POWER CONVERTER CONTROL DEVICE

Abstract

A power converter control device (10, 30, 301, 40) for performing PWM control for power converters (81A, 81B) each supplying power to a plurality of winding groups of a multiplexed winding electric motor (83) having the plurality of winding groups, the power converter control device (10, 30, 301, 40) comprising: - a PWM control unit (18, 38, 48) configured to perform PWM control while switching between an asynchronous PWM control mode using a carrier for PWM control that is not synchronized with the three-phase voltage commands for driving the multiplexed winding electric motor (83) and a synchronous PWM control mode using a carrier for PWM control that is synchronized with the three-phase voltage commands for driving the multiplexed winding electric motor (83), wherein the synchronous PWM control mode includes a first synchronous PWM control mode in which voltage commands for the respective winding groups of the multiplexed winding electric motor (83) are subjected to PWM control using a carrier synchronized with a voltage command for a specific winding group, and a second synchronous PWM control mode in which voltage commands for the respective winding groups of the multiplexed winding electric motor (83) are subjected to PWM control using respective carriers synchronized with the respective voltage commands, and wherein the PWM control unit (18, 38, 48) performs control while selectively switching between the asynchronous PWM control mode and both or one of the first synchronous PWM control mode and the second synchronous PWM control mode based on an operating state of the multiplexed winding electric motor (83).

Description

Technical FieldThe present invention relates to a converter control device that controls a converter for performing power conversion between DC power and AC power using a plurality of semiconductor switching elements. More particularly, it relates to a converter control device that controls a converter for supplying AC power to a plurality of winding groups so as to drive a multiplexed electric motor having an armature winding multiplexed by the plurality of winding groups.Prior ArtA converter is connected between an AC power system and a DC power system, and is used for converting power between DC and AC. In the case where an electric motor is operated using such a converter, pulse width modulation (hereinafter, PWM) control is widely used as a control method thereof. In such PWM control, a voltage command for driving the electric motor is subjected to triangular wave comparison processing using a carrier signal. As a result, a switching pulse for driving a switching element of the power converter is generated.If the rotational speed of the electric motor becomes high while the carrier frequency is constant, the number of times of the voltage command per switching cycle decreases. Therefore, in order to drive the electric motor at high speed, synchronous PWM control in which the phase of the voltage command and the phase of the carrier signal are synchronized and the frequency of the carrier signal is set to an integer multiple of the frequency of the voltage command (which is often an odd multiple such as three times or nine times) can be performed. Applying synchronous PWM control allows the electric motor to be driven with fewer pulses, resulting in a reduction in the switching losses of the converter.In the case where the voltage command is updated in the synchronous PWM control, the calculation therefor needs to be performed between the peak and the peak, between the ground and the ground, or between the peak and the ground of the carrier signal. When the voltage command is changed, the phase of the carrier signal is also changed, so that the time for calculating the voltage command varies.In the case of a multiplexed winding electric motor, a plurality of converters for supplying AC power to each winding group and a voltage command and a carrier signal corresponding to each winding group are needed. In addition, a change in current flowing through one winding group may affect the current flowing through another winding group. Therefore, there is the case where each voltage command is calculated using currents of the plurality of winding groups, but when synchronous PWM control is performed using two carrier signals, the timing for calculating each voltage command becomes complicated.As for PWM control for a plurality of winding groups as described above, the following electric motor control means has already been proposed: In the case where three-phase AC electric motors are driven, the period of two carrier signals having equal periods and for generating two sets of PWM waveforms is set as a counter value of a clock number, and a period set value for forming a phase difference is cumulatively added to the counter value for the period of the one carrier signal until a counter value corresponding to a predetermined phase difference is reached. Thereby, the phase relationship is shifted to each other so as to achieve a desired phase difference (see, for example, JP 2007-252 138 A (Patent Document 1)).In addition, the following example has been proposed: In a synchronous operation mode of an inverter (converter), a synchronous PWM reference signal is transmitted to slave inverters (slaves) in series by a master inverter (master), collectively, and the oscillation frequency of a carrier signal is controlled in synchronization with the received synchronous PWM reference signal on the slave inverter side (see, for example, JP 2007-295 647 A (Patent Document 2)).JP 2017-93 208 A discloses a motor control device configured to drive a dual three-phase winding motor and including a high voltage battery, a capacitor, two converters, current sensors, switching units, a motor, a resolver, and a controller. Besides a temperature signal from temperature sensors and a motor rotation speed signal from the resolver, the controller receives a torque command from a host controller and an abnormality detection signal from the converters. Based on these input signals, the controller provides a gate control signal to the converters. The control device includes a phase difference calculation unit; a rotation speed calculation unit; a torque distribution calculation unit; and control units of the converters. The phase difference calculation unit calculates switching between synchronous PWM control or asynchronous PWM control based on the temperature signals of the temperature sensors and the rotation speed. If the internal temperature of the converters is below a threshold value and the rotational speed is below a threshold value, asynchronous PWM is selected, otherwise synchronous PWM.DE 11 2013 006 977 T5 discloses a converter for driving an AC electric motor, which is a load including an inverter circuit; a DC voltage source unit; a switching signal generation unit including a carrier wave generation unit, a modulated wave generation unit, and a comparison unit; a modulation rate calculation unit; a modulated wave selection unit; a voltage detection unit; and an AC voltage instruction generation unit. In an embodiment, the switching signal generation unit includes an asynchronous PWM switching signal generation unit, a synchronous PWM switching signal generation unit, and a switching signal selection unit that selects gate signals output from the asynchronous PWM switching signal generation unit and the synchronous PWM switching signal generation unit according to the modulation mode.SUMMARY OF THE INVENTIONProblems to be Solved by the InventionHowever, in the electric motor control device described in Patent Document 1, the period and the phase of the carrier signal are controlled using a PWM timer to obtain a desired phase difference. Therefore, this is not applicable to synchronous PWM control that is synchronized with the rotation frequency of the winding groups, and there is a possibility that the electric motor cannot be stably controlled during the high-speed rotation.In the synchronous operation mode of an inverter as described in Patent Document 2, in a range in which there are three switching pulses per one period of the voltage command, when the rotation speed of the electric motor becomes even higher, the phase of the voltage command and the phase of the fundamental wave included in the switching pulses are shifted from each other by synchronizing the carrier signals between the different inverters, and thus control may become unstable.The present invention has been devised to solve the above problem. It is therefore an object of the above invention to provide a converter control device that controls a converter for supplying power to a multiplexed electric motor and enables a multiplexed electric motor to be stably operated.Solution of the ProblemsThe object on which the invention is based is achieved by a converter control device having the features of independent claim 1. Advantageous further developments of the converter control device according to the invention are specified in the dependent claims 2 to 10.Effect of the inventionThe present invention makes it possible to obtain a converter control device that controls a converter for supplying power to a multiplexed electric motor, and makes it possible to stably drive a multiplexed electric motor over a wide span from a low-speed range to a high-speed range.Brief Description of the DrawingsFIG. 1 is a block diagram showing the configuration of a converter control device according to Embodiment 1 of the present invention. FIG. 2 is a hardware configuration diagram of the converter control device according to Embodiment 1 of the present invention. FIG. 3 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 1 of the present invention. FIG. 4 is a timing chart showing the relationship between voltage commands for respective winding groups, an asynchronous carrier signal, and switching pulses for respective winding groups according to Embodiment 1 of the present invention. FIG. 5 is a timing chart showing the relationship between voltage commands for respective winding groups and carrier signals for respective winding groups in 9-pulse synchronous PWM control according to Embodiment 1 of the present invention. FIG. 6 is a timing chart showing the relationship between voltage commands for respective winding groups and carrier signals for respective winding groups in 6-pulse synchronous PWM control according to Embodiment 1 of the present invention. FIG. 7 is a timing chart showing the relationship between voltage commands for respective winding groups and carrier signals for respective winding groups in 3-pulse synchronous PWM control according to Embodiment 1 of the present invention. FIG. 8 shows calculation timings in the case where voltage command calculation is performed between the peaks and bottoms of carrier signals for both winding groups in synchronous PWM control according to Embodiment 1 of the present invention. FIG. 9 shows calculation timing in a case where voltage command calculation is performed between a peak and a bottom of a carrier signal for a single winding group in synchronous PWM control according to Embodiment 1 of the present invention. FIG. 10 shows calculation timings in the case where voltage command calculation is performed between the peaks of carrier signals for both winding groups in synchronous PWM control according to Embodiment 1 of the present invention. FIG. 11 shows calculation timing in the case where voltage command calculation is performed between the peaks of a carrier signal for a single winding group in synchronous PWM control according to Embodiment 1 of the present invention. FIG. 12 shows the relationship between the operation state of a multiplexed electric motor and the type of a carrier signal used in the PWM control according to Embodiment 1 of the present invention. Fig. 13 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 2 of the present invention. FIG. 14 is a block diagram showing the configuration of a converter control device according to Embodiment 3 of the present invention. Fig. 15 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 3 of the present invention. FIG. 16 shows calculation timings in the case where voltage command calculation is performed between the peaks and bottoms of carrier signals for both winding groups in synchronous PWM control according to Embodiment 3 of the present invention. FIG. 17 shows calculation timings in the case where voltage command calculation is performed between the peaks of carrier signals for both winding groups in synchronous PWM control according to Embodiment 3 of the present invention. FIG. 18 shows the relationship between the operation state of a multiplexed winding electric motor and the type of a carrier signal used in the PWM control according to Embodiment 3 of the present invention. FIG. 19 is a block diagram showing the configuration of a converter control device in another example according to Embodiment 3 of the present invention. FIG. 20 is a block diagram showing the configuration of a converter control device according to Embodiment 4 of the present invention. Fig. 21 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 4 of the present invention. FIG. 22 shows calculation timings in the case where voltage command calculation is performed between the peaks and bottoms of carrier signals for both winding groups in synchronous PWM control according to Embodiment 4 of the present invention. Fig. 23 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 5 of the present invention. FIG. 24 is a timing chart showing the relationship between a voltage command for each winding group and asynchronous carrier signals according to Embodiment 5 of the present invention.DESCRIPTION OF EMBODIMENTSEmbodiment 1Hereinafter, Embodiment 1 of the present invention will be described with reference to FIGS. 1 to 12. FIG. 1 is a block diagram showing the configuration of a converter control device according to Embodiment 1. A converter controller 10 controls a first converter 81B and a second converter 81A for supplying AC power to a multiplexed electric motor 83 having, for example, two armature winding groups (not shown) formed of a first group and a group.The multiplexed winding electric motor 83 is, for example, a three-phase AC electric motor that includes a rotor using a permanent magnet or the like and is configured such that the winding groups of the first group and the second group each formed of three-phase windings are provided with a predetermined phase difference (electrical angle) therebetween. The first group winding group, i.e., the first winding group, is connected to the first inverter 81A, and the second group winding group, i.e., the second winding group, is connected to the second inverter 81B, so that the first inverter 81A supplies power to the first group winding group and the second inverter 81B supplies power to the second group winding group.The first converter 81A and the second converter 81B each include a plurality of switching elements (not shown). These switching elements are turned on / off by PWM control so that DC power supplied from a DC power supply (not shown) is converted into three-phase AC power, and three-phase AC voltages VU 1, VV 1, VW 1 of the first group and three-phase AC voltages VU 2, VV 2, VW 2 of the second group are applied to the respective winding groups so as to be supplied with the three-phase AC power.A first current detector 82A and a second current detector 82B are connected between the electric motor 83 having a multiplex winding, and the first inverter 81A and the second inverter 81B. The first current detector 82A and the second current detector 82B detect three-phase AC currents flowing between the multiplexed winding electric motor 83 and the first inverter 81A and the second inverter 81B, and output first three-phase current values iu 1, iv 1, iw 1 for the first group and second three-phase current values iu 2, iv 2, iw 2 for the second group as current information. A position sensor 84 is mounted on the electric motor 83 with multiplex winding.The position sensor 84 detects the magnetic pole position of the rotor of the multiplexed winding electric motor 83 and outputs a magnetic pole position signal θe. As the first current detector 82A and the second current detector 82B, for example, shunt resistors or current transformers are used, and as the position sensor 84, for example, an incremental rotary encoder or a resolver is used. In place of the position sensor 84, a rotation sensor for detecting the magnetic pole position from the rotational speed of the rotor may be used.The inverter controller 10 outputs switching pulses generated based on current commands supplied from the outside to the first inverter 81A and the second inverter 81B. Thereby, the first inverter 81A and the second inverter 81B are controlled. As for the control for an induction electric motor or a synchronous electric motor, generally, the control is performed in a two-phase orthogonal coordinate system represented by the d-axis and the q-axis of a rotational coordinate system. Therefore, in the following description as well, control using a two-phase orthogonal coordinate system will be described.The converter control device 10 includes:a voltage command generation unit 11 for generating first group voltage commands vd 1*, vq 1*, i.e., first voltage commands represented in a two-phase orthogonal coordinate system and for driving the winding group of the first group, and second group voltage commands vd 2*, vq 2*, i.e., second voltage commands represented in a two-phase orthogonal coordinate system and for driving the winding group of the second group; a carrier generation unit 13 for generating a first carrier signal C 1 and a second carrier signal C 2 on the basis of the first group voltage commands vd 1*, vq 1*, the second group voltage commands vd 2*, vq 2*, and the magnetic pole position signal θe of the rotor of the multiplexed winding electric motor 83;a first two-phase to three-phase conversion unit 12A for converting the first group voltage commands vd 1*, vq 1* into first group to three-phase voltage commands vu 1*, vv 1*, vw 1* represented in a three-phase AC coordinate system; and a first switching pulse generation unit 15A that performs triangular wave comparison processing using a first carrier signal C 1 for the first group to three-phase voltage commands vu 1*, vv 1*, vw 1*, so that first group switching pulses UP 1, UN 1, VP 1, VN 1, WP 1, WN 1, i.e., first switching pulses, for driving the switching elements of the first power converter 81A are generated.The inverter control device 10 also includes: a second two-phase to three-phase conversion unit 12B for converting the second group voltage commands vd 2*, vq 2* for the winding group of the second group into second group to three-phase voltage commands vu 2*, vv 2*, vw 2* represented in a three-phase AC coordinate system; and a second-group switching pulse generation unit 15B that performs triangular-wave comparison processing using a second carrier signal C 2 for the second group three-phase voltage commands vu 2*, vv 2*, vw 2* so that second group switching pulses UP 2, UN 2, VP 2, VN 2, WP 2, WN 2, i.e., second switching pulses, for driving the switching elements of the second power converter 81B are generated. The carrier generation unit 13, the first switching pulse generation unit 15A, and the second switching pulse generation unit 15B constitute a PWM control unit 18.The first group switching pulses UP 1, UN 1, VP 1, VN 1, WP 1, WN 1 respectively drive a positive-side U-phase switching element, a negative-side U-phase switching element, a positive-side V-phase switching element, a negative-side V-phase switching element, a positive-side W-phase switching element, and a negative-side W-phase switching element of the first inverter 81A. Similarly, the second group switching pulses UP 2, UN 2, VP 2, VN 2, WP 2, WN 2 respectively drive a positive-side U-phase switching element, a negative-side U-phase switching element, a positive-side V-phase switching element, a negative-side V-phase switching element, a positive-side W-phase switching element, and a negative-side W-phase switching element of the second inverter 81B.The voltage command generation unit 11 performs a predetermined calculation using first group orthogonal two-phase current commands id 1*, iq 1* and second group orthogonal two-phase current commands id 2*, iq 2* supplied from the outside, first group orthogonal two-phase current values id 1, iq 1, and second group orthogonal two-phase current values id 2, iq 2.Thereby, it generates the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2*. The first group orthogonal two-phase current values id 1, iq 1 and the second group orthogonal two-phase current values id 2, iq 2 are obtained by a first three-phase to two-phase conversion unit 14A and a second three-phase to two-phase conversion unit 14B that respectively convert, into a two-phase orthogonal coordinate system, the first group three-phase current values iu 1, iv 1, iw 1 and the second group three-phase current values iu 2, iv 2, iw 2 that are respectively detected by the first current detector 82A and the second current detector 82B.The magnetic pole position signal θe detected by the position sensor 84 is used as rotor phase information to be used for various calculations by the carrier generation unit 13, the first two-phase to three-phase conversion unit 12A, the second two-phase to three-phase conversion unit 12B, the first three-phase to two-phase conversion unit 14A, and the second three-phase to two-phase conversion unit 14B.FIG. 2 is a block diagram in which the converter control device according to Embodiment 1 is implemented by a processor, and shows components configured for calculation processing. The converter control device 10 includes: a processor 91 for performing various calculations depending on the voltage command generation unit 11, the carrier generation unit 13, the first two-phase to three-phase conversion unit 12A, the second two-phase to three-phase conversion unit 12B, the first three-phase to two-phase conversion unit 14A, the second three-phase to two-phase conversion unit 14B, the first switching pulse generation unit 15A, and the second switching pulse generation unit 15B; a storage device 92 including a read only memory (ROM) 921 and a random access memory (RAM) 922, and storing a control program executed by the processor 91 and data required for various calculations; Input circuits 94A, 94B that receive current information from the first current detector 82A and the second current detector 82B and the magnetic pole position signal θe from the position sensor 84; output circuits 93A, 93B that output switching pulses to the first inverter 81A and the second inverter 81B; and a communication circuit 95 that is connected to a master controller 96 and performs communication with the master controller 96 by serial communication or the like. The functions of the respective functional units described above are achieved by executing the control program on the above hardware.Here, the calculation of each voltage command will be described in more detail. In the calculations of the voltage commands in the converter controller 10, a current-locked loop is formed between the converter controller 10 and the first converter 81A and the second converter 81B, so that the first group orthogonal two-phase current values id 2, iq 2 and the second group orthogonal two-phase current values id 1, iq 1 respectively take the first group orthogonal two-phase current commands id 1*, iq 1* and the second group orthogonal two-phase current commands id 2*, iq 2*. Thereby, the above current values are controlled. A matrix for converting current values represented in a three-phase AC coordinate system to a two-phase orthogonal coordinate system is defined as shown in Expression (1). In Expression (1), id and iq are current values on the d-axis and the q-axis, respectively, i.e., orthogonal two-phase current values. [Mathematical Expression 1]From the expression (1), id and iq are calculated as shown in the following expression (2). [Mathematical Expression 2]In Expression (1) and Expression (2), the phase information θ represents the direction of the magnetic pole position of the d-axis, the d-axis being defined as the direction of the magnetic field flux of the rotor of the electric motor. Note that, if three-phase current values for two phases are used to calculate the three-phase current value for the other phase by using, for example, the equation iw=-iu-iv, it is possible to perform conversion from two three-phase current values to orthogonal two-phase current values. Here, as the phase information θ used for the above coordinate system conversion calculation, the magnetic pole position signal θe for the rotor of the multiplexed winding electric motor 83 is used.The calculations of the above expression (1) and expression (2) are performed by the first three-phase to two-phase conversion unit 14A and the second three-phase to two-phase conversion unit 14B. The first three-phase to two-phase conversion unit 14A and the second three-phase to two-phase conversion unit 14B output the calculation results, i.e., the first group orthogonal to two-phase current values id 2, iq 2 and the second group orthogonal to two-phase current values id 2, iq 2, to the voltage command generation unit 11.The voltage command generation unit 11 receives the first group orthogonal two-phase current commands id 1*, iq 1* and the second group orthogonal two-phase current commands id 2*, iq 2* given from a higher-level controller, the first group orthogonal two-phase current values id 1, iq 1 calculated by the first three-phase to two-phase conversion unit 14A, and the second group orthogonal two-phase current values id 2, iq 2 calculated by the second three-phase to two-phase conversion unit 14B.It then calculates the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* by proportional-integral (PI) control. In the PI control, the calculation of Expression (3) is performed, and thus the voltage commands for controlling the winding group of the first group and the voltage commands for controlling the winding group of the second group are calculated. [Mathematical Expression 3]In Expression (3), s is the differential operator of the Laplace transform, Kp is the proportional gain, and Ki is the integral gain.For inter-group interference prevention processing for suppressing deterioration of the control response due to magnetic interference between the two winding groups of the multiplexed winding electric motor 83, the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* may be calculated by the following method: The sums and the differences of the currents flowing through the winding group of the first group and the currents flowing through the winding group of the second group are calculated. Thereby, voltage commands for both the first group and the second group are calculated simultaneously.In this method, by the PI control, as shown in Expression (4) and Expression (5), the sums of the voltage commands for both groups are calculated from the sums of the first group current commands id 1*, iq 1* and the second group current commands id 2*, iq 2*, and the differences between the voltage commands for both groups are calculated from the differences between the first group current commands id 1*, iq 1* and the second group current commands id 2*, iq 2*.In the expression (4), vdsum* is the sum of the voltage commands for both groups in the d-axis direction, and vddiff* is the difference between the voltage commands for both groups in the d-axis direction. In Expression (5), vqsum* is the sum of the voltage commands for both groups in the q-axis direction, and vqdiff* is the difference between the voltage commands for both groups in the q-axis direction. [Mathematical Expression 4] [Mathematical Expression 5]After vdsum*, vddiff*, vqsum*, and vqdiff* are calculated, calculation is performed from Expression (6), thus separating the sums vdsum* and vqsum* of the voltage commands and the differences vddiff* and vqdiff* between the voltage commands into the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2*, which are the voltage commands for the respective winding groups. [Mathematical Expression 6]When the PI control is performed with Expression (4) and Expression (5), four voltage commands vd 1*, vq 1*, vd 2*, vq 2* are calculated using four orthogonal two-phase current values id 1, iq 1, id 2, iq 2. In this case, the first group voltage commands vd 1*, vq 1* are influenced by the values of the three-phase currents flowing through the winding group of the second group, and the second group voltage commands vd 2*, vq 2* are influenced by the values of the three-phase currents flowing through the winding group of the first group.Therefore, it is necessary to update the voltage commands for the first group and the second group simultaneously. Note that the inter-group interference prevention processing may be performed by a different calculation method instead of a calculation method using the sums and differences.After the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* are calculated in the voltage command generation unit 11, the first two-phase to three-phase conversion unit 12A and the second two-phase to three-phase conversion unit 12B respectively convert the first group voltage commands vd 1*, vq 1* into the first group to three-phase voltage commands vu 1*, vv 1*, vw 1* and convert the second group voltage commands vd 2*, vq 2* into the second group to three-phase voltage commands vu 2*, vv 2*, vw 2*. The coordinate conversion for converting the voltage commands in a two-phase orthogonal coordinate system into a three-phase AC voltage command as described above is defined as shown in the expression (7). [Mathematical Expression 7]With the expression (7), the calculation shown in the expression (8) is performed. [Mathematical Expression 8]In Embodiment 1, the magnetic pole position signal θe is used as the phase information θ in the coordinate conversion with Expression (7) and Expression (8), and a predetermined phase difference is set between the magnetic pole position used in the calculation by the first two-phase-to-three-phase conversion unit 12A and the magnetic pole position used in the calculation by the second two-phase-to-three-phase conversion unit 12B. The phase difference is based on a phase difference of the electrical angle between the winding group of the first group and the winding group of the second group.After the calculation of Expression (8) is performed, the first two-phase to three-phase conversion units 12A and the second two-phase to three-phase conversion units 12B output the first group to three-phase voltage commands vu 1*, vv 1*, vw 1*, and the second group to three-phase voltage commands vu 2*, vv 2*, vw 2* to the first switching pulse generation unit 15A and the second switching pulse generation unit 15B, respectively. In addition, the first carrier signal C 1 and the second carrier signal C 2 from the carrier generation unit 13 are also input to the first switching pulse generation unit 15A and the second switching pulse generation unit 15B, respectively.The first switching pulse generation unit 15A performs triangular wave comparison processing using the first carrier signal C 1 for the first group three-phase voltage commands vu 1*, vv 1*, vw 1*, and thus generates the first group switching pulses UP 1, UN 1, VP 1, VN 1, WP 1, WN 1. The second-group switching pulse generation unit 15B performs triangular-wave comparison processing using the second carrier signal C 2 for the second group three-phase voltage commands vu 2*, vv 2*, vw 2*, and thus generates the second group switching pulses UP 2, UN 2, VP 2, VN 2, WP 2, WN 2.The switching pulses generated by the first switching pulse generation unit 15A and the second switching pulse generation unit 15B are provided with times for preventing short-circuiting of the upper and lower arms of the inverter, and then output to the first inverter 81A and the second inverter 81B, so that the respective switching elements of the first inverter 81A and the second inverter 81B are operated.Next, the carrier generation unit 13 will be described in more detail. Fig. 3 is a detailed block diagram showing the configuration of the carrier generation unit shown in Fig. 1. In the carrier generation unit 13, the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* are input to a first group voltage command input terminal 131A and a second group voltage command input terminal 131B, respectively, and the magnetic pole position signal θe is input to a magnetic pole position signal input terminal 132.The first group voltage commands vd 1*, vq 1* input to the carrier generation unit 13 are input to a first control phase angle calculation unit 133A, and the first control phase angle calculation unit 133A calculates a first control phase angle THV 1 from the first group voltage commands vd 1*, vq 1*. The first control phase angle THV 1 and the magnetic pole position signal θe are added by a first adder 134A, and the output θthv 1 from the first adder 134A is input to the first phase correction unit 135A.Meanwhile, the second group voltage commands vd 2*, vq 2* input to the carrier generation unit 13 are input to the second control phase angle calculation unit 133B, and the second control phase angle calculation unit 133B calculates a second control phase angle THV 2 from the second group voltage commands vd 2*, vq 2*. The second control phase angle THV 2 and the magnetic pole position signal θe are added by a second adder 134B, and the output θthv 2 from the second adder 134B is input to the second phase correction unit 135B.The first phase correction unit 135A causes the phase of the output θthv 1 of the first adder 134A to lag 90° and performs predetermined correction thereon. The result is output as a first voltage phase θv 1 to a first sync carrier generation unit 136A. The second phase correction unit 135B causes the phase of the output θthv 2 of the second adder 134B to lag 90° and corrects the phase by an electrical angle phase difference θoffset between the winding group of the first group and the winding group of the second group. The result is output as a second voltage phase θv 2 to a second sync carrier generation unit 136B.The first synchronous carrier generation unit 136A generates a first synchronous carrier signal synchronized with the first group voltage commands vd 1*, vq 1* using the first voltage phase θv 1, and outputs the first synchronous carrier signal to a first selector 138A and a second selector 138B. The second synchronous carrier generation unit 136B generates a second synchronous carrier signal synchronized with the second group voltage commands vd 2*, vq 2* using the second voltage phase θv 2, and outputs the second synchronous carrier signal to the second selector 138B. In addition, the asynchronous carrier generation unit 137 generates an asynchronous carrier signal formed of constant frequency triangular waves, and outputs the asynchronous carrier signal to the first selector 138A and the second selector 138B.The first selector 138A outputs a carrier signal selected from the two carrier signals (the first synchronous carrier signal and the asynchronous carrier signal) as the first carrier signal C 1 to a first output terminal 139A. The first output terminal 139A outputs the first carrier signal C 1 to the first switching pulse generation unit 15A. The second selector 138B outputs a carrier signal selected from among the three carrier signals (the first synchronous carrier signal, the second synchronous carrier signal, and the asynchronous carrier signal) as the second carrier signal C 2 to a second output terminal 139B. The second output terminal 139B outputs the second carrier signal C 2 to the second switching pulse generation unit 15B. Selection of the carrier signals in the first selector 138A and the second selector 138B will be described later.Next, the operation of the carrier generation unit 13 will be described in more detail. In the case where the synchronous carrier signals to be used for PWM control for the winding group of the first group and the winding group of the second group are generated, the first voltage phase θv 1 is generated from the first control phase angle THV 1 and the magnetic pole position signal θe, and the second voltage phase θv 2 is generated from the second control phase angle THV 2 and the magnetic pole position signal θe.In the case where the synchronous carrier signal to be used only for the PWM control for the winding group of the first group is generated and in the case where the synchronous carrier signal to be used commonly for the PWM control for both winding groups is generated, the voltage phase therefor is calculated from the first control phase angle THV 1 and the magnetic pole position signal θe, and the voltage phase for the synchronous carrier signal to be used for the PWM control for the winding group of the second group is calculated from the second control phase angle THV 2 and the magnetic pole position signal θe.The first control phase angle calculation unit 133A calculates the first control phase angle THV 1 by calculation of Expression (9) using the first group voltage commands vd 1*, vq 1*. [Mathematical Expression 9]The second control phase angle calculation unit 133B calculates the second control phase angle THV 2 by calculation of Expression (10) using the second group voltage commands vd 2*, vq 2*. [Mathematical Expression 10]The first voltage phase θv 1 is calculated from Expression (11) using the magnetic pole position signal θe and a phase adjustment value α by the first adder 134A and the first phase correction unit 135A. [Mathematical Expression 11]The setting with the phase setting value α in the first phase correction unit 135A in the calculation by Expression (11) serves to make a correction of the shift of the phase of a common synchronous carrier signal in the case where the carrier signal synchronized with the first group voltage commands vd 1*, vq 1* is used as the common synchronous carrier signal for the synchronous PWM control for the winding group of the first group and the winding group of the second group.With such correction, a specific frequency component included in the second group switching pulses generated by the second group switching pulse generation unit 15B can be adjusted, so that the magnitudes of the fundamental waves included in the first group switching pulse and the second group switching pulse can be made equal to each other.For example, in the case where the phase difference θoffsetof the electrical angle between the winding group of the first group and the winding group of the second group is 30°, the phase adjustment value α is set to 15° (1 / 2 of the phase difference). Note that such correction processing is performed as necessary, and is not essential in the present invention.The second voltage phase θv 2 is calculated from Expression (12) using the magnetic pole position signal θe and the electrical angle phase difference θoffset between the winding group of the first group and the winding group of the second group by the second adder 134B and the second phase correction unit 135B. Therefore, setting as in the above-described setting with the phase setting value α is not performed in the calculation of the second voltage phase θv 2. [Mathematical Expression 12]The carrier signals generated on the basis of the first voltage phase θv 1 and the second voltage phase θv 2 calculated as described above are synchronized with the first group three-phase voltage commands vu 1*, vv 1*, vw 1* and the second group three-phase voltage commands vu 2*, vv 2*, vw 2*, respectively.The first synchronous carrier generation unit 136A generates the first synchronous carrier signal synchronized with the first group three-phase voltage commands vu 1*, vv 1*, vw 1* on the basis of the first voltage phase θv 1 output from the first phase correction unit 135A. The synchronous carrier signal is a triangular wave having a frequency that is, for example, three, six, nine, or fifteen times the frequency of the voltage commands. By performing triangular wave comparison using the above carrier signal, the number of pulses in the switching pulses per one period of the voltage commands can be controlled.The first synchronous carrier generation unit 136A in Embodiment 1 outputs a synchronous carrier signal in which the number of pulses is set to any number of pulses, such as nine, six, three, and the like, according to the rotational speed of the multiplexed winding electric motor 83, the frequency of the voltage commands, and the modulation factor of the voltage commands.As described above, there is a case where the first synchronous carrier signal generated by the first synchronous carrier generation unit 136A is used only for the PWM control of the winding groups of the first group, and a case where the first synchronous carrier signal is used in common for the PWM control of the winding groups of the first group and the second group. Similarly, the second synchronous carrier generation unit 136B generates the second synchronous carrier signal synchronized with the second group three-phase voltage commands vu 2*, vv 2*, vw 2* on the basis of the second voltage phase θv 2 output from the second phase correction unit 135B.The asynchronous carrier generation unit 137 generates an asynchronous carrier signal which is a triangular wave having a constant frequency of, for example, 10 kHz and which is not synchronized with the voltage commands.The first selector 138A selects one of the first synchronous carrier signal and the asynchronous carrier signal, and outputs the selected carrier signal as the first carrier signal C 1. The first carrier signal C 1 is output to the first switching pulse generation unit 15A via the first output terminal 139A. In the selection for the first carrier signal C 1, for example, when it is determined that the operating state of the multiplexed winding electric motor 83 is in a low-speed range, the asynchronous carrier signal is selected, and when it is determined that the operating state is in a medium-speed range or in a high-speed range, the first synchronous carrier signal is selected.The second selector 138B selects one of the first synchronous carrier signal, the second synchronous carrier signal, and the asynchronous carrier signal, and outputs the selected carrier signal as the second carrier signal C 2. The second carrier signal C 2 is output to the second switching pulse generation unit 15B via the second output terminal 139B. In the selection for the second carrier signal C 2, for example, when it is determined that the operation state of the multiplexed winding electric motor 83 is in a low-speed range, the asynchronous carrier signal is selected. When it is determined that the operating state is in a medium speed range, the first synchronous carrier signal is selected. When it is determined that the operating state is in a high-speed range, the second synchronous carrier signal is selected.In the present invention, the control mode in which the first selector 138A and the second selector 138B select the asynchronous carrier signal is referred to as "asynchronous PWM control mode"; the control mode in which the first selector 138A and the second selector 138B select the carrier signal (in Embodiment 1: the first synchronous carrier signal) synchronized with the voltage commands for a specific winding group is referred to as "first synchronous PWM control mode"; and the control mode in which the first selector 138A and the second selector 138B select individual carriers (the first synchronous carrier signal and the second synchronous carrier signal) synchronized with the respective voltage commands is referred to as "second synchronous PWM control mode".The determination as to which range the operating state of the multiplexed winding electric motor 83 belongs to may be performed based on the rotational speed of the multiplexed winding electric motor 83, for example. More specifically, if two threshold values are set, then if the rotational speed of the multiplexed winding electric motor 83 is less than the smaller threshold value, then it is determined that the operating state is in the low-speed range. If the speed is not less than the lower threshold but less than the higher threshold, then it is determined that the operating state is in the medium speed range. If the rotational speed is not less than the larger threshold, it is determined that the operating state is in the high-speed range.With this configuration, the carrier generation unit 13 can selectively output the carrier signal according to the operation state of the electric motor 83 with multiplex winding, under control of the first inverter 81A and the second inverter 81B. The operating state may be determined based on the frequency or the modulation factor of the voltage commands. The determination base or thresholds may be the same for the first selector 138A and the second selector 138B, or may be different therebetween.Generation of switching pulses by the triangular wave comparison processing will now be described. FIG. 4 is a timing chart showing the relationship among the voltage commands for the respective winding groups, the asynchronous carrier signal, and the switching pulses for the respective winding groups in Embodiment 1, and shows the switching pulses generated by triangular wave comparison in the asynchronous PWM control.In FIG. 4, the relationship between the first group U-phase voltage command vu 1*, the second group U-phase voltage command vu 2*, the asynchronous carrier signal, the first group U-phase switching pulse, and the second group U-phase switching pulse is shown, and the phase difference between the voltage commands is set to 30° on the basis of the electrical phase difference between the winding group of the first group and the winding group of the second group in the multiplexed winding electric motor 83. In the asynchronous PWM control shown in FIG. 4, the first group U-phase switching pulse is generated based on the comparison of the strength between the first group U-phase voltage command vu 1* and the asynchronous carrier signal at each time point.In the case that the first group U-phase voltage command vu 1* is larger, the first group U-phase switching pulse is turned on ("ON"), and in the case that the asynchronous carrier signal is larger, the first group U-phase switching pulse is turned off ("OFF"). Similarly, the second group switching pulse is determined based on the comparison of the strength between the second group U-phase voltage command vu 2* and the asynchronous carrier signal at each time point. Note that the phase difference between the first group and the second group shown in FIG. 4 is merely an example, and is not limited thereto.Such a switching pulse is generated for each of the U-phase, the V-phase, and the W-phase of the three-phase voltage commands. However, the switching pulses for the V phase and the W phase have different phases relative to the switching pulse for the U phase merely as a whole, and are the same except for this. Therefore, the description thereof is omitted. The feature that the switching pulse is generated based on the ratio of the magnitudes between the carrier signal and the voltage command as described above also applies to the case of the synchronous PWM control described below.FIG. 5 is a timing chart showing the relationship between the voltage commands for the respective winding groups and the carrier signals for the respective winding groups in the 9-pulse synchronous PWM control in Embodiment 1, and shows the relationship between the first group U-phase voltage command vu 1*, the second group U-phase voltage command vu 2*, the first group synchronous carrier signal, the second group synchronous carrier signal, and the common synchronous carrier signal. The first group U-phase voltage command vu 1*, the first group synchronous carrier signal, and the synchronous carrier signal, which are commonly used for both winding groups, are synchronized with the first voltage phase θv 1.The second group U-phase voltage command vu 2* and the second group synchronous carrier signal are synchronized with the second voltage phase θv 2. In the 9-pulse synchronous PWM control, a synchronous carrier signal having a frequency nine times the frequency of the first voltage command and synchronized with the first voltage phase θv 1 is generated, so that nine on / off switching operations are performed per one period of the first voltage phase θv 1 and the second voltage phase θv 2.In the case where the electrical phase difference between the winding group of the first group and the winding group of the second group is 30° in the control of the multiplexed winding electric motor 83, and the first group voltage command and the second group voltage command are in the same state, if a carrier signal synchronized with the first group U-phase voltage command vu 1* and a carrier signal synchronized with the second group U-phase voltage command vu 2* are independently generated, the first group synchronous carrier signal and the second group synchronous carrier signal have a carrier phase difference of 90° as shown in FIG. 5, with a carrier period defined as 360°.The calculation of the voltage commands for both winding groups in the PI control including the inter-group interference prevention processing needs to complete coordinate conversion from a two-phase orthogonal coordinate system to a three-phase coordinate system and coordinate conversion from a three-phase coordinate system to a two-phase orthogonal coordinate system within a period T 9 scorresponding to the interval between a peak of the first group synchronous carrier signal and a peak of the second group synchronous carrier signal. The period T 9 sis determined by the phase difference between the two carrier signals, and the phase difference between the carrier signals varies when the voltage command for one winding group is updated.Again, in the case where the synchronous PWM control is performed using the common synchronous carrier signal without using the two carrier signals, the voltage command calculation and the like need only be completed between a peak and a bottom of the common synchronous carrier signal, and since the period for calculating the voltage commands is not affected by the phase difference between the two carrier signals, the period for calculating the voltage commands does not vary greatly even when the voltage commands are updated.That is, if the common synchronous carrier signal is used, variation in the time period for calculating the voltage commands can be suppressed, and in performing the inter-group interference prevention processing using the sums and the differences of the currents of both winding groups, the voltage command calculation can be stably performed. As the common synchronous carrier signal, either the carrier signal synchronized with the first group voltage command or the carrier signal synchronized with the second group voltage command may be used. In Embodiment 1, the carrier signal synchronized with the first group voltage command is used.In the case where the first group synchronous carrier signal is used as the common synchronous carrier signal for synchronous PWM control for the winding group of the second group, it is desirable that the first group synchronous carrier signal for which the phase is corrected according to the value of the phase difference between the first group voltage command and the second group voltage command be used. For example, in Embodiment 1, the first group synchronous carrier signal for which the phase is shifted by 15°, which corresponds to 1 / 2 of the phase difference θoffset between the winding groups of the first group and the second group, is used as the common synchronous carrier signal.As described above, the first synchronous carrier signal corrected by a phase correction value determined based on the phase difference of the electrical angle between the winding group of the first group and the winding group of the second group is used as the common synchronous carrier signal, so that the content rate of the fundamental wave component included in the switching pulses by the synchronous PWM control for the first group and the content rate of the fundamental wave component included in the switching pulses by the synchronous PWM control for the second group become equal to each other, and thus it becomes possible to perform more stable control.Note that the synchronous PWM control can be performed using the common synchronous carrier signal even without performing the phase correction as described above. In the case where there is no electrical phase difference between the winding group of the first group and the winding group of the second group, the first group synchronous carrier signal may be used as the common synchronous carrier signal without performing phase correction as described above.FIG. 6 is a timing chart showing the relationship between the voltage commands for the respective winding groups and the carrier signals for the respective winding groups in 6-pulse synchronous PWM control in Embodiment 1. In the case where the electrical phase difference between the winding group of the first group and the winding group of the second group is 30°, and the first group U-phase voltage command vu 1* and the second group U-phase voltage command vu 2* are in the same state, the first group synchronous carrier signal and the second group synchronous carrier signal have a carrier phase difference of 180°, as shown in FIG. 6. Here, the period T 6 sis the interval between a peak of the first group synchronous carrier signal and a peak of the second group synchronous carrier signal, and is a period for calculating the voltage commands as in the case of 9 pulses.In the case of 6-pulse synchronous PWM control, when the calculation is performed in synchronization with the carrier signal, for each winding group, if there is a carrier phase difference of 180°, the timing of a peak (a bottom) of the first group synchronous carrier signal and the timing of a bottom (a peak) of the second group synchronous carrier signal coincide with each other. Consequently, acquisition of the current information and update of the voltage commands are performed twice in a short time, so that the voltage command calculation could not be performed correctly.Therefore, even in the case of the 6-pulse synchronous PWM control, the synchronous PWM control is performed using the common synchronous carrier signal, so that the voltage command update and the like are prevented from being performed twice. In the case where the synchronous PWM control for the second group is performed using the first group synchronous carrier signal, it is desirable that the first group synchronous carrier signal for which the phase is shifted by 15° is used as the common synchronous carrier signal.This serves to equalize the content rates of the fundamental wave components in the first group switching pulse and the second group switching pulse, as in the case of 9 pulses described in Fig. 5. In the case where there is no electrical phase difference between the winding group of the first group and the winding group of the second group, the first group synchronous carrier signal may be used as the common synchronous carrier signal without performing phase correction as described above.FIG. 7 is a timing chart showing the relationship between the voltage commands for the respective winding groups and the carrier signals for the respective winding groups in 3-pulse synchronous PWM control in Embodiment 1. In the case of 3 pulses, the first group synchronous carrier signal and the second group synchronous carrier signal have a carrier phase difference of 90° as in the case of 9 pulses. Meanwhile, in the 3-pulse synchronous PWM control, as compared with the case of 9 pulses and 6 pulses, the period of the carrier signal becomes longer, and the time period T 3 sfor calculating the voltage commands corresponding to the interval between a peak of the first group synchronous carrier signal and a peak of the second group synchronous carrier signal is sufficiently long.Therefore, even if the length of T 3 schanges due to voltage command update, a sufficiently long calculation time for the voltage command calculation, the coordinate conversion, and the inter-group interference prevention processing can be obtained. Therefore, in the 3-pulse synchronous PWM control, even when the voltage command calculation is performed according to two timings of the carrier signal synchronized with the first group voltage command and the carrier signal synchronized with the second group voltage command, the calculation time does not become insufficient, and it is possible to stably control the multiplexed winding electric motor 83.FIG. 8 shows processing timings in the case where voltage command calculation is performed between the peaks and the bottoms of the carrier signals for both winding groups in the synchronous PWM control in Embodiment 1, and shows the case of the 9-pulse synchronous PWM control. In the 9-pulse synchronous PWM control, the first carrier signal C 1 to be used for generating the first group switching pulse and the second carrier signal C 2 to be used for generating the second group switching pulse have a carrier phase difference of 90°.In FIG. 8, first, at the timing of a peak of the second carrier signal C 2, the second group voltage command is updated, and simultaneously a phase current value of the current flowing through the winding group of the second group is detected, so that the next calculation of the second group voltage command is started (including the accompanying coordinate conversion calculation and the like; the same applies hereinafter).Next, at the timing of a peak of the first carrier signal C 1, the first group voltage command is updated, and simultaneously a phase current value of the current flowing through the winding group of the first group is detected, so that the next calculation of the first group voltage command is started. Subsequently, at the timing of a bottom of the second carrier signal C 2 and then at the timing of a bottom of the first carrier signal C 1, voltage command update and detection of the phase current value (start of voltage command calculation) are repeated for the respective winding groups.As described above, in the case where the PI feedback control using the sum and the difference of the current values is used for the inter-group interference prevention processing in the voltage command calculation for both the winding groups of the first group and the second group, the voltage command update for the one winding group affects the voltage command calculation for the other winding group. Therefore, it is necessary that the voltage command calculation, the coordinate conversion calculation, and the like are completed within a calculation time T 1 (equal to T 9 sin the case of 9 pulses) between the timings of the peaks and bottoms of the first carrier signal C 1 and the peaks and bottoms of the second carrier signal C 2.However, as described above, when the current control is performed, the voltage command varies, and when the voltage command varies, the voltage phase also varies. Accordingly, the synchronous carrier signal synchronized with the voltage phase also varies due to the current control. The phase variation in the synchronous carrier signal causes the voltage command calculation time to vary.For example, if the second carrier signal C 2 becomes a carrier signal as shown by C 2* in FIG. 8, the voltage command calculation time T 1 varies and shortens to T 1*, so that there is a possibility that the calculation cannot be completed to the end. Therefore, in the case where it is expected that the calculation time cannot be sufficiently secured, it is necessary to suppress the variation of the calculation time.In addition, even in the case where the inter-group interference prevention processing using the current values of the first group and the second group is not performed, if there is a circuit that operates in coordination with the calculation of voltage commands for the first group and the second group, it is necessary to finish the voltage command calculation for the one winding group before starting the voltage command calculation for the other winding group. In the configuration shown in FIG. 1, for example, it is assumed that in the analog-to-digital (A / D) conversion of data of the phase current, a plurality of current values are subjected to A / D conversion by time-domain multiplexing using a single A / D converter.In this case, if the timings for detecting phase currents of the respective winding groups are synchronized with the respective carrier signals, there is a possibility that the A / D conversion cannot be completed in time depending on the phase relationship between the first carrier signal C 1 and the second carrier signal C 2. This applies not only to the case of A / D conversion, but also to the case where a memory for a look-up table (LUT) or a calculation module for trigonometric functions are shared, for example.FIG. 9 shows calculation processings in the case where voltage command calculation is performed between a peak and a bottom of the carrier signal for one winding group in the synchronous PWM control in Embodiment 1, and shows the operation synchronized with the timings of a peak and a bottom of the carrier signal in the case where the PWM control is performed using the common synchronous carrier signal. In FIG. 9, the voltage command update and the current value detection (start of voltage command calculation) are performed for both the winding group of the first group and the second group at timings based on the first carrier signal C 1.Therefore, the starts and the ends of the voltage commands are synchronized with a single carrier signal. As a result, the variation of the voltage command calculation time Tca is suppressed, and the voltage command calculation can be stably performed. Similarly, even in the case where an AD converter, a memory for a look-up table (LUT) of a trigonometric function, or the like is shared in the calculation for the first group and the second group, the voltage command calculation can be stably performed.FIG. 10 shows processing timings in a case where, unlike FIG. 8, voltage command calculation is performed between the peaks of the carrier signals for both winding groups in the synchronous PWM control in Embodiment 1, and shows timings of current value acquisition (start of voltage command calculation) and voltage command update for both winding groups in a case where the voltage command calculations for both the winding group of the first group and the second group are synchronized with the peaks of the respective carrier signals.In the case where the PI control is performed using the sums and differences of the current values for the voltage command calculation for both the winding group of the first group and the second group for the purpose of the inter-group interference prevention processing, it is necessary to complete the calculations for the voltage commands, the coordinate conversion, and the like between the timing of a peak of the first carrier signal C 1 and the timing of a peak of the second carrier signal C 2.As in the case of FIG. 9, when current control is performed, the synchronous carrier synchronized with the voltage command also varies. When the second carrier signal C 2 becomes a carrier signal as shown by C 2*, the voltage command calculation time T 1 ashown in FIG. 10 varies and the voltage command calculation time is decreased to T 1 a*, so that there is a possibility that the voltage command calculation cannot be completed to the end.FIG. 11 shows processing timings in the case where voltage command calculation is performed between the peaks of the carrier signal for a single winding group in synchronous PWM control in Embodiment 1, and shows timings of current value detection (start of voltage command calculation) and voltage command update for both winding groups in the case where calculations of the first group voltage command and the second group voltage command are synchronized with a peak of the common synchronous carrier signal.In this case, since the phase voltage command update and current value detection (start of voltage command calculation) are performed for both the winding group of the first group and the second group in synchronization with a peak of the first carrier signal C 1, the variation of the voltage command calculation time TCb is suppressed, and the voltage command calculation can be stably performed.FIG. 12 shows the relationship between the operating state of the multiplexed winding electric motor and the type of the carrier signal used in the PWM control in Embodiment 1. In Embodiment 1, the operation state is determined based on the rotational speed of the multiplexed winding electric motor 83.In the case where the rotation speed is low and the operation state is in a low rotation speed range, the control mode is set to the "asynchronous PWM control mode". The first selector 138A and the second selector 138A select the asynchronous carrier signal, and asynchronous PWM control for generating switching pulses using the common asynchronous carrier signal is performed for both the first group and the second group. Voltage command computations are performed in synchronization with the timings of the peaks and bottoms of the lower asynchronous carrier signal. Therefore, the calculation times are constant, and it is possible to stably perform the voltage command calculation including the calculation for the inter-group interference prevention processing.In the case where the rotation speed is medium speed and the operation state is in a medium speed range, the 9-pulse synchronous PWM control or the 6-pulse synchronous PWM control are performed. The control mode is set to the "first synchronous PWM control mode". When the number of pulses is large, the interval between a peak and a ground or between a peak and a peak of a synchronous carrier signal is shortened. Therefore, the first selector 138A and the second selector 138B commonly select the first synchronous carrier signal, and synchronous PWM control for generating switching pulses using the same synchronous carrier signal is performed. As a result, the variation of the voltage command calculation time is suppressed, and the voltage command calculation can be accurately performed.In the case where the rotation speed is further increased and the operation state is in a high rotation speed range, the 3-pulse synchronous PWM control is performed, and the control mode is set to the "second synchronous PWM control mode". The first selector 138A and the second selector 138B select the first synchronous carrier signal and the second synchronous carrier signal, respectively. In this case, the first carrier signal C 1 is synchronized with the first group voltage command, and the second carrier signal C 2 is synchronized with the second group voltage command. Consequently, synchronous carrier signals independent of each other are used for generating the switching pulses for the first group and the second group.As described above, when it is determined that the rotation speed is lower than a predetermined rotation speed and the operation state is in a low rotation speed range, the asynchronous PWM control is selected, and when it is determined that the rotation speed is higher than the predetermined rotation speed and the operation state is in a medium rotation speed range or a high rotation speed range, the first synchronous PWM control mode or the second synchronous PWM control mode is selected.In this example, the control mode is switched based on the rotational speed. However, the above procedure is applicable also to the case where the control mode is switched on the basis of the modulation factor or the frequency of the voltage commands.In the 9-pulse synchronous PWM control in the low-speed range and the medium-speed range, the current control is performed by PI control as shown in Expression (3) to Expression (5), but in the 6-pulse synchronous PWM control in the medium-speed range and the high-speed range, the driving and the control of the multiplexed winding electric motor 83 can also be performed by directly inputting the voltage commands without performing the current control. In this case, the directly input voltage command is compared with the carrier signal and the switching pulse is generated. In addition, in the case where the current control is not performed, the voltage command does not vary greatly, and the interval between a peak and a bottom of the synchronous carrier signal does not vary greatly. Therefore, it is possible to stably perform calculation for coordinate conversion and the like.According to Embodiment 1, the time for calculating the voltage commands for driving the multiplexed winding electric motor is secured, and the multiplexed winding electric motor can be stably controlled. More specifically, as carrier signals for generating switching pulses for driving switching elements of the converter, the first synchronous carrier signal synchronized with the first group voltage command, the second synchronous carrier signal synchronized with the second group voltage command, and the asynchronous carrier signal are generated, and carrier signals of types corresponding to the operation state of the multiplexed winding electric motor are selectively output as the first carrier signal and the second carrier signal.In this way, in the low-speed range in which the asynchronous PWM control is to be performed, switching pulses for the first group and the second group are generated using the common asynchronous carrier signal for the first group and the second group. In the medium speed range in which there is a possibility that the voltage command calculation time becomes insufficient when the inter-group interference prevention processing is performed in the synchronous PWM control, switching pulses for the first group and the second group are generated using the common synchronous carrier signal for the first group and the second group. Thereby, the variation of the voltage command calculation time is suppressed.Consequently, it is possible to accurately perform the voltage command calculation. In the high-speed range in which the period of the synchronous carrier signal for each group is long and there is no possibility that the voltage command calculation time becomes insufficient, switching pulses for the respective groups are generated using independent synchronous carrier signals for the respective groups. Accordingly, in all the operating states in the low-speed range, the medium-speed range, and the high-speed range, the voltage command calculation time is secured, and the multiplexed winding electric motor can be stably controlled.Embodiment 2Hereinafter, Embodiment 2 of the present invention will be described with reference to FIG. 13. In Embodiment 2, the common synchronous carrier signal to be used for synchronous PWM control is generated by being calculated on the basis of the voltage commands for both winding groups, unlike Embodiment 1. FIG. 13 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 2. Components identical or corresponding to those in FIG. 3 are provided with the same reference numerals. In the following description, the components different from FIG. 3 will be mainly described.In a carrier generation unit 23, a first phase correction unit 235A is a phase correction circuit that calculates a first voltage phase θv 1 using: an output θthv 1 of the first adder 134A, a first control phase angle THV 1 output from the first control phase angle calculation unit 133A, and a second control phase angle THV 2 output from the second control phase angle calculation unit 133B. It outputs the first voltage phase θv 1. The other components are the same as in Embodiment 1, and therefore, the description thereof is omitted.Next, the operation will be described. In the first phase correction unit 235A, a voltage phase correction value Δθ is calculated from the difference between the first control phase angle THV 1 and the second control phase angle THV 2, and the output θthv 1 of the first adder 134A is corrected by the voltage phase correction value Δθ. Thereby, the first voltage phase θv 1 is calculated.More specifically, in the calculation of the voltage phase for generating the synchronous carrier signal to be used in the synchronous PWM control commonly for the first group and the second group, the phase correction is performed, in addition to using the electrical phase difference between the winding group of the first group and the winding group of the second group, on the basis of the voltage phase correction value Δθ representing the phase difference between the first group voltage command and the second group voltage command, for example, using Δθ / 2 as the phase correction value.In addition, even in the case where the synchronous PWM control is independently performed for each of the first group and the second group by selecting the first synchronous carrier signal as the first carrier signal C 1 and selecting the second synchronous carrier signal as the second carrier signal C 2, it is possible to perform the correction based on the voltage phase correction value Δθ in the calculation of the first voltage phase θv 1.In this case, limit value processing may be performed using a maximum correction value Δθmax such that the first voltage phase θv 1 does not deviate from the second voltage phase θv 2 by a certain value or greater. That is, if the voltage phase correction value Δθ is equal to or less than the maximum correction value Δθmax, correction is performed using the voltage phase correction value Δθ, and if the voltage phase correction value Δθ is greater than the maximum correction value Δθmax, correction is performed using the maximum correction value Δθmax.In the case where the first voltage phase θv 1 calculated under the above limit processing is used, the relative phase ratio between the first carrier signal C 1 and the second carrier signal C 2 is maintained within a certain value, and thus the calculation time for performing the current control in the synchronous PWM control is stably ensured. Note that the same effect can be obtained even in the case where the phase correction based on the voltage phase correction value Δθ is performed in the calculation of the second voltage phase θv 2 as described above.The other components are the same as those in Embodiment 1, and therefore, the description thereof is omitted.In Embodiment 2, the same effect as in Embodiment 1 can be obtained.In addition, the first voltage phase or the second voltage phase is calculated using a phase correction value based on the difference between the first control phase angle and the second control phase angle. Therefore, the relative phase relationship between the first synchronous carrier signal and the second synchronous carrier signal can be maintained, and thus the plurality of converters for supplying AC power to the multiplexed electric motor can be stably controlled.Embodiment 3Hereinafter, Embodiment 3 of the present invention will be described with reference to FIGS. 14 to 19. In Embodiment 3, the second synchronous carrier signal is generated on the basis of the first voltage phase, so that the second carrier signal is also synchronized with the first voltage phase in the synchronous PWM control. FIG. 14 is a block diagram showing the configuration of a converter control device according to Embodiment 3. Components identical or corresponding to those in FIG. 1 are provided with the same reference numerals. In the following description, the components different from FIG. 1 will be mainly described. A converter control device 30 is different from Embodiment 1 in a carrier generation unit 33.The carrier generation unit 33 receives the first group voltage commands vd 1*, vq 1* from a voltage command generation unit 31 and the magnetic pole position signal θe from the position sensor 84, and outputs a first carrier signal C 1A to the first switching pulse generation unit 15A and a second carrier signal C 1B to the second switching pulse generation unit 15B. The carrier generation unit 33, the first switching pulse generation unit 15A, and the second switching pulse generation unit 15B constitute a PWM control unit 38.FIG. 15 is a detailed block diagram showing the configuration of the carrier generation unit according to Embodiment 3. Components identical or corresponding to those in FIG. 3 are provided with the same reference numerals. In the following description, the components different from FIG. 3 will be mainly described. In the carrier generation unit 33, a second phase correction unit 335B calculates a second voltage phase θv 2* from the first group voltage commands vd 1*, vq 1* and the magnetic pole position signal θe, and outputs the second voltage phase θv 2* to a second synchronous carrier generation unit 336B. The second synchronous carrier generation unit 336B generates a second synchronous carrier signal using the second voltage phase θv 2*.If the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* are equal to each other, the first control phase angle THV 1 and the second control phase angle THV 2 are the same angle. Therefore, assuming that the voltage commands for the first group and the second group are not greatly different from each other, the second voltage phase θv 2* is calculated by calculating the following expression (13) using the phase difference θoffset of the electrical angle between the winding group of the first group and the winding group of the second group. [Mathematical Expression 13]The second synchronous carrier generation unit 336B generates a second synchronous carrier signal using the second voltage phase θv 2* calculated as shown in Expression (13), and outputs the second synchronous carrier signal to the second selector 138B. In this case, the phase of the second synchronous carrier signal becomes a phase obtained by correcting the first control phase angle THV 1 by a phase correction value determined on the basis of the phase difference θoffset of the electrical angle between the winding group of the first group and the winding group of the second group.The first selector 138A selects one of the first synchronous carrier signal received from the first synchronous carrier generation unit 136A and the asynchronous carrier signal received from the asynchronous carrier generation unit 137, and outputs the selected carrier signal as the first carrier signal C 1A. The first carrier signal C 1A is output to the first switching pulse generation unit 15A via a first output terminal 339A.The second selector 338B selects one of the second synchronous carrier signal received from the second synchronous carrier generation unit 336B and the asynchronous carrier signal received from the asynchronous carrier generation unit 137, and outputs the selected carrier signal as the second carrier signal C 1B. The second carrier signal C 1B is output to the second switching pulse generation unit 15B via a second output terminal 339B.FIG. 16 shows processing timings in the case where the voltage command calculation is performed between the peaks and bottoms of the carrier signals for both winding groups in the synchronous PWM control in Embodiment 3, and shows the synchronous carrier signals for the first group and the second group output from the carrier generation unit 33 in the 9-pulse synchronous PWM control. The first carrier signal C 1A is generated to be synchronized with the first voltage phase θv 1. The second carrier signal C 1B is generated to be synchronized with the second voltage phase θv 2* and provided with a certain phase difference θoffset relative to the first voltage phase θv 1.In the voltage command calculation, as in Embodiment 1, for the first group voltage command, the start of the voltage command calculation and the voltage command update are performed in synchronization with the timings of the peaks and bottoms of the first carrier signal C 1A, and for the second group voltage command, the start of the voltage command calculation and the voltage command update are performed in synchronization with the timings of the peaks and bottoms of the second carrier signal C 1B. In fact, the second carrier signal C 1B is synchronized with the first voltage phase θv 1 as shown in Expression (13). Therefore, the voltage command calculation time T 2 becomes almost constant at 1 / 4 of the period of the first group synchronous carrier signal, and thus a stable calculation time can be ensured.In FIG. 16, the case of 9-pulse synchronous PWM control is shown. However, the same applies to the carrier signals in the 3-pulse synchronous PWM control.FIG. 17 shows processing timings in the case where the voltage command calculation is performed between the peaks and bottoms of the carrier signals for both winding groups in the synchronous PWM control in Embodiment 3, and shows the synchronous carrier signals for the first group and the second group output from the carrier generation unit 33 in the 6-pulse synchronous PWM control. The first carrier signal C 1A is generated to be synchronized with the first voltage phase θv 1. The second carrier signal C 1B is generated to be synchronized with the second voltage phase θv 2* and provided with a certain phase difference θoffset relative to the first voltage phase θv 1.In the voltage command calculation, as in Embodiment 1, for the first group voltage command, the start of the voltage command calculation and the voltage command update are performed in synchronization with the timings of the peaks and bottoms of the first carrier signal C 1A, and for the second group voltage command, the start of the voltage command calculation and the voltage command update are performed in synchronization with the timings of the peaks and bottoms of the second carrier signal C 1B. In the example shown in FIG. 17, the carrier phase difference between the first carrier signal C 1A and the second carrier signal C 1B is 180°, and a peak (a bottom) of the first carrier signal C 1A and a bottom (a peak) of the second carrier signal C 1B arrive simultaneously.Therefore, the start of the voltage command calculation and the voltage command update for the first group and the second group are simultaneously performed. In addition, since the second carrier signal C 1B is synchronized with the first voltage phase θv 1, the voltage command calculation time T 3 becomes almost constant at 1 / 2 of the period of the first group synchronous carrier signal.FIG. 18 shows the relationship between the operating state of the multiplexed winding electric motor and the type of the carrier signal used in the PWM control in Embodiment 3. The determination of the operation state of the multiplexed winding electric motor 83 is the same as in Embodiment 1.In the case where the rotation speed is low and the operation state is in a low rotation speed range, as in Embodiment 1, the control mode is set to the "asynchronous PWM control mode", and asynchronous PWM control for generating switching pulses using the common asynchronous carrier is performed for the first group and the second group. In this case, the voltage command calculation time is constant, and it is possible to stably perform the voltage command calculation including the calculation for the inter-group interference prevention processing.In the case where the rotation speed is medium speed and the operation state is in a medium speed range, as in Embodiment 1, the 9-pulse synchronous PWM control or the 6-pulse synchronous PWM control is performed, and the control mode is set to the "first synchronous PWM control mode". In this regard, Embodiment 3 differs from Embodiment 1 in that the respective carrier signals (first carrier signal C 1A and second carrier signal C 1B) are used for the first group and the second group. However, the first carrier signal C 1A and the second carrier signal C 1B are both synchronized with the first voltage phase.In addition, the first carrier signal C 1A and the second carrier signal C 1B are synchronized with the first voltage phase with a certain phase difference θoffset being imparted therebetween. Therefore, a calculation time including the calculation for the inter-group interference prevention processing is ensured, and thus stable current control can be performed. Note that control is also possible in which the voltage command is directly input as a command value from the outside.In the case where the rotation speed is further increased and the operation state is in a high rotation speed range, the 3-pulse synchronous PWM control is performed, and the control mode is set to the "second synchronous PWM control mode". As in Embodiment 1 as well, the respective carrier signals for the first group and the second group are used in the 3-pulse PWM control, but unlike Embodiment 1, the two carrier signals are not independent of each other, and with a certain phase difference θoffset given to the carrier signal for the second group, the two carrier signals are synchronized with the first group voltage command in common.FIG. 19 is a block diagram showing the configuration of a converter control device in another example of Embodiment 3, the converter control device according to Embodiment 3 being implemented using a computing unit of a microcomputer. In FIG. 19, components identical or corresponding to those in FIGS. 1 and 14 are provided with the same reference numerals. In the following description, the components different from FIGS. 1 and 14 will be mainly described.A converter controller 301 is different from the converter controller 30 in that the voltage command generation unit 11, the first two-phase to three-phase conversion unit 12A, the second two-phase to three-phase conversion unit 12B, the first three-phase to two-phase conversion unit 14A, and the second three-phase to two-phase conversion unit 14B in FIGS. 1 and 14 are collectively replaced by a computation unit 39 formed of a microcomputer. Hereinafter, mainly this part will be described.On the basis of the first group current commands id 1*, iq 1* (not illustrated) and the second group current commands id 2*, iq 2* (not illustrated) input from the master controller (not illustrated), and the first group three-phase current values iu 1, iv 1, iw 1, the second group three-phase current values iu 2, iv 2, iw 2, and the magnetic pole position signal θe that are detected, the arithmetic unit 39 performs the same calculation as in Embodiment 1, and outputs: the first group three-phase voltage commands vu 1*, vv 1*, vw 1*, the second group three-phase voltage commands vu 2*, vv 2*, vw 2*, and the first group voltage commands vd 1*, vq 1*, which are represented in an orthogonal two-phase coordinate system.The carrier generation unit 33 receives the first group voltage commands vd 1*, vq 1*, and generates the first carrier signal C 1A and the second carrier signal C 1B to be used for generating switching pulses in the first switching pulse generation unit 15A and the second switching pulse generation unit 15B, respectively. As in Embodiment 1, the first switching pulse generation unit 15A and the second switching pulse generation unit 15B generate the first group switching pulses UP 1, UN 1, VP 1, VN 1, WP 1, WN 1 and the second group switching pulses UP 2, UN 2, VP 2, VN 2, WP 2, WN 2 by triangular wave comparison, and output them to the first inverter 81A and the second inverter 81B. As in Embodiment 3, the carrier generation unit 33 outputs the first carrier signal C 1A to the first switching pulse generation unit 15A and outputs the second carrier signal C 1B to the second switching pulse generation unit 15B.The other components are the same as in Embodiment 1, and therefore, the description thereof is omitted.The configuration that the arithmetic unit 39 uses, as in the inverter control device 301, can also be applied to Embodiment 1 or Embodiment 2.In Embodiment 3, the same effect as in Embodiment 1 can be obtained.The carrier generation unit is configured such that the second synchronous carrier signal is synchronized with the first group voltage command. Therefore, the second group voltage command does not need to be input to the carrier generation unit, and the configuration of the carrier generation unit is simplified.In addition, if a voltage command unit and the like are replaced with the calculation unit formed of a microcomputer, various functional units can be integrated, and thus the inverter control device can be configured at low cost.Embodiment 4Hereinafter, Embodiment 4 of the present invention will be described with reference to FIGS. 20 to 22. Embodiment 4 is different from Embodiment 3 in that, in the generation of the synchronous carrier signals for synchronous PWM control, the second synchronous carrier signal is generated using the voltage phases of both the first group and the second group. FIG. 20 is a block diagram showing the configuration of a converter control device according to Embodiment 4.Components identical or corresponding to those in FIG. 19 are provided with the same reference numerals. In the following description, the components different from FIG. 19 will be mainly described. A converter controller 40 is different from Embodiment 3 in that an arithmetic unit 49 and a carrier generation unit 43 are provided.On the basis of the first group current commands id 1*, iq 1* (not illustrated) and the second group current commands id 2*, iq 2* (not illustrated) input from the master controller (not illustrated), and the first group three-phase current values iu 1, iv 1, iw 1, the second group three-phase current values iu 2, iv 2, iw 2, and the magnetic pole position signal θe that are detected, the arithmetic unit 49 performs the same calculation as in Embodiment 3, and outputs: the first group three-phase voltage commands vu 1*, vv 1*, vw 1*, the second group three-phase voltage commands vu 2*, vv 2*, vw 2*, and the first group voltage commands vd 1*, vq1* and the second group voltage commands vd2*, vq2* represented in a two-phase orthogonal coordinate system. The first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* are input to the carrier generation unit 43.The carrier generation unit 43 outputs a first carrier signal C 1M to the first switching pulse generation unit 15A and a second carrier signal C 1N to the second switching pulse generation unit 15B, on the basis of the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* output from the calculation unit 49 and the magnetic pole position signal θe. As in Embodiment 3, the first switching pulse generation unit 15A and the second switching pulse generation unit 15B generate by triangular wave processing. The carrier generation unit 43, the first switching pulse generation unit 15A, and the second switching pulse generation unit 15B constitute a PWM control unit 48.FIG. 21 is a detailed block diagram showing the configuration of the carrier generation unit according to Embodiment 4. Components identical or corresponding to the carrier generation unit shown in FIG. 3 are provided with the same reference numerals. In the following description, the components different from FIG. 3 will be mainly described. The difference from Embodiment 3 is that the second synchronous carrier signal is generated from the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* represented in a two-phase orthogonal coordinate system and the magnetic pole position signal θe.In the carrier generation unit 43, a second phase correction unit 435B receives the output θthv 1 of the first adder 434A, the first control phase angle THV 1, and the second control phase angle THV 2, and calculates a second voltage phase θv 2** from the output θthv 1 of the first adder 434A, the first control phase angle THV 1, and the second control phase angle THV 2. If the first group voltage commands vd 1*, vq 1* and the second group voltage commands vd 2*, vq 2* are different from each other, the first control phase angle THV 1 and the second control phase angle THV 2 are different from each other. Therefore, in Embodiment 4, a voltage phase correction value Δθ* is calculated from the difference between the first control phase angle THV 1 and the second control phase angle THV 2.Then, in the calculation of the second voltage phase θv 2** on the basis of the output θthv 1 of the first adder 434A, correction is performed by the voltage phase correction value Δθ*. In this case, the second voltage phase θv 2** to which the second synchronous carrier signal is synchronized becomes a phase obtained by correcting the output θthv 1 of the first adder 434A by the voltage phase correction value Δθ* determined on the basis of the phase difference between the first control phase angle THV 1 and the second control phase angle THV 2.In the calculation of the second voltage phase θv 2**, limit value processing may be performed as described in Embodiment 2 such that the first voltage phase θv 1 and the second voltage phase θv 2** do not deviate from each other in electrical phase angle by a certain value or larger. In this case, if the phase correction value Δθ* is equal to or less than the maximum correction value Δθmax*, correction is made by the phase correction value Δθ*, and if the voltage phase correction value Δθ* is greater than the maximum correction value Δθmax*, correction is made by the maximum correction value Δθmax.The first selector 438A selects one of the first synchronous carrier signal received from the first synchronous carrier generation unit 136A and the asynchronous carrier signal received from the asynchronous carrier generation unit 137, and outputs the selected carrier signal as the first carrier signal C 1M. The first carrier signal C 1M is output to the first switching pulse generation unit 15A via a first output terminal 439A.The second selector 438B selects one of the second synchronous carrier signal received from the second synchronous carrier generation unit 436B and the asynchronous carrier signal received from the asynchronous carrier generation unit 137, and outputs the selected carrier signal as the second carrier signal C 1N. The second carrier signal C 1N is output to the second switching pulse generation unit 15B via a second output terminal 439B.FIG. 22 shows processing timings in the case where the voltage command calculation is performed between the peaks and bottoms of the carrier signals for both winding groups in the synchronous PWM control in Embodiment 4, and shows the synchronous carrier signals for the first group and the second group output from the carrier generation unit 43 in the 9-pulse synchronous PWM control. The first carrier signal C 1M is generated to be synchronized with the first voltage phase θv 1. The second carrier signal C 1N is generated to be synchronized with the second voltage phase θv 2** and provided with a certain phase difference θoffset relative to the first voltage phase θv 1.In the voltage command calculation, as in Embodiment 1, for the first group voltage command, the start of the voltage command calculation and the voltage command update are performed in synchronization with the timings of the peaks and bottoms of the first carrier signal C 1M, and for the second group voltage command, the start of the voltage command calculation and the voltage command update are performed in synchronization with the timings of the peaks and bottoms of the second carrier signal C 1N. Since the voltage phase correction value Δθ* is subjected to the limit processing by the maximum correction value Δθmax, the correction of the second voltage phase θv 2** is performed within a range in which the voltage command calculation time T 1M is not made insufficient.In FIG. 22, the case of 9-pulse synchronous PWM control is shown. However, the same applies to the carrier signals in the 3-pulse synchronous PWM control.The other components are the same as in Embodiment 1, and therefore, the description thereof is omitted.In Embodiment 4, the same effect as in Embodiment 1 can be obtained.In the calculation of the second voltage phase, correction is performed by the voltage phase correction value calculated from the difference between the first control phase angle and the second control phase angle. Accordingly, it is possible to obtain a converter control device capable of more stably controlling a multiplexed electric motor.Embodiment 5Hereinafter, Embodiment 5 of the present invention will be described with reference to FIGS. 23 and 24. In Embodiment 5, in the case where the asynchronous PWM control is performed in a low-speed range, different asynchronous carrier signals are used in the PWM control for the winding group of the first group and the PWM control for the winding group of the second group, respectively, unlike Embodiments 1 to 4. Fig. 23 is a detailed block diagram showing the configuration of a carrier generation unit according to Embodiment 5.Components identical or corresponding to those in FIG. 3 are provided with the same reference numerals. In the following description, the components different from FIG. 3 will be mainly described. A carrier generation unit 53 includes an asynchronous carrier generation unit 537 that generates a first asynchronous carrier signal and a second asynchronous carrier signal, and outputs the respective asynchronous carrier signals to the first selector 138A and the second selector 138B.The first selector 138A selects one of the first synchronous carrier signal and the first asynchronous carrier signal, and outputs the selected carrier signal as the first carrier signal C 1. The first carrier signal C 1 is output to the first switching pulse generation unit 15A via the first output terminal 139A. The second selector 138B selects one of the first synchronous carrier signal, the second synchronous carrier signal, and the second asynchronous carrier signal, and outputs the selected carrier signal as the second carrier signal C 2. The second carrier signal C 2 is output to the second switching pulse generation unit 15B via the second output terminal 139B.FIG. 24 is a timing chart showing the relationship between the voltage command for each winding group and the asynchronous carrier signals in Embodiment 5. In FIG. 24, as an example, the first selector 138A and the second selector 138B respectively select the first asynchronous carrier signal and the second asynchronous carrier signal, so that the first asynchronous carrier signal is applied to the PWM control for the first group and the second asynchronous carrier signal is applied to the PWM control for the second group.The first asynchronous carrier signal and the second asynchronous carrier signal have the same frequency and differ from each other only in the carrier phase. In the example shown in FIG. 24, the carrier phase difference is 90°. Note that the carrier phase difference existing between the two asynchronous carrier signals is not limited to 90°, and may be 180° or the like.The other components are the same as in Embodiment 1, and therefore, the description thereof is omitted.In Embodiment 5, the same effect as in Embodiment 1 can be obtained.In addition, in the asynchronous PWM control, the asynchronous carrier signals having the same frequency and different phases are used for the respective winding groups. Accordingly, harmonics occurring in the currents flowing through the winding groups can be suppressed. Thereby, the electric motor with multiplex winding can be controlled more stably.In the above embodiments, the case where a duplex three-phase AC electric motor duplexed using two winding groups is driven is described as an example. However, the present invention is also applicable to multi-group and multi-phase electric motors, and the number of phases and the number of groups is not particularly limited. In addition, the multiplexed winding electric motor used may be a permanent magnet synchronous electric motor or an induction electric motor.Various methods for selecting carrier signals in the PWM control are conceivable, and the type of the combination is not limited to a combination. For example, it is also possible to perform control in which a single carrier signal or two carrier signals having a phase difference are used in asynchronous PWM control and a carrier signal for synchronous PWM control is synchronized with a specific winding group such as the winding group of the second group.In addition, for the correction method for the carrier signals, there are various other methods such as calculation using an approximation formula and calculation related to a table, and thus the method is not limited to a single method. In addition, the carrier signal is not limited to a triangular wave, and a saw-tooth wave carrier signal may be used.It should be noted that, within the scope of the present invention, the above embodiments can be freely combined with each other. In addition, each of the above embodiments can be appropriately modified or simplified.Description of the Reference Numerals10, 30 Converter control device 301, 40 Converter control device 11, 31 Voltage command generation unit 13 Carrier generation unit 23, 33 Carrier generation unit 43, 53 Carrier generation unit 15A First switching pulse generation unit 15B Second switching pulse generation unit 18, 38, 48 PWM control unit 81A First converter 81B Second converter 83 Electric motor with multiplex winding 131A First group voltage command input terminal 131B Second group voltage command input terminal 132 Magnetic pole position signal input terminal 133A First control phase angle calculation unit 133B Second control phase angle calculation unit 135A, 235A First phase correction unit 135B Second phase correction unit 335B, 435b Second phase correction unit 136A First synchronous carrier generation unit 136B Second synchronous carrier generation unit 336B, 436B Second synchronous carrier generation unit 137, 537 Asynchronous carrier generation unit 138A, 438A First selector 138B Second selector 338B, 438B Second selector C1 Second selector C1A, C1M First carrier signal C2 Second carrier signal C1B, C1N Second carrier signal θe Magnetpol position signal

Claims

A converter control device (10, 30, 301, 40) for performing PWM control for converters (81A, 81B) each supplying power to a plurality of winding groups of a multiplexed electric motor (83) having the plurality of winding groups, the converter control device (10, 30, 301, 40) comprising: a PWM control unit (18, 38, 48) configured to perform the PWM control while being synchronized between an asynchronous PWM control mode using a carrier for the PWM control that is not synchronized with the three-phase voltage commands for driving the multiplexed electric motor (83), A synchronous PWM control mode using a carrier for PWM control synchronized with three-phase voltage commands for driving the electric motor (83) with multiplex winding, wherein the synchronous PWM control mode includes a first synchronous PWM control mode in which voltage commands for the respective winding groups of the electric motor (83) with multiplex winding are subjected to PWM control using a carrier synchronized with a voltage command for a specific winding group and a second synchronous PWM control mode in which voltage commands for the respective winding groups of the electric motor (83) with multiplex winding are subjected to PWM control using respective carriers, which are synchronized with the respective voltage commands, and wherein the PWM control unit (18, 38, 48) performs control while selectively switching between the asynchronous PWM control mode and both or one of the first synchronous PWM control mode and the second synchronous PWM control mode, on the basis of an operation state of the multiplexed winding electric motor (83).The converter control device (10, 30, 301, 40) according to claim 1, wherein in the first synchronous PWM control mode, phase correction is performed according to a phase difference between the voltage commands for the plurality of winding groups for the carrier synchronized with the voltage command for the respective winding group of the multiplexed winding electric motor (83).The converter control device (10, 30, 301, 40) according to claim 1, wherein in the first synchronous PWM control mode, phase correction is performed according to an electrical phase difference between the winding groups of the multiplexed winding electric motor (83) for the carrier synchronized with the voltage command for the respective winding group of the multiplexed winding electric motor (83).The converter control device (10, 30, 301, 40) according to claim 1, wherein in the second synchronous PWM control mode, phase correction is performed according to a phase difference between the voltage commands for the plurality of winding groups for each of the carriers synchronized with the voltage commands for the respective winding groups of the multiplexed winding electric motor (83).The converter control device (10, 30, 301, 40) according to any one of claims 1 to 4, wherein the PWM control unit (18, 38, 48) selects the asynchronous PWM control mode in a range in which a frequency of the voltage commands for the plurality of winding groups of the multiplexed winding electric motor (83) is lower than a predetermined frequency, and selects the first synchronous PWM control mode or the second synchronous PWM control mode in a range in which the frequency of the voltage commands for the plurality of winding groups is higher than the predetermined frequency.The converter control device (10, 30, 301, 40) according to any one of claims 1 to 4, wherein the PWM control unit (18, 38, 48) selects the asynchronous PWM control mode in a range in which the rotation speed of the multiplexed electric motor (83) is lower than a predetermined rotation speed, and selects the first synchronous PWM control mode or the second synchronous PWM control mode in a range in which the rotation speed is higher than the predetermined rotation speed.The converter control device (10, 30, 301, 40) according to any one of claims 1 to 4, wherein the PWM control unit (18, 38, 48) selects the asynchronous PWM control mode in a range in which a modulation factor of the voltage commands for the plurality of winding groups of the multiplexed winding electric motor (83) is lower than a predetermined value, and selects the first synchronous PWM control mode or the second synchronous PWM control mode in a range in which the modulation factor of the voltage commands for the plurality of winding groups is higher than the predetermined value.The converter control device (10, 30, 301, 40) according to any one of claims 1 to 4, wherein the PWM control unit (18, 38, 48) is configured to determine which range the operating state of the multiplexed electric motor (83) belongs to, two threshold values being set based on a value of a rotational speed of the multiplexed electric motor (83), the frequency of the voltage commands for the plurality of winding groups of the multiplexed electric motor (83), or the modulation factor of the voltage commands for the plurality of winding groups of the multiplexed electric motor (83), and wherein when the value is less than the smallest one of the two threshold values, the asynchronous PWM control mode is selected when the value is not less than the smaller one of the two threshold values, but smaller than the larger one of the two threshold values, the first synchronous PWM control mode is selected, and when the value is not smaller than the larger one of the two threshold values, the second synchronous PWM control mode is selected.The converter control device (10, 30, 301, 40) according to any one of claims 1 to 8, wherein the PWM control unit (18, 38, 48) comprises: a carrier generation unit (13, 23, 33, 43, 53) configured to output the carrier for performing the PWM control; and a switching pulse generation unit (15A, 15B) configured to generate a switching pulse for driving a switching element of the converter (81A, 81B) using the carrier, and wherein the carrier generation unit (13, 23, 33, 43, 53) outputs the carrier according to each control mode based on the operation state of the multiplexed winding electric motor (83).The converter control device (10, 30, 301, 40) according to any one of claims 1 to 9, wherein in the asynchronous PWM control mode, the PWM control of the voltage commands for the respective winding groups of the multiplexed winding electric motor (83) is performed using a respective one of different carriers having the same frequency but different phase, respectively.

Images