JP7126910B2 - MOTOR CONTROL DEVICE AND MOTOR CONTROL METHOD - Google Patents

Description

More particularly, the present invention relates to a motor control device and a motor control method in which a voltage phase control section controls a conventional flux-weakening control region with low loss.

Electric motors are used as power sources for many home appliances and mechanical equipment. Among them, a permanent magnet is provided on the rotor side, an armature winding is provided on the stator side, and a permanent magnet (PM) motor rotates the rotor by controlling the magnetic field of the armature winding. ) has low loss and high efficiency because there is no field loss, and is often used in large machinery due to the recent trend toward energy saving. As a method of controlling the PM motor, it is common to change the three-phase drive signals Su, Sv, and Sw for switching the inverter according to the torque command value. The drive signals Su, Sv, and Sw are often generated by switching between sine wave control and rectangular wave control according to the operating conditions of the PM motor. In this control method, operation is generally controlled by sine wave control (PWM control) using a sine wave pattern with high motor efficiency in the operating range of medium to low speed rotation, and the output voltage is controlled in the operating range of high speed rotation and high torque. Operation is controlled by rectangular wave control using a rectangular wave pattern that has a high output and a high output.

One example of such a motor control device is the invention described in [Patent Document 1] below. The motor control device described in [Patent Document 1] has a configuration for performing flux-weakening control by sinusoidal wave control in addition to the above-described sinusoidal wave control and rectangular wave control, and indicates the magnetic pole position of the rotor (rotor). A commutation sensor (CS) is used to enable stable motion control even when the speed of the motor changes rapidly.

JP-A-2001-268973

However, the invention described in [Patent Document 1] performs square wave control, sine wave control, and flux-weakening control by sine wave control, so there are many control methods, and torque shock occurs at the time of switching and delay in response. There is a risk that it will occur. In addition, a control system for performing flux-weakening control by sine wave control is required, which may complicate the configuration of the control system and increase the cost of the microcomputer used for control.

The present invention has been made in view of the above circumstances, and aims to provide a motor control device and a motor control method in which a voltage phase control section (rectangular wave control section) covers the conventional flux-weakening control region and suppresses loss. and

The present invention
(1) An inverter 20 that causes three-phase AC drive currents Iu, Iv, and (Iw) to flow to the PM motor 10, and drive current acquisition units 12u and 12v that acquire values of the drive currents Iu, Iv, and (Iw). , an angle acquisition unit 14 for acquiring the electrical angle θ of the PM motor 10, and the drive currents Iu, Iv, and (Iw) acquired by the drive current acquisition units 12u and 12v based on the electrical angle θ are fed back along the d-axis. A three-phase/dq converter 22 that converts a current value Id and a q-axis feedback current value Iq, and a voltage phase ^θv and a voltage command value |Va| A voltage phase control unit 50 that generates a command value Vd and a q-axis voltage command value Vq, and a dq/ A drive signal for switching the inverter 20 by comparing the carrier wave generated by the carrier generation unit 34 with the three-phase voltage command values Vu, Vv, and Vw. In the motor control device 100 having a control signal generator 30 that generates Su, Sv, and Sw,
The voltage phase control unit 50 is
a voltage phase setting unit 502 that sets the voltage phase θv based on the torque command value T ^* ;
a voltage command setting unit 60 that obtains a voltage command value |Va| that takes the minimum current to output a torque T that is substantially the same as the torque command value T ^* in a region below a predetermined upper limit value |Va| _MAX ; and a voltage command value generator 516 for generating a d-axis voltage command value Vd and a q-axis voltage command value Vq based on the voltage phase θv and the voltage command value |Va|. By providing 100, the above problems are solved.
(2) The voltage command setting unit 60b
Id (ref) acquisition units 644a to 644d that acquire the d-axis current value Id (ref) that takes the minimum current to output the torque T that is substantially the same as the torque command value T ^* ;
an Id subtraction unit 648 that acquires the difference ΔId between the d-axis current value Id(ref) and the d-axis feedback current value Id;
When the difference ΔId is negative, the voltage command value |Va| is decreased, and when the difference ΔId is positive, the voltage command value |Va| 66, and by providing the motor control device 100 described in (1) above.
(3) The Id(ref) acquisition unit 644a acquires the d-axis current value Id(ref) based on the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq. The above problem is solved by providing the motor control device 100 described in (2) above.
(4) The Id (ref) acquisition unit 644b obtains the d-axis current value Id (ref) and the q-axis feedback current value corresponding to the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq. A coefficient K _(Id/Iq) that is a ratio to Iq is obtained, and the coefficient K _(Id/Iq) is multiplied by the q-axis feedback current value Iq or the absolute value |Iq| The above problem is solved by providing the motor control device 100 described in (2) above, characterized in that the d-axis current value Id(ref) is obtained by
(5) The Id (ref) acquisition unit 644c acquires the target current phase θi _(base) that takes the minimum current to output the torque T that is substantially the same as the torque command value T ^* , and obtains the d-axis feedback current value Id. and the magnitude |Ia| of the current vector obtained from the q-axis feedback current value Iq or the absolute value | _Iq | ₎ )
Or the following formula Id (ref) = - | Iq | tan (θi _(base) )
The above problem is solved by providing the motor control device 100 described in (2) above, characterized in that the current value Id(ref) is acquired based on the above.
(6) The Id(ref) acquisition unit 644d acquires the d-axis current value Id(ref) based on the torque command value T ^* or the absolute value |T ^* | of the torque command value T ^* . The above problem is solved by providing the motor control device 100 described in (2) above.
(7) The voltage command setting unit 60a has a current phase calculation unit 62, a θi subtraction unit 63, and a current phase control unit 64,
The current phase calculator 62 includes an Ia absolute value calculator 622 that acquires the magnitude |Ia| of the current vector based on the following equation from the d-axis feedback current value Id and the q-axis feedback current value Iq;
|Ia|=(Id ² +Iq ² ) ^1/2
a phase calculator 624 that calculates a current phase θi based on the following equation from the d-axis feedback current value Id and the q-axis feedback current value Iq;
θi=tan ⁻¹ (−Id/Iq)
with
Acquiring a target current phase θi _(base) that corresponds to the magnitude |Ia| of the current vector and takes the minimum current for outputting a torque T that is substantially the same as the torque command value T ^* ;
The θi subtraction unit 63 obtains a difference Δθi by subtracting the current phase θi from the target current phase θi _(base) and outputs the difference Δθi to the current phase control unit 64,
The current phase control unit 64 is
When the q-axis feedback current value Iq is positive and the difference Δθi is positive, the voltage command value |Va| is decreased, and when the q-axis feedback current value Iq is positive and the difference Δθi is negative, the voltage command value | is increased, the voltage command value |Va| is decreased when the q-axis feedback current value Iq is negative and the difference Δθi is negative, and the q-axis feedback current value Iq is negative and the difference Δθi is positive. The above problem is solved by providing the motor control device 100 described in (1) above, characterized in that the voltage command value |Va|
(8) The voltage command setting unit 60a has a q-axis current determination unit 626 that determines whether the q-axis feedback current value Iq is a value near zero, and the q-axis feedback current value Iq is a value near zero. an input switching unit 628 that switches the input to the current phase control unit 64 from the difference Δθi to the d-axis feedback current value Id when it is determined that
The current phase control unit 64 is
The voltage command value |Va| is decreased when the d-axis feedback current value Id is positive, and the voltage command value |Va| 7) The above problems are solved by providing the motor control device 100 described above.
(9) The voltage command setting units 60a and 60b further include a variation monitoring unit 650a that monitors the torque command value T ^* , and detects that the torque command value T ^* has decreased beyond a preset threshold value. When the monitoring unit 650a detects and the integral value of the integral control of the current phase control unit 64 or the d-axis current control unit 66 is larger than a preset threshold value, the integral value is decreased to obtain the voltage command value |Va The above problem is solved by providing the motor control device 100 according to any one of (2) to (8) characterized by generating |.
(10) The voltage command setting units 60a and 60b further include a variation monitoring unit 650b that monitors the voltage phase θv, and the variation monitoring unit 650b detects that the voltage phase θv is at or near a preset upper limit value. Any one of (2) to (8) above, wherein either or both of the control gain of the current phase control section 64 or the d-axis current control section 66 or the integral value of the integral control is increased when The above problem is solved by providing the motor control device 100 described above.
(11) The voltage command setting units 60a and 60b further include a variation monitoring unit 650c that monitors the difference ΔId or the difference Δθi. By providing the motor control device 100 according to any one of (2) to (8) above, characterized in that the control gain of the current phase control section 64 or the d-axis current control section 66 is increased when the , to solve the above problems.
(12) The voltage command setting units 60a and 60b output the voltage command value |Va| output from the current phase control unit 64 or the d-axis current control unit 66 to the voltage command value generation unit 516 through the low-pass filter 652. The above problems are solved by providing the motor control device 100 according to any one of (9) to (11) above.
(13) The voltage phase control unit 50 further includes a carrier setting unit 520, and the carrier setting unit 520 generates carrier setting information Sc based on the voltage phase θv, the electrical angle θ, and the electrical angular velocity ω. Sc intersects the center position of the trailing edge of the carrier wave generated by the carrier wave generation unit 34 with the zero position of the leading edge of the three-phase voltage command values Vu, Vv, and Vw, and the frequency of the carrier wave is set to the three-phase voltage command value Vu, The above problem is solved by providing the motor control device 100 according to any one of the above (1) to (12) characterized in that Vv and Vw are maintained at integer multiples of 3 of odd numbers.
(14) The voltage phase control unit 50 further includes an offset correction unit 70. The offset correction unit 70 generates a d-axis correction voltage ΔVd and a q-axis correction voltage ΔVd based on the d-axis feedback current value Id and the q-axis feedback current value Iq. A correction voltage ΔVq is generated respectively, and the d-axis correction voltage ΔVd and the q-axis correction voltage ΔVq are added to the d-axis voltage command value Vd and the q-axis voltage command value Vq generated by the voltage command value generation unit 516, respectively, and a control signal is generated. The above problem is solved by providing the motor control device 100 according to any one of (1) to (13) characterized by outputting to the generation unit 30 .
(15) An inverter 20 that causes three-phase AC drive currents Iu, Iv, and (Iw) to flow to the PM motor 10, and drive current acquisition units 12u and 12v that acquire values of the drive currents Iu, Iv, and (Iw). , an angle acquisition unit 14 for acquiring the electrical angle θ of the PM motor 10, and the drive currents Iu, Iv, and (Iw) acquired by the drive current acquisition units 12u and 12v based on the electrical angle θ are fed back along the d-axis. A three-phase/dq converter 22 that converts a current value Id and a q-axis feedback current value Iq, and a voltage phase ^θv and a voltage command value |Va| A voltage phase control unit 50 that generates a command value Vd and a q-axis voltage command value Vq, and a dq/ A drive signal for switching the inverter 20 by comparing the carrier wave generated by the carrier generation unit 34 with the three-phase voltage command values Vu, Vv, and Vw. A motor control method for a motor control device 100 having a control signal generator 30 that generates Su, Sv, and Sw,
The voltage phase control unit 50 is
a voltage phase setting step of setting a voltage phase θv based on the torque command value T ^* ;
a voltage command setting step of acquiring a voltage command value |Va| that takes a minimum current to output a torque T that is substantially the same as the torque command value T ^* in a region below a predetermined upper limit value |Va| _MAX ;
a d-axis and q-axis voltage command value generating step of generating a d-axis voltage command value Vd and a q-axis voltage command value Vq based on the voltage phase θv and the voltage command value |Va| The above problems are solved by providing a control method.
(16) The voltage command setting step is
an Id(ref) acquisition step of acquiring a d-axis current value Id(ref) that takes the minimum current to output a torque substantially equal to the torque command value T ^* ;
a ΔId obtaining step of obtaining a difference ΔId between the d-axis current value Id (ref) and the d-axis feedback current value Id;
a voltage command generating step of decreasing the voltage command value |Va| when the difference ΔId is negative and increasing the voltage command value |Va| when the difference ΔId is positive to generate the voltage command value |Va| The above problem is solved by providing the motor control method described in (15) above, characterized by:
(17) The Id(ref) acquisition step acquires the d-axis current value Id(ref) based on the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq. The above problem is solved by providing the motor control method described in (16) above.
(18) The Id(ref) acquisition step corresponds to the q-axis feedback current value Iq or the absolute value |Iq| Further, the coefficient K _{( Id/Iq)} is multiplied by the q-axis feedback current value Iq or the absolute value |Iq| to obtain a negative value. The above problem is solved by providing the motor control method described in (16) above, characterized in that the d-axis current value Id(ref) is obtained by
(19) The Id (ref) acquisition step acquires the target current phase θi _(base) that takes the minimum current to output the torque T that is substantially the same as the torque command value T ^* , and obtains the d-axis feedback current value Id and The magnitude of the current vector |Ia| obtained from the q-axis feedback current value Iq or the absolute value | _Iq | )
Or the following formula Id (ref) = - | Iq | tan (θi _(base) )
The above problem is solved by providing the motor control method described in (16) above, wherein the current value Id(ref) is obtained based on the above.
(20) The Id(ref) acquisition step acquires the d-axis current value Id(ref) based on the torque command value T ^* or the absolute value |T ^* | of the torque command value T ^* . The above problem is solved by providing the motor control method described in (16) above.
(21) the voltage command setting step includes a current phase calculation step, a θi subtraction step, and a current phase control step;
The current phase calculation step acquires the magnitude |Ia| of the current vector based on the following equation from the d-axis feedback current value Id and the q-axis feedback current value Iq,
|Ia|=(Id ² +Iq ² ) ^1/2
A current phase θi is calculated based on the following formula from the d-axis feedback current value Id and the q-axis feedback current value Iq,
θi=tan ⁻¹ (−Id/Iq)
Acquiring a target current phase θi _(base) that corresponds to the magnitude |Ia| of the current vector and takes the minimum current for outputting a torque T that is substantially the same as the torque command value T ^* ;
The θi subtraction step obtains a difference Δθi by subtracting the current phase θi from the target current phase θi _(base) ,
The current phase control step reduces the voltage command value |Va| when the q-axis feedback current value Iq is positive and the difference Δθi is positive, and when the q-axis feedback current value Iq is positive and the difference Δθi is negative. When the q-axis feedback current value Iq is negative and the difference Δθi is negative, the voltage command value |Va| is decreased, and when the q-axis feedback current value Iq is negative The problem is solved by providing the motor control method described in (15) above, characterized in that the voltage command value |Va| is increased when the difference Δθi is positive.
(22) The voltage command setting step includes a q-axis current determination step of determining whether or not the q-axis feedback current value Iq is a value near zero, and a determination that the q-axis feedback current value Iq is a value near zero. an input switching step of switching the input to the current phase control unit 64 from the difference Δθi to the d-axis feedback current value Id when the
The current phase control step reduces the voltage command value |Va| when the d-axis feedback current value Id is positive, and increases the voltage command value |Va| when the d-axis feedback current value Id is negative. The above problem is solved by providing the motor control method described in (21), characterized by:
(23) The voltage command setting step is
a variation monitoring step of detecting that the torque command value T ^* decreases beyond a preset threshold;
When the torque command value T ^* decreases beyond a preset threshold and the integral value of the integral control of the current phase control section 64 or the d-axis current control section 66 is larger than a preset threshold, the integral The above problem is solved by providing the motor control method according to any one of the above (16) to (22), characterized by further comprising a fast response step of decreasing the value.
(24) The voltage command setting step is
a variation monitoring step of detecting that the voltage phase θv has reached or is in the vicinity of a preset upper limit value;
When the voltage phase θv becomes a preset upper limit value or its vicinity, either one or both of the control gain of the current phase control section 64 or the d-axis current control section 66 or the integral value of the integral control is increased. The above problem is solved by providing the motor control method according to any one of the above (16) to (22), further comprising:
(25) The voltage command setting step is
a variation monitoring step of detecting that the difference ΔId or the difference Δθi exceeds a preset value;
and a high-speed response step of increasing the control gain of the current phase control section 64 or the d-axis current control section 66 when the difference ΔId or the difference Δθi exceeds a preset value. The above problem is solved by providing the motor control method according to any one of (16) to (22).
(26) The voltage command setting step outputs the voltage command value |Va| output from the current phase control section 64 or the d-axis current control section 66 to the voltage command value generation section 516 through the low-pass filter 652. The above problem is solved by providing the motor control method according to any one of (23) to (25).
(27) The voltage phase control unit 50 further includes a carrier setting unit 520, and the carrier setting unit 520 generates carrier setting information Sc based on the voltage phase θv, the electrical angle θ, and the electrical angular velocity ω. Sc intersects the center position of the trailing edge of the carrier wave generated by the carrier wave generation unit 34 with the zero position of the leading edge of the three-phase voltage command values Vu, Vv, and Vw, and the frequency of the carrier wave is set to the three-phase voltage command values Vu, Vw, and Vw. The above problem is solved by providing a motor control method according to any one of the above (15) to (26) characterized in that Vv and Vw are maintained at integer multiples of 3 of odd numbers.
(28) further comprising an offset correction step;
In the offset correction step, the d-axis correction voltage ΔVd and the q-axis correction voltage ΔVq are generated based on the d-axis feedback current value Id and the q-axis feedback current value Iq, respectively. The above (15) to (27), wherein the d-axis correction voltage ΔVd and the q-axis correction voltage ΔVq are added to the command value Vd and the q-axis voltage command value Vq, respectively, and output to the control signal generation unit 30. The above problem is solved by providing any one of the motor control methods.

In the motor control device and the motor control method according to the present invention, the voltage phase control section controls the operation of the conventional flux-weakening control region. For this reason, the number of control systems is small, and it is possible to suppress the occurrence of torque shock and delay in response at the time of switching. Also, the configuration of the control system can be simplified, and the cost can be reduced.
Further, in the motor control device and the motor control method according to the present invention, the ^minimum current value | The operation is controlled so as to achieve the target current phase θi _(base) that takes Ia|. Therefore, the PM motor can be operated in an efficient state with little loss. Also, when a sine wave control section is provided, switching from the voltage phase control section to the sine wave control section is performed at a current phase θi equivalent to the target current phase θi _(base) , so smooth switching operation with little torque shock can be achieved. It becomes possible.

1 is a block diagram of a motor control device according to the present invention; FIG. It is a figure explaining the period of the carrier wave of the motor control apparatus which concerns on this invention. 1 is a block diagram showing a voltage command setting section of a first form of a motor control device according to the present invention; FIG. FIG. 4 is a vector diagram explaining a d-axis current value Id(ref) according to the present invention; FIG. 5 is a block diagram showing a voltage command setting section of a second form of the motor control device according to the present invention; FIG. 8 is a block diagram showing another example of the voltage command setting section of the second embodiment of the motor control device according to the present invention; 3 is a block diagram showing an example of a voltage command setting section of a motor control device according to the present invention having a correction voltage calculation section; FIG. FIG. 4 is a block diagram showing a voltage command setting section of the motor control device according to the present invention having a variation monitoring section;

An embodiment of a motor control device 100 and a motor control method according to the present invention will be described with reference to the drawings. Here, FIG. 1 is a block diagram of a motor control device 100 according to the present invention. Although the motor control device 100 including the sine wave control unit 40 will be described here, the motor control device 100 and the motor control method according to the present invention may be _MIN of the voltage command value |Va|, which will be described later, may be optimized so that the sinusoidal wave control unit 40 is not provided.

First, a motor control device 100 according to the present invention controls the operation of a PM motor (permanent magnet motor) 10. An inverter 20 causes three-phase AC drive currents Iu, Iv, and Iw to flow to the PM motor 10. Then, the drive current acquisition units 12u and 12v that acquire the values of the drive currents Iu, Iv, and (Iw), the angle acquisition unit 14 that acquires the electrical angle θ of the PM motor 10, and the drive current acquisition units 12u and 12v are A three-phase/dq conversion unit 22 that converts the acquired drive currents Iu, Iv, and (Iw) into a d-axis feedback current value Id and a q-axis feedback current value Iq a voltage phase control unit 50 that sets the voltage phase θv and the voltage command value |Va| according to the torque command value T ^* , and generates the d-axis voltage command value Vd and the q-axis voltage command value Vq in the voltage phase control mode; , and a control signal generator 30 for generating drive signals Su, Sv, and Sw for switching the inverter 20 based on the d-axis voltage command value Vd and the q-axis voltage command value Vq. Further, in the configuration in which the motor control device 100 includes the sine wave control unit 40, the d-axis current command value Id ^* and the q-axis current command value Iq ^* are set according to the torque command value T ^* instructed from the outside, and the sine wave A sine wave control unit 40 that generates the d-axis voltage command value Vd and the q-axis voltage command value Vq in the control mode, and a switching unit 24 that switches the control of the PM motor 10 between the sine wave control unit 40 and the voltage phase control unit 50. ,have. Further, the control signal generation unit 30 includes a linear correction unit 38 that linearly corrects the d-axis voltage command value Vd and the q-axis voltage command value Vq based on the voltage command value |Va| Based on a dq/three-phase conversion unit 32 that converts the command value Vd and the q-axis voltage command value Vq into three-phase voltage command values Vu, Vv, and Vw of the U-phase, V-phase, and W-phase, and carrier setting information Sc, which will be described later. A carrier wave generation unit 34 that generates a periodic carrier wave, and the three-phase voltage command values Vu, Vv, and Vw output from the dq/three-phase conversion unit 32 are compared with the carrier wave output from the carrier wave generation unit 34, and the inverter 20 and a drive signal generation unit 36 that generates drive signals Su, Sv, and Sw for switching the .

In addition, the inverter 20 that constitutes the motor control device 100 performs a switching operation in accordance with the Hi-Low drive signals Su, Sv, and Sw output from the drive signal generator 36, and supplies direct current from a well-known DC power supply unit 18 such as a battery. The electric power is converted into a three-phase AC voltage based on the drive signals Su, Sv, and Sw and output. As a result, three-phase drive currents Iu, Iv, and Iw whose phases are shifted by 1/3 cycle (2/3π (rad)) flow down through the armature windings of the PM motor 10, respectively.

Further, the PM motor 10 is provided with a permanent magnet on the rotor side as described above, and a three-phase armature winding on the stator side. By causing Iv and Iw to flow down, the magnetic pole and magnetic flux of each armature winding are continuously changed to rotate the rotor. As the PM motor 10, it is preferable to use an IPM (Interior Permanent Magnet) motor in which a permanent magnet is embedded in the rotor.

Further, the drive current acquisition units 12u and 12v can use well-known current sensors capable of non-contact acquisition of the drive currents Iu, Iv and Iw that flow down due to the switching operation of the inverter 20 . Further, the drive current acquisition unit may be configured for current sensorless control in which the drive currents Iu, Iv, and Iw are acquired by calculation from the total drive current or the like. In this example, two drive currents Iu, Iv out of the drive currents Iu, Iv, and Iw are acquired using a known current sensor and converted into d-axis and q-axis feedback current values Id and Iq. there is

Further, as the angle acquisition unit 14, a well-known angle sensor capable of acquiring the angle of the rotor may be used, or an angle sensorless control configuration in which the angle is acquired by calculation from a voltage command value or the like may be employed. Among them, it is particularly preferable to acquire the electrical angle θ of the PM motor 10 using a resolver rotation angle sensor. When the carrier wave is a triangular wave, the electrical angle θ and the driving currents Iu and Iv are obtained at both the peak and trough timings of the triangular wave. It is preferable to use in The electrical angle θ acquired by the angle acquisition unit 14 is also output to the angular velocity computation unit 16, and the angular velocity computation unit 16 calculates an electrical angular velocity ω (rad/s) from the input electrical angle θ. output to each part of

Moreover, it is preferable to provide the PM motor 10 with a well-known cooling mechanism 101 . Here, the cooling mechanism 101 includes, for example, a water jacket 102 that is provided around the PM motor 10 and cools the PM motor 10 by flowing cooling water, and a well-known temperature acquisition means 108 that acquires the water temperature Tw of the cooling water. ,have. Further, the armature winding of the PM motor 10 is provided with a temperature sensor such as a thermistor to acquire the winding temperature Ta. Then, the water temperature Tw and the winding temperature Ta are output to the motor parameter setting unit 110, and the motor parameter setting unit 110 indirectly obtains the temperature of the permanent magnet of the PM motor 10 from the water temperature Tw. Motor parameters (induced voltage constant φa, d-axis inductance Ld, q-axis inductance Lq) corresponding to temperature and winding temperature Ta are obtained from, for example, a data table and output to torque calculation units 404 and 504 .

Further, the three-phase/dq conversion unit 22 calculates the driving currents Iu, Iv, and (Iw) obtained by the driving current obtaining units 12u and 12v based on the electrical angle θ (rad) of the PM motor 10 obtained by the angle obtaining unit 14. 3-phase 2-phase conversion and rotating coordinate conversion are performed on the values of , and the drive currents Iu, Iv, and (Iw) are converted to the d-axis current value Id (flux component current value) and the q-axis current value Iq (torque component current value). Convert. Then, these are output to the switching unit 24 as the d-axis feedback current value Id and the q-axis feedback current value Iq.

The switching unit 24 is a switching circuit that switches the method of generating the d-axis voltage command value Vd and the q-axis voltage command value Vq according to the operating conditions of the PM motor 10. For example, the PM motor 10 operates in a predetermined low speed region. , the PM motor 10 is operated in a sine wave control mode by the sine wave control unit 40, and when the PM motor 10 operates at a predetermined high rotation speed and high torque, the control of the PM motor 10 is performed by voltage phase control. 50 to operate in the voltage phase control mode. In the motor control device 100 according to the present invention, the voltage phase control unit 50 also controls the operation of the conventional weakening magnetic field control region (overmodulation PWM control region). Therefore, when the switching is determined by the voltage command value |Va|, the switching is performed by the voltage command value |Va| including the conventional field-weakening control region. and the electrical angular velocity .omega. are combined, and when both the voltage command value |Va.vertline. and the electrical angular velocity .omega. It is more preferable to perform Also, these switching thresholds may be set in combination with the power supply voltage Vdc and other thresholds. A hysteresis width is given to the threshold when switching from the sine wave control unit 40 to the voltage phase control unit 50 and the threshold when switching from the voltage phase control unit 50 to the sine wave control unit 40, and It is preferable to prevent frequent switching operations of .

Further, when switching from the sine wave control unit 40 to the voltage phase control unit 50, the voltage command value |Va| and used as an initial value and an integral value for integral control in a current phase control section 64 or a d-axis current control section 66, which will be described later. In addition, the voltage phase θv immediately before switching in the sine wave control unit 40 is output to the voltage phase setting unit 502 constituting the voltage phase control unit 50, and used as the initial value in the voltage phase setting unit 502 and the integral value of the integral control. . Further, the current phase θi immediately before switching in the sine wave control section 40 may be output to the phase calculation section 624 described later as the initial value of the current phase θi. In this way, when switching from the sine wave control unit 40 to the voltage phase control unit 50, the voltage command value |Va| By using a value or an integral value, the numerical values before and after switching are taken over, and switching operation with less torque shock can be performed. Furthermore, these voltage command value |Va|, voltage phase θv, and current phase θi may take over the values immediately before switching as described above, or use a low-pass filter or the like (the control period by the sine wave control unit 40 inside) the smoothed value may be handed over to the voltage phase control unit 50 . In this configuration, a smoothed value from which short-term fluctuations have been removed is used for handover, so a more stable handover operation can be performed.

Further, when switching from the voltage phase control unit 50 to the sine wave control unit 40, the d-axis and q-axis feedback current values Id and Iq immediately before switching generated by the voltage phase control unit 50 are output to the current command value setting unit 402. is used as the initial value of the current command value Ia ^* and the integral value for the integral control. Further, when the sine wave control unit 40 has a d-axis low-pass filter 490A and a q-axis low-pass filter 490B, the d-axis and q-axis feedback current values Id and Iq are the initial values and accumulated values of these low-pass filters 490A and 490B. is also used. The d-axis and q-axis voltage command values Vd and Vq immediately before switching generated by the voltage phase control unit 50 are used as initial values of the d-axis and q-axis voltage command values Vd and Vq of the sine wave control unit 40. In addition, values obtained by subtracting interference terms Vd' and Vq' calculated by a non-interference control unit 414, which will be described later, from the d-axis and q-axis voltage command values Vd and Vq are used as integral values of the current integration control unit 410a. . Thus, when switching from the voltage phase control unit 50 to the sine wave control unit 40, the d-axis and q-axis feedback current values Id and Iq of the voltage phase control unit 50 and the d-axis and q-axis voltage command values Vd and Vq are It is output to a predetermined block of the sine wave control section 40 and used as an initial value or an integral value. Further, as will be described later, the voltage phase control unit 50 controls the current phase θi to a value equivalent to the target current phase θi _(base) in a region where the voltage command value |Va| is less than the upper limit value. Switching to the control unit 40 is performed in this state. For this reason, the d-axis and q-axis current values Id and Iq before and after switching are substantially the same, and a switching operation with less torque shock can be performed.

Next, the configuration and operation of the sine wave control section 40 will be described. Note that the configuration of the sine wave control unit 40 described below is an example suitable for the present invention, and is not limited to the configuration described below, and any other sine wave control mechanism may be used.

First, a torque command value T ^* is output from a control unit or the like of a host system (not shown). This torque command value T ^* is a target torque for the operation of the PM motor 10 . When the switching unit 24 selects the sine wave control unit 40, the torque command value T ^* is input to the current command value setting unit 402 of the sine wave control unit 40 via the low-pass filter LPF.

Further, motor parameters (φa, Ld, Lq) corresponding to the permanent magnet temperature and the winding temperature Ta are input from the motor parameter setting unit 110 to the torque calculation unit 404 of the sine wave control unit 40 . Also, the d-axis and q-axis current command values Id ^* and Iq ^* output from the current command value generator 406 are input. Based on these input values, the torque calculation unit 404 calculates the current torque T of the PM motor 10 and outputs it to the current command value setting unit 402 .

Then, the current command value setting unit 402 integrally controls the current command value Ia ^* such that the difference between the torque command value T ^* input through the low-pass filter LPF and the torque T input from the torque calculation unit 404 becomes zero. It is calculated by well-known arithmetic processing such as proportional control, and is output to the current command value generator 406 .

The current command value generation unit 406 refers to, for example, the current-phase angle data map 620 based on the magnitude | ^{Ia *} | of the current command value Ia ^* input from the current command value setting unit 402, Obtain the corresponding target current phase θi _(base) . Note that the target current phase θi _(base) is the current phase angle at which the torque T is maximized for each magnitude |Ia ^* | of the input current command value Ia ^* . converted and recorded in the current-phase angle data map 620 .

Next, the current command value generator 406 calculates the d-axis current command value Id ^* and the q-axis current command value Iq ^* based on the following equations from the obtained |Ia ^* | and the target current phase θi _(base) . do.
Id ^* =Ia ^* .sin(θi _(base) )
Iq ^* =Ia ^* cos(θi _(base) )
The d-axis current command value Id ^* always takes a negative value, and the q-axis current command value Iq ^* takes a value with the same sign as the current command value Ia ^* .

The d-axis current command value Id ^* and the q-axis current command value Iq ^* generated by the current command value generation unit 406 are obtained by the d-axis low-pass filter 490A and the q-axis low-pass filter 490B, respectively. 490A and q-axis low pass filter 490B. At this time, let τd(up) be the time constant of the d-axis low-pass filter 490A when the absolute value |Id ^* | of the d-axis current command value Id ^* increases, and τd(down) be the time constant when it decreases, When the time constant of the q-axis low-pass filter 490B when the absolute value |Iq ^* | of the q-axis current command value Iq ^* increases is τq(up) and the time constant when it decreases is τq(down),
τd(down)>τq(up)>(τq(down), τd(up))
It is preferable to
According to this configuration, when the absolute value |Id ^* | increases or when the absolute value |Iq ^* | decreases, the d-axis current command value Id ^* or q The shaft current command value Iq ^* is rapidly transmitted, and when the q absolute value | ^{Iq *} | When decreasing, the decrease speed is delayed from the increase speed of the q-axis current command value Iq ^* and output. As a result, the d-axis current command value Id ^* , which is the flux-weakening current, can be generated with priority over the q-axis current command value Iq ^{* .} The PM motor 10 can be controlled so that the d-axis current Id does not run short of the axis current Iq.

The d-axis and q-axis current command values Id ^* and Iq ^* that have passed through the d-axis and q-axis low-pass filters 490A and 490B are then input to the voltage command value generator 416 . A preferred example of the voltage command value generator 416 will now be described. First, the d-axis and q-axis current command values Id ^* and Iq ^* input to the voltage command value generation unit 416 are branched into two, one of which is input to the non-interference control unit 414 . On the other hand, the d-axis and q-axis feedback current values Id and Iq are subtracted to obtain deviations ΔId and ΔIq, which are then input to the current control section 410 . Further, the non-interference control unit 414 receives the motor parameters (φa, Ld, Lq) and the electrical angular velocity ω, and calculates the interference terms Vd′ and Vq′ between the d-axis and the q-axis.

Further, the current control section 410 has, for example, a current integral control section 410a and a current proportional control section 410b. It is input to each of the current proportional control section 410b. Well-known current integral control and current proportional control are performed in the current integral control section 410a and the current proportional control section 410b so that the deviation ΔId and the deviation ΔIq become zero, respectively. Then, after the interference terms Vd′ and Vq′ from the non-interference control unit 414 are added to the output of the current integration control unit 410a, the output from the current proportional control unit 410b is added, thereby A d-axis voltage command value Vd and a q-axis voltage command value Vq are generated in consideration of the influence of the interference component of . The d-axis voltage command value Vd and the q-axis voltage command value Vq are output to the control signal generation section 30 via the switching section 24 .

Also, the d-axis and q-axis voltage command values Vd'' and Vq'' in the previous stage to which the output of the current proportional control unit 410b is added are output to the polar coordinate conversion unit 418 of the sine wave control unit 40, and the polar coordinate conversion unit 418 , a voltage phase θv and a voltage command value |Va| are obtained by performing polar coordinate conversion. Then, voltage phase θv is output to carrier setting section 420 . Also, the voltage command value |Va| is output to the linear correction section 38 of the control signal generation section 30 via the switching section 24 .

Further, carrier setting section 420 generates carrier setting information Sc according to voltage phase θv, electrical angular velocity ω, and electrical angle θ obtained by polar coordinate conversion section 418 . The carrier setting information Sc is information for maintaining the carrier wave generated by the carrier wave generator 34 at an appropriate frequency and state, which will be described later. However, in a low-speed region where the PM motor 10 is stopped or rotates at a low speed, the carrier setting unit 420 uses a preset constant cycle value (fixed value) as the carrier setting information Sc. Therefore, in this region, the control signal generator 30 performs a comparison operation between the carrier wave and the three-phase voltage command values Vu, Vv, and Vw in an asynchronous control state to generate the drive signals Su, Sv, and Sw. Incidentally, the comparison operation will be described later. As a result, the rotation of the PM motor 10 is asynchronously controlled. Further, when the electrical angular velocity ω becomes equal to or higher than a preset rotational speed, the carrier setting section 420 generates carrier setting information Sc for synchronization control and outputs it to the carrier wave generating section 34 . Thereby, the PM motor 10 is synchronously controlled. It is preferable that the rotation speed at which synchronous control is switched to asynchronous control and the rotation speed at which asynchronous control is switched to synchronous control have a hysteresis width to prevent frequent switching of the rotation speed at the switching boundary.

Here, the carrier setting information Sc will be explained. First, the carrier setting information Sc maintains the carrier generated by the carrier generator 34 at an appropriate frequency and state. For example, when the carrier wave is a triangular wave as shown in FIG. 2(a), the proper frequency and state of the carrier wave is the center of the trailing edge of the carrier wave as shown at point A in FIG. 2(a). The position crosses the rising zero position of the three-phase voltage command values Vu, Vv, and Vw (Vu at point A), and the frequency of the carrier wave is an odd integer of 3 of the frequencies of the three-phase voltage command values Vu, Vv, and Vw It is multiplied by 9, 15, 21, 27, etc. (hereinafter, these multiples are referred to as synchronization numbers). For example, in the case where the carrier wave is a combination of a rising sawtooth wave and a falling sawtooth wave on the horizontal axis, as shown in FIG. 2(b), point A in FIG. , the center position of the trailing edge of the carrier wave crosses the zero position of the rising edge of the three-phase voltage command values Vu, Vv, and Vw (Vu at point A), and the frequency of the carrier wave having two sawtooth waves as one period is an integer multiple of 3, ie, 9, 15, 21, 27, etc. times the odd number of the frequencies of the three-phase voltage command values Vu, Vv, and Vw. The carrier setting information Sc changes in conjunction with changes in the electrical angular velocity ω, and maintains the carrier wave in the above state. Further, when the electrical angular velocity ω increases or decreases beyond a predetermined value, the synchronization number is increased or decreased by one step to maintain the carrier wave in the above state. As a result, the carrier wave generated by the carrier wave generator 34 is maintained at a frequency that always satisfies the above conditions during synchronization control. The motor control device 100 and the motor control method according to the present invention having this configuration have good continuity when the drive signals Su, Sv, and Sw change from a sine wave pattern (overmodulation pattern) to a rectangular wave pattern. , stable drive signals Su, Sv, and Sw can be generated. In addition, the output line voltages Vuv, Vvw, and Vwu have symmetry, and the PM motor 10 can be controlled stably.

Next, the configuration and operation of the voltage phase control section 50 will be described. The voltage phase control unit 50 according to the present invention can also control the weakened magnetic flux region by this control configuration.

First, when the voltage command value |Va| to the voltage phase control unit 50. At this time, the voltage command value |Va| immediately before switching in the sine wave control section 40 is output to the current phase control section 64 or the d-axis current control section 66 as described above, and the current phase control section 64 or the d-axis current control It is used as an initial value in section 66 and as an integral value for integral control. Also, the voltage phase θv immediately before switching in the sine wave control section 40 is output to the voltage phase setting section 502 and used as the initial value in the voltage phase setting section 502 and the integral value for the integral control. These takeover values may be smoothed values as described above.

Switching from the sine wave control unit 40 to the voltage phase control unit 50 may be performed, for example, when the voltage command value |Va| Also, the threshold value of the voltage command value |Va| at this time may be a value when the voltage utilization rate of the output voltage is close to the upper limit value of the sine wave control. In addition, when the torque command value T ^* is input such that the output voltage is insufficient in the control by the sine wave control unit 40, or when such torque is output, or when such electrical angular velocity is obtained, switching is performed. may be performed. Furthermore, these switching thresholds may be set in combination with the power supply voltage Vdc and other thresholds.

By the switching operation of the switching unit 24, the torque command value T ^* is input to the voltage phase setting unit 502 of the voltage phase control unit 50 via the low-pass filter LPF. In addition, the motor parameters (φa, Ld, Lq) of the PM motor 10 are input from the motor parameter setting unit 110 to the torque calculation unit 504 of the voltage phase control unit 50, and the d axis from the 3-phase/dq conversion unit 22, q-axis feedback current values Id and Iq are input. Then, the torque calculation unit 504 calculates the current torque T of the PM motor 10 from these motor parameters and the d-axis and q-axis feedback current values Id and Iq, and outputs it to the voltage phase setting unit 502 .

Then, the voltage phase setting unit 502 generates the voltage phase θv such that the difference between the input torque command value T ^* and the torque T becomes zero by well-known calculations such as integral control and proportional control (voltage phase setting step ). An upper limit value of the voltage phase θv corresponding to the power supply voltage Vdc and the electrical angular velocity ω is set in the voltage phase setting unit 502, and the voltage phase setting unit 502 sets the voltage phase θv within the range of the upper limit value. Generate. Then, the generated voltage phase θv is output to voltage command value generation section 516 and carrier setting section 520 of voltage phase control section 50 . Then, carrier setting section 520 generates carrier setting information Sc similar to carrier setting section 420 from voltage phase θv, electrical angular velocity ω, and electrical angle θ, and outputs the same to carrier wave generating section 34 .

Further, the voltage phase control unit 50 has a voltage command setting unit 60, to which one or both of the d-axis and q-axis feedback current values Id and Iq are input, and the voltage command value | is within the range of the upper limit value | _Va | _MAX and the lower limit value | ^Va | A voltage command value |Va| that takes a current is acquired and output to the voltage command value generation unit 516 and the linear correction unit 38 (voltage command setting step). When the voltage command value |Va| is equal to the upper limit value |Va| _MAX , the PM motor 10 performs voltage phase control with the voltage phase θv at the maximum output voltage. At this time, when the voltage phase θv increases in order to output a torque T that is substantially the same as the torque command value T ^* , the d-axis current Id increases in the negative direction, thereby enabling control of the flux-weakening region. becomes. In particular, the upper limit value | _Va | When the crossing rectangular wave forming voltage value |Va'| , the PM motor 10 is controlled by this square wave pattern. The configuration and operation of the voltage command setting section 60 will be described later in detail.

Further, voltage command value generation unit 516 generates d-axis voltage command value Vd and q-axis voltage command value from voltage phase θv input from voltage phase setting unit 502 and voltage command value |Va| input from voltage command setting unit 60. Generate a value Vq (d-axis q-axis voltage command value generating step).

The voltage phase control section 50 may have an offset correction section 70 that corrects offsets that occur in the drive currents Iu, Iv, and Iw due to the deviation of the electrical angle θ acquired by the angle acquisition section 14 or the like. Here, an example of the offset correction section 70 is shown below.

The offset correction section 70 shown in this example has a smoothing section 72 , a correction current generation section 74 , a correction voltage generation section 76 , and a voltage command value correction section 78 . Then, the smoothing unit 72 of the offset correction unit 70 smoothes the d-axis and q-axis feedback current values Id and Iq input via the switching unit 24 by, for example, moving average processing or smoothing processing. Note that the smoothing process here means a process of smoothing the input signals (d-axis and q-axis feedback currents Id and Iq) by performing calculations based on the following equations every arbitrary period.
C=B(1−K)+K×A
where A is the input value (d-axis and q-axis feedback currents Id and Iq), B is the output value after smoothing in the previous cycle, K is the smoothing constant, and C is the output value. (estimated d-axis and q-axis current commands Id ^* and Iq ^* to be described later).

By this smoothing process, a pseudo estimated d-axis current command value Id ^* and an estimated q-axis current command value Iq ^* in which fluctuation components caused by offsets and amplitude imbalances of the drive currents Iu, Iv, and Iw are smoothed are obtained. generated. These estimated d-axis and q-axis current command values Id ^* and Iq ^* are output to the correction current generator 74 .

Further, the d-axis feedback current value Id and the q-axis feedback current value Iq are input to the correction current generation unit 74 , and the correction current generation unit 74 calculates the estimated d-axis current command value Id ^* generated by the smoothing unit 72 . , the d-axis feedback current value Id and the q-axis feedback current value Iq are subtracted from the estimated q-axis current command value Iq ^* , respectively. As a result, the d-axis correction current ΔId and the q-axis correction current ΔIq are generated as fluctuation components. Then, these d-axis correction current ΔId and q-axis correction current ΔIq are output to the correction voltage generator 76 . Note that the d-axis correction current ΔId and the q-axis correction current ΔIq are obtained from the estimated d-axis and q-axis current command values Id ^* and Iq ^* in which offset and amplitude unbalance components (fluctuation components) are smoothed. Since it is obtained by subtracting the d-axis and q-axis feedback current values Id and Iq including the balance component (fluctuation component), basically the opposite phase of the fluctuation component is obtained.

Further, the correction voltage generation unit 76 generates the d-axis correction voltage ΔVd and the q-axis correction voltage from the d-axis correction current ΔId and the q-axis correction current ΔIq input from the correction current generation unit 74 by, for example, proportional control using a predetermined correction gain. ΔVq is generated and output to voltage command value correction unit 78 .

Voltage command value correction unit 78 converts d-axis correction voltage ΔVd and q-axis correction voltage ΔVq input from correction voltage generation unit 76 into d-axis voltage command value Vd and q-axis voltage command value Vq output from voltage command value generation unit 516 . are added to each. Therefore, in the d-axis and q-axis voltage command values Vd and Vq thus generated, the opposite voltages (d-axis and q-axis correction voltages ΔVd and ΔVq) of the offset and amplitude imbalance components generated in the drive currents Iu, Iv and Iw are applied. is added. These d-axis voltage command value Vd and q-axis voltage command value Vq are input to the linear correction unit 38 of the control signal generation unit 30 via the switching unit 24 . It should be noted that the d-axis voltage command value Vd and the q-axis voltage command value Vq corrected by the offset correction unit 70 are added with voltages opposite to the offset and the amplitude unbalance component as described above. The offset and the like of the PM motor 10 which is caused by the above are corrected and eliminated (offset correction step).

Next, the configuration and operation of the control signal generator 30 will be described. First, during control by the sine wave control unit 40, the voltage command value |Va| from the polar coordinate conversion unit 418 and the d-axis, q Axial voltage command values Vd and Vq are input. Acquired by the voltage command setting unit 60 and the d-axis and q-axis voltage command values Vd and Vq generated by the voltage command value generation unit 516 are input during control by the voltage phase control unit 50. do. Further, the linear correction unit 38 is preset with table data of the magnification using the voltage command value |Va| as an argument. Linear correction is performed by multiplying the voltage command value Vd and the q-axis voltage command value Vq. This linear correction corrects and linearizes the nonlinearity between the change in the voltage command value |Va| and the voltage output by the inverter 20 .

The electrical angle θ from the angle acquisition unit 14 and the electrical angular velocity ω from the angular velocity calculator 16 are input to the dq/three-phase converter 32 constituting the control signal generator 30, and the electrical angle θ and the electrical angular velocity ω are input. , and the d-axis and q-axis voltage command values Vd and Vq are calculated as three-phase voltage command values based on this predicted electrical angle θ'. They are converted into Vu, Vv, and Vw and output to the drive signal generator 36 .

Further, the drive signal generator 36 has a carrier wave generator 34. The carrier setting information Sc is input to the carrier wave generator 34, and the carrier wave having the above-described period is generated based on the carrier setting information Sc. . The drive signal generator 36 compares the carrier wave with the three-phase voltage command values Vu, Vv, and Vw, respectively, and thereby generates Hi-Low drive signals Su, Sv, and Sw. Incidentally, the comparison operation means that the drive signal generator 36 compares the magnitudes of the carrier wave and the three-phase voltage commands Vu, Vv, and Vw, and determines the Hi-Low of the drive signals Su, Sv, and Sw according to the magnitude relationship. This is a setting operation, which generates drive signals Su, Sv, and Sw that switch between Hi and Low at respective intersections of the carrier wave and the three-phase voltage commands Vu, Vv, and Vw.

The inverter 20 turns on and off internal switching elements according to the driving signals Su, Sv, and Sw output from the driving signal generating section 36, and the DC power from the DC power supply section 18 is generated based on the driving signals Su, Sv, and Sw. Converts to AC voltage and outputs. As a result, AC driving currents Iu, Iv, and Iw whose phases are shifted by 1/3 cycle (2/3π (rad)) flow down through the armature windings of the PM motor 10, respectively. As a result, the PM motor 10 rotates with a torque corresponding to the torque command value T ^* .

Next, the voltage command setting unit 60 and the voltage command setting step, which are characteristic configurations of the motor control device 100 and the motor control method according to the present invention, will be described. First, when the voltage command value | ^Va | is within the range between the upper limit value | _Va | _MAX and the lower limit value |Va| obtains the voltage command value |Va| that takes the minimum current value |Ia| Further, when the voltage command value |Va| takes the upper limit value |Va| _MAX (when it is limited by the upper limit value |Va| _MAX ), the maximum Voltage phase control is performed using the voltage phase θv with the output voltage.

Next, the configuration of the voltage command setting section 60a of the first mode corresponding to Claims 7, 8, 21 and 22 of the present application will be described with reference to FIG. A voltage command setting unit 60a according to the first embodiment of the present invention shown in FIG. is doing. The current phase calculator 62 also has an Ia absolute value calculator 622 , a phase calculator 624 , and a θi subtractor 63 . Then, the d-axis feedback current value Id and the q-axis feedback current value Iq from the three-phase/dq conversion unit 22 are input to the Ia absolute value calculation unit 622, and the Ia absolute value calculation unit 622 calculates the input d-axis, q The magnitude |Ia| of the current vector is calculated from the shaft feedback current values Id and Iq based on the following equation (1).
|Ia|=(Id ² +Iq ² ) ^1/2 (1)

The d-axis feedback current value Id and the q-axis feedback current value Iq are input to the phase calculation unit 624, and the phase calculation unit 624 calculates the following (2 ) is used to calculate the current phase θi.
θi=tan ⁻¹ (−Id/Iq) (2)
Note that the d-axis and q-axis feedback current values Id and Iq input to the voltage command setting unit 60a of the first mode and the voltage command setting units 60b and 60c of the second and third modes to be described later are obtained by low-pass filtering or averaging. A value pre-smoothed by processing may be used.

Next, the current phase calculator 62 refers to the current-phase angle data map 620 using |Ia| acquired by the Ia absolute value calculator 622 as an argument, and the target current phase θi _{(base )} is read and obtained. Further, this target current phase θi _(base) may be obtained by calculation using a formula for calculating the target current phase θi _(base) as a function of the absolute value |Ia| (the above corresponds to the current phase calculation step). . When the target current phase θi _(base) is obtained by reading from the data map, the current-phase angle data map 620 shows the maximum torque T for each magnitude |Ia ^* | of the current command value Ia ^* as described above. The target current phase θi _(base) is recorded as table data, and the target current phase θi _(base) can be obtained by reading directly from the current-phase angle data map 620 . Further, when obtaining the target current phase θi _(base) by calculation, the current phase calculator 62 has a calculation formula for the target current phase θi _(base) , and this calculation formula is obtained in advance by experiments or the like, for example. Based on the data of |Ia| and the target current phase θi ₍ base), a formula for calculating the target current phase θi _(base) as a function of |Ia| By substituting the value of |Ia| obtained by the Ia absolute value calculator 622 into this calculation formula, the target current phase θi _(base) is obtained. As a method of creating the calculation formula, for example, after dividing | _Ia | However, these may be selected and used according to the value (the interval) of |Ia| acquired by the Ia absolute value calculator 622 . Also, for example, |Ia| and the target current phase θi _(base) are plotted on a graph, and an approximate curve formula (generally a high-order polynomial function) for this plot is obtained for the target current phase θi _(base) . A calculation formula may be used.

Next, the current phase calculator 62 confirms the sign (positive or negative) of the q-axis feedback current value Iq, and sets the sign of the target current phase θi _(base) to be the same as the q-axis feedback current value Iq. The signs of the current phase θi and the target current phase θi _(base) are positive in the counterclockwise direction and negative in the clockwise direction with reference to the q-axis current Iq in FIG. 4 described later.

Next, the target current phase θi _(base) and the current phase θi are input to the θi subtractor 63, and the current phase θi is subtracted from the target current phase θi _(base) as shown in the following formula, and the difference Δθi is calculated (θi subtraction step).
Δθi = θi _(base) - θi
Then, this difference Δθi is output to the input switching section 628 .

Also, the q-axis feedback current value Iq is input to the q-axis current determination unit 626, and the q-axis current determination unit 626 determines whether the input q-axis feedback current value Iq is within a preset range of values near zero. (q-axis current determination step). Then, when it is determined that the q-axis feedback current value Iq is not within the range of values near zero, the input switching section 628 outputs the difference Δθi to the current phase control section 64 . Then, the current phase control unit 64 acquires the voltage command value |Va| based on the difference Δθi, which will be described later. Further, when it is determined that the q-axis feedback current value Iq is within the range of values near zero, the input switching unit 628 switches the output to the current phase control unit 64 from the difference Δθi to the d-axis feedback current value Id ( input switching step). Then, the current phase control unit 64 acquires the voltage command value |Va| based on the near-zero process described later.

Next, a process for acquiring the voltage command value |Va| based on the difference Δθi will be described. First, when the control of the PM motor 10 is switched from the sine wave control unit 40 to the voltage phase control unit 50, the voltage command value |Va| Enter as a value. Also, the q-axis feedback current value Iq is input to the current phase control section 64, and the current phase control section 64 confirms whether the q-axis feedback current value Iq is positive or negative.

Also, the current phase control unit 64 is set with an upper limit value |Va| _MAX and a lower limit value |Va| _MIN of the voltage command value |Va| _. Va| _MIN is also used as the upper limit value and the lower limit value of the integral value in the integral control of the current phase control section 64 . Note that the upper limit value |Va| _MAX and the lower limit value | _Va | For example, the upper limit value |Va| _MAX may be a rectangular wave forming voltage value |Va'| such that the driving signals Su, Sv, and Sw form a rectangular wave pattern of one pulse. When not controlled, it may be set to a value lower than the rectangular wave forming voltage value |Va'|. When the upper limit value |Va| _MAX is set based on the rectangular wave forming voltage value |Va'|, the upper limit value |Va| _MAX changes according to the number of synchronizations. Alternatively, a data map of the upper limit value |Va| _MAX may be created for each of the power supply voltage Vdc of the DC power supply unit 18 and the electrical angular velocity ω, and this data map may be read out and set. Further, these methods for setting the upper limit value |Va| _MAX may be used in combination as appropriate. The lower limit value |Va| _MIN may be a fixed value, _or a data map of the lower limit value |Va| Also good. If the value at which the induced voltage by the rotor and the output voltage of the inverter 20 are equal is set as the lower limit value |Va| _MIN in table data, the voltage command value |Va| becomes the lower limit value |Va| _MIN . It is possible to minimize the d-axis current (flux-weakening current) in the reduced state. Thereby, the loss in this operating region can be suppressed.

Then, when the difference Δθi is input from the current phase calculator 62, the current phase controller 64 performs well-known proportional control and integral control based on the value of the difference Δθi. is positive, the voltage command value │Va│ is decreased within the range of the lower limit value │Va│ _MIN . When the sign of the q-axis feedback current value Iq is positive and the sign of the difference Δθi is negative, the voltage command value |Va| is increased within the range of the upper limit value |Va| _MAX . When the sign of the q-axis feedback current value Iq is negative and the sign of the difference Δθi is negative, the voltage command value |Va| is decreased within the range of the lower limit value |Va| _MIN . Further, when the sign of the q-axis feedback current value Iq is negative and the sign of the difference Δθi is positive, the voltage command value |Va| is increased within the range of the upper limit value |Va| _MAX (current phase control step・When the difference Δθi is input). Then, this voltage command value |Va| is output to the voltage command value generation unit 516 and the linear correction unit 38 .

Here, for example, when the sign of the q-axis feedback current value Iq is positive and the sign of the difference Δθi is positive, the current phase θi is positioned on the q-axis side of the target current phase θi _(base) . This state means that a current larger than the minimum target current phase .theta.i _(base) is flowing, and is not a preferable state because the copper loss is large especially at low load when the torque command value T ^* is small. The reason for such a state is thought to be that the output voltage of the _inverter 20 is large. It is assumed to be greater than T ^* . Therefore, by decreasing the voltage command value |Va|, |Ia| is reduced and the voltage phase θv is moved to the d-axis side, so that the current phase θi is equal to the target current phase θi _(base) .

Thus, the current phase control unit 64 increases or decreases the voltage command value |Va| depending on whether the q-axis feedback current value Iq is positive or negative and the difference Δθi is positive or negative. The axis current Id and the q-axis current Iq change, and the torque T changes accordingly. Then, the voltage phase setting unit 502 increases or decreases the voltage phase θv so that the torque T matches the torque command value T ^* , and the d-axis currents Id and q corresponding to the voltage command value |Va| The current phase θi due to the shaft current Iq changes so as to approach the target current phase θi _(base) . As a result, the d-axis and q-axis voltage command values Vd and Vq are equal to the target current phase θi _(base) , which takes the minimum current value |Ia| to output the same torque T as the torque command value T ^* . The phase θi is controlled, and as a result, the PM motor 10 operates in an efficient state with little loss.

Further, when the q-axis current determination unit 626 determines that the q-axis feedback current value Iq is within the range of values near zero, the input switching unit 628 switches the output to the current phase control unit 64 from the difference Δθi to the d-axis Switch to the feedback current value Id. It is preferable that the determination threshold in the q-axis current determining section 626 has a hysteresis width to prevent frequent switching at the boundary point. Moreover, at the time of this switching, the control gains of the proportional control, the integral control, etc. in the current phase control section 64 may be switched to those for near-zero processing.

Here, in the process of obtaining the voltage command value |Va| based on the difference Δθi, the current phase θi is calculated by the above equation (2). However, since the term (-Id/Iq) exists in the above formula (2), when the q-axis feedback current value Iq, which is the denominator, is zero or near zero, the numerical value of this term increases to infinity. A computational error occurs. Further, when the q-axis feedback current value Iq is near zero and the d-axis feedback current value Id also alternates between positive and negative values near zero, the current phase θi calculated by the above equation (2) corresponding to the positive/negative value is approximately − The voltage command value |Va| may fluctuate between 90° and about +90°, causing hunting. Therefore, when the q-axis feedback current value Iq is close to zero, the q-axis current determination unit 626 and the input switching unit 628 switch the output to the current phase control unit 64 from the difference Δθi to the d-axis feedback current value Id. The voltage command value |Va| is generated without using the phase θi. As a result, the term (-Id/Iq) is not involved in the operation of the current phase control unit 64, and problems in calculation can be avoided.

In the case of near-zero processing, the current phase control unit 64 performs well-known proportional control and integral control based on the input d-axis feedback current value Id. Decrease the voltage command value |Va| within the range of the lower limit value |Va| _MIN . Further, when the sign of the d-axis feedback current value Id is negative, the voltage command value |Va| is increased within the range of the upper limit value |Va| _MAX (current phase control step/near-zero processing). Then, it outputs to the voltage command value generation unit 516 and the linear correction unit 38 .

Here, when the q-axis feedback current value Iq is near zero, it is necessary to flow the d-axis current as the flux-weakening current. However, when the sign of the d-axis feedback current value Id is positive, the output voltage of the inverter 20 is larger than the induced voltage by the rotor magnet (permanent magnet of the rotor). becomes a current that increases At this time, the current phase control unit 64 operates to decrease the voltage command value |Va|. As a result, when the d-axis current drops, particularly when the lower limit value | _Va | . Further, when the sign of the d-axis feedback current value Id is negative, the current phase control unit 64 increases the voltage command value |Va| to increase the output voltage of the inverter 20, thereby reducing the excessive d-axis current (flux-weakening current). Suppress. As a result, the PM motor 10 can be operated efficiently with little loss. In the configuration shown here, even when both the q-axis feedback current value Iq and the d-axis feedback current value Id take values close to zero, the operation continues to the voltage phase control unit 50 without switching to the sine wave control unit 40. can be controlled. Thereby, continuous and smooth operation control can be performed.

Furthermore, in the near-zero process, when the d-axis feedback current value Id is also within a predetermined range near zero, the voltage command value |Va| may be maintained without being changed. Here, in the near-zero process, the direction of increase or decrease of the voltage command value |Va| changes depending on whether the d-axis feedback current value Id is positive or negative. Therefore, when the d-axis feedback current value Id alternates between positive and negative in the vicinity of zero, the increase processing and decrease processing for the voltage command value |Va| are frequently switched accordingly. In this regard, in a configuration in which the original voltage command value |Va| can be suppressed, and the torque change of the PM motor 10 can be prevented.

Further, as shown in FIG. 3B, the offset value (-a) is added to the d-axis feedback current value Id input to the current phase control unit 64, and the determination threshold value near zero of the d-axis feedback current value Id is set to It may be shifted corresponding to this offset value (-a). According to this configuration, by optimizing the value of the offset value (-a), the d-axis feedback current value input to the current phase control unit 64 when the q-axis feedback current value Iq takes a value near zero. Id can always be negative. In this configuration, when the input to the current phase control unit 64 is switched between the d-axis feedback current value Id and the difference Δθi, the current phase θi immediately after switching is offset by the offset value (−a), and the target current phase θi _{( base)} and relatively smooth switching can be performed.

Next, the configuration of the voltage command setting unit 60b of the second embodiment of the present invention corresponding to claims 1 to 6 and claims 15 to 20 of the present application will be described. Note that the voltage command setting unit 60b of the second embodiment has a d-axis current value Id (ref) that takes the target current phase θi _{(base) shown} in the vector diagram of FIG. 4 instead of the target current phase θi (base). is used to change the voltage command value |Va|.

As shown in FIGS. 5(a) to 5(d) and 6, the voltage command setting unit 60b of the second embodiment according to the present invention has a minimum voltage for outputting a torque T substantially the same as the torque command value T ^*. The Id (ref) acquisition units 644a to 644d acquire the d-axis current value Id (ref) at the target current phase θi _(base) that takes the current |Ia| An Id subtraction unit 648 that subtracts the d-axis feedback current value Id from the obtained d-axis current value Id(ref) to obtain a difference ΔId, and a d-axis voltage command value |Va| that changes the voltage command value |Va| and a current control unit 66 . Note that the d-axis current value Id(ref) basically indicates a negative value.

Here, Id (ref), coefficient K _(Id/Iq) , target current phase θi _(base) , etc. will be described by reading from the data map. You can In this case, the Id (ref) acquisition units 644a to 644d have calculation formulas for Id (ref), coefficient K _(Id/Iq) , target current phase θi _(base) , etc. Based on the acquired data, as in the case of the current phase calculation unit 62, linear functions in a plurality of sections are connected and configured, or created and recorded by a known method such as obtaining from an approximate curve formula. It is preferable to keep

Here, the Id (ref) acquisition unit 644a of the first form has, for example, an Id (ref) data map unit 646a. , and its absolute value |Iq| as an argument, the d-axis current value Id(ref) is read out and acquired. When the Id(ref) data map unit 646a reads the d-axis current value Id(ref) using the absolute value |Iq| of the q-axis feedback current value Iq as an argument, the Id(ref) acquisition unit of the first form 644a, as shown in FIG. 5A, has an Iq absolute value calculator 642 that calculates the absolute value |Iq| of the q-axis feedback current value Iq. In this case, the Id (ref) data map unit 646a has table data of the d-axis current value Id (ref) that takes the target current phase θi _(base) for each absolute value |Iq|. The d-axis current value Id(ref) is read out from the absolute value |Iq| When the Id(ref) data map unit 646a reads out the d-axis current value Id(ref) using the q-axis feedback current value Iq itself as an argument, the Id(ref) data map unit 646a It has table data of the d-axis current value Id (ref) corresponding to Iq. Then, the Id(ref) data map unit 646a uses the q-axis feedback current value Iq input to the Id(ref) acquisition unit 644a as a direct argument to read and acquire the d-axis current value Id(ref).

These table data of the Id (ref) data map section 646a are obtained by experiments or the like with the data of |Ia| and the target current phase θi _(base) , and the q-axis feedback The relationship between the current value Iq or the absolute value |Iq| and the d-axis current value Id(ref) is obtained and can be obtained by converting it into table data.
Iq=|Ia|·cos(θi _(base) )
Id(ref)=−|Ia|·sin(θi _(base) )
According to the configuration of the Id(ref) acquisition unit 644a of the first embodiment, the d-axis current value Id(ref) is read directly using the q-axis feedback current value Iq or its absolute value |Iq| as an argument. It is possible to reduce the load of arithmetic processing related to acquisition of the d-axis current value Id(ref). Note that the Id(ref) acquisition unit 644a uses the formula for calculating the d-axis current value Id(ref) as a function of the q-axis feedback current value Iq or its absolute value |Iq| The current value Id(ref) may be acquired.

Further, the Id (ref ₎ acquisition unit 644b of the second form corresponds to, for example, the q-axis feedback current value Iq or its absolute value |Iq|, and the q-axis It has a coefficient data map section 646b having table data of a coefficient K _(Id/Iq) which is the ratio of the current value Iq and the d-axis current value Id(ref). Note that the coefficient data map unit 646b obtains the q-axis current value Iq and the d-axis current value Id(ref) in the same manner as the table data of Id(ref) described above, and converts the d-axis current value Id(ref) into the q-axis current value. It can be obtained by dividing by the value Iq to calculate the coefficient K _(Id/Iq) and converting it into table data. Then, when the coefficient data map unit 646b reads the coefficient K _(Id/Iq) using the absolute value |Iq| As shown in (b), an Iq absolute value calculator 642 for calculating the absolute value |Iq| of the q-axis feedback current value Iq is provided. |Iq| is calculated and output, and the coefficient data map unit 646b reads the coefficient K _(Id/Iq) using the absolute value |Iq| Then, the Id(ref) acquisition unit 644b multiplies the read coefficient K _(Id/Iq) by the absolute value |Iq| get.

When the coefficient data map unit 646b reads the coefficient K (Id/Iq) using the q-axis feedback current value Iq itself as an argument, the coefficient data map unit 646b reads the coefficient K _(Id/Iq) corresponding to both positive and negative q-axis feedback current values Iq. It has table data of _(Id/Iq) . Then, the coefficient data map unit 646b uses the q-axis feedback current value Iq input to the Id(ref) acquisition unit 644b as a direct argument to read and acquire the coefficient K _(Id/Iq) . Then, the Id(ref) obtaining unit 644b obtains the d-axis current value Id(ref) by multiplying the read coefficient K _(Id/Iq) by the q-axis feedback current value Iq.

Note that the d-axis current value Id(ref) must be a negative value regardless of the argument. Therefore, when the d-axis current value Id (ref) is a positive value in terms of calculation, it is either multiplied by -1 to make it a negative value, or the coefficient K _(Id/Iq) of the positive value region is set to It is preferable to record as a negative value. In the configuration of the Id(ref) acquisition unit 644b of the second form, although the number of arithmetic operations increases compared to the Id(ref) acquisition unit 644a of the first form, the area where the absolute value |Iq| Since the coefficient K _(Id/Iq) is approximately proportional to the coefficient K (Id/Iq) in this region, the data of the coefficient K _(Id/Iq) can be thinned out and the capacity of the table data in the coefficient data map section 646b can be reduced. It becomes possible. Note that the Id (ref) acquisition unit 644b uses the formula for calculating the coefficient K _(Id/Iq) as a function of the q-axis feedback current value Iq or its absolute value |Iq|, and calculates the coefficient K _{( Id/Iq)} may be acquired.

Also, the Id (ref) acquisition unit 644c of the third form shown in FIG. have. Then, the Ia absolute value calculation unit 622′ calculates the current vector from the d-axis and q-axis feedback current values Id and Iq by Equation (1) in the same manner as the Ia absolute value calculation unit 622 of the voltage command setting unit 60a of the first embodiment. The magnitude |Ia| of is calculated. Then, for example, using the current-phase angle data map 620, with |Ia| as an argument, the target current phase θi _(base) is read out and output to the Id(ref) calculator 646c. Also, the Iq absolute value calculator 642 calculates the absolute value |Iq| of the q-axis feedback current value Iq and outputs it to the Id(ref) calculator 646c. Then, the Id(ref) calculator 646c calculates the d-axis current value Id(ref) based on the following equation.
Id(ref)=−|Iq|・tan(θi _(base) ) (3)
Alternatively, the d-axis current value Id(ref) may be calculated based on the following equation without providing the Iq absolute value calculator 642 .
Id(ref)=−|Ia|·sin(θi _(base) )
In the Id (ref) acquisition unit 644c of the third embodiment, the current-phase angle data map 620 is used to eliminate the need for a data map for acquiring the d-axis current value Id (ref). can be reduced.

Furthermore, the Id (ref) acquisition unit _644c ′ of the third embodiment uses the absolute value |Iq| of the q-axis feedback current value Iq as an argument, for example, as shown in FIG. ₎ is provided, the target current phase θi _(base) is read using the absolute value |Iq| calculated by the Iq absolute value calculator 642 as an argument, and the d-axis current value Id ( ref) may be calculated. Further, a data map of the target current phase θi _(base) is provided using the positive and negative q-axis feedback current value Iq as an argument, and the target current phase θi _(base) is read out using the q-axis feedback current value Iq directly as an argument. (3) may be used to calculate the d-axis current value Id(ref). Note that the Id (ref) acquisition units 644c and 644c′ calculate the target current phase θi _(base) as a function of the current vector magnitude |Ia| or the q-axis feedback current value Iq or its absolute value |Iq| may be used to obtain the target current phase θi _(base) by calculation as described above.

Further, the Id (ref) acquisition unit 644d of the fourth embodiment reads out and acquires the d-axis current value Id (ref) using, for example, the torque command value T ^* or its absolute value |T ^* | as an argument T-Id ( ref) It has a data map section 469 . When the T-Id(ref) data map unit 469 reads the d-axis current value Id(ref) using the absolute value |T ^* | of the torque command value T ^* as an argument, Id(ref) of the fourth form As shown in FIG. 6, the acquisition unit 644d has a torque absolute value calculation unit 643 that calculates the absolute value |T ^* | of the torque command value T ^* . In this case, the T-Id (ref) data map section 469 has table data of the d-axis current value Id (ref) that takes the target current phase θi _(base) at each absolute value |T ^* | , absolute value |T ^* | Further, when the T-Id (ref) data map unit 469 reads the d-axis current value Id (ref) using the torque command value T ^* itself as an argument, the T-Id (ref) data map unit 469 can read both positive and negative values, for example. It has table data corresponding to the torque command value T ^* , torque command value T ^* for powering operation, and table data for each torque command value T ^* for regenerative operation. Id(ref) is obtained by reading out the command value T ^* as a direct argument. Table data of T-Id(ref) is obtained in advance by experiment or the like. Note that the Id(ref) acquisition unit 644d uses the formula for calculating the d-axis current value Id(ref) as a function of the torque command value T ^* or its absolute value |T ^* | The current value Id(ref) may be acquired. The operations of these Id(ref) acquisition units 644a to 644d correspond to the Id(ref) acquisition step.

The d-axis current value Id(ref) thus acquired is input to the Id subtraction unit 648, where the d-axis feedback current value Id is subtracted to obtain a difference ΔId (ΔId acquisition step). The difference ΔId calculated by the Id subtraction unit 648 is input to the d-axis current control unit 66, and the d-axis current control unit 66 performs well-known proportional control and integral control based on the value of the difference ΔId to determine the sign of the difference ΔId. is negative, the voltage command value │Va│ is decreased within the range of the lower limit value │Va│ _MIN . When the sign of the difference ΔId is positive, the voltage command value |Va| is increased within the range of the upper limit value |Va| _MAX (voltage command generation step). This operation brings the current phase .theta.i closer to the target current phase ^.theta.i _(base) as in the voltage command setting unit 60a of the first embodiment. is controlled at a current phase .theta.i equivalent to the target current phase .theta.i _(base) that takes the minimum current value .vertline.Ia.vertline. Note that the upper limit value |Va| _MAX and the lower limit value | _Va | Also, these upper limit value |Va| _MAX and lower limit value | _Va |

Here, the voltage command setting unit 60b of the second form does not use the current phase θi unlike the configuration of the voltage command setting part 60a of the first form. Therefore, it is possible to simplify the control system by eliminating the need for a special switching operation when the q-axis feedback current value Iq takes a value near zero. Further, since consistent control can be performed with the same configuration using the difference ΔId, signal continuity can be maintained and smooth control can be performed.

Also, the voltage command setting units 60a and 60b of the first and second forms may be configured to include a correction voltage calculation unit 515 as shown in FIG. Here, in the voltage command setting units 60a and 60b having the correction voltage calculation unit 515, the correction voltage calculation unit 515 generates the correction voltage |Va''| and outputs it to the current phase control unit 64 and the d-axis current control unit 66. do. Then, the current phase control unit 64 and the d-axis current control unit 66 add the corrected voltage |Va''| to the voltage command value generated by themselves to obtain the final voltage command value |Va|. 516. At this time, the upper and lower limit values of the integral value in the integral control of the current phase control section 64 and the d-axis current control section 66 are obtained by subtracting the correction voltage |Va''| from the original upper and lower limit values. restrictions. Also, the upper limit value |Va| _MAX and the lower limit value |Va| _MIN of the voltage command setting units 60a and 60b are set to Limit.

The method of generating the correction voltage |Va''| by the correction voltage calculation unit 515 is, for example, the induced voltage constant φa of the permanent magnet, the d-axis current value Id (d-axis feedback current value Id), or the q-axis current value Iq (q It is preferably generated based on at least one of the shaft feedback current values Iq) and the electrical angular velocity ω. Here, as a specific method of generating the correction voltage |Va''|, for example, it may be calculated based on the following formula considering the induced voltage of the permanent magnet corresponding to the change in the electrical angular velocity ω.
|Va''|=|ω・φa|
It may be calculated based on the following formula considering the q-axis current value Iq.
|Va''|=|ω·Lq·Iq| (Lq: q-axis inductance)
Alternatively, it may be calculated based on the following formula, which considers the flux-weakening control by the d-axis current value Id with respect to the induced voltage by the permanent magnet.
|Va''|=|ω·φa+ω·Ld·Id| (Ld: d-axis inductance)
Furthermore, by synthesizing the above equations, the correction voltage |Va''| You can do it.
Vd'=-ω·Lq·Iq
Vq′=ω·φa+ω·Ld·Id
|Va″|=(Vd′ ² +Vq′ ² ) ^1/2
Note that the d-axis current value Id and the q-axis current value Iq in these equations may be values obtained by smoothing the feedback current values Id and Iq using a low-pass filter or the like.

In the configuration of FIG. 7 including these correction voltage calculation units 515, the correction voltage |Va''| is generated based on the electrical angular velocity ω, and the correction voltages |Va''| Since |Va| is generated, it is possible to generate a more appropriate voltage command value |Va| corresponding to changes in the electrical angular velocity ω compared with the voltage command value generated only by proportional control and integral control. Furthermore, in a configuration where the d-axis current value Id(ref) can be obtained, the correction voltage |Va''| may be calculated using Id(ref) instead of the d-axis current value Id in the above equation. . With this configuration, it is possible to generate the correction voltage |Va''| corresponding to the value of the d-axis current value Id(ref), which is the target value. can generate a more appropriate voltage command value |Va|. In particular, in the configuration using the voltage command setting unit ^60b of the second embodiment shown in FIG. 6, the correction voltage |Va ''| can be generated, a more appropriate voltage command value |Va| can be generated. In the configuration including the correction voltage calculation unit 515, when switching from the sine wave control unit 40 to the voltage phase control unit 50, the correction voltage |Va''| may be used as the initial value of the current phase control section 64 or the d-axis current control section 66 and the integral value of the integral control.

Furthermore, the current phase control unit 64 and the d-axis current control unit 66 of the voltage command setting units 60a and 60b of the first and second forms have a shortage of the current torque value T with respect to the torque command value T ^* . In this case, the operation to reduce the voltage command value |Va| may not be performed. In this configuration, when a voltage command value |Va| larger than the current value is required, such as when increasing the torque T, the operation of reducing the voltage command value |Va| is not performed. Therefore, especially when the torque T of the PM motor 10 is insufficient, it is possible to prevent the response delay of the torque T due to the processing of the current phase control section 64 and the d-axis current control section 66 . Further, the current phase control unit 64 and the d-axis current control unit 66 do not increase the voltage command value |Va| when the current torque value T is excessive with respect to the torque command value T ^* . can be In this configuration, when a voltage command value |Va| smaller than the current value is required, such as when reducing the torque T, the voltage command value |Va| is not increased. This makes it possible to prevent a delay in the response of the torque T due to the processing of the current phase control section 64 and the d-axis current control section 66, particularly when the torque T of the PM motor 10 is excessive.

Furthermore, even when the input values are the same, the current phase control unit 64 and the d-axis current control unit 66 control the change width of the output voltage command value |Va| You may make it change according to the value of (omega). That is, when the electrical angular velocity ω is low, the variation width of the voltage command value |Va| is decreased, and when the electrical angular velocity ω is high, the variation width of the voltage command value |Va| is increased. In this configuration, since the amount of change in the voltage command value |Va| is small at a low electrical angular velocity ω, it is possible to stably control the operation during low-speed rotation. Further, since the amount of change in the voltage command value |Va| increases when the electrical angular velocity ω is high, it is possible to improve the response speed of the voltage command value |Va| during high-speed rotation. Thereby, stable control can be performed from a low electrical angular velocity ω to a high electrical angular velocity ω. As a method of changing the variation width of the voltage command value |Va|, for example, there is a method of increasing or decreasing control gains of proportional control and integral control according to the electrical angular velocity ω. Further, for example, the difference Δθi or the difference ΔId input to the current phase control unit 64 and the d-axis current control unit 66 is corrected by multiplying by a coefficient that increases or decreases according to the electrical angular velocity ω, and based on this corrected difference Δθi or difference ΔId. is used to calculate the voltage command value |Va|. Alternatively, for example, a method of providing a limiter to the change width of each control cycle of the proportional control and the integral control of the current phase control unit 64 and the d-axis current control unit 66, and increasing or decreasing the limit value of this limiter according to the electrical angular velocity ω. etc.

Furthermore, the current phase control unit 64 and the d-axis current control unit 66 of the voltage command setting units 60a and 60b of the first and second forms are provided in order to improve responsiveness to sudden changes in the torque command value T ^* . , may have the following configuration. First, as the first ^{configuration} , as ^shown in FIG. A decrease exceeding a preset threshold value is detected (first variation monitoring step). The threshold value of the torque command value T ^* at this time may be a fixed value, or may be set corresponding to either or both of the power supply voltage Vdc and the electrical angular velocity ω. Further, the low-load torque command value T ^* detected by the fluctuation monitoring unit 650a may be the torque command value T ^* for regenerative operation. At this time, the threshold value of the torque command value T ^* may be set to a different value for the power running torque command value T ^* and the regeneration torque command value T ^* , or may be set to a common value for both. can be

Then, when the fluctuation monitoring unit 650a detects a decrease exceeding the threshold value of the torque command value T ^* , the current phase control unit 64 and the d-axis current control unit 66 confirm the integral value of their own integral control. Decrease this integral value according to a predetermined method when it is greater than a predetermined threshold (fast response step). For example, when the torque command value T ^* is relatively high and the integrated value of the current phase control unit 64 and the d-axis current control unit 66 is greater than 90% of the upper limit value | _Va | When a torque command value T ^* exceeding a low load is input, the current phase control unit 64 and the d-axis current control unit 66 integrate a preset small value, for example, a value that is 90% of the upper limit value |Va| _MAX . A value is substituted to calculate the voltage command value |Va|. The integral value to be substituted may be a fixed value, or may be set based on the upper limit value |Va| _MAX . Further, the torque command value T ^* , the power supply voltage Vdc, and the electrical angular velocity ω may all be set, or the torque command value T ^* and the electrical angular velocity ω, or the torque command value T ^* and the power supply voltage Vdc may be set. You can set it accordingly. Furthermore, when the upper limit value |Va| _MAX changes with the number of synchronizations, the integration value may be set after considering the number of synchronizations for these parameters. Furthermore, instead of changing the _integral value, the upper limit value | _Va | By limiting Va| to a small value, the integrated value and the voltage command value |Va| may be decreased. Then, according to the above configuration having the variation monitoring unit 650a, when the torque command value T ^* decreases beyond the threshold value and the integral value is large, the integral value is decreased to calculate the voltage command value |Va| . As a result, the value of the voltage command value |Va| can be quickly decreased, and the PM motor 10 can be quickly operated in an efficient state with little loss.

Note that the integrated value of the current phase control unit 64 or the d-axis current control unit 66 when the torque command value T ^* takes a value near the threshold value of the fluctuation monitoring unit 650a is sufficiently smaller than the threshold value of the integration value. It is preferable to set the threshold value of the torque command value T ^* so that In this configuration, when the torque command value T ^* is near the threshold and decreases beyond the threshold (because the integrated value of the current phase control unit 64 or the d-axis current control unit 66 is smaller than the threshold), integration No value decrementation is performed. That is, when the torque command value T ^* fluctuates around the threshold value, the integral value is calculated by normal integral control to avoid frequent reduction operations. As a result, it is possible to prevent the control of the PM motor 10 from becoming unstable and torque fluctuation from occurring.

Further, as a second configuration, as shown in FIG. 8B, a variation monitoring unit 650b provided in the voltage command setting units 60a and 60b monitors the voltage phase θv output from the voltage phase setting unit 502, When this voltage phase θv becomes the upper limit value of θv or near the upper limit value (fluctuation monitoring step of the second form), the control gain (integral control , one or both of proportional control), the integral value of the integral control of the current phase control section 64 and the d-axis current control section 66, or both the control gain and the integral value are increased (fast response step). When increasing the integrated value of the current phase control unit 64 and the d-axis current control unit ⁶⁶ , the increase width of the integrated value per one control cycle may be a fixed value. may be set based on the difference between Further, when the current torque T is small with respect to the torque command value T ^* and the difference between the two is large, the increment of the integrated value may be increased. Further, when the integral value is increased, the output of the proportional control of the current phase control section 64 and the d-axis current control section 66 and the change in the integral value due to the integral control may be ignored.

Here, for example, when the torque command value T ^* is a low load command value, the voltage command value |Va| output by the voltage command setting units 60a and 60b is a low value. If a torque command value T ^* for a high load is input in this state, the voltage command value |Va| In this case, the PM motor 10 operates with insufficient torque until the voltage command value |Va| follows the torque command value T ^* . However, in the configuration including the fluctuation monitoring unit 650b, the gain of the current phase control unit 64 and the d-axis current control unit 66, for example, the integral control, is increased when the voltage phase θv becomes the upper limit value or near the upper limit value. Or, operate by increasing the integral value of the integral control. As a result, the speed of increase of the voltage command value |Va| increases, and the responsiveness of the voltage command value |Va| improves. As a result, the time during which the PM motor 10 operates with insufficient torque can be shortened. When the voltage phase θv drops to some extent from the upper limit value, the fluctuation monitoring unit 650b detects this, and the control gain or integral value processing of the current phase control unit 64 and the d-axis current control unit 66 returns to normal. do. As a result, fluctuations in the voltage command value |Va| are reduced, and stable control is performed.

Further, as a third configuration, as shown in FIG. The value of the difference ΔId is monitored, and when the difference Δθi or the absolute value of the difference ΔId indicates a value greater than a preset threshold value (the variation monitoring step of the third form), the current phase control unit 64 controls the d-axis current Increase the control gain of the control unit 66 (fast response step). Here, when the torque command value T ^* changes greatly or the power supply voltage Vdc or the electrical angular velocity ω changes suddenly, the difference Δθi or the difference ΔId increases or decreases. Here, when the response speed of the current phase control unit 64 and the d-axis current control unit 66 is slow, a torque shortage or a large loss state occurs until the voltage command value |Va| follows. However, in the voltage command setting units 60a and 60b having the fluctuation monitoring unit 650c, when the absolute value of the difference Δθi or the difference ΔId indicates a value larger than a preset threshold value, the current phase control unit 64 and the d-axis current control unit Since the control gain of 66 is increased, the responsiveness of the voltage command value |Va| becomes faster, and the above-mentioned unfavorable operating state can be quickly resolved to operate the PM motor 10 in an efficient state with less loss. can. When the fluctuation range of the difference Δθi or the difference ΔId decreases to some extent, the fluctuation monitor 650c detects this, and the control gains of the current phase controller 64 and the d-axis current controller 66 return to normal. As a result, fluctuations in the voltage command value |Va| are reduced, and stable control is performed.

In the configuration including the fluctuation monitoring units 650a to 650c, a low-pass filter 652 is further provided for the output of the current phase control unit 64 and the d-axis current control unit 66. Through this low-pass filter 652, the voltage command value |Va| It is preferable to output to the value generator 516 . Here, in the configuration including the fluctuation monitoring units 650a to 650c, when the torque command value T ^* or the like changes suddenly, the integral value and the control gain are changed to improve the responsiveness of the voltage command value |Va|. Therefore, in this operating region, the fluctuation width of the voltage command value |Va| In this regard, in the configuration provided with the low-pass filter 652, the low-pass filter 652 suppresses abrupt changes in the voltage command value |Va|. A sudden change in the voltage command value |Va| can be suppressed. This allows the fluctuation monitoring units 650a to 650c to function more effectively.

As described above, in the motor control device 100 and the motor control method according to the present invention, the voltage command value | _Va | When it is within the range of the value |Va| _MIN , it is controlled to take the current phase θi equivalent to the target current phase θi _(base) . Therefore, the PM motor 10 can be operated efficiently with little loss. At this time, the output voltage has an overmodulation pattern (or a sine wave pattern), so current waveform distortion is improved compared to conventional square wave control, especially at low loads, suppressing torque vibration and noise. be able to. Furthermore, since the voltage phase control unit 50 covers the area of the flux-weakening control, the number of switching of the control method can be reduced, and the configuration of the control system can be simplified. can be planned. of the voltage command value | _Va | of the voltage phase control section 50 (the current phase control section 64 and the d-axis current control section 66). By setting the ^limit value | _Va | The same control method can cover overmodulation control regions using overmodulation patterns, and even sine wave control regions using low-load sine wave patterns. As a result, the number of times the control method is switched can be reduced, and the occurrence of torque shock and response delay at the time of switching can be reduced.

In addition, since the current phase θi is controlled to be equal to the target current phase θi _(base) , the current phase θi is equal before and after switching from the voltage phase control section 50 to the sine wave control section 40, and torque shock is prevented. Less smooth switching operation becomes possible.

It should be noted that the motor control device 100 and the motor control method shown in this example are examples, and the configuration and operation of each unit such as the control signal generation unit 30, the sine wave control unit 40, the voltage phase control unit 50, the voltage command setting unit 60, etc. , and the configuration of each step can be changed without departing from the gist of the present invention.

10 PM motor
12u, 12v drive current acquisition unit
14 angle acquisition unit
20 inverter
22 3-phase/dq converter
30 control signal generator
32 dq/3 phase converter
34 carrier wave generator
50 voltage phase control unit
502 voltage phase setting unit
516 voltage command value generator
520 carrier setting unit
60, 60a, 60b Voltage command setting unit
62 current phase calculator
63 θi subtraction unit
64 current phase controller
66 d-axis current controller
624 phase calculator
626 q-axis current determination unit
628 Input switching unit
644a-644d Id (ref) acquisition unit
648 Id Subtractor
650a-650c Fluctuation monitor
652 low pass filter
70 offset correction unit
100 motor controller
Id d-axis feedback current value
Iq q-axis feedback current value
Sc Carrier setting information
T ^* Torque command value
|Va| Voltage command value
Vd d-axis voltage command value
Vq q-axis voltage command value
θi current phase
θi _(base) target current phase
θv voltage phase
Δθi difference
ΔId difference

Claims (28)

an inverter that causes a three-phase AC drive current to flow down to the PM motor;
a drive current acquisition unit that acquires the value of the drive current;
an angle acquisition unit that acquires the electrical angle of the PM motor;
a three-phase/dq converter that converts the drive current acquired by the drive current acquirer based on the electrical angle into a d-axis feedback current value Id and a q-axis feedback current value Iq;
a voltage phase control unit that sets a voltage phase θv and a voltage command value |Va| according to an external torque command value and generates a d-axis voltage command value Vd and a q-axis voltage command value Vq;
A dq/three-phase converter for converting the d-axis voltage command value Vd and the q-axis voltage command value Vq into three-phase voltage command values; A motor control device comprising: a control signal generator that compares a phase voltage command value and generates a drive signal for switching the inverter,
The voltage phase control unit is
a voltage phase setting unit that sets a voltage phase θv based on the torque command value;
a voltage command setting unit for acquiring a voltage command value |Va| that takes a minimum current to output a torque substantially equal to the torque command value in a region below a predetermined upper limit value; and the voltage phase θv and the voltage command. and a voltage command value generator that generates a d-axis voltage command value Vd and a q-axis voltage command value Vq based on a value |Va|. The voltage command setting part
an Id(ref) acquisition unit that acquires a d-axis current value Id(ref) that takes a minimum current to output a torque substantially equal to the torque command value;
an Id subtraction unit that obtains a difference ΔId between the d-axis current value Id(ref) and the d-axis feedback current value Id;
a d-axis current control unit that reduces the voltage command value |Va| when the difference ΔId is negative, increases the voltage command value |Va| when the difference ΔId is positive, and outputs the voltage command value |Va| to a voltage command value generation unit; 2. The motor control device according to claim 1, further comprising: The Id(ref) acquisition unit acquires the d-axis current value Id(ref) based on the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq. 3. The motor control device according to 2. The Id(ref) obtaining unit obtains the ratio of the d-axis current value Id(ref) corresponding to the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq to the q-axis feedback current value Iq. is obtained, and the coefficient K _{( Id/Iq)} is multiplied by the q-axis feedback current value Iq or the absolute value |Iq| to obtain a negative value d 3. The motor control device according to claim 2, wherein the shaft current value Id (ref) is obtained. An Id (ref) acquisition unit acquires a target current phase θi _(base) that takes a minimum current to output a torque approximately equal to the torque command value, and obtains a d-axis feedback current value Id and a q-axis feedback current value Iq. or the absolute value of the q-axis feedback current value | _Iq |
Or the following formula Id (ref) = - | Iq | tan (θi _(base) )
3. The motor control device according to claim 2, wherein the current value Id (ref) is obtained based on 3. The motor control device according to claim 2, wherein the Id(ref) acquiring unit acquires the d-axis current value Id(ref) based on the torque command value or the absolute value of the torque command value. the voltage command setting unit has a current phase calculation unit, a θi subtraction unit, and a current phase control unit;
The current phase calculator includes an Ia absolute value calculator that acquires the current vector magnitude |Ia| based on the following equation from the d-axis feedback current value Id and the q-axis feedback current value Iq;
|Ia|=(Id ² +Iq ² ) ^1/2
a phase calculator that calculates a current phase θi based on the following equation from the d-axis feedback current value Id and the q-axis feedback current value Iq;
θi=tan ⁻¹ (−Id/Iq)
with
obtaining a target current phase θi _(base) that corresponds to the magnitude |Ia| of the current vector and takes the minimum current for outputting a torque substantially equal to the torque command value;
The θi subtraction unit obtains a difference Δθi by subtracting the current phase θi from the target current phase θi _(base) and outputs the difference Δθi to the current phase control unit;
The current phase control unit is
reducing the voltage command value |Va| when the q-axis feedback current value Iq is positive and the difference Δθi is positive,
increasing the voltage command value |Va| when the q-axis feedback current value Iq is positive and the difference Δθi is negative;
reducing the voltage command value |Va| when the q-axis feedback current value Iq is negative and the difference Δθi is negative,
2. The motor control device according to claim 1, wherein the voltage command value |Va| is increased when the q-axis feedback current value Iq is negative and the difference .DELTA..theta.i is positive. A voltage command setting unit, a q-axis current determination unit that determines whether or not the q-axis feedback current value Iq is a value near zero; an input switching unit for switching the input to the current phase control unit from the difference Δθi to the d-axis feedback current value Id,
The current phase control unit is
decreasing the voltage command value |Va| when the d-axis feedback current value Id is positive,
8. The motor control device according to claim 7, wherein the voltage command value |Va| is increased when the d-axis feedback current value Id is negative. the voltage command setting unit further has a variation monitoring unit that monitors the torque command value,
The variation monitoring unit detects that the torque command value has decreased beyond a preset threshold, and the integral value of the integral control of the current phase control unit or the d-axis current control unit is lower than the preset threshold. 9. The motor control device according to claim 2, wherein the voltage command value |Va| is generated by decreasing the integrated value when the integrated value is large. the voltage command setting unit further has a variation monitoring unit that monitors the voltage phase θv,
When the voltage phase θv becomes a preset upper limit value or its vicinity, the fluctuation monitoring unit controls either the control gain of the current phase control unit or the d-axis current control unit or the integral value of the integral control. 9. The motor control device according to any one of claims 2 to 8, wherein both are increased. the voltage command setting unit further has a variation monitoring unit that monitors the difference ΔId or the difference Δθi,
2. The variation monitoring unit increases the control gain of the current phase control unit or the d-axis current control unit when the difference ΔId or the difference Δθi exceeds a preset value. Item 9. The motor control device according to any one of items 8. The voltage command setting unit outputs the voltage command value |Va| output from the current phase control unit or the d-axis current control unit to the voltage command value generation unit through a low-pass filter. A motor control device according to any one of the preceding claims. The voltage phase control section further has a carrier setting section, the carrier setting section generates carrier setting information based on the voltage phase θv, the electrical angle, and the electrical angular velocity, and the carrier setting information is the carrier wave generated by the carrier wave generating section. 2. The center position of the falling edge crosses the zero position of the rising edge of the three-phase voltage command value, and the frequency of the carrier wave is maintained at an odd integral multiple of 3 of the three-phase voltage command value. 13. The motor control device according to any one of claims 1 to 12. The voltage phase control unit further has an offset correction unit,
The offset correction unit generates a d-axis correction voltage ΔVd and a q-axis correction voltage ΔVq based on the d-axis feedback current value Id and the q-axis feedback current value Iq, respectively. 14. The d-axis correction voltage ΔVd and the q-axis correction voltage ΔVq are added to the value Vd and the q-axis voltage command value Vq, respectively, and output to the control signal generator. A motor controller as described. an inverter that causes a three-phase AC drive current to flow down to the PM motor;
a drive current acquisition unit that acquires the value of the drive current;
an angle acquisition unit that acquires the electrical angle of the PM motor;
a three-phase/dq converter that converts the drive current acquired by the drive current acquirer based on the electrical angle into a d-axis feedback current value Id and a q-axis feedback current value Iq;
a voltage phase control unit that sets a voltage phase θv and a voltage command value |Va| according to an external torque command value and generates a d-axis voltage command value Vd and a q-axis voltage command value Vq;
A dq/three-phase converter for converting the d-axis voltage command value Vd and the q-axis voltage command value Vq into three-phase voltage command values; A motor control method for a motor control device comprising: a control signal generation unit that compares the phase voltage command value and generates a drive signal for switching the inverter,
The voltage phase control unit is
a voltage phase setting step of setting a voltage phase θv based on the torque command value;
a voltage command setting step of obtaining a voltage command value |Va| that takes a minimum current to output a torque substantially equal to the torque command value in a region below a predetermined upper limit value;
a d-axis and q-axis voltage command value generating step of generating a d-axis voltage command value Vd and a q-axis voltage command value Vq based on the voltage phase θv and the voltage command value |Va| control method. The voltage command setting step is
an Id(ref) acquisition step of acquiring a d-axis current value Id(ref) that takes a minimum current to output a torque approximately equal to the torque command value;
a ΔId obtaining step of obtaining a difference ΔId between the d-axis current value Id (ref) and the d-axis feedback current value Id;
a voltage command generating step of decreasing the voltage command value |Va| when the difference ΔId is negative and increasing the voltage command value |Va| when the difference ΔId is positive to generate the voltage command value |Va| 16. The motor control method according to claim 15, further comprising: The Id(ref) acquisition step acquires the d-axis current value Id(ref) based on the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq. 17. The motor control method according to 16. The Id(ref) acquisition step corresponds to the q-axis feedback current value Iq or the absolute value |Iq| of the q-axis feedback current value Iq, and the ratio of the d-axis current value Id(ref) to the q-axis feedback current value Iq is obtained, and the coefficient K _{( Id/Iq)} is multiplied by the q-axis feedback current value Iq or the absolute value |Iq| to obtain a negative value d 17. The motor control method according to claim 16, wherein the shaft current value Id(ref) is obtained. An Id (ref) acquisition step acquires a target current phase θi _(base) that takes the minimum current for outputting approximately the same torque as the torque command value, and obtains the d-axis feedback current value Id and the q-axis feedback current value Iq. or the absolute value of the q-axis feedback current value | _Iq |
Or the following formula Id (ref) = - | Iq | tan (θi _(base) )
17. The motor control method according to claim 16, wherein the current value Id(ref) is obtained based on 17. The motor control method according to claim 16, wherein the Id(ref) acquisition step acquires the d-axis current value Id(ref) based on the torque command value or the absolute value of the torque command value. the voltage command setting step includes a current phase calculation step, a θi subtraction step, and a current phase control step;
The current phase calculation step acquires the magnitude |Ia| of the current vector based on the following equation from the d-axis feedback current value Id and the q-axis feedback current value Iq,
|Ia|=(Id ² +Iq ² ) ^1/2
A current phase θi is calculated based on the following formula from the d-axis feedback current value Id and the q-axis feedback current value Iq,
θi=tan ⁻¹ (−Id/Iq)
Acquiring a target current phase θi _(base) corresponding to the magnitude |Ia| of the current vector and taking the minimum current for outputting substantially the same torque as the torque command value,
The θi subtraction step obtains a difference Δθi by subtracting the current phase θi from the target current phase θi _(base) ,
The current phase control step includes:
reducing the voltage command value |Va| when the q-axis feedback current value Iq is positive and the difference Δθi is positive,
increasing the voltage command value |Va| when the q-axis feedback current value Iq is positive and the difference Δθi is negative;
reducing the voltage command value |Va| when the q-axis feedback current value Iq is negative and the difference Δθi is negative,
16. The motor control method according to claim 15, wherein the voltage command value |Va| is increased when the q-axis feedback current value Iq is negative and the difference Δθi is positive. The voltage command setting step includes a q-axis current determination step of determining whether or not the q-axis feedback current value Iq is a value near zero, and when it is determined that the q-axis feedback current value Iq is a value near zero. an input switching step of switching the input to the current phase control unit from the difference Δθi to the d-axis feedback current value Id,
The current phase control step is
decreasing the voltage command value |Va| when the d-axis feedback current value Id is positive,
22. The motor control method according to claim 21, wherein the voltage command value |Va| is increased when the d-axis feedback current value Id is negative. The voltage command setting step is
a fluctuation monitoring step of detecting that the torque command value decreases beyond a preset threshold;
When the torque command value decreases beyond a preset threshold and the integral value of integral control of the current phase control section or the d-axis current control section is larger than a preset threshold, the integral value is decreased. 23. The motor control method according to any one of claims 16 to 22, further comprising: a fast response step. The voltage command setting step is
a variation monitoring step of detecting that the voltage phase θv has reached or is near a preset upper limit value;
When the voltage phase θv becomes a preset upper limit value or its vicinity, the control gain of the current phase control section or the d-axis current control section or the integral value of the integral control or both are increased at high speed. 23. A motor control method according to any one of claims 16 to 22, further comprising a responding step. The voltage command setting step is
a variation monitoring step of detecting that the difference ΔId or the difference Δθi exceeds a preset value;
16. The method further comprises a high-speed response step of increasing a control gain of the current phase controller or the d-axis current controller when the difference ΔId or the difference Δθi exceeds a preset value. 23. A motor control method according to any one of claims 1 to 22. The voltage command setting step outputs the voltage command value |Va| output from the current phase control section or the d-axis current control section to the voltage command value generation section through a low-pass filter. A motor control method according to any one of the above. The voltage phase control section further has a carrier setting section, the carrier setting section generates carrier setting information based on the voltage phase θv, the electrical angle, and the electrical angular velocity, and the carrier setting information is the carrier wave generated by the carrier wave generating section. 15. The center position of the falling edge crosses the zero position of the rising edge of the three-phase voltage command value, and the frequency of the carrier wave is maintained at an odd integer multiple of 3 of the three-phase voltage command value. 27. The motor control method according to claim 26. further comprising an offset correction step;
In the offset correction step, a d-axis correction voltage ΔVd and a q-axis correction voltage ΔVq are generated based on the d-axis feedback current value Id and the q-axis feedback current value Iq, respectively. 28. Any one of claims 15 to 27, wherein the d-axis correction voltage ΔVd and the q-axis correction voltage ΔVq are added to the value Vd and the q-axis voltage command value Vq, respectively, and output to the control signal generator. Motor control method as described.

Images