INVERTER CONTROL METHOD AND MOTOR CONTROL DEVICE

Abstract

An inverter control method is a method for controlling an inverter that outputs an application voltage, which is a voltage to be applied to a motor that drives a load by using rotation of a shaft. The method includes: causing the inverter to output the application voltage having an amplitude smaller than a first maximum and causing the motor to rotate at a first speed and drive the load which is predetermined; and causing the inverter to output the application voltage having an amplitude of a second maximum and causing the motor to rotate at a second speed and drive the predetermined load. The first maximum is a possible maximum value of an amplitude of the application voltage when the motor drives the predetermined load at the first speed. The first speed is a maximum of a speed of rotation of the motor when the motor drives the predetermined load. The second maximum is a possible maximum value of the amplitude of the application voltage when the motor drives the predetermined load at the second speed. The second speed is lower than the first speed.

Description

TECHNICAL FIELD

The present disclosure relates to a technique for controlling an inverter and a technique for controlling a motor.

BACKGROUND ART

Japanese Patent Application Laid-Open No. 2015-211561 proposes a technique for reducing vibration attributed to a temporal second-order component of radial electromagnetic force of a three-phase synchronous motor. The “temporal second-order component of radial electromagnetic force” is explained as a radial electromagnetic force that is two times as much as the fundamental frequency of current flowing in the motor. It is also explained that, by vibration attributed to a temporal second-order component of radial electromagnetic force, deformation into an elliptical form or a deformation mode into a square form occurs in the three-phase synchronous motor. In this technique, for a motor in which d-axis inductance and q-axis inductance differ, a q-axis current is increased, and also, a negative d-axis current is increased.

SUMMARY OF INVENTION

Problem to be Solved by the Invention

Japanese Patent Application Laid-Open No. 2015-211561 focuses on the temporal second-order component of radial electromagnetic force. Thus, there is no explicit mention of a so-called “uneven contact”, which is an event in which a radial stress given from a motor shaft to a bearing that rotatably supports the motor shaft becomes strong at a specific rotation angle. The uneven contact is a factor for damaging the bearing. The present disclosure provides a technique for reducing the radial stress generated by the shaft when the motor rotates.

Means to Solve the Problem

An inverter control method of this disclosure is a method for controlling an inverter (210b) that outputs an application voltage (Vs), which is a voltage to be applied to a motor (1) that drives a load (20) by using rotation of a shaft (10).

According to a first aspect of the inverter control method, the inverter (210b) is caused to output the application voltage (Vs) having an amplitude smaller than a first maximum (V_max_ω_MAX) and the motor is caused to rotate at a first speed (ω_MAX) and drive the load which is predetermined. The inverter (210b) is caused to output the application voltage (Vs) having an amplitude of a second maximum (V_max_ω3) and the motor is caused to rotate at a second speed (ω3) and drive the predetermined load. The second speed (ω3) is lower than the first speed (ω_MAX).

The first maximum (V_max_ω_MAX) is a possible maximum value of an amplitude (|Vs|) of the application voltage when the motor drives the predetermined load at the first speed (ω_MAX). The first speed (ω_MAX) is a maximum of a speed (ω_m) of rotation of the motor when the motor drives the predetermined load.

The second maximum (V_max_ω_MAX) is a possible maximum value of the amplitude (|Vs|) of the application voltage when the motor drives the predetermined load at the second speed (ω3).

According to a second aspect of the inverter control method, in a case in which a speed (ω_m) of rotation of the motor when the motor outputs a predetermined torque is higher than or equal to a base speed (ωb) of the motor when the motor outputs the predetermined torque, the inverter (210b) is caused to output the application voltage (Vs) having an amplitude obtained by multiplying a first maximum (V_max_ω1) by a first ratio, the motor is caused to rotate at a first speed (ω1), and the motor is caused to output the predetermined torque. The inverter (210b) is caused to output the application voltage (Vs) having an amplitude obtained by multiplying a second maximum (V_max_ω2) by a second ratio, the motor is caused to rotate at a second speed (ω2), and the motor is caused to output the predetermined torque.

The first maximum (V_max_ω1) is a possible maximum value of an amplitude (|Vs|) of the application voltage when the motor outputs the predetermined torque at the first speed (ω1). The second maximum (V_max_ω2) is a possible maximum value of the amplitude (|Vs|) of the application voltage when the motor outputs the predetermined torque at the second speed (ω2).

The second speed (ω2) is higher than the first speed (ω1), and the second ratio is smaller than the first ratio.

According to a third aspect of the inverter control method of this disclosure, in the second aspect thereof, in which the second speed is a possible maximum (ω_max) of the speed (ω_m) when the motor outputs the predetermined torque.

A motor control device of this disclosure includes: an inverter (210b) that outputs an application voltage (Vs), which is a voltage to be applied to a motor (1) that drives a load (20) by using rotation of a shaft (10); and a controller (209) that controls operation of the inverter.

According to a first aspect of the motor control device, the controller causes the inverter (210b) to output the application voltage (Vs) having an amplitude smaller than a first maximum (V_max_ω_MAX) and causes the motor to rotate at a first speed (ω_MAX) and drive the load which is predetermined, the controller causes the inverter (210b) to output the application voltage (Vs) having an amplitude of a second maximum (V_max_ω3) and causes the motor to rotate at a second speed (ω3) and drive the predetermined load. The second speed (ω3) is lower than the first speed (ω_MAX).

The first maximum (V_max_ω_MAX) is a possible maximum value of an amplitude (|Vs|) of the application voltage when the motor drives the predetermined load at the first speed (ω_MAX), and the first speed (ω_MAX) is a maximum of a speed (ω_m) of rotation of the motor when the motor drives the predetermined load.

The second maximum (V_max_ω3) is a possible maximum value of the amplitude (|Vs|) of the application voltage when the motor drives the predetermined load at the second speed (ω3).

According to a second aspect of the motor control device, in a case in which a speed (ω_m) of rotation of the motor when the motor outputs a predetermined torque is higher than or equal to a base speed (ωb) of the motor when the motor outputs the predetermined torque, the controller causes the inverter (210b) to output the application voltage (Vs) having an amplitude obtained by multiplying a first maximum (V_max_ω1) by a first ratio, causes the motor to rotate at a first speed (ω1), and causes the motor to output the predetermined torque, and causes the inverter (210b) to output the application voltage (Vs) having an amplitude obtained by multiplying a second maximum (V_max_ω2) by a second ratio, causes the motor to rotate at a second speed (ω2), and causes the motor to output the predetermined torque.

The first maximum (V_max_ω1) is a possible maximum value of an amplitude (|Vs|) of the application voltage when the motor outputs the predetermined torque at the first speed (ω1), and the second maximum (V_max_ω2) is a possible maximum value of the amplitude (|Vs|) of the application voltage when the motor outputs the predetermined torque at the second speed (ω2).

The second speed (ω2) is higher than the first speed (ω1), and the second ratio is smaller than the first ratio.

According to a third aspect of the motor control device of this disclosure, in the second aspect thereof, the second speed is a possible maximum (ω_max) of the speed (ω_m) when the motor outputs the predetermined torque.

The inverter control method and the motor control device according to the present disclosure reduces the radial stress given from the shaft to the bearing when the motor rotates.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a sectional view illustrating an example of a structure of a compressor:

FIG. 2 is a block diagram illustrating a motor and a configuration of a motor control apparatus that drives the motor;

FIG. 3 shows graphs illustrating relationships between control employed in an embodiment and a rotation speed as solid lines;

FIG. 4 is a flowchart illustrating control of an output circuit by a controller;

FIG. 5 is a graph illustrating a relationship between an axial deviation and an amplitude of an application voltage in which the rotation speed is a parameter;

FIG. 6 is a graph illustrating a relationship between a current amplitude and the axial deviation in which the rotation speed is a parameter;

FIG. 7 is a graph illustrating a relationship between the rotation speed and the current amplitude when a torque is a predetermined value;

FIG. 8 is a graph illustrating a relationship between a phase of a current vector and the axial deviation in which the rotation speed is a parameter:

FIG. 9 is a graph illustrating a relationship between the rotation speed and the phase when the torque is a predetermined value;

FIG. 10 is a graph illustrating a relationship between a d-axis current and the axial deviation in which the rotation speed is a parameter;

FIG. 11 is a graph illustrating a relationship between a q-axis current and the axial deviation δ_C in which the rotation speed is a parameter;

FIG. 12 is a graph illustrating a relationship between the rotation speed and the q-axis current when the torque is a predetermined value;

FIG. 13 is a vector diagram illustrating a relationship between a field magnetic flux vector, a magnetic flux vector attributed to an armature reaction, and a primary magnetic flux vector;

FIG. 14 is a graph illustrating a relationship between a T-axis current and the axial deviation in which the rotation speed is a parameter;

FIG. 15 is a graph illustrating a relationship between the rotation speed and the T-axis current when the torque is a predetermined value;

FIG. 16 is a graph illustrating a relationship between a primary magnetic flux and the axial deviation in which the rotation speed is a parameter:

FIG. 17 is a graph illustrating a relationship between the rotation speed and the primary magnetic flux when the torque is a predetermined value;

FIG. 18 is a graph illustrating a relationship between a load angle and the axial deviation in which the rotation speed is a parameter;

FIG. 19 is a graph illustrating a relationship between the rotation speed and the load angle when the torque is a predetermined value;

FIG. 20 is a graph illustrating a relationship between an instantaneous real power and the axial deviation in which the rotation speed is a parameter;

FIG. 21 is a graph illustrating a relationship between the rotation speed and the instantaneous real power when the torque is a predetermined value;

FIG. 22 is a block diagram illustrating a first modification of the controller; and

FIG. 23 is a block diagram illustrating a second modification of the controller.

DESCRIPTION OF EMBODIMENTS

FIG. 1 is a sectional view illustrating an example of a structure of a compressor 100 employed for a refrigeration circuit, such as a heat pump. The compressor 100 includes a compression mechanism 20, a motor 1, a bearing 14, and a casing 15. The compression mechanism 20 compresses refrigerant (omitted from illustration). For example, a swing type is employed for the compression mechanism 20, and refrigerant is compressed by rotation transferred from the motor 1 via a shaft 10. The compression mechanism 20 is a load driven by the motor 1.

The motor 1 includes a stator 11 and a rotor 12. For example, the stator 11 and the rotor 12 are an armature and a field element, respectively. For example, the motor 1 is an inner-rotor type interior magnet synchronous motor, and the rotor 12 has a permanent magnet (omitted from illustration) that generates a field magnetic flux.

The shaft 10 is attached to the rotor 12, and is rotatably attached to the casing 15 by the bearing 14.

A balance weight 13a is provided on the compression mechanism 20 side of the rotor 12 in the direction of the shaft 10 (hereinafter “axial direction”). A balance weight 13c is provided on the opposite side to the compression mechanism 20 of the rotor 12 in the axial direction. For convenience of description of the structure, above the sectional view in FIG. 1, a top view of the rotor 12 (view of the rotor 12 seen from the opposite side to the compression mechanism 20 in the axial direction) is illustrated by being combined with the section of the rotor 12 by four imaginary lines, which are chain lines.

Rotation of the rotor 12 (hereinafter also referred to as rotation of the motor 1) causes centrifugal forces F_A and F_C to act on the balance weights 13a and 13c, respectively. Unbalanced magnetic pull F_B acts on the shaft 10. The unbalanced magnetic pull F_B is a component in a radial direction. i.e., a component in the direction orthogonal to the axial direction, attributed to an imbalance in magnetic pull between the stator 11 and the rotor 12. Only this component is focused herein because a deflection amount (hereinafter referred to as “axial deviation”) generated by the centrifugal forces F_A and F_C acting in the radial direction and also a stress applied in the radial direction on the shaft 10 is studied. For convenience, the unbalanced magnetic pull F_B is illustrated as acting at a position B of the shaft 10 in the center of the rotor 12 in the axial direction.

As the speed of rotation (hereinafter also referred to as “rotation speed”) of the motor 1 is higher, the centrifugal forces F_A and F_C are larger. As the rotation speed is higher, the axial deviation is larger. The axial deviation is a factor of a so-called uneven contact, which is an event in which a radial stress given from the shaft 10 to the bearing 14 becomes strong at a specific rotation angle.

In order to enhance the performance of the refrigeration circuit, the rotation speed is desirably high. In other words, a small axial deviation is advantageous in enhancing the performance of the refrigeration circuit.

In the following embodiment, a motor driving technique for reducing the axial deviation is introduced. FIG. 2 is a block diagram illustrating the motor 1 and a configuration of a motor control device 200 that drives the motor 1. Herein, an example of a case in which the motor 1 is a three-phase interior magnet synchronous motor (denoted as IPMSM in the figure) is illustrated. The motor control device 200 converts three-phase alternating currents Iu, Iv, and Iw flowing in the motor 1 into a d-axis component (hereinafter “d-axis current”) id, a q-axis component (hereinafter “q-axis current”) i_q and performs vector control. The “d-axis” and “q-axis” herein respectively indicate coordinate axe that is in the same phase as the field magnetic flux of the motor 1 and coordinate axe that advances 90 degrees with respect to the field magnetic flux. The d-axis current is contributes to the field magnetic flux, and the q-axis current i_q contributes to a torque output from the motor 1.

The motor control device 200 includes an output circuit 210 and a controller 209 that controls operation of the output circuit 210. The output circuit 210 outputs, to the motor 1, an application voltage Vs to be applied to the motor 1. The motor 1 is, for example, driven with the rotation speed controlled by the application voltage Vs. For example, the output circuit 210 performs DC/AC conversion on a DC voltage Vdc and outputs the three-phase application voltage Vs to the motor 1. The output circuit 210 supplies the three-phase alternating currents Iu, Iv, and Iw to the motor 1.

The output circuit 210 includes a pulse-width modulation circuit (displayed as “PWM circuit” in the figure) 210a and a voltage control type PWM inverter 210b. The pulse-width modulation circuit 210a receives three-phase voltage command values v_u*, v_v*, and v_w* and generates agate signal G for controlling operation of the PWM inverter 210b. Note that an inverter of other modulation type may also be employed instead of the PWM inverter 210b.

The DC voltage Vdc is supplied to the PWM inverter 210b from a DC power source. The PWM inverter 210b performs operation controlled by the gate signal G, converts the DC voltage Vdc into the application voltage Vs, and applies it to the motor 1. The three-phase alternating currents Iu, Iv, and Iw are supplied from the PWM inverter 210b to the motor 1. The voltage command values v_u*, v_v*, and v_w* are command values of the application voltage Vs.

Although the power source that supplies the DC voltage Vdc is provided outside the motor control device 200 in FIG. 2, the power source may alternatively be included in the motor control device 200. The power source can be realized, for example, using an AC/DC converter.

The controller 209 includes, for example, a current command generating unit 211, a current controller 212, coordinate converters 213 and 214, a position sensor 215, a multiplier 216, and a speed calculator 217.

Current sensors 218u and 218v sense the alternating currents Iu and Iv, respectively. The controller 20) may alternatively include the current sensors 218u and 218v. The position sensor 215 senses the rotation position of the motor 1 as a rotation angle θ_m which is a mechanical angle of the motor 1. The multiplier 216 multiplies the rotation angle θ_m by a number of pole pairs P_n to obtain a rotation angle θ as an electric angle. The coordinate converter 214 receives the values of the alternating currents Iu and Iv and the rotation angle θ and obtains the d-axis current i_d and the q-axis current i_q.

The speed calculator 217 obtains, from the rotation angle θ_m, a rotation speed ωm based on a mechanical angle. The current command generating unit 211 receives a torque command τ* or receives the rotation speed ω_m and its command value ω_m*, and obtains, from these, a command value i_d* of the d-axis current is and a command value i_q* of the q-axis current i_q. The torque command τ* is a command value of a torque T output from the motor 1.

From the d-axis current i_d and its command value i_d* and the q-axis current i_q and its command value i_q*, the current controller 212 obtains a command value v_d* of a d-axis voltage v_d and a command value v_q* of a q-axis voltage v_q. For example, the command values v_d* and v_q* can be obtained by feedback control for making the deviation between the d-axis current is and its command value i_d* and the deviation between the q-axis current i_q and its command value i_q* close to zero.

From the command value v_d* of the d-axis voltage v_d, the command value v_q* of the q-axis voltage v_q, and the rotation angle θ, the coordinate converter 213 generates the three-phase voltage command values v_u*, v_v*, and v_w*.

In this embodiment, the position sensor 215 is not necessarily provided. It is also possible to employ a so-called sensorless type in which the rotation angle θ_m is obtained from the alternating currents Iu and Iv and the application voltage Vs.

FIG. 3 shows graphs illustrating relationships between control employed in this embodiment and the rotation speed ω_m as solid lines. Each of the upper graph, the middle graph, and the lower graph in FIG. 3 employs the rotation speed ω_m on the horizontal axis, and the torque command τ* is fixed to a certain value.

In FIG. 3, the upper graph employs an amplitude |Vs| of the application voltage Vs on the vertical axis, the middle graph employs an axial deviation δ_C on the vertical axis, and the lower graph employs the d-axis current is on the vertical axis. Herein, the axial deviation δ_C is an axial deviation at a position C (FIG. 1) on the end of the shaft 10 on the balance weight 13c side in the axial direction.

When the rotation speed ω_m is lower than or equal to a rotation speed v1 (also simply referred to as “speed v1”: the same applies to other rotation speeds), as the rotation speed ω_m is higher, the amplitude |Vs| is larger. For example, as such control, maximum torque/current control or maximum efficiency control can be employed. FIG. 3 illustrates an example of a case in which the maximum torque/current control is performed when the rotation speed ω_m is lower than or equal to the speed v1. In addition, the amplitude |Vs| when the rotation speed ω_m is at the speed v1 is illustrated as a voltage value Vmax.

When the rotation speed ω_m is higher than a speed v2, the amplitude |Vs| is less than the voltage value Vmax. The speed v2 is higher than or equal to the speed v1. Such control is provisionally called “voltage reduction control” for convenience in this embodiment. As its example, the upper graph in FIG. 3 illustrates a case in which v2>v1 and the amplitude |Vs| is smaller as the rotation speed (Om is higher.

When the rotation speed ω_m is higher than the speed v1 and lower than or equal to the speed v2, the amplitude |Vs| is equal to the amplitude |Vs| (=Vmax) at the speed v1 regardless of the rotation speed ω_m. At this time, so-called flux-weakening control is performed on the motor 1. When v1=v2, a phenomenon in which the rotation speed ω_m is higher than the speed v1 and lower than or equal to the speed v2 does not occur, and the flux-weakening control is not performed.

For dependency of the application voltage Vs on the rotation speed ω_m, the controller 209 causes the output circuit 210 to output the application voltage Vs. Specifically, the controller 209 generates the voltage command values v_u*, v_v*, and v_w* by which the output circuit 210 outputs the application voltage Vs in accordance with the rotation speed ω_m, and outputs the voltage command values to the output circuit 210.

FIG. 4 is a flowchart illustrating control of the output circuit 210 by the controller 209. This flowchart is a routine for controlling the application voltage Vs. This routine is, for example, interrupt processing for a main routine that is not illustrated, which starts by interrupt processing and returns to the main routine upon ending of the routine. The routine is, for example, performed together with the main routine by the controller 209.

In step S401, the rotation speed ω_m is compared with the speed v1 and the speed v2. If it is determined in step S401 that ω_m≤v1, the process proceeds to step S402. If it is determined in step S401 that v1<ω_m≤v2, the process proceeds to step S403. If it is determined in step S401 that v2<ω_m, the process proceeds to step S404.

In step S402, the maximum torque/current control is performed. Alternatively, instead of the maximum torque/current control, in step S402, the maximum efficiency control may be performed. Alternatively, in step S402, the maximum torque/current control and the maximum efficiency control may be performed by being switched therebetween.

In step S403, the voltage value Vmax is employed as the amplitude |Vs|, and, for example, the flux-weakening control is performed. In step S404, the voltage reduction control is performed, and a value less than the voltage value Vmax is employed as the amplitude |Vs|.

In FIG. 3, for comparison with this embodiment, the broken line represents a case in which the flux-weakening control is maintained without employing the “voltage reduction control” even if the rotation speed ω_m is higher than the speed v2. In any case in which any of the maximum torque/current control, the maximum efficiency control, and the flux-weakening control is employed, as the rotation speed ω_m is higher, the axial deviation δ_C is larger.

FIG. 3 illustrates an upper limit value δ_Co of the axial deviation δc. The speed v2 at which the axial deviation δ_C becomes the upper limit value δ_Co by the maximum torque/current control, the maximum efficiency control, or the flux-weakening control is actually measured or calculated in advance. Herein, an example of a case is illustrated in which, even if the rotation speed ω_m is increased to exceed the speed v1 and the control is switched from the maximum torque/current control to the flux-weakening control, the axial deviation δ_C is less than the upper limit value δ_Co when the rotation speed ω_m is lower than the speed v2. That is, an example of a case is illustrated in which, when the rotation speed ω_m is lower than or equal to the speed v2, even if the amplitude |Vs| is maintained at the voltage value Vmax, the axial deviation δ_C is less than the upper limit value δ_Co.

When the rotation speed ω_m exceeds the speed v2, the amplitude |Vs| becomes a value less than the voltage value Vmax. Thus, even if the rotation speed ω_m is high, the axial deviation δ_C can be suppressed to be less than or equal to the upper limit value δ_Co.

For example, the voltage value Vmax is the maximum value of an AC voltage into which the PWM inverter 210b can convert the DC voltage Vdc. Since the maximum torque/current control is employed herein, the speed v1 at which the amplitude |Vs| becomes the voltage value Vmax corresponds with abase speed. The base speed herein is the maximum value of the rotation speed of the motor 1 at which the motor 1 can generate the torque τ by the maximum torque/current control. In a case in which the maximum efficiency control is employed, the speed v1 is higher than the base speed.

FIG. 5 is a graph illustrating a relationship between the axial deviation δ_C and the amplitude |Vs| in which the rotation speed ω_m is a parameter. FIGS. 3 and 5 illustrate cases in which the same torque command τ* is used. Hereinafter, reasons why the axial deviation δ_C can be suppressed to be less than or equal to the upper limit value δ_Co by the voltage reduction control will be described with reference to FIG. 5.

FIG. 5 illustrates a relationship between the axial deviation δ_C and the amplitude |Vs| when the rotation speed ω_m becomes a speed v1, v5, v6, or v7 (where v1<v2<v5<v6<v7). When the torque τ is maintained, as the amplitude |Vs| for achieving the rotation speed ω_m is larger, the axial deviation δ_C is larger. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 5, when the rotation speed ω_m becomes a speed v3, v4, or v2 (where v3<v4<v1<v2), the value of the axial deviation δ_C with the amplitude |Vs| employed during the maximum torque/current control and the flux-weakening control is additionally plotted. The thick line in FIG. 5 indicates that, in the direction of arrowheads attached thereto, the amplitude |Vs| employed in this embodiment changes in accordance with increase in the rotation speed ω_m.

In accordance with increase in the rotation speed ω_m to the speeds v3, v4, and v1, the amplitude |Vs| and the axial deviation δ_C increase. When the rotation speed ω_m reaches the speed v1, the amplitude |Vs| reaches the voltage value Vmax. Thus, even if the rotation speed ω_m is more increased, the amplitude |Vs| is no more increased.

Until the rotation speed ω_m reaches the speed v2, the amplitude |Vs| is maintained at the voltage value Vmax (the thick-line arrow in FIG. 5 directs upward from bottom in parallel to the vertical axis). In this case, the flux-weakening control is performed, and the axial deviation δ_C increases.

When the rotation speed ω_m reaches the speed v2, the axial deviation δ_C reaches the upper limit value δ_Co, and when the rotation speed ω_m exceeds the speed v2, the voltage reduction control is performed. Thus, even if the rotation speed ω_m is high, the axial deviation δ_C is maintained at the upper limit value δ_Co.

It is needless to say that the axial deviation δ_C is not necessarily maintained at the upper limit value δ_Co even if the amplitude |Vs| is decreased. However, if the amplitude |Vs| is decreased to be less than the voltage value Vmax, the axial deviation δ_C is more reduced than that in a case in which the amplitude |Vs| is maintained at the voltage value Vmax. As for the middle graph in FIG. 3, when the voltage reduction control is employed, the solid-line curve is always below the broken-line curve. In other words, the radial stress at a specific rotation angle when the motor 1 is rotating is reduced. This contributes to reduction of the uneven contact of the shaft 10 to the bearing 14.

As described above, the axial deviation δ_C may become less than the upper limit value δ_Co by decrease in the amplitude |Vs|. For example, the amplitude |Vs| in the voltage reduction control can be a fixed value that is lower than the voltage value represented by the solid line in the upper graph in FIG. 3.

In the lower graph in FIG. 3, also in the voltage reduction control as in the flux-weakening control, the d-axis current is decreases (since the d-axis current i_d is a negative value, the absolute value thereof increases). Note that the inclination of decrease in the d-axis current is with respect to increase in the rotation speed ω_m is more obvious in the voltage reduction control than in the flux-weakening control.

Note that in the voltage reduction control, unlike in the simple flux-weakening control, the amplitude |Vs| becomes a value lower than the maximum thereof.

Hereinafter, the d-axis current i_d for making the axial deviation δ_C less than or equal to the upper limit value δ_Co will be described by using expressions.

Name                             Symbol
d-axis permeance coefficient     Pd0, Pd1
q-axis permeance coefficient     Pq0, Pq1
d-axis gap permeance per unit area Pgd = Pd0 + 2Pd1cos(2Pnθrm)
q-axis gap permeance per unit area Pgq = Pq0 + 2Pq1cos(2Pnθrm)
permanent magnet magnetomotive   FM
force constant
armature current magnetomotive   FD
force constant
permanent magnet magnetomotive   fM = FMcos(Pnθrm)
force
armature current magnetomotive   fD = FD(idcos[Pnθrm] + iqsin[Pnθrm])
force
number of pole pairs             Pn
arbitrary phase on rotor (based on θrm
d-axis, mechanical angle)
d-axis current                   id
q-axis current                   iq
air permeability                 μ0
offset amount of shaft 10        x
average gap length between       g
stator 11 and rotor 12
unbalanced magnetic pull         FB
centrifugal force acting on balance FA, FC
weights 13a and 13c
axial deflection (axial stress) at δC
point C
axial stress predetermined value δCo
constant determined by material  kA, kB, kC
physical property and shape of
shaft
mass of balance weights 13a and  mA, mC
13c
center of gravity (rotation center rA, rC
basis) of balance weights 13a and
13c
mechanical angular speed         ωm

The axial deviation δ_C can be expressed as Expression (1) based on an elasticity equation of beam deflection.

δ_C=k_AF_A+k_BF_B+k_CF_C  (1)

As an armature winding included in the armature of the motor 1, a case in which a plurality of coils are connected in series for each phase is employed as an example. In this case, the unbalanced magnetic pull F_B is expressed as Expression (2).

$\begin{array}{cc}{F}_B=\frac{x}{g}·\frac{\pi }{2{\mu }_0}·\left\{\begin{array}{c}{\left(F_D{i}_d+F_M\right)}^2\left[{\left(P_{d0}+P_{d1}\right)}^2+P_{d1}^2+\right]\\ {\left(F_D{i}_q\right)}^2\left[{\left(P_{q0}-P_{q1}\right)}^2+P_{q1}^2\right]\end{array}\right\}={\mathrm{ai}}_d^2+{\mathrm{bi}}_d+c& \left(2\right)\\ \mathrm{where}\{\begin{array}{c}a=\frac{x}{g}·\frac{\pi }{2{\mu }_0}·\left(F_D^2\right)\left[{\left(P_{d0}+P_{d1}\right)}^2+P_{d1}^2\right]\\ b=\frac{x}{g}·\frac{\pi }{2{\mu }_0}·\left(2F_D{F}_M\right)\left[{\left(P_{d0}+P_{d1}\right)}^2+P_{d1}^2\right]\\ c=\frac{x}{g}·\frac{\pi }{2{\mu }_0}·\left\{\begin{array}{c}+\left(F_M^2\right)\left[{\left(P_{d0}+P_{d1}\right)}^2+P_{d1}^2\right]\\ +{\left(F_D{i}_q\right)}^2\left[{\left(P_{q0}-P_{q1}\right)}^2+P_{q1}^2\right]\end{array}\right\}\end{array}& \end{array}$

The centrifugal forces F_A and F_C are expressed as Expression (3), and Expression (4) is derived from Expressions (1), (2), and (3).

$\begin{array}{cc}{F}_A=m_A{r}_A{\omega }_m^2,F_C=m_C{r}_C{\omega }_m^2& \left(3\right)\\ {\omega }_m^2=-\frac{k_B{\delta }_c}{k_A{m}_A{r}_A+k_C{m}_C{r}_C}\left({\mathrm{ai}}_d^2+{\mathrm{bi}}_d+c\right)& \left(4\right)\end{array}$

In a case in which the q-axis current i_q is fixed, not only values a and b, but also a value c is fixed. Thus, from a relationship illustrated in Expression (5) obtained by setting δ_C=δ_Co in Expression (4), it is found that the square of the rotation speed ω_m is in direct proportion to a quadratic expression of the d-axis current i_d. That is, by determining the d-axis current is in accordance with the rotation speed ω_m according to Expression (5), the axial deviation δ_C can be less than or equal to the upper limit value δ_Co.

$\begin{array}{cc}{\omega }_m^2=-\frac{k_B{\delta }_{\mathrm{co}}}{k_A{m}_A{r}_A+k_C{m}_C{r}_C}\left({\mathrm{ai}}_d^2+{\mathrm{bi}}_d+c\right)& \left(5\right)\end{array}$

As understood from Expression (5), when the d-axis current i_d is larger than the value (−b/2a), as the d-axis current i_d is smaller, the axial deviation δ_C is also smaller. When the d-axis current i_d is smaller than the value (−b/2a), as the d-axis current i_d is smaller, the axial deviation δ_C is larger. Thus, in order to reduce the axial deviation δ_C as much as possible, the d-axis current i_d is desirably the value (−b/2a).

FIG. 6 is a graph illustrating a relationship between a current amplitude ia (arbitrary unit) and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque τ is fixed. Herein, ia=[i_d²+i_q²]^1/2 and is the amplitude of a current vector Ia when the alternating currents Iu, Iv, and Iw are expressed as the current vector Ia.

FIG. 6 illustrates a relationship between the axial deviation δ_C and the current amplitude ia when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. When the torque τ is maintained, as the current amplitude ia for achieving the rotation speed (o, is larger, the axial deviation δ_C is smaller. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 6, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the current amplitude ia employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the current amplitude ia becomes a value ia{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the current amplitude ia becomes a value ia0.

FIG. 7 is a graph illustrating a relationship between the rotation speed win and the current amplitude ia (arbitrary unit: the same unit as in FIG. 6) when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

Thus, when the rotation speed ω_m exceeds the speed v2, the current amplitude ia becomes a value larger than the value employed during the flux-weakening control (this is larger than the value ia{circumflex over ( )}), so that the above-described voltage reduction control can be performed.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the alternating currents Iu, Iv. and Iw from which the current vector Ia with the current amplitude ia larger than the value of the current amplitude ia employed during the flux-weakening control (this is larger than the value ia{circumflex over ( )}) is obtained.

The value of the current amplitude ia employed during the flux-weakening control can be obtained as follows. Expressions (6), (7), (8), and (9) are satisfied by adopting a rotation speed ω as an electric angle, the torque τ (this may be substituted by the torque command τ*), d-axis inductance L_d and q-axis inductance L_q of the motor 1, a field magnetic flux ψ_a generated by a permanent magnet of the field element included in the motor 1, an electric resistance R_a of the motor 1, the d-axis voltage v_d and the q-axis voltage v_q (these may be substituted by the respective command value v_d* and v_q*), and a differential operator p.

$\begin{array}{cc}V\max =\sqrt{v_d^2+v_q^2}& \left(6\right)\\ \left(\begin{array}{c}{v}_d\\ v_q\end{array}\right)\left(\begin{array}{cc}{R}_a+{\mathrm{pL}}_d& -\omega L_q\\ \omega L_d& R_a+{\mathrm{pL}}_q\end{array}\right)\left(\begin{array}{c}{i}_d\\ i_q\end{array}\right)+\left(\begin{array}{c}0\\ {\mathrm{\omega \Psi }}_a\end{array}\right)& \left(7\right)\\ \tau =P_n{\Psi }_a{i}_q+P_n\left(L_d-L_q\right)i_d{i}_q& \left(8\right)\\ \mathrm{ia}=\sqrt{i_d^2+i_q^2}& \left(9\right)\end{array}$

The rotation speed ω is obtained by the product of the rotation speed ω_m and the number of pole pairs P_n. Thus, the current amplitude ia obtained from simultaneous equations of Expressions (6), (7), (8), and (9) where ω=P_n·ω_m is the value of the current amplitude ia employed during the flux-weakening control. The current amplitude ia obtained from simultaneous equations of Expressions (6), (7), (8), and (9) in which the left side of Expression (6) is ω=P_n·v1 is the value ia0.

FIG. 8 is a graph illustrating a relationship between a phase β of the current vector Ia with respect to the q-axis and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque τ is fixed. There is a relationship of Expression (10).

$\begin{array}{cc}\beta ={\tan }^{-1}\left(-\frac{i_d}{i_q}\right)& \left(10\right)\end{array}$

FIG. 8 illustrates a relationship between the axial deviation δ_C and the phase β when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. When the torque τ is maintained, as the phase β for achieving the rotation speed ω_m is larger, the axial deviation δ_C is smaller. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 8, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the phase β employed during the flux-weakening control when the rotation speed Om becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the phase β becomes a value β{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the phase β becomes a value β0.

FIG. 9 is a graph illustrating a relationship between the rotation speed (on and the phase β when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

Thus, when the rotation speed ω_m exceeds the speed v2, the phase β becomes a value larger than the value employed during the flux-weakening control (this is larger than the value β{circumflex over ( )}), so that the above-described voltage reduction control can be performed.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the alternating currents Iu, Iv, and Iw with which the phase β larger than the value of the phase β employed during the flux-weakening control is obtained.

The phase β obtained from simultaneous equations of Expressions (6), (7), (8), and (10) where ω=P_n·ω_m is the value of the phase β employed during the flux-weakening control. The phase β obtained from simultaneous equations of Expressions (6), (7), (8), and (10) in which the left side of Expression (6) is ω=P_n·v1 is the value β0.

FIG. 10 is a graph illustrating a relationship between the d-axis current i_d (<0; arbitrary unit) and the axial deviation δ_C in which the rotation speed ω_m is a parameter.

FIG. 10 illustrates a relationship between the axial deviation δ_C and the d-axis current i_d when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. Note that the torque τ is fixed. When the torque τ is maintained, as the d-axis current is for achieving the rotation speed ω_m is larger (the absolute value is smaller), the axial deviation δ_C is larger. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 10, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the d-axis current is employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the d-axis current is becomes a value i_d{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the d-axis current i_d becomes a value i_d0 (refer to the lower graph in FIG. 3).

Thus, when the rotation speed ω_m exceeds the speed v2, the d-axis current is becomes a value smaller (the absolute value is larger) than a value employed during the flux-weakening control (this is smaller than the value i_d{circumflex over ( )}), so that the above-described voltage reduction control can be performed.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the alternating currents Iu, Iv, and Iw having a d-axis component the value of which is smaller than the value of the d-axis current is employed during the flux-weakening control.

FIG. 11 is a graph illustrating a relationship between the q-axis current i_q (arbitrary unit) and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque τ is fixed.

FIG. 11 illustrates a relationship with the axial deviation δ_C when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. When the torque τ is maintained, as the q-axis current i_q for achieving the rotation speed ω_m is larger, the axial deviation δ_C is larger. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 11, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the q-axis current i_q employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the q-axis current i_q becomes a value i_q{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the q-axis current i_q becomes a value i_q0.

Thus, when the rotation speed ω_m exceeds the speed v2, the q-axis current i_q becomes a value smaller than the value employed during the flux-weakening control (this is smaller than the value i_q{circumflex over ( )}), so that the above-described voltage reduction control can be performed.

FIG. 12 is a graph illustrating a relationship between the rotation speed ω_m and the q-axis current i_q (arbitrary unit: the same unit as in FIG. 11) when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the alternating currents Iu. Iv, and Iw having a q-axis component the value of which is smaller than the value of the q-axis current i_q employed during the flux-weakening control.

The d-axis current i_d and the q-axis current i_q obtained from simultaneous equations of Expressions (6), (7), and (8) where ω=P_n·ω_m are a d-axis current and a q-axis current, respectively, employed during the flux-weakening control. The d-axis current is and the q-axis current i_q obtained from simultaneous equations of Expressions (6), (7), and (8) in which the left side of Expression (6) is ω=P_n·v1 is the values i_d0 and i_q0.

FIG. 13 is a vector diagram illustrating a relationship between a field magnetic flux vector ψ_a, a magnetic flux vector ψ_b attributed to an armature reaction, and a primary magnetic flux vector λ₀. In FIG. 13, in order to explicitly indicate that these magnetic flux vectors ψ_a, ψ_b, and λ₀ are vectors, arrows are shown for the respective symbols. Note that, by using the same symbols, amplitudes of these vectors are also referred to as field magnetic flux ψ_a, magnetic flux ψ_b, and primary magnetic flux λ₀ in the description of this embodiment.

The primary magnetic flux vector λ₀ is a composite of a magnetic flux vector (−ψ_b) and the field magnetic flux vector ψ_a. A load angle δ₀ is a phase of the primary magnetic flux vector λ₀ with respect to the field magnetic flux vector ψ_a. The primary magnetic flux λ₀ is represented as Expression (11). There is a relationship of Expression (12) between the primary magnetic flux λ₀ and the load angle δ₀.

$\begin{array}{cc}{\lambda }_o=\sqrt{{\left({\Psi }_a+L_d{i}_d\right)}^2+{\left(L_q{i}_q\right)}^2}& \left(11\right)\\ \{\begin{array}{c}{\lambda }_o\cos {\delta }_o=L_d{i}_d+{\Psi }_a\\ {\lambda }_o\sin {\delta }_o=L_q{i}_q\end{array}& \left(12\right)\end{array}$

An α-axis and a β-axis are coordinate axes of a fixed coordinate system in the motor 1. The d-axis and the q-axis are coordinate axes of a rotary coordinate system, meaning of each of which is described above. The field magnetic flux vector ψ_a and the d-axis have the same phase and the same direction in a vector diagram. An M-axis and a T-axis indicate coordinate axes that advance in the same phase as the primary magnetic flux vector λ₀ and 90 degrees with respect to this, respectively. The primary magnetic flux vector λ₀ and the M-axis have the same direction in a vector diagram. Hereinafter, an M-axis component and a T-axis component of the three-phase alternating currents Iu, Iv, and Iw output to the motor 1 are also referred to as M-axis current i_M and T-axis current i_T. The T-axis current i_T is represented as Expression (13).

i _r =−i _d sin δ_o+i_q cos δ_o  (13)

FIG. 14 is a graph illustrating a relationship between the T-axis current i_T (arbitrary unit) and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque τ is fixed.

FIG. 14 illustrates a relationship between the axial deviation δ_C and the T-axis current it when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. When the torque τ is maintained, as the T-axis current i_T for achieving the rotation speed ω_m is larger, the axial deviation δ_C is smaller. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 14, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the T-axis current i_T employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the T-axis current i_T becomes a value i_T{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the T-axis current i_T becomes a value i_T0.

Thus, when the rotation speed ω_m exceeds the speed v2, the T-axis current i_T becomes a value larger than the value employed during the flux-weakening control (this is larger than the value i_T{circumflex over ( )}), so that the above-described voltage reduction control can be performed.

FIG. 15 is a graph illustrating a relationship between the rotation speed ω_m and the T-axis current i_T (arbitrary unit: the same unit as in FIG. 14) when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the alternating currents Iu, Iv, and Iw having a T-axis component (T-axis current i_T) the value of which is larger than the value of the T-axis component of the alternating currents Iu, Iv, and Iw output to the motor 1 in a case in which the flux-weakening control is performed at the speed.

The T-axis current i_T obtained from simultaneous equations of Expressions (6), (7), (8), (12), and (13) where ω=P_n·ω_m is the value of the T-axis current i_T in a case in which the flux-weakening control is performed. The T-axis current i_T obtained from simultaneous equations of Expressions (6), (7), (8), (12), and (13) in which the left side of Expression (6) is ω=P_n·v1 is the value i_T0.

FIG. 16 is a graph illustrating a relationship between the primary magnetic flux λ₀ (arbitrary unit) and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque τ is fixed.

FIG. 16 illustrates a relationship between the axial deviation δ_C and the primary magnetic flux λ₀ when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. When the torque τ is maintained, as the primary magnetic flux λ₀ for achieving the rotation speed ω_m is larger, the axial deviation δ_C is larger. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 16, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the primary magnetic flux λ₀ employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the primary magnetic flux λ₀ becomes a value λ₀{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the primary magnetic flux λ₀ becomes a value λ₀0.

Thus, when the rotation speed ω_m exceeds the speed v2, the primary magnetic flux λ₀ having a value smaller than the value of the primary magnetic flux in a case in which the flux-weakening control is performed is generated, so that the above-described voltage reduction control can be performed.

FIG. 17 is a graph illustrating a relationship between the rotation speed ω_m and the primary magnetic flux λ₀ (arbitrary unit: the same unit as in FIG. 16) when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the alternating currents Iu, Iv, and Iw that causes the motor 1 to generate the primary magnetic flux λ₀ smaller than the value of the primary magnetic flux in a case in which the flux-weakening control is performed.

The primary magnetic flux λ₀ obtained from simultaneous equations of Expressions (6), (7), (8), and (11) where ω=P_n·ω_m is the value of the primary magnetic flux λ₀ in a case in which the flux-weakening control is performed. The primary magnetic flux λ₀ obtained from simultaneous equations of Expressions (6), (7), (8), and (11) in which the left side of Expression (6) is ω=P_n·v1 is the value λ₀0.

FIG. 18 is a graph illustrating a relationship between the load angle δ₀ and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque is fixed.

FIG. 18 illustrates a relationship between the axial deviation δ_C and the load angle δ₀ when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. When the torque τ is maintained, as the load angle δ₀ for achieving the rotation speed ω_m is larger, the axial deviation δ_C is smaller. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 18, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the load angle δ₀ employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the load angle δ₀ becomes a value δ₀{circumflex over ( )} obtained as described later. When the rotation speed ω_m is lower than or equal to the speed v1 during the maximum torque/current control, the load angle δ₀ becomes a value δ₀0.

Thus, when the rotation speed ω_m exceeds the speed v2, the load angle δ₀ becomes a value larger than the value of the load angle in a case in which the flux-weakening control is performed, so that the above-described voltage reduction control can be performed.

FIG. 19 is a graph illustrating a relationship between the rotation speed ω_m and the load angle δ₀ when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output the alternating currents Iu, Iv, and Iw that causes the motor 1 to generate the load angle δ₀ larger than the value of the load angle in a case in which the flux-weakening control is performed.

The load angle δ₀ obtained from simultaneous equations of Expressions (6), (7), (8), and (12) where ω=P_n·ω_m is the value of the load angle δ₀ in a case in which the flux-weakening control is performed. The load angle δ₀ obtained from simultaneous equations of Expressions (6), (7), (8), and (12) in which the left side of Expression (6) is ω=P_n·v1 is the value δ₀0.

FIG. 20 is a graph illustrating a relationship between an instantaneous real power Po (arbitrary unit) and the axial deviation δ_C in which the rotation speed ω_m is a parameter. Note that the torque is fixed.

FIG. 20 illustrates a relationship between the axial deviation δ_C and the instantaneous real power Po when the rotation speed ω_m becomes a speed v1, v5, v6, or v7. The instantaneous real power Po is an instantaneous real power supplied from the output circuit 210 to the motor 1. In other words, the instantaneous real power Po is an instantaneous real power generated by the motor 1. Po=vd·id+vq·iq and, for example, can be calculated by v_d*·i_d+v_q*·i_q by using the command values v_d* and v_q*.

When the torque τ is maintained, as the instantaneous real power Po for achieving the rotation speed ω_m is larger, the axial deviation δ_C is smaller. As the rotation speed ω_m is higher, the axial deviation δ_C is larger.

In FIG. 20, the value of the axial deviation δ_C (this corresponds to the upper limit value δ_Co) with the instantaneous real power Po employed during the flux-weakening control when the rotation speed ω_m becomes a speed v2 is additionally plotted. In this case, the amplitude |Vs| becomes the voltage value Vmax, and the instantaneous real power Po becomes a value Po{circumflex over ( )}(=v_d*·i_d{circumflex over ( )}+v_q*·i_q{circumflex over ( )}). When the rotation speed ω_m is lower than or equal to the speed v during the maximum torque/current control, the instantaneous real power Po becomes less than or equal to a value Po0 (=v_d*·i_d0+v_q*·i_q0).

Thus, when the rotation speed ω_m exceeds the speed v2, the instantaneous real power Po becomes a value larger than the value of the instantaneous real power in a case in which the flux-weakening control is performed, so that the above-described voltage reduction control can be performed.

FIG. 21 is a graph illustrating a relationship between the rotation speed ω_m and the instantaneous real power Po (arbitrary unit: the same unit as in FIG. 20) when the torque τ is a predetermined value. The solid line illustrates a case in which ω_m>v2 and the voltage reduction control is employed, and the broken line illustrates a case in which ω_m>v2 and the flux-weakening control is employed. In the illustrated cases, the maximum torque/current control is employed when ω_m≤v1 and the flux-weakening control is employed when v1<ω_m≤v2.

That is, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the output circuit 210 to output, to the motor 1, the instantaneous real power Po larger than the value of the instantaneous real power in a case in which the flux-weakening control is performed.

FIG. 22 is a block diagram illustrating a first modification of the controller 209. The first modification extracts and illustrates only the current command generating unit 211, the current controller 212 and a periphery thereof illustrated in FIG. 2. In the first modification, a limiter 219 is provided between the current command generating unit 211 and the current controller 212 in the controller 209, and the command value i_d* of the d-axis current i_d is limited to less than or equal to an upper limit value i_dlim. Specifically, if the command value i_d* obtained from the current command generating unit 211 exceeds the upper limit value i_dlim, the limiter 219 inputs the upper limit value i_dlim as the command value i_d* to the current controller 212.

In the first modification, an upper-limit-value calculating unit 220 is further provided in the controller 209. The upper-limit-value calculating unit 220 calculates the upper limit value i_dlim by using the command value i_q* of the q-axis current i_q, the command value Om of the rotation speed ω_m, and the upper limit value δ_Co of the axial deviation δ_C. Expression (5) can be modified into Expression (14).

$\begin{array}{cc}{i}_d=\frac{-b+\sqrt{b^2-4a\left(c+\frac{{\omega }_m^2\left(k_A{m}_A{r}_A+k_C{m}_C{r}_C\right)}{k_B{\delta }_{\mathrm{co}}}\right)}}{2a}& \left(14\right)\end{array}$

In Expression (14), the upper limit value i_dlim can be calculated as the value of the d-axis current i_d obtained by employing the command value ω_m* as the rotation speed ω_m.

As described above, in order to reduce the axial deviation δ_C as much as possible, the d-axis current i_d is desirably the value (−b/2a). Thus, it is desirable not to satisfy i_dlim<(−b/2a). If i_dlim<(−b/2a), for example, control for reducing the command value ω_m* (droop control) is desirably performed.

FIG. 23 is a block diagram illustrating a second modification of the controller 209. The second modification can be employed for so-called primary magnetic flux control in which the primary magnetic flux λ₀ is controlled.

The controller 209 includes, for example, a voltage command generating unit 221, coordinate converters 223 and 224, and an angle calculating unit 227.

From a command value ω* of a rotation speed ω as an electric angle and the T-axis current i_T, by using a known method, the angle calculating unit 227 obtains a rotation speed ω_OC of the M-axis and further obtains a position θ_OC of the M-axis. From the values of the alternating currents Iu and Iv and the position θ_OC, the coordinate converter 224 obtains the M-axis current i_M and the T-axis current i_T.

The voltage command generating unit 221 obtains the M-axis current i_M, the T-axis current i_T, and a command value λ₀* of the primary magnetic flux λ⁰, and, from the rotation speed ω_OC, a command value v_T* of a T-axis voltage v_T and a command value v_M* of the M-axis voltage v_M.

From the command values v_T* and v_M* and the position θ_OC, the coordinate converter 223 generates the three-phase voltage command values v_u*, v_v*, and v_w*.

The controller 209 further includes a limiter 229, the upper-limit-value calculating unit 220, and an upper-limit-value calculating unit 225. The limiter 229 limits the command value λ₀* of the primary magnetic flux λ₀ to less than or equal to an upper limit value λ_0lim. Specifically, if the command value λ₀* exceeds the upper limit value λ_0lim, the limiter 229 inputs the upper limit value λ_0lim as the command value λ₀* to the voltage command generating unit 221.

The upper-limit-value calculating unit 220 can calculate the upper limit value i_dlim as the value of the d-axis current i_d obtained by employing the command value ω_m* as the rotation speed ω_m and an estimated value i_qe as the q-axis current i_q in Expression (14).

The upper-limit-value calculating unit 225 can calculate the upper limit value λ_0lim as the value of the primary magnetic flux λ₀ obtained by employing i_d=i_dlim and i_q=i_qe in Expression (11).

Expression (12) can be modified into Expression (15). From Expressions (4) and (15), Expression (16) can be obtained. From Expression (16), if the load angle δ₀ and the axial deviation δ_C are fixed, it is found that the square of the rotation speed ω_m is in direct proportion to a quadratic expression of the primary magnetic flux λ₀.

$\begin{array}{cc}\{\begin{array}{c}{i}_d=\frac{{\lambda }_o\cos {\delta }_o-{\Psi }_a}{L_d}\\ i_q=\frac{{\lambda }_o\sin {\delta }_o}{L_q}\end{array}& \left(15\right)\\ {\omega }_m^2=\frac{\text{?}\text{?}}{k_A{m}_A{r}_A+k_C{m}_C{r}_C}\left(a{\left\{\frac{{\lambda }_o\cos {\delta }_o-{\Psi }_a}{L_d}\right\}}^2+b\left\{\frac{{\lambda }_o\cos \text{?}-{\Psi }_a}{\text{?}}\right\}+\frac{x}{g}·\frac{\pi }{2{\mu }_0}·\left\{\begin{array}{c}+\left(\text{?}\right)\left[{\left(\text{?}+\text{?}\right)}^2+\text{?}\right]\\ +{\left(\text{?}\frac{\text{?}\sin \text{?}}{\text{?}}\right)}^2\left[{\left(\text{?}-\text{?}\right)}^2+\text{?}\right]\end{array}\right\}\right)=-\frac{\text{?}\text{?}}{k_A{m}_A{r}_A+k_C{m}_C{r}_C}\left(a{\left\{\frac{{\lambda }_o\cos {\delta }_o-{\Psi }_a}{L_d}\right\}}^2+b\left\{\frac{{\lambda }_o\cos {\delta }_o-{\Psi }_a}{L_d}\right\}+d+e{\left\{\frac{{\lambda }_o\sin {\delta }_o}{L_q}\right\}}^2\right)=-\frac{\text{?}\text{?}}{k_A{m}_A{r}_A+k_C{m}_C{r}_C}\left\{\left(\frac{{\mathrm{acos}}^2\text{?}}{L_d^2}+\frac{?{\sin }^2\text{?}}{L_q^2}\right)\text{?}+\left(-\frac{a2\cos {\delta }_o{\Psi }_a}{L_d^2}+\frac{b\cos {\delta }_o}{L_d}\right){\lambda }_o+\left(\frac{a{\Psi }_o^2}{L_d^2}-\frac{b{\Psi }_a}{L_d}+d\right)\right\}& \left(16\right)\\ \mathrm{where}\{\begin{array}{c}d=\frac{x}{g}·\frac{\pi }{2{\mu }_0}\left(F_M^2\right)\left[{\left(\text{?}+P_{d1}\right)}^2+\text{?}\right]\\ e=\frac{x}{g}·\frac{\pi }{2{\mu }_0}\left[{\left(\text{?}-\text{?}\right)}^2+\text{?}\right]\end{array}\text{?}\text{indicates text missing or illegible when filed}& \end{array}$

The upper limit value λ_0lim may also be obtained according to Expression (16) in which δ_C=δ_Co and ω_m=ω_m*.

As described above, the motor control device 200 includes the PWM inverter 210b and the controller 209. The PWM inverter 210b outputs, to the motor 1, the application voltage Vs to be applied to the motor 1. The controller 209 controls operation of the PWM inverter 210b. The motor 1 drives the compression mechanism 20, which is the load, by using rotation of the shaft 10. The PWM inverter 210b is included in the output circuit 210.

In the above-described embodiment, for example, in a case in which the predetermined torque τ is caused to be output from the motor 1,

(i) when the rotation speed ω_m is lower than or equal to the speed v1, as the rotation speed ω_m is higher, the amplitude |Vs| is larger (e.g., the maximum torque/current control or the maximum efficiency control);

(ii) the amplitude |Vs| when the rotation speed ω_m is higher than the speed v2 (≥v1) is less than the voltage value Vmax of the amplitude |Vs| at the speed v1 (the voltage reduction control); and

(iii) the amplitude |Vs| when the rotation speed ω_m is higher than the speed v1 and lower than or equal to the speed v2 is the voltage value Vmax (e.g., the flux-weakening control).

For example, during the voltage reduction control, when the rotation speed ω_m is higher than the speed v2, as the rotation speed ω_m is higher, the amplitude |Vs| is smaller.

In a case in which the motor 1 is caused to generate the predetermined torque T, when the rotation speed ω_m exceeds the speed v2, the controller 209 causes the PWM inverter 210b to, for example:

(iia) output, to the motor 1, the alternating currents Iu, Iv, and Iw with the phase β larger than the phase β of the alternating currents Iu, Iv, and Iw output to the motor 1 when the flux-weakening control is applied at the rotation speed;

(iib) output, to the motor 1, the alternating currents Iu, Iv, and Iw from which the current vector Ia with the current amplitude ia larger than the current amplitude ia of the current vector Ia of the alternating currents Iu, Iv, and Iw output to the motor 1 when the flux-weakening control is applied at the rotation speed can be obtained;

(iic) output, to the motor 1, the alternating currents Iu, Iv, and Iw having a d-axis component (d-axis current i_d) smaller than the d-axis component (value is of the d-axis current i_d) of the alternating currents Iu, Iv, and Iw output to the motor 1 when the flux-weakening control is applied at the rotation speed;

(iid) output, to the motor 1, the alternating currents Iu, Iv, and Iw having a q-axis component (q-axis current i_q) smaller than the q-axis component (value i_q of the q-axis current i_q) of the alternating currents Iu, Iv, and Iw output to the motor 1 when the flux-weakening control is applied at the rotation speed;

(iie) output, to the motor 1, the alternating currents Iu, Iv, and Iw having a T-axis component (T-axis current i_T) larger than the T-axis component (value it of the T-axis current i_T) of the alternating currents Iu, Iv, and Iw output to the motor 1 when the flux-weakening control is applied at the rotation speed;

(iif) output, to the motor 1, the alternating currents Iu, Iv, and Iw that causes the motor 1 to generate the primary magnetic flux λ₀ with an amplitude smaller than the primary magnetic flux (more strictly, the value λ₀ of the amplitude) generated in the motor 1 when the flux-weakening control is applied at the rotation speed;

(iig) output, to the motor 1, the alternating currents Iu, Iv, and Iw that causes the motor 1 to generate the primary magnetic flux λ₀ with a load angle δ₀ larger than the load angle δ₀ of the primary magnetic flux λ₀ generated in the motor 1 when the flux-weakening control is applied at the rotation speed; or

(iih) output, to the motor 1, the instantaneous real power Po larger than the instantaneous real power generated in the motor 1 when the flux-weakening control is applied at the rotation speed.

It is not always necessary to employ the maximum torque/current control, the maximum efficiency control, or the flux-weakening control. Typically, the maximum rotation speed of a motor employed in a product system is determined depending on the product system. The product system herein includes, in terms of the embodiment, the motor 1, the motor control device 200, and the compression mechanism 20 driven by the motor 1. The maximum amplitude |Vs| depends on the rotation speed ω_m.

For convenience of the following description, various quantities are defined. The maximum rotation speed ω_m of the motor 1 determined depending on the product system is a speed ω_MAX. A possible maximum value of the amplitude |Vs| when the motor 1 rotates at the speed ω_MAX is a voltage value V_max_ω_MAX. A possible maximum value of the amplitude |Vs| when the motor 1 rotates at a speed ω3 lower than the speed ω_MAX is a voltage value V_max_ω3.

As described above, as the rotation speed is higher, the axial deviation δ_C is larger. The axial deviation δ_C can be decreased by decreasing the amplitude |Vs|. Thus, when the motor 1 rotates at the speed ω_MAX, the PWM inverter 210b desirably outputs the application voltage Vs smaller than the voltage value V_Max_ω_Max.

On the other hand, in order to reduce current to be consumed, the amplitude |Vs| desirably becomes the possible maximum when the motor 1 rotates. Thus, at the at least one speed ω3, the PWM inverter 210b desirably outputs the application voltage Vs with the amplitude |Vs| of the voltage value V_max_ω3.

These can be summarized and expressed as follows:

(a) the PWM inverter 210b is caused to output the application voltage Vs having the amplitude |Vs| smaller than the voltage value V_max_ω_MAX, and the motor 1 is caused to rotate at the speed ω_MAX and drive a load (e.g., the compression mechanism 20); and

(b) the PWM inverter 210b is caused to output the application voltage Vs having the amplitude |Vs| of the voltage value V_max_ω3, and the motor 1 is caused to rotate at the speed ω3 (<ω_MAX) and drive the load; in which

(c) the voltage value V_max_ω_MAX is a possible maximum value of the amplitude |Vs| when the motor drives the load at the speed ω_MAX;

(d) the speed ω_MAX is a maximum of the rotation speed ω_m when the motor 1 drives the load;

(e) the voltage value V_max_ω3 is a possible maximum value of the amplitude |Vs| when the motor 1 drives the load at the speed ω3; and

(f) the speed ω3 is lower than the speed ω_MAX (the above conditions are not necessarily satisfied at all rotation speeds ω_m smaller than the speed ω_MAX).

In other words:

(g) at the speed ω_MAX, the ratio of the amplitude |Vs| to the voltage value V_max_ω_MAX is smaller than 1; and

(h) at the speed ω3 lower than the speed ω_MAX, the ratio of the amplitude |Vs| to the voltage value V_max_ω3 is equal to 1.

Without limitation to when the motor 1 rotates at the speed ω_MAX, as the rotation speed ω_m is higher, the axial deviation δ_C is larger. In addition, the voltage reduction control is performed at the rotation speed ω_m higher than or equal to the base speed (defined as a maximum rotation speed of the motor 1 at which the motor 1 can generate the torque τ during the maximum torque/current control or the maximum efficiency control). Thus, by adopting the base speed ωb when the motor 1 outputs the predetermined torque τ, the speeds ω1 (≥ωb) and ω2 (>ω1), the voltage value V_max_ω1 as the possible maximum of the amplitude |Vs| at rotation at the speed ω1, and the voltage value V_max_ω2 as the possible maximum of the amplitude |Vs| at rotation at the speed ω2, there may be a relationship as follows.

When the motor 1 outputs the predetermined torque τ,

(i) the ratio of the amplitude |Vs| to the voltage value V_max_ω1 at the certain speed ω1, which is higher than or equal to the base speed ωb when the predetermined torque τ is output is a first ratio;

(j) the ratio of the amplitude |Vs| to the voltage value V_max_ω2 at the certain speed ω2 higher than the speed ω1 is a second ratio; and

(k) the second ratio is smaller than the first ratio (the above conditions are not necessarily satisfied at all of two rotation speeds ω_m higher than or equal to the base speed ωb).

In other words, in a case in which the rotation speed ω_m when the motor 1 outputs the predetermined torque τ is higher than or equal to the base speed ωb when the predetermined torque τ is output:

(l) the PWM inverter 210b is caused to output the application voltage Vs having the amplitude |Vs| obtained by multiplying the voltage value V_max_ω1 by the first ratio, the motor 1 is caused to rotate at the speed ω1, and the motor 1 is caused to output the torque τ;

(m) the PWM inverter 210b is caused to output the application voltage Vs having the amplitude |Vs| obtained by multiplying the voltage value V_max_ω2 by the second ratio, the motor 1 is caused to rotate at the speed ω2, and the motor 1 is caused to output the torque T;

(n) the voltage value V_max_ω1 is a possible maximum value of the amplitude |Vs| when the motor 1 outputs the torque τ at the speed ω1:

(o) the voltage value V_max_ω2 is a possible maximum value of the amplitude |Vs| when the motor 1 outputs the torque τ at the speed ω2:

(p) the speed ω2 is higher than the speed ω1; and

(q) the second ratio is smaller than the first ratio.

For the relationship ω2>ω1>ωb, the speed ω2 may be a possible maximum ω_max of the rotation speed ω_m when the motor 1 outputs the torque τ. When ω1=v1, V_max=V_max_ω1 is satisfied.

A case in which v2>ωb and the torque τ is maintained is described with reference to FIG. 3 as an example. As described above, v6>v5>v2 is satisfied.

(l′) The motor 1 is caused to rotate at the speed v5, and the amplitude |Vs| in this case has a value obtained by multiplying the first voltage value by the first ratio;

(m′) the motor 1 is caused to rotate at the speed v6, and the amplitude |Vs| in this case has a value obtained by multiplying the second voltage value by the second ratio;

(n′) the first voltage value is the possible maximum value of the amplitude |Vs| when the motor 1 outputs the torque τ at the speed v5;

(o′) the second voltage value is the possible maximum value of the amplitude |Vs| when the motor 1 outputs the torque τ at the speed v6:

(p′) the speed v6 is higher than the speed v5; and

(q′) the second ratio is smaller than the first ratio.

By the above-described control, the radial stress at a specific rotation angle when the motor 1 is rotating is reduced. This contributes to reduction of the uneven contact of the shaft 10 to the bearing 14.

Although the power source that supplies the DC voltage Vdc is provided outside the motor control device 200, the power source may alternatively be included in the motor control device 200. The power source can be realized, for example, an AC/DC converter. The amplitude |Vs| of the application voltage Vs output from the PWM inverter 210b in such a case will be described below.

The converter converts an AC voltage Vin into the DC voltage Vdc. In this conversion, an alternating current Tin flows into the converter and a direct current Idc is output. A power factor cos(Din on the input side of the converter and a loss Ploss1 at the time of conversion of the converter are adopted.

In the following description, the PWM inverter 210b outputs an AC voltage Vout and an alternating current Iout. A power factor cos Φ out on the output side of the PWM inverter 210b and a loss Ploss2 at the time of conversion of the PWM inverter 210b are adopted.

Regarding the converter, the following Expression (17) is satisfied based on the law of the conservation of energy. In the first expression, the second term on the right side indicates voltage drop attributed to a converter loss. A transformer ratio a of the converter is adopted.

Vdc=Vin×a−Ploss1/Idc, a=Iin×cos Φin/Idc  (17)

Regarding the PWM inverter 210b, the following Expression (18) is satisfied based on the law of the conservation of energy. In the first expression, the second term on the right side indicates voltage drop attributed to a converter loss. A modulation index b of the PWM inverter 210b is adopted.

Vout=Vdc×b−Ploss2/(Iout×cos Φout), b=Idc/Iout/cos Φout   (18)

From Expressions (17) and (18), the following expression is satisfied.

$\begin{array}{cc}\mathrm{Vout}=\left(\mathrm{Vin}\times a-\mathrm{Ploss}1/\mathrm{Idc}\right)\times b-\mathrm{Ploss}2/\left(\mathrm{Iout}\times \cos \Phi \mathrm{out}\right)=\mathrm{Vin}\times a\times b-b\times \mathrm{Ploss}1/\mathrm{Idc}-\mathrm{Ploss}2/\left(\mathrm{Iout}\times \cos \Phi \mathrm{out}\right)& \left(19\right)\end{array}$

From Expression (19), the AC voltage Vout output from the PWM inverter 210b is uniquely determined by the AC voltage Vin converted by the converter, the transformer ratio a, the modulation index b, the loss Ploss1 of the converter, the loss Ploss2 of the PWM inverter 210b, the direct current Idc input to the PWM inverter 210b, the alternating current Iout output from the PWM inverter 210b, and the power factor cos Φ out of the PWM inverter 210b. Note that the transformer ratio a, the modulation index b, the losses Ploss1 and Ploss2, the direct current Idc, the alternating current Iout, and the power factor cos Φ out are uniquely determined if the product system that employs the motor to which voltage is applied from the PWM inverter 210b, and the torque and rotation speed of the motor are determined.

Thus, the amplitude |Vs| in the above embodiment is uniquely determined if the power source voltage, the product system, the torque t, and the rotation speed ω_m are determined. However, in a case in which the power source that supplies the DC voltage Vdc is an AC/DC converter, the amplitude |Vs| is also dependent on the AC voltage Vin input to the converter.

The maximum of the amplitude |Vs| is further described. From Expression (19), the AC voltage Vout becomes a maximum when the transformer ratio a and the modulation index b are maximums. When maximums aMAX and bMAX of the transformer ratio a and the modulation index b, respectively, are adopted, a maximum VoutMAX of the AC voltage Vout is determined according to the following Expression (20).

VoutMAX=Vin×aMAX×bMAX−bMAX×Ploss1/Idc−Ploss2/(Iout×cos Φout)  (20)

The maximums aMAX and bMAX are each uniquely determined according to product system. As described above, the amplitude |Vs| is uniquely determined if the power source voltage, the product system, the torque τ, and the rotation speed ω_m are determined. Thus, the maximum of the amplitude |Vs| is also uniquely determined if the power source voltage, the product system, the torque τ, and the rotation speed (o, are determined. For example, when the same torque τ is maintained in a certain product system at a certain power source voltage, the voltage values V_max_ω1, V_max_ω2, V_max_ω3, and V_max_ω_MAX are uniquely determined by the speeds ω1, ω2, ω3, and ω_MAX, respectively.

However, in a case in which the power source that supplies the DC voltage Vdc is an AC/DC converter, these voltage values are also dependent on the AC voltage Vin input to the converter.

Although the embodiment has been described above, it should be understood that various modifications can be made for forms or details without departing from the spirit and scope of the claims. The above various embodiment and modifications can be mutually combined.

Claims

1. An inverter control method, which is a method for controlling an inverter that outputs an application voltage, which is a voltage to be applied to a motor that drives a load by using rotation of a shaft, comprising: causing the inverter to output the application voltage having an amplitude smaller than a first maximum and causing the motor to rotate at a first speed and drive the load which is predetermined; andcausing the inverter to output the application voltage having an amplitude of a second maximum and causing the motor to rotate at a second speed and drive the predetermined load,wherein the first maximum is a possible maximum value of an amplitude of the application voltage when the motor drives the predetermined load at the first speed,wherein the first speed is a maximum of a speed of rotation of the motor when the motor drives the predetermined load,wherein the second maximum is a possible maximum value of the amplitude of the application voltage when the motor drives the predetermined load at the second speed, andwherein the second speed is lower than the first speed.

2. An inverter control method, which is a method for controlling an inverter that outputs an application voltage, which is a voltage to be applied to a motor that drives a load by using rotation of a shaft, comprising: in a case in which a speed of rotation of the motor when the motor outputs a predetermined torque is higher than or equal to a base speed of the motor when the motor outputs the predetermined torque,causing the inverter to output the application voltage having an amplitude obtained by multiplying a first maximum by a first ratio, causing the motor to rotate at a first speed, and causing the motor to output the predetermined torque; andcausing the inverter to output the application voltage having an amplitude obtained by multiplying a second maximum by a second ratio, causing the motor to rotate at a second speed, and causing the motor to output the predetermined torque,wherein the first maximum is a possible maximum value of an amplitude of the application voltage when the motor outputs the predetermined torque at the first speed,wherein the second maximum is a possible maximum value of the amplitude of the application voltage when the motor outputs the predetermined torque at the second speed,wherein the second speed is higher than the first speed, andwherein the second ratio is smaller than the first ratio.

3. The inverter control method according to claim 2, wherein the second speed is a possible maximum of the speed when the motor outputs the predetermined torque.

4. A motor control device comprising: an inverter that outputs an application voltage, which is a voltage to be applied to a motor that drives a load by using rotation of a shaft; anda controller that controls operation of the inverter,wherein the controller causes the inverter to output the application voltage having an amplitude smaller than a first maximum and causes the motor to rotate at a first speed and drive the load which is predetermined,wherein the controller causes the inverter to output the application voltage having an amplitude of a second maximum and causes the motor to rotate at a second speed and drive the predetermined load,wherein the first maximum is a possible maximum value of an amplitude of the application voltage when the motor drives the predetermined load at the first speed,wherein the first speed is a maximum of a speed of rotation of the motor when the motor drives the predetermined load,wherein the second maximum is a possible maximum value of the amplitude of the application voltage when the motor drives the predetermined load at the second speed, andwherein the second speed is lower than the first speed.

5. A motor control device comprising: an inverter that outputs an application voltage, which is a voltage to be applied to a motor that drives a load by using rotation of a shaft; anda controller that controls operation of the inverter,wherein, in a case in which a speed of rotation of the motor when the motor outputs a predetermined torque is higher than or equal to a base speed of the motor when the motor outputs the predetermined torque,the controller causes the inverter to output the application voltage having an amplitude obtained by multiplying a first maximum by a first ratio, causes the motor to rotate at a first speed, and causes the motor to output the predetermined torque, andcauses the inverter to output the application voltage having an amplitude obtained by multiplying a second maximum by a second ratio, causes the motor to rotate at a second speed, and causes the motor to output the predetermined torque,wherein the first maximum is a possible maximum value of an amplitude of the application voltage when the motor outputs the predetermined torque at the first speed,wherein the second maximum is a possible maximum value of the amplitude of the application voltage when the motor outputs the predetermined torque at the second speed,wherein the second speed is higher than the first speed, andwherein the second ratio is smaller than the first ratio.

6. The motor control device according to claim 5, wherein the second speed is a possible maximum of the speed when the motor outputs the predetermined torque.