Motor device

Abstract

Provided is a motor device capable of synchronizing a phase of a PWM cycle of a first system with a phase of a PWM cycle of a second system without communication between a control circuit of the first system and a control circuit of the second system. The present invention is a motor device wherein: an excitation winding and two output windings of a first system and an excitation winding and two output windings of a second system magnetically interfere with each other; and a control circuit 6B of the second system extracts a component of the first system that is caused by the magnetic interference and that is included in each of detection values of output voltages of the two output windings of the second system and synchronizes a phase of a PWM cycle of the second system with a phase of a PWM cycle of the first system on the basis of the extracted component of the first system.

Description

Motor device

[0001] This disclosure relates to a motor device.

[0002] In recent years, motor systems that drive a dual three-phase motor, in which two sets of three-phase armature windings are wound on a single stator and driven by two inverters, have become widespread. The reason for this widespread adoption is the demand for improved reliability and fault tolerance (the ability to continue operation even if a part fails) of the motor system.

[0003] Patent Document 1 describes a first control circuit for controlling a first inverter and a second control circuit for controlling a second inverter. The switching timing of the first and second inverters is synchronized based on synchronization information. A master circuit, which is one of the first and second control circuits, outputs its own synchronization information to a slave circuit, which is the other of the first and second control circuits. The slave circuit performs a synchronization process to synchronize the switching timing of the first and second inverters based on the synchronization information obtained from the master circuit and its own corresponding synchronization information. In the sixth embodiment of Patent Document 1, as shown in Figure 10A, microcontroller 1 (corresponding to the first control circuit) transmits a communication signal CRR_cpu1 to microcontroller 2 (corresponding to the second control circuit), and as shown in Figure 10B, microcontroller 2 (corresponding to the second control circuit) synchronizes its own carrier signal CRR_cpu2 with CRR_cpu1 obtained via the communication line, thereby synchronizing the first inverter and the second inverter.

[0004] International Publication No. 2019 / 044728

[0005] However, in the technology described in Patent Document 1, a communication line is required for the first control circuit and the second control circuit to synchronize, which increases costs. In addition, the volume of the device increases due to the provision of the communication line.

[0006] Therefore, the object of this disclosure is to provide a motor device that can synchronize the phase of the PWM period of the first system and the phase of the PWM period of the second system without the first system control circuit and the second system control circuit communicating with each other.

[0007] The first motor device according to this disclosure includes a motor having a motor stator around which a first system of armature windings and a second system of armature windings are wound, and a motor rotor; a resolver having a resolver stator around which a first system of excitation windings, two first system of output windings, a second system of excitation windings, and two second system of output windings are wound, and a resolver rotor that rotates integrally with the motor rotor; a first system of inverter having a plurality of switching elements for converting the power supplied to the first system of armature windings; and a second system of inverter having a plurality of switching elements for converting the power supplied to the second system of armature windings. A first control circuit that calculates a first voltage command value to be applied to the first armature winding, generates a first PWM signal with a first PWM period based on the first voltage command value, turns on and off the plurality of switching elements of the first inverter based on the first PWM signal, applies a first AC voltage to the first excitation winding with a phase synchronized with the phase of the first PWM period, detects the output voltages of the two output windings of the first system, and calculates the rotation angle of the resolver rotor based on the detected output voltages of the two output windings of the first system, The system includes a second control circuit that calculates a second voltage command value to be applied to the second armature winding, generates a second PWM signal with a second PWM period based on the second voltage command value, turns on and off the plurality of switching elements of the second inverter based on the second PWM signal, applies a second AC voltage to the second excitation winding whose phase is synchronized with the second PWM period, detects the output voltages of the two output windings of the second system, and calculates the rotation angle of the resolver rotor based on the detected output voltages of the two output windings of the second system, wherein the first excitation winding and the two output windings of the first system and the second excitation winding and the two output windings of the second system magnetically interfere with each other.The second control circuit extracts the component of the first system caused by magnetic interference from each of the detected output voltage values ​​of the two output windings of the second system, and synchronizes the phase of the PWM period of the second system with the phase of the PWM period of the first system based on the extracted components of the first system of the two output windings of the second system.

[0008] A second motor device according to the present disclosure includes a motor having a motor stator around which a first system of armature windings and a second system of armature windings are wound, and a motor rotor; a first system of inverter having a plurality of switching elements for converting power supplied to the first system of armature windings; a second system of inverter having a plurality of switching elements for converting power supplied to the second system of armature windings; a first system of control circuit that calculates a first system of voltage command values ​​to be applied to the first system of armature windings, generates a first system of PWM signals with a first system of PWM period based on the first system of voltage command values, and turns the plurality of switching elements of the first system of inverters on and off based on the first system of PWM signals; The inverter comprises a second control circuit that calculates a second voltage command value to be applied to the second armature winding, generates a second PWM signal with a second PWM period based on the second voltage command value, and turns on and off the plurality of switching elements of the second inverter based on the second PWM signal, wherein the first armature winding and the second armature winding magnetically interfere with each other, the second control circuit detects the second current flowing through the second armature winding, extracts the first component caused by magnetic interference included in the detected value of the second current, and synchronizes the phase of the second PWM period with the phase of the first PWM period based on the extracted first component.

[0009] According to the first motor device of this disclosure, the magnetic interference between the first resolver winding and the second resolver winding is utilized to extract the component of the first system caused by magnetic interference from each of the detected values ​​of the output voltages of the two output windings of the second system. Based on the extracted components of the first system from the two output windings of the second system, information on the relative phase of the phase of the first system's AC voltage with respect to the phase of the second system's AC voltage with respect to the phase of the second system's AC voltage with respect to the phase of the second system's PWM period can be obtained, and information on the relative phase of the phase of the first system's PWM period with respect to the phase of the second system's PWM period can be obtained. Therefore, the phase of the second system's PWM period can be synchronized to the phase of the first system's PWM period using the relative phase information. Since there is no need to perform communication for synchronization between the first system's control circuit and the second system's control circuit, the cost of the device can be reduced, and reliability against communication failures can be improved.

[0010] According to the second motor device of this disclosure, by utilizing the magnetic interference between the first and second armature windings, it is possible to extract the component of the first system caused by magnetic interference from the detected current of the second armature winding. Based on the extracted component of the first system included in the detected current of the second armature winding, information on the relative phase of the first system component with respect to the phase of the second system's PWM period can be obtained, and information on the relative phase of the first system's PWM period with respect to the phase of the second system's PWM period can be obtained. Therefore, the phase of the second system's PWM period can be synchronized to the phase of the first system's PWM period using the relative phase information. Since there is no need to perform communication for synchronization between the first system's control circuit and the second system's control circuit, the cost of the device can be reduced and reliability against communication failures can be improved.

[0011] This is a schematic diagram of the motor device according to Embodiment 1. This is a schematic diagram of the three-phase armature winding of the first system and the three-phase armature winding of the second system according to Embodiment 1. This is a schematic diagram of the resolver according to Embodiment 1. This is a side view of the resolver according to Embodiment 1, viewed in the axial direction. This is a time chart for explaining the detection timing of the first system according to Embodiment 1. This is a block diagram of the control circuit of the first system according to Embodiment 1. This is a hardware configuration diagram of the control circuit of the first system according to Embodiment 1. This is a time chart for explaining the PWM control of the first system according to Embodiment 1. This is a diagram for explaining the resolver signal of the first system according to Embodiment 1. This is a block diagram of the first system removal processing unit according to Embodiment 1. This is a block diagram of the control circuit of the second system according to Embodiment 1. This is a hardware configuration diagram of the control circuit of the second system according to Embodiment 1. This is a time chart for explaining the PWM control of the second system according to Embodiment 1. This is a diagram for explaining the resolver signal of the second system according to Embodiment 1. This is a block diagram of the second system removal processing unit according to Embodiment 1. This is a block diagram of the first system component extraction unit according to Embodiment 1. This is a time chart illustrating the behavior of the sum of squares according to Embodiment 1. This is a flowchart of the PWM periodic synchronization unit according to Embodiment 1. This is a time chart illustrating the synchronization process according to Embodiment 1. This is a time chart illustrating multiple detection timings according to Embodiment 2. This is a flowchart of the PWM periodic synchronization unit according to Embodiment 2. This is a time chart illustrating the synchronization process according to Embodiment 2. This is a time chart illustrating the behavior of the deviation according to Embodiment 3. This is a flowchart of the PWM periodic synchronization unit according to Embodiment 3. This is a time chart illustrating the synchronization process according to Embodiment 3. This is a time chart illustrating the synchronization process according to Embodiment 3. This is a schematic configuration diagram of the motor device according to Embodiment 4. This is a block diagram of the first control circuit according to Embodiment 4. This is a block diagram of the second control circuit according to Embodiment 4. This is a schematic diagram of the three-phase armature winding of the first system and the three-phase armature winding of the second system according to Embodiment 4. This is a time chart illustrating the lower two-phase modulation according to Embodiment 4.This is a time chart illustrating the generation of voltage pulses by lower two-phase modulation according to Embodiment 4. This is a time chart illustrating the generation of voltage pulses by normal modulation according to Embodiment 4. This is a block diagram of the first system component extraction unit according to Embodiment 4. This is a block diagram of the d-axis equivalent circuit representing magnetic interference according to Embodiment 6. This is a time chart showing the behavior when the d-axis voltages are synchronized between systems according to Embodiment 6. This is a time chart showing the behavior when the d-axis voltages are not synchronized between systems according to Embodiment 6. This is a flowchart of the PWM period synchronization unit according to Embodiment 6. This is a flowchart of the PWM period synchronization unit according to Embodiment 6. This is a flowchart of the PWM period synchronization unit according to Embodiment 6. This is a schematic configuration diagram of the motor device according to Embodiment 7. This is a time chart showing the behavior of PWM control when the first system is lower two-phase modulation and the second system is upper two-phase modulation according to Embodiment 7. This is a time chart showing the behavior of PWM control when the first system is lower two-phase modulation and the second system is lower two-phase modulation according to Embodiment 8.

[0012] 1. Embodiment 1 The motor device according to Embodiment 1 will be described with reference to the drawings. Figure 1 is a schematic diagram of the motor device according to this embodiment. The motor device includes a motor 1, a resolver 2, a first inverter 4A, a second inverter 4B, a first control circuit 6A, and a second control circuit 6B, etc.

[0013] 1-1. Motor 1 Motor 1 comprises a motor stator and a motor rotor. The motor stator is wound with a first system of multi-phase armature windings N1 and a second system of multi-phase armature windings N2. The first system of multi-phase armature windings N1 and the second system of multi-phase armature windings N2 magnetically interfere with each other.

[0014] In this embodiment, as shown in Figure 2, the first armature winding N1 is provided with a three-phase armature winding consisting of a U1 phase armature winding Cu1, a V1 phase armature winding Cv1, and a W1 phase armature winding Cw1. The second armature winding N2 is provided with a three-phase armature winding consisting of a U2 phase armature winding Cu2, a V2 phase armature winding Cv2, and a W2 phase armature winding Cw2. The three-phase armature windings of each system may be star-connected or delta-connected. A phase difference (for example, 30 + 60 × H degrees (H is an integer) in electrical angles) may be provided between the three-phase armature winding N1 of the first system and the three-phase armature winding N2 of the second system.

[0015] The motor rotor may be equipped with permanent magnets, and the motor 1 may be a permanent magnet synchronous motor; the motor rotor may be equipped with electromagnets, and the motor 1 may be a wound-field synchronous motor; or the motor rotor may not be equipped with magnets, and the motor 1 may be an induction motor or a synchronous reluctance motor. The example described below will focus on the case where the motor rotor is equipped with permanent magnets.

[0016] 1-2. Resolver 2 The resolver 2 is a sensor for detecting the rotational position of the motor rotor. As shown in Figure 3, the resolver 2 has a first excitation winding 10A, two first output windings 111A and 112A (also referred to as the first output winding 111A and the second output winding 112A of the first system), a second excitation winding 10B, and two second output windings 111B and 112B of the second system (also referred to as the first output winding 111B and the second output winding 112B of the second system). Magnetic interference occurs between the first resolver windings 10A, 111A, and 112A and the second resolver windings 10B, 111B, and 112B. In other words, the magnetic flux generated by the excitation winding 10A of the first system induces voltages not only in the two output windings 111A and 112A of the first system, but also in the two output windings 111B and 112B of the second system, and the magnetic flux generated by the excitation winding 10B of the second system induces voltages not only in the two output windings 111B and 112B of the second system, but also in the two output windings 111A and 112A of the first system.

[0017] As shown in Figure 4, the excitation winding 10A of the first system, the two output windings 111A and 112A of the first system, the excitation winding 10B of the second system, and the two output windings 111B and 112B of the second system are wound on the same resolver stator 13. The resolver stator 13 has 12 teeth TE1 to TE12 that are evenly spaced in the circumferential direction. A resolver rotor 14 is positioned radially inward of the resolver stator 13. The resolver rotor 14 is mounted on the rotation axis of the motor rotor so as to rotate together with the motor rotor. The resolver rotor 14 has a plurality of protrusions evenly spaced in the circumferential direction on its outer circumference. The height of the radially outward protrusions of the protrusions is formed so that the gap permeance between the resolver stator 13 and the resolver rotor 14 changes sinusoidally in response to rotation. In other words, the resolver 2 is a variable reluctance (VR) type resolver. This embodiment is provided with five protrusions, and the axial angle doubler is set to 5. Therefore, for every one rotation of the motor rotor in mechanical angle, it rotates 5 times in electrical angle.

[0018] As shown in Figure 5, which illustrates an example assuming no magnetic interference between systems, when the motor rotor rotates with an AC voltage VRA applied to the excitation winding 10A of the first system, the amplitude of the AC voltage V1A induced in the first output winding 111A of the first system and the amplitude of the AC voltage V2A induced in the second output winding 112A of the first system change sinusoidally (or cosinely) according to the rotation angle (gap perminance) of the motor rotor. The first output winding 111A and the second output winding 112A of the first system are wound around the resolver stator 13 at circumferential positions such that the amplitudes of their AC voltages differ from each other by 90 degrees in electrical angle. Similarly, the first output winding 111B and the second output winding 112B of the second system are wound around the resolver stator 13 at circumferential positions such that the amplitudes of their induced AC voltages differ from each other by 90 degrees in electrical angle.

[0019] The first excitation winding 10A, wound around multiple teeth, is connected in series between the teeth, and the two terminals of the first excitation winding 10A connected in series are connected to the first control circuit 6A, which will be described later. Similarly, the two terminals of the first output winding 111A of the first system, which is connected in series between the teeth, are connected to the first control circuit 6A. The two terminals of the second output winding 112A of the first system, which is connected in series between the teeth, are connected to the first control circuit 6A. The two terminals of the second excitation winding 10B, which is connected in series, are connected to the second control circuit 6B, which will be described later. Similarly, the two terminals of the first output winding 111B of the second system, which is connected in series between the teeth, are connected to the second control circuit 6B. The two terminals of the second output winding 112B of the second system, which is connected in series between the teeth, are connected to the second control circuit 6B.

[0020] The number of protrusions (axis double angle) and the number of teeth may be set to any number. The first resolver winding and the second resolver winding do not have to be divided into two in the circumferential direction, but may be distributed in the circumferential direction.

[0021] 1-3. Inverter 4A of the first system The inverter 4A of the first system has multiple switching elements that convert the power supplied to the three-phase armature winding N1 of the first system. The inverter 4A of the first system converts the DC power of the DC power supply 3A of the first system and the AC power supplied to the three-phase armature winding N1 of the first system.

[0022] The first inverter 4A has three series circuits, each corresponding to the armature winding of a three-phase system. These series circuits consist of a high-potential switching element Sp connected in series to the high-potential terminal of the first DC power supply 3A, and a low-potential switching element Sn connected in series to the low-potential terminal of the first DC power supply 3A. The connection point of the two switching elements in each series circuit is connected to the armature winding of the corresponding phase. Switching elements include IGBTs (Insulated Gate Bipolar Transistors) with diodes connected in antiparallel, bipolar transistors with diodes connected in antiparallel, MOSFETs (Metal Oxide Semiconductor Field Effect Transistors), etc. The gate terminal of each switching element is connected to the first control circuit 6A via a gate drive circuit, etc. Therefore, each switching element is turned on or off by a PWM signal output from the first control circuit 6A.

[0023] The first inverter 4A is equipped with a first current sensor 5A for detecting the current flowing through the armature windings of each phase of the first system. The first current sensor 5A is equipped with a shunt resistor connected in series with the switching element on the low-potential side of each phase. The potential difference across the shunt resistors of each phase is input to the first control circuit 6A. The first current sensor 5A may be provided on the wire connecting the series circuit of the switching elements of each phase to the armature winding. Alternatively, the first current sensor 5A may be provided on the wire connecting the first DC power supply 3A and the first inverter 4A, and the current in the armature windings of each phase may be detected by the known "busbar 1 shunt method".

[0024] The first DC power supply 3A outputs the first DC voltage Vdc1 to the first inverter 4A. The first DC power supply 3A can be any device that outputs a DC voltage, such as a battery, DC-DC converter, diode rectifier, or PWM rectifier. A voltage sensor for detecting the first DC voltage Vdc1 may be provided, and the output signal of the voltage sensor may be input to the first control circuit 6A.

[0025] 1-4. Second System Inverter 4B The second system inverter 4B has multiple switching elements that convert the power supplied to the three-phase armature winding N2 of the second system. The second system inverter 4B converts the DC power of the second system DC power supply 3B and the AC power supplied to the three-phase armature winding N2 of the second system.

[0026] The second inverter 4B has three series circuits, each corresponding to the armature winding of a three-phase system. These series circuits consist of a high-potential switching element Sp connected in series to the high-potential terminal of the second DC power supply 3B, and a low-potential switching element Sn connected in series to the low-potential terminal of the second DC power supply 3B. The connection point of the two switching elements in each series circuit is connected to the armature winding of the corresponding phase. The gate terminal of each switching element is connected to the second control circuit 6B via a gate drive circuit, etc. Therefore, each switching element is turned on or off by a PWM signal output from the second control circuit 6B.

[0027] The second inverter 4B is equipped with a second current sensor 5B for detecting the current flowing through the armature windings of each phase of the second system. The second current sensor 5B is equipped with a shunt resistor connected in series with the switching element on the low-potential side of each phase. The potential difference across the shunt resistors of each phase is input to the second control circuit 6B. The second current sensor 5B may be provided on the wire connecting the series circuit of the switching elements of each phase to the armature winding. Alternatively, the second current sensor 5B may be provided on the wire connecting the second DC power supply 3B and the second inverter 4B, and the current in the armature windings of each phase may be detected by the known "busbar 1 shunt method".

[0028] The DC power supply 3B of the second system outputs the DC voltage Vdc2 of the second system to the inverter 4B of the second system. As the DC power supply 3B of the second system, any device that outputs a DC voltage, such as a battery, a DC-DC converter, a diode rectifier, a PWM rectifier, etc., can be used. A voltage sensor for detecting the DC voltage Vdc2 of the second system may be provided, and the output signal of the voltage sensor may be input to the control circuit 6B of the second system. The DC power supply 3A of the first system and the DC power supply 3B of the second system may be the same.

[0029] 1-5. The control circuit 6A of the first system As shown in FIG. 6, the control circuit 6A of the first system includes functional units such as a first-system excitation application unit 30, a first-system output voltage detection unit 31, a first-system removal processing unit 32, a first-system angle calculation unit 33, a first-system voltage command value calculation unit 34, and a first-system voltage application unit 35.

[0030] The functions of the functional units 30 to 35 and the like of the control circuit 6A of the first system are realized by a processing circuit provided in the control circuit 6A of the first system. Specifically, as shown in FIG. 7, the control circuit 6A of the first system includes, as a processing circuit, an arithmetic processing device 90 (computer) such as a CPU (Central Processing Unit), a storage device 91 that exchanges data with the arithmetic processing device 90, an input circuit 92 that inputs an external signal to the arithmetic processing device 90, an output circuit 93 that outputs a signal from the arithmetic processing device 90 to the outside, and a communication device 94 that performs data communication with an external device 50.

[0031] As the arithmetic processing device 90, an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), various logic circuits, and various signal processing circuits may be provided. Also, as the arithmetic processing device 90, a plurality of the same type or different types may be provided, and each process may be executed in a shared manner. As the storage device 91, a RAM (Random Access Memory) and a ROM (Read Only Memory) and the like are provided.

[0032] Various sensors such as the first output winding 111A of the first system, the second output winding 112A of the first system, and the current sensor 5A of the first system are connected to the input circuit 92. The input circuit 92 includes an A / D converter or the like that inputs the output voltage of each output winding and the output voltage of the sensor to the arithmetic processing unit 90. The excitation winding 10A of the first system is connected to the output circuit 93, and the output circuit 93 includes a drive circuit such as a switching element for applying an AC voltage VRA to this excitation winding. Further, an electrical load such as a gate drive circuit for on / off driving a plurality of switching elements of the inverter 4A of the first system is connected to the output circuit 93, and the output circuit 93 includes a drive circuit or the like that outputs a control signal from the arithmetic processing unit 90 to these electrical loads. The communication device 94 communicates with the external device 50.

[0033] And each function of each functional unit 30 to 35 and the like included in the control circuit 6A of the first system is realized by the arithmetic processing unit 90 executing software (program) stored in the storage device 91 such as a ROM and cooperating with other hardware of the control circuit 6A of the first system such as the storage device 91, the input circuit 92, the output circuit 93, and the communication device 94. The setting data used by each functional unit 30 to 35 and the like is stored in the storage device 91 such as a ROM.

[0034] 1-5-1. Voltage application process for the first system The first system voltage command value calculation unit 34 calculates the three-phase voltage command values Vu1_ref, Vv1_ref, Vw1_ref of the first system to be applied to the three-phase armature windings N1 of the first system, generates the first system PWM signal of the first system PWM period Tpwm1 for each switching element based on the three-phase voltage command values Vu1_ref, Vv1_ref, Vw1_ref of the first system, and turns on and off each switching element based on the first system PWM signal.

[0035] In this embodiment, the first system voltage command value calculation unit 34 detects the current flowing through the three-phase armature windings of the first system at a current detection timing synchronized with the phase of the first system's PWM period Tpwm1, and calculates the multi-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system based on the detected current values ​​of the three-phase armature windings of the first system. In this embodiment, the current detection timing of the first system is set to the start timing of the first system's PWM period Tpwm1 (the peak of the first system's carrier wave CA1). Alternatively, the current detection timing of the first system may be set to the center timing of the first system's PWM period Tpwm1 (the trough of the first system's carrier wave CA1). Alternatively, the current detection timing for the first system may be set to the start timing of the first system's PWM period Tpwm1 every two cycles (the peak of the first system's carrier wave CA1). In any case, the current detection timing for the first system coincides with either or both of the start timing and center timing of the first system's PWM period Tpwm1 (the peaks and troughs of the first system's carrier wave CA1).

[0036] In this embodiment, the voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref for the three phases of the first system are calculated by current feedback control on a dq axis rotation coordinate system that rotates in synchronization with the angle θ1 of the first system. The first system voltage command value calculation unit 34 includes a current command value calculation unit 341, a current control unit 342, a current coordinate transformation unit 343, and a voltage coordinate transformation unit 344.

[0037] The dq-axis rotation coordinate is defined as a rotation coordinate consisting of a d-axis defined in the direction of the magnetic flux of the motor rotor, and a q-axis defined in a direction advanced by π / 2 electrical angle from the d-axis. In this embodiment, the direction of the magnetic flux of the motor rotor is defined as the direction of the north pole of the permanent magnet provided on the motor rotor.

[0038] As shown in Figure 6, the current command value calculation unit 341 calculates the d-axis current command value Id1_ref and the q-axis current command value Iq1_ref for the first system. The current command value calculation unit 341 multiplies the output command value OPc by the load factor of the first system to calculate the output command value OPc1 for the first system. For example, the load factor of the first system is set to 0.5 when both the first and second systems are normal, and the load factor of the first system is set to a value greater than 0.5 when the first system is normal and the second system is abnormal.

[0039] The output command value OPc is determined according to the system in which the motor 1 is used, and may be the torque command value of the motor 1, the current command value of the motor 1, the mechanical power command value of the motor 1, or the power command value of the rotating electric machine. The output command value OPc may be transmitted from an external device 50, or it may be calculated within the first control circuit 6A.

[0040] The current command value calculation unit 341 calculates the dq axis current command values ​​Id1_ref and Iq1_ref of the first system according to current vector control methods such as maximum efficiency control, maximum torque current control, weakened flux control, and Id=0 control, based on the output command value OPc1 of the first system, the DC voltage Vdc1 of the first system, and the angular velocity ω1 of the first system.

[0041] The current coordinate transformation unit 343 takes the current detection values ​​Iu1, Iv1, and Iw1 flowing through the three-phase armature windings of the first system, detected by the first system current sensor 5A, and performs three-phase to two-phase conversion and rotational coordinate transformation based on the angle θ1 of the first system to convert them into the d-axis current detection value Id1 and the q-axis current detection value Iq1 of the first system, expressed in a dq-axis rotational coordinate system.

[0042] The current control unit 342 performs feedback control by changing the d-axis voltage command value Vd1_ref and q-axis voltage command value Vq1_ref of the first system, using PI control or the like, so that the detected d-axis current value Id1 and q-axis current value Iq1 of the first system approach the d-axis current command value Id1_ref and q-axis current command value Iq1_ref of the first system.

[0043] The voltage coordinate transformation unit 344 transforms the d-axis voltage command value Vd1_ref and the q-axis voltage command value Vq1_ref of the first system into three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system by performing fixed coordinate transformation and two-phase to three-phase transformation based on the angle θ1 of the first system.

[0044] For the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system, amplitude reduction modulation may be applied, such as third harmonic superposition, the min-max method (pseudo-third harmonic superposition), and two-phase modulation, which superimpose a zero-sequence voltage that reduces the amplitude of the AC voltage command value while maintaining the line voltage. In third harmonic superposition, a sinusoidal zero-sequence voltage with a rotation period three times that of the electrical angle is superimposed. In the min-max method (pseudo-third harmonic superposition), a triangular wave zero-sequence voltage with a rotation period three times that of the electrical angle is superimposed. For the sake of explanation, the modulated voltage command value will also be simply referred to as the voltage command value.

[0045] The first system voltage application unit 35 generates PWM signals for each switching element of the first system inverter 4A based on the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system, and turns each switching element of the first system on and off based on each PWM signal of the first system.

[0046] As shown in Figure 8, the first system voltage application unit 35 generates PWM signals to turn each switching element on and off by comparing the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system with the carrier wave CA1 of the first system. Each PWM signal is input to the gate terminal of each switching element of the inverter 4A of the first system via a gate drive circuit, turning each switching element on or off. The carrier wave CA1 of the first system is a triangular wave having the amplitude of the DC voltage Vdc1 of the first system and oscillating with the PWM period Tpwm1 of the first system. Other types of waves, such as sawtooth waves, may also be used.

[0047] For each phase, the first system voltage application unit 35 turns on the PWM signal Gp of the switching element on the high potential side (1 in this example) when the carrier wave CA1 of the first system falls below the voltage command value, thereby turning on the switching element on the high potential side. If the carrier wave CA1 of the first system exceeds the voltage command value, the first system voltage application unit 35 turns off the PWM signal Gp of the switching element on the high potential side (0 in this example), thereby turning off the switching element on the high potential side. On the other hand, for each phase, the first system voltage application unit 35 turns off the PWM signal Gn of the switching element on the low potential side (0 in this example) when the carrier wave CA1 of the first system falls below the voltage command value, thereby turning off the switching element on the low potential side. If the carrier wave CA1 of the first system exceeds the voltage command value, the first system voltage application unit 35 turns on the PWM signal Gn of the switching element on the low potential side (1 in this example), thereby turning on the switching element on the low potential side. Furthermore, for each phase, a short-circuit prevention period (dead time) may be provided between the on-period of the high-potential switching element and the on-period of the low-potential switching element, during which both the high-potential and low-potential switching elements are turned off.

[0048] 1-5-2. Angle Detection Processing of the First System <First System Excitation Application Unit 30> The first system excitation application unit 30 applies the first system AC voltage VRA (in this example, a sinusoidal AC voltage VRA) to the first system excitation winding 10A, with the phase synchronized with the phase of the first system PWM period Tpwm1. In this embodiment, as shown in Figure 9, the timing of the peak (maximum value) of the first system AC voltage VRA is synchronized to match the timing of the peak (maximum value) of the first system carrier wave CA1. The first period TA is set to be the same as the first system PWM period Tpwm1. For each first system PWM period Tpwm1, the timing of the peak (maximum value) of the first system AC voltage VRA matches the timing of the peak (maximum value) of the first system carrier wave CA1. For example, the first period TA is set to 50 μs.

[0049] The first excitation application unit 30 may generate a square wave signal with a first period TA that alternately outputs two voltage values, "H" level (e.g., 5V) and "L" level (e.g., 0V), using a drive circuit, input its output to a low-pass filter circuit, and apply the output of the low-pass filter circuit as an AC voltage VRA to the first excitation winding 10A. In this case as well, the phase of the square wave signal relative to the first PWM signal CA is set so that the timing of the peak (maximum value) of the first AC voltage VRA matches the timing of the peak (maximum value) of the output of the low-pass filter circuit.

[0050] <First System Output Voltage Detection Unit 31> The first system output voltage detection unit 31 detects the output voltages V1A_S and V2A_S of the two output windings 111A and 112A of the first system at a single timing synchronized with the phase of the PWM period Tpwm1 of the first system. The first system output voltage detection unit 31 detects the output voltages V1A and V2A of the two output windings of the first system at the start timing of the PWM period Tpwm1 of the first system (the peak of the carrier wave CA1 of the first system, the maximum value of the AC voltage VRA of the first system). Therefore, the detection timing of the first system is set to the timing of each PWM period Tpwm1 (= first period TA) of the first system.

[0051] In this embodiment, as shown in Figure 9, the first system output voltage detection unit 31 detects the output voltages V1A_S and V2A_S of the two output windings of the first system at the timing of each PWM period Tpwm1 of the first system where the carrier wave CA1 of the first system reaches a peak (maximum value).

[0052] <First System Removal Processing Unit 32> As shown in Figure 9, an example of the output voltage V1A of the first output winding 111A of the first system, the output voltages V1A_S and V2A_S of the two output windings 111A and 112A of the first system are superimposed with second-period components V1A_TB and V2A_TB, respectively, which are induced by the magnetic flux of the second period TB excited in the excitation winding 10B of the second system due to magnetic interference between the systems. The third graph from the top in Figure 9 shows the output voltage V1A of the first output winding 111A of the first system, the fourth graph shows the first period component V1A_TA, which is included in the output voltage V1A of the first output winding 111A of the first system and induced by the magnetic flux of the excitation winding 10A of the first system, and the fifth graph shows the second period component V1A_TB, which is included in the output voltage V1A of the first output winding 111A of the first system and induced by the magnetic flux of the excitation winding 10B of the second system. The output voltage V1A of the first output winding 111A of the first system is a signal obtained by summing the first period component V1A_TA and the second period component V1A_TB. In this embodiment, the second period TB is set to twice the first period TA.

[0053] Therefore, if the angle is calculated based on the output voltage V1A of the first output winding of the first system and the output voltage V2A of the second output winding of the first system, which are superimposed with the second period components V1A_TB and V2A_TB, a detection error will occur. For this reason, in order to suppress the angle detection error, it is necessary to remove the second period components V1A_TB and V2A_TB from the output voltage V1A of the first output winding of the first system and the output voltage V2A of the second output winding of the first system.

[0054] Therefore, the first system removal processing unit 32 performs a second-period component removal process on the detected output voltage values ​​V1A_S and V2A_S of the two output windings of the first system, removing (reducing) the second-period component.

[0055] As shown in the fifth graph from the top in Figure 9, the second period component V1A_TB of the output voltage of the first output winding of the first system undergoes a phase reversal and a sign reversal at a period obtained by adding an integer multiple of the second period TB to half a period of the second period TB / 2 (for example, half a period of the second period TB / 2).

[0056] Therefore, as shown in Figure 10, the first system removal processing unit 32 includes a first delay unit 3211 that outputs the detected value V1A_S of the output voltage of the first output winding of the first system with a delay of ΔT1. The detected value V1A_S of the output voltage of the first output winding of the first system and the output V1A_Sold of the first delay unit 3211 are added together to calculate the detected value V1A_F of the output voltage of the first output winding of the first system after the removal of the second period component. Similarly, the first system removal processing unit 32 includes a second delay unit 3212 that outputs the detected value V2A_S of the output voltage of the second output winding of the first system with a delay of ΔT1. The detected value V2A_S of the output voltage of the second output winding of the first system and the output V2A_Sold of the second delay unit 3212 are added together to calculate the detected value V2A_F of the output voltage of the second output winding of the first system after the removal of the second period component. The first system removal processing interval ΔT1 is set to the first period TA. Alternatively, ΔT1 may be set to ΔT1 = TB / 2 + TB × M, where M is a non-negative integer.

[0057] In this configuration, two second-period components with opposite positive and negative signs are added together, and the two second-period components cancel each other out. Therefore, the second-period components are removed from the detected output voltage values ​​V1A_F and V2A_F of the two output windings of the first system after addition. Based on the detected values ​​after the removal of the second-period components, the angle θ1 of the first system can be calculated with accuracy.

[0058] Alternatively, the first system removal processing unit 32 may be configured to perform a bandwidth reduction filter process, such as a high-pass filter process or a bandstop filter process, to reduce the second-period component as a removal process for the second-period component.

[0059] <First System Angle Calculation Unit 33> The first system angle calculation unit 33 calculates the angle θ1 of the first system based on the detected output voltages V1A_F and V2A_F of the two output windings of the first system after the second period component removal process. The first system angle calculation unit 33 calculates the angle θ1 of the first system by calculating the arctangent (inverse tangent function) of the ratio of the detected output voltage V1A_F of the first output winding of the first system after the second period component removal process to the detected output voltage V2A_F of the second output winding of the first system, as shown in the following equation. In addition, the angular velocity ω1 of the first system is also calculated based on the angle θ1 of the first system. θ1 = tan -1 (V1A_F / V2A_F) ...(1)

[0060] 1-6. Second Control Circuit 6B As shown in Figure 11, the second control circuit 6B includes functional units such as a second excitation application unit 40, a second output voltage detection unit 41, a second removal processing unit 42, a second angle calculation unit 43, a second voltage command value calculation unit 44, a second voltage application unit 45, a first component extraction unit 46, and a PWM period synchronization unit 47.

[0061] The functions of each functional unit 40 to 47 of the second control circuit 6B are realized by the processing circuit provided in the second control circuit 6B. Specifically, as shown in Figure 12, the second control circuit 6B includes a processing circuit consisting of a arithmetic processing unit 80 (computer) such as a CPU, a storage device 81 that exchanges data with the arithmetic processing unit 80, an input circuit 82 that inputs external signals to the arithmetic processing unit 80, an output circuit 83 that outputs signals from the arithmetic processing unit 80 to the outside, and a communication device 84 that communicates data with an external device 50.

[0062] The arithmetic processing unit 80 may include an ASIC, IC, DSP, FPGA, various logic circuits, and various signal processing circuits. Furthermore, multiple arithmetic processing units 80 of the same or different types may be provided, with each unit performing a portion of the processing. The storage device 81 may include RAM and ROM.

[0063] The input circuit 82 is connected to various sensors, including the second system's first output winding 111B, the second system's second output winding 112B, and the second system's current sensor 5B. The input circuit 82 includes an A / D converter and the like that inputs the output voltages of each output winding and the sensor output voltages to the arithmetic processing unit 80. The output circuit 83 is connected to the second system's excitation winding 10B and includes a drive circuit for a switching element and the like that for applying an AC voltage VRB to this excitation winding. The output circuit 83 is also connected to electrical loads such as a gate drive circuit that drives multiple switching elements of the second system's inverter 4B on and off, and includes a drive circuit and the like that that outputs control signals from the arithmetic processing unit 80 to these electrical loads. The communication device 84 communicates with the external device 50.

[0064] The functions of the second control circuit 6B, such as the functional units 40-47, are realized by the arithmetic processing unit 80 executing software (programs) stored in a storage device 81 such as a ROM, and cooperating with other hardware of the second control circuit 6B, such as the storage device 81, input circuit 82, output circuit 83, and communication device 84. The setting data used by each functional unit 40-47 is stored in the storage device 81 such as a ROM.

[0065] 1-6-1. Second System Voltage Application Processing The second system voltage command value calculation unit 44 calculates the second system's three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref to be applied to the second system's three-phase armature winding N2, generates a second system PWM signal with a second system PWM period Tpwm2 for each switching element based on the second system's three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref, and turns each switching element on or off based on the second system PWM signal.

[0066] In this embodiment, the second system voltage command value calculation unit 44 detects the current flowing through the three-phase armature windings of the second system at a current detection timing synchronized with the phase of the second system's PWM period Tpwm2, and calculates the three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref for the second system based on the detected current values ​​of the three-phase armature windings of the second system. In this embodiment, the current detection timing for the second system is set to the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2). Alternatively, the current detection timing for the second system may be set to the center timing of the second system's PWM period Tpwm2 (the trough of the second system's carrier wave CA2). Alternatively, the current detection timing for the second system may be set to the start timing of the second system's PWM period Tpwm2 every two cycles (the peak of the second system's carrier wave CA2). In any case, the current detection timing for the second system coincides with either or both of the start timing and / or center timing of the second system's PWM period Tpwm2 (the peaks and troughs of the second system's carrier wave CA2).

[0067] In this embodiment, the voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the three phases of the second system are calculated by current feedback control on a dq axis rotation coordinate system that rotates in synchronization with the angle θ2 of the second system. The second system voltage command value calculation unit 44 includes a current command value calculation unit 441, a current control unit 442, a current coordinate transformation unit 443, and a voltage coordinate transformation unit 444.

[0068] As shown in Figure 11, the current command value calculation unit 441 calculates the d-axis current command value Id2_ref and the q-axis current command value Iq2_ref for the second system. The current command value calculation unit 441 multiplies the output command value OPc by the load factor of the second system to calculate the output command value OPc2 for the second system. For example, when both the first and second systems are normal, the load factor of the second system is set to 0.5, and when the second system is normal and the first system is abnormal, the load factor of the second system is set to a value greater than 0.5.

[0069] The output command value OPc is determined according to the system in which the motor 1 is used, and may be the torque command value of the motor 1, the current command value of the motor 1, the mechanical power command value of the motor 1, or the power command value of the rotating electric machine. The output command value OPc may be transmitted from an external device 50, or it may be calculated within the second control circuit 6B.

[0070] The current command value calculation unit 441 calculates the dq axis current command values ​​Id2_ref and Iq2_ref for the second system according to current vector control methods such as maximum efficiency control, maximum torque current control, flux weakening control, and Id=0 control, based on the output command value OPc2 of the fourth system, the DC voltage Vdc2 of the second system, and the angular velocity ω2 of the second system.

[0071] The current coordinate transformation unit 443 takes the current detection values ​​Iu2, Iv2, and Iw2 flowing through the three-phase armature windings of the second system, detected by the second system current sensor 5B, and performs three-phase to two-phase conversion and rotational coordinate transformation based on the angle θ2 of the second system to convert them into the d-axis current detection value Id2 and the q-axis current detection value Iq2 of the second system, expressed in a dq-axis rotational coordinate system.

[0072] The current control unit 442 performs feedback control by changing the d-axis voltage command value Vd2_ref and q-axis voltage command value Vq2_ref of the second system, using PI control or the like, so that the d-axis current detection value Id2 and q-axis current detection value Iq2 of the second system approach the d-axis current command value Id2_ref and q-axis current command value Iq2_ref of the second system.

[0073] The voltage coordinate transformation unit 444 performs fixed coordinate transformation and two-phase to three-phase transformation on the d-axis voltage command value Vd2_ref and q-axis voltage command value Vq2_ref of the second system based on the angle θ2 of the second system, and converts them into three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the second system.

[0074] For the three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the second system, amplitude reduction modulation may be applied, such as third harmonic superposition, the min-max method (pseudo-third harmonic superposition), and two-phase modulation, which superimpose a zero-sequence voltage that reduces the amplitude of the AC voltage command value while maintaining the line voltage. For the sake of explanation, the modulated voltage command value will also be simply referred to as the voltage command value.

[0075] The second system voltage application unit 45 generates PWM signals for each switching element of the second system inverter 4B based on the three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the second system, and turns each switching element of the second system on and off based on the PWM signals of the second system.

[0076] As shown in Figure 13, the second system voltage application unit 45 generates PWM signals to turn each switching element on and off by comparing the second system's three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref with the second system's carrier wave CA2. Each PWM signal is input to the gate terminal of each switching element of the second system inverter 4B via a gate drive circuit, turning each switching element on or off. The second system's carrier wave CA2 is a triangular wave having the amplitude of the second system's DC voltage Vdc2 and oscillating with the second system's PWM period Tpwm2. Other types of waves, such as sawtooth waves, may also be used. The second system's PWM period Tpwm2 is set to the same period as the first system's PWM period Tpwm1, except during relative phase changes described later.

[0077] The generation of PWM signals for each switching element is the same as that for the first voltage application unit 35, so the explanation is omitted.

[0078] 1-6-2. Angle Detection Processing of the Second System <Second System Excitation Application Unit 40> The second system excitation application unit 40 applies the second system AC voltage VRB (in this example, a sinusoidal AC voltage VRB) to the second system excitation winding 10B, with the phase of the second system PWM period Tpwm2 synchronized with that of the second system AC voltage VRB. The timing of the peak (maximum value) of the second system AC voltage VRB is synchronized to match the timing of the peak (maximum value) of the second system carrier wave CA2. In this embodiment, the second system PWM period Tpwm2 is the same as the first system PWM period Tpwm1, and the second period TB is twice the first period TA. The second period TB is set to twice the second system PWM period Tpwm2. Every twice the period of the second PWM system Tpwm2, the timing of the peak (maximum value) of the AC voltage VRB of the second system coincides with the timing of the peak (maximum value) of the carrier wave CA2 of the second system. For example, the second period TB is set to 100 μs.

[0079] In addition, similar to the first system, the second AC voltage VRB may be generated by inputting a square wave signal of the second period TB into a low-pass filter circuit.

[0080] <Second System Output Voltage Detection Unit 41> The second system output voltage detection unit 41 detects the output voltages V1B_S and V2B_S of the two output windings 111B and 112B of the second system at a single timing synchronized with the phase of the second system's PWM period Tpwm2. The second system output voltage detection unit 41 detects the output voltages V1B and V2B of the two output windings of the first system at the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2, the maximum or minimum value of the second system's AC voltage VRB). Therefore, the detection timing for the second system is set to the timing of each second system's PWM period Tpwm2 (= second period TB / 2, first period TA).

[0081] <Second System Removal Processing Unit 42> As shown in Figure 14, an example of the output voltage V1B of the first output winding 111B of the second system, the output voltages V1B_S and V2B_S of the two output windings 111B and 112B of the second system are superimposed with first-period components V1B_TA and V2B_TA, respectively, which are induced by the magnetic flux of the first period TA excited in the excitation winding 10A of the first system due to magnetic interference between the systems. The third graph from the top in Figure 14 shows the output voltage V1B of the second system's first output winding 111B, the fourth graph shows the second period component V1B_TB, which is included in the output voltage V1B of the second system's first output winding 111B and induced by the magnetic flux of the second system's excitation winding 10B, and the fifth graph shows the first period component V1B_TA, which is included in the output voltage V1B of the second system's first output winding 111B and induced by the magnetic flux of the first system's excitation winding 10A. The output voltage V1B of the second system's first output winding 111B is a signal obtained by summing the second period component V1B_TB and the first period component V1B_TA. In this embodiment, the second period TB is set to twice the first period TA.

[0082] Therefore, if the angle is calculated based on the output voltage V1B of the first output winding of the second system and the output voltage V2B of the second output winding of the second system, which are superimposed with the components V1B_TA and V2B_TA of the first period, a detection error will occur. For this reason, in order to suppress the angle detection error, it is necessary to remove the components V1B_TA and V2B_TA of the first period from the output voltage V1B of the first output winding of the second system and the output voltage V2B of the second output winding of the second system.

[0083] Therefore, the second system removal processing unit 42 performs a first-period component removal process on the detected output voltage values ​​V1B_S and V2B_S of the two output windings of the second system, removing (reducing) the first-period component.

[0084] As shown in the fifth graph from the top in Figure 14, the first period component V1B_TA of the output voltage of the first output winding of the second system has the same phase and the same sign (positive and negative) at periods that are integer multiples of the first period TA (for example, the first period TA), and these values ​​are equivalent.

[0085] Therefore, as shown in Figure 15, the second system removal processing unit 42 includes a first delay unit 4211 that outputs the detected output voltage V1B_S of the second system's first output winding with a delay of ΔT2. The output V1B_Sold of the first delay unit 4211 is subtracted from the detected output voltage V1B_S of the second system's first output winding to calculate the detected output voltage V1B_F of the second system's first output winding after the removal of the first periodic component. Similarly, the second system removal processing unit 42 includes a second delay unit 4212 that outputs the detected output voltage V2B_S of the second system's second output winding with a delay of ΔT2. The output V2A_Sold of the second delay unit 4212 is subtracted from the detected output voltage V2B_S of the second system's second output winding to calculate the detected output voltage V2B_F of the second system's second output winding after the removal of the first periodic component. The second system removal processing interval ΔT2 is set to the first period TA. Alternatively, ΔT2 may be set to TA × P, where P is an integer greater than or equal to 1.

[0086] In this configuration, two first-period components with the same positive and negative signs and equivalent values ​​are subtracted, causing the two first-period components to cancel each other out. Therefore, the first-period components are removed from the detected output voltage values ​​V1B_F and V2B_F of the two output windings of the second system after subtraction. Based on the detected values ​​after the removal of the first-period components, the angle θ2 of the second system can be calculated with accuracy.

[0087] Alternatively, the second system removal processing unit 42 may be configured to perform a bandwidth reduction filter process, such as a low-pass filter process or a band-stop filter process, to reduce the components of the first period as a removal process for the first period component.

[0088] <Second System Angle Calculation Unit 43> The second system angle calculation unit 43 calculates the angle θ2 of the second system based on the detected output voltage values ​​V1B_F and V2B_F of the two output windings of the second system after the removal of the first periodic component. The second system angle calculation unit 43 calculates the angle θ2 of the second system by calculating the arctangent (inverse tangent function) of the ratio between the detected output voltage value V1B_F of the first output winding of the second system after the removal of the first periodic component and the detected output voltage value V2B_F of the second output winding of the second system, as shown in the following equation. In addition, the angular velocity ω2 of the second system is also calculated based on the angle θ2 of the second system. θ2 = tan -1 (V1B_F / V2B_F) ...(2)

[0089] 1-6-3. Phase-Synchronized Processing of PWM Period <First System Component Extraction Unit 46> The first system component extraction unit 46 extracts the first system components V1B_TA and V2B_TA, which are generated by magnetic interference and are included in the detected output voltage values ​​V1B_S and V2B_S of the two output windings of the second system.

[0090] In this embodiment, the second system output voltage detection unit 41 detects the output voltages V1B_S and V2B_S of the two output windings of the second system at a single timing synchronized with the phase of the second system's PWM period Tpwm2. As described above, the second system output voltage detection unit 41 detects the output voltages V1B and V2B of the two output windings of the second system at the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2, or the maximum or minimum value of the second system's AC voltage VRB). Therefore, the detection timing for the second system is set to the timing of each second system's PWM period Tpwm2 (= second period TB / 2, first period TA).

[0091] The first system component extraction unit 46 extracts the components V1B_TA and V2B_TA of the first period TA contained in the detected output voltage values ​​V1B_S and V2B_S of the two output windings of the second system, respectively, as the first system components V1B_TA and V2B_TA of the two output windings of the second system.

[0092] As shown in the fourth graph from the top in Figure 14, the second period component V1B_TB included in the output voltage of the second output winding undergoes a phase reversal and a sign reversal at a period obtained by adding an integer multiple of the second period TB to the half-period of the second period TB / 2 (for example, the half-period of the second period TB / 2). Therefore, by adding the output voltages of the two second systems, which are shifted by a period obtained by adding an integer multiple of the second period TB to the half-period of the second period TB / 2, the components of the second period TB can be canceled out, and the component V1B_TA of the first period TA can be extracted.

[0093] Therefore, as shown in Figure 16, the first system component extraction unit 46 is equipped with a first extraction delay unit 461 that outputs the detected value V1B_S of the output voltage of the first output winding of the second system with a delay of the first period extraction processing interval ΔTe2. The output V1B_Sold of the first extraction delay unit 461 is added to the detected value V1B_S of the output voltage of the first output winding of the second system to calculate the first system component V1B_TA included in the detected value V1B_S of the output voltage of the first output winding of the second system. Similarly, the first system component extraction unit 46 includes a second extraction delay unit 462 that outputs the detected value V2B_S of the output voltage of the second output winding of the second system with a delay of ΔTe2. The output V2B_Sold of the second extraction delay unit 462 is added to the detected value V2B_S of the output voltage of the second output winding of the second system to calculate the first system component V2B_TA included in the detected value V2B_S of the output voltage of the second output winding of the second system. The first period extraction interval ΔTe2 is set to half a period of the second period TB / 2 (= first period TA). Alternatively, ΔTe2 may be set to TB / 2 + TB × Q, where Q is a non-negative integer.

[0094] <PWM Period Synchronization Unit 47> The PWM period synchronization unit 47 synchronizes the phase of the PWM period Tpwm2 of the second system to the phase of the PWM period Tpwm1 of the first system, based on the components of the first system of the two output windings of the second system that have been extracted.

[0095] With this configuration, based on the components of the first system from the two output windings of the extracted second system, information on the relative phase of the AC voltage VRA of the first system, which is synchronized with the phase of the AC voltage VRB of the second system, with respect to the phase of the AC voltage VRB of the second system, which is synchronized with the phase of the AC voltage VRB of the second system, with respect to the phase of the AC voltage VRB of the second system, which is synchronized with the phase of the AC voltage VRB of the second system, can be obtained. Similarly, information on the relative phase of the phase of the AC voltage VRA of the first system with respect to the phase of the AC voltage Tpwm1 of the second system, with respect to the phase of the AC voltage Tpwm2 of the second system, can be obtained. Therefore, the phase of the AC voltage Tpwm2 of the second system can be synchronized with the phase of the AC voltage Tpwm1 of the first system using the relative phase information. Since there is no need to perform communication for synchronization between the control circuit 6A of the first system and the control circuit 6B of the second system, the cost of the device can be reduced, and the reliability against communication abnormalities can be improved.

[0096] If the on / off timing of the first PWM signal coincides with the current detection timing of the second system, the on / off switching of the applied voltage to the armature winding of the first system will disrupt the current detection value of the armature winding of the second system, degrading the current detection accuracy. Therefore, by synchronizing the phase of the second system's PWM period Tpwm2 with the phase of the first system's PWM period Tpwm1, the current detection timing of the second system can be prevented from overlapping with the on / off timing of the first system's PWM signal, thereby improving the accuracy of the current detection value of the armature winding of the second system. Similarly, the current detection timing of the first system can be prevented from overlapping with the on / off timing of the second system's PWM signal, thereby improving the accuracy of the current detection value of the armature winding of the first system. Thus, the control accuracy of the motor can be improved.

[0097] Specifically, as shown in Figure 8, the on / off timing of the first PWM signal of each switching element is usually located between the start timing of the first PWM period Tpwm1 (the peak of the first carrier wave CA1) and the center timing of the first PWM period Tpwm1 (the trough of the first carrier wave CA1), and between the center timing of the first PWM period Tpwm1 (the trough of the first carrier wave CA1) and the end timing of the first PWM period Tpwm1 (the peak of the first carrier wave CA1). Furthermore, even when two-phase modulation is applied, the on / off timing of the two-phase PWM signals that are not attached to +Vdc1 / 2 or -Vdc1 / 2 is usually located between the peaks and troughs, and between the troughs and peaks, of the first carrier wave CA1.

[0098] Furthermore, the current detection timing of the second system is synchronized with the phase of the second system's PWM period Tpwm2 (in this example, it is set to the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2)). Therefore, by synchronizing the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2) with the start timing or center timing of the first system's PWM period Tpwm1 (the peak or trough of the first system's carrier wave CA1), the current detection timing of the second system can be prevented from overlapping with the on / off timing of the first system's PWM signal, thereby improving the accuracy of the current detection value of the second system's armature winding.

[0099] In this embodiment, the PWM period synchronization unit 47 calculates the sum of the squares of the first system components V1B_TA and V2B_TA of the two extracted second system output windings, Amp1, as shown in the following equation: Amp1 = V1B_TA 2 +V2B_TA 2 ... (3)

[0100] In this embodiment, as described above, the phases of the output voltages of the first output winding of the second system and the second output winding of the second system are configured to be different by 90 degrees in electrical angle. An AC voltage VRA of the first period TA is applied to the excitation winding 10A of the first system. Therefore, as shown in FIG. 17 and the following equations, the components V1B_TA and V2B_TA of the first system included in the output voltages of the two output windings of the second system oscillate in the first period TA, and the amplitude of V1B_TA changes in a sine wave shape (sin θ) according to the rotor angle θ, and the amplitude of V1B_TA changes in a cosine wave shape (cos θ) according to the rotor angle θ. Here, A is the amplitude.V1B_TA = A × sin θ × sin(2π / TA × t)V2B_TA = A × cos θ × sin(2π / TA × t) ··· (4) Amp1 = A 2 × sin 2 (2π / TA × t) = A 2 × {1 - cos(4π / TA × t)} / 2

[0101] The sum of squares Amp1 oscillates in a period that is 1 / 2 times the first period TA from 0 to the maximum value A. And the timing when the sum of squares Amp1 becomes the maximum value A is the timing when the AC voltage VRA of the first period TA becomes a peak or a valley. Therefore, according to the phase shift amount of the detection timing of the output voltages of the two output windings of the second system with respect to the timing of the peak or valley of the AC voltage VRA of the first period TA, the sum of squares Amp1 changes between 0 and the maximum value A. 2 2 to A. 2 <00, in the case where the sum of squares Amp1 is the maximum value A, the timing of the peak or valley of the carrier wave CA1 of the first system coincides with the timing of the peak of the carrier wave CA2 of the second system. Therefore, based on the sum of squares Amp, the relative phase between the phase of the PWM period of the first system and the phase of the PWM period of the second system can be determined.

[0102] 2 In this embodiment, the timings of the peaks and valleys of the AC voltage VRA of the first period TA coincide with the timings of the peaks and valleys of the carrier wave CA1 of the first system, respectively. The detection timings of the output voltages of the two output windings of the second system coincide with the timing of the peak of the carrier wave CA2 of the second system. Therefore, when the sum of squares Amp1 is the maximum value A, the timing of the peak or valley of the carrier wave CA1 of the first system coincides with the timing of the peak of the carrier wave CA2 of the second system. Therefore, based on the sum of squares Amp1, the relative phase between the phase of the PWM period of the first system and the phase of the PWM period of the second system can be determined.

[0103] The PWM period synchronization unit 47 sets the sum of squares Amp1 to the maximum value A of the sum of squares Amp1. 2 The phase of the second PWM period Tpwm2 is changed to approach this value.

[0104] In this configuration, the sum of squares Amp1 is the maximum value A 2 As the signal approaches this point, the peak of the second carrier wave CA2 is synchronized with the peak or trough of the first carrier wave CA1. Therefore, the current detection timing of the second system, which is set to the peak of the second carrier wave CA2, can be prevented from overlapping with the on / off timing of the first system's PWM signal, which is set to points other than the peaks and troughs of the first carrier wave CA1, thereby improving the accuracy of the current detection value of the second system's armature winding.

[0105] In this embodiment, the PWM period synchronization unit 47 sets the sum of squares Amp1 to the maximum value A of the sum of squares Amp1. 2 If it is determined that the value has decreased, the PWM period Tpwm2 of the second system is changed from the specified period Tpwm20, which is set to the same period as the PWM period Tpwm1 of the first system, and the sum of squares Amp1 is changed to the maximum value A of the sum of squares Amp1. 2 If it is determined that the value has not decreased, the PWM period Tpwm2 of the second system is set to the specified period Tpwm20. Tpwm20 = Tpwm1.

[0106] For example, the system is configured as shown in the flowchart of Figure 18. The processing in the flowchart of Figure 18 is performed for each PWM period Tpwm2 of the second system (the peak of the carrier wave CA2 of the second system). In step S01, the PWM period synchronization unit 47 calculates the sum of squares Amp1 based on the components V1B_TA and V2B_TA of the first system of the two output windings of the second system, which were detected and calculated in the current calculation period.

[0107] Then, in step S02, the PWM period synchronization unit 47 sets the sum of squares Amp1 to the maximum value A of the sum of squares Amp1. 2It is determined whether the sum of squares Amp1 is greater than or equal to the synchronization judgment value Ath, which is set lower than the specified value. If the sum of squares Amp1 is greater than or equal to the synchronization judgment value Ath, the process proceeds to step S03. If the sum of squares Amp1 is less than the synchronization judgment value Ath, the process proceeds to step S04. Based on this determination, the sum of squares Amp1 is determined to be greater than or equal to the maximum value A of the sum of squares A. 2 It is determined whether the value has come closer to the threshold value. The synchronization threshold value Ath is the maximum value A of the sum of squares Amp1. 2 It is set by multiplying by a coefficient smaller than 1 (for example, 0.95). The synchronization judgment value Ath is the maximum value A of the previously measured sum of squares Amp1. 2 The coefficients may be set in advance based on the above, or they may be set adaptively based on the maximum value of multiple previously calculated sums of squares Amp1. The coefficients are determined to avoid misjudgments, taking into account detection calculation errors and variability.

[0108] In step S03, since the timing of the peak of the second carrier wave CA2 is closer than or equal to the timing of the peak or trough of the first carrier wave CA1 by a predetermined value, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a predetermined period Tpwm20, which is the same period as the PWM period Tpwm1 of the first PWM signal (Tpwm2 = Tpwm20 = Tpwm1), and ends the relative phase change.

[0109] In step S04, if the timing of the peak of the second carrier wave CA2 is shifted by more than a certain value from the timing of the peak or trough of the first carrier wave CA1, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a period that is changed by a period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 + ΔT = Tpwm1 + ΔT), and performs a relative phase change. In this example, the period change amount ΔT is set to a negative value, but it may also be set to a positive value. The absolute value of the period change amount ΔT is set to a value of 10% or less or 1% or less of the specified period Tpwm20. The changed PWM period Tpwm2 of the second system is reflected in the generation of the second carrier wave CA2, the generation of the second AC voltage VRB, and the setting of each detection timing.

[0110] As shown in the time chart in Figure 19, when the PWM period Tpwm2 of the second system changes by an amount ΔT from the specified period Tpwm20, which is set to be the same period as the PWM period Tpwm1 of the first system, the relative phase of the timing of the peak of the carrier wave CA2 of the second system with respect to the timing of the peak or trough of the carrier wave CA1 of the first system changes by an amount of relative phase change corresponding to the period change amount ΔT. Then, at time t5, when the sum of squares Amp1 becomes greater than or equal to the synchronization criterion Ath and the relative phase approaches 0, the PWM period Tpwm2 of the second system is set to the specified period Tpwm20, which is set to be the same period as the PWM period Tpwm1 of the first system, the relative phase is fixed in a state close to 0, and the change in relative phase ends.

[0111] Furthermore, variations in the timer function (clock circuit) of the first control circuit 6A and the timer function (clock circuit) of the second control circuit 6B result in a difference between the PWM period Tpwm1 of the first system and the PWM period Tpwm2 of the second system. As a result, the relative phase gradually increases even after the relative phase change has finished. In this case as well, when the sum of squares Amp1 falls below the synchronization judgment value Ath again, the PWM period Tpwm2 of the second system is set to Tpwm20 + ΔT, the relative phase change resumes, and synchronization is performed.

[0112] Furthermore, the second period TB may be set to the same period as the second PWM period Tpwm2, and the first period TA may be set to twice the period of the second period TB. In this case, the sum of squares Amp1 oscillates with a period that is half the period of the first period TA, i.e., the second period TB and the second PWM period Tpwm2. Therefore, by changing the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches its maximum value, the peak of the carrier wave CA2 of the second system can be synchronized with the peak of the carrier wave CA1 of the first system.

[0113] Furthermore, if the second period TB is set to the same period as the second system's PWM period Tpwm2, and the first period TA is set to twice the period of the second period TB, the PWM period synchronization unit 47 may change the phase of the second system's PWM period Tpwm2 so that the sum of squares Amp1 approaches the minimum value of the sum of squares Amp1. In this case, the peaks of the second system's carrier wave CA2 can be synchronized with the troughs of the first system's carrier wave CA1.

[0114] 2. Embodiment 2 Next, a motor device according to Embodiment 2 will be described. The same components as in Embodiment 1 will not be described. The basic configuration and processing of the motor device according to this embodiment are the same as in Embodiment 1, but the processing of the second output voltage detection unit 41 and the PWM period synchronization unit 47 differs from that of Embodiment 1.

[0115] In this embodiment, the second system output voltage detection unit 41 detects the output voltages V1A_S and V2A_S of the two output windings of the first system at multiple timings synchronized with the phase of the second system's PWM period Tpwm2.

[0116] As shown in Figure 20, the multiple timings are set to three timings: a reference timing, a pre-timing immediately preceding the reference timing, and a post-timing immediately following the reference timing. The reference timing is set to the start timing of the second PWM period Tpwm2 (the peak of the second carrier wave CA2, or the maximum or minimum value of the second AC voltage VRB), similar to Embodiment 1. The pre-timing and post-timing are set symmetrically with respect to the reference timing. Note that the multiple timings only need to be set to two or more timings, including the reference timing.

[0117] The first system component extraction unit 46 extracts the first system components V1B_TA and V2B_TA, which are generated by magnetic interference and are included in the detected output voltage values ​​V1B_S and V2B_S of the two output windings of the second system detected at each timing. The extraction process is the same as in Embodiment 1, so the explanation is omitted. In addition, the angle θ2 of the second system is calculated using the output voltages V1A_S and V2A_S detected at the reference timing, as in Embodiment 1.

[0118] The PWM periodic synchronization unit 47 calculates the sum of squares Amp1 of the first system components V1B_TA and V2B_TA of the two output windings of the second system for each timing. The calculation of the sum of squares Amp1 is the same as in Embodiment 1, so the explanation is omitted.

[0119] The PWM period synchronization unit 47 changes the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the maximum value of the sum of squares Amp1, based on the relative magnitudes of multiple sums of squares Amp1 corresponding to multiple timings.

[0120] By examining the relative magnitudes of multiple sums of squares Amp1 corresponding to multiple timings, it is possible to obtain information on whether the sum of squares Amp1 of multiple timings is increasing, decreasing, or near its maximum value. Accordingly, by advancing, delaying, or maintaining the phase of the second PWM period Tpwm2, the sum of squares Amp1 of the reference timing can be brought closer to its maximum value, and the phase of the second PWM period Tpwm2 can be synchronized with the phase of the first PWM period Tpwm1.

[0121] In this embodiment, the PWM period synchronization unit 47 determines the sum of squares Amp1 based on the relative magnitudes of the multiple sums of squares Amp1 corresponding to multiple timings, and the maximum value of the sum of squares A 2 Determine whether the period falls within a period containing the specified value, a period in which the sum of squares Amp1 decreases, or a period in which the sum of squares Amp1 increases, and then determine if the sum of squares is at its maximum value A 2 If the period includes the specified period, the PWM period Tpwm2 of the second system is set to a specified period Tpwm20, which is the same period as the PWM period of the first system's PWM signal. If the period includes the period in which the sum of squares Amp1 decreases, the PWM period Tpwm2 of the second system is set to a period lower than the specified period Tpwm20. If the period includes the period in which the sum of squares Amp1 increases, the PWM period Tpwm2 of the second system is set to a period higher than the specified period Tpwm20.

[0122] For example, the system is configured as shown in the flowchart of Figure 21. The processing in the flowchart of Figure 21 is performed for each second PWM period Tpwm2 (peak of the second carrier wave CA2). In step S11, the PWM period synchronization unit 47 calculates the sum of the squares of three timings Amp1 based on the first system components V1B_TA and V2B_TA of the two output windings of the second system, which are detected and calculated in synchronization with the start timing of the current second PWM period Tpwm2 (peak of the second carrier wave CA2). Specifically, based on the first system components V1B_TA and V2B_TA of the reference timing, the sum of squares Amp10 of the reference timing is calculated; based on the first system components V1B_TA and V2B_TA of the previous timing, the sum of squares Amp1bf of the previous timing is calculated; and based on the first system components V1B_TA and V2B_TA of the later timing, the sum of squares Amp1af of the later timing is calculated.

[0123] Then, in step S12, the PWM period synchronization unit 47 determines that if the sum of squares of the reference timings Amp10 is greater than or equal to the sum of squares of the subsequent timings Amp1af, and the sum of squares of the reference timings Amp10 is greater than or equal to the sum of squares of the preceding timings Amp1bf, then the sum of squares of the reference timings Amp10 is equal to the maximum value of the sum of squares A 2 If it is determined that the period includes the specified period, proceed to step S13; otherwise, proceed to step S14.

[0124] In step S13, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a predetermined period Tpwm20, which is the same period as the PWM period Tpwm1 of the first system PWM signal (Tpwm2 = Tpwm20 = Tpwm1), and ends the relative phase change.

[0125] In step S14, the PWM period synchronization unit 47 determines that it is in a period in which the sum of squares of the reference timings Amp10 is decreasing if the sum of squares of the reference timings Amp10 is greater than or equal to the sum of squares of the subsequent timings Amp1af and less than or equal to the sum of squares of the previous timings Amp1bf, or if the sum of squares of the subsequent timings Amp1af is greater than or equal to the sum of squares of the reference timings Amp10 and less than or equal to the sum of squares of the previous timings Amp1bf, and proceeds to step S15. Otherwise, it determines that it is in a period in which the sum of squares of the reference timings Amp10 is increasing, and proceeds to step S16.

[0126] As shown in Figure 22, when the period in which the sum of squares of the reference timings Amp10 is decreasing is in which the sum of squares of the reference timings Amp10 is in phase leading relative to the maximum value, in step S15, the PWM period synchronization unit 47 sets the second system PWM period Tpwm2 to a period that is decreased by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 - ΔT = Tpwm1 - ΔT, ΔT > 0), and performs a relative phase change that delays the phase of the second system PWM period Tpwm2. As a result, the sum of squares of the reference timings Amp10 increases and approaches the maximum value. In the example in Figure 22, in the next second system PWM period Tpwm2, the sum of squares of the reference timings Amp10 approaches the maximum value of the sum of squares A 2 It is determined that the system is located within the specified period, and the PWM period Tpwm2 of the second system is set to the specified period Tpwm20, and the relative phase change has ended. The period change amount ΔT is set to a value of 10% or less or 1% or less of the specified period Tpwm20.

[0127] On the other hand, as shown in Figure 23, when the period is in which the sum of squares of the reference timings Amp10 is increasing, the sum of squares of the reference timings Amp10 is in phase lag relative to the maximum value. Therefore, in step S16, the PWM period synchronization unit 47 sets the second system PWM period Tpwm2 to a period increased by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 + ΔT = Tpwm1 + ΔT, ΔT > 0), and performs a relative phase change that advances the phase of the second system PWM period Tpwm2. As a result, the sum of squares of the reference timings Amp10 increases and approaches the maximum value. In the example in Figure 23, in the next second system PWM period Tpwm2, the sum of squares of the reference timings Amp10 approaches the maximum value of the sum of squares A 2 It is determined that the period includes the specified time, and the PWM period Tpwm2 of the second system is set to the specified period Tpwm20, and the relative phase change has ended.

[0128] The modified PWM period Tpwm2 of the second system is reflected in the generation of the carrier wave CA2 of the second system, the generation of the AC voltage VRB of the second system, and the setting of each detection timing.

[0129] In this embodiment, unlike in Embodiment 1, a synchronization judgment value Ath set lower than the maximum value is not used. Therefore, there is no risk of the judgment accuracy deteriorating due to setting errors in the synchronization judgment value Ath. Thus, robustness against variability factors is improved.

[0130] Furthermore, the second period TB may be set to the same period as the second PWM period Tpwm2, and the first period TA may be set to twice the period of the second period TB. In this case, the sum of squares Amp1 oscillates with a period that is half the period of the first period TA, i.e., the second period TB and the second PWM period Tpwm2. Therefore, by changing the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches its maximum value, the peak of the carrier wave CA2 of the second system can be synchronized with the peak of the carrier wave CA1 of the first system.

[0131] Furthermore, if the second period TB is set to the same period as the second system's PWM period Tpwm2, and the first period TA is set to twice the period of the second period TB, then, similar to Embodiment 6 described later, the PWM period synchronization unit 47 may change the phase of the second system's PWM period Tpwm2 based on the relative magnitudes of multiple sums of squares Amp1 corresponding to multiple timings, so that the sum of squares Amp1 approaches the minimum value of the sum of squares Amp1. In this case, the peaks of the second system's carrier wave CA2 can be synchronized with the troughs of the first system's carrier wave CA1.

[0132] 3. Embodiment 3 Next, a motor device according to Embodiment 3 will be described. The same components as in Embodiment 1 will not be described. The basic configuration and processing of the motor device according to this embodiment are the same as in Embodiment 1, but the processing of the second output voltage detection unit 41 and the PWM period synchronization unit 47 differs from that of Embodiment 1.

[0133] In this embodiment, similar to embodiment 2, the second system output voltage detection unit 41 detects the output voltages V1A_S and V2A_S of the two output windings of the first system at multiple timings synchronized with the phase of the second system's PWM period Tpwm2.

[0134] Similar to Embodiment 2, the multiple timings are set to three timings: a reference timing, a pre-timing immediately preceding the reference timing, and a post-timing immediately following the reference timing. The reference timing is set to the start timing of the second PWM period Tpwm2 (the peak of the second carrier wave CA2, or the maximum or minimum value of the second AC voltage VRB), similar to Embodiment 1. The pre-timing and post-timing are set symmetrically with respect to the reference timing. Note that the multiple timings only need to be set to two or more timings, including the reference timing.

[0135] The first system component extraction unit 46 extracts the first system components V1B_TA and V2B_TA, which are generated by magnetic interference and are included in the detected output voltage values ​​V1B_S and V2B_S of the two output windings of the second system detected at each timing. The extraction process is the same as in Embodiment 1, so the explanation is omitted. In addition, the angle θ2 of the second system is calculated using the output voltages V1A_S and V2A_S detected at the reference timing, as in Embodiment 1.

[0136] In this embodiment, the PWM periodic synchronization unit 47 calculates the deviation Diff of the first system components V1B_TA and V2B_TA of the two output windings of the second system at each timing. In this embodiment, as shown in the following equation, for each timing, the PWM periodic synchronization unit 47 calculates the deviation Diff by subtracting the first system component V1B_TA of the first output winding of the second system from the first system component V2B_TA of the second output winding of the second system. Diff = V2B_TA - V1B_TA ... (5)

[0137] Figure 24 shows a time chart of the AC voltage VRA, the components V1B_TA and V2B_TA of the first system of the two output windings, and the deviation Diff. The deviation Diff oscillates in the first period TA, similar to the components V1B_TA and V2B_TA of the first system of the two output windings, and its amplitude changes sinusoidally according to the rotor angle θ.

[0138] In this embodiment, the timing of the peaks and troughs of the AC voltage VRA in the first period TA coincides with the timing of the peaks and troughs of the carrier wave CA1 of the first system, respectively. The detection timing of the output voltages of the two output windings of the second system coincides with the timing of the peaks of the carrier wave CA2 of the second system. Therefore, when the deviation Diff is at a peak (maximum value, local maximum), the timing of the peak of the carrier wave CA1 of the first system and the timing of the peak of the carrier wave CA2 of the second system coincide. Thus, the relative phase between the phase of the PWM period of the first system and the phase of the PWM period of the second system can be determined based on the deviation Diff. However, since the amplitude of the deviation Diff oscillating in the first period TA varies according to the rotor angle θ, it is not possible to determine it using a constant determination value as in Embodiment 1. Furthermore, in the angular interval of the rotor angle θ where the amplitude of the deviation Diff oscillating in the first period TA is small, the determination accuracy using the deviation Diff deteriorates due to the S / N ratio, etc.

[0139] The PWM period synchronization unit 47 changes the phase of the second PWM period Tpwm2 based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, so that the deviation Diff approaches the maximum value of the deviation Diff.

[0140] By examining the relative magnitudes of multiple deviations (Diff) corresponding to multiple timings, it is possible to obtain information on whether the deviations (Diff) of multiple timings are increasing, decreasing, or near their maximum value (maximum value). Accordingly, by advancing, delaying, or maintaining the phase of the second PWM period Tpwm2, the deviation (Diff) of the reference timing can be brought closer to its maximum value, and the phase of the second PWM period Tpwm2 can be synchronized with the phase of the first PWM period Tpwm1.

[0141] In this embodiment, the PWM period synchronization unit 47 determines, based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, whether the deviation Diff is in a period containing the maximum value of the deviation Diff, a period in which the deviation Diff is decreasing, or a period in which the deviation Diff is increasing. If the deviation Diff is in a period containing the maximum value, the PWM period Tpwm2 of the second system is set to a predetermined period Tpwm20, which is the same period as the PWM period of the first system PWM signal. If the deviation Diff is in a period of decreasing deviation, the PWM period Tpwm2 of the second system is reduced to less than the predetermined period Tpwm20. If the deviation Diff is in a period of increasing deviation, the PWM period Tpwm2 of the second system is increased to less than the predetermined period Tpwm20.

[0142] For example, the system is configured as shown in the flowchart of Figure 25. The processing in the flowchart of Figure 25 is performed for each second PWM period Tpwm2 (peak of the second carrier wave CA2). In step S21, the PWM period synchronization unit 47 calculates the three timing deviations Diff based on the first system components V1B_TA and V2B_TA of the two output windings of the second system, which are detected and calculated in synchronization with the start timing of the current second PWM period Tpwm2 (peak of the second carrier wave CA2). Specifically, the deviation Diff0 of the reference timing is calculated based on the first system components V1B_TA and V2B_TA of the reference timing, the deviation Diffbf of the previous timing is calculated based on the first system components V1B_TA and V2B_TA of the previous timing, and the deviation Diffaf of the later timing is calculated based on the first system components V1B_TA and V2B_TA of the later timing.

[0143] Then, in step S22, the PWM period synchronization unit 47 determines whether the absolute value of the reference timing deviation Diff0, the absolute value of the preceding timing deviation Diffbf, and the absolute value of the succeeding timing deviation Diffaf are all less than the judgment permission threshold Dth. If they are less than the judgment permission threshold Dth, the unit proceeds to step S23; otherwise, the unit proceeds to step S24. In step S23, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a specified period Tpwm20, which is set to be the same period as the PWM period Tpwm1 of the first system PWM signal (Tpwm2 = Tpwm20 = Tpwm1), and does not change the relative phase.

[0144] As mentioned above, in the angular interval of rotor angle θ where the amplitude of the deviation Diff oscillating in the first period TA is small, the judgment accuracy using the deviation Diff deteriorates due to the S / N ratio, etc. When the absolute values ​​of the deviations Diff0, Diffbf, and Diffaf at each timing are close to 0, the amplitude of the deviation Diff is small, and the judgment accuracy deteriorates, so the relative phase change is interrupted to suppress the deterioration of synchronization accuracy.

[0145] In step S24, the PWM period synchronization unit 47 determines that if the deviation of the reference timing Diff0 is greater than or equal to the deviation of the subsequent timing Diffaf, and the deviation of the reference timing Diff0 is greater than or equal to the deviation of the previous timing Diffbf, then the deviation of the reference timing Diff0 is in a period that includes the maximum value of the deviation Diff, and proceeds to step S23; otherwise, proceeds to step S25.

[0146] In step S23, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a predetermined period Tpwm20, which is the same period as the PWM period Tpwm1 of the first system PWM signal, and ends the relative phase change.

[0147] In step S25, the PWM period synchronization unit 47 determines that it is in a period in which the deviation of the reference timing, Diff0, decreases if the deviation of the reference timing, Diff0, is greater than or equal to the deviation of the later timing, Diffaf, and less than or equal to the deviation of the earlier timing, or if the deviation of the later timing, Diffaf, is greater than or equal to the deviation of the reference timing, Diff0, and less than or equal to the deviation of the earlier timing, and proceeds to step S26. Otherwise, it determines that it is in a period in which the deviation of the reference timing, Diff0, decreases, and proceeds to step S27.

[0148] As shown in Figure 26, when the reference timing deviation Diff0 is in a period of decrease, the reference timing deviation Diff0 is in a phase leading position relative to its maximum value. Therefore, in step S26, the PWM period synchronization unit 47 sets the second system's PWM period Tpwm2 to a period that is a positive period change amount ΔT less than the specified period Tpwm20 (Tpwm2 = Tpwm20 - ΔT = Tpwm1 - ΔT, ΔT > 0), and performs a relative phase change that delays the phase of the second system's PWM period Tpwm2. As a result, the reference timing deviation Diff0 increases and approaches its maximum value. In the example in Figure 26, in the next second system's PWM period Tpwm2, it is determined that the reference timing deviation Diff0 is in a period that includes the maximum value of the deviation Diff, and the second system's PWM period Tpwm2 is set to the specified period Tpwm20, and the relative phase change is completed. The periodic change ΔT is set to a value of 10% or less, or 1% or less, of the specified period Tpwm20.

[0149] On the other hand, as shown in Figure 27, when the reference timing deviation Diff0 is in an increasing phase, the reference timing deviation Diff0 is in a phase lag relative to its maximum value. Therefore, in step S27, the PWM period synchronization unit 47 sets the second system PWM period Tpwm2 to a period increased by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 + ΔT = Tpwm1 + ΔT, ΔT > 0), and performs a relative phase change that advances the phase of the second system PWM period Tpwm2. As a result, the reference timing deviation Diff0 increases and approaches its maximum value. In the example shown in Figure 27, it is determined that the reference timing deviation Diff0 is located within the period containing the maximum value of the deviation Diff, at the next PWM period Tpwm2 of the second system. As a result, the PWM period Tpwm2 of the second system is set to the specified period Tpwm20, and the relative phase change is completed.

[0150] The modified PWM period Tpwm2 of the second system is reflected in the generation of the carrier wave CA2 of the second system, the generation of the AC voltage VRB of the second system, and the setting of each detection timing.

[0151] Furthermore, similar to Embodiment 6 described later, the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, so that the deviation Diff approaches the minimum value of the deviation Diff.

[0152] Alternatively, the PWM period synchronization unit 47 may calculate the deviation Diff by subtracting the first system component V2B_TA of the second output winding of the second system from the first system component V1B_TA of the first output winding of the second system (Diff = V1B_TA - V2B_TA).

[0153] Furthermore, the PWM period synchronization unit 47 may use the first period component V1B_TA of the output voltage of the second output winding of the second system, or the first period component V2B_TA of the output voltage of the second output winding of the second system, instead of the deviation Diff.

[0154] 4. Embodiment 4 Next, a motor device according to Embodiment 4 will be described. The same components as in Embodiment 1 will not be described. The basic configuration and operation of the motor device according to this embodiment are the same as in Embodiment 1, but it differs from Embodiment 1 in that it uses magnetic interference between the first armature winding and the second armature winding to synchronize the phase of the PWM period of the second system with the phase of the PWM period of the first system. The following will focus on the differences from Embodiment 1.

[0155] Figure 28 shows a schematic configuration diagram of the motor device according to this embodiment, Figure 29 shows a block diagram of the first control circuit 6A according to this embodiment, and Figure 30 shows a block diagram of the second control circuit 6B according to this embodiment.

[0156] Similar to Embodiment 1, the first armature winding and the second armature winding magnetically interfere with each other. Specifically, as shown in Figure 31, there is mutual inductance between the systems. There is mutual inductance Mu between the armature winding Cu1 of the U1 phase and the armature winding Cu2 of the U2 phase, there is mutual inductance Mv between the armature winding Cv1 of the V1 phase and the armature winding Cv2 of the V2 phase, and there is mutual inductance Mw between the armature winding Cw1 of the W1 phase and the armature winding Cw2 of the W2 phase.

[0157] In this embodiment, an MR sensor 21 is used as the rotational position sensor instead of the resolver 2. The MR sensor 21 outputs the sine value Sin1 and cosine value Cos1 of the first system to the first control circuit 6A, and outputs the sine value Sin2 and cosine value Cos2 of the second system to the second control circuit 6B.

[0158] A Hall element or an encoder may be used as the rotational position sensor. Alternatively, a resolver 2 similar to that in Embodiment 1 may be used as the rotational position sensor, with the first control circuit 6A calculating the first system angle θ1 similarly to Embodiment 1, and the second control circuit 6B calculating the second system angle θ2 similarly to Embodiment 2. Or, without a rotational position sensor, each control circuit 6A and 6B may be configured to estimate the rotational angle based on current information obtained by superimposing harmonic components on the current command value (a so-called sensorless method).

[0159] Unlike Embodiment 1, the first control circuit 6A does not include a first-system excitation application unit 30, a first-system output voltage detection unit 31, and a first-system removal processing unit 32. The first-system angle calculation unit 33 calculates the first-system angle θ1 based on the first-system sine value Sin1 and cosine value Cos1 output from the MR sensor 21 using the following equation: θ1 = tan -1 (Sin1 / Cos1) ... (6)

[0160] Unlike the first embodiment, the second control circuit 6A does not include a second excitation application unit 40, a second output voltage detection unit 41, and a second removal processing unit 42. The second angle calculation unit 43 calculates the second angle θ2 based on the sine value Sin2 and cosine value Cos2 of the second system output from the MR sensor 21 using the following equation: θ2 = tan -1 (Sin2 / Cos2) ... (7)

[0161] <First System Voltage Command Value Calculation Unit 34> The first system voltage command value calculation unit 34 shifts the average value of the three phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system at the rotation period of the electrical angle from the center voltage (the oscillation center value of the carrier wave CA1 of the first system, 0V in this example) to the lower potential side or the higher potential side. For example, two-phase modulation is applied to the sinusoidal three phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref.

[0162] In this embodiment, a downward two-phase modulation is applied to shift the average value to the lower potential side. Specifically, the first system voltage command value calculation unit 34 determines the minimum value Vmin1 of the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system, as shown in the following equation, adds half the DC voltage of the first system Vdc1 / 2 to the minimum value Vmin1 to calculate the zero-sequence voltage V01, subtracts the zero-sequence voltage V01 from the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref to calculate the modulated three-phase voltage command values ​​Vu1_ref', Vv1_ref', and Vw1_ref'. Carrier comparison PWM control is performed using the modulated three-phase voltage command values ​​to generate a PWM signal. Vmin1=MIN(Vu1_ref, Vv1_ref, Vw1_ref) V01=Vmin1+Vdc1 / 2 Vu1_ref'=Vu1_ref-V01...(8) Vv1_ref'=Vv1_ref-V01 Vw1_ref'=Vw1_ref-V01

[0163] As shown in Figure 32, which illustrates the voltage command values ​​of the three phases before and after the two-phase modulation described below, the voltage command values ​​of the three phases are shifted to the lower potential side so that the voltage command value of the phase with the lowest voltage matches -Vdc1 / 2.

[0164] Alternatively, a superposition two-phase modulation may be applied to shift the average value to the higher potential side. Specifically, the first system voltage command value calculation unit 34 determines the maximum value Vmax1 of the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system, subtracts half the DC voltage of the first system, Vdc1 / 2, from the maximum value Vmax1 to calculate the zero-sequence voltage V01, and adds the zero-sequence voltage V01 to the three-phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref to calculate the modulated three-phase voltage command values ​​Vu1_ref', Vv1_ref', and Vw1_ref'. In this case, the three-phase voltage command values ​​are shifted to the higher potential side so that the voltage command value of the phase with the maximum voltage matches +Vdc1 / 2. Vmax1=MAX(Vu1_ref, Vv1_ref, Vw1_ref) V01=Vmax1-Vdc1 / 2 Vu1_ref'=Vu1_ref-V01...(9) Vv1_ref'=Vv1_ref-V01 Vw1_ref'=Vw1_ref-V01

[0165] Furthermore, the lower two-phase modulation and the upper two-phase modulation may be switched alternately. This reduces the imbalance between the heat generated by the switching elements on the high-potential side and the heat generated by the switching elements on the low-potential side.

[0166] <Second System Voltage Command Value Calculation Unit 44> The second system voltage command value calculation unit 44 matches the average value of the three phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the second system at the rotation period of the electrical angle to the center voltage (the oscillation center value of the carrier wave CA2 of the second system, which is 0V in this example). For example, no modulation is applied to the sinusoidal three phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref, or modulation is applied by superimposing a zero-sequence voltage at which the average value at the rotation period of the electrical angle becomes 0 (for example, third harmonic superposition, min-max method (pseudo third harmonic superposition)).

[0167] <First System Component Extraction Unit 46> In this embodiment, the first system component extraction unit 46 extracts the first system component caused by magnetic interference, which is included in the current detection value of the second system armature winding. Then, the PWM period synchronization unit 47 synchronizes the phase of the second system's PWM period Tpwm1 to the phase of the first system's PWM period Tpwm2 based on the extracted first system component. In this embodiment, the first system's PWM period Tpwm1 and the second system's PWM period Tpwm2 are the same PWM period.

[0168] With this configuration, based on the component of the first system included in the extracted current detection value of the second system's armature winding, information on the relative phase of the first system's component synchronized with the phase of the first system's PWM period Tpwm1 with respect to the phase of the second system's PWM period Tpwm2 can be obtained, and information on the relative phase of the first system's PWM period Tpwm1 with respect to the phase of the second system's PWM period Tpwm2 can be obtained. Therefore, the phase of the second system's PWM period Tpwm2 can be synchronized with the phase of the first system's PWM period Tpwm1 using the relative phase information. Since there is no need to perform communication for synchronization between the first system's control circuit 6A and the second system's control circuit 6B, the cost of the device can be reduced, and reliability against communication abnormalities can be improved.

[0169] As described in Embodiment 1, when the on / off timing of the first PWM signal coincides with the current detection timing of the second system, the on / off switching of the applied voltage to the armature winding of the first system disrupts the current detection value of the armature winding of the second system, degrading the current detection accuracy. Therefore, by synchronizing the phase of the second system's PWM period Tpwm2 with the phase of the first system's PWM period Tpwm1, the current detection timing of the second system can be prevented from overlapping with the on / off timing of the first system's PWM signal, thereby improving the accuracy of the current detection value of the armature winding of the second system. Similarly, the current detection timing of the first system can be prevented from overlapping with the on / off timing of the second system's PWM signal, thereby improving the accuracy of the current detection value of the armature winding of the first system. Thus, the control accuracy of the motor can be improved.

[0170] Similar to Embodiment 1, the second system voltage command value calculation unit 44 detects the current flowing through the three-phase armature windings of the second system at a current detection timing synchronized with the phase of the second system's PWM period Tpwm2. The current detection timing for the second system is set to the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2). Alternatively, the current detection timing for the second system may be set to the center timing of the second system's PWM period Tpwm2 (the trough of the second system's carrier wave CA2). Or, the current detection timing for the second system may be set to the start timing and center timing of the second system's PWM period Tpwm2 (the peak and trough of the second system's carrier wave CA2). Or, the current detection timing for the second system may be set to the start timing of the second system's PWM period Tpwm2 every two cycles (the peak of the second system's carrier wave CA2). In any case, the current detection timing of the second system coincides with either or both of the start timing and / or center timing of the second system's PWM period Tpwm2 (the peaks and troughs of the second system's carrier wave CA2).

[0171] Specifically, as shown in Figure 8, the on / off timing of the first PWM signal of each switching element is usually located between the start timing of the first PWM period Tpwm1 (the peak of the first carrier wave CA1) and the center timing of the first PWM period Tpwm1 (the trough of the first carrier wave CA1), and between the center timing of the first PWM period Tpwm1 (the trough of the first carrier wave CA1) and the end timing of the first PWM period Tpwm1 (the peak of the first carrier wave CA1). Furthermore, even when two-phase modulation is applied, the on / off timing of the two-phase PWM signals that are not attached to +Vdc1 / 2 or -Vdc1 / 2 is usually located between the peaks and troughs, and between the troughs and peaks, of the first carrier wave CA1.

[0172] Furthermore, the current detection timing of the second system is synchronized with the phase of the second system's PWM period Tpwm2 (in this example, it is set to the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2)). Therefore, by synchronizing the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2) with the start timing or center timing of the first system's PWM period Tpwm1 (the peak or trough of the first system's carrier wave CA1), the current detection timing of the second system can be prevented from overlapping with the on / off timing of the first system's PWM signal, thereby improving the accuracy of the current detection value of the second system's armature winding.

[0173] Figure 33 shows the voltages Vu1, Vv1, Vw1 and the phase-to-phase voltages Vu1-Vv1, Vv1-Vw1, Vw1-Vu1 at point A (90 degrees) in Figure 32 of the first system, where two-phase modulation is being performed. Point A is the point when the voltage command values, which are stuck at -Vdc1 / 2, are swapped, and the two-phase voltage command values ​​coincide with -Vdc1 / 2. At the timing of the trough of the carrier wave CA1 of the first system, one voltage pulse is generated in the two phase-to-phase voltages Vu1-Vv1, Vw1-Vu1. As a result, due to magnetic interference, the component of the PWM period Tpwm1 of the first system becomes larger in the current detection value of the second system, and the peak (mountain) of that component coincides with the timing of the trough of the carrier wave CA1 of the first system. Similarly, when the upper two-phase modulation is performed, the component of the first system's PWM period Tpwm1 becomes larger, but the peak (mountain) of that component coincides with the timing of the peak of the first system's carrier wave CA1.

[0174] Figure 34 shows the voltages Vu2, Vv2, Vw2 applied to the armature windings of each phase of the second system by the PWM signal, and the inter-phase voltages Vu2-Vv2, Vv2-Vw2, Vw2-Vu2, and Vw2-Vu2 at the 90-degree point of the second system when no shift has occurred. At the 90-degree point, the voltage command value Vv2_ref for the V2 phase and the voltage command value Vw2_ref for the W2 phase are the same. Between the peaks and troughs, and between the troughs and peaks, two voltage pulses are generated in the two inter-phase voltages Vu2-Vv2 and Vw2-Vu2. As a result, in the current detection value of the second system, the component with a period twice that of the second system's PWM period Tpwm2 becomes large.

[0175] In this way, by shifting the voltage command value of the first system to the lower or higher potential side, while not shifting the voltage command value of the second system, the periods of the components of each system superimposed on the current detection value of the second system will differ, and the periodic components of each system can be extracted.

[0176] Therefore, the first system component extraction unit 46 extracts the component of the second system's PWM period Tpwm2, which is included in the detected value of the current of the second system, as a component of the first system. Then, the PWM period synchronization unit 47 synchronizes the phase of the second system's PWM period Tpwm1 to the phase of the first system's PWM period Tpwm2 based on the extracted first system component.

[0177] <In the case of a single timing> As described above, the second system voltage command value calculation unit 44 detects the three-phase currents Iu2, Iv2, and Iw2 flowing through the three-phase armature windings of the second system at a single timing synchronized with the phase of the second system's PWM period Tpwm2. The second system voltage command value calculation unit 44 detects the current at the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2).

[0178] The first system component extraction unit 46 extracts the components of the first system from the detected values ​​Iu2, Iv2, and Iw2 of the three phase currents of the second system, and the PWM period synchronization unit 47 calculates the sum of the squares of the extracted three phases of the first system, Amp1.

[0179] In this embodiment, as shown in Figure 35 and the following equation, the first system component extraction unit 46 processes the current detection values ​​Iu2, Iv2, and Iw2 of each phase with a bandpass filter BPF that passes the component of the second system's PWM period Tpwm2 (= Tpwm1), and calculates the current detection values ​​Iu2f, Iv2f, and Iw2f after BPF processing of each phase. Then, the PWM period synchronization unit 47 squares the current detection values ​​Iu2f, Iv2f, and Iw2f after BPF processing of each phase, sums the squared values ​​of each phase, and calculates the sum of squares Amp1. Iu2f = BPF(Iu2) Iv2f = BPF(Iv2) Iw2f = BPF(Iw2) Amp1 = Iu2f 2 +Iv2f 2 +Iw2f 2 ... (10)

[0180] The PWM period synchronization unit 47 changes the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the maximum value of the sum of squares Amp1.

[0181] With this configuration, when the sum of squares Amp1 approaches its maximum value, the peak of the second carrier wave CA2 is synchronized with the trough of the first carrier wave CA1. Therefore, the current detection timing of the second system, which is set to the peak of the second carrier wave CA2, can be prevented from overlapping with the on / off timing of the first system's PWM signal, which is set to a point other than the peaks and troughs of the first carrier wave CA1, thereby improving the accuracy of the current detection value of the second system's armature winding. Note that when the peak of the second carrier wave CA2 is synchronized with the trough of the first carrier wave CA1, it coincides with the timing of the voltage pulse explained using Figure 33, but the voltage pulse in Figure 33 occurs every 120 degrees, so the frequency is not high.

[0182] Furthermore, when superimposed two-phase modulation is performed, the peak of the second carrier wave CA2 can be synchronized with the peak of the first carrier wave CA1. In this case, since the peak of the second carrier wave CA2 will not coincide with the timing of the voltage pulse explained using Figure 33, the accuracy of the current detection value can be further improved.

[0183] In this embodiment, the PWM period synchronization unit 47 determines that the sum of squares Amp1 has decreased from its maximum value, and changes the PWM period Tpwm2 of the second system from a predetermined period Tpwm20, which is set to the same period as the PWM period Tpwm1 of the first system. If it determines that the sum of squares Amp1 has not decreased from its maximum value, it sets the PWM period Tpwm2 of the second system to the predetermined period Tpwm20. Tpwm20 = Tpwm1.

[0184] This synchronization process is the same as the flowchart in Figure 18 of Embodiment 1, except for step S01, so its explanation will be omitted. In step S01, the PWM period synchronization unit 47 calculates the sum of squares Amp1 based on the components Iu2f, Iv2f, and Iw2f of the first system extracted from the three-phase current detection values ​​of the second system detected and calculated in the current calculation period. The synchronization determination value Ath is set by multiplying the maximum value of the sum of squares Amp1 by a coefficient less than 1 (for example, 0.95). The synchronization determination value Ath may be set in advance based on the maximum value of the sum of squares Amp1 measured in advance, or it may be set adaptively based on the maximum value of a plurality of calculated values ​​of the sum of squares Amp1 calculated in the past, or it may be set using map data in which the relationship between the operating state such as the output command value of the second system and the angular velocity ω2 of the second system and the synchronization determination value Ath is set in advance. The coefficient is determined so as not to cause misjudgment, taking into account detection calculation errors and variations. The synchronization process may be performed every 120 degrees when the voltage pulse shown in Figure 33 is generated.

[0185] Furthermore, similar to Embodiment 6 described later, the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the minimum value of the sum of squares Amp1.

[0186] <In the case of multiple timings> Alternatively, similar to Embodiment 2, the second system voltage command value calculation unit 44 may detect the three-phase currents Iu2, Iv2, and Iw2 flowing through the three-phase armature windings of the second system at multiple timings synchronized with the phase of the second system's PWM period Tpwm2. The first system component extraction unit 46 may then extract the first system component from each of the detected values ​​Iu2, Iv2, and Iw2 of the three-phase currents of the second system detected at each timing, and the PWM period synchronization unit 47 may calculate the sum of the squares of the extracted three-phase first system components Amp1 at each timing.

[0187] Similar to Embodiment 2, the multiple timings are set to three timings: a reference timing, a pre-timing immediately preceding the reference timing, and a post-timing immediately following the reference timing. The reference timing is set to the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2). The pre-timing and post-timing are set symmetrically with respect to the reference timing. Note that the multiple timings only need to be set to two or more timings, including the reference timing.

[0188] Furthermore, the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the maximum value of the sum of squares Amp1, based on the relative magnitudes of the multiple sums of squares Amp1 corresponding to the multiple timings.

[0189] By examining the relative magnitudes of multiple sums of squares Amp1 corresponding to multiple timings, it is possible to obtain information on whether the sum of squares Amp1 of multiple timings is increasing, decreasing, or near its maximum value. Accordingly, by advancing, delaying, or maintaining the phase of the second PWM period Tpwm2, the sum of squares Amp1 of the reference timing can be brought closer to its maximum value, and the phase of the second PWM period Tpwm2 can be synchronized with the phase of the first PWM period Tpwm1.

[0190] Similar to Embodiment 2, the PWM period synchronization unit 47 determines, based on the relative magnitudes of a plurality of sums of squares Amp1 corresponding to a plurality of timings, whether the sum of squares Amp1 is in a period containing the maximum value of the sum of squares, a period in which the sum of squares Amp1 decreases, or a period in which the sum of squares Amp1 increases. If the period contains the maximum value of the sum of squares, the PWM period Tpwm2 of the second system is set to a predetermined period Tpwm20, which is the same period as the PWM period of the first system PWM signal. If the period is in which the sum of squares Amp1 decreases, the PWM period Tpwm2 of the second system is set to a period lower than the predetermined period Tpwm20. If the period is in which the sum of squares Amp1 increases, the PWM period Tpwm2 of the second system is set to a period higher than the predetermined period Tpwm20.

[0191] This synchronization process is the same as the flowchart in Figure 21 of Embodiment 2, except for step S11, so its explanation will be omitted. In step S11, the sum of the squares of the three timings, Amp1, is calculated based on the components Iu2f, Iv2f, and Iw2f of the first system of the three timings that are detected and calculated in synchronization with the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2). Specifically, the sum of the squares of the reference timings, Amp10, is calculated based on the components Iu2f, Iv2f, and Iw2f of the first system of the reference timing, the sum of the squares of the previous timings, Amp1bf, is calculated based on the components Iu2f, Iv2f, and Iw2f of the first system of the previous timing, and the sum of the squares of the later timings, Amp1af, is calculated based on the components Iu2f, Iv2f, and Iw2f of the first system of the later timing. The synchronization process may be performed every 120 degrees when the voltage pulse shown in Figure 33 is generated.

[0192] Furthermore, similar to Embodiment 6 described later, the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 based on the relative magnitudes of multiple sums of squares Amp1 corresponding to multiple timings, so that the sum of squares Amp1 approaches the minimum value of the sum of squares.

[0193] <When calculating Diff> Alternatively, similar to Embodiment 3, the second system voltage command value calculation unit 44 may detect the three-phase currents Iu2, Iv2, and Iw2 flowing through the three-phase armature windings of the second system at multiple timings synchronized with the phase of the second system's PWM period Tpwm2. The first system component extraction unit 46 then extracts the first system component from each of the detected values ​​Iu2, Iv2, and Iw2 of the three-phase currents of the second system detected at each timing, and the PWM period synchronization unit 47 converts the multi-phase first system components detected at each timing into two-phase first system components and calculates the deviation Diff of the two-phase first system components at each timing.

[0194] Similar to Embodiment 2, the multiple timings are set to three timings: a reference timing, a pre-timing immediately before the reference timing, and a post-timing immediately after the reference timing.

[0195] The PWM periodic synchronization unit 47 performs a known three-phase αβ conversion (Clark conversion) using the following equations to convert the components Iu2f, Iv2f, and Iw2f of the first three-phase system to the components Iα2f and Iβ2f of the first two-phase system. Iα2f = Iu2f - 0.5 × Iv2f - 0.5 × Iw2f Iβ2f = 0.866 × Iv2f - 0.866 × Iw2f ... (11)

[0196] Then, the PWM period synchronization unit 47 calculates the deviation Diff by subtracting the first component Iβ2f of the β phase from the first component Iα2f of the α phase, as shown in the following equation: Diff = Iα2f - Iβ2f ... (12)

[0197] When the deviation Diff is at a peak (maximum value, local maximum), the timing of the trough of the first carrier wave CA1 coincides with the timing of the peak of the second carrier wave CA2. Therefore, the relative phase between the phase of the first PWM period and the phase of the second PWM period can be determined based on the deviation Diff. However, if the amplitude of the deviation Diff is small, the accuracy of the determination using the deviation Diff deteriorates due to factors such as the signal-to-noise ratio.

[0198] Similar to Embodiment 3, the PWM period synchronization unit 47 changes the phase of the second PWM period Tpwm2 based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, so that the deviation Diff approaches the maximum value of the deviation Diff.

[0199] By examining the relative magnitudes of multiple deviations (Diff) corresponding to multiple timings, it is possible to obtain information on whether the deviations (Diff) of multiple timings are increasing, decreasing, or near their maximum value (maximum value). Accordingly, by advancing, delaying, or maintaining the phase of the second PWM period Tpwm2, the deviation (Diff) of the reference timing can be brought closer to its maximum value, and the phase of the second PWM period Tpwm2 can be synchronized with the phase of the first PWM period Tpwm1.

[0200] Similar to Embodiment 3, the PWM period synchronization unit 47 determines, based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, whether the deviation Diff is in a period containing the maximum value of the deviation Diff, a period in which the deviation Diff is decreasing, or a period in which the deviation Diff is increasing. If the deviation Diff is in a period containing the maximum value, the PWM period Tpwm2 of the second system is set to a predetermined period Tpwm20, which is the same period as the PWM period of the first system PWM signal. If the deviation Diff is in a period of decreasing deviation, the PWM period Tpwm2 of the second system is reduced to less than the predetermined period Tpwm20. If the deviation Diff is in a period of increasing deviation, the PWM period Tpwm2 of the second system is increased to less than the predetermined period Tpwm20.

[0201] This synchronization process is the same as the flowchart in Figure 25 of Embodiment 3, except for step S21, so its explanation will be omitted. In step S21, based on the components Iu2f, Iv2f, and Iw2f of the first system at three timings, which are detected and calculated in synchronization with the start timing of the second system's PWM period Tpwm2 (the peak of the second system's carrier wave CA2), the two-phase components Iα2f, Iβ2f, and the deviation Diff of the first system at three timings are calculated. Specifically, based on the first system components Iu2f, Iv2f, and Iw2f of the reference timing, the first system components Iα2f0, Iβ2f0, and the deviation Diff0 of the two phases of the reference timing are calculated; based on the first system components Iu2f, Iv2f, and Iw2f of the previous timing, the first system components Iα2fbf, Iβ2fbf, and the deviation Diffbf of the two phases of the previous timing are calculated; and based on the first system components Iu2f, Iv2f, and Iw2f of the later timing, the first system components Iα2faf, Iβ2faf, and the deviation Diffaf of the two phases of the later timing are calculated. Note that the synchronization process may be performed every 120 degrees when the voltage pulse shown in Figure 33 occurs.

[0202] Furthermore, similar to Embodiment 6 described later, the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, so that the deviation Diff approaches the minimum value of the deviation Diff.

[0203] Alternatively, the PWM period synchronization unit 47 may calculate the deviation Diff by subtracting the first component Iα2f of the α phase from the first component Iβ2f of the β phase (Diff = Iβ2f - Iα2f).

[0204] Furthermore, the PWM period synchronization unit 47 may use the first component of the α phase, Iα2f, or the first component of the β phase, Iβ2f, instead of the deviation Diff.

[0205] <Swap of voltage command value shifting method between systems> Alternatively, the first system voltage command value calculation unit 34 may match the average value of the three phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system at the rotation period of the electrical angle to the center voltage (the oscillation center value of the carrier wave CA1 of the first system, which is 0V in this example). For example, no modulation may be applied to the sinusoidal three phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref, or a modulation may be applied that superimposes a zero-sequence voltage at which the average value at the rotation period of the electrical angle becomes 0 (for example, third harmonic superposition, min-max method (pseudo third harmonic superposition)).

[0206] Furthermore, the second system voltage command value calculation unit 44 may shift the average value of the three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the first system at the rotation period of the electrical angle from the center voltage (the oscillation center value of the carrier wave CA1 of the first system, 0V in this example) to the lower potential side or the higher potential side. For example, lower two-phase modulation or upper two-phase modulation may be applied to the sinusoidal three-phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref.

[0207] In this case, in the first system, two voltage pulses are generated in the two phase voltages between the peaks and troughs of the carrier wave CA1 of the first system, and between the troughs and peaks. On the other hand, in the second system, one voltage pulse is generated in the two phase voltages at the timing of the trough of the carrier wave CA2 of the second system. Therefore, the component of the first system superimposed on the current detection value of the second system due to magnetic interference will have a larger component with a period twice that of the PWM period Tpwm2 of the second system, and the peak (mountain) of that component will coincide with the peaks and troughs of the carrier wave CA1 of the first system.

[0208] Therefore, the first system component extraction unit 46 extracts a component of the first system that has a period twice that of the second system's PWM period Tpwm2 (= Tpwm1 × 2), which is included in the detected value of the current of the second system.

[0209] The first system component extraction unit 46 processes the current detection values ​​Iu2, Iv2, and Iw2 of each phase using a bandpass filter BPF that passes a component with a period twice that of the second system's PWM period Tpwm2, and calculates the current detection values ​​Iu2f, Iv2f, and Iw2f of each phase after BPF processing.

[0210] The PWM period synchronization unit 47 then synchronizes the phase of the PWM period Tpwm1 of the second system to the phase of the PWM period Tpwm2 of the first system based on the extracted components of the first system. The synchronization method used is the method using the sum of squares of one timing Amp1, the method using the sum of squares of multiple timings Amp1, or the method using the Diff of multiple timings, as described above, so the explanation is omitted. However, since the components of the first system are components with a period twice that of the PWM period Tpwm2 of the second system, the peaks of the carrier wave CA2 of the second system are synchronized to the peaks or troughs of the carrier wave CA1 of the first system.

[0211] 5. Embodiment 5 Next, a motor device according to Embodiment 5 will be described. The same components as in Embodiment 4 described above will be omitted from the description. The basic configuration and processing of the motor device according to this embodiment are the same as in Embodiment 4, but the PWM period Tpwm1 of the first system and the PWM period Tpwm2 of the second system are different periods, and accordingly the processing of the first system component extraction unit 46, the first system voltage command value calculation unit 34, and the second system voltage command value calculation unit 44 differs from that of Embodiment 4.

[0212] In this embodiment, the PWM period Tpwm1 of the first system and the PWM period Tpwm2 of the second system are set to different periods. For example, the PWM period Tpwm1 of the first system is set to twice the period of the PWM period Tpwm2 of the second system.

[0213] With this configuration, as in Embodiment 4, the first system component can be extracted from the current detection value of the armature winding of the second system without differentiating the modulation method between systems between two-phase modulation and other modulation methods. In other words, in this embodiment, the modulation method of each system may be set to any modulation method or without modulation.

[0214] Therefore, the first system component extraction unit 46 extracts the component of the first system's PWM period Tpwm1 from each of the detected values ​​Iu2, Iv2, and Iw2 of the three-phase current of the second system as the component of the first system. For example, similar to Embodiment 4, a bandpass filter BPF is applied to pass the component of the first system's PWM period Tpwm1.

[0215] The PWM period synchronization unit 47 then synchronizes the phase of the second system's PWM period Tpwm1 to the phase of the first system's PWM period Tpwm2 based on the extracted first system's components. The synchronization method is the same as in Embodiment 4, using the sum of squares of one timing Amp1, using the sum of squares of multiple timings Amp1, or using the Diff of multiple timings, so the explanation is omitted. However, if the first system's PWM period Tpwm1 is set to twice the period of the second system's PWM period Tpwm2, the components of the first system will be components with twice the period of the second system's PWM period Tpwm2, so the peaks of the second system's carrier wave CA2 are synchronized to the peaks or troughs of the first system's carrier wave CA1.

[0216] 6. Embodiment 6 Next, a motor device according to Embodiment 6 will be described. The same components as in Embodiment 4 will not be described. The basic configuration and processing of the motor device according to this embodiment are the same as in Embodiment 4, but the modulation method of the first system and the modulation method of the second system are the same, and accordingly, the processing of the first system component extraction unit 46 and the PWM period synchronization unit 47 differs from that of Embodiment 4.

[0217] Similar to Embodiment 4, the PWM period Tpwm1 of the first system and the PWM period Tpwm2 of the second system are the same PWM period.

[0218] However, unlike Embodiment 4, the modulation method that applies a zero-sequence voltage to the voltage command value of the first system and the modulation method that applies a zero-sequence voltage to the voltage command value of the second system are the same modulation method.

[0219] In this embodiment, the phases are synchronized by utilizing the fact that when the voltage command value of the first system and the voltage command value of the second system are the same, and the phase of the PWM period of the first system and the phase of the PWM period of the second system coincide, the component of the PWM period or the component with twice the period of the PWM period included in the current detection value of the second system becomes the minimum.

[0220] <Explanation of Principle> Figure 36 shows a block diagram of the equivalent circuit of the d-axis when the motor voltage equation is expressed in a dq-axis rotating coordinate system. Vd1 is the d-axis voltage of the first system, Vd2 is the d-axis voltage of the second system, Id1 is the d-axis current of the first system, and Id2 is the d-axis current of the second system. Vid12 is the magnetic interference component of the d-axis voltage of the first system caused by the d-axis current Id2 of the second system. Vid21 is the magnetic interference component of the d-axis voltage of the second system caused by the d-axis current Id1 of the first system. s is the Laplace operator, R is the winding resistance of the motor, Ld is the d-axis self-inductance, and Md is the d-axis mutual inductance. The equivalent circuit of the q-axis is not shown in the diagram, but it is the same as the d-axis, just by replacing d with q.

[0221] Next, in Figure 36, Figures 37 and 38 show the time waveforms of each value when pulse voltages with a time width sufficiently shorter than the electrical time constant determined by Ld / R are applied to Vd1 and Vd2. The horizontal axis is time. Figure 37 shows the waveforms when pulse voltages of the same sign and amplitude are applied to Vd1 and Vd2. When a positive voltage is applied to Vd1 and Vd2, Id1 and Id2 tend to increase in the positive direction. However, because the magnetic interference components Vid12 and Vid21 of the d-axis voltage due to Id1 and Id2 are generated in the positive direction (because they are the derivatives of the increasing Id1 and Id2, their signs are positive), the magnetic interference components Vid12 and Vid21 act to weaken Vd1 and Vd2. As a result, Vd1-Vid12 and Vd2-Vid21 input to the first-order lag system 1 / (R+sLd) decrease compared to Vd1 and Vd2, and the amount of change in Id1 and Id2 becomes smaller than that shown in Figure 38, which will be described later.

[0222] Figure 38 shows the waveform when a pulse voltage of the same amplitude and opposite sign as Vd2 is applied to Vd1. Vd1 is the same as Vd1 in Figure 37, and Vd2 has the opposite sign to Vd2 in Figure 37. In this case, the application of Vd2 in the negative direction increases Id2 in the negative direction, and as a result, Vid12 is generated in the negative direction. Therefore, unlike Figure 37, the magnetic interference components Vid12 and Vid21 act in a direction that strengthens Vd1 and Vd2, and the Vd1-Vid12 and Vd2-Vid21 input to 1 / (R+sLd) increase compared to Vd1 and Vd2, and the amount of change in Id1 and Id2 is larger than in Figure 37. The phenomena in Figures 37 and 38 occur similarly for the q-axis voltages Vq1 and Vq2.

[0223] From the above discussion, when Vd1 and Vd2 are the same, and Vq1 and Vq2 are the same, that is, when the voltage command value of the first system and the voltage command value of the second system are the same, and the phase of the PWM period of the first system and the phase of the PWM period of the second system are the same, the fluctuation of the current in the armature winding of the first system and the fluctuation of the current in the armature winding of the second system are minimized. Conversely, by changing the phase of the PWM period of the second system so that the fluctuation of the current in the armature winding of the first system is minimized, the phase of the PWM period of the second system is synchronized with the phase of the PWM period of the first system.

[0224] <Periodic components according to the modulation method> The same modulation method is used for the modulation method in which a zero-sequence voltage is added to the voltage command value of the first system and the modulation method in which a zero-sequence voltage is added to the voltage command value of the second system. On the other hand, as explained in Embodiment 4, when lower two-phase modulation or upper two-phase modulation is performed, the PWM period component included in the current detection value of the armature winding becomes large. On the other hand, when no modulation is performed, or when modulation is performed in which the average value is matched to the center voltage (for example, third harmonic superposition, min-max method (pseudo third harmonic superposition)), the period component of twice the PWM period included in the current detection value of the armature winding becomes large.

[0225] Therefore, the first system component extraction unit 46 changes the period to be extracted according to the modulation methods of the first and second systems. That is, when the modulation methods of the first and second systems are modulations that shift the average value to a lower or higher potential than the center voltage (for example, lower two-phase modulation or upper two-phase modulation), the first system component extraction unit 46 extracts the component of the second system's PWM period Tpwm2 included in the detected value of the second system's current as the component of the first system. On the other hand, when the modulation methods of the first and second systems are no modulation or modulation that matches the average value to the center voltage (for example, third harmonic superposition, min-max method (pseudo third harmonic superposition)), the first system component extraction unit 46 extracts the component of the second system's PWM period Tpwm2 included in the detected value of the second system's current as the component of the first system. In this embodiment, the extracted first system component also includes the second system component. Furthermore, when lower two-phase modulation or upper two-phase modulation is used, the PWM period component increases every 120 degrees, and at other angles, the component with twice the period of the PWM period increases, so the component with twice the period of the PWM period may be extracted.

[0226] The processing in the first system component extraction unit 46 is the same as in Embodiment 4, so the explanation will be omitted.

[0227] <PWM Period Synchronization Unit 47> The PWM period synchronization unit 47 calculates the sum of squares Amp1 of the extracted three-phase first system components. The method for calculating the sum of squares Amp1 is the same as in Embodiment 4, so the explanation is omitted.

[0228] The PWM period synchronization unit 47 changes the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the minimum value of the sum of squares Amp1.

[0229] In this embodiment, the PWM period synchronization unit 47 determines that the sum of squares Amp1 has increased from its minimum value, and changes the PWM period Tpwm2 of the second system from a predetermined period Tpwm20, which is set to the same period as the PWM period Tpwm1 of the first system. If it determines that the sum of squares Amp1 has not increased from its minimum value, it sets the PWM period Tpwm2 of the second system to the predetermined period Tpwm20. Tpwm20 = Tpwm1.

[0230] For example, the system is configured as shown in the flowchart of Figure 39. The processing in the flowchart of Figure 39 is performed for each PWM period Tpwm2 of the second system (the peak of the carrier wave CA2 of the second system). In step S31, the PWM period synchronization unit 47 calculates the sum of squares Amp1 based on the components Iu2f, Iv2f, and Iw2f of the first system extracted from the three-phase current detection values ​​of the second system that were detected and calculated in the current calculation period.

[0231] Then, in step S32, the PWM period synchronization unit 47 determines whether the sum of squares Amp1 is less than or equal to a synchronization determination value Ath, which is set to be higher than the minimum value of the sum of squares Amp1. If the sum of squares Amp1 is less than or equal to the synchronization determination value Ath, the process proceeds to step S33; if the sum of squares Amp1 is greater than the synchronization determination value Ath, the process proceeds to step S34. This determination determines whether the sum of squares Amp1 has come closer to the minimum value of the sum of squares Amp1 by the determination value or more. The synchronization determination value Ath is set by multiplying the minimum value of the sum of squares Amp1 by a coefficient greater than 1 (for example, 1.05). The synchronization determination value Ath may be set in advance based on the minimum value of the sum of squares Amp1 measured in advance, or it may be set adaptively based on the minimum value of multiple calculated sums of squares Amp1 calculated in the past, or it may be set using map data in which the relationship between the second system's output command value and the operating state of the second system, such as the angular velocity ω2, and the synchronization determination value Ath is set in advance. The coefficient is determined so as not to cause misjudgment, taking into account detection calculation errors and variations. The synchronization process may be performed every 120 degrees when the voltage pulse shown in Figure 33 is generated.

[0232] In step S33, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a predetermined period Tpwm20, which is the same period as the PWM period Tpwm1 of the first system PWM signal (Tpwm2 = Tpwm20 = Tpwm1), and ends the relative phase change.

[0233] In step S34, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a period that is changed by a period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 + ΔT = Tpwm1 + ΔT), and performs a relative phase change. In this example, the period change amount ΔT is set to a negative value, but it may also be set to a positive value. The absolute value of the period change amount ΔT is set to a value of 10% or less or 1% or less of the specified period Tpwm20.

[0234] <In the case of multiple timings> Alternatively, similar to Embodiment 4, the second system voltage command value calculation unit 44 may detect the three-phase currents Iu2, Iv2, and Iw2 flowing through the three-phase armature windings of the second system at multiple timings synchronized with the phase of the second system's PWM period Tpwm2. The first system component extraction unit 46 may then extract the first system component from each of the detected values ​​Iu2, Iv2, and Iw2 of the three-phase currents of the second system detected at each timing, and the PWM period synchronization unit 47 may calculate the sum of the squares of the extracted three-phase first system components Amp1 at each timing.

[0235] Furthermore, the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the minimum value of the sum of squares Amp1, based on the relative magnitudes of the multiple sums of squares Amp1 corresponding to the multiple timings.

[0236] By examining the relative magnitudes of the multiple sums of squares Amp1 corresponding to multiple timings, it is possible to obtain information on whether the sum of squares Amp1 of multiple timings is increasing, decreasing, or near its minimum value. Accordingly, by advancing, delaying, or maintaining the phase of the second PWM period Tpwm2, the sum of squares Amp1 of the reference timing can be brought closer to its minimum value, and the phase of the second PWM period Tpwm2 can be synchronized with the phase of the first PWM period Tpwm1.

[0237] The PWM period synchronization unit 47 determines, based on the relative magnitudes of multiple sums of squares Amp1 corresponding to multiple timings, whether the sum of squares Amp1 is in a period containing the minimum value of the sum of squares, a period in which the sum of squares Amp1 is increasing, or a period in which the sum of squares Amp1 is decreasing. If the period contains the minimum value of the sum of squares, the PWM period Tpwm2 of the second system is set to a predetermined period Tpwm20, which is the same period as the PWM period of the first system's PWM signal. If the period is in which the sum of squares Amp1 is increasing, the PWM period Tpwm2 of the second system is reduced from the predetermined period Tpwm20. If the period is in which the sum of squares Amp1 is decreasing, the PWM period Tpwm2 of the second system is increased from the predetermined period Tpwm20.

[0238] For example, the system is configured as shown in the flowchart of Figure 40. The processing in the flowchart of Figure 40 is performed for each second PWM period Tpwm2 (peak of the second carrier wave CA2). In step S41, the PWM period synchronization unit 47 calculates the sum of the squares of three timings, Amp1, based on the first system components Iu2f, Iv2f, and Iw2f of the three timings detected and calculated in synchronization with the start timing of the current second PWM period Tpwm2 (peak of the second carrier wave CA2). Specifically, the sum of squares of the reference timings, Amp10, is calculated based on the first system components Iu2f, Iv2f, and Iw2f of the reference timing; the sum of squares of the previous timings, Amp1bf, is calculated based on the first system components Iu2f, Iv2f, and Iw2f of the previous timing; and the sum of squares of the later timings, Amp1af, is calculated based on the first system components Iu2f, Iv2f, and Iw2f of the later timing.

[0239] Then, in step S42, the PWM period synchronization unit 47 determines that if the sum of squares of the reference timings Amp10 is less than or equal to the sum of squares of the subsequent timings Amp1af, and the sum of squares of the reference timings Amp10 is less than or equal to the sum of squares of the preceding timings Amp1bf, then the sum of squares of the reference timings Amp10 is in a period that includes the minimum value of the sum of squares, and proceeds to step S43; otherwise, proceeds to step S44.

[0240] In step S43, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a predetermined period Tpwm20, which is the same period as the PWM period Tpwm1 of the first system PWM signal (Tpwm2 = Tpwm20 = Tpwm1), and ends the relative phase change.

[0241] In step S44, the PWM period synchronization unit 47 determines that it is in a period in which the sum of squares of the reference timings Amp10 is increasing if the sum of squares of the reference timings Amp10 is less than or equal to the sum of squares of the subsequent timings Amp1af and greater than or equal to the sum of squares of the previous timings Amp1bf, or if the sum of squares of the subsequent timings Amp1af is less than or equal to the sum of squares of the reference timings Amp10 and greater than or equal to the sum of squares of the previous timings Amp1bf, and proceeds to step S45. Otherwise, it determines that it is in a period in which the sum of squares of the reference timings Amp10 is decreasing, and proceeds to step S46.

[0242] In step S45, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a period that is reduced by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 - ΔT = Tpwm1 - ΔT, ΔT > 0), and performs a relative phase change that delays the phase of the PWM period Tpwm2 of the second system.

[0243] In step S46, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a period that is increased by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 + ΔT = Tpwm1 + ΔT, ΔT > 0), and performs a relative phase change that advances the phase of the PWM period Tpwm2 of the second system.

[0244] <When calculating Diff> Alternatively, similar to Embodiment 4, the second system voltage command value calculation unit 44 may detect the three-phase currents Iu2, Iv2, and Iw2 flowing through the three-phase armature windings of the second system at multiple timings synchronized with the phase of the second system's PWM period Tpwm2. The first system component extraction unit 46 then extracts the first system component from each of the detected values ​​Iu2, Iv2, and Iw2 of the three-phase currents of the second system detected at each timing, and the PWM period synchronization unit 47 converts the multi-phase first system component detected at each timing into a two-phase first system component, and calculates the deviation Diff of the two-phase first system component at each timing. The same conversion and deviation Diff calculation as in Embodiment 4 are used, so the explanation is omitted.

[0245] Similar to Embodiment 4, the PWM period synchronization unit 47 changes the phase of the second PWM period Tpwm2 based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, so that the deviation Diff approaches the minimum value of the deviation Diff.

[0246] By examining the relative magnitudes of multiple deviations (Diff) corresponding to multiple timings, it is possible to obtain information on whether the deviations (Diff) of multiple timings are decreasing, increasing, or near their minimum values ​​(local minimums). Accordingly, by advancing, delaying, or maintaining the phase of the second PWM period Tpwm2, the deviation (Diff) of the reference timing can be brought closer to its minimum value, and the phase of the second PWM period Tpwm2 can be synchronized with the phase of the first PWM period Tpwm1.

[0247] Similar to Embodiment 4, the PWM period synchronization unit 47 determines, based on the relative magnitudes of multiple deviations Diff corresponding to multiple timings, whether the deviation Diff is in a period containing the minimum value of the deviation Diff, a period in which the deviation Diff is increasing, or a period in which the deviation Diff is decreasing. If the deviation Diff is in a period containing the minimum value, the PWM period Tpwm2 of the second system is set to a predetermined period Tpwm20, which is the same period as the PWM period of the first system PWM signal. If the deviation Diff is in a period in which it is increasing, the PWM period Tpwm2 of the second system is reduced from the predetermined period Tpwm20. If the deviation Diff is in a period in which it is decreasing, the PWM period Tpwm2 of the second system is increased from the predetermined period Tpwm20.

[0248] For example, the system is configured as shown in the flowchart of Figure 41. The processing in the flowchart of Figure 41 is performed for each second PWM period Tpwm2 (peak of the second carrier wave CA2). In step S51, the PWM period synchronization unit 47 calculates the two-phase first system components Iα2f, Iβ2f and the deviation Diff of the three timings based on the first system components Iu2f, Iv2f, and Iw2f of the three timings that have been detected and calculated in synchronization with the start timing of the current second PWM period Tpwm2 (peak of the second carrier wave CA2). Specifically, based on the components Iu2f, Iv2f, and Iw2f of the first system at the reference timing, the components Iα2f0, Iβ2f0, and the deviation Diff0 of the first system of the second phase at the reference timing are calculated; based on the components Iu2f, Iv2f, and Iw2f of the first system at the previous timing, the components Iα2fbf, Iβ2fbf, and the deviation Diffbf of the first system of the second phase at the previous timing are calculated; and based on the components Iu2f, Iv2f, and Iw2f of the first system at the later timing, the components Iα2faf, Iβ2faf, and the deviation Diffaf of the first system of the second phase at the later timing are calculated.

[0249] Then, in step S52, the PWM period synchronization unit 47 determines whether the absolute value of the reference timing deviation Diff0, the absolute value of the preceding timing deviation Diffbf, and the absolute value of the succeeding timing deviation Diffaf are all less than the judgment permission threshold Dth. If they are less than the judgment permission threshold Dth, the unit proceeds to step S53; otherwise, the unit proceeds to step S54. In step S53, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a specified period Tpwm20, which is set to be the same period as the PWM period Tpwm1 of the first system PWM signal (Tpwm2 = Tpwm20 = Tpwm1), and does not change the relative phase.

[0250] In step S54, the PWM period synchronization unit 47 determines that if the deviation of the reference timing Diff0 is less than or equal to the deviation of the subsequent timing Diffaf, and the deviation of the reference timing Diff0 is less than or equal to the deviation of the previous timing Diffbf, then the deviation of the reference timing Diff0 is in a period that includes the minimum value of the deviation Diff, and proceeds to step S53; otherwise, proceeds to step S55.

[0251] In step S53, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a specified period Tpwm20, which is the same period as the PWM period Tpwm1 of the first system PWM signal, and ends the relative phase change.

[0252] In step S55, the PWM period synchronization unit 47 determines that it is in a period in which the deviation of the reference timing, Diff0, is increasing if the deviation of the reference timing, Diff0, is less than or equal to the deviation of the subsequent timing, Diffaf, and greater than or equal to the deviation of the previous timing, or if the deviation of the subsequent timing, Diffaf, is less than or equal to the deviation of the reference timing, Diff0, and greater than or equal to the deviation of the previous timing, and proceeds to step S56. Otherwise, it determines that it is in a period in which the deviation of the reference timing, Diff0, is decreasing, and proceeds to step S57.

[0253] In step S56, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a period that is reduced by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 - ΔT = Tpwm1 - ΔT, ΔT > 0), and performs a relative phase change that delays the phase of the PWM period Tpwm2 of the second system.

[0254] In step S57, the PWM period synchronization unit 47 sets the PWM period Tpwm2 of the second system to a period that is increased by a positive period change amount ΔT from the specified period Tpwm20 (Tpwm2 = Tpwm20 + ΔT = Tpwm1 + ΔT, ΔT > 0), and performs a relative phase change that advances the phase of the PWM period Tpwm2 of the second system.

[0255] 7. Embodiment 7 Next, a motor device according to Embodiment 7 will be described. The same components as in Embodiments 1 to 3 described above will be omitted from the description. The basic configuration and processing of the motor device according to this embodiment are the same as in Embodiments 1 to 3, but it differs from Embodiment 1 in that the modulation method between systems is different.

[0256] In this embodiment, as shown in Figure 42, the first DC power supply and the second DC power supply are the same DC power supply 3, and the same smoothing capacitor 20 is used, connected in parallel to the DC power supply 3. Therefore, the DC voltage Vdc1 of the first system and the DC voltage Vdc2 of the second system are the same DC voltage Vdc.

[0257] In this embodiment, the PWM period Tpwm1 of the first system and the PWM period Tpwm2 of the second system are the same PWM period. The PWM period synchronization unit 47 synchronizes the phase of the PWM period of the second system to the same phase as the phase of the PWM period of the first system. The timing of the peaks of the carrier wave CA1 of the first system and the timing of the peaks of the carrier wave CA2 of the second system coincide.

[0258] For example, the synchronization method using Diff of Embodiment 3 can be used. Alternatively, in the synchronization method of Embodiment 1 or 2, the second period TB may be set to the same period as the second PWM period Tpwm2, the first period TA may be set to twice the period of the second period TB, and the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches the maximum value of the sum of squares Amp1. In this case, the peak of the second carrier wave CA2 can be synchronized with the peak of the first carrier wave CA1.

[0259] The first system voltage command value calculation unit 34 shifts the average values ​​of the three phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system at the rotation period of the electrical angle from the center voltage (the oscillation center value of the carrier wave CA1 of the first system, 0V in this example) to one side of the low potential side and the high potential side. The first system voltage command value calculation unit 34 also shifts the average values ​​of the three phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the second system at the rotation period of the electrical angle from the center voltage (the oscillation center value of the carrier wave CA2 of the second system, 0V in this example) to the other side of the low potential side and the high potential side.

[0260] For example, the first system voltage command value calculation unit 34 performs lower two-phase modulation, and the second system voltage command value calculation unit 44 performs upper two-phase modulation. Alternatively, the first system voltage command value calculation unit 34 performs upper two-phase modulation, and the second system voltage command value calculation unit 44 performs lower two-phase modulation.

[0261] Figure 43 shows the behavior of PWM control when the first system is a lower-side two-phase modulation system and the second system is an upper-side two-phase modulation system. The timing of charging or discharging of the bus current flowing through the DC power supply and smoothing capacitor differs between the systems. If charging or discharging occurs at the same timing between the systems, the charging bus current or discharging bus current will be large, requiring an increase in the capacitance of the smoothing capacitor. By reversing the shift direction between the systems, the overlap of charging or discharging timings between the systems can be reduced, and the capacitance of the smoothing capacitor can be reduced.

[0262] 8. Embodiment 8 Next, a motor device according to Embodiment 8 will be described. The same components as in Embodiments 1 to 3 described above will be omitted from the description. The basic configuration and processing of the motor device according to this embodiment are the same as in Embodiments 1 to 3, but it differs from Embodiment 1 in that the modulation method between systems is different.

[0263] Similar to Embodiment 7, the first DC power supply and the second DC power supply are the same DC power supply 3, and the same smoothing capacitor 20 is used, connected in parallel to the DC power supply 3. Therefore, the DC voltage Vdc1 of the first system and the DC voltage Vdc2 of the second system are the same DC voltage Vdc.

[0264] In this embodiment, the PWM period Tpwm1 of the first system and the PWM period Tpwm2 of the second system are the same PWM period. The PWM period synchronization unit 47 synchronizes the phase of the PWM period of the second system to the opposite phase of the PWM period of the first system. The timing of the trough of the carrier wave CA1 of the first system and the timing of the peak of the carrier wave CA2 of the second system coincide.

[0265] For example, a synchronization method is used in which the phase of the second PWM period is changed so that the deviation Diff approaches its minimum value. Alternatively, in the synchronization method of Embodiment 1 or 2, the second period TB is set to the same period as the second PWM period Tpwm2, the first period TA is set to twice the period of the second period TB, and the PWM period synchronization unit 47 may change the phase of the second PWM period Tpwm2 so that the sum of squares Amp1 approaches its minimum value. In this case, the peaks of the second carrier wave CA2 can be synchronized with the troughs of the first carrier wave CA1.

[0266] The first system voltage command value calculation unit 34 shifts the average values ​​of the three phase voltage command values ​​Vu1_ref, Vv1_ref, and Vw1_ref of the first system at the rotation period of the electrical angle from the center voltage (the oscillation center value of the carrier wave CA1 of the first system, 0V in this example) to one side of the low potential side and one side of the high potential side. The first system voltage command value calculation unit 34 also shifts the average values ​​of the three phase voltage command values ​​Vu2_ref, Vv2_ref, and Vw2_ref of the second system at the rotation period of the electrical angle from the center voltage (the oscillation center value of the carrier wave CA2 of the second system, 0V in this example) to one side of the low potential side and one side of the high potential side. In other words, the shift direction is the same between the systems.

[0267] For example, the first system voltage command value calculation unit 34 performs lower two-phase modulation, and the second system voltage command value calculation unit 44 performs lower two-phase modulation. Alternatively, the first system voltage command value calculation unit 34 performs upper two-phase modulation, and the second system voltage command value calculation unit 44 performs upper two-phase modulation.

[0268] Figure 44 shows the behavior of PWM control when the first system uses downward two-phase modulation and the second system uses downward two-phase modulation. Although the same modulation method is used, the phase of the PWM period of the second system and the phase of the PWM period of the first system are in opposite phases, so the timing of charging or discharging of the bus current flowing through the DC power supply and smoothing capacitor differs between the systems. If charging or discharging occurs at the same timing between the systems, the charging bus current or discharging bus current will be large, requiring an increase in the capacitance of the smoothing capacitor. By making the shift direction the same between the systems and making the PWM periods in opposite phases, the overlap in the timing of charging or discharging between the systems can be reduced, and the capacitance of the smoothing capacitor can be reduced.

[0269] Furthermore, when two-phase modulation is used in both the first and second systems, the on-time of the switching element on the low-potential side can be ensured. Therefore, when a lower-arm shunt system is used in which the current sensor is connected in series with the switching element on the low-potential side, current detection accuracy can be ensured.

[0270] [Other Embodiments] (1) In each of the embodiments described above, the case in which the second control circuit 6B performs synchronization processing was explained as an example. However, the first and second systems may be swapped, and the control circuit 6A of the first system may perform synchronization processing.

[0271] (2) Variations of synchronization have been described in each of the above embodiments. Depending on the purpose and preconditions, the PWM period synchronization unit 47 may change the phase of the second PWM period so that the start timing of the second PWM period (the peak of the second carrier wave CA2), which is set to the current detection timing of the second armature winding, coincides with one or both of the start timing of the first PWM period (the peak of the first carrier wave CA1) and the center timing (the trough of the first carrier wave CA1). To this end, depending on the preconditions, the PWM period synchronization unit 47 may change the phase of the second PWM period so that the sum of squares Amp1 or the deviation Diff approaches its maximum or minimum value.

[0272] While this disclosure describes various exemplary embodiments and examples, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but are applicable individually or in various combinations to the embodiments. Accordingly, countless variations not illustrated are envisioned within the scope of the art disclosed in this disclosure. For example, these include modifying, adding or omitting at least one component, or extracting at least one component and combining it with a component from another embodiment.

[0273] 1: Motor, 2: Resolver, 4A: Inverter of the first system, 4B: Inverter of the second system, 6A: Control circuit of the first system, 6B: Control circuit of the second system, Amp1: Sum of squares, Diff: Deviation, TA: First period, TB: Second period, Tpwm1: PWM period of the first system, Tpwm2: PWM period of the second system, Tpwm20: Standard period

Claims

1. A motor having a motor stator around which a first system of armature windings and a second system of armature windings are wound, and a motor rotor; a resolver having a resolver stator around which a first system of excitation windings, two first system of output windings, a second system of excitation windings, and two second system of output windings are wound, and a resolver rotor that rotates integrally with the motor rotor; a first system of inverter having a plurality of switching elements for converting the power supplied to the first system of armature windings; a second system of inverter having a plurality of switching elements for converting the power supplied to the second system of armature windings; A first control circuit that calculates a first voltage command value to be applied to the first armature winding, generates a first PWM signal with a first PWM period based on the first voltage command value, turns on and off the plurality of switching elements of the first inverter based on the first PWM signal, applies a first AC voltage to the first excitation winding with a phase synchronized with the phase of the first PWM period, detects the output voltages of the two output windings of the first system, and calculates the rotation angle of the resolver rotor based on the detected output voltages of the two output windings of the first system, The system includes a second control circuit that calculates a second voltage command value to be applied to the second armature winding, generates a second PWM signal with a second PWM period based on the second voltage command value, turns on and off the plurality of switching elements of the second inverter based on the second PWM signal, applies a second AC voltage to the second excitation winding whose phase is synchronized with the second PWM period, detects the output voltages of the two output windings of the second system, and calculates the rotation angle of the resolver rotor based on the detected output voltages of the two output windings of the second system, wherein the first excitation winding and the two output windings of the first system and the second excitation winding and the two output windings of the second system magnetically interfere with each other.The second control circuit extracts the component of the first system caused by magnetic interference from each of the detected output voltage values ​​of the two output windings of the second system, and synchronizes the phase of the PWM period of the second system with the phase of the PWM period of the first system based on the extracted components of the first system of the two output windings of the second system.

2. The motor device according to claim 1, wherein the control circuit of the first system applies an AC voltage of the first period, synchronized with the PWM period of the first system, to the excitation winding of the first system as the AC voltage of the first system; the control circuit of the second system applies an AC voltage of the second period, synchronized with the PWM period of the second system, to the excitation winding of the second system, which is different from the first period; and the component of the first period included in the detected output voltage values ​​of the two output windings of the second system is extracted as the component of the first system for the two output windings of the second system.

3. The motor device according to claim 2, wherein the PWM period of the first system and the PWM period of the second system are the same PWM period, the first period is the same as the PWM period and the second period is twice the period of the first period, or the second period is the same as the PWM period and the first period is twice the period of the second period.

4. The motor device according to claim 1, wherein the control circuit of the second system detects the output voltages of the two output windings of the second system at a single timing synchronized with the phase of the PWM period of the second system, extracts the components of the first system contained in each of the detected values ​​of the output voltages of the two output windings of the second system, calculates the sum of the squares of the extracted components of the first system of the two output windings of the second system, and changes the phase of the PWM period of the second system so that the sum of squares approaches a target value of the maximum or minimum value of the sum of squares.

5. The motor device according to claim 1, wherein the second control circuit detects the output voltages of the two output windings of the second system at multiple timings synchronized with the phase of the second system's PWM period, extracts the components of the first system contained in each of the detected values ​​of the output voltages of the two output windings of the second system detected at each timing, calculates the sum of the squares of the components of the first system of the two output windings of the second system at each timing, and changes the phase of the second system's PWM period based on the relative magnitudes of the multiple sums of squares corresponding to the multiple timings, such that the sum of squares approaches a target value of the maximum or minimum value of the sum of squares.

6. The motor device according to claim 1, wherein the second control circuit detects the output voltages of the two output windings of the second system at multiple timings synchronized with the phase of the second system's PWM period, extracts the components of the first system contained in each of the detected values ​​of the output voltages of the two output windings of the second system detected at each timing, calculates the deviation of the extracted components of the first system of the two output windings of the second system, and changes the phase of the second system's PWM period based on the relative magnitudes of the multiple deviations corresponding to the multiple timings, such that the deviation approaches a target value of the maximum or minimum value of the deviation.

7. The motor device according to claim 1, wherein the first DC power supply connected to the first inverter and the second DC power supply connected to the second inverter are the same DC power supply, the PWM period of the first system and the PWM period of the second system are the same PWM period, the control circuit of the first system shifts the average value of the voltage command value of the first system in the rotation period of the electrical angle to one side of the center voltage to the low potential side and the high potential side, the control circuit of the second system shifts the average value of the voltage command value of the second system in the rotation period of the electrical angle to the other side of the center voltage to the low potential side and the high potential side, and the phase of the PWM period of the second system is synchronized to be in phase with the phase of the PWM period of the first system.

8. The motor device according to claim 1, wherein the first DC power supply connected to the first inverter and the second DC power supply connected to the second inverter are the same DC power supply, the PWM period of the first system and the PWM period of the second system are the same PWM period, the control circuit of the first system shifts the average value of the voltage command value of the first system in the rotation period of the electrical angle to one side of the center voltage to the lower potential side and the higher potential side, the control circuit of the second system shifts the average value of the voltage command value of the second system to one side of the center voltage to the lower potential side and the higher potential side, and the phase of the PWM period of the second system is synchronized to be in opposite phase to the phase of the PWM period of the first system.

9. A motor having a motor stator and a motor rotor around which a first system of armature windings and a second system of armature windings are wound; a first system inverter having a plurality of switching elements for converting the power supplied to the first system of armature windings; a second system inverter having a plurality of switching elements for converting the power supplied to the second system of armature windings; a first system control circuit that calculates a first system voltage command value to be applied to the first system of armature windings, generates a first system PWM signal with a first system PWM period based on the first system voltage command value, and turns the plurality of switching elements of the first system inverter on and off based on the first system PWM signal; A motor device comprising: a second control circuit that calculates a second voltage command value to be applied to the second armature winding, generates a second PWM signal with a second PWM period based on the second voltage command value, and turns on and off the plurality of switching elements of the second inverter based on the second PWM signal, wherein the first armature winding and the second armature winding magnetically interfere with each other, the second control circuit detects the second current flowing through the second armature winding, extracts the first component caused by magnetic interference included in the detected value of the second current, and synchronizes the phase of the second PWM period with the phase of the first PWM period based on the extracted first component.

10. The PWM period of the first system and the PWM period of the second system are the same PWM period, the control circuits of the first system and the control circuits of the second system perform either first voltage control or second voltage control, in the first voltage control, the control circuit of the first system shifts the average value of the voltage command value of the first system at the rotation period of the electrical angle from the center voltage to the lower potential side or the higher potential side, the control circuit of the second system matches the average value of the applied voltage to the armature winding of the second system at the rotation period of the electrical angle to the center voltage, and extracts the PWM period component included in the detected current value of the second system as a component of the first system, in the second voltage control, the control circuit of the first system matches the average value of the voltage command value of the first system at the rotation period of the electrical angle to the center voltage, The motor device according to claim 9, wherein the second control circuit shifts the average value of the voltage command value of the second system at the rotation period of the electrical angle to a lower potential side or a higher potential side from the center voltage, and extracts a component with a period twice that of the PWM period included in the detected value of the current of the second system as a component of the first system.

11. The motor device according to claim 9, wherein the second control circuit detects the multi-phase currents of the second system flowing through the multi-phase armature windings of the second system at one timing synchronized with the phase of the second system's PWM period, extracts the component of the first system from each of the detected values ​​of the multi-phase currents of the second system, calculates the sum of the squares of the extracted multi-phase components of the first system, and changes the phase of the second system's PWM period so that the sum of squares approaches a target value of the maximum or minimum value of the sum of squares.

12. The motor device according to claim 9, wherein the second control circuit detects the multi-phase currents of the second system flowing through the multi-phase armature windings of the second system at multiple timings synchronized with the phase of the second system's PWM period, extracts the components of the first system from each of the detected values ​​of the multi-phase currents of the second system detected at each timing, calculates the sum of the squares of the multi-phase components of the first system extracted at each timing, and changes the phase of the second system's PWM period based on the relative magnitudes of the multiple sums of squares corresponding to the multiple timings, such that the sum of squares approaches a target value of the maximum or minimum value of the sum of squares.

13. The motor device according to claim 9, wherein the second control circuit detects the multi-phase current of the second system flowing through the multi-phase armature windings of the second system at multiple timings synchronized with the phase of the second system's PWM period, extracts the components of the first system contained in each of the detected values ​​of the multi-phase current of the second system detected at each timing, converts the multi-phase components of the first system detected at each timing into two-phase components of the first system, calculates the deviation of the two-phase components of the first system at each timing, and changes the phase of the second system's PWM period so that the deviation approaches a target value of the maximum or minimum value of the deviation based on the relative magnitudes of the multiple deviations corresponding to the multiple timings.

14. The motor device according to claim 1 or 9, wherein the control circuit of the second system changes the phase of the PWM period of the second system so that the start timing of the PWM period of the second system, which is set to the current detection timing of the armature winding of the second system, coincides with one or both of the start timing and the center timing of the PWM period of the first system.

15. The motor device according to claim 4 or 11, wherein the control circuit of the second system determines that the sum of squares deviates from the target value, and changes the PWM period of the second system from a predetermined period which is the same as the PWM period of the first system, and determines that the sum of squares does not deviate from the target value, and sets the PWM period of the second system to the predetermined period.

16. The control circuit of the second system determines, based on the relative magnitudes of the multiple sums of squares corresponding to the multiple timings, whether the sum of squares is in a period including the target value, a period in which the sum of squares decreases, or a period in which the sum of squares increases, and if the sum of squares is in a period including the target value, it sets the PWM period of the second system to a predetermined period set to the same period as the PWM period of the first system; if the target value is the maximum value of the sum of squares, if the sum of squares is in a period of decrease, it decreases the PWM period of the second system below the predetermined period; if the sum of squares is in a period of increase, it increases the PWM period of the second system above the predetermined period; if the target value is the minimum value of the sum of squares, The motor device according to claim 5 or 12, wherein if the sum of squares falls within the period of increase, the PWM period of the second system is reduced to less than the specified period, and if the sum of squares falls within the period of decrease, the PWM period of the second system is increased to less than the specified period.

17. The motor device according to claim 6 or 13, wherein the second control circuit determines, based on the relative magnitudes of the multiple deviations corresponding to the multiple timings, whether the deviation is in a period including the target value, a period in which the deviation decreases, or a period in which the deviation increases; if the deviation is in a period including the target value, the PWM period of the second system is set to a predetermined period that is the same as the PWM period of the first system; if the target value is the maximum value of the deviation, and the deviation is in a period of decrease, the PWM period of the second system is reduced to less than the predetermined period; if the deviation is in a period of increase, the PWM period of the second system is increased to less than the predetermined period; if the target value is the minimum value of the deviation, and the deviation is in a period of increase, the PWM period of the second system is reduced to less than the predetermined period; if the deviation is in a period of decrease, the PWM period of the second system is increased to less than the predetermined period.

18. The motor device according to claim 9, wherein the PWM period of the first system and the PWM period of the second system are different periods, and the control circuit of the second system extracts the component of the PWM period of the first system that is included in the detected value of the current of the second system as a component of the first system.

19. The motor device according to claim 9, wherein the PWM period of the first system and the PWM period of the second system are the same PWM period, the modulation method of adding a zero-sequence voltage to the voltage command value of the first system and the modulation method of adding a zero-sequence voltage to the voltage command value of the second system are the same modulation method, and the control circuit of the second system changes the phase of the PWM period of the second system such that the component of the PWM period or the component of twice the period of the PWM period included in the detected value of the current of the second system approaches its minimum value.

Images