DESCRIPTION MOTOR CONTROL DEVICE
Field
[0001] The present invention relates to a motor control
device.
Background
[0002] In a compressor used in an air conditioner or the like, load torque periodically fluctuates during one rotation of a rotor of a motor that drives the compressor. The fluctuations in load torque are caused by changes in pressure of gas refrigerant in each step of suction, compression, and discharge. The periodic fluctuations in load torque causes fluctuations in rotational speed of the motor, which in turn causes vibration and noise. In particular, in a single rotary compressor, vibration tends to increase in a low rotation region. In a case where the compressor having such fluctuations in load torque during one rotation of the rotor is driven, torque control
(periodic disturbance suppression control) is performed to reduce the fluctuations in motor speed.
[0003] The torque control is usually performed in a normal control region such as maximum torque/current control for controlling the low rotation region where vibration is noticeable. However, depending on the specifications of an inverter and a motor and load conditions, vibration occurs even in a voltage saturation region such as in weak magnetic flux control, leading to increase in peak current of the motor. The increase in vibration causes damage to piping and noise in the air conditioner or the like, also causes reduction in efficiency and demagnetization of a compressor motor due to

the increase in peak current of the motor, and a protective function of the inverter to prevent the demagnetization works, so that the motor stops or the like. [0004] Therefore, optimizing fluctuations in output torque for reducing the periodic fluctuations with respect to the fluctuations in speed improves the vibration efficiency of the torque control in the voltage saturation region.
Citation List
Patent Literature
[0005] Patent Literature 1: JP 2019-180173 A
Summary
Technical Problem
[0006] In a case where the torque control is performed in both the normal control region and the voltage saturation region, a current command value generated in the normal control region and a current command value generated in the voltage saturation region are greatly different from each other, and thus, a current deviation inputted to a voltage command generator increases when the control is switched. This causes a problem that the output of the voltage command generator greatly changes and the torque control becomes unstable.
[0007] The present invention has been made in view of the above, and an object of the present invention is to reduce a difference between current command values at a time when a control region is switched between a normal control region and a voltage saturation region, and to prevent instability of the torque control.
Solution to Problem

[0008] In one aspect of the disclosed embodiment, in a motor control device including a current command generator that generates a motor current command value on a basis of a velocity command value and a velocity of a motor, the current command generator includes a first current command generator that generates a current command value in a normal control region, and a second current command generator that generates a current command value in a voltage saturation region. In a case where a control region is switched from the normal control region to the voltage saturation region, or, from the voltage saturation region to the normal control region, the current command generator uses the first current command generator and the second current command generator together to generate the motor current command value.
Advantageous Effects of Invention
[0009] According to an example of an embodiment of the present invention, it is possible to prevent the instability of the torque control at a time when the control region is switched between the normal control region and the voltage saturation region.
Brief Description of Drawings
[0010] FIG. 1 is a block diagram illustrating an example of a motor control device according to an embodiment.
FIG. 2 is a diagram for explaining smoothing processing by a current command smoothing processing unit.
FIG. 3 is a block diagram illustrating an example of the current command smoothing processing unit.
FIG. 4 is diagrams illustrating a current vector while smoothing processing is performed.
FIG. 5 is a diagram illustrating transition of a

current command value while smoothing processing is performed.
Description of Embodiments
[0011] Hereinafter, an example of an embodiment of a motor control device according to the disclosed technology is described with reference to the accompanying drawings. The embodiment described below relates to a motor control device used in, for example, an air conditioner or a low-temperature storage device that performs torque control of a permanent magnet synchronous motor (PMSM) for driving a compressor with periodic fluctuations in load torque by position sensorless vector control. The disclosed technology is, however, widely applicable to a motor control device that performs torque control of a motor for driving load having periodic fluctuations in load torque. Note that a velocity simply referred to in the following description means an angular velocity unless otherwise specified.
[0012] Note that the embodiment described below does not limit the disclosed technology. In addition, the embodiment described below mainly illustrates the configuration and processing according to the disclosed technology, and description of other configuration and processing is simplified or omitted.
[0013] A list of descriptions of main symbols used below is indicated in the following (Table 1).
[0014]
Table 1

Symbol Meaning
C0e Electrical angle estimated angular velocity
COe* Electrical angular velocity command value
C0m Mechanical angle estimated angular velocity
COm* Mechanical angular velocity command value

To* Average torque command value
AT Variable torque command value (correction torque)
■J* Total torque command value
id d-axis current detection value
-Ld mtpi MTPI d-axis current command value
Id fw* Weak magnetic flux d-axis current command value
Iq q-axis current detection value
-Lq mtpi MTPI q-axis current command value
lq fw Weak magnetic flux q-axis current command value
-Ld after d-axis current command value after state transition
-Lq after q-axis current command value after state transition
-Ld before d-axis current command value before state transition
-Lq before q-axis current command value before state transition
Id* d-axis current command value
T *
lq q-axis current command value
Vd-pi Temporary d-axis voltage command value
Vq-pi Temporary q-axis voltage command value
tset Smoothing set time
t Smoothing elapsed time
Gbefore Smoothing gain before
Gafter Smoothing gain after
vd* d-axis voltage command value
Vq* q-axis voltage command value
ee Electrical angle phase (dq-axis phase)
0m Mechanical angle phase
lu U-phase current
Iv V-phase current
-Lw W-phase current
Vu* U-phase output voltage command value
Vv* V-phase output voltage command value
v„* W-phase output voltage command value
[0015] (Embodiment)
[Motor Control Device According to Embodiment] FIG. 1 is a block diagram illustrating an example of a motor control device according to an embodiment. A motor control device 100 according to the embodiment includes

subtracters 11, 18, 19, and 38, a velocity controller 12, adders 13, 16, 17, 21, and 22, a current command generator 14, a control switching determination unit 15, a voltage command generator 20, a d-q/u, v, w converter (two-phase to three-phase converter) 23, a pulse width modulation (PWM) modulator 24, and an intelligent power module (IPM) 25.
[0016] The motor control device 100 also includes a shunt resistor 26, current sensors 27a and 27b, and a 3(() current calculator 28. It is sufficient that the motor control device 100 includes the shunt resistor 26 or the current sensors 27a and 27b.
[0017] The motor control device 100 also includes a u, v, w/d-q converter (three-phase to two-phase converter) 29, an axial error computing unit 30, a phase locked loop (PLL) controller 31, a position estimator 32, a 1/Pn processing unit 33, a correction torque generator 34, infinite impulse responses (iirs) 35a and 35b, a decoupling controller 36, and a current error correction generator 37.
[0018] The subtractor 11 outputs, to the velocity controller 12, an angular velocity deviation Aco obtained by subtracting a mechanical angle estimated angular velocity com, which is the current estimated angular velocity outputted by the 1/Pn processing unit 33, from a mechanical angular velocity command value com* inputted from the outside (for example, a high-level controller) to the motor control device 100.
[0019] The velocity controller 12 generates and outputs an average torque command value To* such that the average of the angular velocity deviation Aco inputted from the subtractor 11 approaches 0 (zero). The adder 13 outputs a total torque command value T* obtained by adding the average torque command value To* outputted by the velocity

controller 12 and a variable torque command value AT outputted by the correction torque generator 34. [0020] The current command generator 14 generates and outputs a d-axis current command value Id* and a q-axis current command value Iq* on the basis of the total torque command value T* outputted by the adder 13 in each of the normal control region and the voltage saturation region. However, when switching between the normal control region and the voltage saturation region is performed, the current command generator 14 generates and outputs a current command value that gradually changes, for a predetermined time after switching, from a current command value before switching to a current command value after switching. The current command generator 14 includes a normal control region current command generator 14a (corresponding to a first current command generator), a voltage saturation region current command generator 14b (corresponding to a second current command generator), and a current command smoothing processing unit 14c (corresponding to a third current command generator).
[0021] In the normal control region, the normal control region current command generator 14a generates and outputs an MTPI d-axis current command value Id_mtpi* and an MTPI q-axis current command value Iq_mtpi* which are intersections of a constant torque curve of the total torque command value T* outputted by the adder 13 and a maximum torque/current (MTPI) curve that reaches the maximum torque at the same current.
[0022] In the voltage saturation region, the voltage saturation region current command generator 14b generates and outputs a weak magnetic flux d-axis current command value Id_fw* and a weak magnetic flux q-axis current command value Iq_fW* that satisfy the total torque command value T*

outputted by the adder 13. The voltage saturation region current command generator 14b generates and outputs the weak magnetic flux d-axis current command value Id_fw* and the weak magnetic flux q-axis current command value Iq_fW* by using a d-axis current detection value Id, a q-axis current detection value Iq, an output voltage limit value Vdq_iimit, a mechanical angle phase 9m, and an electrical angle estimated angular velocity coe. The output voltage limit value Vdq_iimit is obtained by converting a DC voltage Vdc that is supplied from the outside (a power supply converter (not illustrated), for example) to the IPM 25 into a voltage value in a dq rotational coordinate axis system which is a control system.
[0023] The current command smoothing processing unit 14c performs, in a case where switching between the normal control region and the voltage saturation region is performed, smoothing processing on the command value of the current command generator 14 by using the current command values before and after switching, and outputs the resultant. The current command smoothing processing unit 14c outputs a final d-axis current command value Id* and a final q-axis current command value Iq* using the d-axis current command value and the q-axis current command value outputted by the normal control region current command generator 14a and the voltage saturation region current command generator 14b. Note that the details of the current command smoothing processing unit 14c are described later.
[0024] The control switching determination unit 15 determines whether the current control region of a motor 10 is the normal control region or the voltage saturation region on the basis of the output voltage limit value Vdq_iimit, a d-axis voltage command value Vd* as a result of

correcting a temporary d-axis voltage command value Vd_Pi with a d-axis decoupling correction value Vda, and a q-axis voltage command value Vq* as a result of correcting a temporary q-axis voltage command value Vq_pi with a q-axis decoupling correction value Vqa. Then, the control switching determination unit 15 outputs CONTROL_TYPE: A (normal control) in a case where the current control region of the motor 10 is the normal control region, and outputs CONTROL_TYPE: B (voltage saturation control) in a case where the current control region of the motor 10 is the voltage saturation region.
[0025] The adder 16 outputs a d-axis current correction command value Id_FF* that is the result of addition obtained by adding the d-axis current command value Id* outputted by the current command generator 14 and a d-axis current error correction value Aid outputted by the current error correction generator 37. The adder 17 outputs a q-axis current correction command value Iq_FF* that is the result of addition obtained by adding the q-axis current command value Iq* outputted by the current command generator 14 and a q-axis current error correction value Alq outputted by the current error correction generator 37. [0026] The subtractor 18 outputs a d-axis current deviation Id_dif obtained by subtracting the d-axis current detection value Id of the motor 10 outputted by the u, v, w/d-q converter 29 from the d-axis current correction command value Id_FF* outputted by the adder 16. The subtractor 19 outputs a q-axis current deviation Iq_dif obtained by subtracting the q-axis current detection value Iq of the motor 10 outputted by the u, v, w/d-q converter 29 from the q-axis current correction command value Iq_FF* outputted by the adder 17. [0027] The voltage command generator 20 performs

proportional integral (PI) control on each of the d-axis current deviation Id_dif outputted by the subtractor 18 and the q-axis current deviation Iq_dif outputted by the subtractor 19, and outputs a temporary d-axis voltage command value Vd_Pi and a temporary q-axis voltage command value Vq_pi with fluctuation errors reduced.
[0028] The adder 21 outputs a d-axis voltage command value Vd* obtained by adding the temporary d-axis voltage command value Vd_Pi outputted by the voltage command generator 20 and the d-axis decoupling correction value Vda outputted by the decoupling controller 36. Further, the adder 22 outputs a q-axis voltage command value Vq* obtained by adding the temporary q-axis voltage command value Vq_pi outputted by the voltage command generator 2 0 and the q-axis decoupling correction value Vqa outputted by the decoupling controller 36.
[0029] The d-q/u, v, w converter 23 converts two-phase d-axis voltage command value Vd* and q-axis voltage command value Vq* outputted by the adders 21 and 22 to three-phase U-phase output voltage command value Vu*, V-phase output voltage command value Vv*, and W-phase output voltage command value Vw* on the basis of an electrical angle phase
(dq-axis phase) 9e that is the current rotor position outputted by the position estimator 32. The d-q/u, v, w converter 23 then outputs the U-phase output voltage command value Vu*, the V-phase output voltage command value Vv*, and the W-phase output voltage command value Vw* to the PWM modulator 24. The PWM modulator 24 generates six-phase PWM signals on the basis of the U-phase output voltage command value Vu*, the V-phase output voltage command value Vv*, the W-phase output voltage command value Vw*, and a PWM carrier signal (not illustrated), and outputs the six-phase PWM signals to the IPM 25.

[0030] The IPM 25 converts the DC voltage Vdc supplied from the outside on the basis of the six-phase PWM signals outputted by the PWM modulator 24, generates an AC voltage to be applied to each of the U-phase, the V-phase, and the W-phase of the motor 10, and applies each of the AC voltages to the U-phase, the V-phase, and the W-phase of the motor 10.
[0031] In a case where a bus current is detected by a one-shunt method using the shunt resistor 26, the 3(() current calculator 28 calculates a U-phase current value Iu, a V-phase current value Iv, and a W-phase current value Iw of the motor 10 on the basis of the six-phase PWM switching information outputted by the PWM modulator 24 and the detected bus current.
[0032] Alternatively, the method for detecting a current is not limited to the one-shunt method of detecting the bus current, and for example, another method is possible in which a current transformer (CT) is used as the two current sensors, the current sensor 27a detects the U-phase current of the motor 10, and the current sensor 27b detects the V-phase current of the motor 10. In a case where the current sensors 27a and 27b detect the U-phase current and the V-phase current, the 3$ current calculator 28 calculates the remaining W-phase current value Iw by using Kirchhoff's law of Iu + Iv + Iw = 0 (zero) . The 3(() current calculator 28 outputs the calculated phase current values Iu, Iv, and Iw of the respective phases to the u, v, w/d-q converter 29. [0033] The u, v, w/d-q converter 29 converts the three-phase U-phase current value Iu, V-phase current value Iv, and W-phase current value Iw outputted by the 3$ current calculator 28 to two-phase d-axis current detection value Id and q-axis current detection value Iq on the basis of an

electrical angle phase 9e that is the current rotor position outputted by the position estimator 32. The u, v, w/d-q converter 29 then outputs the d-axis current detection value Id to the current command generator 14, the subtractor 18, the axial error computing unit 30, the iir 35a, and the current error correction generator 37, and outputs the q-axis current detection value Iq to the current command generator 14, the subtractor 19, the axial error computing unit 30, the iir 35b, and the current error correction generator 37.
[0034] The axial error computing unit 30 calculates an axial error A9 (difference between an estimated rotation axis and the actual rotation axis) using the d-axis voltage command value Vd* outputted by the adder 21, the q-axis voltage command value Vq* outputted by the adder 22, the d-axis current detection value Id and the q-axis current detection value Iq outputted by the u, v, w/d-q converter 29. The axial error A9 thus calculated is outputted to the PLL controller 31.
[0035] The PLL controller 31 calculates an electrical angle estimated angular velocity coe, which is the current estimated angular velocity, on the basis of the axial error A9 outputted by the axial error computing unit 30, and outputs the electrical angle estimated angular velocity coe to the current command generator 14, the position estimator 32, and the 1/Pn processing unit 33. [0036] The position estimator 32 estimates the
electrical angle phase 9e and the mechanical angle phase 9m on the basis of the electrical angle estimated angular velocity coe outputted by the PLL controller 31. The position estimator 32 then outputs the estimated electrical angle phase 9e to each of the d-q/u, v, w converter 23 and

the u, v, w/d-q converter 29. The position estimator 32 also outputs the estimated mechanical angle phase 9m to each of the voltage saturation region current command generator 14b of the current command generator 14, the current error correction generator 37, and the correction torque generator 34.
[0037] The 1/Pn processing unit 33 calculates a mechanical angle estimated angular velocity com by dividing the electrical angle estimated angular velocity coe outputted by the PLL controller 31 by the number of pole pairs Pn of the motor 10, and outputs the mechanical angle estimated angular velocity com to each of the subtractor 11 and the correction torque generator 34.
[0038] The subtractor 38 calculates a mechanical angle estimated angular velocity fluctuation Acom by subtracting the mechanical angular velocity command value com* from the mechanical angle estimated angular velocity com outputted by the 1/Pn processing unit 33, and outputs the mechanical angle estimated angular velocity fluctuation Acom thus calculated to the correction torque generator 34.
[0039] The correction torque generator 34 generates, on the basis of a velocity fluctuation tolerance |Acom|% which is stored in the motor control device 100 and is a velocity fluctuation range in which the vibrations of the motor 10 are tolerable, the mechanical angle estimated angular velocity fluctuation Acom outputted by the subtractor 38, and the mechanical angle phase 9m outputted by the position estimator 32, a variable torque command value AT for reducing the mechanical angle estimated angular velocity fluctuation (velocity fluctuation) Acom, which is periodic velocity fluctuation, to be equal to or less than the

velocity fluctuation tolerance | Acom|*. The variable torque command value AT is adjusted in light of reduction in power consumption, prevention of demagnetization of the motor 10, and the like. Note that the mechanical angle estimated angular velocity fluctuation (velocity fluctuation) Acom is different from the value of the angular velocity deviation Aco described above merely in positive and negative signs. [0040] The iir 35a is a filter that receives an input of the d-axis current detection value Id outputted by the u, v, w/d-q converter 29, removes noise, and outputs a d-axis response current Id_iir. The iir 35b is a filter that receives an input of the q-axis current detection value Iq outputted by the u, v, w/d-q converter 29, removes noise, and outputs a q-axis response current Iq_iir. Each iir is an infinite impulse response filter and an example of a noise reduction filter.
[0041] The decoupling controller 36 generates a d-axis decoupling correction value Vda for correcting the temporary d-axis voltage command value Vd_Pi on the basis of an electrical angular velocity command value coe* supplied from the outside of the motor control device 100 (for example, a high-level controller) and the q-axis response current Iq_iir. The decoupling controller 36 also generates a q-axis decoupling correction value Vqa for correcting the temporary q-axis voltage command value Vq_pi on the basis of the electrical angular velocity command value coe* and the d-axis response current Id_iir. The d-axis decoupling correction value Vda and the q-axis decoupling correction value Vqa are correction values for feeding forward, in advance, the interference terms between the dq axes to cancel the interference caused by current control. Here, regarding the calculation of the interference terms, in

order to achieve stable control, the decoupling correction values are desirably direct-current values. For this reason, the calculation is made, for example, in which the
electrical angular velocity command value coe* is used for the velocity and the d-axis response current Id_iir and the q-axis response current Iq_ iir, which are output values of the iir from which the fluctuation component is excluded, are used for the d-axis current detection value Id and the q-axis current detection value Iq.
[0042] The current error correction generator 37 integrates fluctuation errors (phase error and amplitude error) caused in a case where the dq-axis current fails to follow the current command value due to a response delay of the current command generator 14 or the interference of the dq-axis, and generates an inverted output of the integrated value as current error correction values (d-axis current error correction value Aid and q-axis current error correction value Alq) . Here, the d-axis current error correction value Aid is a current feedforward component for correcting a fluctuation error between the d-axis current command value Id* and the d-axis current detection value Id flowing through the motor 10, and the q-axis current error
correction value Alq is a current feedforward component for correcting a fluctuation error between the q-axis current command value Iq* and the q-axis current detection value Iq flowing through the motor 10.
[0043] That is, the current error correction generator 37 generates and outputs the d-axis current error correction value Aid and the q-axis current error correction value Alq on the basis of the d-axis current command value Id* and the q-axis current command value Iq* outputted by the current command generator 14, the d-axis

current detection value Id and the q-axis current detection value Iq outputted by the u, v, w/d-q converter 29, and the mechanical angle phase 9m outputted by the position estimator 32.
[0044] [Current Command Smoothing Processing Unit According to Embodiment]
The description goes on to the current command smoothing processing unit 14c. Usually, since the current command value generated in the normal control region is different from the current command value generated in the voltage saturation region, the current command value rapidly changes in response to the control region switched, which makes the rotational speed control and the torque control of the motor unstable. In order to prevent the control from becoming unstable as described above, a transition period is provided in response to the control region switched, and the smoothing processing is performed in the transition period. The smoothing processing is processing of gradually changing the current command value from the current command value of the control region before switching to the current command value of the control region after switching by simultaneously using the normal control region current command generator 14a and the voltage saturation region current command generator 14b when the control region is switched. FIG. 2 is a diagram for explaining the smoothing processing on the current command value by the current command smoothing processing unit 14c. In FIG. 2, when the control region is switched between the normal control region and the voltage saturation region, Gafter (smoothing gain_after) is a coefficient multiplied by the current command value of the control region after switching, and Gbefore (smoothing gain_before) is a coefficient multiplied by the current

command value of the control region before switching.
[0045] The range of values for Gafter and Gbefore is 0
(zero) to 1, and Gafter + Gbefore = 1. Assuming that a set time (smoothing processing set time) set as a time for the smoothing processing (transition period) is denoted by tSet
(one second, for example) and an elapsed time since the start of the smoothing processing (smoothing processing elapsed time) is denoted by t, Gafter = t/tSet and Gbefore = 1 - t/tset are satisfied. Therefore, the rate of increase in Gafter and the rate of decrease in Gbefore are the same constant rate.
[0046] In a case where the elapsed time t is 0 (zero) at a time when the control region is switched, the current command smoothing processing unit 14c outputs only the current command value before switching. The current command smoothing processing unit 14c then reduces the ratio of the current command value before switching with increasing the elapsed time t, increases the ratio of the current command value after switching, and then outputs the resultant. The current command smoothing processing unit 14c then outputs only the current command value after switching in response to the elapsed time t reaching the set time tSet.
[0047] FIG. 3 is a block diagram illustrating an example of the current command smoothing processing unit 14c. The current command smoothing processing unit 14c includes a smoothing gain_after calculator 14cl (corresponding to a post-gain calculator), post-multipliers 14c2 and 14c3, pre-multipliers 14c5 and 14c6, a smoothing gain_before calculator 14c4 (corresponding to a pre-gain calculator), and adders 14c7 and 14c8.
[0048] The smoothing gain_after calculator 14cl calculates and outputs Gafter within the set time tSet after

the control region is switched. The post-multiplier 14c2 multiplies the Gafter outputted by the smoothing gain_after calculator 14cl by a d-axis current command value after switching, and outputs the resultant as a post-d-axis current command value Id_after*. The post-multiplier 14c3 multiplies the Gafter outputted by the smoothing gain_after calculator 14cl by a q-axis current command value after switching, and outputs the resultant as a post-q-axis current command value Iq_after*.
[0049] Specifically, in a case where switching is performed from the normal control region to the voltage saturation region, the post-multiplier 14c2 outputs Id_after* = Gafter x Id_fw% and the post-multiplier 14c3 outputs Iq_after* = Gafter x Iq_fw*. On the other hand, in a case where switching is performed from the voltage saturation region to the normal control region, the post-multiplier 14c2 outputs Id_atter* = Gafter x Id_ mtpi*, and the post-multiplier
14c3 OUtputS Iq_after * = Gafter X Iq_mtpi*.
[0050] The current command smoothing processing unit 14c uses a switch 14c9, for example, to change the input to the post-multipliers 14c2 and 14c3 to either the output (Id_fw% Iq_fw*) of the voltage saturation region current command generator 14b or the output (Id_mtpi% Iq_mtPi*) of the normal control region current command generator 14a.
[0051] The smoothing gain_before calculator 14c4 calculates and outputs Gbefore within the set time tSet after the control region is switched. The pre-multiplier 14c5 multiplies the Gbefore outputted by the smoothing gain_before calculator 14c4 by a d-axis current command value before switching, and outputs the resultant as a pre-d-axis current command value Idjoefore*. The pre-multiplier 14c6 multiplies the Gbefore outputted by the smoothing gain_before calculator 14c4 by a q-axis current command value before

switching, and outputs the resultant as a pre-q-axis current command value lqjoefore*.
[0052] Specifically, in a case where switching is performed from the normal control region to the voltage saturation region, the pre-multiplier 14c5 outputs Idjoefore* = Gbefore x Id_mtpi% and the pre-multiplier 14c6 outputs Iq_before* = Gbefore x Iq_mtPi*. On the other hand, in a case where switching is performed from the voltage saturation region to the normal control region, the pre-multiplier 14c5 outputs Id_before* = Gbefore x Id_ fw*, and the pre-
multiplier 14c6 OUtputS lqjoefore* = Gbefore X Iq_fw* .
[0053] The current command smoothing processing unit 14c uses the switch 14c9, for example, to change the input to the pre-multipliers 14c5 and 14c6 to either the output
(Id_mtpi% Iq_mtPi*) of the normal control region current command generator 14a or the output (Id_fw% Iq_fw*) of the voltage saturation region current command generator 14b.
[0054] The adder 14c7 adds the output of the post-multiplier 14c2 and the output of the pre-multiplier 14c5 and outputs the d-axis current command value Id* that has been subjected to the smoothing processing. That is, the adder 14c7 outputs the d-axis current command value Id* = Id_after* + Idjoefore*. The adder 14c8 adds the output of the post-multiplier 14c3 and the output of the pre-multiplier 14c6 and outputs the q-axis current command value Iq* that has been subjected to the smoothing processing. That is, the adder 14c8 outputs the q-axis current command value Iq*
— -Lq_after + -Lq_before •
[0055] [Current Command Value during Smoothing Processing]
FIG. 4 is diagrams illustrating a current vector while smoothing processing is performed. In FIG. 4, Vo is an induced voltage, Ya is an interlinkage magnetic flux of the

motor 10, Ld is a d-axis inductance of the motor 10, and M is the center of a constant induced voltage ellipse. FIG. 4 illustrates a case where the control region is switched from the normal control region to the voltage saturation region. FIG. 4(a) illustrates a case where 0% of the set time tset has elapsed, FIG. 4(b) illustrates a case where 50% of the set time tset has elapsed, and FIG. 4 (c) illustrates a case where 100% of the set time tset has elapsed.
[0056] As illustrated in FIG. 4(a), in the case where 0% of the set time tset has elapsed, the current command value is within the normal control region, and thus the current command value periodically varies on the maximum torque/current (MTPI) curve according to the fluctuations in torque. The d-axis current command value Id* decreases in a case where the q-axis current command value Iq* increases, and the d-axis current command value Id* increases in a case where the q-axis current command value Iq* decreases. Further, the range of fluctuations in the q-axis current command value Iq* is larger than the range of fluctuations in the d-axis current command value Id*.
[0057] As illustrated in FIG. 4(b), in the case where 50% of the set time tset has elapsed, the current command value is averaged between the current command value at the intersection of the maximum torque/current (MTPI) curve in the normal control region and the constant torque curve of the total torque command value T* and the current command value at the intersection of the constant induced voltage ellipse in the voltage saturation region and the constant torque curve of the total torque command value T* at the intermediate point where the control region is switched from the normal control region to the voltage saturation region, and is used as the d-axis current command value Id*

and the q-axis current command value Iq* corresponding to this point. This point periodically varies according to the fluctuations in torque. At this time, as compared with the normal control region of FIG. 4(a), the range of fluctuations in the d-axis current command value Id* is large, and the range of fluctuations in the q-axis current command value Iq* is small. Further, as compared with the voltage saturation region of FIG. 4(c) described later, the range of fluctuations in the d-axis current command value Id* is small, and the range of fluctuations in the q-axis current command value Iq* is large.
[0058] Further, as illustrated in FIG. 4(c), in the case where 100% of the set time tSet has elapsed, the current command value is within the voltage saturation region, and thus the current command value periodically varies on the constant induced voltage ellipse according to the fluctuations in torque. The d-axis current command value Id* decreases in a case where the q-axis current command value Iq* increases, and the d-axis current command value Id* increases in a case where the q-axis current command value Iq* decreases. However, unlike the normal control region, the range of fluctuations in the q-axis current command value Iq* is smaller than the range of fluctuations in the d-axis current command value Id*.
[0059] FIG. 5 is a diagram illustrating transition of a current command value while smoothing processing is performed. In FIG. 5, the horizontal axis represents a time (s: second), and the vertical axis represents a current (A: ampere) . The set time tset is one second. The upper part of the diagram illustrates the q-axis current, and the lower part thereof illustrates the d-axis current. FIG. 5 illustrates a case where the control region is switched from the normal control region to the voltage

saturation region.
[0060] As illustrated in FIG. 5, the d-axis current command value Id* and the q-axis current command value Iq* are always in opposite phase. Further, in the case where 0% of the set time tset has elapsed (at 0 (zero) on the horizontal axis), the range of fluctuations in the Iq_mtpi* is larger than the range of fluctuations in the Id_mtPi*, so that the range of fluctuations in the Iq* is larger than the range of fluctuations in the Id*. This is consistent with FIG. 4 (a). Further, in the case where 100% of the set time tset has elapsed (at 1 on the horizontal axis), the range of fluctuations in the Iq_fW* is smaller than the range of fluctuations in the Id_tw*, so that the range of fluctuations in the Iq* is smaller than the range of fluctuations in the Id*. This is consistent with FIG. 4
(c) .
[0061] As described above, in the motor control device 100 according to this embodiment, the current command smoothing processing unit 14c uses the current command value before switching immediately after the control region is switched, and reduces, over time, the current command value before switching by using Gbetore. At the same time, in the motor control device 100 according to this embodiment, the current command smoothing processing unit 14c increases the current command value after switching by using Gatter and, after the set time has elapsed, uses the current command value after switching completely. As a result, in a case where the control region is switched, the motor control device 100 according to this embodiment can realize smooth state transition by gradually reducing the current command value of the control region before switching while performing the torque control, and at the same time, by gradually increasing the current command

value of the control region after switching, which prevents instability of the torque control when the control region is switched.
Reference Signs List [0062] 10 MOTOR
11, 18, 19 SUBTRACTOR
12 VELOCITY CONTROLLER
13, 16, 17, 21, 22 ADDER
14 CURRENT COMMAND GENERATOR
14a NORMAL CONTROL REGION CURRENT COMMAND GENERATOR (FIRST CURRENT COMMAND GENERATOR)
14b VOLTAGE SATURATION REGION CURRENT COMMAND GENERATOR (SECOND CURRENT COMMAND GENERATOR)
14c CURRENT COMMAND SMOOTHING PROCESSING UNIT (THIRD CURRENT COMMAND GENERATOR)
14cl SMOOTHING GAIN_after CALCULATOR (POST-GAIN CALCULATOR)
14c2, 14c3 POST-MULTIPLIER
14c5, 14c6 PRE-MULTIPLIER
14c4 SMOOTHING GAINJoefore CALCULATOR (PRE-GAIN CALCULATOR)
14c7, 14c8 ADDER
15 CONTROL SWITCHING DETERMINATION UNIT
2 0 VOLTAGE COMMAND GENERATOR
23 d-q/u, v, w CONVERTER
2 4 PWM MODULATOR
25 IPM
2 6 SHUNT RESISTOR
2 7a, 2 7b CURRENT SENSOR
2 8 3(|) CURRENT CALCULATOR
2 9 u, v, w/d-q CONVERTER
3 0 AXIAL ERROR COMPUTING UNIT

4 PLL CONTROLLER
5 POSITION ESTIMATOR
6 1/Pn PROCESSING UNIT
7 CORRECTION TORQUE GENERATOR 35a, 35b iir
3 6 DECOUPLING CONTROLLER
37 CURRENT ERROR CORRECTION GENERATOR
100 MOTOR CONTROL DEVICE

CLAIMS
1. A motor control device comprising a current command
generator that generates a motor current command value on a
basis of a velocity command value and a velocity of a
motor, wherein
the current command generator includes
a first current command generator that generates a current command value in a normal control region, and
a second current command generator that generates a current command value in a voltage saturation region, and
in a case where a control region is switched from the normal control region to the voltage saturation region, or, from the voltage saturation region to the normal control region, the current command generator uses the first current command generator and the second current command generator together to generate the motor current command value.
2. The motor control device according to claim 1, wherein
the current command generator includes a third current
command generator that generates the motor current command
value on a basis of the current command values outputted by
each of the first current command generator and the second
current command generator.
3. The motor control device according to claim 2, wherein
the third current command generator
uses the current command value generated by the first current command generator and the current command value generated by the second current command generator to calculate a pre-current command value based on a current command value of a control region before switching and a post-current command value based on a current command value

of a control region after switching, and uses the pre-current command value and the post-current command value calculated to generate the motor current command value.
4. The motor control device according to claim 3, wherein
the third current command generator includes
a pre-gain calculator that calculates a pre-gain for determining a ratio of the current command value of the control region before switching among current command values used in a transition period in a case where the control region is switched,
a pre-multiplier that multiplies the pre-gain calculated by the pre-gain calculator by the current command value of the control region before switching and outputs the pre-current command value,
a post-gain calculator that calculates a post-gain for determining a ratio of the current command value of the control region after switching among the current command values used in the transition period in a case where the control region is switched,
a post-multiplier that multiplies the post-gain calculated by the post-gain calculator by the current command value of the control region after switching and outputs the post-current command value, and
an adder that adds the pre-current command value outputted by the pre-multiplier and the post-current command value outputted by the post-multiplier to generate the motor current command value.
5. The motor control device according to claim 4, wherein
the pre-gain calculator reduces the pre-gain from 1 to
0 (zero) at a constant rate within a predetermined set time after switching the control region starts, and

the post-gain calculator increases the post-gain from 0 (zero) to 1 at the constant rate within the set time after switching the control region starts.