown in a state after a sufficient time has passed since power on of the power supply 1, that is, a state after the control unit 6 has controlled the ON time and the OFF time of the drive signal Sa so that the load 9 and the output voltage Vdc have a constant output. [0029] Furthermore, FIG. 3 is an example in which the "ON" time to "OFF" time ratio (duty ratio) of the drive signal Sa is substantially constant. That is, the "ON" time to "OFF" time ratio is set constant, assuming the case where the power supply 1 is a DC power supply. When the power supply 1 is an AC power supply, for example, the "ON" time to "OFF" time ratio (duty ratio) of the drive signal Sa only needs to be adjusted by proportional-integral control or the like so .that the voltage on the DC side becomes constant. Furthermore, FIG. 3 is a waveform example when the pulse width of the drive signal Sb is fixed. A case where the pulse width of the drive signal Sb is set to be variable is explained separately. [0030] I1 represents a current flowing in the reactor 21. I1 is branched at the point A in FIG. 1, and divides into a current I2 flowing in the switch 22 and a current I3 flowing toward the rectifier 23. Therefore, the relation of these current values is shown by the following equation (2) . I1=I2+I3 (2) [0031] I3 is branched at the point B, and divides into a current I4 flowing in the rectifier 23 and a current I5 flowing toward a secondary wining of the transformer 41 and the rectifier 42. Therefore, the relation of these current values is shown by the following equation (3). I3=l4+l5 (3) [0032] When the drive signal Sa is ON in the state where the forward current flows through the rectifier 23, conduction is established between the point A and the point D, and thus the potential at the point B becomes substantially equal to the potential at the point D (the point A and the point B have the same potential). For example, when an insulated gate bipolar transistor (IGBT) or a field-effect transistor (FET) is used as the switch 22, an ON voltage of these elements causes a potential difference between the point B and the point D (the potential of the point B becomes substantially equal to a negative side potential of the power supply 1). Meanwhile, the potential of the point C is substantially held in a state of charging potential (a charging potential of the capacitor constituting the smoothing circuit 3) by the smoothing circuit 3. Therefore, at this time, a reverse bias is applied to the rectifier 23 by as much as the potential difference between the point C and the point B, and the rectifier 23 shifts to an OFF operation. [0033] Generally, a pn junction diode is used as the rectifier 23. In this case, a short-circuit current flows in the path from the rectifier 23 to the switch 22, until reverse recovery of the rectifier 23 is complete (hereinafter, "recovery current"). Therefore, to prevent an increase in a circuit loss (a power loss) due to the recovery current, the control unit 6 turns on the commutation signal Sb of the commutation unit 4 in a predetermined period immediately before turning on the drive signal Sa. Accordingly, the current flowing through the rectifier 23 is commutated toward the commutation unit 4 (commutated to the rectifier 42 via the transformer 41) (see FIG. 2D). [0034] An element that can endure a peak current repeatedly but has a small current capacity (rated) (an element having a high voltage resistance but a small current capacity), as compared with the rectifier 23, is used as the rectifier 42. [0035] Generally, in a rectifier, an element having a smaller current capacity has a smaller amount of accumulated carriers than an element having a larger current capacity. Accordingly, as the current capacity decreases, the time until the reverse recovery is complete can be reduced, and the recovery current also decreases. The amount of accumulated carriers of the rectifier depends on the magnitude of the forward current. As the applied reverse bias voltage decreases, the recovery current decreases. Consequently, by turning on the commutation signal Sb to commutate the current flowing through the rectifier 23 toward the rectifier 42 before turning on the drive signal Sa, the recovery current flowing in the path from the rectifier 23 to the switch 22 can be reduced. [0036] The rectifier 42 can be formed of a wide bandgap semiconductor formed of Sic, GaN, diamond, or the like. The wide bandgap semiconductor has a lower conduction loss and a lower switching loss than conventional semiconductors (non-wide bandgap semiconductors), and thus making it possible to enhance the efficiency of the power conversion device. Furthermore, because the wide bandgap semiconductor has a high voltage resistance and a high allowable current density, the rectifier can be downsized, and by using the downsized rectifier, the device can be also downsized. [0037] In the above explanations, the duty ratio of the drive signal Sa is explained as being constant; however, a highly efficient operation may become possible by changing the output of the booster circuit as appropriate, due to such a reason that, for example, when the load 9 is an electric motor, a generated inductive voltage varies depending on a rotation frequency of the electric motor. Furthermore, because the required output voltage of the booster circuit is different depending on the specification of the electric motor, a required load torque, and operating conditions, the "ON" time to "OFF" time ratio (duty ratio) of the drive signal Sa is adjusted as appropriate. The adjustment process is performed in the control unit 6. For example, the adjustment process is realized by using a controller that executes proportional-integral (PI) control to execute proportional-integral control, using, as inputs, the actual output voltage Vdc obtained by the voltage detection unit 5 and a target voltage Vdc* (a command value) set in the control unit 6. [0038] As a result, feedback control is executed so that the actual output voltage Vdc approaches the target value Vdc*, and the ON time of the drive signal Sa is successively corrected and set. After a certain time has passed, the Vdc and the Vdc* become substantially the same, excluding a steady-state deviation. [0039] The Vdc* may be mapped as an internal memory, and the value thereof may be changed according to the operating conditions. Alternatively, control may be executed by storing the Vdc* in an external memory and reading the Vdc* into the control unit 6. [0040] When current control is required, a reference signal (duty) of the drive signal Sa may be generated, taking into consideration the current value Idc obtained by the current detection element 10. The adjustment process is performed in the control unit 6. For example, two controllers that execute proportional-integral control are used. First, a first controller receives inputs of the actual output voltage Vdc obtained by the voltage detection unit 5 and the target voltage Vdc* (the command value) set in the control .unit 6, to execute the proportional-integral control and outputs a current command value Idc*. Next, a second controller receives inputs of the current command value Idc* and a current detection value Idc to execute feedback control so that the actual output current Idc approaches the target value Idc*, to correct and set the ON time of the drive signal Sa successively. Also in this case, the Vdc and the Idc become substantially the target value (excluding a steady-state deviation) after a certain time has passed. Furthermore, by appropriately adjusting the Idc* depending on the power supply voltage, a power factor of the power supply can be improved and a high-frequency current can be suppressed. [0041] A control dead time or the like needs to be considered according to use conditions. Therefore, the controllers may execute PID control combined with a derivative control action according to the conditions. [0042] The Idc* may be mapped as an internal memory and the value thereof may be changed according to the operating conditions. That is, instead of obtaining the Idc* by the proportional-integral control or the like using the output voltage Vdc and the target voltage Vdc* (the command value), a plurality of Idc* values may be prepared, and an appropriate Idc* value according to the operating conditions may be used. Furthermore, control may be executed by storing the Idc* in an external memory and reading the Idc* into the control unit 6. Alternatively, control may be executed with an alternative amount such as electric power instead of the electric current. [0043] In the above explanations, a method of performing the commutation operation is shown from the aspect of the drive signals of the switches 22 and 24. Meanwhile, the actual switching speed of the switches 22 and 24 changes according to the type of the element and various conditions of the drive circuit (setting of the constant of a gate peripheral circuit and the like) . Therefore, even if a rise timing (an on-timing) of the drive signal Sb (SB) of the switch 44 and a fall timing (an off-timing) of the drive signal Sa (SA) of the switch 22 are set the same, actual switching timings of the switches 44 and 22 do not always coincide with each other. Furthermore, with regard to the drive signal Sb (SB) of the switch 22, it can be considered that the recovery suppression effect can be increased by adjusting a pulse width of the signal Sb (SB) in order to ensure the required commutation time. Therefore, a circuit having high versatility that can change the ON/OFF timing of the switches 44 and 22 is explained below. [0044] FIG. 4 is an example of a switch control unit that generates switch drive signals (the drive signals Sa and Sb). The switch control unit is provided in the control unit 6. The switch control unit shown in FIG. 4 includes reference-signal generation units 201i to 2013 that generate respective reference signals having different levels (fixed reference values), a triangle-wave-signal generation unit 202 that generates a triangle-wave signal, a state-memory enable-signal generation unit 203 that generates a state-memory enable signal (details thereof are described later), comparators 2111 to 2II3 that compare two input signals, and an arithmetic-logic unit 220 including a logic inversion unit 221, a logical-product computation unit 222, and a state storage unit 223. [0045] FIG. 5A is an example of a drive signal generated by a switch control unit having the configuration shown in FIG. 4. Details of a switch drive-signal generation operation in the switch control unit are explained below with reference to FIGS. 4 and 5A. [0046] In the generation operation of the switch drive signal, the reference-signal generation units 2011 to 2013 first generate a reference signal S1 (a first reference signal), a reference signal S2 (a second reference signal), a reference signal S3 (a third reference signal), respectively, as a threshold. It is assumed that these reference signals have a relation of S3Sl), outputs the Hi-level drive signal Sa (ON). On the other hand, when Sc is smaller than S1 (ScS2), outputs the Hi-level signal Sy. On the other hand, when Sc is smaller than S2 (ScS3), outputs the HI-level signal Sz. On the other hand, when Sc is smaller than S3 (ScSl), outputs the Hi-level drive signal Sa (ON). On the other hand, when Scl is smaller than S1 (ScKSl) , the comparator 2121 outputs the LO-level drive signal Sa (OFF). [0060] Similarly, the comparator 2122 generates the signal Sy based on the second triangle-wave signal Sc2 and the reference signal S1. Specifically, the comparator 2122 compares the second triangle-wave signal Sc2 with the reference signal S1, and when Sc2 is equal to or larger than S1 (Sc2>Sl), outputs the Hi-level signal Sy. On the other hand, when Sc2 is smaller than S1 (Sc2Sl), outputs the Hi-level signal Sz. On the other hand, when Sc3 is smaller than S1 (Sc3