JP7393763B2 - Rotating electrical machine control system - Google Patents

Description

The present invention relates to a rotating electrical machine control system that performs current feedback control of a permanent magnet type rotating electrical machine in an orthogonal vector coordinate system of a d-axis along the direction of field magnetic flux by a permanent magnet and a q-axis orthogonal to the d-axis.

In a permanent magnet rotating electrical machine driven by alternating current, circumferential excitation force (torque ripple, which is torque fluctuation in the circumferential direction) and radial excitation force (attraction and repulsion forces generated between the permanent magnet and the stator) force) may cause vibration. It is important to reduce the vibration of the rotating electric machine because this vibration can cause problems such as becoming a source of audible noise. In the case of a three-phase AC rotating electric machine, it is known that the influence of the circumferential excitation force and the radial excitation force caused by the sixth harmonic of the electrical angle is large. Japanese Patent Laid-Open No. 2017-118726 describes such a system in which a current flowing through a rotating electric machine is feedback-controlled in a vector coordinate system of the d-axis, which is the direction of the magnetic field of a permanent magnet, and the q-axis, which is orthogonal to the d-axis. A technique for reducing the sixth-order circumferential excitation force and radial excitation force has been disclosed. According to this, the current command for one of the d and q axes is corrected using a correction value that suppresses the radial excitation force, and the current command for the d and q axes is corrected using a correction value that corrects the circumferential excitation force. ([0088] to [0091], [0098], [0105], etc.).

JP 2017-118726 Publication

However, in this case, even if the radial excitation force is reduced, the amount of reduction in the radial excitation force may decrease or become worse due to the current superimposed to reduce the circumferential excitation force. There is. For example, if the excitation force added to reduce the circumferential excitation force is large compared to the radial excitation force to be reduced, the radial excitation force may increase. Ta.

In view of the above background, it is desired to provide a technique for appropriately reducing both the circumferential excitation force and the radial excitation force.

In view of the above, a permanent magnet type rotating electric machine driven by N-phase alternating current (N is an arbitrary natural number) has two directions: a d-axis along the direction of field magnetic flux by the permanent magnets, and a q-axis perpendicular to the d-axis. One aspect of a rotating electrical machine control system that performs current feedback control in an orthogonal vector coordinate system is as follows:
a current command calculation unit that calculates a d-axis current command and a q-axis current command as a current command, which is a command value of a current to be passed through the rotating electrical machine, based on a target torque of the rotating electrical machine;
As a correction current command superimposed on the current command in order to reduce a radial excitation force that is a radial variation of the rotating electrical machine and a circumferential excitation force that is a circumferential torque variation of the rotating electrical machine, a correction current command calculation unit that calculates a d-axis correction current command to be superimposed on the d-axis current command and a q-axis correction current command to be superimposed on the q-axis current command,
Among the harmonic torque components extracted from the actual torque of the rotating electric machine, a torque having the opposite phase of the (2NM)-order harmonic torque component (M is any natural number) of the circumferential excitation force is used as an excitation force reduction torque. year,
The peak value on the larger side of the excitation force reduction torque is defined as a first torque,
The smaller peak value of the excitation force reduction torque is defined as a second torque;
A curve representing a combination of a basic d-axis current and a q-axis current for outputting an arbitrary torque in the orthogonal vector coordinate system is a prescribed basic control line,
A curve representing a combination of the d-axis current and the q-axis current capable of outputting a constant torque in the orthogonal vector coordinate system is an equal torque line,
In the orthogonal vector coordinate system, the intersection of the basic control line and the equal torque line of the target torque is set as a reference point,
In the orthogonal vector coordinate system, a straight line passing through the reference point is a correction straight line,
The intersection of the first equal torque line, which is the equal torque line of the first torque, and the correction straight line is the first intersection, and the second equal torque line, which is the equal torque line of the second torque, intersects with the correction straight line. With the intersection as the second intersection,
The correction current command calculation unit calculates the d-axis current and the q-axis current between the first intersection point and the second intersection point of the correction straight line, which can be set in plurality, to reduce the excitation force reduction torque. The correction straight line having a slope such that both the radial excitation force and the circumferential excitation force are reduced compared to before correction when vibrated at a frequency is set as a target correction straight line, and the above-mentioned about the target correction straight line is The d-axis current and the q-axis current, which oscillate at the frequency of the excitation force reduction torque with peaks at the first intersection and the second intersection, are used as the correction current command.

In the orthogonal vector coordinate system, there are multiple correction straight lines passing through the reference point. Therefore, there are also a plurality of d-axis currents and q-axis currents that oscillate at the frequency of the excitation force reduction torque with peaks at the first intersection point and the second intersection point. Therefore, depending on the d-axis current and q-axis current that are superimposed on the current command as a correction current command, even if either the radial excitation force or the circumferential excitation force is reduced, the other may increase. be. According to this configuration, among the plurality of correction straight lines, the correction straight line in which both the radial excitation force and the circumferential excitation force are reduced compared to before correction is set as the target correction straight line. Then, the d-axis current and the q-axis current that oscillate at the frequency of the excitation force reduction torque with peaks at the first and second intersections of the target correction straight line become correction current commands. Therefore, when this correction current command is superimposed on the current command, both the radial excitation force and the circumferential excitation force can be reduced. Thus, according to this configuration, both the circumferential excitation force and the radial excitation force can be appropriately reduced.

Further features and advantages of the rotating electrical machine control system will become clear from the following description of an embodiment with reference to the drawings.

Block diagram showing an example of a rotating electric machine drive system Block diagram showing an example of a rotating electric machine control device Waveform diagram showing an example of the actual torque of a rotating electric machine Waveform diagram of the 6th harmonic torque component and its opposite phase excitation force reduction torque Diagram showing the relationship between torque and current command in the dq-axis orthogonal coordinate system Waveform diagram showing an example of q-axis correction current command Waveform diagram showing an example of d-axis correction current command Diagram showing an example of multiple correction straight lines that can be set in the dq-axis orthogonal coordinate system Explanatory diagram showing how to determine the target correction straight line Diagram showing an example of multiple correction straight lines that can be set in the dq-axis orthogonal coordinate system Explanatory diagram showing how to determine the target correction straight line Composite waveform diagram of 6th, 12th, 18th, and 24th harmonic torque components Waveform diagram of excitation force reduction torque of 6th, 12th, 18th, and 24th harmonics Diagram showing the effect of reducing radial excitation force Diagram showing the effect of reducing circumferential excitation force

Hereinafter, embodiments of a rotating electrical machine control system will be described based on the drawings. A rotating electric machine control system, for example, drives and controls a rotating electric machine that serves as a driving force source for a vehicle. The block diagram in FIG. 1 schematically shows the system configuration of a rotating electrical machine drive device 100 including a rotating electrical machine control device 10 (MG-CTRL). In a broad sense, the rotating electrical machine drive device 100 corresponds to a rotating electrical machine control system, and in a narrow sense, the rotating electrical machine control device 10 corresponds to a rotating electrical machine control system.

A rotating electrical machine 80 to be driven by a rotating electrical machine control system includes a stator 81 in which a stator coil 83 of N phase (N is an arbitrary natural number) is arranged in a stator core 85, and a rotor 82 in which a permanent magnet 84 is arranged in a rotor core 86. This is a permanent magnet synchronous motor (PMSM) with Although FIG. 1 illustrates a four-pole (two-pole pair) rotor 82 having four magnetic poles (two north poles and two south poles), this is merely a schematic illustration and is not intended for the invention. It is not limited. The same applies to the stator 81, and although FIG. 1 illustrates a configuration in which three-phase stator coils 83 are short-circuited at the neutral point, the number of phases, the wiring method, and the winding method of the stator coil 83, etc. It does not limit the invention. Note that the rotating electric machine 80 can function both as an electric motor and as a generator.

As shown in FIG. 1, the rotating electrical machine drive device 100 includes an inverter 50 that converts power between DC power and multi-phase AC power. The inverter 50 is connected to an AC rotating electric machine 80 and a DC power source 41, and converts power between multiple phases of AC and DC. The DC power source 41 is configured by, for example, a rechargeable secondary battery (battery) such as a lithium ion battery, an electric double layer capacitor, or the like. When the rotating electric machine 80 is a driving force source for a vehicle, the DC power supply 41 is a high voltage, large capacity DC power supply, and the rated power supply voltage is, for example, 200 to 400 [V]. The DC side of the inverter 50 is equipped with a smoothing capacitor (DC link capacitor 42) that smoothes the voltage between the positive and negative electrodes (DC link voltage Vdc).

The inverter 50 includes a plurality of switching elements 51. The switching element 51 includes an IGBT (Insulated Gate Bipolar Transistor), a power MOSFET (Metal Oxide Semiconductor Field Effect Transistor), a SiC-MOSFET (Silicon Carbide - Metal Oxide Semiconductor FET), a SiC-SIT (SiC - Static Induction Transistor), and a GaN - It is preferable to apply a power semiconductor element that can operate at high frequencies, such as a MOSFET (Gallium Nitride - MOSFET). FIG. 1 illustrates an example in which an IGBT is used as the switching element 51. Note that each switching element 51 is provided with a freewheel diode 53 in parallel with the direction from the negative electrode to the positive electrode (direction from the lower stage side to the upper stage side) as the forward direction.

As shown in FIG. 1, the inverter 50 is controlled by the rotating electric machine control device 10. The rotating electric machine control device 10 is constructed using a logic circuit such as a microcomputer as a core component. For example, the rotating electrical machine control device 10 receives a target torque (torque command) for the rotating electrical machine 80 provided as a request signal from another control device such as a vehicle control device 90 (VHL-CTRL), which is one of the upper control devices. T ^* : see FIG. 2, etc.), current feedback control using a vector control method is performed to control the rotating electric machine 80 via the inverter 50. In the vector control method, currents flowing through the stator coils 83 of each of the three phases (N phase) of an AC motor (Iu, Iv, Iw: see FIG. 2) are controlled by the magnetic field generated by the permanent magnet 84 disposed on the rotor 82. Feedback control is performed by converting the coordinates into vector components of the d-axis, which is the direction, and the q-axis, which is the direction perpendicular to the d-axis (the direction advanced by π/2 in electrical angle with respect to the direction of the magnetic field).

The actual current flowing through the stator coil 83 of each phase of the rotating electrical machine 80 is detected by the current sensor 43, and the rotating electrical machine control device 10 acquires the detection result. Note that this example shows a configuration in which three-phase alternating current is detected, but for example, in the case of three-phase alternating current, the three phases are balanced and the sum of their instantaneous values is zero, so only two phases can be detected. The remaining one phase may be acquired by the rotating electric machine control device 10 through calculation after detecting the current. Further, the magnetic pole position (electrical angle θ) and rotational speed (angular velocity ω) of the rotor 82 of the rotating electrical machine 80 at each point in time are detected by a rotation sensor 44 such as a resolver, and the rotating electrical machine control device 10 Get detection results. The rotating electrical machine control device 10 uses the detection results of the current sensor 43 and rotation sensor 44 to execute current feedback control.

As shown in FIG. 2, the rotating electrical machine control device 10 is configured to have various functional units for current feedback control, and each functional unit is configured using hardware such as a microcomputer and software (program). This will be realized through collaboration with In this embodiment, the rotating electrical machine control device 10 includes a current command calculation section 1, a voltage command calculation section 2, a two-phase three-phase coordinate conversion section 3, a three-phase two-phase coordinate conversion section 4, and a modulation section 5. , a corrected current command map 6, and a corrected current command calculation section 7.

The current command calculation unit 1 calculates a target current (current command I ^* ) to be passed through the stator coil 83 of the rotating electrical machine 80 based on the torque command T ^* (required torque) transmitted from the vehicle control device 90. As described above, since the rotating electrical machine control device 10 performs feedback control on the rotating electrical machine 80 in the dq-axis orthogonal vector coordinate system, the current command calculation unit 1 calculates the d-axis current command Id* and the q-axis current command I ^* as the current command I ^* . Calculate current command Iq ^* .

The voltage command calculation unit 2 generates a voltage command to be applied to the inverter 50 based on the deviation between the current command I ^* and the current flowing through the stator coil 83 (U-phase current Iu, V-phase current Iv, W-phase current Iw). A certain voltage command V ^* is calculated. Here, an example is shown in which the voltage command calculation section 2 is configured with a proportional-integral controller (PI), but the voltage command calculation section 2 is configured with a proportional-differential-integral controller (PID). You can leave it there.

The three-phase currents (U-phase current Iu, V-phase current Iv, W-phase current Iw) flowing through the stator coil 83 detected by the current sensor 43 (SEN-I) are converted to the dq-axis in the three-phase two-phase coordinate conversion unit 4. It is converted into two-phase currents (d-axis current Id, q-axis current Iq) in a vector coordinate system. The three-phase two-phase coordinate conversion unit 4 performs coordinate conversion based on the magnetic pole position (electrical angle θ) of the rotor 82 at each point in time detected by the rotation sensor 44 (SEN-R). In FIG. 2, between the current command calculation unit 1 and the voltage command calculation unit 2, two-phase currents (d-axis current Id, q-axis current Iq) converted by the three-phase two-phase coordinate conversion unit 4, and d Although an example is shown in which the deviation between the axis current command Id ^* and the q-axis current command Iq ^* is calculated, the calculation of the deviation may be performed by the voltage command calculation unit 2. The voltage command calculation unit 2 calculates the d-axis voltage command Vd* based on the deviation between the d-axis current command Id ^* and the d-axis current Id and the rotation speed (angular velocity ω), and also calculates the d-axis voltage command Vd ^* and the q-axis current command Iq ^* and q. A q-axis voltage command Vq ^* is calculated based on the deviation from the shaft current Iq and the rotational speed (angular velocity ω).

The 2-phase 3-phase coordinate conversion unit 3 converts the 2-phase voltage command V ^* (d-axis voltage command Vd ^* , q-axis voltage command Vq ^* ) of the dq-axis vector coordinate system into a 3-phase voltage command corresponding to the 3-phase inverter 50. The coordinates are converted into voltage commands V ^* (U-phase voltage commands Vu ^* , V-phase voltage commands Vv ^* , W-phase voltage commands Vw ^* ). ^The modulator 5 generates three ^- phase switching ^control signals ⁽ A U-phase switching control signal Su, a V-phase switching control signal Sv, and a W-phase switching control signal Sw) are generated. Here, an example is shown in which the modulator 5 generates the switching control signal by pulse width modulation (PWM) control. In FIG. 2, the switching control signals are simplified to three (Su, Sv, Sw), but the modulation section 5 outputs six switching control signals (Su, Sv, Sw) corresponding to the six switching elements 51 of the inverter 50. A U-phase upper-stage switching control signal, a U-phase lower-stage switching control signal, etc.) are generated.

A control terminal (for example, a gate terminal of an IGBT) of each switching element 51 constituting the inverter 50 is connected to the rotating electric machine control device 10 via the drive circuit 20, and the switching of each switching element 51 is individually controlled. . As described above, the rotating electric machine control device 10 that generates switching control signals is configured with a microcomputer as its core, and its operating voltage is, for example, 5 [V] or 3.3 [V]. On the other hand, as described above, the inverter 50 is connected to the DC power supply 41 whose rated power supply voltage is, for example, 200 to 400 [V], and the control terminal of the switching element 51 is connected to a drive voltage of, for example, 15 to 20 [V]. It is necessary to input a signal. The drive circuit 20 increases the driving capability (for example, the ability to operate subsequent circuits, such as voltage amplitude and output current) of the switching control signal generated by the rotating electrical machine control device 10 and relays the signal to the inverter 50 .

By the way, in the rotating electric machine 80, when the rotor 82 rotates, the number of magnetic flux linkages changes, causing fluctuations in torque. This excitation force (torque ripple) generated in the circumferential direction C shown in FIG. 1 is referred to as a circumferential excitation force ("Fc" shown in FIG. 9). Further, due to the attractive force and repulsive force between the stator core 85 and the permanent magnets 84, fluctuations in torque occur also in the radial direction R shown in FIG. This excitation force in the radial direction R is referred to as a radial excitation force ("Fr" shown in FIG. 9). If the rotor 82 vibrates due to these excitation forces, problems such as generation of audible sounds may occur, so it is preferable to reduce these excitation forces. These excitation forces can be reduced, for example, by generating torque that cancels out these excitation forces. However, if one of the circumferential excitation force and the radial excitation force is reduced, the other may increase, so it is necessary to use an appropriate torque that can reduce both the circumferential excitation force and the radial excitation force. It is required to generate

As shown in FIG. 2, the rotating electrical machine control device 10 of the present embodiment calculates a correction current command Iac (d-axis correction current command Idac, q-axis correction current command Iqac) for generating such appropriate torque. The corrected current command calculation unit 7 (CRCT) is provided. The corrected current command calculation unit 7 calculates the corrected current command Iac based on the torque command T ^* and the rotational speed (angular velocity ω) with reference to the corrected current command map 6 (MAP). The d-axis corrected current command Idac and the q-axis corrected current command Iqac are superimposed on the d-axis current command Id ^* and the q-axis current command Iq ^* calculated by the current command calculation unit 1, respectively, to obtain the corrected d-axis current command Id ^{*. *} and the corrected q-axis current command Iq ^** . The voltage command calculation unit 2 calculates the d-axis voltage command Vd ^* based on the deviation between the corrected d-axis current command Id ^** and the d-axis current Id and the rotational speed (angular velocity ω). At the same time, a q-axis voltage command Vq ^* is calculated based on the deviation between the corrected q-axis current command Iq ^** and the q-axis current Iq and the rotational speed (angular velocity ω). Thereby, the rotating electrical machine 80 can output torque for reducing the excitation force. In other words, the correction current command calculation unit 7 calculates the radial excitation force, which is the variation in the radial direction R of the rotating electrical machine 80 (the variation in force in the radial direction R), and the circumferential direction, which is the torque variation in the circumferential direction C of the rotating electrical machine 80. The d-axis correction current command Idac is superimposed on the d-axis current command Id ^* and the q-axis correction is superimposed on the q-axis current command Iq ^* as the correction current command Iac superimposed on the current command I ^* to reduce the excitation force. This is a functional unit that calculates the current command Iqac.

The torque for reducing the excitation force (excitation force reduction torque) is the harmonic torque component extracted from the actual torque T of the rotating electric machine 80 as shown in FIG. ) harmonic torque component (M is an arbitrary natural number). In this embodiment, since N indicating the number of AC phases is 3, for example, when "M=1", of the harmonic torque components extracted from the actual torque T, the 6th harmonic torque component The opposite phase torque becomes the excitation force reduction torque. The solid line waveform in FIG. 4 represents the 6th harmonic torque component extracted from the actual torque T shown in FIG. Indicates torque.

Here, the peak value on the larger side of the excitation force reduction torque is defined as a first torque T1, and the peak value on the smaller side of the excitation force reduction torque is defined as a second torque T2. Further, the average value of the excitation force reduction torque is defined as average torque Tav. Note that this average torque Tav corresponds to the excitation force reduction torque and the DC component of the actual torque T of the rotating electric machine 80, so if the current feedback control based on the torque command T ^* is appropriately executed, the torque The value is approximately equal to the command T ^* (target torque).

FIG. 5 shows the relationship between torque and current command I ^* in the dq-axis orthogonal coordinate system. The curve indicated by the symbol “30” in FIG. 5 represents the combination of the d-axis current and the q-axis current that can output a constant torque in the dq-axis orthogonal vector coordinate system (the vector locus of the current in the dq-orthogonal vector coordinate system). This is the equal torque line. The symbol "31" is a first equal torque line that is the equal torque line 30 of the first torque T1 mentioned above, and the symbol "32" is a second equal torque line that is the equal torque line 30 of the second torque T2. Further, the reference numeral "33" is a reference equal torque line which is the equal torque line 30 of the target torque (torque command T ^* : see FIG. 2, etc.) of the rotating electric machine 80 mentioned above.

The code “60” indicates the combination of the d-axis current and the q-axis current (the combination of the current in the dq orthogonal vector coordinate system) when controlling the rotating electric machine 80 under standard conditions (hereinafter this control is referred to as “basic control”). This is the basic control line that shows the vector trajectory). Generally, the basic control line 60 is a vector locus indicating an optimal combination of d-axis current and q-axis current to output an arbitrary torque in a dq-axis orthogonal vector coordinate system. As an example, the basic control line 60 can be a maximum torque line or a maximum efficiency line that indicates a vector locus of a combination of a d-axis current and a q-axis current that can output each torque with the highest efficiency.

When the circumferential excitation force and the radial excitation force are not considered, that is, when the d-axis current command Id ^* and the q-axis current command Iq ^* are simply set according to the torque command T ^* , as shown in FIG. The current value at the reference point P0 shown is set. This reference point P0 is the intersection of the basic control line 60 and the equal torque line (in this case, the reference equal torque line 33) corresponding to the torque command T ^* (target torque) in the dq-axis orthogonal vector coordinate system. In this embodiment, in order to reduce the circumferential excitation force and the radial excitation force, as described above with reference to FIG. 4, the correction current command Iac that can output the excitation force reduction torque is It is superimposed on the current command Id ^* and the q-axis current command Iq ^* . In other words, for each of the d-axis current command Id ^* and the q-axis current command Iq ^* (current command I ^* according to the torque command T ^* ), which are composed of DC components, the AC component (here, the 6th harmonic component) is ) are respectively superimposed on the d-axis correction current command Idac and the q-axis correction current command Iqac.

The d-axis correction current command Idac and the q-axis correction current command Iqac are an excitation force reduction torque that oscillates between the first torque T1 and the second torque T2 via the average torque Tav (target torque) (see FIG. 4). ) are the d-axis current and q-axis current that can be output. Therefore, the vector locus of these d-axis currents and q-axis currents in the dq-axis orthogonal vector coordinate system is a straight line (line segment) that passes through the reference point P0 and connects the first equal torque line 31 and the second equal torque line 32. ) is preferable. This straight line passing through the reference point P0 is hereinafter referred to as a corrected straight line K. Further, the intersection between the first equal torque line 31 and the correction straight line K is called a first intersection P1, and the intersection between the second equal torque line 32 and the correction straight line K is called a second intersection P2.

In principle, the correction straight line K can be set infinitely. FIG. 5 illustrates three correction straight lines K (K11, K12, K13). The correction straight line K shown as "K11" changes the torque by changing the current along the q-axis, and the correction straight line K shown as "K12" changes the torque by changing the current along the d-axis. The corrected straight line K indicated by "K13" shows a form in which the torque is changed by changing the current along a direction inclined with respect to the d-axis and the q-axis. FIG. 6 shows an example of the q-axis current (q-axis correction current command Iqac) in which these three correction straight lines K (K11, K12, K13) form a vector locus, and FIG. K11, K12, K13) illustrate the d-axis current (d-axis correction current command Idac) whose vector locus is the vector locus.

When the vector locus of the correction current command Iac becomes a correction straight line K shown by "K11", the q-axis correction current command Iqac is divided into "q11a" and "q11b" with "qdc" in between, as shown by the two-dot chain line in FIG. ” It becomes a waveform that oscillates between Here, "qdc" is the value of the q-axis current command Iq ^* according to the torque command T ^* (the value of the q-axis current at the reference point P0). Since the correction straight line K indicated by "K11" is parallel to the q-axis, ^the d-axis correction current command Idac is a constant value, and its value is ^the value ( The value of the d-axis current at the reference point P0 is "ddc".

When the vector locus of the correction current command Iac becomes a correction straight line K indicated by "K12", the d-axis correction current command Idac is divided into "d12a" and "d12b" with "ddc" in between, as shown by the dashed line in FIG. ” It becomes a waveform that oscillates between Since the correction straight line K indicated by "K12" is parallel to the d-axis, the q-axis correction current command Iqac is a constant value, and the value is "qdc" which is the value of the q-axis current at the reference point P0. .

When the vector locus of the correction current command Iac becomes a correction straight line K indicated by "K13", the q-axis correction current command Iqac is divided into "q13a" and "q13b" with "qdc" in between, as shown by the solid line in FIG. The waveform oscillates between . The amplitude A (wave height a) of the q-axis correction current command Iqac corresponding to the correction straight line K indicated by "K13" is the amplitude A (wave height a) of the q-axis correction current command Iqac corresponding to the correction straight line K indicated by "K11". It is smaller than . Further, the d-axis correction current command Idac has a waveform that oscillates between "d13a" and "d13b" with "ddc" in between, as shown by the solid line in FIG. The amplitude B (wave height b) of the d-axis correction current command Idac corresponding to the correction straight line K indicated by "K13" is the amplitude B (wave height b) of the d-axis correction current command Idac corresponding to the correction straight line K indicated by "K12". It is smaller than .

Here, the q-axis correction current command Iqac and the d-axis correction current command Idac are expressed by the following equations (1) and (2). “φ6” indicates the phase of the sixth harmonic.

Iqac=Acos(6θ+φ6+π)...(1)
Idac=Bcos(6θ+φ6)...(2)

As described above, the q-axis current command Iq ^* and the d-axis current command Id ^* calculated in the current command calculation unit 1 are "Iqdc" and "Iddc", respectively. Therefore, the corrected current command (corrected q-axis current command Iq ^** and corrected d-axis current command Id ^** ) based on the corrected current command Iac is expressed by the following equations (3) and (4).

Iq ^** = Iq ^* + Iqac = Iqdc + Acos (6θ + φ6 + π) ... (3)
Id ^** = Id ^* + Idac = Iddc + Bcos (6θ + φ6) ... (4)

As described above, the waveforms of the d-axis correction current command Idac and the q-axis correction current command Iqac become different depending on the correction straight line K, which can be set in plural numbers. The degree of influence of the circumferential excitation force and the radial excitation force differs between the d-axis current and the q-axis current. Therefore, depending on the combination of the d-axis correction current command Idac and the q-axis correction current command Iqac, even if one of the circumferential excitation force and the radial excitation force can be reduced, the other may increase, or the other There is a possibility that the reduction may not be sufficient. Therefore, it is preferable to appropriately set a combination of the d-axis correction current command Idac and the q-axis correction current command Iqac that can reduce both the circumferential excitation force and the radial excitation force.

In this embodiment, the actual torque T of the rotating electric machine 80 is obtained through experiment or simulation, and the harmonic torque component to be reduced is extracted, and furthermore, the actual torque T of the rotating electric machine 80 is acquired through experiment or simulation, and the appropriate torque T for outputting the excitation force reduction torque is determined through experiment or simulation. A combination of the d-axis correction current command Idac and the q-axis correction current command Iqac is defined. This combination is stored, for example, in the corrected current command map 6, and the corrected current command calculation unit 7 refers to the corrected current command map 6 and calculates the torque command T ^* , the rotational speed (angular velocity ω) of the rotating electric machine 80, the magnetic pole position ( A d-axis correction current command Idac and a q-axis correction current command Iqac are calculated based on the electrical angle θ) and the like.

The principle of determining an appropriate combination of the d-axis correction current command Idac and the q-axis correction current command Iqac to output the excitation force reduction torque will be described below with reference to FIGS. 8 and 9 as well. FIG. 8 shows an example of a plurality of correction straight lines K that can be set in the dq-axis orthogonal coordinate system, and FIG. 9 shows an example of the d-axis correction current command Idac and the q-axis correction current command A method for determining a target correction straight line KT serving as a vector locus of Iqac is shown.

FIG. 8 illustrates five correction straight lines K (K2, K3, K4, K5, K6) passing through the reference point P0. The correction straight line K indicated by "K2" is a form in which the torque is changed by changing the current along the basic control line 60 (for example, a form in which the torque is changed to be most efficient). Similar to "K11" in FIG. 5, the correction straight line K shown as "K3" has a form in which the torque is changed by changing the current along the q-axis, and the correction straight line K shown as "K6" has a form in which the torque is changed by changing the current along the q axis. Similar to "K12", this is a form in which the current is changed along the d-axis to change the torque. The correction straight lines K indicated by "K4" and "K5", similar to "K13" in FIG. It shows. Note that the correction straight line K (K2) in which the torque changes along the basic control line 60 is also a form in which the torque is changed by changing the current along the direction inclined with respect to the d-axis and the q-axis.

FIG. 9 shows the radial direction when the rotating electrical machine 80 is controlled by superimposing the correction current command Iac corresponding to the five correction straight lines K (K2, K3, K4, K5, K6) on the current command I ^* . The relationship between the excitation force Fr and each correction straight line K and the relationship between the circumferential excitation force Fc and each correction straight line K are shown. Each correction straight line K (K2, K3, K4, K5, K6) can be specified by the slope of the line segment, but in FIGS. 8 and 9, the q-axis is used as an element equivalent to the slope of the correction straight line K. Each correction straight line K (K2, K3, K4, K5, K6) is specified based on the angle with the correction straight line K in the counterclockwise direction (deviation angle α with respect to the q axis).

As shown in FIG. 8, the deviation angle α of the correction straight line “K3” is “α3=0”, the deviation angle α of the correction straight line “K2” is “α2”, and the deviation angle α of the correction straight line “K4” is "α4", the deviation angle α of the correction straight line "K5" is "α5", and the deviation angle of the correction straight line "K6" is "α6=π/2". In Figure 9, the excitation force on each correction straight line K (K2, K3, K4, K5, K6) is plotted in correspondence with the deviation angle α, and the deviation angle α is between “0 and π/2”. It is possible to estimate the change in the excitation force when the excitation force changes. FIG. 9 also shows the radial excitation force Fr and the circumferential excitation force Fc when the rotating electric machine 80 is controlled by the DC component current command I ^* at the reference point P0 without superimposing the correction current command Iac. is shown by a line parallel to the horizontal axis, so that the reduction amount (radial direction reduction amount Yr, circumferential direction reduction amount Yc) of the excitation force in each correction straight line K (K2, K3, K4, K5, K6) can be compared. I have to.

The deviation angle α is calculated using the amplitude (A, B) or wave height (a, b) of the correction current command Iac described above with reference to FIGS. 6 and 7 and equations (1) to (4). can be expressed by the following formula (5) or formula (6). Since the deviation angle α has a value according to the ratio of the amplitude (A, B) or the wave height (a, b), it can be said to be an index equivalent to the slope of the correction straight line K.

α = arctan(B/A)...(5)
α = arctan(b/a)...(6)

As shown in FIG. 9, the radial excitation force Fr is the smallest when the correction current command Iac corresponds to the correction straight line K indicated by "K5". In this case, compared to the case where the rotating electrical machine 80 is controlled without superimposing the correction current command Iac, the radial excitation force Fr is reduced by the radial reduction amount Yr. Further, the circumferential excitation force Fc is the smallest in the case of the correction current command Iac corresponding to the correction straight line K indicated by "K5" and "K6". In this case, the circumferential excitation force Fc is reduced by the circumferential reduction amount Yc compared to the case where the rotating electric machine 80 is controlled without superimposing the correction current command Iac. Here, both the radial excitation force Fr and the circumferential excitation force Fc are reduced in the case of the correction current command Iac corresponding to the correction straight line K indicated by "K5". Therefore, the correction straight line K indicated by "K5" is set as the target correction straight line KT, and the d-axis current and q-axis current whose target correction straight line KT is the vector trajectory are the d-axis correction current command Idac and the q-axis correction current The command Iqac is set.

Note that if there is a correction straight line K in which both the radial excitation force Fr and the circumferential excitation force Fc are reduced to the same extent, any correction straight line K may be used as the target correction straight line KT. Further, in such a case, the correction straight line K having the slope (deviation angle α) closest to the correction straight line K along the basic control line 60 may be used as the target correction straight line KT.

In addition, if the slope of the correction straight line K where the radial excitation force Fr is the smallest and the slope of the correction straight line K where the circumferential excitation force Fc is the smallest are different, the radial excitation force Fr and the circumferential excitation force The correction straight line K having the inclination (deviation angle α) that gives the smallest sum of the vibration force Fc may be used as the target correction straight line KT. Alternatively, if there is a difference in the reduction effect between the radial excitation force Fr and the circumferential excitation force Fc, for example, the slope (deviation angle α) at which (γ×Fr+δ×Fc) is the smallest is determined. The correction straight line K may be used as the target correction straight line KT. Here, γ and δ are arbitrary coefficients. For example, if it is preferable to reduce the circumferential excitation force Fc rather than the radial excitation force Fr, the coefficient may be set so that δ>γ. Conversely, if it is preferable to reduce the radial excitation force Fr rather than the circumferential excitation force Fc, the coefficient may be set so that γ>δ.

Also, referring to FIG. 9, when the corrected current command Iac that oscillates along the corrected straight line K is superimposed on the current command I ^* , the circumferential It can be seen that the directional excitation force Fc is reduced. On the other hand, depending on the deviation angle α, the radial excitation force Fr may increase compared to the case where the correction current command Iac is not superimposed on the current command I ^* . That is, since the excitation force reduction torque is in the opposite phase of the harmonic torque component of the circumferential excitation force Fc, a reduction effect tends to appear on the circumferential excitation force Fc regardless of the deviation angle α. Therefore, if the slope of the correction straight line K where the radial excitation force Fr is the smallest is different from the slope of the correction straight line K where the circumferential excitation force Fc is the smallest, the effect of reducing the radial excitation force Fr can be reduced. Priority may be given to the correction straight line K where the radial excitation force Fr is the smallest as the target correction straight line KT.

Furthermore, as described above, depending on the deviation angle α, the radial excitation force Fr may increase compared to the case where the correction current command Iac is not superimposed on the current command I ^* . be. In view of this point, if the slope of the correction straight line K where the radial excitation force Fr is the smallest is different from the slope of the correction straight line K where the circumferential excitation force Fc is the smallest, the circumferential excitation force Fc is the smallest, and the radial excitation force Fr does not deteriorate (the radial excitation force Fr does not increase compared to the case where the correction current command Iac is not superimposed on the current command I ^* ).The correction straight line K is the target correction straight line. It may also be KT.

In the embodiment described above with reference to FIGS. 8 and 9, the correction straight line "K5" set as the target correction straight line KT is not the correction straight line K along the basic control line 60. In this way, the correction straight line K along which both the radial excitation force Fr and the circumferential excitation force Fc are reduced is not necessarily along the basic control line 60. Furthermore, as shown in FIG. 9, for example, the radial excitation force Fr may actually increase due to the superimposition of the correction current command Iac (such as in the case of correction straight lines K indicated by "K2", "K3", and "K6"). ). Further, although illustration is omitted, the deviation angle α that specifies the correction straight line K where both the radial excitation force Fr and the circumferential excitation force Fc are reduced varies depending on the operating state of the rotating electric machine 80. Therefore, it is preferable that the target correction straight line KT is appropriately set according to the operating state of the rotating electric machine 80. For example, it is preferable that different target correction straight lines KT be set depending on the operating point that is a combination of the rotational speed (angular velocity ω) and target torque (torque command T ^* ) of the rotating electric machine 80.

In this embodiment, as described above, the corrected current command map 6 is constructed by acquiring the actual torque of the rotating electric machine 80 through experiment or simulation. The correction current command calculation unit 7 refers to the correction current command map 6 and calculates the d-axis current and the q-axis current between the first intersection P1 and the second intersection P2 among the correction straight lines K that can be set in plurality. The slope (deviation angle α) at which both the radial excitation force Fr and the circumferential excitation force Fc are reduced compared to before correction when vibrating at the frequency (1/φ6) of the excitation force reduction torque is calculated. The corrected current command Iac is determined by using the corrected straight line K as the target corrected straight line KT. That is, the correction current command calculation unit 7 generates a d-axis current (Idac) that oscillates at the frequency (1/φ6) of the excitation force reduction torque with peaks at the first intersection P1 and the second intersection P2 for the target correction straight line KT. and the q-axis current (Iqac) as a correction current command Iac.

By the way, in the above, since the influence of the radial excitation force Fr and the circumferential excitation force Fc is large on the 6th harmonic torque component, the excitation force is a torque with the opposite phase of the 6th harmonic torque component. An example is given in which the correction current command Iac is set based on the reduced torque. However, the correction current command calculation unit 7 may calculate the correction current command Iac for each of a plurality of harmonic torque components having different values of M in the (2NM)-order harmonic torque components. The 6th harmonic torque component is used when "M = 1", but for example, harmonic torque components (12th, 18th, 24th) of "M = 2, 3, 4" may also be used. The corrected current command Iac may be calculated by

In the above, the correction straight line K and the target correction straight line KT for the sixth harmonic torque component have been explained with reference to FIGS. 8 and 9. For example, the correction current command Iac can be similarly calculated for the 12th harmonic torque component based on the optimum target correction straight line TK. Hereinafter, the correction straight line K and the target correction straight line KT for the 12th harmonic torque component will be explained with reference to FIGS. 10 and 11. FIG. 10 corresponds to FIG. 8 and shows an example of a plurality of correction straight lines K that can be set in the dq-axis orthogonal coordinate system. Further, FIG. 11 corresponds to FIG. 9 and shows a method for determining the target correction straight line KT.

Similarly to FIG. 8, FIG. 10 illustrates correction straight lines K (here, seven lines: K2, K3, K4, K5, K6, K7, K8) passing through the reference point P0. FIG. 9 shows a case where a rotating electric machine 80 is controlled by superimposing a correction current command Iac corresponding to these seven correction straight lines K (K2, K3, K4, K5, K6, K7, K8) on a current command I ^* . The relationship between the radial excitation force Fr and each correction straight line K and the relationship between the circumferential excitation force Fc and each correction straight line K are shown in FIG. Similarly to FIGS. 8 and 9, in FIGS. 10 and 11, as an element equivalent to the inclination of the correction straight line K, the angle with the correction straight line K in the counterclockwise direction with the q-axis as a reference (deviation from the q-axis) Each correction straight line K (K2, K3, K4, K5, K6, K7, K8) is specified by the angle α).

Similar to FIG. 9, in FIG. 11, the excitation force on each correction straight line K (K2, K3, K4, K5, K6, K7, K8) is plotted corresponding to the deviation angle α, and the plotted points A characteristic curve is drawn by connecting them. The characteristic curve shows the change in the excitation force when the deflection angle α changes between “0 and π/2”. In FIG. 11, the broken line parallel to the horizontal axis indicates the radial excitation force Fr and the circumferential direction when the rotating electric machine 80 is controlled by the DC component current command I ^* at the reference point P0 without superimposing the correction current command Iac. It shows the excitation force Fc. Based on this broken line, the reduction amount of the excitation force (radial direction reduction amount Yr, circumferential direction reduction amount Yc) in each correction straight line K is compared.

As shown in FIG. 11, the radial excitation force Fr is the smallest when the correction current command Iac corresponds to the correction straight line K indicated by "K7". In this case, compared to the case where the rotating electrical machine 80 is controlled without superimposing the correction current command Iac, the radial excitation force Fr is reduced by the radial reduction amount Yr. Further, the circumferential excitation force Fc is the smallest when the correction current command Iac corresponds to the correction straight line K indicated by "K3" or "K6". In this case, the circumferential excitation force Fc is reduced by the circumferential reduction amount Yc compared to the case where the rotating electric machine 80 is controlled without superimposing the correction current command Iac.

In this way, when the slope of the correction straight line K where the radial excitation force Fr is the smallest and the slope of the correction straight line K where the circumferential excitation force Fc is the smallest are different, as described above, one aspect is possible. As such, the correction straight line K where the radial excitation force Fr is the smallest is set as the target correction straight line KT, giving priority to the effect of reducing the radial excitation force Fr. In this case, it is "K7". In addition, as mentioned above, as another aspect, the circumferential excitation force Fc is the smallest, and the radial excitation force Fr does not deteriorate (compared to the case where the correction current command Iac is not superimposed on the current command I ^* . A correction straight line K in which the radial excitation force Fr does not increase is set as the target correction straight line KT. In this case, the correction straight line K of "K3, K4, K5, K6, K8" where the radial excitation force Fr increases is excluded, and the target correction straight line KT is set from "K2, K7". The correction straight line "K2" is the correction straight line K where the circumferential excitation force Fc is the smallest among the seven correction straight lines K, and is the correction straight line K where the circumferential excitation force Fc is smaller than the correction straight line "K7". It is. However, in the correction straight line "K2", the radial excitation force Fr hardly decreases. Therefore, in this case, the correction straight line "K7" is selected as the target correction straight line KT.

In this way, the correction current command calculation unit 7 may calculate the correction current command Iac for each of a plurality of harmonic torque components having different values of M in the (2NM)-order harmonic torque components. Here, the case of "M=1" (6th order) and the case of "M=2" (12th order) are illustrated, but harmonics such as "M=3, 4" (18th order, 24th order) etc. The corrected current command Iac may also be calculated using the torque component. That is, the correction current command calculation unit 7 calculates each (2NM)-order harmonic based on the target correction straight line KT for each of a plurality of harmonic torque components having different values of M of the (2NM)-order harmonic torque component. It is preferable to calculate the corrected current command Iac for the torque component. That is, it is preferable to obtain different target correction straight lines KT depending on the value of "M" and calculate the correction current command Iac based on each target correction straight line KT.

FIG. 12 shows waveforms of 6th, 12th, 18th, and 24th harmonic torque components extracted from the actual torque T of the rotating electric machine 80 shown in FIG. 3 (6th, 12th, 18th, 24th order synthetic component). Further, FIG. 13 shows the excitation force reduction torque of the opposite phase of the composite component of the 6th, 12th, 18th, and 24th harmonic torque components. FIG. 14 shows the magnitude of the 6th harmonic component of the radial excitation force Fr before correction (reference point P0), the magnitude when only the 6th harmonic component (6f) is used, and the 6th and 12th harmonic components. This is a diagram comparing the magnitudes when using the next, 18th, and 24th harmonic components (6f+12f+18f+24f), and FIG. 15 shows the magnitude of the 6th harmonic component of the circumferential excitation force Fc before correction (reference P0), the magnitude when only the 6th harmonic component (6f) is used, and the magnitude when the 6th, 12th, 18th, and 24th harmonic components (6f + 12f + 18f + 24f) are used are compared. It is a diagram.

As shown in FIG. 14, the radial excitation force Fr is 6th, 12th, 18th, 24th, The second radial reduction amount Yr2 when using the harmonic component (6f+12f+18f+24f) is larger. In addition, as shown in FIG. 15, the circumferential excitation force Fc is also of the 6th, 12th, and 18th order with respect to the first circumferential reduction amount Yc1 when only the 6th harmonic component (6f) is used. , the second circumferential direction reduction amount Yc2 when using the 24th harmonic component (6f+12f+18f+24f) is larger. That is, by using the correction current command Iac corresponding to a plurality of harmonic torque components, the excitation force can be further reduced.

Note that the q-axis correction current command Iqac and the d-axis correction current command Idac corresponding to the above-mentioned equations (1) and (2) are expressed by the following equations (7) and (8). Regarding amplitude A, amplitude B, and phase φ, the 6th harmonic component is indicated by A6, B6, and φ6, the 12th harmonic component is indicated by A12, B12, and φ12, and the 18th harmonic component is indicated by A18, B18. , φ18, and the 24th harmonic components are shown as A24, B24, and φ24. Since the corrected current commands (Id ^** , Iq ^** ) can be easily understood from equations (3), (4), (7), and (8), their description will be omitted.

Iqac=A6cos(6θ+φ6+π)+
A12cos(12θ+φ12+π)+
A18cos(18θ+φ18+π)+
A24cos(24θ+φ24+π)...(7)
Idac=A6cos(6θ+φ6)+
A12cos(12θ+φ12)+
A18cos(18θ+φ18)+
A24cos(24θ+φ24)...(8)

[Overview of embodiment]
The outline of the rotating electric machine control system (10) described above will be briefly described below.

In one embodiment, a permanent magnet type rotating electrical machine (80) driven by N-phase alternating current (N is an arbitrary natural number) is connected to a d-axis along the direction of field magnetic flux by a permanent magnet (84) and the d-axis. A rotating electrical machine control system (10) that performs current feedback control in a vector coordinate system orthogonal to a q-axis that is perpendicular to
The d-axis current command (Id ^* ) and the q-axis current command (Iq ^* ) as the current command (I ^* ), which is the command value of the current flowing through the rotating electrical machine (80), are used as the target torque of the rotating electrical machine (80). (T ^* ); a current command calculation unit (1) that calculates based on (T*);
A radial excitation force (Fr) that is a variation in the radial direction (R) of the rotating electrical machine (80) and a circumferential excitation force (Fc) that is a torque variation in the circumferential direction (C) of the rotating electrical machine (80). The d-axis correction current command (Idac) and the q-axis current command are superimposed on the d-axis current command (Id ^* ) as a correction current command (Iac) superimposed on the current command (I ^* ) to reduce a correction current command calculation unit (7) that calculates a q-axis correction current command (Iqac) to be superimposed on (Iq ^* );
Among the harmonic torque components extracted from the actual torque (T) of the rotating electric machine (80), the inverse of the (2NM)-order harmonic torque component (M is any natural number) of the circumferential excitation force (Fc). Let the phase torque be the excitation force reduction torque,
Let the peak value of the larger side of the excitation force reduction torque be a first torque (T1),
The smaller peak value of the excitation force reduction torque is defined as a second torque (T2),
A curve representing a combination of a basic d-axis current and a q-axis current for outputting an arbitrary torque in the orthogonal vector coordinate system is a prescribed basic control line (60),
A curve representing a combination of the d-axis current and the q-axis current that can output a constant torque in the orthogonal vector coordinate system is an equal torque line (30),
In the orthogonal vector coordinate system, the intersection of the basic control line (60) and the equal torque line (30 (33)) of the target torque (T ^* ) is set as a reference point (P0),
In the orthogonal vector coordinate system, a straight line passing through the reference point (P0) is a correction straight line (K),
The intersection of the first equal torque line (31), which is the equal torque line (30) of the first torque (T1), and the correction straight line (K) is the first intersection (P1), and the second torque (T2) The intersection of the second equal torque line (32) which is (30) in the equal torque line and the correction straight line (K) is set as a second intersection (P2),
The correction current command calculation unit (7) calculates the d-axis current and the a slope at which both the radial excitation force (Fr) and the circumferential excitation force (Fc) are reduced compared to before correction when the q-axis current is vibrated at the frequency of the excitation force reduction torque; The correction straight line (K) having the target correction straight line (KT) is set as a target correction straight line (KT), and the excitation force is reduced with the first intersection point (P1) and the second intersection point (P2) of the target correction straight line (KT) as peaks. The d-axis current and the q-axis current that oscillate at the frequency of torque are the correction current command (Iac).

In the orthogonal vector coordinate system, there are multiple correction straight lines (K) passing through the reference point (P0). Therefore, there are also a plurality of d-axis currents and q-axis currents that oscillate at the frequency of the excitation force reduction torque with peaks at the first intersection point (P1) and the second intersection point (P2). Therefore, depending on the d-axis current and q-axis current superimposed on the current command (I ^* ) as the correction current command (Iac), either the radial excitation force (Fr) or the circumferential excitation force (Fc) Even if one is reduced, the other may increase. According to this configuration, among the plurality of correction straight lines (K), the correction straight line (K) in which both the radial excitation force (Fr) and the circumferential excitation force (Fc) are reduced compared to before correction ) is set as the target correction straight line (KT). Then, the d-axis current and q-axis current, which oscillate at the frequency of the excitation force reduction torque with peaks at the first intersection (P1) and second intersection (P2) of the target correction straight line (KT), are adjusted to the correction current command ( Iac). Therefore, when this correction current command (Iac) is superimposed on the current command (I ^* ), both the radial excitation force (Fr) and the circumferential excitation force (Fc) can be reduced. Thus, according to this configuration, both the circumferential excitation force (Fc) and the radial excitation force (Fr) can be appropriately reduced.

Here, the correction current command calculation unit (7) calculates the correction current command (Iac) for each of the plurality of harmonic torque components having different values of M of the (2NM)-order harmonic torque components. suitable.

It is known that torque fluctuations become particularly large in specific harmonic torque components. Therefore, it is preferable that the correction current command (Iac) is calculated so as to be able to output an excitation force reduction torque that targets at least the harmonic torque component of the order that has the greatest influence. However, the torque fluctuation is a combination of multiple harmonic torque components. According to this configuration, since the correction current command (Iac) is calculated for each of a plurality of harmonic torque components having different values of M in the (2NM)-order harmonic torque component, it is possible to further reduce torque fluctuations. can.

Further, the correction current command calculation unit (7) is configured to perform, based on the target correction straight line (KT) for each of the plurality of harmonic torque components having different values of M of the (2NM) harmonic torque components, It is preferable to calculate the correction current command (Iac) for each of the (2NM) harmonic torque components.

Since torque fluctuation is a combination of multiple harmonic torque components, the correction current command (Iac) is Preferably, it is calculated. The correction current command (Iac) is calculated based on the target correction straight line (KT) from among the plurality of correction straight lines (K) as described above. The slope of the target correction straight line (KT) for calculating the correction current command (Iac) for reducing the harmonic torque component of each order, that is, the slope of the target correction straight line (KT) is the harmonic of each order. It may differ depending on the torque component. When the correction current command calculation unit (7) calculates a correction current command (Iac) for each harmonic torque component of each order based on the target correction straight line (KT) for the harmonic torque component of each order, further Torque fluctuations can be appropriately reduced.

1: Current command calculation unit 7: Correction current command calculation unit 10: Rotating electric machine control device (rotating electric machine control system)
30: Equal torque line 31: First equal torque line 32: Second equal torque line 33: Reference equal torque line 60: Basic control line 80: Rotating electric machine 84: Permanent magnet 100: Rotating electric machine drive device (rotating electric machine control system)
C: Circumferential direction Fc: Circumferential excitation force Fr: Radial excitation force Iac: Correction current command I ^* : Current command Id ^* : d-axis current command Iq ^* : q-axis current command Idac: d-axis correction current command Iqac : q-axis correction current command K : Correction straight line KT : Target correction straight line P0 : Reference point P1 : First intersection P2 : Second intersection R : Radial direction T : Actual torque T ^* : Torque command (required torque)
T1: First torque T2: Second torque α: Deviation angle

Claims (3)

A permanent magnet type rotating electric machine driven by N-phase alternating current (N is any natural number) is defined by an orthogonal vector coordinate system with a d-axis along the direction of field magnetic flux by the permanent magnet and a q-axis orthogonal to the d-axis. A rotating electrical machine control system that performs current feedback control in
a current command calculation unit that calculates a d-axis current command and a q-axis current command as a current command, which is a command value of a current to be passed through the rotating electrical machine, based on a target torque of the rotating electrical machine;
As a correction current command superimposed on the current command in order to reduce a radial excitation force that is a radial variation of the rotating electrical machine and a circumferential excitation force that is a circumferential torque variation of the rotating electrical machine, a correction current command calculation unit that calculates a d-axis correction current command to be superimposed on the d-axis current command and a q-axis correction current command to be superimposed on the q-axis current command,
Among the harmonic torque components extracted from the actual torque of the rotating electric machine, a torque having the opposite phase of the (2NM)-order harmonic torque component (M is any natural number) of the circumferential excitation force is used as an excitation force reduction torque. year,
The peak value on the larger side of the excitation force reduction torque is defined as a first torque,
The smaller peak value of the excitation force reduction torque is defined as a second torque;
A curve representing a combination of a basic d-axis current and a q-axis current for outputting an arbitrary torque in the orthogonal vector coordinate system is a prescribed basic control line,
A curve representing a combination of the d-axis current and the q-axis current capable of outputting a constant torque in the orthogonal vector coordinate system is an equal torque line,
In the orthogonal vector coordinate system, the intersection of the basic control line and the equal torque line of the target torque is set as a reference point,
In the orthogonal vector coordinate system, a straight line passing through the reference point is a correction straight line,
The intersection of the first equal torque line, which is the equal torque line of the first torque, and the correction straight line is the first intersection, and the second equal torque line, which is the equal torque line of the second torque, intersects with the correction straight line. With the intersection as the second intersection,
The correction current command calculation unit calculates the d-axis current and the q-axis current between the first intersection point and the second intersection point of the correction straight line, which can be set in plurality, to reduce the excitation force reduction torque. The correction straight line having a slope such that both the radial excitation force and the circumferential excitation force are reduced compared to before correction when vibrated at a frequency is set as a target correction straight line, and the above-mentioned about the target correction straight line is A rotating electric machine control system in which the d-axis current and the q-axis current, which oscillate at the frequency of the excitation force reduction torque with peaks at the first intersection point and the second intersection point, are used as the correction current command. The rotating electric machine according to claim 1, wherein the correction current command calculation unit calculates the correction current command for each of the plurality of harmonic torque components having different values of M of the (2NM)-order harmonic torque components. control system. The correction current command calculation unit calculates each of the (2NM)-order harmonics based on the target correction straight line for each of the plurality of harmonic torque components having different values of M of the (2NM)-order harmonic torque components. The rotating electric machine control system according to claim 2, wherein the correction current command of the wave torque component is calculated.

Images