JP7544656B2 - Motor Control Device - Google Patents

Description

The present invention relates to a motor control device equipped with two inverters.

Conventionally, a "one battery, one capacitor, two inverter system" has been known in which a battery is used as one power source and a capacitor is used as the other power source in a motor drive system that drives a three-phase open winding motor using two power sources and two inverters. Hereinafter, the term "motor" in this specification refers to a motor generator that functions as an electric motor and a generator in a hybrid vehicle or electric vehicle.

For example, Patent Document 1 discloses a method for driving a one-battery, one-capacitor, two-inverter system using a triangular wave comparison PWM modulation method. Non-Patent Document 1 discloses a method for driving a one-battery, one-capacitor, two-inverter system using a space vector modulation method.

Patent No. 6174498

Y. Oto, "Space Vector Modulation of Dual-Inverter System Focusing on Improvement of Multilevel Voltage Waveforms," IEEE Transactions on Industrial Electronics, vol.66, No.12, pp.9139-9148, Dec.2019

The conventional technology described in Patent Document 1 aims to reduce losses and deterioration due to the internal resistance of the battery by using a capacitor to repeatedly charge and discharge the battery when braking and driving the motor in a one-battery, one-capacitor, two-inverter system driven by asynchronous PWM.

Furthermore, in a one-battery, one-capacitor, two-inverter system, by making the capacitor voltage smaller than the battery voltage, it is expected that motor loss can be reduced through multi-level driving. However, in the driving method of Patent Document 1, in the capacitor-side driving mode and battery-side driving mode, only one of the two inverters is driven. Also, in the capacitor charging mode, the two inverters are driven based on the fundamental wave of the same phase, and the legs of each phase are switched simultaneously, so the driving voltage waveform does not become multi-level.

In contrast, the conventional technology in Non-Patent Document 1 controls the capacitor voltage to half the battery voltage when driving with asynchronous PWM in a one-battery, one-capacitor, two-inverter system, making it possible to reduce motor loss through multi-level driving over the entire range of voltage modulation rates.

However, in the space vector modulation method disclosed in Non-Patent Document 1, the capacitor voltage is fixed to half the battery voltage, and the condition is that the charging voltage vector and the discharging voltage vector are the same. Therefore, it is necessary to charge the capacitor voltage to half the battery voltage immediately before driving the motor after starting the vehicle. If multi-level driving is performed when the capacitor voltage is less than half the battery voltage, the motor phase current will be distorted and the torque ripple will become large.

The present invention was created in light of these points, and its purpose is to provide a motor control device that does not require initial charging of the capacitor and suppresses torque ripple in a system that drives a three-phase open-winding motor using one battery, one capacitor, and two inverters.

The present invention is a control device that controls the energization of a motor in a motor drive system that drives a motor (80) having three-phase windings (81, 82, 83) with an open neutral point using two inverters that receive DC power separately from a first power source (11) that outputs a first power source voltage (V1) and a second power source (12) that is chargeable and dischargeable and outputs a second power source voltage (V2) that changes according to the charge state. Typically, the first power source is a battery and the second power source is a capacitor.

The motor control device includes a first inverter (60), a second inverter (70), a control circuit (30), and drive circuits (56, 57). The first inverter converts DC power input from a first power source into AC power and supplies it to one end of the three-phase winding. The second inverter converts DC power input from a second power source into AC power and supplies it to the other end of the three-phase winding.

The control circuit generates gate signals to be output to the two inverters using space vector modulation. The drive circuit switches the two inverters based on the gate signals generated by the control circuit. The control circuit has a command voltage calculation unit (34, 35), a current polarity mode determination unit (37), a charge/discharge mode determination unit (38), a voltage vector selection unit (41), and a duty ratio calculation unit (42).

The command voltage calculation unit calculates a command voltage by current feedback control based on a command torque to the motor. The current polarity mode determination unit determines a current polarity mode indicating a polarity state of the current based on a phase current flowing through the motor. The charge/discharge mode determination unit determines one of the following charge/discharge modes based on the second power supply voltage: a charge mode in which charging is required for the second power supply, a discharge mode in which discharging is required for the second power supply, or a hold mode in which neither charging nor discharging is required for the second power supply.

The voltage vector selection unit determines a voltage vector set to be used from a charging voltage vector, which is a voltage vector that charges the second power source, a discharging voltage vector, which is a voltage vector that discharges the second power source, and a holding voltage vector, which is a voltage vector that does not charge or discharge the second power source, in space vector coordinates, based on the command voltage, the current polarity mode, and the charge/discharge mode.

The duty ratio calculation unit calculates the duty ratio, which is the "ratio of application time of each voltage vector in one period of space vector modulation," for the voltage vector set determined by the voltage vector selection unit, based on the command voltage, the first power supply voltage, and the second power supply voltage, which is variable within the range of "0 or more and the first power supply voltage or less."

If the first power source is a battery and the second power source is a capacitor, in the present invention, when asynchronous PWM is driven in a one-battery, one-capacitor, two-inverter system, the control range of the capacitor voltage is variable within the range of "0 or more and battery voltage V1 or less", and a space vector modulation method corresponding to the capacitor voltage fluctuation is used. In addition, the duty ratio calculation unit calculates the duty ratio for the voltage vector set determined based on the charge/discharge mode based not only on the command voltage and battery voltage, but also on the capacitor voltage. This makes it possible to perform multi-level driving with reduced torque ripple compared to the conventional technology of Non-Patent Document 1, without the need for initial charging of the capacitor. Furthermore, the distortion of the motor phase current is suppressed, and the torque ripple can be reduced.

1 is an overall configuration diagram of a motor drive system according to an embodiment. FIG. 2 is a configuration diagram of a control circuit according to an embodiment. FIG. 3 is a diagram showing the configuration of a space vector modulation unit in FIG. 2 . FIG. 4 is a diagram for explaining the definition of a current polarity mode. 5 is a flowchart showing an algorithm for determining a current polarity mode. 1A shows an example of a current path in a charge mode, FIG. 1B shows an example of a discharge mode, and FIG. 1C shows an example of a current path in a hold mode. 4 is a flowchart showing an algorithm for determining a charge/discharge mode. FIG. 13 is a diagram of hysteresis when switching between a charge/discharge mode and a hold mode. FIG. 4 is a diagram showing changes in charge/discharge modes and the like in accordance with an image of a vehicle running. FIG. 4 is a diagram showing the relationship between a capacitor voltage and a system loss in charge/discharge control. FIG. 2 is a diagram for explaining the definition of a voltage vector. FIG. 13 is a diagram showing the arrangement of charge/discharge voltage vectors in current polarity modes A and D. FIG. 13 is a diagram showing the arrangement of charge/discharge voltage vectors in current polarity modes B and E. FIG. 13 is a diagram showing the arrangement of charge/discharge voltage vectors in current polarity modes C and F. FIG. 13 is a diagram showing voltage vectors, etc. used in the “A-charge or D-discharge” mode. FIG. 13 is a diagram showing voltage vectors etc. used in the “B-charge or E-discharge” mode. FIG. 13 is a diagram showing voltage vectors, etc. used in the “C-charge or F-discharge” mode. A diagram showing voltage vectors etc. used in the "D-charge or A-discharge" mode. FIG. 13 is a diagram showing voltage vectors, etc. used in the “E-charge or B-discharge” mode. FIG. 13 is a diagram showing voltage vectors, etc. used in the “F-charge or C-discharge” mode. FIG. 1 is a diagram for explaining αβ→ab coordinate transformation. 5A and 5B are diagrams showing the relationship between charge/discharge modes and voltage vector arrangement patterns. A diagram of a voltage vector arrangement pattern (e.g., P, p) for charging and discharging. A diagram of an intermediate voltage vector arrangement pattern (e.g., Q, q). Diagram of intermediate voltage vector arrangement pattern (e.g. R, r). A diagram of an intermediate voltage vector arrangement pattern (e.g., S, s). 13 is a time chart illustrating voltage vector selection in (a) the basic form, (b) application form 1, and (c) application form 2. Diagram of voltage vector arrangement pattern for mixed charging and discharging mode. FIG. 1A is a diagram for explaining a command voltage vector, (b) a subsector (triangular region), and (c) a voltage vector at a vertex of a subsector. 11 is a flowchart for determining a command subsector. 1A is a diagram for explaining an example of a calculation formula for voltage vectors m and n in a charging mode, and FIG. 1B is a diagram for explaining an example of a calculation formula for voltage vectors m and n in a discharging mode. 11A and 11B are diagrams for explaining why a capacitor can be charged from a voltage of 0. 11A and 11B are diagrams showing the feasibility of multilevel driving according to Comparative Example 1, Comparative Example 2, and the present embodiment. FIG. 13 is a diagram showing a voltage vector arrangement in Comparative Example 2 (V2=V1/2). FIG. 13 is a diagram for explaining a command voltage vector in Comparative Example 2 (V2=V1/2). FIG. 11 is a diagram comparing torque ripples between Comparative Example 2 and the present embodiment.

(One embodiment)
An embodiment of a motor control device according to the present invention will now be described with reference to the drawings. The motor control device of this embodiment is a device that controls the energization of a motor generator (hereinafter referred to as "motor"), which is a power source for a hybrid vehicle or an electric vehicle, in a system that drives the motor generator using two power sources and two inverters.

[Motor drive system]
1 shows the overall configuration of a motor drive system using two power sources 11, 12 and two inverters 60, 70. The motor 80 is a three-phase AC motor having a U-phase winding 81, a V-phase winding 82, and a W-phase winding 83. The motor may be an IPM or an induction motor. When applied to a hybrid vehicle, the motor 80 functions as an electric motor that generates torque for driving the drive wheels, and also functions as a generator that can generate electricity using the kinetic energy of the vehicle transmitted from the engine and the drive wheels.

Motor 80 is an open winding motor in which the neutral points of three-phase windings 81, 82, 83 are not connected within motor 80, i.e., the neutral point is open. One end 811, 821, 831 of three-phase windings 81, 82, 83 is connected to each phase output terminal of first inverter 60, and the other end 812, 822, 832 is connected to each phase output terminal of second inverter 70. U-phase voltage VU1, V-phase voltage VV1, and W-phase voltage VW1 are applied to the first inverter 60 side of three-phase windings 81, 82, 83. U-phase voltage VU2, V-phase voltage VV2, and W-phase voltage VW2 are applied to the second inverter 70 side of three-phase windings 81, 82, 83.

For example, a current sensor 84 is provided in the power path from the first inverter 60 to the motor 80 to detect phase currents flowing through the three-phase windings 81, 82, and 83. In the example of FIG. 1, a U-phase current Iu and a W-phase current Iw are detected, but any two or three phase currents may be detected. The current sensor 84 may be provided in the power path from the second inverter 70 to the motor 80. The rotation angle sensor 85 is composed of a resolver or the like, and detects a rotor position signal θm of the motor 80.

The first power source 11 and the second power source 12 are two independent power sources insulated from each other. In this embodiment, the first power source 11 is composed of a battery such as a lithium ion battery, and outputs a first power source voltage V1. The first power source 11 may be composed of multiple batteries connected in series or parallel, and may output a voltage via a boost converter. The second power source 12 is composed of a capacitor such as an electric double layer capacitor. The second power source 12 is chargeable and dischargeable, and outputs a second power source voltage V2 that changes depending on the charging state.

In this embodiment, the "first power source" is the battery 11, and the "second power source" is the capacitor 12. Other examples of power source configurations are described in the "Other embodiments" section. The "first power source voltage" is the battery voltage V1, and the "second power source voltage" is the capacitor voltage V2. The two inverters 60, 70 receive DC power from the battery 11 and the capacitor 12, respectively.

The first voltage sensor 16 detects the battery voltage V1. The second voltage sensor 17 detects the capacitor voltage V2. The control range of the capacitor voltage V2 is variable within the range of 0 to the battery voltage V1 (0≦V2≦V1). This is different from the conventional technology of Non-Patent Document 1, in which the capacitor voltage V2 is fixed to half the battery voltage V1.

The motor control device 100 includes a first inverter 60, a second inverter 70, a control circuit 30, and drive circuits 56 and 57. The first inverter 60 has six upper and lower arm switching elements 61-66 bridge-connected. The switching elements 61, 62, and 63 are upper arm elements of the U-phase, V-phase, and W-phase, respectively, and the switching elements 64, 65, and 66 are lower arm elements of the U-phase, V-phase, and W-phase, respectively. A smoothing capacitor 13 is connected to the input section of the first inverter 60. The first inverter 60 converts DC power input from the battery 11 into AC power, and supplies it to one end 811, 821, and 831 of the three-phase windings 81, 82, and 83.

The second inverter 70 has six upper and lower arm switching elements 71-76 connected in a bridge configuration. The switching elements 71, 72, and 73 are the upper arm elements of the U-phase, V-phase, and W-phase, respectively, and the switching elements 74, 75, and 76 are the lower arm elements of the U-phase, V-phase, and W-phase, respectively. The second inverter 20 converts the DC power input from the capacitor 12 into AC power, and supplies it to the other ends 812, 822, and 832 of the three-phase windings 81, 82, and 83.

The switching elements 61-66, 71-76 operate as power elements. To prevent short circuits between the high-potential side wiring P1, P2 and the low-potential side wiring N1, N2, the upper arm element and the lower arm element of each phase are controlled to be turned on and off complementarily, that is, when one is on, the other is off. The switching elements 61-66, 71-76 are composed of, for example, IGBTs, and a freewheeling diode that allows current to flow from the low-potential side to the high-potential side is connected in parallel. Alternatively, the switching elements 61-66, 71-76 may be composed of other power elements such as MOSFETs.

The control circuit 30 is configured with a microcomputer and includes a CPU, ROM, RAM, I/O, and bus lines connecting these components (not shown). The control circuit 30 executes software processing by having the CPU execute a program prestored in a physical memory device (i.e., a readable non-transitory tangible recording medium) such as a ROM, or performs control through hardware processing using a dedicated electronic circuit.

The control circuit 30 acquires information such as the phase currents Iu, Iw, the battery voltage V1, the capacitor voltage V2, and the rotor position signal θm detected by each sensor. In addition, a command torque T ^* is input to the control circuit 30 from the outside. When the motor 80 is driven, the control circuit 30 generates gate signals to be output to the two inverters 60 and 70 based on the information, thereby controlling the motor currents Iu, Iv, Iw and the capacitor voltage V2. In particular, the control circuit 30 of this embodiment generates the gate signals using space vector modulation. The first drive circuit 56 and the second drive circuit 57 perform switching operations of the first inverter 60 and the second inverter 70, respectively, based on the gate signals generated by the control circuit 30.

Next, the detailed configuration of the control circuit 30 will be described with reference to Figures 2 and 3. As shown in Figure 2, the control circuit 30 includes an electrical angle conversion unit 31, a command current generation unit 32, a current dq conversion unit 33, a dq axis current control unit 34, a voltage αβ conversion unit 35, a current polarity mode determination unit 37, a charge/discharge mode determination unit 38, a space vector modulation unit 40, and a gate signal generation unit 43.

The electrical angle conversion unit 31 converts the rotor position signal θm of the motor 80 into an electrical angle θe. The command current generation unit 32 calculates dq-axis current command values Id ^* , Iq ^* from a command torque T ^* provided by an external controller or the like using a map or formula. The current dq conversion unit 33 calculates dq-axis currents Id, Iq from phase currents (e.g., U-phase current Iu and W-phase current Iw) flowing through the motor 80 and the electrical angle θe.

The dq-axis current control unit 34 calculates dq-axis voltage command values Vd ^* , Vq ^* so that the fed-back dq-axis currents Id, Iq follow the dq-axis current command values Id ^* , Iq ^* . The voltage αβ conversion unit 35 uses the dq-axis voltage command values Vd ^* , Vq ^* and the electrical angle θe to calculate αβ-axis voltage command values Vα ^* , Vβ ^*, which are two-axis orthogonal coordinates in a fixed coordinate system. The dq-axis current control unit 34 and the voltage αβ conversion unit 35 correspond to a "command voltage calculation unit that calculates a command voltage by current feedback control based on a command torque to the motor."

The current polarity mode determination unit 37 determines the current polarity mode, which indicates the polarity state of the current, based on the phase currents (e.g., U-phase current Iu and W-phase current Iw) flowing through the motor 80. Details of the current polarity mode determination will be described later with reference to Figures 4 and 5.

The charge/discharge mode determination unit 38 determines which of the charge/discharge modes, the charge mode, the discharge mode, or the hold mode, is to be selected based on the command torque T ^* , the motor rotation speed N, and the capacitor voltage V2. The charge mode is a mode in which charging of the capacitor 12 is required, and the discharge mode is a mode in which discharging of the capacitor 12 is required. The hold mode is a mode in which neither charging nor discharging of the capacitor 12 is required. Details of the charge/discharge mode determination will be described later with reference to Figs. 6 to 10.

Based on the αβ-axis voltage command values Vα ^* , Vβ ^* , battery voltage V1, capacitor voltage V2, current polarity mode and charge/discharge mode, the space vector modulation unit 40 generates a switching pattern and its duty ratio for generating gate signals for the inverters 60, 70. Based on the switching pattern and duty ratio generated by the space vector modulation unit 40, the gate signal generation unit 43 generates gate signals for driving the power elements of the inverters 60, 70 so as to achieve both motor control and capacitor charge/discharge control.

As shown in FIG. 3, the space vector modulation unit 40 includes a voltage vector selection unit 41 that selects a voltage vector to be used in the space vector modulation, and a duty ratio calculation unit 42 that calculates a duty ratio for outputting the voltage vector selected by the voltage vector selection unit 41. In space vector modulation, a desired voltage vector is realized by alternately applying multiple voltage vector components at a predetermined time ratio. A set of combined voltage vectors is called a "voltage vector set." The duty ratio is the ratio of application time of each voltage vector in one period of the space vector modulation.

The voltage vector selection unit 41 determines a voltage vector set to be used from a charging voltage vector, a discharging voltage vector, and a holding voltage vector in space vector coordinates based on the αβ-axis voltage command values Vα ^* , Vβ ^* , which are command voltages, the current polarity mode, and the charge/discharge mode. The charging voltage vector is a voltage vector that charges the capacitor 12, and the discharging voltage vector is a voltage vector that discharges the capacitor 12. The holding voltage vector is a voltage vector that does not charge or discharge the capacitor 12.

The duty ratio calculation unit 42 calculates a duty ratio for the voltage vector set determined by the voltage vector selection unit 41, based on the command voltages Vα ^* , Vβ ^* , the battery voltage V1, and the capacitor voltage V2.

Details of the algorithms used by the voltage vector selection unit 41 and the duty ratio calculation unit 42 will be described later with reference to Figures 11 to 32.

[Current polarity mode determination]
The definition of the current polarity mode is shown in Fig. 4. The motor currents Iu, Iv, Iw are defined as positive when they flow from the first inverter 60 to the second inverter 70. When the phases of the current vectors Iu, Iv, Iw of the U, V, and W phases are set to 0°, 120°, and 240°, six types of current polarity modes A to F are defined, which correspond to the six regions divided by the following formula. The long dashed lines in the figure indicate the boundaries of each current polarity mode.
(60n±30)° (n=0, 1, 2, 3, 4, 5)

The current polarity mode determination unit 37 determines the current polarity mode based on the positive or negative current flowing through each phase of the motor 80. The voltage vector selection unit 41 selects the charge voltage vector or the discharge voltage vector based on the current polarity mode. In each area except on the boundary line, the current in one phase is positive and the current in two phases is negative, or the current in one phase is negative and the current in two phases is positive. On the boundary line, the current in one phase is zero. The relationship between the positive or negative current of each phase and the current polarity modes A to F is as follows:

When Iu≧0, Iv<0, Iw<0, current polarity mode A.
When Iu≧0, Iv≧0, Iw<0, current polarity mode B.
When Iu<0, Iv≧0, Iw<0, current polarity mode C.
When Iu<0, Iv≧0, Iw≧0, current polarity mode D.
When Iu<0, Iv<0, Iw≧0, current polarity mode E.
Current polarity mode F when Iu≧0, Iv<0, Iw≧0.

The flowchart in Figure 5 shows an example of an algorithm for determining the current polarity mode. In explaining the flowchart, the symbol "S" means step. In S11, it is determined whether "Iu ≧ 0". In S12 and S14, it is determined whether "Iv ≧ 0". In S13 and S15, it is determined whether "Iw ≧ 0". Current polarity modes A to F are determined based on the combination of each determination result. For example, when S11 is YES and S12 is YES, the current polarity mode is determined to be B.

[Charge/discharge mode determination]
6A shows an example of a current path in the charging mode. The upper arm elements of the first inverter 60 are U-phase ON, V-phase OFF, and W-phase OFF, and the upper arm elements of the second inverter 70 are U-phase ON, V-phase OFF, and W-phase OFF. As will be described later, the voltage vector of the first inverter 60 in this state is represented as (100), and the voltage vector of the second inverter 70 is represented as (100)'. In the charging mode, a current flows from the positive electrode of the capacitor 12 to the negative electrode.

Figure 6 (b) shows an example of a current path in discharge mode. The upper arm elements of the first inverter 60 are U-phase off, V-phase off, and W-phase off, and the upper arm elements of the second inverter 70 are U-phase off, V-phase on, and W-phase on. In this state, the voltage vector of the first inverter 60 is represented as (000), and the voltage vector of the second inverter 70 is represented as (011)'. In discharge mode, current flows from the negative pole to the positive pole of the capacitor 12.

Figure 6 (c) shows an example of a current path in the hold mode. The upper arm elements of the first inverter 60 are U-phase on, V-phase off, and W-phase off, and the upper arm elements of the second inverter 70 are U-phase off, V-phase off, and W-phase off. In this state, the voltage vector of the first inverter 60 is represented as (100), and the voltage vector of the second inverter 70 is represented as (000)'. In the hold mode, no current flows through the capacitor 12.

The charge/discharge mode determination unit 38 freely controls the charging and discharging of the capacitor 12 while controlling the energization of the motor 80 in accordance with the command torque T ^* and the motor rotation speed N by switching between the charging mode, the discharging mode, and the holding mode. Note that the motor energization control and the capacitor charging and discharging control can be performed simultaneously not only during the powering operation of the motor 80 but also during the regenerative operation.

The algorithm for determining the charge/discharge mode is shown in the flowchart of Fig. 7. The thresholds for torque, rotation speed, and voltage in Fig. 7 correspond to the thresholds shown in Fig. 8. The thresholds with the symbol suffix "_H" are set to have absolute values larger than those with the symbol suffix "_L". In S21, it is determined whether the previous charge/discharge mode determination was the hold mode. If "the previous charge/discharge mode determination was the hold mode and YES in S21", the absolute value of the torque command value |T*| and the motor rotation speed N are compared with the respective thresholds in S22, and if "the previous charge/discharge mode determination was the charge mode or discharge mode and NO in S21", the absolute value of the torque command value |T ^* | and the motor rotation speed N are compared with the respective thresholds in S23.

In S22, it is determined whether the absolute value |T ^* | of the torque command value is equal to or less than a low threshold value Tth_L and the motor rotation speed N is equal to or more than a high threshold value Nth_H. If the answer is NO in S22, it is determined in S24 that the charge/discharge control is "outside the region in which the charge/discharge control is performed." If the answer is YES in S22, it is determined in S26 that the charge/discharge control is "within the region in which the charge/discharge control is performed," and the previous hold mode is changed to the charge mode or the discharge mode.

In S23, it is determined whether the absolute value |T ^* | of the torque command value is equal to or less than a high threshold value Tth_H and the motor rotation speed N is equal to or more than a low threshold value Nth_L. If the answer is NO in S23, it is determined in S24 that the charge/discharge control is "outside the implementation range". If the answer is YES in S23, it is determined in S26 that the charge/discharge control is "within the implementation range", and the charge mode or the discharge mode is selected.

8 illustrates the algorithms of S22 and S23. Hysteresis is provided for each threshold value for the absolute value |T ^* | of the torque command value and the number of revolutions N to prevent hunting of the control. For the absolute value |T ^* | of the torque command value, the low torque side state in hysteresis switching is set to "1" and the high torque side state is set to "0". For the number of revolutions N, the high revolution side state in hysteresis switching is set to "1" and the low revolution side state is set to "0". When the absolute value |T ^* | of the torque command value and the number of revolutions N are both "1", it is determined that the charge/discharge control is "within the implementation range", and otherwise it is determined that the charge/discharge control is "outside the implementation range".

If it is determined in S24 that the charge/discharge control is "outside the implementation range," then in S25 it is determined whether the capacitor voltage V2 is equal to or lower than an ultra-low threshold Vth_LL near 0V. If the answer is YES in S25, the hold mode is selected, and if the answer is NO, the discharge mode is selected. The ultra-low threshold Vth_LL is set to a value that takes into account detection errors and ensures that the voltage is positive. This prevents the voltage from becoming negative due to discharge.

If it is determined in S26 that the "charge/discharge control is within the implementation range," it is determined in S27 whether the capacitor voltage V2 is equal to or higher than the high threshold value Vth_H, and in S28 whether the capacitor voltage V2 is equal to or lower than the low threshold value Vth_L. If S27 is YES, the discharge mode is selected, and if S28 is YES, the charge mode is selected. If both S27 and S28 are NO, the previous charge mode or discharge mode is maintained.

The threshold values Vth_H and Vth_L of the capacitor voltage V2 may be set to fixed values or may be set variably according to the state of the system. By setting the voltage threshold values Vth_H and Vth_L to fixed values with a hysteresis width centered on half the battery voltage V1, hysteresis control can be performed to maintain the capacitor voltage V2 at approximately half the battery voltage V1, as in the conventional technology of Non-Patent Document 1. Alternatively, the voltage threshold values Vth_H and Vth_L may be set variably according to the magnitude and phase of the command voltage vectors Vα ^* and Vβ ^* , thereby maximizing the motor loss reduction effect achieved by multi-level driving.

With reference to Figure 9, the flow of determining the charge/discharge mode of the capacitor 12 will be explained through changes in the charge/discharge mode, capacitor voltage V2, motor torque T, and motor rotation speed N associated with an image of vehicle driving. The vehicle's driving state changes in the following order: <1> starting, <2> steady driving, <3> uphill driving, <4> regenerative driving downhill, and <5> deceleration/stop.

<1> At start-up, the capacitor voltage V2, motor torque T, and motor rotation speed N are all 0, and the charge/discharge mode is hold mode. <2> When moving to steady-state driving, the motor rotation speed N exceeds the high threshold value Nth_H, and the mode switches to charge mode. When the capacitor voltage V2 reaches the high threshold value Vth_H, the mode switches to discharge mode, and charge/discharge control begins. Thereafter, charging and discharging are repeated within the range of the voltage hysteresis width, so this is also called voltage hysteresis control.

<3> When switching from steady driving to uphill driving, if the motor torque T exceeds the high threshold +Tth_H, the capacitor 12 begins to discharge. If the capacitor voltage V2 drops to the ultra-low threshold Vth_LL, the discharge mode stops and the charge/discharge control ends. <4> When switching to regenerative driving and the motor torque T falls below the low threshold +Tth_L, the charge/discharge control (voltage hysteresis control) starts again. <5> When switching to deceleration/stop and the motor rotation speed N falls below the low threshold Nth_L, the discharge mode is maintained. If the capacitor voltage V2 drops to the ultra-low threshold Vth_LL, the discharge mode stops and the charge/discharge control ends.

Next, referring to FIG. 10, the relationship between the capacitor voltage V2 and system loss during charge/discharge control will be described. As described above, the capacitor voltage V2 is variable within the range of 0 to the battery voltage V1 (0≦V2≦V1). However, within that voltage range, there is an optimum value in terms of system loss. FIG. 10 shows the relationship between the capacitor voltage V2 and system loss. System loss is the sum of the power losses generated by the inverters 60, 70 and the motor 80.

When the capacitor voltage V2 is 0 and equal to the battery voltage V1, the system loss is maximum, and when the capacitor voltage V2 is half the battery voltage V1, the system loss is minimum. This is because the current ripple generated in the inverters 60, 70 is minimized, and as a result, the losses generated in the inverters 60, 70 and the motor 80 are minimized. Therefore, it is preferable that the target value of the capacitor voltage V2 is set to half the battery voltage V1. Specifically, it is preferable that the voltage thresholds Vth_H and Vth_L for switching between the discharge mode and the charge mode in the voltage hysteresis control are set to values close to half the battery voltage V1.

[Space Vector Modulation (1) Selection of Voltage Vector]
(Definition of voltage vector)
With reference to FIG. 11, the definition of the voltage vector synthesized by the two inverters 60 and 70 will be described. The voltage vector follows the notation method of general space vector modulation. However, the voltage vector of the first inverter 60 on the battery 11 side is distinguished as (100), and the voltage vector of the second inverter 70 on the capacitor 12 side is distinguished as (100)', and the synthesized voltage vector is represented as (100)(100)'. "1" indicates that the upper arm element is on and the lower arm element is off, and "0" indicates that the upper arm element is off and the lower arm element is on. The zero vector of the first inverter 60 is represented in two ways, (000) and (111), but when indicating the origin in the following figures, only (000) is written and (111) is omitted.

In the system configuration of this embodiment, the direction of the voltage vector of the second inverter 70 is opposite to that of the voltage vector of the first inverter 60. Therefore, even if the voltage vector (100) of the first inverter 60 and the voltage vector (100)' of the second inverter 70 are expressed in the same way, the directions of the voltage vectors are reversed by 180°.

Under the condition "0<V2<V1", which excludes the case where the upper and lower limits are equal from "0≦V2≦V1", the range of the voltage vector of the first inverter 60 is represented by a relatively large regular hexagon, and the range of the voltage vector of the second inverter 70 is represented by a relatively small regular hexagon. The voltage vectors outside the voltage vector range of the first inverter 60 marked with an x can only be used to either charge or discharge the capacitor 12, and are therefore not used.

Each voltage vector is classified as a charging voltage vector, a discharging voltage vector, or a holding voltage vector depending on the charging and discharging state of the capacitor 12. Excluding unused voltage vectors, voltage vectors when the second inverter 70 is not a zero vector are indicated with a circle. Of the voltage vectors indicated with a circle, the voltage vector that charges the capacitor 12 is defined as a charging voltage vector, and the voltage vector that discharges the capacitor 12 is defined as a discharging voltage vector. On the other hand, voltage vectors when the second inverter 70 is a zero vector are indicated with a square. The voltage vectors indicated with a square are defined as holding voltage vectors.

In the charge mode illustrated in FIG. 6(a), a charge voltage vector (100)(100)' is used, and in the discharge mode illustrated in FIG. 6(b), a discharge voltage vector (000)(011)' is used. In the hold mode illustrated in FIG. 6(c), a hold voltage vector (100)(000)' in which the second inverter 70 is a zero vector is used. At this time, no current flows through the capacitor 12, so it is not charged or discharged.

(Definition of charge/discharge voltage vector)
The charge voltage vector and the discharge voltage vector are determined by the current polarity mode. As an example, Table 1 shows the types of voltage vectors of the second inverter 70 in the case of current polarity mode A. In current polarity mode A, the U-phase current Iu has the largest absolute value among the three-phase motor currents. Since the U-phase current Iu has a positive sign, it flows from the first inverter 60 to the second inverter 70.

When the upper arm element 71 of only the U phase of the second inverter 70 is on and the voltage vector is (100)', a charging current of +|Iu| flows into the capacitor 12. When the upper arm elements 71, 72 of the U phase and V phase of the second inverter 70 are on, or the upper arm elements 71, 73 of the U phase and W phase are on and the voltage vector is (110)' or (101)', a charging current of +|Iu|-|Iv|, +|Iu|-|Iw|, respectively, flows into the capacitor 12.

When the lower arm element 74 of only the U phase of the second inverter 70 is on and the voltage vector is (011)', a discharge current of -|Iu| flows out from the capacitor 12. When the lower arm elements 74, 75 of the U phase and V phase of the second inverter 70 are on, or the lower arm elements 74, 76 of the U phase and W phase are on and the voltage vector is (001)' or (010)', a discharge current of -|Iu|+|Iv| or -|Iu|+|Iw| flows out from the capacitor 12, respectively.

In summary, in the case of current polarity mode A, if the U-phase upper arm element 71 of the second inverter 70 is on, it becomes a charging voltage vector, and if the U-phase lower arm element 74 is on, it becomes a discharging voltage vector.

(Charge/Discharge Voltage Vector Arrangement)
12 to 14 show the arrangement of charge/discharge voltage vectors for all current polarity modes A to F. Once the current polarity mode is determined, it can be determined whether the second inverter 70 is a charge voltage vector or a discharge voltage vector for a voltage vector that is not a zero vector. For the voltage vector of the first inverter 60, six equilateral triangular regions with a voltage phase of 60° each are called "sectors." In FIG. 12, the sectors with voltage phases of 0° to 60° in current polarity mode A are hatched with dashed lines. These sectors are cited as examples in FIG. 23 to FIG. 26.

(How to select voltage vector)
A method for selecting a voltage vector considered based on the definition of the voltage vector will be described. Hereinafter, a plurality of voltage vectors selected in combination will be referred to as a "voltage vector set." The voltage vector selection unit 41 determines a voltage vector set according to the charge/discharge mode and the current polarity mode according to the following two selection rules.

[Selection rule 1]
In order to match the control period of the capacitor charge/discharge control with the control period of the motor control, the charge/discharge mode is not switched during the control period.

[Selection rule 2]
<Basic type>
In the charging mode, the voltage vector selection unit 41 determines a voltage vector set from a charging voltage vector and a holding voltage vector. In the discharging mode, the voltage vector selection unit 41 determines a voltage vector set from a discharging voltage vector and a holding voltage vector. A specific example of voltage vector selection according to the basic form will be described later with reference to FIG. 27(a) and FIG. 23.

<Application>
In the charge mode, the voltage vector selection unit 41 combines the charge voltage vector and the hold voltage vector, or combines the charge voltage vector, the discharge voltage vector, and the hold voltage vector to determine a voltage vector set that is a charge as one cycle of the space vector modulation. In the discharge mode, the voltage vector selection unit 41 combines the discharge voltage vector and the hold voltage vector, or combines the charge voltage vector, the discharge voltage vector, and the hold voltage vector to determine a voltage vector set that is a discharge as one cycle of the space vector modulation. A specific example of voltage vector selection that mixes the charge voltage vector and the discharge voltage vector in the applied form will be described later with reference to Figures 27(b), (c), and 28.

<Common to basic and applied types>
In the holding mode (no charging or discharging), the voltage vector selection unit 41 determines a voltage vector set from only the holding voltage vector or by combining the charging voltage vector, the discharging voltage vector, and the holding voltage vector. When combining the charging voltage vector and the discharging voltage vector, the voltage vector selection unit 41 determines a voltage vector set such that charging and discharging are cancelled as one period of the space vector modulation.

Based on the above selection rules, Figures 15 to 20 show the voltage vectors selected in the charge and discharge modes of all current polarity modes A to F. For example, the same voltage vector is selected in the charge mode of current polarity A and the discharge mode of current polarity D. The charge and discharge mode in this case is written as "A-charge or D-discharge". Figure 15 shows the voltage vectors selected in the charge and discharge modes of "A-charge or D-discharge", Figure 16 shows "B-charge or E-discharge", Figure 17 shows "C-charge or F-discharge", Figure 18 shows "D-charge or A-discharge", Figure 19 shows "E-charge or B-discharge", and Figure 20 shows "F-charge or C-discharge".

Each of Figures 15 to 20 shows the current paths that are dominant during charging and discharging, the elements of the second inverter 70 that are turned on during charging and discharging, the voltage vectors of the second inverter 70 used during charging and discharging, and the voltage vectors of the first inverter 60 and the second inverter 70. The meaning of the arrangement pattern in which the symbols "P, Q, R, p, q, r" are attached to sectors every 60 degrees will be described later with reference to Figure 22.

In the charge mode, the second inverter 70 always places the charge voltage vector between the holding voltage vectors, which are zero vectors, so three-level drive is possible. The same is true for the discharge mode. The holding voltage vectors used in the holding mode are as shown by the square marks in Figures 11 to 14. In the holding mode, instead of using only the holding voltage vectors, two-level drive is used.

(αβ→ab conversion)
In the next explanation of "Duty ratio calculation", we will discuss every 60° of the voltage phase based on the positive and negative directions of the UVW phases. In preparation for this, we will divide the αβ-axis voltage phase into six sectors of 60° each, and introduce the ab-axis voltages. Figure 21 shows the definitions of the voltage phases and ab-axis voltages of each sector.

(Voltage vector arrangement pattern for each sector)
The six voltage vector arrangement patterns in six sectors for every 60° of the αβ-axis voltage phase are represented by the symbols "P, Q, R, p, q, r". For example, the voltage vectors used in the "A-charge or D-discharge" mode shown in Figure 15 are arranged as follows: sectors with voltage phases of 0° to 60° are pattern P, sectors with voltage phases of 60° to 120° are pattern Q, sectors with voltage phases of 120° to 180° are pattern R, sectors with voltage phases of 2180° to 240° are pattern p, sectors with voltage phases of 240° to 300° are pattern q, and sectors with voltage phases of 300° to 360° are pattern r. Figure 22 shows the relationship between the charge and discharge modes and the voltage vector arrangement patterns.

Figures 23 to 25 show six types of voltage vector arrangement patterns, taking as examples the sectors of voltage phases 0° to 60° in current polarity mode A (areas hatched with dashed lines in Figure 12). Here, one sector is divided into four "subsectors (triangular areas)" #1 to #4. A sector is an equilateral triangular area with the three voltage vectors of the first inverter 60 as vertices. A subsector is a triangular area obtained by dividing a sector by a line connecting the voltage vectors of the second inverter. A detailed definition and meaning of a subsector will be described later with reference to Figure 29.

As shown in FIG. 23, the P pattern is a charging arrangement pattern that includes subsector #4 with three charging voltage vectors (100)(100)', (110)(110)', and (100)(101)' as vertices. The P pattern is a discharging arrangement pattern that includes subsector #4 with three discharging voltage vectors (000)(011)', (000)(001)', and (110)(010)' as vertices.

In contrast to these, an arrangement pattern in which both charging voltage vectors and discharging voltage vectors are mixed at the three vertices of a triangle is called an intermediate arrangement pattern. In an intermediate arrangement pattern, the charging or discharging of the capacitor 12 is determined by the duty ratio and the motor current.

As shown in FIG. 24, the Q pattern includes subsector #4 with two charging voltage vectors (100)(100)', (100)(101)' and one discharging voltage vector (000)(001)' as vertices. The q pattern includes subsector #4 with two discharging voltage vectors (000)(011)', (110)(010)' and one charging voltage vector (110)(110)' as vertices.

As shown in FIG. 25, the R pattern includes subsector #4 with two discharge voltage vectors (000) (011)', (000) (001)' and one charge voltage vector (100) (101)' as vertices. The r pattern includes subsector #4 with two charge voltage vectors (100) (100)', (110) (110)' and one discharge voltage vector (110) (010)' as vertices.

As arrangement patterns different from the above six types, FIG. 26 shows intermediate arrangement patterns S and s. Subsector #4 of S and s patterns is represented by an equilateral triangle connecting every other vertex in a regular hexagonal region surrounded by three regular hexagons that indicate the voltage vector range of the second inverter 70.

The S pattern includes subsector #4 with two discharge voltage vectors (000) (001)', (110) (010)' and one charge voltage vector (100) (100)' as vertices. The S pattern includes subsector #4 with two charge voltage vectors (110) (110)', (100) (101)' and one discharge voltage vector (000) (011)' as vertices. In this way, by using a voltage vector arrangement pattern in which both charge voltage vectors and discharge voltage vectors are mixed, the number of charge and discharge repetitions during hysteresis control is reduced.

(Voltage vector selection in charge/discharge mode)
As shown in Fig. 27(a), when the capacitor voltage V2 rises and exceeds the upper threshold in the charge mode, the mode is switched to the discharge mode in the next control period. When the capacitor voltage V2 falls and falls below the lower threshold in the discharge mode, the mode is switched to the charge mode in the next control period. The selection of the voltage vector used in the charge mode and the discharge mode will be described using the sector of the voltage phase of 0° to 60° in the current polarity mode A as an example, as in Figs. 23 to 26.

In the basic form of voltage vector selection, the voltage vector arrangement of the P pattern (top of Figure 23) is selected in charge mode, and the p pattern (bottom of Figure 23) is selected in discharge mode. In the P pattern and p pattern, a sector is divided into four subsectors. The vertices of subsectors #1 to #3 of the P pattern consist of charging voltage vectors and holding voltage vectors, and the vertex of subsector #4 consists of only charging voltage vectors. The vertices of subsectors #1 to #3 of the p pattern consist of discharging voltage vectors and holding voltage vectors, and the vertex of subsector #4 consists of only discharging voltage vectors.

As shown in FIG. 27(b), in the applied form 1 of the voltage vector selection, the area between the lower threshold and the upper threshold is defined as the intermediate region. In the mixed charge/discharge mode used in the intermediate region, the sector is divided into five subsectors as shown in FIG. 28. In the example in the top of FIG. 28, subsector #1 of the r pattern (bottom of FIG. 25) is divided into two subsectors with vertices (000) (001)'. In the example in the bottom of FIG. 28, subsector #3 of the s pattern (bottom of FIG. 26) is divided into two regions with vertices (110) (010)'. The vertices of all five subsectors contain both the charge voltage vector and the discharge voltage vector. This lengthens the period of time that the voltage remains in the intermediate region, and reduces the ripple of the capacitor voltage V2.

As shown in Figure 27 (c), in application form 2 of voltage vector selection, a center value is defined between a lower threshold and an upper threshold. When the capacitor voltage V2 crosses the center value in the charge mode or discharge mode, the mixed charge/discharge mode is started, and the voltage vector in the example of the upper or lower part of Figure 28 is used. This extends the period of time that the voltage remains in the intermediate region, as in application form 2, and reduces the ripple of the capacitor voltage V2.

[Space Vector Modulation (2) Duty Ratio Calculation]
Fig. 29 shows (a) command voltage vectors, (b) subsectors (triangular regions), and (c) voltage vectors at the vertices of subsectors in the voltage vector arrangement of the P pattern. The command voltage vector V ^* in Fig. 29(a) is expressed as the sum of voltage vectors in the a-axis and b-axis directions. If the unit voltage vectors on the ab axes are Ea and Eb, and the coefficients of the command voltage vector are defined as Va and Vb, the command voltage vector V ^* is expressed by the following formula.
V ^* =VaEa+VbEb

As shown in FIG. 29(b), one sector is divided into four subsectors (triangular regions) #1 to #4. The region closest to the origin is defined as subsector #1, the region farthest from the origin in the a-axis direction as subsector #2, and the region farthest from the origin in the b-axis direction as subsector #3. Furthermore, the region surrounded by subsectors #1, #2, and #3 is defined as subsector #4. The boundary between each subsector differs for each voltage vector arrangement pattern P, Q, R, p, q, and r. The boundary can be expressed by a straight line equation connecting the voltage vectors at both ends using the coefficients Va and Vb of the command voltage vector, the battery voltage V1, and the capacitor voltage V2.

In FIG. 29(a), the sector that includes the end point of the command voltage vector is defined as the "command sector", and the subsector that includes the end point of the command voltage vector is defined as the "command subsector". The voltage vector selection unit 41 determines the command sector from the command voltage vector, and then determines the command subsector. The voltage vector selection unit 41 then determines the voltage vector set to be used based on the voltage vector whose end point is the vertex of the command subsector.

The flowchart in FIG. 30 shows an algorithm by which the voltage vector selection unit 41 determines the command subsector. In S30, the boundaries between subsectors #1-#4, #2-#4, and #3-#4 are set according to the voltage vector arrangement patterns P, Q, R, p, q, and r. In branch 1 of S31, it is determined whether the end point of the command voltage vector is inside the boundary line between #1 and #4 (i.e., on the side of the origin). In branch 2 of S32, it is determined whether the end point of the command voltage vector is outside the boundary line between #2 and #4 (i.e., on the opposite side to the origin). In branch 3 of S33, it is determined whether the end point of the command voltage vector is outside the boundary line between #3 and #4. If S31 is YES, #1 is determined as the command subsector; if S32 is YES, #2 is determined; if S33 is YES, #3 is determined; and if S31, S32, and S33 are all NO, #4 is determined as the command subsector.

Returning to FIG. 29(c), the three voltage vector sets forming the vertices of each of subsectors #1 to #4 are defined as Vl, Vm, and Vn. In subsectors #1 to #3, the voltage vector of the vertex on the origin side is Vl, the voltage vector of the vertex on the a-axis side is Vm, and the voltage vector of the vertex on the b-axis side is Vn. Conversely, in subsector #4, the voltage vector of the vertex opposite the origin is Vl, the voltage vector of the vertex opposite the a-axis is Vm, and the voltage vector of the vertex opposite the b-axis is Vn.

The duty ratios output by each voltage vector Vl, Vm, Vn forming the vertices of the command subsector are defined as l, m, and n (0≦l≦1, 0≦m≦1, 0≦n≦1, l+m+n=1), respectively. The duty ratios l, m, and n are the ratios of the application time of each voltage vector Vl, Vm, and Vn in one period of space vector modulation. Once the duty ratios m and n of the voltage vectors Vm and Vn have been calculated, the duty ratio l of the voltage vector Vl is calculated by "l=1-m-n". Note that "l" is a lowercase "L", and care should be taken not to misread it as the number "1".

Figure 31 shows an example of a calculation formula for the duty ratio when the command subsector in the sector with a voltage phase of 0° to 60° in current polarity mode A is subsector #2. The duty ratio calculation unit 42 calculates the duty ratio of the command voltage vector based on the coefficients Va and Vb of the command voltage vector, the battery voltage V1, and the capacitor voltage V2 according to the charge/discharge mode.

As shown in FIG. 31A, in a charging mode in which the voltage vector arrangement is a P pattern (upper part of FIG. 23), the duty ratios m and n of the voltage vectors Vm and Vn are calculated by the following formula.
m={Va-(V1-V2)}/V2
n=Vb/V2

As shown in FIG. 31B, in the discharge mode in which the voltage vector arrangement is p pattern (lower part of FIG. 23), the duty ratios m and n of the voltage vectors Vm and Vn are calculated by the following formula.
m=(Va-V2)/(V1-V2)
n=Vb/(V1-V2)

The duty ratios m and n of the voltage vectors Vm and Vn are calculated in the same way when the command subsector is a subsector other than #2 or when the voltage vector arrangement is another pattern. The duty ratio calculation unit 42 calculates the duty ratios for all command voltage vectors using calculation formulas for a total of 24 patterns, with six voltage vector arrangement patterns for each of the four subsectors #1 to #4.

[Space Vector Modulation: Why a capacitor can be charged from zero voltage]
With reference to Fig. 32, the reason why the capacitor voltage V2 can be charged from a state of 0 in space vector modulation will be described. When the capacitor voltage V2 is 0, the algorithm selects subsector #1 as the command subsector. At this time, as shown in the upper part of Fig. 32, the vertex voltage vectors Vm and Vn become charging voltage vectors. As a result, the capacitor 12 is charged regardless of the magnitude and phase of the command voltage vector.

As the capacitor voltage V2 is charged, as shown in the lower part of Figure 32, the voltage vector Vm moves along the a-axis toward the origin and along the outer edge of the sector toward the b-axis. The voltage vector Vn moves along the b-axis toward the origin. This results in a voltage vector arrangement of the P pattern (top of Figure 23).

[Effects of this embodiment]
33 to 36, the effects of this embodiment will be described in comparison with Comparative Example 1, which corresponds to the conventional technology of Patent Document 1, and Comparative Example 2, which corresponds to the conventional technology of Non-Patent Document 1. First, the common points and similarities between Comparative Examples 1 and 2 and this embodiment are that the topology of the main circuit of the two-power-supply two-inverter system is the same. Also, they are the same in that power can be transferred between the battery 11, which is the first power supply, and the capacitor 12, which is the second power supply, while driving the motor. On the other hand, the differences between Comparative Examples 1 and 2 and this embodiment are (1) whether multilevel driving is possible and the control range of the capacitor voltage V2 relative to the battery voltage V1.

Figure 33 shows whether or not multilevel driving is possible according to Comparative Example 1, Comparative Example 2, and this embodiment. As is clear from Figure 4 (c) of Patent Document 1, in Comparative Example 1, the winding voltage of each phase is three levels, and multilevel driving is not achieved. On the other hand, in Comparative Example 2 and this embodiment, the winding voltage of each phase is five levels, and multilevel driving is achieved. Therefore, it is possible to reduce motor loss by multilevel driving over the entire range of voltage modulation rates.

The voltage vector arrangement and command voltage vector of Comparative Example 2 will be described with reference to Figures 34 and 35. Figures 34 and 35 correspond to Figures 31(a) and (b) in Figure 12, respectively, which relate to current polarity mode A of this embodiment. In Comparative Example 2, the capacitor voltage V2 is fixed to half the battery voltage V1 (V2 = V1/2). Therefore, as shown in Figure 34, the voltage vectors during charging and discharging are the same.

As shown in FIG. 35, when the command subsector is subsector #2, the duty ratios m, n of the voltage vectors Vm, Vn are calculated from the coefficients Va, Vb of the command voltage vector and the battery voltage V1 according to the following equation, regardless of the charge/discharge mode.
m=(Va-V2)/V2=(2Va-V1)/V1
n=Vb/V2=2Vb/V1

In contrast, in this embodiment, the control range of the capacitor voltage V2 is variable between 0 and the battery voltage V1 (0≦V2≦V1). Therefore, as shown in FIG. 12, the voltage vectors during charging and discharging do not match. Also, as described above with reference to FIG. 31(a) and (b), the duty ratio of the command voltage vector is calculated based on the charge/discharge mode and the capacitor voltage V2 in addition to the coefficients Va and Vb of the command voltage vector and the battery voltage V1.

Referring to Figure 36, the torque ripple when the motor 80 is driven is compared between Comparative Example 2 and this embodiment. The capacitor voltage V2 changes asynchronously with the period of the motor current (U-phase current Iu is shown as an example). In the space vector modulation method of Comparative Example 2, the capacitor voltage V2 is fixed to half the battery voltage V1, and the condition is that the charging voltage vector and the discharging voltage vector are the same.

Therefore, immediately before driving the motor 80 after starting the vehicle, it is necessary to charge the capacitor voltage V2 to half the battery voltage V1. If multi-level driving is performed when the capacitor voltage V2 is less than half the battery voltage V1, the motor phase current will be distorted and the torque ripple TRc will become large. Note that the U-phase current Iu in Figure 36 is shown for the purpose of showing the current period, and the distortion of the phase current in Comparative Example 2 will not be shown.

In contrast, in this embodiment, the control range of the capacitor voltage V2 is variable within a range from 0 to the battery voltage V1. In addition, the duty ratio calculation unit 42 calculates the duty ratio for the voltage vector set determined based on the charge/discharge mode based not only on the command voltage and the battery voltage V1 but also on the capacitor voltage V2. This eliminates the need for initial charging of the capacitor 12, and enables multi-level driving with reduced torque ripple compared to Comparative Example 2. Furthermore, the distortion of the motor phase current is suppressed, and the torque ripple TRp can be reduced. As described above, in this embodiment, the target value of the capacitor voltage V2 is set to half the battery voltage V1, which further reduces system loss.

Other Embodiments
(a) The configuration of the "first power source" and the "second power source" is not limited to the configuration in which the "first power source" is the battery 11 and the "second power source" is the capacitor 12 as in the above embodiment, and the configuration of the "first power source" and the "second power source" may be interpreted in an expanded manner. The "first power source" may be constituted by a lithium ion battery or a fuel cell or the like. The "second power source" may be any power source that is chargeable and dischargeable and outputs a second power source voltage that changes depending on the charging state, and may be constituted by an electric double layer capacitor or a lithium ion battery or a lithium ion capacitor.

(b) In the control circuit 30, the general control configuration other than the configuration specific to this embodiment, such as the current polarity mode determination unit 37, the charge/discharge mode determination unit 38, and the space vector modulation unit 40, may be modified as appropriate based on well-known techniques related to motor control. For example, torque feedback control may be used instead of current feedback control.

(c) The motor control device of the present invention may be applied to any three-phase motor that is powered by an inverter, not limited to motor generators that are the power sources for hybrid and electric vehicles.

The present invention is not limited to the above-mentioned embodiment, and can be implemented in various forms without departing from the spirit of the invention.

11: battery (first power source); 12: capacitor (second power source);
30...Control circuit,
34: dq axis current control unit (command voltage calculation unit),
35: Voltage αβ conversion unit (command voltage calculation unit),
37: Current polarity mode determination unit; 38: Charge/discharge mode determination unit;
41: Voltage vector selection unit; 42: Duty ratio calculation unit;
56, 57 ... drive circuit,
60: first inverter; 70: second inverter;
80: motor; 81, 82, 83: three-phase winding;
V1: battery voltage (first power supply voltage),
V2: Capacitor voltage (second power supply voltage).

Claims (7)

A control device for controlling energization of a motor (80) having three-phase windings (81, 82, 83) with an open neutral point is used in a motor drive system that drives a motor (80) having three phase windings (81, 82, 83) with an open neutral point, using two inverters to which DC power is input separately from a first power source (11) that outputs a first power source voltage (V1) and a second power source (12) that is chargeable and dischargeable and outputs a second power source voltage (V2) that changes according to a charging state, the control device comprising:
a first inverter (60) that converts DC power input from the first power source into AC power and supplies the AC power to one end of the three-phase winding;
a second inverter (70) that converts DC power input from the second power source into AC power and supplies the AC power to the other end of the three-phase winding;
A control circuit (30) for generating gate signals to be output to the two inverters using space vector modulation;
A drive circuit (56, 57) that switches the two inverters based on a gate signal generated by the control circuit;
Equipped with
The control circuit includes:
a command voltage calculation unit (34, 35) that calculates a command voltage by current feedback control based on a command torque to the motor;
a current polarity mode determination unit (37) for determining a current polarity mode indicating a polarity state of a current based on a phase current flowing through the motor;
a charge/discharge mode determination unit (38) that determines, based on the second power supply voltage, a charge mode that requires charging of the second power supply, a discharge mode that requires discharging of the second power supply, or a hold mode that requires neither charging nor discharging of the second power supply;
a voltage vector selection unit (41) that determines a voltage vector set to be used from a charging voltage vector that is a voltage vector for charging the second power source, a discharging voltage vector that is a voltage vector for discharging the second power source, and a holding voltage vector that is a voltage vector that does not charge or discharge the second power source, in a space vector coordinate system, based on the command voltage, the current polarity mode, and the charge/discharge mode;
a duty ratio calculation unit (42) that calculates a duty ratio, which is a ratio of application times of each voltage vector in one period of space vector modulation, based on the command voltage, the first power supply voltage, and the second power supply voltage that is variable within a range of 0 or more and the first power supply voltage or less, for the voltage vector set determined by the voltage vector selection unit;
A motor control device having the above configuration. The voltage vector selection unit is
In the charging mode, determining a voltage vector set from the charging voltage vector and the holding voltage vector;
The motor control device according to claim 1 , wherein in the discharge mode, a voltage vector set is determined from the discharge voltage vector and the holding voltage vector. The voltage vector selection unit is
In the charging mode, the charging voltage vector and the holding voltage vector are combined, or the charging voltage vector, the discharging voltage vector and the holding voltage vector are combined to determine a voltage vector set that is a charging voltage as one period of space vector modulation;
2. The motor control device according to claim 1, wherein in the discharge mode, the discharge voltage vector and the holding voltage vector are combined, or the charging voltage vector, the discharge voltage vector and the holding voltage vector are combined, to determine a voltage vector set that will be discharged as one period of space vector modulation. The voltage vector selection unit is
4. The motor control device according to claim 2, wherein in the holding mode, a voltage vector set is determined from only the holding voltage vector or from a combination of the charging voltage vector, the discharging voltage vector and the holding voltage vector. The motor control device according to any one of claims 1 to 4, wherein the target value of the second power supply voltage is set to half the first power supply voltage. The current polarity mode determination unit
When the phases of the current vectors of the U phase, V phase, and W phase are set to 0°, 120°, and 240°,
(60n±30)° (n=0, 1, 2, 3, 4, 5)
determining six types of current polarity modes corresponding to six regions divided by
6. The motor control device according to claim 1, wherein the voltage vector selection unit selects a charge voltage vector or a discharge voltage vector based on the current polarity mode. The voltage vector selection unit is
determining a command sector in which an end point of a command voltage vector of the first inverter exists among six sectors that are equilateral triangular regions each having a voltage phase of 60°;
Furthermore, a command subsector in which an end point of a command voltage vector exists is determined from among four subsectors that are triangular regions obtained by dividing the sector by a line connecting the voltage vector of the second inverter;
7. The motor control device according to claim 1, wherein a voltage vector set to be used is determined based on a voltage vector at a vertex of the command subsector.

Images